xref: /netbsd-src/external/gpl3/gcc.old/dist/gcc/fortran/intrinsic.texi (revision 4c3eb207d36f67d31994830c0a694161fc1ca39b)
1@ignore
2Copyright (C) 2005-2020 Free Software Foundation, Inc.
3This is part of the GNU Fortran manual.
4For copying conditions, see the file gfortran.texi.
5
6Permission is granted to copy, distribute and/or modify this document
7under the terms of the GNU Free Documentation License, Version 1.3 or
8any later version published by the Free Software Foundation; with the
9Invariant Sections being ``Funding Free Software'', the Front-Cover
10Texts being (a) (see below), and with the Back-Cover Texts being (b)
11(see below).  A copy of the license is included in the gfdl(7) man page.
12
13
14Some basic guidelines for editing this document:
15
16  (1) The intrinsic procedures are to be listed in alphabetical order.
17  (2) The generic name is to be used.
18  (3) The specific names are included in the function index and in a
19      table at the end of the node (See ABS entry).
20  (4) Try to maintain the same style for each entry.
21
22
23@end ignore
24
25@tex
26\gdef\acosd{\mathop{\rm acosd}\nolimits}
27\gdef\asind{\mathop{\rm asind}\nolimits}
28\gdef\atand{\mathop{\rm atand}\nolimits}
29\gdef\acos{\mathop{\rm acos}\nolimits}
30\gdef\asin{\mathop{\rm asin}\nolimits}
31\gdef\atan{\mathop{\rm atan}\nolimits}
32\gdef\acosh{\mathop{\rm acosh}\nolimits}
33\gdef\asinh{\mathop{\rm asinh}\nolimits}
34\gdef\atanh{\mathop{\rm atanh}\nolimits}
35\gdef\cosd{\mathop{\rm cosd}\nolimits}
36@end tex
37
38
39@node Intrinsic Procedures
40@chapter Intrinsic Procedures
41@cindex intrinsic procedures
42
43@menu
44* Introduction:         Introduction to Intrinsics
45* @code{ABORT}:         ABORT,     Abort the program
46* @code{ABS}:           ABS,       Absolute value
47* @code{ACCESS}:        ACCESS,    Checks file access modes
48* @code{ACHAR}:         ACHAR,     Character in @acronym{ASCII} collating sequence
49* @code{ACOS}:          ACOS,      Arccosine function
50* @code{ACOSD}:         ACOSD,     Arccosine function, degrees
51* @code{ACOSH}:         ACOSH,     Inverse hyperbolic cosine function
52* @code{ADJUSTL}:       ADJUSTL,   Left adjust a string
53* @code{ADJUSTR}:       ADJUSTR,   Right adjust a string
54* @code{AIMAG}:         AIMAG,     Imaginary part of complex number
55* @code{AINT}:          AINT,      Truncate to a whole number
56* @code{ALARM}:         ALARM,     Set an alarm clock
57* @code{ALL}:           ALL,       Determine if all values are true
58* @code{ALLOCATED}:     ALLOCATED, Status of allocatable entity
59* @code{AND}:           AND,       Bitwise logical AND
60* @code{ANINT}:         ANINT,     Nearest whole number
61* @code{ANY}:           ANY,       Determine if any values are true
62* @code{ASIN}:          ASIN,      Arcsine function
63* @code{ASIND}:         ASIND,     Arcsine function, degrees
64* @code{ASINH}:         ASINH,     Inverse hyperbolic sine function
65* @code{ASSOCIATED}:    ASSOCIATED, Status of a pointer or pointer/target pair
66* @code{ATAN}:          ATAN,      Arctangent function
67* @code{ATAND}:         ATAND,     Arctangent function, degrees
68* @code{ATAN2}:         ATAN2,     Arctangent function
69* @code{ATAN2D}:        ATAN2D,    Arctangent function, degrees
70* @code{ATANH}:         ATANH,     Inverse hyperbolic tangent function
71* @code{ATOMIC_ADD}:    ATOMIC_ADD, Atomic ADD operation
72* @code{ATOMIC_AND}:    ATOMIC_AND, Atomic bitwise AND operation
73* @code{ATOMIC_CAS}:    ATOMIC_CAS, Atomic compare and swap
74* @code{ATOMIC_DEFINE}: ATOMIC_DEFINE, Setting a variable atomically
75* @code{ATOMIC_FETCH_ADD}: ATOMIC_FETCH_ADD, Atomic ADD operation with prior fetch
76* @code{ATOMIC_FETCH_AND}: ATOMIC_FETCH_AND, Atomic bitwise AND operation with prior fetch
77* @code{ATOMIC_FETCH_OR}: ATOMIC_FETCH_OR, Atomic bitwise OR operation with prior fetch
78* @code{ATOMIC_FETCH_XOR}: ATOMIC_FETCH_XOR, Atomic bitwise XOR operation with prior fetch
79* @code{ATOMIC_OR}:     ATOMIC_OR, Atomic bitwise OR operation
80* @code{ATOMIC_REF}:    ATOMIC_REF, Obtaining the value of a variable atomically
81* @code{ATOMIC_XOR}:    ATOMIC_XOR, Atomic bitwise OR operation
82* @code{BACKTRACE}:     BACKTRACE, Show a backtrace
83* @code{BESSEL_J0}:     BESSEL_J0, Bessel function of the first kind of order 0
84* @code{BESSEL_J1}:     BESSEL_J1, Bessel function of the first kind of order 1
85* @code{BESSEL_JN}:     BESSEL_JN, Bessel function of the first kind
86* @code{BESSEL_Y0}:     BESSEL_Y0, Bessel function of the second kind of order 0
87* @code{BESSEL_Y1}:     BESSEL_Y1, Bessel function of the second kind of order 1
88* @code{BESSEL_YN}:     BESSEL_YN, Bessel function of the second kind
89* @code{BGE}:           BGE,       Bitwise greater than or equal to
90* @code{BGT}:           BGT,       Bitwise greater than
91* @code{BIT_SIZE}:      BIT_SIZE,  Bit size inquiry function
92* @code{BLE}:           BLE,       Bitwise less than or equal to
93* @code{BLT}:           BLT,       Bitwise less than
94* @code{BTEST}:         BTEST,     Bit test function
95* @code{C_ASSOCIATED}:  C_ASSOCIATED, Status of a C pointer
96* @code{C_F_POINTER}:   C_F_POINTER, Convert C into Fortran pointer
97* @code{C_F_PROCPOINTER}: C_F_PROCPOINTER, Convert C into Fortran procedure pointer
98* @code{C_FUNLOC}:      C_FUNLOC,  Obtain the C address of a procedure
99* @code{C_LOC}:         C_LOC,     Obtain the C address of an object
100* @code{C_SIZEOF}:      C_SIZEOF,  Size in bytes of an expression
101* @code{CEILING}:       CEILING,   Integer ceiling function
102* @code{CHAR}:          CHAR,      Integer-to-character conversion function
103* @code{CHDIR}:         CHDIR,     Change working directory
104* @code{CHMOD}:         CHMOD,     Change access permissions of files
105* @code{CMPLX}:         CMPLX,     Complex conversion function
106* @code{CO_BROADCAST}:  CO_BROADCAST, Copy a value to all images the current set of images
107* @code{CO_MAX}:        CO_MAX,    Maximal value on the current set of images
108* @code{CO_MIN}:        CO_MIN,    Minimal value on the current set of images
109* @code{CO_REDUCE}:     CO_REDUCE, Reduction of values on the current set of images
110* @code{CO_SUM}:        CO_SUM,    Sum of values on the current set of images
111* @code{COMMAND_ARGUMENT_COUNT}: COMMAND_ARGUMENT_COUNT, Get number of command line arguments
112* @code{COMPILER_OPTIONS}: COMPILER_OPTIONS, Options passed to the compiler
113* @code{COMPILER_VERSION}: COMPILER_VERSION, Compiler version string
114* @code{COMPLEX}:       COMPLEX,   Complex conversion function
115* @code{CONJG}:         CONJG,     Complex conjugate function
116* @code{COS}:           COS,       Cosine function
117* @code{COSD}:          COSD,      Cosine function, degrees
118* @code{COSH}:          COSH,      Hyperbolic cosine function
119* @code{COTAN}:         COTAN,     Cotangent function
120* @code{COTAND}:        COTAND,    Cotangent function, degrees
121* @code{COUNT}:         COUNT,     Count occurrences of TRUE in an array
122* @code{CPU_TIME}:      CPU_TIME,  CPU time subroutine
123* @code{CSHIFT}:        CSHIFT,    Circular shift elements of an array
124* @code{CTIME}:         CTIME,     Subroutine (or function) to convert a time into a string
125* @code{DATE_AND_TIME}: DATE_AND_TIME, Date and time subroutine
126* @code{DBLE}:          DBLE,      Double precision conversion function
127* @code{DCMPLX}:        DCMPLX,    Double complex conversion function
128* @code{DIGITS}:        DIGITS,    Significant digits function
129* @code{DIM}:           DIM,       Positive difference
130* @code{DOT_PRODUCT}:   DOT_PRODUCT, Dot product function
131* @code{DPROD}:         DPROD,     Double product function
132* @code{DREAL}:         DREAL,     Double real part function
133* @code{DSHIFTL}:       DSHIFTL,   Combined left shift
134* @code{DSHIFTR}:       DSHIFTR,   Combined right shift
135* @code{DTIME}:         DTIME,     Execution time subroutine (or function)
136* @code{EOSHIFT}:       EOSHIFT,   End-off shift elements of an array
137* @code{EPSILON}:       EPSILON,   Epsilon function
138* @code{ERF}:           ERF,       Error function
139* @code{ERFC}:          ERFC,      Complementary error function
140* @code{ERFC_SCALED}:   ERFC_SCALED, Exponentially-scaled complementary error function
141* @code{ETIME}:         ETIME,     Execution time subroutine (or function)
142* @code{EVENT_QUERY}: EVENT_QUERY, Query whether a coarray event has occurred
143* @code{EXECUTE_COMMAND_LINE}: EXECUTE_COMMAND_LINE, Execute a shell command
144* @code{EXIT}:          EXIT,      Exit the program with status.
145* @code{EXP}:           EXP,       Exponential function
146* @code{EXPONENT}:      EXPONENT,  Exponent function
147* @code{EXTENDS_TYPE_OF}: EXTENDS_TYPE_OF,  Query dynamic type for extension
148* @code{FDATE}:         FDATE,     Subroutine (or function) to get the current time as a string
149* @code{FGET}:          FGET,      Read a single character in stream mode from stdin
150* @code{FGETC}:         FGETC,     Read a single character in stream mode
151* @code{FINDLOC}:       FINDLOC,   Search an array for a value
152* @code{FLOOR}:         FLOOR,     Integer floor function
153* @code{FLUSH}:         FLUSH,     Flush I/O unit(s)
154* @code{FNUM}:          FNUM,      File number function
155* @code{FPUT}:          FPUT,      Write a single character in stream mode to stdout
156* @code{FPUTC}:         FPUTC,     Write a single character in stream mode
157* @code{FRACTION}:      FRACTION,  Fractional part of the model representation
158* @code{FREE}:          FREE,      Memory de-allocation subroutine
159* @code{FSEEK}:         FSEEK,     Low level file positioning subroutine
160* @code{FSTAT}:         FSTAT,     Get file status
161* @code{FTELL}:         FTELL,     Current stream position
162* @code{GAMMA}:         GAMMA,     Gamma function
163* @code{GERROR}:        GERROR,    Get last system error message
164* @code{GETARG}:        GETARG,    Get command line arguments
165* @code{GET_COMMAND}:   GET_COMMAND, Get the entire command line
166* @code{GET_COMMAND_ARGUMENT}: GET_COMMAND_ARGUMENT, Get command line arguments
167* @code{GETCWD}:        GETCWD,    Get current working directory
168* @code{GETENV}:        GETENV,    Get an environmental variable
169* @code{GET_ENVIRONMENT_VARIABLE}: GET_ENVIRONMENT_VARIABLE, Get an environmental variable
170* @code{GETGID}:        GETGID,    Group ID function
171* @code{GETLOG}:        GETLOG,    Get login name
172* @code{GETPID}:        GETPID,    Process ID function
173* @code{GETUID}:        GETUID,    User ID function
174* @code{GMTIME}:        GMTIME,    Convert time to GMT info
175* @code{HOSTNM}:        HOSTNM,    Get system host name
176* @code{HUGE}:          HUGE,      Largest number of a kind
177* @code{HYPOT}:         HYPOT,     Euclidean distance function
178* @code{IACHAR}:        IACHAR,    Code in @acronym{ASCII} collating sequence
179* @code{IALL}:          IALL,      Bitwise AND of array elements
180* @code{IAND}:          IAND,      Bitwise logical and
181* @code{IANY}:          IANY,      Bitwise OR of array elements
182* @code{IARGC}:         IARGC,     Get the number of command line arguments
183* @code{IBCLR}:         IBCLR,     Clear bit
184* @code{IBITS}:         IBITS,     Bit extraction
185* @code{IBSET}:         IBSET,     Set bit
186* @code{ICHAR}:         ICHAR,     Character-to-integer conversion function
187* @code{IDATE}:         IDATE,     Current local time (day/month/year)
188* @code{IEOR}:          IEOR,      Bitwise logical exclusive or
189* @code{IERRNO}:        IERRNO,    Function to get the last system error number
190* @code{IMAGE_INDEX}:   IMAGE_INDEX, Cosubscript to image index conversion
191* @code{INDEX}:         INDEX intrinsic, Position of a substring within a string
192* @code{INT}:           INT,       Convert to integer type
193* @code{INT2}:          INT2,      Convert to 16-bit integer type
194* @code{INT8}:          INT8,      Convert to 64-bit integer type
195* @code{IOR}:           IOR,       Bitwise logical or
196* @code{IPARITY}:       IPARITY,   Bitwise XOR of array elements
197* @code{IRAND}:         IRAND,     Integer pseudo-random number
198* @code{IS_CONTIGUOUS}:  IS_CONTIGUOUS, Test whether an array is contiguous
199* @code{IS_IOSTAT_END}:  IS_IOSTAT_END, Test for end-of-file value
200* @code{IS_IOSTAT_EOR}:  IS_IOSTAT_EOR, Test for end-of-record value
201* @code{ISATTY}:        ISATTY,    Whether a unit is a terminal device
202* @code{ISHFT}:         ISHFT,     Shift bits
203* @code{ISHFTC}:        ISHFTC,    Shift bits circularly
204* @code{ISNAN}:         ISNAN,     Tests for a NaN
205* @code{ITIME}:         ITIME,     Current local time (hour/minutes/seconds)
206* @code{KILL}:          KILL,      Send a signal to a process
207* @code{KIND}:          KIND,      Kind of an entity
208* @code{LBOUND}:        LBOUND,    Lower dimension bounds of an array
209* @code{LCOBOUND}:      LCOBOUND,  Lower codimension bounds of an array
210* @code{LEADZ}:         LEADZ,     Number of leading zero bits of an integer
211* @code{LEN}:           LEN,       Length of a character entity
212* @code{LEN_TRIM}:      LEN_TRIM,  Length of a character entity without trailing blank characters
213* @code{LGE}:           LGE,       Lexical greater than or equal
214* @code{LGT}:           LGT,       Lexical greater than
215* @code{LINK}:          LINK,      Create a hard link
216* @code{LLE}:           LLE,       Lexical less than or equal
217* @code{LLT}:           LLT,       Lexical less than
218* @code{LNBLNK}:        LNBLNK,    Index of the last non-blank character in a string
219* @code{LOC}:           LOC,       Returns the address of a variable
220* @code{LOG}:           LOG,       Logarithm function
221* @code{LOG10}:         LOG10,     Base 10 logarithm function
222* @code{LOG_GAMMA}:     LOG_GAMMA, Logarithm of the Gamma function
223* @code{LOGICAL}:       LOGICAL,   Convert to logical type
224* @code{LONG}:          LONG,      Convert to integer type
225* @code{LSHIFT}:        LSHIFT,    Left shift bits
226* @code{LSTAT}:         LSTAT,     Get file status
227* @code{LTIME}:         LTIME,     Convert time to local time info
228* @code{MALLOC}:        MALLOC,    Dynamic memory allocation function
229* @code{MASKL}:         MASKL,     Left justified mask
230* @code{MASKR}:         MASKR,     Right justified mask
231* @code{MATMUL}:        MATMUL,    matrix multiplication
232* @code{MAX}:           MAX,       Maximum value of an argument list
233* @code{MAXEXPONENT}:   MAXEXPONENT, Maximum exponent of a real kind
234* @code{MAXLOC}:        MAXLOC,    Location of the maximum value within an array
235* @code{MAXVAL}:        MAXVAL,    Maximum value of an array
236* @code{MCLOCK}:        MCLOCK,    Time function
237* @code{MCLOCK8}:       MCLOCK8,   Time function (64-bit)
238* @code{MERGE}:         MERGE,     Merge arrays
239* @code{MERGE_BITS}:    MERGE_BITS, Merge of bits under mask
240* @code{MIN}:           MIN,       Minimum value of an argument list
241* @code{MINEXPONENT}:   MINEXPONENT, Minimum exponent of a real kind
242* @code{MINLOC}:        MINLOC,    Location of the minimum value within an array
243* @code{MINVAL}:        MINVAL,    Minimum value of an array
244* @code{MOD}:           MOD,       Remainder function
245* @code{MODULO}:        MODULO,    Modulo function
246* @code{MOVE_ALLOC}:    MOVE_ALLOC, Move allocation from one object to another
247* @code{MVBITS}:        MVBITS,    Move bits from one integer to another
248* @code{NEAREST}:       NEAREST,   Nearest representable number
249* @code{NEW_LINE}:      NEW_LINE,  New line character
250* @code{NINT}:          NINT,      Nearest whole number
251* @code{NORM2}:         NORM2,     Euclidean vector norm
252* @code{NOT}:           NOT,       Logical negation
253* @code{NULL}:          NULL,      Function that returns an disassociated pointer
254* @code{NUM_IMAGES}:    NUM_IMAGES, Number of images
255* @code{OR}:            OR,        Bitwise logical OR
256* @code{PACK}:          PACK,      Pack an array into an array of rank one
257* @code{PARITY}:        PARITY,    Reduction with exclusive OR
258* @code{PERROR}:        PERROR,    Print system error message
259* @code{POPCNT}:        POPCNT,    Number of bits set
260* @code{POPPAR}:        POPPAR,    Parity of the number of bits set
261* @code{PRECISION}:     PRECISION, Decimal precision of a real kind
262* @code{PRESENT}:       PRESENT,   Determine whether an optional dummy argument is specified
263* @code{PRODUCT}:       PRODUCT,   Product of array elements
264* @code{RADIX}:         RADIX,     Base of a data model
265* @code{RAN}:           RAN,       Real pseudo-random number
266* @code{RAND}:          RAND,      Real pseudo-random number
267* @code{RANDOM_INIT}:   RANDOM_INIT, Initialize pseudo-random number generator
268* @code{RANDOM_NUMBER}: RANDOM_NUMBER, Pseudo-random number
269* @code{RANDOM_SEED}:   RANDOM_SEED, Initialize a pseudo-random number sequence
270* @code{RANGE}:         RANGE,     Decimal exponent range
271* @code{RANK} :         RANK,      Rank of a data object
272* @code{REAL}:          REAL,      Convert to real type
273* @code{RENAME}:        RENAME,    Rename a file
274* @code{REPEAT}:        REPEAT,    Repeated string concatenation
275* @code{RESHAPE}:       RESHAPE,   Function to reshape an array
276* @code{RRSPACING}:     RRSPACING, Reciprocal of the relative spacing
277* @code{RSHIFT}:        RSHIFT,    Right shift bits
278* @code{SAME_TYPE_AS}:  SAME_TYPE_AS,  Query dynamic types for equality
279* @code{SCALE}:         SCALE,     Scale a real value
280* @code{SCAN}:          SCAN,      Scan a string for the presence of a set of characters
281* @code{SECNDS}:        SECNDS,    Time function
282* @code{SECOND}:        SECOND,    CPU time function
283* @code{SELECTED_CHAR_KIND}: SELECTED_CHAR_KIND,  Choose character kind
284* @code{SELECTED_INT_KIND}: SELECTED_INT_KIND,  Choose integer kind
285* @code{SELECTED_REAL_KIND}: SELECTED_REAL_KIND,  Choose real kind
286* @code{SET_EXPONENT}:  SET_EXPONENT, Set the exponent of the model
287* @code{SHAPE}:         SHAPE,     Determine the shape of an array
288* @code{SHIFTA}:        SHIFTA,    Right shift with fill
289* @code{SHIFTL}:        SHIFTL,    Left shift
290* @code{SHIFTR}:        SHIFTR,    Right shift
291* @code{SIGN}:          SIGN,      Sign copying function
292* @code{SIGNAL}:        SIGNAL,    Signal handling subroutine (or function)
293* @code{SIN}:           SIN,       Sine function
294* @code{SIND}:          SIND,      Sine function, degrees
295* @code{SINH}:          SINH,      Hyperbolic sine function
296* @code{SIZE}:          SIZE,      Function to determine the size of an array
297* @code{SIZEOF}:        SIZEOF,    Determine the size in bytes of an expression
298* @code{SLEEP}:         SLEEP,     Sleep for the specified number of seconds
299* @code{SPACING}:       SPACING,   Smallest distance between two numbers of a given type
300* @code{SPREAD}:        SPREAD,    Add a dimension to an array
301* @code{SQRT}:          SQRT,      Square-root function
302* @code{SRAND}:         SRAND,     Reinitialize the random number generator
303* @code{STAT}:          STAT,      Get file status
304* @code{STORAGE_SIZE}:  STORAGE_SIZE, Storage size in bits
305* @code{SUM}:           SUM,       Sum of array elements
306* @code{SYMLNK}:        SYMLNK,    Create a symbolic link
307* @code{SYSTEM}:        SYSTEM,    Execute a shell command
308* @code{SYSTEM_CLOCK}:  SYSTEM_CLOCK, Time function
309* @code{TAN}:           TAN,       Tangent function
310* @code{TAND}:          TAND,      Tangent function, degrees
311* @code{TANH}:          TANH,      Hyperbolic tangent function
312* @code{THIS_IMAGE}:    THIS_IMAGE, Cosubscript index of this image
313* @code{TIME}:          TIME,      Time function
314* @code{TIME8}:         TIME8,     Time function (64-bit)
315* @code{TINY}:          TINY,      Smallest positive number of a real kind
316* @code{TRAILZ}:        TRAILZ,    Number of trailing zero bits of an integer
317* @code{TRANSFER}:      TRANSFER,  Transfer bit patterns
318* @code{TRANSPOSE}:     TRANSPOSE, Transpose an array of rank two
319* @code{TRIM}:          TRIM,      Remove trailing blank characters of a string
320* @code{TTYNAM}:        TTYNAM,    Get the name of a terminal device.
321* @code{UBOUND}:        UBOUND,    Upper dimension bounds of an array
322* @code{UCOBOUND}:      UCOBOUND,  Upper codimension bounds of an array
323* @code{UMASK}:         UMASK,     Set the file creation mask
324* @code{UNLINK}:        UNLINK,    Remove a file from the file system
325* @code{UNPACK}:        UNPACK,    Unpack an array of rank one into an array
326* @code{VERIFY}:        VERIFY,    Scan a string for the absence of a set of characters
327* @code{XOR}:           XOR,       Bitwise logical exclusive or
328@end menu
329
330@node Introduction to Intrinsics
331@section Introduction to intrinsic procedures
332
333The intrinsic procedures provided by GNU Fortran include all of the
334intrinsic procedures required by the Fortran 95 standard, a set of
335intrinsic procedures for backwards compatibility with G77, and a
336selection of intrinsic procedures from the Fortran 2003 and Fortran 2008
337standards.  Any conflict between a description here and a description in
338either the Fortran 95 standard, the Fortran 2003 standard or the Fortran
3392008 standard is unintentional, and the standard(s) should be considered
340authoritative.
341
342The enumeration of the @code{KIND} type parameter is processor defined in
343the Fortran 95 standard.  GNU Fortran defines the default integer type and
344default real type by @code{INTEGER(KIND=4)} and @code{REAL(KIND=4)},
345respectively.  The standard mandates that both data types shall have
346another kind, which have more precision.  On typical target architectures
347supported by @command{gfortran}, this kind type parameter is @code{KIND=8}.
348Hence, @code{REAL(KIND=8)} and @code{DOUBLE PRECISION} are equivalent.
349In the description of generic intrinsic procedures, the kind type parameter
350will be specified by @code{KIND=*}, and in the description of specific
351names for an intrinsic procedure the kind type parameter will be explicitly
352given (e.g., @code{REAL(KIND=4)} or @code{REAL(KIND=8)}).  Finally, for
353brevity the optional @code{KIND=} syntax will be omitted.
354
355Many of the intrinsic procedures take one or more optional arguments.
356This document follows the convention used in the Fortran 95 standard,
357and denotes such arguments by square brackets.
358
359GNU Fortran offers the @option{-std=f95} and @option{-std=gnu} options,
360which can be used to restrict the set of intrinsic procedures to a
361given standard.  By default, @command{gfortran} sets the @option{-std=gnu}
362option, and so all intrinsic procedures described here are accepted.  There
363is one caveat.  For a select group of intrinsic procedures, @command{g77}
364implemented both a function and a subroutine.  Both classes
365have been implemented in @command{gfortran} for backwards compatibility
366with @command{g77}.  It is noted here that these functions and subroutines
367cannot be intermixed in a given subprogram.  In the descriptions that follow,
368the applicable standard for each intrinsic procedure is noted.
369
370
371
372@node ABORT
373@section @code{ABORT} --- Abort the program
374@fnindex ABORT
375@cindex program termination, with core dump
376@cindex terminate program, with core dump
377@cindex core, dump
378
379@table @asis
380@item @emph{Description}:
381@code{ABORT} causes immediate termination of the program.  On operating
382systems that support a core dump, @code{ABORT} will produce a core dump.
383It will also print a backtrace, unless @code{-fno-backtrace} is given.
384
385@item @emph{Standard}:
386GNU extension
387
388@item @emph{Class}:
389Subroutine
390
391@item @emph{Syntax}:
392@code{CALL ABORT}
393
394@item @emph{Return value}:
395Does not return.
396
397@item @emph{Example}:
398@smallexample
399program test_abort
400  integer :: i = 1, j = 2
401  if (i /= j) call abort
402end program test_abort
403@end smallexample
404
405@item @emph{See also}:
406@ref{EXIT}, @gol
407@ref{KILL}, @gol
408@ref{BACKTRACE}
409@end table
410
411
412
413@node ABS
414@section @code{ABS} --- Absolute value
415@fnindex ABS
416@fnindex CABS
417@fnindex DABS
418@fnindex IABS
419@fnindex ZABS
420@fnindex CDABS
421@fnindex BABS
422@fnindex IIABS
423@fnindex JIABS
424@fnindex KIABS
425@cindex absolute value
426
427@table @asis
428@item @emph{Description}:
429@code{ABS(A)} computes the absolute value of @code{A}.
430
431@item @emph{Standard}:
432Fortran 77 and later, has overloads that are GNU extensions
433
434@item @emph{Class}:
435Elemental function
436
437@item @emph{Syntax}:
438@code{RESULT = ABS(A)}
439
440@item @emph{Arguments}:
441@multitable @columnfractions .15 .70
442@item @var{A} @tab The type of the argument shall be an @code{INTEGER},
443@code{REAL}, or @code{COMPLEX}.
444@end multitable
445
446@item @emph{Return value}:
447The return value is of the same type and
448kind as the argument except the return value is @code{REAL} for a
449@code{COMPLEX} argument.
450
451@item @emph{Example}:
452@smallexample
453program test_abs
454  integer :: i = -1
455  real :: x = -1.e0
456  complex :: z = (-1.e0,0.e0)
457  i = abs(i)
458  x = abs(x)
459  x = abs(z)
460end program test_abs
461@end smallexample
462
463@item @emph{Specific names}:
464@multitable @columnfractions .20 .20 .20 .25
465@item Name            @tab Argument            @tab Return type       @tab Standard
466@item @code{ABS(A)}   @tab @code{REAL(4) A}    @tab @code{REAL(4)}    @tab Fortran 77 and later
467@item @code{CABS(A)}  @tab @code{COMPLEX(4) A} @tab @code{REAL(4)}    @tab Fortran 77 and later
468@item @code{DABS(A)}  @tab @code{REAL(8) A}    @tab @code{REAL(8)}    @tab Fortran 77 and later
469@item @code{IABS(A)}  @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab Fortran 77 and later
470@item @code{BABS(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
471@item @code{IIABS(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
472@item @code{JIABS(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
473@item @code{KIABS(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
474@item @code{ZABS(A)}  @tab @code{COMPLEX(8) A} @tab @code{REAL(8)} @tab GNU extension
475@item @code{CDABS(A)} @tab @code{COMPLEX(8) A} @tab @code{REAL(8)} @tab GNU extension
476@end multitable
477@end table
478
479
480
481@node ACCESS
482@section @code{ACCESS} --- Checks file access modes
483@fnindex ACCESS
484@cindex file system, access mode
485
486@table @asis
487@item @emph{Description}:
488@code{ACCESS(NAME, MODE)} checks whether the file @var{NAME}
489exists, is readable, writable or executable. Except for the
490executable check, @code{ACCESS} can be replaced by
491Fortran 95's @code{INQUIRE}.
492
493@item @emph{Standard}:
494GNU extension
495
496@item @emph{Class}:
497Inquiry function
498
499@item @emph{Syntax}:
500@code{RESULT = ACCESS(NAME, MODE)}
501
502@item @emph{Arguments}:
503@multitable @columnfractions .15 .70
504@item @var{NAME} @tab Scalar @code{CHARACTER} of default kind with the
505file name. Tailing blank are ignored unless the character @code{achar(0)}
506is present, then all characters up to and excluding @code{achar(0)} are
507used as file name.
508@item @var{MODE} @tab Scalar @code{CHARACTER} of default kind with the
509file access mode, may be any concatenation of @code{"r"} (readable),
510@code{"w"} (writable) and @code{"x"} (executable), or @code{" "} to check
511for existence.
512@end multitable
513
514@item @emph{Return value}:
515Returns a scalar @code{INTEGER}, which is @code{0} if the file is
516accessible in the given mode; otherwise or if an invalid argument
517has been given for @code{MODE} the value @code{1} is returned.
518
519@item @emph{Example}:
520@smallexample
521program access_test
522  implicit none
523  character(len=*), parameter :: file  = 'test.dat'
524  character(len=*), parameter :: file2 = 'test.dat  '//achar(0)
525  if(access(file,' ') == 0) print *, trim(file),' is exists'
526  if(access(file,'r') == 0) print *, trim(file),' is readable'
527  if(access(file,'w') == 0) print *, trim(file),' is writable'
528  if(access(file,'x') == 0) print *, trim(file),' is executable'
529  if(access(file2,'rwx') == 0) &
530    print *, trim(file2),' is readable, writable and executable'
531end program access_test
532@end smallexample
533@end table
534
535
536
537@node ACHAR
538@section @code{ACHAR} --- Character in @acronym{ASCII} collating sequence
539@fnindex ACHAR
540@cindex @acronym{ASCII} collating sequence
541@cindex collating sequence, @acronym{ASCII}
542
543@table @asis
544@item @emph{Description}:
545@code{ACHAR(I)} returns the character located at position @code{I}
546in the @acronym{ASCII} collating sequence.
547
548@item @emph{Standard}:
549Fortran 77 and later, with @var{KIND} argument Fortran 2003 and later
550
551@item @emph{Class}:
552Elemental function
553
554@item @emph{Syntax}:
555@code{RESULT = ACHAR(I [, KIND])}
556
557@item @emph{Arguments}:
558@multitable @columnfractions .15 .70
559@item @var{I}    @tab The type shall be @code{INTEGER}.
560@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
561expression indicating the kind parameter of the result.
562@end multitable
563
564@item @emph{Return value}:
565The return value is of type @code{CHARACTER} with a length of one.
566If the @var{KIND} argument is present, the return value is of the
567specified kind and of the default kind otherwise.
568
569@item @emph{Example}:
570@smallexample
571program test_achar
572  character c
573  c = achar(32)
574end program test_achar
575@end smallexample
576
577@item @emph{Note}:
578See @ref{ICHAR} for a discussion of converting between numerical values
579and formatted string representations.
580
581@item @emph{See also}:
582@ref{CHAR}, @gol
583@ref{IACHAR}, @gol
584@ref{ICHAR}
585@end table
586
587
588
589@node ACOS
590@section @code{ACOS} --- Arccosine function
591@fnindex ACOS
592@fnindex DACOS
593@cindex trigonometric function, cosine, inverse
594@cindex cosine, inverse
595
596@table @asis
597@item @emph{Description}:
598@code{ACOS(X)} computes the arccosine of @var{X} (inverse of @code{COS(X)}).
599
600@item @emph{Standard}:
601Fortran 77 and later, for a complex argument Fortran 2008 or later
602
603@item @emph{Class}:
604Elemental function
605
606@item @emph{Syntax}:
607@code{RESULT = ACOS(X)}
608
609@item @emph{Arguments}:
610@multitable @columnfractions .15 .70
611@item @var{X} @tab The type shall either be @code{REAL} with a magnitude that is
612less than or equal to one - or the type shall be @code{COMPLEX}.
613@end multitable
614
615@item @emph{Return value}:
616The return value is of the same type and kind as @var{X}.
617The real part of the result is in radians and lies in the range
618@math{0 \leq \Re \acos(x) \leq \pi}.
619
620@item @emph{Example}:
621@smallexample
622program test_acos
623  real(8) :: x = 0.866_8
624  x = acos(x)
625end program test_acos
626@end smallexample
627
628@item @emph{Specific names}:
629@multitable @columnfractions .20 .20 .20 .25
630@item Name            @tab Argument         @tab Return type     @tab Standard
631@item @code{ACOS(X)}  @tab @code{REAL(4) X} @tab @code{REAL(4)}  @tab Fortran 77 and later
632@item @code{DACOS(X)} @tab @code{REAL(8) X} @tab @code{REAL(8)}  @tab Fortran 77 and later
633@end multitable
634
635@item @emph{See also}:
636Inverse function: @gol
637@ref{COS} @gol
638Degrees function: @gol
639@ref{ACOSD}
640@end table
641
642
643
644@node ACOSD
645@section @code{ACOSD} --- Arccosine function, degrees
646@fnindex ACOSD
647@fnindex DACOSD
648@cindex trigonometric function, cosine, inverse, degrees
649@cindex cosine, inverse, degrees
650
651@table @asis
652@item @emph{Description}:
653@code{ACOSD(X)} computes the arccosine of @var{X} in degrees (inverse of
654@code{COSD(X)}).
655
656This function is for compatibility only and should be avoided in favor of
657standard constructs wherever possible.
658
659@item @emph{Standard}:
660GNU extension, enabled with @option{-fdec-math}
661
662@item @emph{Class}:
663Elemental function
664
665@item @emph{Syntax}:
666@code{RESULT = ACOSD(X)}
667
668@item @emph{Arguments}:
669@multitable @columnfractions .15 .70
670@item @var{X} @tab The type shall either be @code{REAL} with a magnitude that is
671less than or equal to one - or the type shall be @code{COMPLEX}.
672@end multitable
673
674@item @emph{Return value}:
675The return value is of the same type and kind as @var{X}.
676The real part of the result is in degrees and lies in the range
677@math{0 \leq \Re \acos(x) \leq 180}.
678
679@item @emph{Example}:
680@smallexample
681program test_acosd
682  real(8) :: x = 0.866_8
683  x = acosd(x)
684end program test_acosd
685@end smallexample
686
687@item @emph{Specific names}:
688@multitable @columnfractions .20 .20 .20 .25
689@item Name            @tab Argument         @tab Return type     @tab Standard
690@item @code{ACOSD(X)}  @tab @code{REAL(4) X} @tab @code{REAL(4)}  @tab GNU extension
691@item @code{DACOSD(X)} @tab @code{REAL(8) X} @tab @code{REAL(8)}  @tab GNU extension
692@end multitable
693
694@item @emph{See also}:
695Inverse function: @gol
696@ref{COSD} @gol
697Radians function: @gol
698@ref{ACOS} @gol
699@end table
700
701
702
703@node ACOSH
704@section @code{ACOSH} --- Inverse hyperbolic cosine function
705@fnindex ACOSH
706@fnindex DACOSH
707@cindex area hyperbolic cosine
708@cindex inverse hyperbolic cosine
709@cindex hyperbolic function, cosine, inverse
710@cindex cosine, hyperbolic, inverse
711
712@table @asis
713@item @emph{Description}:
714@code{ACOSH(X)} computes the inverse hyperbolic cosine of @var{X}.
715
716@item @emph{Standard}:
717Fortran 2008 and later
718
719@item @emph{Class}:
720Elemental function
721
722@item @emph{Syntax}:
723@code{RESULT = ACOSH(X)}
724
725@item @emph{Arguments}:
726@multitable @columnfractions .15 .70
727@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
728@end multitable
729
730@item @emph{Return value}:
731The return value has the same type and kind as @var{X}. If @var{X} is
732complex, the imaginary part of the result is in radians and lies between
733@math{ 0 \leq \Im \acosh(x) \leq \pi}.
734
735@item @emph{Example}:
736@smallexample
737PROGRAM test_acosh
738  REAL(8), DIMENSION(3) :: x = (/ 1.0, 2.0, 3.0 /)
739  WRITE (*,*) ACOSH(x)
740END PROGRAM
741@end smallexample
742
743@item @emph{Specific names}:
744@multitable @columnfractions .20 .20 .20 .25
745@item Name             @tab Argument          @tab Return type       @tab Standard
746@item @code{DACOSH(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
747@end multitable
748
749@item @emph{See also}:
750Inverse function: @gol
751@ref{COSH}
752@end table
753
754
755
756@node ADJUSTL
757@section @code{ADJUSTL} --- Left adjust a string
758@fnindex ADJUSTL
759@cindex string, adjust left
760@cindex adjust string
761
762@table @asis
763@item @emph{Description}:
764@code{ADJUSTL(STRING)} will left adjust a string by removing leading spaces.
765Spaces are inserted at the end of the string as needed.
766
767@item @emph{Standard}:
768Fortran 90 and later
769
770@item @emph{Class}:
771Elemental function
772
773@item @emph{Syntax}:
774@code{RESULT = ADJUSTL(STRING)}
775
776@item @emph{Arguments}:
777@multitable @columnfractions .15 .70
778@item @var{STRING} @tab The type shall be @code{CHARACTER}.
779@end multitable
780
781@item @emph{Return value}:
782The return value is of type @code{CHARACTER} and of the same kind as
783@var{STRING} where leading spaces are removed and the same number of
784spaces are inserted on the end of @var{STRING}.
785
786@item @emph{Example}:
787@smallexample
788program test_adjustl
789  character(len=20) :: str = '   gfortran'
790  str = adjustl(str)
791  print *, str
792end program test_adjustl
793@end smallexample
794
795@item @emph{See also}:
796@ref{ADJUSTR}, @gol
797@ref{TRIM}
798@end table
799
800
801
802@node ADJUSTR
803@section @code{ADJUSTR} --- Right adjust a string
804@fnindex ADJUSTR
805@cindex string, adjust right
806@cindex adjust string
807
808@table @asis
809@item @emph{Description}:
810@code{ADJUSTR(STRING)} will right adjust a string by removing trailing spaces.
811Spaces are inserted at the start of the string as needed.
812
813@item @emph{Standard}:
814Fortran 90 and later
815
816@item @emph{Class}:
817Elemental function
818
819@item @emph{Syntax}:
820@code{RESULT = ADJUSTR(STRING)}
821
822@item @emph{Arguments}:
823@multitable @columnfractions .15 .70
824@item @var{STR} @tab The type shall be @code{CHARACTER}.
825@end multitable
826
827@item @emph{Return value}:
828The return value is of type @code{CHARACTER} and of the same kind as
829@var{STRING} where trailing spaces are removed and the same number of
830spaces are inserted at the start of @var{STRING}.
831
832@item @emph{Example}:
833@smallexample
834program test_adjustr
835  character(len=20) :: str = 'gfortran'
836  str = adjustr(str)
837  print *, str
838end program test_adjustr
839@end smallexample
840
841@item @emph{See also}:
842@ref{ADJUSTL}, @gol
843@ref{TRIM}
844@end table
845
846
847
848@node AIMAG
849@section @code{AIMAG} --- Imaginary part of complex number
850@fnindex AIMAG
851@fnindex DIMAG
852@fnindex IMAG
853@fnindex IMAGPART
854@cindex complex numbers, imaginary part
855
856@table @asis
857@item @emph{Description}:
858@code{AIMAG(Z)} yields the imaginary part of complex argument @code{Z}.
859The @code{IMAG(Z)} and @code{IMAGPART(Z)} intrinsic functions are provided
860for compatibility with @command{g77}, and their use in new code is
861strongly discouraged.
862
863@item @emph{Standard}:
864Fortran 77 and later, has overloads that are GNU extensions
865
866@item @emph{Class}:
867Elemental function
868
869@item @emph{Syntax}:
870@code{RESULT = AIMAG(Z)}
871
872@item @emph{Arguments}:
873@multitable @columnfractions .15 .70
874@item @var{Z} @tab The type of the argument shall be @code{COMPLEX}.
875@end multitable
876
877@item @emph{Return value}:
878The return value is of type @code{REAL} with the
879kind type parameter of the argument.
880
881@item @emph{Example}:
882@smallexample
883program test_aimag
884  complex(4) z4
885  complex(8) z8
886  z4 = cmplx(1.e0_4, 0.e0_4)
887  z8 = cmplx(0.e0_8, 1.e0_8)
888  print *, aimag(z4), dimag(z8)
889end program test_aimag
890@end smallexample
891
892@item @emph{Specific names}:
893@multitable @columnfractions .20 .20 .20 .25
894@item Name               @tab Argument            @tab Return type     @tab Standard
895@item @code{AIMAG(Z)}    @tab @code{COMPLEX Z}    @tab @code{REAL}     @tab Fortran 77 and later
896@item @code{DIMAG(Z)}    @tab @code{COMPLEX(8) Z} @tab @code{REAL(8)}  @tab GNU extension
897@item @code{IMAG(Z)}     @tab @code{COMPLEX Z}    @tab @code{REAL}     @tab GNU extension
898@item @code{IMAGPART(Z)} @tab @code{COMPLEX Z}    @tab @code{REAL}     @tab GNU extension
899@end multitable
900@end table
901
902
903
904@node AINT
905@section @code{AINT} --- Truncate to a whole number
906@fnindex AINT
907@fnindex DINT
908@cindex floor
909@cindex rounding, floor
910
911@table @asis
912@item @emph{Description}:
913@code{AINT(A [, KIND])} truncates its argument to a whole number.
914
915@item @emph{Standard}:
916Fortran 77 and later
917
918@item @emph{Class}:
919Elemental function
920
921@item @emph{Syntax}:
922@code{RESULT = AINT(A [, KIND])}
923
924@item @emph{Arguments}:
925@multitable @columnfractions .15 .70
926@item @var{A}    @tab The type of the argument shall be @code{REAL}.
927@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
928expression indicating the kind parameter of the result.
929@end multitable
930
931@item @emph{Return value}:
932The return value is of type @code{REAL} with the kind type parameter of the
933argument if the optional @var{KIND} is absent; otherwise, the kind
934type parameter will be given by @var{KIND}.  If the magnitude of
935@var{X} is less than one, @code{AINT(X)} returns zero.  If the
936magnitude is equal to or greater than one then it returns the largest
937whole number that does not exceed its magnitude.  The sign is the same
938as the sign of @var{X}.
939
940@item @emph{Example}:
941@smallexample
942program test_aint
943  real(4) x4
944  real(8) x8
945  x4 = 1.234E0_4
946  x8 = 4.321_8
947  print *, aint(x4), dint(x8)
948  x8 = aint(x4,8)
949end program test_aint
950@end smallexample
951
952@item @emph{Specific names}:
953@multitable @columnfractions .20 .20 .20 .25
954@item Name           @tab Argument         @tab Return type      @tab Standard
955@item @code{AINT(A)} @tab @code{REAL(4) A} @tab @code{REAL(4)}   @tab Fortran 77 and later
956@item @code{DINT(A)} @tab @code{REAL(8) A} @tab @code{REAL(8)}   @tab Fortran 77 and later
957@end multitable
958@end table
959
960
961
962@node ALARM
963@section @code{ALARM} --- Execute a routine after a given delay
964@fnindex ALARM
965@cindex delayed execution
966
967@table @asis
968@item @emph{Description}:
969@code{ALARM(SECONDS, HANDLER [, STATUS])} causes external subroutine @var{HANDLER}
970to be executed after a delay of @var{SECONDS} by using @code{alarm(2)} to
971set up a signal and @code{signal(2)} to catch it. If @var{STATUS} is
972supplied, it will be returned with the number of seconds remaining until
973any previously scheduled alarm was due to be delivered, or zero if there
974was no previously scheduled alarm.
975
976@item @emph{Standard}:
977GNU extension
978
979@item @emph{Class}:
980Subroutine
981
982@item @emph{Syntax}:
983@code{CALL ALARM(SECONDS, HANDLER [, STATUS])}
984
985@item @emph{Arguments}:
986@multitable @columnfractions .15 .70
987@item @var{SECONDS} @tab The type of the argument shall be a scalar
988@code{INTEGER}. It is @code{INTENT(IN)}.
989@item @var{HANDLER} @tab Signal handler (@code{INTEGER FUNCTION} or
990@code{SUBROUTINE}) or dummy/global @code{INTEGER} scalar. The scalar
991values may be either @code{SIG_IGN=1} to ignore the alarm generated
992or @code{SIG_DFL=0} to set the default action. It is @code{INTENT(IN)}.
993@item @var{STATUS}  @tab (Optional) @var{STATUS} shall be a scalar
994variable of the default @code{INTEGER} kind. It is @code{INTENT(OUT)}.
995@end multitable
996
997@item @emph{Example}:
998@smallexample
999program test_alarm
1000  external handler_print
1001  integer i
1002  call alarm (3, handler_print, i)
1003  print *, i
1004  call sleep(10)
1005end program test_alarm
1006@end smallexample
1007This will cause the external routine @var{handler_print} to be called
1008after 3 seconds.
1009@end table
1010
1011
1012
1013@node ALL
1014@section @code{ALL} --- All values in @var{MASK} along @var{DIM} are true
1015@fnindex ALL
1016@cindex array, apply condition
1017@cindex array, condition testing
1018
1019@table @asis
1020@item @emph{Description}:
1021@code{ALL(MASK [, DIM])} determines if all the values are true in @var{MASK}
1022in the array along dimension @var{DIM}.
1023
1024@item @emph{Standard}:
1025Fortran 90 and later
1026
1027@item @emph{Class}:
1028Transformational function
1029
1030@item @emph{Syntax}:
1031@code{RESULT = ALL(MASK [, DIM])}
1032
1033@item @emph{Arguments}:
1034@multitable @columnfractions .15 .70
1035@item @var{MASK} @tab The type of the argument shall be @code{LOGICAL} and
1036it shall not be scalar.
1037@item @var{DIM}  @tab (Optional) @var{DIM} shall be a scalar integer
1038with a value that lies between one and the rank of @var{MASK}.
1039@end multitable
1040
1041@item @emph{Return value}:
1042@code{ALL(MASK)} returns a scalar value of type @code{LOGICAL} where
1043the kind type parameter is the same as the kind type parameter of
1044@var{MASK}.  If @var{DIM} is present, then @code{ALL(MASK, DIM)} returns
1045an array with the rank of @var{MASK} minus 1.  The shape is determined from
1046the shape of @var{MASK} where the @var{DIM} dimension is elided.
1047
1048@table @asis
1049@item (A)
1050@code{ALL(MASK)} is true if all elements of @var{MASK} are true.
1051It also is true if @var{MASK} has zero size; otherwise, it is false.
1052@item (B)
1053If the rank of @var{MASK} is one, then @code{ALL(MASK,DIM)} is equivalent
1054to @code{ALL(MASK)}.  If the rank is greater than one, then @code{ALL(MASK,DIM)}
1055is determined by applying @code{ALL} to the array sections.
1056@end table
1057
1058@item @emph{Example}:
1059@smallexample
1060program test_all
1061  logical l
1062  l = all((/.true., .true., .true./))
1063  print *, l
1064  call section
1065  contains
1066    subroutine section
1067      integer a(2,3), b(2,3)
1068      a = 1
1069      b = 1
1070      b(2,2) = 2
1071      print *, all(a .eq. b, 1)
1072      print *, all(a .eq. b, 2)
1073    end subroutine section
1074end program test_all
1075@end smallexample
1076@end table
1077
1078
1079
1080@node ALLOCATED
1081@section @code{ALLOCATED} --- Status of an allocatable entity
1082@fnindex ALLOCATED
1083@cindex allocation, status
1084
1085@table @asis
1086@item @emph{Description}:
1087@code{ALLOCATED(ARRAY)} and @code{ALLOCATED(SCALAR)} check the allocation
1088status of @var{ARRAY} and @var{SCALAR}, respectively.
1089
1090@item @emph{Standard}:
1091Fortran 90 and later.  Note, the @code{SCALAR=} keyword and allocatable
1092scalar entities are available in Fortran 2003 and later.
1093
1094@item @emph{Class}:
1095Inquiry function
1096
1097@item @emph{Syntax}:
1098@multitable @columnfractions .80
1099@item @code{RESULT = ALLOCATED(ARRAY)}
1100@item @code{RESULT = ALLOCATED(SCALAR)}
1101@end multitable
1102
1103@item @emph{Arguments}:
1104@multitable @columnfractions .15 .70
1105@item @var{ARRAY}    @tab The argument shall be an @code{ALLOCATABLE} array.
1106@item @var{SCALAR}   @tab The argument shall be an @code{ALLOCATABLE} scalar.
1107@end multitable
1108
1109@item @emph{Return value}:
1110The return value is a scalar @code{LOGICAL} with the default logical
1111kind type parameter.  If the argument is allocated, then the result is
1112@code{.TRUE.}; otherwise, it returns @code{.FALSE.}
1113
1114@item @emph{Example}:
1115@smallexample
1116program test_allocated
1117  integer :: i = 4
1118  real(4), allocatable :: x(:)
1119  if (.not. allocated(x)) allocate(x(i))
1120end program test_allocated
1121@end smallexample
1122@end table
1123
1124
1125
1126@node AND
1127@section @code{AND} --- Bitwise logical AND
1128@fnindex AND
1129@cindex bitwise logical and
1130@cindex logical and, bitwise
1131
1132@table @asis
1133@item @emph{Description}:
1134Bitwise logical @code{AND}.
1135
1136This intrinsic routine is provided for backwards compatibility with
1137GNU Fortran 77.  For integer arguments, programmers should consider
1138the use of the @ref{IAND} intrinsic defined by the Fortran standard.
1139
1140@item @emph{Standard}:
1141GNU extension
1142
1143@item @emph{Class}:
1144Function
1145
1146@item @emph{Syntax}:
1147@code{RESULT = AND(I, J)}
1148
1149@item @emph{Arguments}:
1150@multitable @columnfractions .15 .70
1151@item @var{I} @tab The type shall be either a scalar @code{INTEGER}
1152type or a scalar @code{LOGICAL} type or a boz-literal-constant.
1153@item @var{J} @tab The type shall be the same as the type of @var{I} or
1154a boz-literal-constant. @var{I} and @var{J} shall not both be
1155boz-literal-constants.  If either @var{I} or @var{J} is a
1156boz-literal-constant, then the other argument must be a scalar @code{INTEGER}.
1157@end multitable
1158
1159@item @emph{Return value}:
1160The return type is either a scalar @code{INTEGER} or a scalar
1161@code{LOGICAL}.  If the kind type parameters differ, then the
1162smaller kind type is implicitly converted to larger kind, and the
1163return has the larger kind.  A boz-literal-constant is
1164converted to an @code{INTEGER} with the kind type parameter of
1165the other argument as-if a call to @ref{INT} occurred.
1166
1167@item @emph{Example}:
1168@smallexample
1169PROGRAM test_and
1170  LOGICAL :: T = .TRUE., F = .FALSE.
1171  INTEGER :: a, b
1172  DATA a / Z'F' /, b / Z'3' /
1173
1174  WRITE (*,*) AND(T, T), AND(T, F), AND(F, T), AND(F, F)
1175  WRITE (*,*) AND(a, b)
1176END PROGRAM
1177@end smallexample
1178
1179@item @emph{See also}:
1180Fortran 95 elemental function: @gol
1181@ref{IAND}
1182@end table
1183
1184
1185
1186@node ANINT
1187@section @code{ANINT} --- Nearest whole number
1188@fnindex ANINT
1189@fnindex DNINT
1190@cindex ceiling
1191@cindex rounding, ceiling
1192
1193@table @asis
1194@item @emph{Description}:
1195@code{ANINT(A [, KIND])} rounds its argument to the nearest whole number.
1196
1197@item @emph{Standard}:
1198Fortran 77 and later
1199
1200@item @emph{Class}:
1201Elemental function
1202
1203@item @emph{Syntax}:
1204@code{RESULT = ANINT(A [, KIND])}
1205
1206@item @emph{Arguments}:
1207@multitable @columnfractions .15 .70
1208@item @var{A}    @tab The type of the argument shall be @code{REAL}.
1209@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
1210expression indicating the kind parameter of the result.
1211@end multitable
1212
1213@item @emph{Return value}:
1214The return value is of type real with the kind type parameter of the
1215argument if the optional @var{KIND} is absent; otherwise, the kind
1216type parameter will be given by @var{KIND}.  If @var{A} is greater than
1217zero, @code{ANINT(A)} returns @code{AINT(X+0.5)}.  If @var{A} is
1218less than or equal to zero then it returns @code{AINT(X-0.5)}.
1219
1220@item @emph{Example}:
1221@smallexample
1222program test_anint
1223  real(4) x4
1224  real(8) x8
1225  x4 = 1.234E0_4
1226  x8 = 4.321_8
1227  print *, anint(x4), dnint(x8)
1228  x8 = anint(x4,8)
1229end program test_anint
1230@end smallexample
1231
1232@item @emph{Specific names}:
1233@multitable @columnfractions .20 .20 .20 .25
1234@item Name            @tab Argument         @tab Return type      @tab Standard
1235@item @code{AINT(A)}  @tab @code{REAL(4) A} @tab @code{REAL(4)}   @tab Fortran 77 and later
1236@item @code{DNINT(A)} @tab @code{REAL(8) A} @tab @code{REAL(8)}   @tab Fortran 77 and later
1237@end multitable
1238@end table
1239
1240
1241
1242@node ANY
1243@section @code{ANY} --- Any value in @var{MASK} along @var{DIM} is true
1244@fnindex ANY
1245@cindex array, apply condition
1246@cindex array, condition testing
1247
1248@table @asis
1249@item @emph{Description}:
1250@code{ANY(MASK [, DIM])} determines if any of the values in the logical array
1251@var{MASK} along dimension @var{DIM} are @code{.TRUE.}.
1252
1253@item @emph{Standard}:
1254Fortran 90 and later
1255
1256@item @emph{Class}:
1257Transformational function
1258
1259@item @emph{Syntax}:
1260@code{RESULT = ANY(MASK [, DIM])}
1261
1262@item @emph{Arguments}:
1263@multitable @columnfractions .15 .70
1264@item @var{MASK} @tab The type of the argument shall be @code{LOGICAL} and
1265it shall not be scalar.
1266@item @var{DIM}  @tab (Optional) @var{DIM} shall be a scalar integer
1267with a value that lies between one and the rank of @var{MASK}.
1268@end multitable
1269
1270@item @emph{Return value}:
1271@code{ANY(MASK)} returns a scalar value of type @code{LOGICAL} where
1272the kind type parameter is the same as the kind type parameter of
1273@var{MASK}.  If @var{DIM} is present, then @code{ANY(MASK, DIM)} returns
1274an array with the rank of @var{MASK} minus 1.  The shape is determined from
1275the shape of @var{MASK} where the @var{DIM} dimension is elided.
1276
1277@table @asis
1278@item (A)
1279@code{ANY(MASK)} is true if any element of @var{MASK} is true;
1280otherwise, it is false.  It also is false if @var{MASK} has zero size.
1281@item (B)
1282If the rank of @var{MASK} is one, then @code{ANY(MASK,DIM)} is equivalent
1283to @code{ANY(MASK)}.  If the rank is greater than one, then @code{ANY(MASK,DIM)}
1284is determined by applying @code{ANY} to the array sections.
1285@end table
1286
1287@item @emph{Example}:
1288@smallexample
1289program test_any
1290  logical l
1291  l = any((/.true., .true., .true./))
1292  print *, l
1293  call section
1294  contains
1295    subroutine section
1296      integer a(2,3), b(2,3)
1297      a = 1
1298      b = 1
1299      b(2,2) = 2
1300      print *, any(a .eq. b, 1)
1301      print *, any(a .eq. b, 2)
1302    end subroutine section
1303end program test_any
1304@end smallexample
1305@end table
1306
1307
1308
1309@node ASIN
1310@section @code{ASIN} --- Arcsine function
1311@fnindex ASIN
1312@fnindex DASIN
1313@cindex trigonometric function, sine, inverse
1314@cindex sine, inverse
1315
1316@table @asis
1317@item @emph{Description}:
1318@code{ASIN(X)} computes the arcsine of its @var{X} (inverse of @code{SIN(X)}).
1319
1320@item @emph{Standard}:
1321Fortran 77 and later, for a complex argument Fortran 2008 or later
1322
1323@item @emph{Class}:
1324Elemental function
1325
1326@item @emph{Syntax}:
1327@code{RESULT = ASIN(X)}
1328
1329@item @emph{Arguments}:
1330@multitable @columnfractions .15 .70
1331@item @var{X} @tab The type shall be either @code{REAL} and a magnitude that is
1332less than or equal to one - or be @code{COMPLEX}.
1333@end multitable
1334
1335@item @emph{Return value}:
1336The return value is of the same type and kind as @var{X}.
1337The real part of the result is in radians and lies in the range
1338@math{-\pi/2 \leq \Re \asin(x) \leq \pi/2}.
1339
1340@item @emph{Example}:
1341@smallexample
1342program test_asin
1343  real(8) :: x = 0.866_8
1344  x = asin(x)
1345end program test_asin
1346@end smallexample
1347
1348@item @emph{Specific names}:
1349@multitable @columnfractions .20 .20 .20 .25
1350@item Name            @tab Argument          @tab Return type       @tab Standard
1351@item @code{ASIN(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab Fortran 77 and later
1352@item @code{DASIN(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab Fortran 77 and later
1353@end multitable
1354
1355@item @emph{See also}:
1356Inverse function: @gol
1357@ref{SIN} @gol
1358Degrees function: @gol
1359@ref{ASIND}
1360@end table
1361
1362
1363
1364@node ASIND
1365@section @code{ASIND} --- Arcsine function, degrees
1366@fnindex ASIND
1367@fnindex DASIND
1368@cindex trigonometric function, sine, inverse, degrees
1369@cindex sine, inverse, degrees
1370
1371@table @asis
1372@item @emph{Description}:
1373@code{ASIND(X)} computes the arcsine of its @var{X} in degrees (inverse of
1374@code{SIND(X)}).
1375
1376This function is for compatibility only and should be avoided in favor of
1377standard constructs wherever possible.
1378
1379@item @emph{Standard}:
1380GNU extension, enabled with @option{-fdec-math}.
1381
1382@item @emph{Class}:
1383Elemental function
1384
1385@item @emph{Syntax}:
1386@code{RESULT = ASIND(X)}
1387
1388@item @emph{Arguments}:
1389@multitable @columnfractions .15 .70
1390@item @var{X} @tab The type shall be either @code{REAL} and a magnitude that is
1391less than or equal to one - or be @code{COMPLEX}.
1392@end multitable
1393
1394@item @emph{Return value}:
1395The return value is of the same type and kind as @var{X}.
1396The real part of the result is in degrees and lies in the range
1397@math{-90 \leq \Re \asin(x) \leq 90}.
1398
1399@item @emph{Example}:
1400@smallexample
1401program test_asind
1402  real(8) :: x = 0.866_8
1403  x = asind(x)
1404end program test_asind
1405@end smallexample
1406
1407@item @emph{Specific names}:
1408@multitable @columnfractions .20 .20 .20 .25
1409@item Name            @tab Argument          @tab Return type       @tab Standard
1410@item @code{ASIND(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab GNU extension
1411@item @code{DASIND(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
1412@end multitable
1413
1414@item @emph{See also}:
1415Inverse function: @gol
1416@ref{SIND} @gol
1417Radians function: @gol
1418@ref{ASIN}
1419@end table
1420
1421
1422
1423@node ASINH
1424@section @code{ASINH} --- Inverse hyperbolic sine function
1425@fnindex ASINH
1426@fnindex DASINH
1427@cindex area hyperbolic sine
1428@cindex inverse hyperbolic sine
1429@cindex hyperbolic function, sine, inverse
1430@cindex sine, hyperbolic, inverse
1431
1432@table @asis
1433@item @emph{Description}:
1434@code{ASINH(X)} computes the inverse hyperbolic sine of @var{X}.
1435
1436@item @emph{Standard}:
1437Fortran 2008 and later
1438
1439@item @emph{Class}:
1440Elemental function
1441
1442@item @emph{Syntax}:
1443@code{RESULT = ASINH(X)}
1444
1445@item @emph{Arguments}:
1446@multitable @columnfractions .15 .70
1447@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
1448@end multitable
1449
1450@item @emph{Return value}:
1451The return value is of the same type and kind as  @var{X}. If @var{X} is
1452complex, the imaginary part of the result is in radians and lies between
1453@math{-\pi/2 \leq \Im \asinh(x) \leq \pi/2}.
1454
1455@item @emph{Example}:
1456@smallexample
1457PROGRAM test_asinh
1458  REAL(8), DIMENSION(3) :: x = (/ -1.0, 0.0, 1.0 /)
1459  WRITE (*,*) ASINH(x)
1460END PROGRAM
1461@end smallexample
1462
1463@item @emph{Specific names}:
1464@multitable @columnfractions .20 .20 .20 .25
1465@item Name             @tab Argument          @tab Return type       @tab Standard
1466@item @code{DASINH(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension.
1467@end multitable
1468
1469@item @emph{See also}:
1470Inverse function: @gol
1471@ref{SINH}
1472@end table
1473
1474
1475
1476@node ASSOCIATED
1477@section @code{ASSOCIATED} --- Status of a pointer or pointer/target pair
1478@fnindex ASSOCIATED
1479@cindex pointer, status
1480@cindex association status
1481
1482@table @asis
1483@item @emph{Description}:
1484@code{ASSOCIATED(POINTER [, TARGET])} determines the status of the pointer
1485@var{POINTER} or if @var{POINTER} is associated with the target @var{TARGET}.
1486
1487@item @emph{Standard}:
1488Fortran 90 and later
1489
1490@item @emph{Class}:
1491Inquiry function
1492
1493@item @emph{Syntax}:
1494@code{RESULT = ASSOCIATED(POINTER [, TARGET])}
1495
1496@item @emph{Arguments}:
1497@multitable @columnfractions .15 .70
1498@item @var{POINTER} @tab @var{POINTER} shall have the @code{POINTER} attribute
1499and it can be of any type.
1500@item @var{TARGET} @tab (Optional) @var{TARGET} shall be a pointer or
1501a target.  It must have the same type, kind type parameter, and
1502array rank as @var{POINTER}.
1503@end multitable
1504The association status of neither @var{POINTER} nor @var{TARGET} shall be
1505undefined.
1506
1507@item @emph{Return value}:
1508@code{ASSOCIATED(POINTER)} returns a scalar value of type @code{LOGICAL(4)}.
1509There are several cases:
1510@table @asis
1511@item (A) When the optional @var{TARGET} is not present then
1512@code{ASSOCIATED(POINTER)} is true if @var{POINTER} is associated with a target; otherwise, it returns false.
1513@item (B) If @var{TARGET} is present and a scalar target, the result is true if
1514@var{TARGET} is not a zero-sized storage sequence and the target associated with @var{POINTER} occupies the same storage units.  If @var{POINTER} is
1515disassociated, the result is false.
1516@item (C) If @var{TARGET} is present and an array target, the result is true if
1517@var{TARGET} and @var{POINTER} have the same shape, are not zero-sized arrays,
1518are arrays whose elements are not zero-sized storage sequences, and
1519@var{TARGET} and @var{POINTER} occupy the same storage units in array element
1520order.
1521As in case(B), the result is false, if @var{POINTER} is disassociated.
1522@item (D) If @var{TARGET} is present and an scalar pointer, the result is true
1523if @var{TARGET} is associated with @var{POINTER}, the target associated with
1524@var{TARGET} are not zero-sized storage sequences and occupy the same storage
1525units.
1526The result is false, if either @var{TARGET} or @var{POINTER} is disassociated.
1527@item (E) If @var{TARGET} is present and an array pointer, the result is true if
1528target associated with @var{POINTER} and the target associated with @var{TARGET}
1529have the same shape, are not zero-sized arrays, are arrays whose elements are
1530not zero-sized storage sequences, and @var{TARGET} and @var{POINTER} occupy
1531the same storage units in array element order.
1532The result is false, if either @var{TARGET} or @var{POINTER} is disassociated.
1533@end table
1534
1535@item @emph{Example}:
1536@smallexample
1537program test_associated
1538   implicit none
1539   real, target  :: tgt(2) = (/1., 2./)
1540   real, pointer :: ptr(:)
1541   ptr => tgt
1542   if (associated(ptr)     .eqv. .false.) call abort
1543   if (associated(ptr,tgt) .eqv. .false.) call abort
1544end program test_associated
1545@end smallexample
1546
1547@item @emph{See also}:
1548@ref{NULL}
1549@end table
1550
1551
1552
1553@node ATAN
1554@section @code{ATAN} --- Arctangent function
1555@fnindex ATAN
1556@fnindex DATAN
1557@cindex trigonometric function, tangent, inverse
1558@cindex tangent, inverse
1559
1560@table @asis
1561@item @emph{Description}:
1562@code{ATAN(X)} computes the arctangent of @var{X}.
1563
1564@item @emph{Standard}:
1565Fortran 77 and later, for a complex argument and for two arguments
1566Fortran 2008 or later
1567
1568@item @emph{Class}:
1569Elemental function
1570
1571@item @emph{Syntax}:
1572@multitable @columnfractions .80
1573@item @code{RESULT = ATAN(X)}
1574@item @code{RESULT = ATAN(Y, X)}
1575@end multitable
1576
1577@item @emph{Arguments}:
1578@multitable @columnfractions .15 .70
1579@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX};
1580if @var{Y} is present, @var{X} shall be REAL.
1581@item @var{Y} @tab The type and kind type parameter shall be the same as @var{X}.
1582@end multitable
1583
1584@item @emph{Return value}:
1585The return value is of the same type and kind as @var{X}.
1586If @var{Y} is present, the result is identical to @code{ATAN2(Y,X)}.
1587Otherwise, it the arcus tangent of @var{X}, where the real part of
1588the result is in radians and lies in the range
1589@math{-\pi/2 \leq \Re \atan(x) \leq \pi/2}.
1590
1591@item @emph{Example}:
1592@smallexample
1593program test_atan
1594  real(8) :: x = 2.866_8
1595  x = atan(x)
1596end program test_atan
1597@end smallexample
1598
1599@item @emph{Specific names}:
1600@multitable @columnfractions .20 .20 .20 .25
1601@item Name            @tab Argument          @tab Return type       @tab Standard
1602@item @code{ATAN(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab Fortran 77 and later
1603@item @code{DATAN(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab Fortran 77 and later
1604@end multitable
1605
1606@item @emph{See also}:
1607Inverse function: @gol
1608@ref{TAN} @gol
1609Degrees function: @gol
1610@ref{ATAND}
1611@end table
1612
1613
1614
1615@node ATAND
1616@section @code{ATAND} --- Arctangent function, degrees
1617@fnindex ATAND
1618@fnindex DATAND
1619@cindex trigonometric function, tangent, inverse, degrees
1620@cindex tangent, inverse, degrees
1621
1622@table @asis
1623@item @emph{Description}:
1624@code{ATAND(X)} computes the arctangent of @var{X} in degrees (inverse of
1625@ref{TAND}).
1626
1627This function is for compatibility only and should be avoided in favor of
1628standard constructs wherever possible.
1629
1630@item @emph{Standard}:
1631GNU extension, enabled with @option{-fdec-math}.
1632
1633@item @emph{Class}:
1634Elemental function
1635
1636@item @emph{Syntax}:
1637@multitable @columnfractions .80
1638@item @code{RESULT = ATAND(X)}
1639@item @code{RESULT = ATAND(Y, X)}
1640@end multitable
1641
1642@item @emph{Arguments}:
1643@multitable @columnfractions .15 .70
1644@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX};
1645if @var{Y} is present, @var{X} shall be REAL.
1646@item @var{Y} @tab The type and kind type parameter shall be the same as @var{X}.
1647@end multitable
1648
1649@item @emph{Return value}:
1650The return value is of the same type and kind as @var{X}.
1651If @var{Y} is present, the result is identical to @code{ATAND2(Y,X)}.
1652Otherwise, it is the arcus tangent of @var{X}, where the real part of
1653the result is in degrees and lies in the range
1654@math{-90 \leq \Re \atand(x) \leq 90}.
1655
1656@item @emph{Example}:
1657@smallexample
1658program test_atand
1659  real(8) :: x = 2.866_8
1660  x = atand(x)
1661end program test_atand
1662@end smallexample
1663
1664@item @emph{Specific names}:
1665@multitable @columnfractions .20 .20 .20 .25
1666@item Name            @tab Argument          @tab Return type       @tab Standard
1667@item @code{ATAND(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab GNU extension
1668@item @code{DATAND(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
1669@end multitable
1670
1671@item @emph{See also}:
1672Inverse function: @gol
1673@ref{TAND} @gol
1674Radians function: @gol
1675@ref{ATAN}
1676@end table
1677
1678
1679
1680@node ATAN2
1681@section @code{ATAN2} --- Arctangent function
1682@fnindex ATAN2
1683@fnindex DATAN2
1684@cindex trigonometric function, tangent, inverse
1685@cindex tangent, inverse
1686
1687@table @asis
1688@item @emph{Description}:
1689@code{ATAN2(Y, X)} computes the principal value of the argument
1690function of the complex number @math{X + i Y}.  This function can
1691be used to transform from Cartesian into polar coordinates and
1692allows to determine the angle in the correct quadrant.
1693
1694@item @emph{Standard}:
1695Fortran 77 and later
1696
1697@item @emph{Class}:
1698Elemental function
1699
1700@item @emph{Syntax}:
1701@code{RESULT = ATAN2(Y, X)}
1702
1703@item @emph{Arguments}:
1704@multitable @columnfractions .15 .70
1705@item @var{Y} @tab The type shall be @code{REAL}.
1706@item @var{X} @tab The type and kind type parameter shall be the same as @var{Y}.
1707If @var{Y} is zero, then @var{X} must be nonzero.
1708@end multitable
1709
1710@item @emph{Return value}:
1711The return value has the same type and kind type parameter as @var{Y}. It
1712is the principal value of the complex number @math{X + i Y}.  If @var{X}
1713is nonzero, then it lies in the range @math{-\pi \le \atan (x) \leq \pi}.
1714The sign is positive if @var{Y} is positive.  If @var{Y} is zero, then
1715the return value is zero if @var{X} is strictly positive, @math{\pi} if
1716@var{X} is negative and @var{Y} is positive zero (or the processor does
1717not handle signed zeros), and @math{-\pi} if @var{X} is negative and
1718@var{Y} is negative zero.  Finally, if @var{X} is zero, then the
1719magnitude of the result is @math{\pi/2}.
1720
1721@item @emph{Example}:
1722@smallexample
1723program test_atan2
1724  real(4) :: x = 1.e0_4, y = 0.5e0_4
1725  x = atan2(y,x)
1726end program test_atan2
1727@end smallexample
1728
1729@item @emph{Specific names}:
1730@multitable @columnfractions .20 .20 .20 .25
1731@item Name                @tab Argument            @tab Return type    @tab Standard
1732@item @code{ATAN2(X, Y)}  @tab @code{REAL(4) X, Y} @tab @code{REAL(4)} @tab Fortran 77 and later
1733@item @code{DATAN2(X, Y)} @tab @code{REAL(8) X, Y} @tab @code{REAL(8)} @tab Fortran 77 and later
1734@end multitable
1735
1736@item @emph{See also}:
1737Alias: @gol
1738@ref{ATAN} @gol
1739Degrees function: @gol
1740@ref{ATAN2D}
1741@end table
1742
1743
1744
1745@node ATAN2D
1746@section @code{ATAN2D} --- Arctangent function, degrees
1747@fnindex ATAN2D
1748@fnindex DATAN2D
1749@cindex trigonometric function, tangent, inverse, degrees
1750@cindex tangent, inverse, degrees
1751
1752@table @asis
1753@item @emph{Description}:
1754@code{ATAN2D(Y, X)} computes the principal value of the argument
1755function of the complex number @math{X + i Y} in degrees.  This function can
1756be used to transform from Cartesian into polar coordinates and
1757allows to determine the angle in the correct quadrant.
1758
1759This function is for compatibility only and should be avoided in favor of
1760standard constructs wherever possible.
1761
1762@item @emph{Standard}:
1763GNU extension, enabled with @option{-fdec-math}.
1764
1765@item @emph{Class}:
1766Elemental function
1767
1768@item @emph{Syntax}:
1769@code{RESULT = ATAN2D(Y, X)}
1770
1771@item @emph{Arguments}:
1772@multitable @columnfractions .15 .70
1773@item @var{Y} @tab The type shall be @code{REAL}.
1774@item @var{X} @tab The type and kind type parameter shall be the same as @var{Y}.
1775If @var{Y} is zero, then @var{X} must be nonzero.
1776@end multitable
1777
1778@item @emph{Return value}:
1779The return value has the same type and kind type parameter as @var{Y}. It
1780is the principal value of the complex number @math{X + i Y}.  If @var{X}
1781is nonzero, then it lies in the range @math{-180 \le \atan (x) \leq 180}.
1782The sign is positive if @var{Y} is positive.  If @var{Y} is zero, then
1783the return value is zero if @var{X} is strictly positive, @math{180} if
1784@var{X} is negative and @var{Y} is positive zero (or the processor does
1785not handle signed zeros), and @math{-180} if @var{X} is negative and
1786@var{Y} is negative zero.  Finally, if @var{X} is zero, then the
1787magnitude of the result is @math{90}.
1788
1789@item @emph{Example}:
1790@smallexample
1791program test_atan2d
1792  real(4) :: x = 1.e0_4, y = 0.5e0_4
1793  x = atan2d(y,x)
1794end program test_atan2d
1795@end smallexample
1796
1797@item @emph{Specific names}:
1798@multitable @columnfractions .20 .20 .20 .25
1799@item Name                @tab Argument            @tab Return type    @tab Standard
1800@item @code{ATAN2D(X, Y)}  @tab @code{REAL(4) X, Y} @tab @code{REAL(4)} @tab GNU extension
1801@item @code{DATAN2D(X, Y)} @tab @code{REAL(8) X, Y} @tab @code{REAL(8)} @tab GNU extension
1802@end multitable
1803
1804@item @emph{See also}:
1805Alias: @gol
1806@ref{ATAND} @gol
1807Radians function: @gol
1808@ref{ATAN2}
1809@end table
1810
1811
1812
1813@node ATANH
1814@section @code{ATANH} --- Inverse hyperbolic tangent function
1815@fnindex ATANH
1816@fnindex DATANH
1817@cindex area hyperbolic tangent
1818@cindex inverse hyperbolic tangent
1819@cindex hyperbolic function, tangent, inverse
1820@cindex tangent, hyperbolic, inverse
1821
1822@table @asis
1823@item @emph{Description}:
1824@code{ATANH(X)} computes the inverse hyperbolic tangent of @var{X}.
1825
1826@item @emph{Standard}:
1827Fortran 2008 and later
1828
1829@item @emph{Class}:
1830Elemental function
1831
1832@item @emph{Syntax}:
1833@code{RESULT = ATANH(X)}
1834
1835@item @emph{Arguments}:
1836@multitable @columnfractions .15 .70
1837@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
1838@end multitable
1839
1840@item @emph{Return value}:
1841The return value has same type and kind as @var{X}. If @var{X} is
1842complex, the imaginary part of the result is in radians and lies between
1843@math{-\pi/2 \leq \Im \atanh(x) \leq \pi/2}.
1844
1845@item @emph{Example}:
1846@smallexample
1847PROGRAM test_atanh
1848  REAL, DIMENSION(3) :: x = (/ -1.0, 0.0, 1.0 /)
1849  WRITE (*,*) ATANH(x)
1850END PROGRAM
1851@end smallexample
1852
1853@item @emph{Specific names}:
1854@multitable @columnfractions .20 .20 .20 .25
1855@item Name             @tab Argument          @tab Return type       @tab Standard
1856@item @code{DATANH(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
1857@end multitable
1858
1859@item @emph{See also}:
1860Inverse function: @gol
1861@ref{TANH}
1862@end table
1863
1864
1865
1866@node ATOMIC_ADD
1867@section @code{ATOMIC_ADD} --- Atomic ADD operation
1868@fnindex ATOMIC_ADD
1869@cindex Atomic subroutine, add
1870
1871@table @asis
1872@item @emph{Description}:
1873@code{ATOMIC_ADD(ATOM, VALUE)} atomically adds the value of @var{VAR} to the
1874variable @var{ATOM}. When @var{STAT} is present and the invocation was
1875successful, it is assigned the value 0. If it is present and the invocation
1876has failed, it is assigned a positive value; in particular, for a coindexed
1877@var{ATOM}, if the remote image has stopped, it is assigned the value of
1878@code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image has
1879failed, the value @code{STAT_FAILED_IMAGE}.
1880
1881@item @emph{Standard}:
1882TS 18508 or later
1883
1884@item @emph{Class}:
1885Atomic subroutine
1886
1887@item @emph{Syntax}:
1888@code{CALL ATOMIC_ADD (ATOM, VALUE [, STAT])}
1889
1890@item @emph{Arguments}:
1891@multitable @columnfractions .15 .70
1892@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
1893type with @code{ATOMIC_INT_KIND} kind.
1894@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
1895is different, the value is converted to the kind of @var{ATOM}.
1896@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
1897@end multitable
1898
1899@item @emph{Example}:
1900@smallexample
1901program atomic
1902  use iso_fortran_env
1903  integer(atomic_int_kind) :: atom[*]
1904  call atomic_add (atom[1], this_image())
1905end program atomic
1906@end smallexample
1907
1908@item @emph{See also}:
1909@ref{ATOMIC_DEFINE}, @gol
1910@ref{ATOMIC_FETCH_ADD}, @gol
1911@ref{ISO_FORTRAN_ENV}, @gol
1912@ref{ATOMIC_AND}, @gol
1913@ref{ATOMIC_OR}, @gol
1914@ref{ATOMIC_XOR}
1915@end table
1916
1917
1918
1919
1920@node ATOMIC_AND
1921@section @code{ATOMIC_AND} --- Atomic bitwise AND operation
1922@fnindex ATOMIC_AND
1923@cindex Atomic subroutine, AND
1924
1925@table @asis
1926@item @emph{Description}:
1927@code{ATOMIC_AND(ATOM, VALUE)} atomically defines @var{ATOM} with the bitwise
1928AND between the values of @var{ATOM} and @var{VALUE}. When @var{STAT} is present
1929and the invocation was successful, it is assigned the value 0. If it is present
1930and the invocation has failed, it is assigned a positive value; in particular,
1931for a coindexed @var{ATOM}, if the remote image has stopped, it is assigned the
1932value of @code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote
1933image has failed, the value @code{STAT_FAILED_IMAGE}.
1934
1935@item @emph{Standard}:
1936TS 18508 or later
1937
1938@item @emph{Class}:
1939Atomic subroutine
1940
1941@item @emph{Syntax}:
1942@code{CALL ATOMIC_AND (ATOM, VALUE [, STAT])}
1943
1944@item @emph{Arguments}:
1945@multitable @columnfractions .15 .70
1946@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
1947type with @code{ATOMIC_INT_KIND} kind.
1948@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
1949is different, the value is converted to the kind of @var{ATOM}.
1950@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
1951@end multitable
1952
1953@item @emph{Example}:
1954@smallexample
1955program atomic
1956  use iso_fortran_env
1957  integer(atomic_int_kind) :: atom[*]
1958  call atomic_and (atom[1], int(b'10100011101'))
1959end program atomic
1960@end smallexample
1961
1962@item @emph{See also}:
1963@ref{ATOMIC_DEFINE}, @gol
1964@ref{ATOMIC_FETCH_AND}, @gol
1965@ref{ISO_FORTRAN_ENV}, @gol
1966@ref{ATOMIC_ADD}, @gol
1967@ref{ATOMIC_OR}, @gol
1968@ref{ATOMIC_XOR}
1969@end table
1970
1971
1972
1973@node ATOMIC_CAS
1974@section @code{ATOMIC_CAS} --- Atomic compare and swap
1975@fnindex ATOMIC_DEFINE
1976@cindex Atomic subroutine, compare and swap
1977
1978@table @asis
1979@item @emph{Description}:
1980@code{ATOMIC_CAS} compares the variable @var{ATOM} with the value of
1981@var{COMPARE}; if the value is the same, @var{ATOM} is set to the value
1982of @var{NEW}. Additionally, @var{OLD} is set to the value of @var{ATOM}
1983that was used for the comparison.  When @var{STAT} is present and the invocation
1984was successful, it is assigned the value 0. If it is present and the invocation
1985has failed, it is assigned a positive value; in particular, for a coindexed
1986@var{ATOM}, if the remote image has stopped, it is assigned the value of
1987@code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image has
1988failed, the value @code{STAT_FAILED_IMAGE}.
1989
1990@item @emph{Standard}:
1991TS 18508 or later
1992
1993@item @emph{Class}:
1994Atomic subroutine
1995
1996@item @emph{Syntax}:
1997@code{CALL ATOMIC_CAS (ATOM, OLD, COMPARE, NEW [, STAT])}
1998
1999@item @emph{Arguments}:
2000@multitable @columnfractions .15 .70
2001@item @var{ATOM}    @tab Scalar coarray or coindexed variable of either integer
2002type with @code{ATOMIC_INT_KIND} kind or logical type with
2003@code{ATOMIC_LOGICAL_KIND} kind.
2004@item @var{OLD}     @tab Scalar of the same type and kind as @var{ATOM}.
2005@item @var{COMPARE} @tab Scalar variable of the same type and kind as
2006@var{ATOM}.
2007@item @var{NEW}     @tab Scalar variable of the same type as @var{ATOM}. If kind
2008is different, the value is converted to the kind of @var{ATOM}.
2009@item @var{STAT}    @tab (optional) Scalar default-kind integer variable.
2010@end multitable
2011
2012@item @emph{Example}:
2013@smallexample
2014program atomic
2015  use iso_fortran_env
2016  logical(atomic_logical_kind) :: atom[*], prev
2017  call atomic_cas (atom[1], prev, .false., .true.))
2018end program atomic
2019@end smallexample
2020
2021@item @emph{See also}:
2022@ref{ATOMIC_DEFINE}, @gol
2023@ref{ATOMIC_REF}, @gol
2024@ref{ISO_FORTRAN_ENV}
2025@end table
2026
2027
2028
2029@node ATOMIC_DEFINE
2030@section @code{ATOMIC_DEFINE} --- Setting a variable atomically
2031@fnindex ATOMIC_DEFINE
2032@cindex Atomic subroutine, define
2033
2034@table @asis
2035@item @emph{Description}:
2036@code{ATOMIC_DEFINE(ATOM, VALUE)} defines the variable @var{ATOM} with the value
2037@var{VALUE} atomically. When @var{STAT} is present and the invocation was
2038successful, it is assigned the value 0. If it is present and the invocation
2039has failed, it is assigned a positive value; in particular, for a coindexed
2040@var{ATOM}, if the remote image has stopped, it is assigned the value of
2041@code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image has
2042failed, the value @code{STAT_FAILED_IMAGE}.
2043
2044@item @emph{Standard}:
2045Fortran 2008 and later; with @var{STAT}, TS 18508 or later
2046
2047@item @emph{Class}:
2048Atomic subroutine
2049
2050@item @emph{Syntax}:
2051@code{CALL ATOMIC_DEFINE (ATOM, VALUE [, STAT])}
2052
2053@item @emph{Arguments}:
2054@multitable @columnfractions .15 .70
2055@item @var{ATOM}   @tab Scalar coarray or coindexed variable of either integer
2056type with @code{ATOMIC_INT_KIND} kind or logical type with
2057@code{ATOMIC_LOGICAL_KIND} kind.
2058
2059@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2060is different, the value is converted to the kind of @var{ATOM}.
2061@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2062@end multitable
2063
2064@item @emph{Example}:
2065@smallexample
2066program atomic
2067  use iso_fortran_env
2068  integer(atomic_int_kind) :: atom[*]
2069  call atomic_define (atom[1], this_image())
2070end program atomic
2071@end smallexample
2072
2073@item @emph{See also}:
2074@ref{ATOMIC_REF}, @gol
2075@ref{ATOMIC_CAS}, @gol
2076@ref{ISO_FORTRAN_ENV}, @gol
2077@ref{ATOMIC_ADD}, @gol
2078@ref{ATOMIC_AND}, @gol
2079@ref{ATOMIC_OR}, @gol
2080@ref{ATOMIC_XOR}
2081@end table
2082
2083
2084
2085@node ATOMIC_FETCH_ADD
2086@section @code{ATOMIC_FETCH_ADD} --- Atomic ADD operation with prior fetch
2087@fnindex ATOMIC_FETCH_ADD
2088@cindex Atomic subroutine, ADD with fetch
2089
2090@table @asis
2091@item @emph{Description}:
2092@code{ATOMIC_FETCH_ADD(ATOM, VALUE, OLD)} atomically stores the value of
2093@var{ATOM} in @var{OLD} and adds the value of @var{VAR} to the
2094variable @var{ATOM}. When @var{STAT} is present and the invocation was
2095successful, it is assigned the value 0. If it is present and the invocation
2096has failed, it is assigned a positive value; in particular, for a coindexed
2097@var{ATOM}, if the remote image has stopped, it is assigned the value of
2098@code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image has
2099failed, the value @code{STAT_FAILED_IMAGE}.
2100
2101@item @emph{Standard}:
2102TS 18508 or later
2103
2104@item @emph{Class}:
2105Atomic subroutine
2106
2107@item @emph{Syntax}:
2108@code{CALL ATOMIC_FETCH_ADD (ATOM, VALUE, old [, STAT])}
2109
2110@item @emph{Arguments}:
2111@multitable @columnfractions .15 .70
2112@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
2113type with @code{ATOMIC_INT_KIND} kind.
2114@code{ATOMIC_LOGICAL_KIND} kind.
2115
2116@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2117is different, the value is converted to the kind of @var{ATOM}.
2118@item @var{OLD}    @tab Scalar of the same type and kind as @var{ATOM}.
2119@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2120@end multitable
2121
2122@item @emph{Example}:
2123@smallexample
2124program atomic
2125  use iso_fortran_env
2126  integer(atomic_int_kind) :: atom[*], old
2127  call atomic_add (atom[1], this_image(), old)
2128end program atomic
2129@end smallexample
2130
2131@item @emph{See also}:
2132@ref{ATOMIC_DEFINE}, @gol
2133@ref{ATOMIC_ADD}, @gol
2134@ref{ISO_FORTRAN_ENV}, @gol
2135@ref{ATOMIC_FETCH_AND}, @gol
2136@ref{ATOMIC_FETCH_OR}, @gol
2137@ref{ATOMIC_FETCH_XOR}
2138@end table
2139
2140
2141
2142@node ATOMIC_FETCH_AND
2143@section @code{ATOMIC_FETCH_AND} --- Atomic bitwise AND operation with prior fetch
2144@fnindex ATOMIC_FETCH_AND
2145@cindex Atomic subroutine, AND with fetch
2146
2147@table @asis
2148@item @emph{Description}:
2149@code{ATOMIC_AND(ATOM, VALUE)} atomically stores the value of @var{ATOM} in
2150@var{OLD} and defines @var{ATOM} with the bitwise AND between the values of
2151@var{ATOM} and @var{VALUE}. When @var{STAT} is present and the invocation was
2152successful, it is assigned the value 0. If it is present and the invocation has
2153failed, it is assigned a positive value; in particular, for a coindexed
2154@var{ATOM}, if the remote image has stopped, it is assigned the value of
2155@code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image has
2156failed, the value @code{STAT_FAILED_IMAGE}.
2157
2158@item @emph{Standard}:
2159TS 18508 or later
2160
2161@item @emph{Class}:
2162Atomic subroutine
2163
2164@item @emph{Syntax}:
2165@code{CALL ATOMIC_FETCH_AND (ATOM, VALUE, OLD [, STAT])}
2166
2167@item @emph{Arguments}:
2168@multitable @columnfractions .15 .70
2169@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
2170type with @code{ATOMIC_INT_KIND} kind.
2171@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2172is different, the value is converted to the kind of @var{ATOM}.
2173@item @var{OLD}    @tab Scalar of the same type and kind as @var{ATOM}.
2174@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2175@end multitable
2176
2177@item @emph{Example}:
2178@smallexample
2179program atomic
2180  use iso_fortran_env
2181  integer(atomic_int_kind) :: atom[*], old
2182  call atomic_fetch_and (atom[1], int(b'10100011101'), old)
2183end program atomic
2184@end smallexample
2185
2186@item @emph{See also}:
2187@ref{ATOMIC_DEFINE}, @gol
2188@ref{ATOMIC_AND}, @gol
2189@ref{ISO_FORTRAN_ENV}, @gol
2190@ref{ATOMIC_FETCH_ADD}, @gol
2191@ref{ATOMIC_FETCH_OR}, @gol
2192@ref{ATOMIC_FETCH_XOR}
2193@end table
2194
2195
2196
2197@node ATOMIC_FETCH_OR
2198@section @code{ATOMIC_FETCH_OR} --- Atomic bitwise OR operation with prior fetch
2199@fnindex ATOMIC_FETCH_OR
2200@cindex Atomic subroutine, OR with fetch
2201
2202@table @asis
2203@item @emph{Description}:
2204@code{ATOMIC_OR(ATOM, VALUE)} atomically stores the value of @var{ATOM} in
2205@var{OLD} and defines @var{ATOM} with the bitwise OR between the values of
2206@var{ATOM} and @var{VALUE}. When @var{STAT} is present and the invocation was
2207successful, it is assigned the value 0. If it is present and the invocation has
2208failed, it is assigned a positive value; in particular, for a coindexed
2209@var{ATOM}, if the remote image has stopped, it is assigned the value of
2210@code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image has
2211failed, the value @code{STAT_FAILED_IMAGE}.
2212
2213@item @emph{Standard}:
2214TS 18508 or later
2215
2216@item @emph{Class}:
2217Atomic subroutine
2218
2219@item @emph{Syntax}:
2220@code{CALL ATOMIC_FETCH_OR (ATOM, VALUE, OLD [, STAT])}
2221
2222@item @emph{Arguments}:
2223@multitable @columnfractions .15 .70
2224@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
2225type with @code{ATOMIC_INT_KIND} kind.
2226@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2227is different, the value is converted to the kind of @var{ATOM}.
2228@item @var{OLD}    @tab Scalar of the same type and kind as @var{ATOM}.
2229@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2230@end multitable
2231
2232@item @emph{Example}:
2233@smallexample
2234program atomic
2235  use iso_fortran_env
2236  integer(atomic_int_kind) :: atom[*], old
2237  call atomic_fetch_or (atom[1], int(b'10100011101'), old)
2238end program atomic
2239@end smallexample
2240
2241@item @emph{See also}:
2242@ref{ATOMIC_DEFINE}, @gol
2243@ref{ATOMIC_OR}, @gol
2244@ref{ISO_FORTRAN_ENV}, @gol
2245@ref{ATOMIC_FETCH_ADD}, @gol
2246@ref{ATOMIC_FETCH_AND}, @gol
2247@ref{ATOMIC_FETCH_XOR}
2248@end table
2249
2250
2251
2252@node ATOMIC_FETCH_XOR
2253@section @code{ATOMIC_FETCH_XOR} --- Atomic bitwise XOR operation with prior fetch
2254@fnindex ATOMIC_FETCH_XOR
2255@cindex Atomic subroutine, XOR with fetch
2256
2257@table @asis
2258@item @emph{Description}:
2259@code{ATOMIC_XOR(ATOM, VALUE)} atomically stores the value of @var{ATOM} in
2260@var{OLD} and defines @var{ATOM} with the bitwise XOR between the values of
2261@var{ATOM} and @var{VALUE}. When @var{STAT} is present and the invocation was
2262successful, it is assigned the value 0. If it is present and the invocation has
2263failed, it is assigned a positive value; in particular, for a coindexed
2264@var{ATOM}, if the remote image has stopped, it is assigned the value of
2265@code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image has
2266failed, the value @code{STAT_FAILED_IMAGE}.
2267
2268@item @emph{Standard}:
2269TS 18508 or later
2270
2271@item @emph{Class}:
2272Atomic subroutine
2273
2274@item @emph{Syntax}:
2275@code{CALL ATOMIC_FETCH_XOR (ATOM, VALUE, OLD [, STAT])}
2276
2277@item @emph{Arguments}:
2278@multitable @columnfractions .15 .70
2279@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
2280type with @code{ATOMIC_INT_KIND} kind.
2281@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2282is different, the value is converted to the kind of @var{ATOM}.
2283@item @var{OLD}    @tab Scalar of the same type and kind as @var{ATOM}.
2284@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2285@end multitable
2286
2287@item @emph{Example}:
2288@smallexample
2289program atomic
2290  use iso_fortran_env
2291  integer(atomic_int_kind) :: atom[*], old
2292  call atomic_fetch_xor (atom[1], int(b'10100011101'), old)
2293end program atomic
2294@end smallexample
2295
2296@item @emph{See also}:
2297@ref{ATOMIC_DEFINE}, @gol
2298@ref{ATOMIC_XOR}, @gol
2299@ref{ISO_FORTRAN_ENV}, @gol
2300@ref{ATOMIC_FETCH_ADD}, @gol
2301@ref{ATOMIC_FETCH_AND}, @gol
2302@ref{ATOMIC_FETCH_OR}
2303@end table
2304
2305
2306
2307@node ATOMIC_OR
2308@section @code{ATOMIC_OR} --- Atomic bitwise OR operation
2309@fnindex ATOMIC_OR
2310@cindex Atomic subroutine, OR
2311
2312@table @asis
2313@item @emph{Description}:
2314@code{ATOMIC_OR(ATOM, VALUE)} atomically defines @var{ATOM} with the bitwise
2315AND between the values of @var{ATOM} and @var{VALUE}. When @var{STAT} is present
2316and the invocation was successful, it is assigned the value 0. If it is present
2317and the invocation has failed, it is assigned a positive value; in particular,
2318for a coindexed @var{ATOM}, if the remote image has stopped, it is assigned the
2319value of @code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote
2320image has failed, the value @code{STAT_FAILED_IMAGE}.
2321
2322@item @emph{Standard}:
2323TS 18508 or later
2324
2325@item @emph{Class}:
2326Atomic subroutine
2327
2328@item @emph{Syntax}:
2329@code{CALL ATOMIC_OR (ATOM, VALUE [, STAT])}
2330
2331@item @emph{Arguments}:
2332@multitable @columnfractions .15 .70
2333@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
2334type with @code{ATOMIC_INT_KIND} kind.
2335@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2336is different, the value is converted to the kind of @var{ATOM}.
2337@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2338@end multitable
2339
2340@item @emph{Example}:
2341@smallexample
2342program atomic
2343  use iso_fortran_env
2344  integer(atomic_int_kind) :: atom[*]
2345  call atomic_or (atom[1], int(b'10100011101'))
2346end program atomic
2347@end smallexample
2348
2349@item @emph{See also}:
2350@ref{ATOMIC_DEFINE}, @gol
2351@ref{ATOMIC_FETCH_OR}, @gol
2352@ref{ISO_FORTRAN_ENV}, @gol
2353@ref{ATOMIC_ADD}, @gol
2354@ref{ATOMIC_OR}, @gol
2355@ref{ATOMIC_XOR}
2356@end table
2357
2358
2359
2360@node ATOMIC_REF
2361@section @code{ATOMIC_REF} --- Obtaining the value of a variable atomically
2362@fnindex ATOMIC_REF
2363@cindex Atomic subroutine, reference
2364
2365@table @asis
2366@item @emph{Description}:
2367@code{ATOMIC_DEFINE(ATOM, VALUE)} atomically assigns the value of the
2368variable @var{ATOM} to @var{VALUE}. When @var{STAT} is present and the
2369invocation was successful, it is assigned the value 0. If it is present and the
2370invocation has failed, it is assigned a positive value; in particular, for a
2371coindexed @var{ATOM}, if the remote image has stopped, it is assigned the value
2372of @code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote image
2373has failed, the value @code{STAT_FAILED_IMAGE}.
2374
2375
2376@item @emph{Standard}:
2377Fortran 2008 and later; with @var{STAT}, TS 18508 or later
2378
2379@item @emph{Class}:
2380Atomic subroutine
2381
2382@item @emph{Syntax}:
2383@code{CALL ATOMIC_REF(VALUE, ATOM [, STAT])}
2384
2385@item @emph{Arguments}:
2386@multitable @columnfractions .15 .70
2387@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2388is different, the value is converted to the kind of @var{ATOM}.
2389@item @var{ATOM}   @tab Scalar coarray or coindexed variable of either integer
2390type with @code{ATOMIC_INT_KIND} kind or logical type with
2391@code{ATOMIC_LOGICAL_KIND} kind.
2392@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2393@end multitable
2394
2395@item @emph{Example}:
2396@smallexample
2397program atomic
2398  use iso_fortran_env
2399  logical(atomic_logical_kind) :: atom[*]
2400  logical :: val
2401  call atomic_ref (atom, .false.)
2402  ! ...
2403  call atomic_ref (atom, val)
2404  if (val) then
2405    print *, "Obtained"
2406  end if
2407end program atomic
2408@end smallexample
2409
2410@item @emph{See also}:
2411@ref{ATOMIC_DEFINE}, @gol
2412@ref{ATOMIC_CAS}, @gol
2413@ref{ISO_FORTRAN_ENV}, @gol
2414@ref{ATOMIC_FETCH_ADD}, @gol
2415@ref{ATOMIC_FETCH_AND}, @gol
2416@ref{ATOMIC_FETCH_OR}, @gol
2417@ref{ATOMIC_FETCH_XOR}
2418@end table
2419
2420
2421@node ATOMIC_XOR
2422@section @code{ATOMIC_XOR} --- Atomic bitwise OR operation
2423@fnindex ATOMIC_XOR
2424@cindex Atomic subroutine, XOR
2425
2426@table @asis
2427@item @emph{Description}:
2428@code{ATOMIC_AND(ATOM, VALUE)} atomically defines @var{ATOM} with the bitwise
2429XOR between the values of @var{ATOM} and @var{VALUE}. When @var{STAT} is present
2430and the invocation was successful, it is assigned the value 0. If it is present
2431and the invocation has failed, it is assigned a positive value; in particular,
2432for a coindexed @var{ATOM}, if the remote image has stopped, it is assigned the
2433value of @code{ISO_FORTRAN_ENV}'s @code{STAT_STOPPED_IMAGE} and if the remote
2434image has failed, the value @code{STAT_FAILED_IMAGE}.
2435
2436@item @emph{Standard}:
2437TS 18508 or later
2438
2439@item @emph{Class}:
2440Atomic subroutine
2441
2442@item @emph{Syntax}:
2443@code{CALL ATOMIC_XOR (ATOM, VALUE [, STAT])}
2444
2445@item @emph{Arguments}:
2446@multitable @columnfractions .15 .70
2447@item @var{ATOM}   @tab Scalar coarray or coindexed variable of integer
2448type with @code{ATOMIC_INT_KIND} kind.
2449@item @var{VALUE}  @tab Scalar of the same type as @var{ATOM}. If the kind
2450is different, the value is converted to the kind of @var{ATOM}.
2451@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
2452@end multitable
2453
2454@item @emph{Example}:
2455@smallexample
2456program atomic
2457  use iso_fortran_env
2458  integer(atomic_int_kind) :: atom[*]
2459  call atomic_xor (atom[1], int(b'10100011101'))
2460end program atomic
2461@end smallexample
2462
2463@item @emph{See also}:
2464@ref{ATOMIC_DEFINE}, @gol
2465@ref{ATOMIC_FETCH_XOR}, @gol
2466@ref{ISO_FORTRAN_ENV}, @gol
2467@ref{ATOMIC_ADD}, @gol
2468@ref{ATOMIC_OR}, @gol
2469@ref{ATOMIC_XOR}
2470@end table
2471
2472
2473@node BACKTRACE
2474@section @code{BACKTRACE} --- Show a backtrace
2475@fnindex BACKTRACE
2476@cindex backtrace
2477
2478@table @asis
2479@item @emph{Description}:
2480@code{BACKTRACE} shows a backtrace at an arbitrary place in user code. Program
2481execution continues normally afterwards. The backtrace information is printed
2482to the unit corresponding to @code{ERROR_UNIT} in @code{ISO_FORTRAN_ENV}.
2483
2484@item @emph{Standard}:
2485GNU extension
2486
2487@item @emph{Class}:
2488Subroutine
2489
2490@item @emph{Syntax}:
2491@code{CALL BACKTRACE}
2492
2493@item @emph{Arguments}:
2494None
2495
2496@item @emph{See also}:
2497@ref{ABORT}
2498@end table
2499
2500
2501
2502@node BESSEL_J0
2503@section @code{BESSEL_J0} --- Bessel function of the first kind of order 0
2504@fnindex BESSEL_J0
2505@fnindex BESJ0
2506@fnindex DBESJ0
2507@cindex Bessel function, first kind
2508
2509@table @asis
2510@item @emph{Description}:
2511@code{BESSEL_J0(X)} computes the Bessel function of the first kind of
2512order 0 of @var{X}. This function is available under the name
2513@code{BESJ0} as a GNU extension.
2514
2515@item @emph{Standard}:
2516Fortran 2008 and later
2517
2518@item @emph{Class}:
2519Elemental function
2520
2521@item @emph{Syntax}:
2522@code{RESULT = BESSEL_J0(X)}
2523
2524@item @emph{Arguments}:
2525@multitable @columnfractions .15 .70
2526@item @var{X} @tab The type shall be @code{REAL}.
2527@end multitable
2528
2529@item @emph{Return value}:
2530The return value is of type @code{REAL} and lies in the
2531range @math{ - 0.4027... \leq Bessel (0,x) \leq 1}. It has the same
2532kind as @var{X}.
2533
2534@item @emph{Example}:
2535@smallexample
2536program test_besj0
2537  real(8) :: x = 0.0_8
2538  x = bessel_j0(x)
2539end program test_besj0
2540@end smallexample
2541
2542@item @emph{Specific names}:
2543@multitable @columnfractions .20 .20 .20 .25
2544@item Name            @tab Argument          @tab Return type       @tab Standard
2545@item @code{DBESJ0(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}   @tab GNU extension
2546@end multitable
2547@end table
2548
2549
2550
2551@node BESSEL_J1
2552@section @code{BESSEL_J1} --- Bessel function of the first kind of order 1
2553@fnindex BESSEL_J1
2554@fnindex BESJ1
2555@fnindex DBESJ1
2556@cindex Bessel function, first kind
2557
2558@table @asis
2559@item @emph{Description}:
2560@code{BESSEL_J1(X)} computes the Bessel function of the first kind of
2561order 1 of @var{X}. This function is available under the name
2562@code{BESJ1} as a GNU extension.
2563
2564@item @emph{Standard}:
2565Fortran 2008
2566
2567@item @emph{Class}:
2568Elemental function
2569
2570@item @emph{Syntax}:
2571@code{RESULT = BESSEL_J1(X)}
2572
2573@item @emph{Arguments}:
2574@multitable @columnfractions .15 .70
2575@item @var{X} @tab The type shall be @code{REAL}.
2576@end multitable
2577
2578@item @emph{Return value}:
2579The return value is of type @code{REAL} and lies in the
2580range @math{ - 0.5818... \leq Bessel (0,x) \leq 0.5818 }. It has the same
2581kind as @var{X}.
2582
2583@item @emph{Example}:
2584@smallexample
2585program test_besj1
2586  real(8) :: x = 1.0_8
2587  x = bessel_j1(x)
2588end program test_besj1
2589@end smallexample
2590
2591@item @emph{Specific names}:
2592@multitable @columnfractions .20 .20 .20 .25
2593@item Name             @tab Argument          @tab Return type       @tab Standard
2594@item @code{DBESJ1(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
2595@end multitable
2596@end table
2597
2598
2599
2600@node BESSEL_JN
2601@section @code{BESSEL_JN} --- Bessel function of the first kind
2602@fnindex BESSEL_JN
2603@fnindex BESJN
2604@fnindex DBESJN
2605@cindex Bessel function, first kind
2606
2607@table @asis
2608@item @emph{Description}:
2609@code{BESSEL_JN(N, X)} computes the Bessel function of the first kind of
2610order @var{N} of @var{X}. This function is available under the name
2611@code{BESJN} as a GNU extension.  If @var{N} and @var{X} are arrays,
2612their ranks and shapes shall conform.
2613
2614@code{BESSEL_JN(N1, N2, X)} returns an array with the Bessel functions
2615of the first kind of the orders @var{N1} to @var{N2}.
2616
2617@item @emph{Standard}:
2618Fortran 2008 and later, negative @var{N} is allowed as GNU extension
2619
2620@item @emph{Class}:
2621Elemental function, except for the transformational function
2622@code{BESSEL_JN(N1, N2, X)}
2623
2624@item @emph{Syntax}:
2625@multitable @columnfractions .80
2626@item @code{RESULT = BESSEL_JN(N, X)}
2627@item @code{RESULT = BESSEL_JN(N1, N2, X)}
2628@end multitable
2629
2630@item @emph{Arguments}:
2631@multitable @columnfractions .15 .70
2632@item @var{N} @tab Shall be a scalar or an array of type  @code{INTEGER}.
2633@item @var{N1} @tab Shall be a non-negative scalar of type  @code{INTEGER}.
2634@item @var{N2} @tab Shall be a non-negative scalar of type  @code{INTEGER}.
2635@item @var{X} @tab Shall be a scalar or an array of type  @code{REAL};
2636for @code{BESSEL_JN(N1, N2, X)} it shall be scalar.
2637@end multitable
2638
2639@item @emph{Return value}:
2640The return value is a scalar of type @code{REAL}. It has the same
2641kind as @var{X}.
2642
2643@item @emph{Note}:
2644The transformational function uses a recurrence algorithm which might,
2645for some values of @var{X}, lead to different results than calls to
2646the elemental function.
2647
2648@item @emph{Example}:
2649@smallexample
2650program test_besjn
2651  real(8) :: x = 1.0_8
2652  x = bessel_jn(5,x)
2653end program test_besjn
2654@end smallexample
2655
2656@item @emph{Specific names}:
2657@multitable @columnfractions .20 .20 .20 .25
2658@item Name                @tab Argument            @tab Return type       @tab Standard
2659@item @code{DBESJN(N, X)} @tab @code{INTEGER N}    @tab @code{REAL(8)}    @tab GNU extension
2660@item                     @tab @code{REAL(8) X}    @tab                   @tab
2661@end multitable
2662@end table
2663
2664
2665
2666@node BESSEL_Y0
2667@section @code{BESSEL_Y0} --- Bessel function of the second kind of order 0
2668@fnindex BESSEL_Y0
2669@fnindex BESY0
2670@fnindex DBESY0
2671@cindex Bessel function, second kind
2672
2673@table @asis
2674@item @emph{Description}:
2675@code{BESSEL_Y0(X)} computes the Bessel function of the second kind of
2676order 0 of @var{X}. This function is available under the name
2677@code{BESY0} as a GNU extension.
2678
2679@item @emph{Standard}:
2680Fortran 2008 and later
2681
2682@item @emph{Class}:
2683Elemental function
2684
2685@item @emph{Syntax}:
2686@code{RESULT = BESSEL_Y0(X)}
2687
2688@item @emph{Arguments}:
2689@multitable @columnfractions .15 .70
2690@item @var{X} @tab The type shall be @code{REAL}.
2691@end multitable
2692
2693@item @emph{Return value}:
2694The return value is of type @code{REAL}. It has the same kind as @var{X}.
2695
2696@item @emph{Example}:
2697@smallexample
2698program test_besy0
2699  real(8) :: x = 0.0_8
2700  x = bessel_y0(x)
2701end program test_besy0
2702@end smallexample
2703
2704@item @emph{Specific names}:
2705@multitable @columnfractions .20 .20 .20 .25
2706@item Name            @tab Argument          @tab Return type       @tab Standard
2707@item @code{DBESY0(X)}@tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
2708@end multitable
2709@end table
2710
2711
2712
2713@node BESSEL_Y1
2714@section @code{BESSEL_Y1} --- Bessel function of the second kind of order 1
2715@fnindex BESSEL_Y1
2716@fnindex BESY1
2717@fnindex DBESY1
2718@cindex Bessel function, second kind
2719
2720@table @asis
2721@item @emph{Description}:
2722@code{BESSEL_Y1(X)} computes the Bessel function of the second kind of
2723order 1 of @var{X}. This function is available under the name
2724@code{BESY1} as a GNU extension.
2725
2726@item @emph{Standard}:
2727Fortran 2008 and later
2728
2729@item @emph{Class}:
2730Elemental function
2731
2732@item @emph{Syntax}:
2733@code{RESULT = BESSEL_Y1(X)}
2734
2735@item @emph{Arguments}:
2736@multitable @columnfractions .15 .70
2737@item @var{X} @tab The type shall be @code{REAL}.
2738@end multitable
2739
2740@item @emph{Return value}:
2741The return value is of type @code{REAL}. It has the same kind as @var{X}.
2742
2743@item @emph{Example}:
2744@smallexample
2745program test_besy1
2746  real(8) :: x = 1.0_8
2747  x = bessel_y1(x)
2748end program test_besy1
2749@end smallexample
2750
2751@item @emph{Specific names}:
2752@multitable @columnfractions .20 .20 .20 .25
2753@item Name            @tab Argument          @tab Return type       @tab Standard
2754@item @code{DBESY1(X)}@tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
2755@end multitable
2756@end table
2757
2758
2759
2760@node BESSEL_YN
2761@section @code{BESSEL_YN} --- Bessel function of the second kind
2762@fnindex BESSEL_YN
2763@fnindex BESYN
2764@fnindex DBESYN
2765@cindex Bessel function, second kind
2766
2767@table @asis
2768@item @emph{Description}:
2769@code{BESSEL_YN(N, X)} computes the Bessel function of the second kind of
2770order @var{N} of @var{X}. This function is available under the name
2771@code{BESYN} as a GNU extension.  If @var{N} and @var{X} are arrays,
2772their ranks and shapes shall conform.
2773
2774@code{BESSEL_YN(N1, N2, X)} returns an array with the Bessel functions
2775of the first kind of the orders @var{N1} to @var{N2}.
2776
2777@item @emph{Standard}:
2778Fortran 2008 and later, negative @var{N} is allowed as GNU extension
2779
2780@item @emph{Class}:
2781Elemental function, except for the transformational function
2782@code{BESSEL_YN(N1, N2, X)}
2783
2784@item @emph{Syntax}:
2785@multitable @columnfractions .80
2786@item @code{RESULT = BESSEL_YN(N, X)}
2787@item @code{RESULT = BESSEL_YN(N1, N2, X)}
2788@end multitable
2789
2790@item @emph{Arguments}:
2791@multitable @columnfractions .15 .70
2792@item @var{N} @tab Shall be a scalar or an array of type  @code{INTEGER} .
2793@item @var{N1} @tab Shall be a non-negative scalar of type  @code{INTEGER}.
2794@item @var{N2} @tab Shall be a non-negative scalar of type  @code{INTEGER}.
2795@item @var{X} @tab Shall be a scalar or an array of type  @code{REAL};
2796for @code{BESSEL_YN(N1, N2, X)} it shall be scalar.
2797@end multitable
2798
2799@item @emph{Return value}:
2800The return value is a scalar of type @code{REAL}. It has the same
2801kind as @var{X}.
2802
2803@item @emph{Note}:
2804The transformational function uses a recurrence algorithm which might,
2805for some values of @var{X}, lead to different results than calls to
2806the elemental function.
2807
2808@item @emph{Example}:
2809@smallexample
2810program test_besyn
2811  real(8) :: x = 1.0_8
2812  x = bessel_yn(5,x)
2813end program test_besyn
2814@end smallexample
2815
2816@item @emph{Specific names}:
2817@multitable @columnfractions .20 .20 .20 .25
2818@item Name               @tab Argument            @tab Return type     @tab Standard
2819@item @code{DBESYN(N,X)} @tab @code{INTEGER N} @tab @code{REAL(8)}  @tab GNU extension
2820@item                    @tab @code{REAL(8) X} @tab                 @tab
2821@end multitable
2822@end table
2823
2824
2825
2826@node BGE
2827@section @code{BGE} --- Bitwise greater than or equal to
2828@fnindex BGE
2829@cindex bitwise comparison
2830
2831@table @asis
2832@item @emph{Description}:
2833Determines whether an integral is a bitwise greater than or equal to
2834another.
2835
2836@item @emph{Standard}:
2837Fortran 2008 and later
2838
2839@item @emph{Class}:
2840Elemental function
2841
2842@item @emph{Syntax}:
2843@code{RESULT = BGE(I, J)}
2844
2845@item @emph{Arguments}:
2846@multitable @columnfractions .15 .70
2847@item @var{I} @tab Shall be of @code{INTEGER} type.
2848@item @var{J} @tab Shall be of @code{INTEGER} type, and of the same kind
2849as @var{I}.
2850@end multitable
2851
2852@item @emph{Return value}:
2853The return value is of type @code{LOGICAL} and of the default kind.
2854
2855@item @emph{See also}:
2856@ref{BGT}, @gol
2857@ref{BLE}, @gol
2858@ref{BLT}
2859@end table
2860
2861
2862
2863@node BGT
2864@section @code{BGT} --- Bitwise greater than
2865@fnindex BGT
2866@cindex bitwise comparison
2867
2868@table @asis
2869@item @emph{Description}:
2870Determines whether an integral is a bitwise greater than another.
2871
2872@item @emph{Standard}:
2873Fortran 2008 and later
2874
2875@item @emph{Class}:
2876Elemental function
2877
2878@item @emph{Syntax}:
2879@code{RESULT = BGT(I, J)}
2880
2881@item @emph{Arguments}:
2882@multitable @columnfractions .15 .70
2883@item @var{I} @tab Shall be of @code{INTEGER} type.
2884@item @var{J} @tab Shall be of @code{INTEGER} type, and of the same kind
2885as @var{I}.
2886@end multitable
2887
2888@item @emph{Return value}:
2889The return value is of type @code{LOGICAL} and of the default kind.
2890
2891@item @emph{See also}:
2892@ref{BGE}, @gol
2893@ref{BLE}, @gol
2894@ref{BLT}
2895@end table
2896
2897
2898
2899@node BIT_SIZE
2900@section @code{BIT_SIZE} --- Bit size inquiry function
2901@fnindex BIT_SIZE
2902@cindex bits, number of
2903@cindex size of a variable, in bits
2904
2905@table @asis
2906@item @emph{Description}:
2907@code{BIT_SIZE(I)} returns the number of bits (integer precision plus sign bit)
2908represented by the type of @var{I}.  The result of @code{BIT_SIZE(I)} is
2909independent of the actual value of @var{I}.
2910
2911@item @emph{Standard}:
2912Fortran 90 and later
2913
2914@item @emph{Class}:
2915Inquiry function
2916
2917@item @emph{Syntax}:
2918@code{RESULT = BIT_SIZE(I)}
2919
2920@item @emph{Arguments}:
2921@multitable @columnfractions .15 .70
2922@item @var{I} @tab The type shall be @code{INTEGER}.
2923@end multitable
2924
2925@item @emph{Return value}:
2926The return value is of type @code{INTEGER}
2927
2928@item @emph{Example}:
2929@smallexample
2930program test_bit_size
2931    integer :: i = 123
2932    integer :: size
2933    size = bit_size(i)
2934    print *, size
2935end program test_bit_size
2936@end smallexample
2937@end table
2938
2939
2940
2941@node BLE
2942@section @code{BLE} --- Bitwise less than or equal to
2943@fnindex BLE
2944@cindex bitwise comparison
2945
2946@table @asis
2947@item @emph{Description}:
2948Determines whether an integral is a bitwise less than or equal to
2949another.
2950
2951@item @emph{Standard}:
2952Fortran 2008 and later
2953
2954@item @emph{Class}:
2955Elemental function
2956
2957@item @emph{Syntax}:
2958@code{RESULT = BLE(I, J)}
2959
2960@item @emph{Arguments}:
2961@multitable @columnfractions .15 .70
2962@item @var{I} @tab Shall be of @code{INTEGER} type.
2963@item @var{J} @tab Shall be of @code{INTEGER} type, and of the same kind
2964as @var{I}.
2965@end multitable
2966
2967@item @emph{Return value}:
2968The return value is of type @code{LOGICAL} and of the default kind.
2969
2970@item @emph{See also}:
2971@ref{BGT}, @gol
2972@ref{BGE}, @gol
2973@ref{BLT}
2974@end table
2975
2976
2977
2978@node BLT
2979@section @code{BLT} --- Bitwise less than
2980@fnindex BLT
2981@cindex bitwise comparison
2982
2983@table @asis
2984@item @emph{Description}:
2985Determines whether an integral is a bitwise less than another.
2986
2987@item @emph{Standard}:
2988Fortran 2008 and later
2989
2990@item @emph{Class}:
2991Elemental function
2992
2993@item @emph{Syntax}:
2994@code{RESULT = BLT(I, J)}
2995
2996@item @emph{Arguments}:
2997@multitable @columnfractions .15 .70
2998@item @var{I} @tab Shall be of @code{INTEGER} type.
2999@item @var{J} @tab Shall be of @code{INTEGER} type, and of the same kind
3000as @var{I}.
3001@end multitable
3002
3003@item @emph{Return value}:
3004The return value is of type @code{LOGICAL} and of the default kind.
3005
3006@item @emph{See also}:
3007@ref{BGE}, @gol
3008@ref{BGT}, @gol
3009@ref{BLE}
3010@end table
3011
3012
3013
3014@node BTEST
3015@section @code{BTEST} --- Bit test function
3016@fnindex BTEST
3017@fnindex BBTEST
3018@fnindex BITEST
3019@fnindex BJTEST
3020@fnindex BKTEST
3021@cindex bits, testing
3022
3023@table @asis
3024@item @emph{Description}:
3025@code{BTEST(I,POS)} returns logical @code{.TRUE.} if the bit at @var{POS}
3026in @var{I} is set.  The counting of the bits starts at 0.
3027
3028@item @emph{Standard}:
3029Fortran 90 and later, has overloads that are GNU extensions
3030
3031@item @emph{Class}:
3032Elemental function
3033
3034@item @emph{Syntax}:
3035@code{RESULT = BTEST(I, POS)}
3036
3037@item @emph{Arguments}:
3038@multitable @columnfractions .15 .70
3039@item @var{I} @tab The type shall be @code{INTEGER}.
3040@item @var{POS} @tab The type shall be @code{INTEGER}.
3041@end multitable
3042
3043@item @emph{Return value}:
3044The return value is of type @code{LOGICAL}
3045
3046@item @emph{Example}:
3047@smallexample
3048program test_btest
3049    integer :: i = 32768 + 1024 + 64
3050    integer :: pos
3051    logical :: bool
3052    do pos=0,16
3053        bool = btest(i, pos)
3054        print *, pos, bool
3055    end do
3056end program test_btest
3057@end smallexample
3058
3059@item @emph{Specific names}:
3060@multitable @columnfractions .20 .20 .20 .25
3061@item Name           @tab Argument         @tab Return type             @tab Standard
3062@item @code{BTEST(I,POS)} @tab @code{INTEGER I,POS} @tab @code{LOGICAL} @tab Fortran 95 and later
3063@item @code{BBTEST(I,POS)} @tab @code{INTEGER(1) I,POS} @tab @code{LOGICAL(1)} @tab GNU extension
3064@item @code{BITEST(I,POS)} @tab @code{INTEGER(2) I,POS} @tab @code{LOGICAL(2)} @tab GNU extension
3065@item @code{BJTEST(I,POS)} @tab @code{INTEGER(4) I,POS} @tab @code{LOGICAL(4)} @tab GNU extension
3066@item @code{BKTEST(I,POS)} @tab @code{INTEGER(8) I,POS} @tab @code{LOGICAL(8)} @tab GNU extension
3067@end multitable
3068@end table
3069
3070@node C_ASSOCIATED
3071@section @code{C_ASSOCIATED} --- Status of a C pointer
3072@fnindex C_ASSOCIATED
3073@cindex association status, C pointer
3074@cindex pointer, C association status
3075
3076@table @asis
3077@item @emph{Description}:
3078@code{C_ASSOCIATED(c_ptr_1[, c_ptr_2])} determines the status of the C pointer
3079@var{c_ptr_1} or if @var{c_ptr_1} is associated with the target @var{c_ptr_2}.
3080
3081@item @emph{Standard}:
3082Fortran 2003 and later
3083
3084@item @emph{Class}:
3085Inquiry function
3086
3087@item @emph{Syntax}:
3088@code{RESULT = C_ASSOCIATED(c_ptr_1[, c_ptr_2])}
3089
3090@item @emph{Arguments}:
3091@multitable @columnfractions .15 .70
3092@item @var{c_ptr_1} @tab Scalar of the type @code{C_PTR} or @code{C_FUNPTR}.
3093@item @var{c_ptr_2} @tab (Optional) Scalar of the same type as @var{c_ptr_1}.
3094@end multitable
3095
3096@item @emph{Return value}:
3097The return value is of type @code{LOGICAL}; it is @code{.false.} if either
3098@var{c_ptr_1} is a C NULL pointer or if @var{c_ptr1} and @var{c_ptr_2}
3099point to different addresses.
3100
3101@item @emph{Example}:
3102@smallexample
3103subroutine association_test(a,b)
3104  use iso_c_binding, only: c_associated, c_loc, c_ptr
3105  implicit none
3106  real, pointer :: a
3107  type(c_ptr) :: b
3108  if(c_associated(b, c_loc(a))) &
3109     stop 'b and a do not point to same target'
3110end subroutine association_test
3111@end smallexample
3112
3113@item @emph{See also}:
3114@ref{C_LOC}, @gol
3115@ref{C_FUNLOC}
3116@end table
3117
3118
3119@node C_F_POINTER
3120@section @code{C_F_POINTER} --- Convert C into Fortran pointer
3121@fnindex C_F_POINTER
3122@cindex pointer, convert C to Fortran
3123
3124@table @asis
3125@item @emph{Description}:
3126@code{C_F_POINTER(CPTR, FPTR[, SHAPE])} assigns the target of the C pointer
3127@var{CPTR} to the Fortran pointer @var{FPTR} and specifies its shape.
3128
3129@item @emph{Standard}:
3130Fortran 2003 and later
3131
3132@item @emph{Class}:
3133Subroutine
3134
3135@item @emph{Syntax}:
3136@code{CALL C_F_POINTER(CPTR, FPTR[, SHAPE])}
3137
3138@item @emph{Arguments}:
3139@multitable @columnfractions .15 .70
3140@item @var{CPTR}  @tab scalar of the type @code{C_PTR}. It is
3141@code{INTENT(IN)}.
3142@item @var{FPTR}  @tab pointer interoperable with @var{cptr}. It is
3143@code{INTENT(OUT)}.
3144@item @var{SHAPE} @tab (Optional) Rank-one array of type @code{INTEGER}
3145with @code{INTENT(IN)}. It shall be present
3146if and only if @var{fptr} is an array. The size
3147must be equal to the rank of @var{fptr}.
3148@end multitable
3149
3150@item @emph{Example}:
3151@smallexample
3152program main
3153  use iso_c_binding
3154  implicit none
3155  interface
3156    subroutine my_routine(p) bind(c,name='myC_func')
3157      import :: c_ptr
3158      type(c_ptr), intent(out) :: p
3159    end subroutine
3160  end interface
3161  type(c_ptr) :: cptr
3162  real,pointer :: a(:)
3163  call my_routine(cptr)
3164  call c_f_pointer(cptr, a, [12])
3165end program main
3166@end smallexample
3167
3168@item @emph{See also}:
3169@ref{C_LOC}, @gol
3170@ref{C_F_PROCPOINTER}
3171@end table
3172
3173
3174@node C_F_PROCPOINTER
3175@section @code{C_F_PROCPOINTER} --- Convert C into Fortran procedure pointer
3176@fnindex C_F_PROCPOINTER
3177@cindex pointer, C address of pointers
3178
3179@table @asis
3180@item @emph{Description}:
3181@code{C_F_PROCPOINTER(CPTR, FPTR)} Assign the target of the C function pointer
3182@var{CPTR} to the Fortran procedure pointer @var{FPTR}.
3183
3184@item @emph{Standard}:
3185Fortran 2003 and later
3186
3187@item @emph{Class}:
3188Subroutine
3189
3190@item @emph{Syntax}:
3191@code{CALL C_F_PROCPOINTER(cptr, fptr)}
3192
3193@item @emph{Arguments}:
3194@multitable @columnfractions .15 .70
3195@item @var{CPTR}  @tab scalar of the type @code{C_FUNPTR}. It is
3196@code{INTENT(IN)}.
3197@item @var{FPTR}  @tab procedure pointer interoperable with @var{cptr}. It is
3198@code{INTENT(OUT)}.
3199@end multitable
3200
3201@item @emph{Example}:
3202@smallexample
3203program main
3204  use iso_c_binding
3205  implicit none
3206  abstract interface
3207    function func(a)
3208      import :: c_float
3209      real(c_float), intent(in) :: a
3210      real(c_float) :: func
3211    end function
3212  end interface
3213  interface
3214     function getIterFunc() bind(c,name="getIterFunc")
3215       import :: c_funptr
3216       type(c_funptr) :: getIterFunc
3217     end function
3218  end interface
3219  type(c_funptr) :: cfunptr
3220  procedure(func), pointer :: myFunc
3221  cfunptr = getIterFunc()
3222  call c_f_procpointer(cfunptr, myFunc)
3223end program main
3224@end smallexample
3225
3226@item @emph{See also}:
3227@ref{C_LOC}, @gol
3228@ref{C_F_POINTER}
3229@end table
3230
3231
3232@node C_FUNLOC
3233@section @code{C_FUNLOC} --- Obtain the C address of a procedure
3234@fnindex C_FUNLOC
3235@cindex pointer, C address of procedures
3236
3237@table @asis
3238@item @emph{Description}:
3239@code{C_FUNLOC(x)} determines the C address of the argument.
3240
3241@item @emph{Standard}:
3242Fortran 2003 and later
3243
3244@item @emph{Class}:
3245Inquiry function
3246
3247@item @emph{Syntax}:
3248@code{RESULT = C_FUNLOC(x)}
3249
3250@item @emph{Arguments}:
3251@multitable @columnfractions .15 .70
3252@item @var{x} @tab Interoperable function or pointer to such function.
3253@end multitable
3254
3255@item @emph{Return value}:
3256The return value is of type @code{C_FUNPTR} and contains the C address
3257of the argument.
3258
3259@item @emph{Example}:
3260@smallexample
3261module x
3262  use iso_c_binding
3263  implicit none
3264contains
3265  subroutine sub(a) bind(c)
3266    real(c_float) :: a
3267    a = sqrt(a)+5.0
3268  end subroutine sub
3269end module x
3270program main
3271  use iso_c_binding
3272  use x
3273  implicit none
3274  interface
3275    subroutine my_routine(p) bind(c,name='myC_func')
3276      import :: c_funptr
3277      type(c_funptr), intent(in) :: p
3278    end subroutine
3279  end interface
3280  call my_routine(c_funloc(sub))
3281end program main
3282@end smallexample
3283
3284@item @emph{See also}:
3285@ref{C_ASSOCIATED}, @gol
3286@ref{C_LOC}, @gol
3287@ref{C_F_POINTER}, @gol
3288@ref{C_F_PROCPOINTER}
3289@end table
3290
3291
3292@node C_LOC
3293@section @code{C_LOC} --- Obtain the C address of an object
3294@fnindex C_LOC
3295@cindex procedure pointer, convert C to Fortran
3296
3297@table @asis
3298@item @emph{Description}:
3299@code{C_LOC(X)} determines the C address of the argument.
3300
3301@item @emph{Standard}:
3302Fortran 2003 and later
3303
3304@item @emph{Class}:
3305Inquiry function
3306
3307@item @emph{Syntax}:
3308@code{RESULT = C_LOC(X)}
3309
3310@item @emph{Arguments}:
3311@multitable @columnfractions .10 .75
3312@item @var{X} @tab  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 parameters, or be a scalar, nonpolymorphic variable with no length type parameters.
3313
3314@end multitable
3315
3316@item @emph{Return value}:
3317The return value is of type @code{C_PTR} and contains the C address
3318of the argument.
3319
3320@item @emph{Example}:
3321@smallexample
3322subroutine association_test(a,b)
3323  use iso_c_binding, only: c_associated, c_loc, c_ptr
3324  implicit none
3325  real, pointer :: a
3326  type(c_ptr) :: b
3327  if(c_associated(b, c_loc(a))) &
3328     stop 'b and a do not point to same target'
3329end subroutine association_test
3330@end smallexample
3331
3332@item @emph{See also}:
3333@ref{C_ASSOCIATED}, @gol
3334@ref{C_FUNLOC}, @gol
3335@ref{C_F_POINTER}, @gol
3336@ref{C_F_PROCPOINTER}
3337@end table
3338
3339
3340@node C_SIZEOF
3341@section @code{C_SIZEOF} --- Size in bytes of an expression
3342@fnindex C_SIZEOF
3343@cindex expression size
3344@cindex size of an expression
3345
3346@table @asis
3347@item @emph{Description}:
3348@code{C_SIZEOF(X)} calculates the number of bytes of storage the
3349expression @code{X} occupies.
3350
3351@item @emph{Standard}:
3352Fortran 2008
3353
3354@item @emph{Class}:
3355Inquiry function of the module @code{ISO_C_BINDING}
3356
3357@item @emph{Syntax}:
3358@code{N = C_SIZEOF(X)}
3359
3360@item @emph{Arguments}:
3361@multitable @columnfractions .15 .70
3362@item @var{X} @tab The argument shall be an interoperable data entity.
3363@end multitable
3364
3365@item @emph{Return value}:
3366The return value is of type integer and of the system-dependent kind
3367@code{C_SIZE_T} (from the @code{ISO_C_BINDING} module). Its value is the
3368number of bytes occupied by the argument.  If the argument has the
3369@code{POINTER} attribute, the number of bytes of the storage area pointed
3370to is returned.  If the argument is of a derived type with @code{POINTER}
3371or @code{ALLOCATABLE} components, the return value does not account for
3372the sizes of the data pointed to by these components.
3373
3374@item @emph{Example}:
3375@smallexample
3376   use iso_c_binding
3377   integer(c_int) :: i
3378   real(c_float) :: r, s(5)
3379   print *, (c_sizeof(s)/c_sizeof(r) == 5)
3380   end
3381@end smallexample
3382The example will print @code{T} unless you are using a platform
3383where default @code{REAL} variables are unusually padded.
3384
3385@item @emph{See also}:
3386@ref{SIZEOF}, @gol
3387@ref{STORAGE_SIZE}
3388@end table
3389
3390
3391@node CEILING
3392@section @code{CEILING} --- Integer ceiling function
3393@fnindex CEILING
3394@cindex ceiling
3395@cindex rounding, ceiling
3396
3397@table @asis
3398@item @emph{Description}:
3399@code{CEILING(A)} returns the least integer greater than or equal to @var{A}.
3400
3401@item @emph{Standard}:
3402Fortran 95 and later
3403
3404@item @emph{Class}:
3405Elemental function
3406
3407@item @emph{Syntax}:
3408@code{RESULT = CEILING(A [, KIND])}
3409
3410@item @emph{Arguments}:
3411@multitable @columnfractions .15 .70
3412@item @var{A} @tab The type shall be @code{REAL}.
3413@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
3414expression indicating the kind parameter of the result.
3415@end multitable
3416
3417@item @emph{Return value}:
3418The return value is of type @code{INTEGER(KIND)} if @var{KIND} is present
3419and a default-kind @code{INTEGER} otherwise.
3420
3421@item @emph{Example}:
3422@smallexample
3423program test_ceiling
3424    real :: x = 63.29
3425    real :: y = -63.59
3426    print *, ceiling(x) ! returns 64
3427    print *, ceiling(y) ! returns -63
3428end program test_ceiling
3429@end smallexample
3430
3431@item @emph{See also}:
3432@ref{FLOOR}, @gol
3433@ref{NINT}
3434@end table
3435
3436
3437
3438@node CHAR
3439@section @code{CHAR} --- Character conversion function
3440@fnindex CHAR
3441@cindex conversion, to character
3442
3443@table @asis
3444@item @emph{Description}:
3445@code{CHAR(I [, KIND])} returns the character represented by the integer @var{I}.
3446
3447@item @emph{Standard}:
3448Fortran 77 and later
3449
3450@item @emph{Class}:
3451Elemental function
3452
3453@item @emph{Syntax}:
3454@code{RESULT = CHAR(I [, KIND])}
3455
3456@item @emph{Arguments}:
3457@multitable @columnfractions .15 .70
3458@item @var{I} @tab The type shall be @code{INTEGER}.
3459@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
3460expression indicating the kind parameter of the result.
3461@end multitable
3462
3463@item @emph{Return value}:
3464The return value is of type @code{CHARACTER(1)}
3465
3466@item @emph{Example}:
3467@smallexample
3468program test_char
3469    integer :: i = 74
3470    character(1) :: c
3471    c = char(i)
3472    print *, i, c ! returns 'J'
3473end program test_char
3474@end smallexample
3475
3476@item @emph{Specific names}:
3477@multitable @columnfractions .18 .18 .24 .25
3478@item Name           @tab Argument         @tab Return type             @tab Standard
3479@item @code{CHAR(I)} @tab @code{INTEGER I} @tab @code{CHARACTER(LEN=1)} @tab Fortran 77 and later
3480@end multitable
3481
3482@item @emph{Note}:
3483See @ref{ICHAR} for a discussion of converting between numerical values
3484and formatted string representations.
3485
3486@item @emph{See also}:
3487@ref{ACHAR}, @gol
3488@ref{IACHAR}, @gol
3489@ref{ICHAR}
3490
3491@end table
3492
3493
3494
3495@node CHDIR
3496@section @code{CHDIR} --- Change working directory
3497@fnindex CHDIR
3498@cindex system, working directory
3499
3500@table @asis
3501@item @emph{Description}:
3502Change current working directory to a specified path.
3503
3504This intrinsic is provided in both subroutine and function forms; however,
3505only one form can be used in any given program unit.
3506
3507@item @emph{Standard}:
3508GNU extension
3509
3510@item @emph{Class}:
3511Subroutine, function
3512
3513@item @emph{Syntax}:
3514@multitable @columnfractions .80
3515@item @code{CALL CHDIR(NAME [, STATUS])}
3516@item @code{STATUS = CHDIR(NAME)}
3517@end multitable
3518
3519@item @emph{Arguments}:
3520@multitable @columnfractions .15 .70
3521@item @var{NAME}   @tab The type shall be @code{CHARACTER} of default
3522kind and shall specify a valid path within the file system.
3523@item @var{STATUS} @tab (Optional) @code{INTEGER} status flag of the default
3524kind.  Returns 0 on success, and a system specific and nonzero error code
3525otherwise.
3526@end multitable
3527
3528@item @emph{Example}:
3529@smallexample
3530PROGRAM test_chdir
3531  CHARACTER(len=255) :: path
3532  CALL getcwd(path)
3533  WRITE(*,*) TRIM(path)
3534  CALL chdir("/tmp")
3535  CALL getcwd(path)
3536  WRITE(*,*) TRIM(path)
3537END PROGRAM
3538@end smallexample
3539
3540@item @emph{See also}:
3541@ref{GETCWD}
3542@end table
3543
3544
3545
3546@node CHMOD
3547@section @code{CHMOD} --- Change access permissions of files
3548@fnindex CHMOD
3549@cindex file system, change access mode
3550
3551@table @asis
3552@item @emph{Description}:
3553@code{CHMOD} changes the permissions of a file.
3554
3555This intrinsic is provided in both subroutine and function forms; however,
3556only one form can be used in any given program unit.
3557
3558@item @emph{Standard}:
3559GNU extension
3560
3561@item @emph{Class}:
3562Subroutine, function
3563
3564@item @emph{Syntax}:
3565@multitable @columnfractions .80
3566@item @code{CALL CHMOD(NAME, MODE[, STATUS])}
3567@item @code{STATUS = CHMOD(NAME, MODE)}
3568@end multitable
3569
3570@item @emph{Arguments}:
3571@multitable @columnfractions .15 .70
3572
3573@item @var{NAME} @tab Scalar @code{CHARACTER} of default kind with the
3574file name. Trailing blanks are ignored unless the character
3575@code{achar(0)} is present, then all characters up to and excluding
3576@code{achar(0)} are used as the file name.
3577
3578@item @var{MODE} @tab Scalar @code{CHARACTER} of default kind giving the
3579file permission. @var{MODE} uses the same syntax as the @code{chmod} utility
3580as defined by the POSIX standard. The argument shall either be a string of
3581a nonnegative octal number or a symbolic mode.
3582
3583@item @var{STATUS} @tab (optional) scalar @code{INTEGER}, which is
3584@code{0} on success and nonzero otherwise.
3585@end multitable
3586
3587@item @emph{Return value}:
3588In either syntax, @var{STATUS} is set to @code{0} on success and nonzero
3589otherwise.
3590
3591@item @emph{Example}:
3592@code{CHMOD} as subroutine
3593@smallexample
3594program chmod_test
3595  implicit none
3596  integer :: status
3597  call chmod('test.dat','u+x',status)
3598  print *, 'Status: ', status
3599end program chmod_test
3600@end smallexample
3601@code{CHMOD} as function:
3602@smallexample
3603program chmod_test
3604  implicit none
3605  integer :: status
3606  status = chmod('test.dat','u+x')
3607  print *, 'Status: ', status
3608end program chmod_test
3609@end smallexample
3610
3611@end table
3612
3613
3614
3615@node CMPLX
3616@section @code{CMPLX} --- Complex conversion function
3617@fnindex CMPLX
3618@cindex complex numbers, conversion to
3619@cindex conversion, to complex
3620
3621@table @asis
3622@item @emph{Description}:
3623@code{CMPLX(X [, Y [, KIND]])} returns a complex number where @var{X} is converted to
3624the real component.  If @var{Y} is present it is converted to the imaginary
3625component.  If @var{Y} is not present then the imaginary component is set to
36260.0.  If @var{X} is complex then @var{Y} must not be present.
3627
3628@item @emph{Standard}:
3629Fortran 77 and later
3630
3631@item @emph{Class}:
3632Elemental function
3633
3634@item @emph{Syntax}:
3635@code{RESULT = CMPLX(X [, Y [, KIND]])}
3636
3637@item @emph{Arguments}:
3638@multitable @columnfractions .15 .70
3639@item @var{X} @tab The type may be @code{INTEGER}, @code{REAL},
3640or @code{COMPLEX}.
3641@item @var{Y} @tab (Optional; only allowed if @var{X} is not
3642@code{COMPLEX}.)  May be @code{INTEGER} or @code{REAL}.
3643@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
3644expression indicating the kind parameter of the result.
3645@end multitable
3646
3647@item @emph{Return value}:
3648The return value is of @code{COMPLEX} type, with a kind equal to
3649@var{KIND} if it is specified.  If @var{KIND} is not specified, the
3650result is of the default @code{COMPLEX} kind, regardless of the kinds of
3651@var{X} and @var{Y}.
3652
3653@item @emph{Example}:
3654@smallexample
3655program test_cmplx
3656    integer :: i = 42
3657    real :: x = 3.14
3658    complex :: z
3659    z = cmplx(i, x)
3660    print *, z, cmplx(x)
3661end program test_cmplx
3662@end smallexample
3663
3664@item @emph{See also}:
3665@ref{COMPLEX}
3666@end table
3667
3668
3669
3670@node CO_BROADCAST
3671@section @code{CO_BROADCAST} --- Copy a value to all images the current set of images
3672@fnindex CO_BROADCAST
3673@cindex Collectives, value broadcasting
3674
3675@table @asis
3676@item @emph{Description}:
3677@code{CO_BROADCAST} copies the value of argument @var{A} on the image with
3678image index @code{SOURCE_IMAGE} to all images in the current team.  @var{A}
3679becomes defined as if by intrinsic assignment.  If the execution was
3680successful and @var{STAT} is present, it is assigned the value zero.  If the
3681execution failed, @var{STAT} gets assigned a nonzero value and, if present,
3682@var{ERRMSG} gets assigned a value describing the occurred error.
3683
3684@item @emph{Standard}:
3685Technical Specification (TS) 18508 or later
3686
3687@item @emph{Class}:
3688Collective subroutine
3689
3690@item @emph{Syntax}:
3691@code{CALL CO_BROADCAST(A, SOURCE_IMAGE [, STAT, ERRMSG])}
3692
3693@item @emph{Arguments}:
3694@multitable @columnfractions .20 .65
3695@item @var{A}            @tab INTENT(INOUT) argument; shall have the same
3696dynamic type and type parameters on all images of the current team. If it
3697is an array, it shall have the same shape on all images.
3698@item @var{SOURCE_IMAGE} @tab a scalar integer expression.
3699It shall have the same the same value on all images and refer to an
3700image of the current team.
3701@item @var{STAT}         @tab (optional) a scalar integer variable
3702@item @var{ERRMSG}       @tab (optional) a scalar character variable
3703@end multitable
3704
3705@item @emph{Example}:
3706@smallexample
3707program test
3708  integer :: val(3)
3709  if (this_image() == 1) then
3710    val = [1, 5, 3]
3711  end if
3712  call co_broadcast (val, source_image=1)
3713  print *, this_image, ":", val
3714end program test
3715@end smallexample
3716
3717@item @emph{See also}:
3718@ref{CO_MAX}, @gol
3719@ref{CO_MIN}, @gol
3720@ref{CO_SUM}, @gol
3721@ref{CO_REDUCE}
3722@end table
3723
3724
3725
3726@node CO_MAX
3727@section @code{CO_MAX} --- Maximal value on the current set of images
3728@fnindex CO_MAX
3729@cindex Collectives, maximal value
3730
3731@table @asis
3732@item @emph{Description}:
3733@code{CO_MAX} determines element-wise the maximal value of @var{A} on all
3734images of the current team.  If @var{RESULT_IMAGE} is present, the maximum
3735values are returned in @var{A} on the specified image only and the value
3736of @var{A} on the other images become undefined.  If @var{RESULT_IMAGE} is
3737not present, the value is returned on all images.  If the execution was
3738successful and @var{STAT} is present, it is assigned the value zero.  If the
3739execution failed, @var{STAT} gets assigned a nonzero value and, if present,
3740@var{ERRMSG} gets assigned a value describing the occurred error.
3741
3742@item @emph{Standard}:
3743Technical Specification (TS) 18508 or later
3744
3745@item @emph{Class}:
3746Collective subroutine
3747
3748@item @emph{Syntax}:
3749@code{CALL CO_MAX(A [, RESULT_IMAGE, STAT, ERRMSG])}
3750
3751@item @emph{Arguments}:
3752@multitable @columnfractions .20 .65
3753@item @var{A}            @tab shall be an integer, real or character variable,
3754which has the same type and type parameters on all images of the team.
3755@item @var{RESULT_IMAGE} @tab (optional) a scalar integer expression; if
3756present, it shall have the same the same value on all images and refer to an
3757image of the current team.
3758@item @var{STAT}         @tab (optional) a scalar integer variable
3759@item @var{ERRMSG}       @tab (optional) a scalar character variable
3760@end multitable
3761
3762@item @emph{Example}:
3763@smallexample
3764program test
3765  integer :: val
3766  val = this_image ()
3767  call co_max (val, result_image=1)
3768  if (this_image() == 1) then
3769    write(*,*) "Maximal value", val  ! prints num_images()
3770  end if
3771end program test
3772@end smallexample
3773
3774@item @emph{See also}:
3775@ref{CO_MIN}, @gol
3776@ref{CO_SUM}, @gol
3777@ref{CO_REDUCE}, @gol
3778@ref{CO_BROADCAST}
3779@end table
3780
3781
3782
3783@node CO_MIN
3784@section @code{CO_MIN} --- Minimal value on the current set of images
3785@fnindex CO_MIN
3786@cindex Collectives, minimal value
3787
3788@table @asis
3789@item @emph{Description}:
3790@code{CO_MIN} determines element-wise the minimal value of @var{A} on all
3791images of the current team.  If @var{RESULT_IMAGE} is present, the minimal
3792values are returned in @var{A} on the specified image only and the value
3793of @var{A} on the other images become undefined.  If @var{RESULT_IMAGE} is
3794not present, the value is returned on all images.  If the execution was
3795successful and @var{STAT} is present, it is assigned the value zero.  If the
3796execution failed, @var{STAT} gets assigned a nonzero value and, if present,
3797@var{ERRMSG} gets assigned a value describing the occurred error.
3798
3799@item @emph{Standard}:
3800Technical Specification (TS) 18508 or later
3801
3802@item @emph{Class}:
3803Collective subroutine
3804
3805@item @emph{Syntax}:
3806@code{CALL CO_MIN(A [, RESULT_IMAGE, STAT, ERRMSG])}
3807
3808@item @emph{Arguments}:
3809@multitable @columnfractions .20 .65
3810@item @var{A}            @tab shall be an integer, real or character variable,
3811which has the same type and type parameters on all images of the team.
3812@item @var{RESULT_IMAGE} @tab (optional) a scalar integer expression; if
3813present, it shall have the same the same value on all images and refer to an
3814image of the current team.
3815@item @var{STAT}         @tab (optional) a scalar integer variable
3816@item @var{ERRMSG}       @tab (optional) a scalar character variable
3817@end multitable
3818
3819@item @emph{Example}:
3820@smallexample
3821program test
3822  integer :: val
3823  val = this_image ()
3824  call co_min (val, result_image=1)
3825  if (this_image() == 1) then
3826    write(*,*) "Minimal value", val  ! prints 1
3827  end if
3828end program test
3829@end smallexample
3830
3831@item @emph{See also}:
3832@ref{CO_MAX}, @gol
3833@ref{CO_SUM}, @gol
3834@ref{CO_REDUCE}, @gol
3835@ref{CO_BROADCAST}
3836@end table
3837
3838
3839
3840@node CO_REDUCE
3841@section @code{CO_REDUCE} --- Reduction of values on the current set of images
3842@fnindex CO_REDUCE
3843@cindex Collectives, generic reduction
3844
3845@table @asis
3846@item @emph{Description}:
3847@code{CO_REDUCE} determines element-wise the reduction of the value of @var{A}
3848on all images of the current team.  The pure function passed as @var{OPERATOR}
3849is used to pairwise reduce the values of @var{A} by passing either the value
3850of @var{A} of different images or the result values of such a reduction as
3851argument.  If @var{A} is an array, the deduction is done element wise. If
3852@var{RESULT_IMAGE} is present, the result values are returned in @var{A} on
3853the specified image only and the value of @var{A} on the other images become
3854undefined.  If @var{RESULT_IMAGE} is not present, the value is returned on all
3855images.  If the execution was successful and @var{STAT} is present, it is
3856assigned the value zero.  If the execution failed, @var{STAT} gets assigned
3857a nonzero value and, if present, @var{ERRMSG} gets assigned a value describing
3858the occurred error.
3859
3860@item @emph{Standard}:
3861Technical Specification (TS) 18508 or later
3862
3863@item @emph{Class}:
3864Collective subroutine
3865
3866@item @emph{Syntax}:
3867@code{CALL CO_REDUCE(A, OPERATOR, [, RESULT_IMAGE, STAT, ERRMSG])}
3868
3869@item @emph{Arguments}:
3870@multitable @columnfractions .20 .65
3871@item @var{A}            @tab is an @code{INTENT(INOUT)} argument and shall be
3872nonpolymorphic. If it is allocatable, it shall be allocated; if it is a pointer,
3873it shall be associated.  @var{A} shall have the same type and type parameters on
3874all images of the team; if it is an array, it shall have the same shape on all
3875images.
3876@item @var{OPERATOR}     @tab pure function with two scalar nonallocatable
3877arguments, which shall be nonpolymorphic and have the same type and type
3878parameters as @var{A}.  The function shall return a nonallocatable scalar of
3879the same type and type parameters as @var{A}.  The function shall be the same on
3880all images and with regards to the arguments mathematically commutative and
3881associative.  Note that @var{OPERATOR} may not be an elemental function, unless
3882it is an intrisic function.
3883@item @var{RESULT_IMAGE} @tab (optional) a scalar integer expression; if
3884present, it shall have the same the same value on all images and refer to an
3885image of the current team.
3886@item @var{STAT}         @tab (optional) a scalar integer variable
3887@item @var{ERRMSG}       @tab (optional) a scalar character variable
3888@end multitable
3889
3890@item @emph{Example}:
3891@smallexample
3892program test
3893  integer :: val
3894  val = this_image ()
3895  call co_reduce (val, result_image=1, operator=myprod)
3896  if (this_image() == 1) then
3897    write(*,*) "Product value", val  ! prints num_images() factorial
3898  end if
3899contains
3900  pure function myprod(a, b)
3901    integer, value :: a, b
3902    integer :: myprod
3903    myprod = a * b
3904  end function myprod
3905end program test
3906@end smallexample
3907
3908@item @emph{Note}:
3909While the rules permit in principle an intrinsic function, none of the
3910intrinsics in the standard fulfill the criteria of having a specific
3911function, which takes two arguments of the same type and returning that
3912type as result.
3913
3914@item @emph{See also}:
3915@ref{CO_MIN}, @gol
3916@ref{CO_MAX}, @gol
3917@ref{CO_SUM}, @gol
3918@ref{CO_BROADCAST}
3919@end table
3920
3921
3922
3923@node CO_SUM
3924@section @code{CO_SUM} --- Sum of values on the current set of images
3925@fnindex CO_SUM
3926@cindex Collectives, sum of values
3927
3928@table @asis
3929@item @emph{Description}:
3930@code{CO_SUM} sums up the values of each element of @var{A} on all
3931images of the current team.  If @var{RESULT_IMAGE} is present, the summed-up
3932values are returned in @var{A} on the specified image only and the value
3933of @var{A} on the other images become undefined.  If @var{RESULT_IMAGE} is
3934not present, the value is returned on all images.  If the execution was
3935successful and @var{STAT} is present, it is assigned the value zero.  If the
3936execution failed, @var{STAT} gets assigned a nonzero value and, if present,
3937@var{ERRMSG} gets assigned a value describing the occurred error.
3938
3939@item @emph{Standard}:
3940Technical Specification (TS) 18508 or later
3941
3942@item @emph{Class}:
3943Collective subroutine
3944
3945@item @emph{Syntax}:
3946@code{CALL CO_MIN(A [, RESULT_IMAGE, STAT, ERRMSG])}
3947
3948@item @emph{Arguments}:
3949@multitable @columnfractions .20 .65
3950@item @var{A}            @tab shall be an integer, real or complex variable,
3951which has the same type and type parameters on all images of the team.
3952@item @var{RESULT_IMAGE} @tab (optional) a scalar integer expression; if
3953present, it shall have the same the same value on all images and refer to an
3954image of the current team.
3955@item @var{STAT}         @tab (optional) a scalar integer variable
3956@item @var{ERRMSG}       @tab (optional) a scalar character variable
3957@end multitable
3958
3959@item @emph{Example}:
3960@smallexample
3961program test
3962  integer :: val
3963  val = this_image ()
3964  call co_sum (val, result_image=1)
3965  if (this_image() == 1) then
3966    write(*,*) "The sum is ", val ! prints (n**2 + n)/2,
3967                                  ! with n = num_images()
3968  end if
3969end program test
3970@end smallexample
3971
3972@item @emph{See also}:
3973@ref{CO_MAX}, @gol
3974@ref{CO_MIN}, @gol
3975@ref{CO_REDUCE}, @gol
3976@ref{CO_BROADCAST}
3977@end table
3978
3979
3980
3981@node COMMAND_ARGUMENT_COUNT
3982@section @code{COMMAND_ARGUMENT_COUNT} --- Get number of command line arguments
3983@fnindex COMMAND_ARGUMENT_COUNT
3984@cindex command-line arguments
3985@cindex command-line arguments, number of
3986@cindex arguments, to program
3987
3988@table @asis
3989@item @emph{Description}:
3990@code{COMMAND_ARGUMENT_COUNT} returns the number of arguments passed on the
3991command line when the containing program was invoked.
3992
3993@item @emph{Standard}:
3994Fortran 2003 and later
3995
3996@item @emph{Class}:
3997Inquiry function
3998
3999@item @emph{Syntax}:
4000@code{RESULT = COMMAND_ARGUMENT_COUNT()}
4001
4002@item @emph{Arguments}:
4003@multitable @columnfractions .15 .70
4004@item None
4005@end multitable
4006
4007@item @emph{Return value}:
4008The return value is an @code{INTEGER} of default kind.
4009
4010@item @emph{Example}:
4011@smallexample
4012program test_command_argument_count
4013    integer :: count
4014    count = command_argument_count()
4015    print *, count
4016end program test_command_argument_count
4017@end smallexample
4018
4019@item @emph{See also}:
4020@ref{GET_COMMAND}, @gol
4021@ref{GET_COMMAND_ARGUMENT}
4022@end table
4023
4024
4025
4026@node COMPILER_OPTIONS
4027@section @code{COMPILER_OPTIONS} --- Options passed to the compiler
4028@fnindex COMPILER_OPTIONS
4029@cindex flags inquiry function
4030@cindex options inquiry function
4031@cindex compiler flags inquiry function
4032
4033@table @asis
4034@item @emph{Description}:
4035@code{COMPILER_OPTIONS} returns a string with the options used for
4036compiling.
4037
4038@item @emph{Standard}:
4039Fortran 2008
4040
4041@item @emph{Class}:
4042Inquiry function of the module @code{ISO_FORTRAN_ENV}
4043
4044@item @emph{Syntax}:
4045@code{STR = COMPILER_OPTIONS()}
4046
4047@item @emph{Arguments}:
4048None
4049
4050@item @emph{Return value}:
4051The return value is a default-kind string with system-dependent length.
4052It contains the compiler flags used to compile the file, which called
4053the @code{COMPILER_OPTIONS} intrinsic.
4054
4055@item @emph{Example}:
4056@smallexample
4057   use iso_fortran_env
4058   print '(4a)', 'This file was compiled by ', &
4059                 compiler_version(), ' using the options ', &
4060                 compiler_options()
4061   end
4062@end smallexample
4063
4064@item @emph{See also}:
4065@ref{COMPILER_VERSION}, @gol
4066@ref{ISO_FORTRAN_ENV}
4067@end table
4068
4069
4070
4071@node COMPILER_VERSION
4072@section @code{COMPILER_VERSION} --- Compiler version string
4073@fnindex COMPILER_VERSION
4074@cindex compiler, name and version
4075@cindex version of the compiler
4076
4077@table @asis
4078@item @emph{Description}:
4079@code{COMPILER_VERSION} returns a string with the name and the
4080version of the compiler.
4081
4082@item @emph{Standard}:
4083Fortran 2008
4084
4085@item @emph{Class}:
4086Inquiry function of the module @code{ISO_FORTRAN_ENV}
4087
4088@item @emph{Syntax}:
4089@code{STR = COMPILER_VERSION()}
4090
4091@item @emph{Arguments}:
4092None
4093
4094@item @emph{Return value}:
4095The return value is a default-kind string with system-dependent length.
4096It contains the name of the compiler and its version number.
4097
4098@item @emph{Example}:
4099@smallexample
4100   use iso_fortran_env
4101   print '(4a)', 'This file was compiled by ', &
4102                 compiler_version(), ' using the options ', &
4103                 compiler_options()
4104   end
4105@end smallexample
4106
4107@item @emph{See also}:
4108@ref{COMPILER_OPTIONS}, @gol
4109@ref{ISO_FORTRAN_ENV}
4110@end table
4111
4112
4113
4114@node COMPLEX
4115@section @code{COMPLEX} --- Complex conversion function
4116@fnindex COMPLEX
4117@cindex complex numbers, conversion to
4118@cindex conversion, to complex
4119
4120@table @asis
4121@item @emph{Description}:
4122@code{COMPLEX(X, Y)} returns a complex number where @var{X} is converted
4123to the real component and @var{Y} is converted to the imaginary
4124component.
4125
4126@item @emph{Standard}:
4127GNU extension
4128
4129@item @emph{Class}:
4130Elemental function
4131
4132@item @emph{Syntax}:
4133@code{RESULT = COMPLEX(X, Y)}
4134
4135@item @emph{Arguments}:
4136@multitable @columnfractions .15 .70
4137@item @var{X} @tab The type may be @code{INTEGER} or @code{REAL}.
4138@item @var{Y} @tab The type may be @code{INTEGER} or @code{REAL}.
4139@end multitable
4140
4141@item @emph{Return value}:
4142If @var{X} and @var{Y} are both of @code{INTEGER} type, then the return
4143value is of default @code{COMPLEX} type.
4144
4145If @var{X} and @var{Y} are of @code{REAL} type, or one is of @code{REAL}
4146type and one is of @code{INTEGER} type, then the return value is of
4147@code{COMPLEX} type with a kind equal to that of the @code{REAL}
4148argument with the highest precision.
4149
4150@item @emph{Example}:
4151@smallexample
4152program test_complex
4153    integer :: i = 42
4154    real :: x = 3.14
4155    print *, complex(i, x)
4156end program test_complex
4157@end smallexample
4158
4159@item @emph{See also}:
4160@ref{CMPLX}
4161@end table
4162
4163
4164
4165@node CONJG
4166@section @code{CONJG} --- Complex conjugate function
4167@fnindex CONJG
4168@fnindex DCONJG
4169@cindex complex conjugate
4170
4171@table @asis
4172@item @emph{Description}:
4173@code{CONJG(Z)} returns the conjugate of @var{Z}.  If @var{Z} is @code{(x, y)}
4174then the result is @code{(x, -y)}
4175
4176@item @emph{Standard}:
4177Fortran 77 and later, has an overload that is a GNU extension
4178
4179@item @emph{Class}:
4180Elemental function
4181
4182@item @emph{Syntax}:
4183@code{Z = CONJG(Z)}
4184
4185@item @emph{Arguments}:
4186@multitable @columnfractions .15 .70
4187@item @var{Z} @tab The type shall be @code{COMPLEX}.
4188@end multitable
4189
4190@item @emph{Return value}:
4191The return value is of type @code{COMPLEX}.
4192
4193@item @emph{Example}:
4194@smallexample
4195program test_conjg
4196    complex :: z = (2.0, 3.0)
4197    complex(8) :: dz = (2.71_8, -3.14_8)
4198    z= conjg(z)
4199    print *, z
4200    dz = dconjg(dz)
4201    print *, dz
4202end program test_conjg
4203@end smallexample
4204
4205@item @emph{Specific names}:
4206@multitable @columnfractions .20 .20 .20 .25
4207@item Name             @tab Argument             @tab Return type       @tab Standard
4208@item @code{DCONJG(Z)} @tab @code{COMPLEX(8) Z}  @tab @code{COMPLEX(8)} @tab GNU extension
4209@end multitable
4210@end table
4211
4212
4213
4214@node COS
4215@section @code{COS} --- Cosine function
4216@fnindex COS
4217@fnindex DCOS
4218@fnindex CCOS
4219@fnindex ZCOS
4220@fnindex CDCOS
4221@cindex trigonometric function, cosine
4222@cindex cosine
4223
4224@table @asis
4225@item @emph{Description}:
4226@code{COS(X)} computes the cosine of @var{X}.
4227
4228@item @emph{Standard}:
4229Fortran 77 and later, has overloads that are GNU extensions
4230
4231@item @emph{Class}:
4232Elemental function
4233
4234@item @emph{Syntax}:
4235@code{RESULT = COS(X)}
4236
4237@item @emph{Arguments}:
4238@multitable @columnfractions .15 .70
4239@item @var{X} @tab The type shall be @code{REAL} or
4240@code{COMPLEX}.
4241@end multitable
4242
4243@item @emph{Return value}:
4244The return value is of the same type and kind as @var{X}. The real part
4245of the result is in radians. If @var{X} is of the type @code{REAL},
4246the return value lies in the range @math{ -1 \leq \cos (x) \leq 1}.
4247
4248@item @emph{Example}:
4249@smallexample
4250program test_cos
4251  real :: x = 0.0
4252  x = cos(x)
4253end program test_cos
4254@end smallexample
4255
4256@item @emph{Specific names}:
4257@multitable @columnfractions .20 .20 .20 .25
4258@item Name            @tab Argument            @tab Return type       @tab Standard
4259@item @code{COS(X)}   @tab @code{REAL(4) X}    @tab @code{REAL(4)}    @tab Fortran 77 and later
4260@item @code{DCOS(X)}  @tab @code{REAL(8) X}    @tab @code{REAL(8)}    @tab Fortran 77 and later
4261@item @code{CCOS(X)}  @tab @code{COMPLEX(4) X} @tab @code{COMPLEX(4)} @tab Fortran 77 and later
4262@item @code{ZCOS(X)}  @tab @code{COMPLEX(8) X} @tab @code{COMPLEX(8)} @tab GNU extension
4263@item @code{CDCOS(X)} @tab @code{COMPLEX(8) X} @tab @code{COMPLEX(8)} @tab GNU extension
4264@end multitable
4265
4266@item @emph{See also}:
4267Inverse function: @gol
4268@ref{ACOS} @gol
4269Degrees function: @gol
4270@ref{COSD}
4271@end table
4272
4273
4274
4275@node COSD
4276@section @code{COSD} --- Cosine function, degrees
4277@fnindex COSD
4278@fnindex DCOSD
4279@fnindex CCOSD
4280@fnindex ZCOSD
4281@fnindex CDCOSD
4282@cindex trigonometric function, cosine, degrees
4283@cindex cosine, degrees
4284
4285@table @asis
4286@item @emph{Description}:
4287@code{COSD(X)} computes the cosine of @var{X} in degrees.
4288
4289This function is for compatibility only and should be avoided in favor of
4290standard constructs wherever possible.
4291
4292@item @emph{Standard}:
4293GNU extension, enabled with @option{-fdec-math}.
4294
4295@item @emph{Class}:
4296Elemental function
4297
4298@item @emph{Syntax}:
4299@code{RESULT = COSD(X)}
4300
4301@item @emph{Arguments}:
4302@multitable @columnfractions .15 .70
4303@item @var{X} @tab The type shall be @code{REAL} or
4304@code{COMPLEX}.
4305@end multitable
4306
4307@item @emph{Return value}:
4308The return value is of the same type and kind as @var{X}. The real part
4309of the result is in degrees.  If @var{X} is of the type @code{REAL},
4310the return value lies in the range @math{ -1 \leq \cosd (x) \leq 1}.
4311
4312@item @emph{Example}:
4313@smallexample
4314program test_cosd
4315  real :: x = 0.0
4316  x = cosd(x)
4317end program test_cosd
4318@end smallexample
4319
4320@item @emph{Specific names}:
4321@multitable @columnfractions .20 .20 .20 .25
4322@item Name            @tab Argument            @tab Return type       @tab Standard
4323@item @code{COSD(X)}   @tab @code{REAL(4) X}    @tab @code{REAL(4)}    @tab GNU extension
4324@item @code{DCOSD(X)}  @tab @code{REAL(8) X}    @tab @code{REAL(8)}    @tab GNU extension
4325@item @code{CCOSD(X)}  @tab @code{COMPLEX(4) X} @tab @code{COMPLEX(4)} @tab GNU extension
4326@item @code{ZCOSD(X)}  @tab @code{COMPLEX(8) X} @tab @code{COMPLEX(8)} @tab GNU extension
4327@item @code{CDCOSD(X)} @tab @code{COMPLEX(8) X} @tab @code{COMPLEX(8)} @tab GNU extension
4328@end multitable
4329
4330@item @emph{See also}:
4331Inverse function: @gol
4332@ref{ACOSD} @gol
4333Radians function: @gol
4334@ref{COS}
4335@end table
4336
4337
4338
4339@node COSH
4340@section @code{COSH} --- Hyperbolic cosine function
4341@fnindex COSH
4342@fnindex DCOSH
4343@cindex hyperbolic cosine
4344@cindex hyperbolic function, cosine
4345@cindex cosine, hyperbolic
4346
4347@table @asis
4348@item @emph{Description}:
4349@code{COSH(X)} computes the hyperbolic cosine of @var{X}.
4350
4351@item @emph{Standard}:
4352Fortran 77 and later, for a complex argument Fortran 2008 or later
4353
4354@item @emph{Class}:
4355Elemental function
4356
4357@item @emph{Syntax}:
4358@code{X = COSH(X)}
4359
4360@item @emph{Arguments}:
4361@multitable @columnfractions .15 .70
4362@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
4363@end multitable
4364
4365@item @emph{Return value}:
4366The return value has same type and kind as @var{X}. If @var{X} is
4367complex, the imaginary part of the result is in radians. If @var{X}
4368is @code{REAL}, the return value has a lower bound of one,
4369@math{\cosh (x) \geq 1}.
4370
4371@item @emph{Example}:
4372@smallexample
4373program test_cosh
4374  real(8) :: x = 1.0_8
4375  x = cosh(x)
4376end program test_cosh
4377@end smallexample
4378
4379@item @emph{Specific names}:
4380@multitable @columnfractions .20 .20 .20 .25
4381@item Name            @tab Argument          @tab Return type       @tab Standard
4382@item @code{COSH(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab Fortran 77 and later
4383@item @code{DCOSH(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab Fortran 77 and later
4384@end multitable
4385
4386@item @emph{See also}:
4387Inverse function: @gol
4388@ref{ACOSH}
4389@end table
4390
4391
4392
4393@node COTAN
4394@section @code{COTAN} --- Cotangent function
4395@fnindex COTAN
4396@fnindex DCOTAN
4397@cindex trigonometric function, cotangent
4398@cindex cotangent
4399
4400@table @asis
4401@item @emph{Description}:
4402@code{COTAN(X)} computes the cotangent of @var{X}. Equivalent to @code{COS(x)}
4403divided by @code{SIN(x)}, or @code{1 / TAN(x)}.
4404
4405This function is for compatibility only and should be avoided in favor of
4406standard constructs wherever possible.
4407
4408@item @emph{Standard}:
4409GNU extension, enabled with @option{-fdec-math}.
4410
4411@item @emph{Class}:
4412Elemental function
4413
4414@item @emph{Syntax}:
4415@code{RESULT = COTAN(X)}
4416
4417@item @emph{Arguments}:
4418@multitable @columnfractions .15 .70
4419@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
4420@end multitable
4421
4422@item @emph{Return value}:
4423The return value has same type and kind as @var{X}, and its value is in radians.
4424
4425@item @emph{Example}:
4426@smallexample
4427program test_cotan
4428  real(8) :: x = 0.165_8
4429  x = cotan(x)
4430end program test_cotan
4431@end smallexample
4432
4433@item @emph{Specific names}:
4434@multitable @columnfractions .20 .20 .20 .25
4435@item Name            @tab Argument          @tab Return type     @tab Standard
4436@item @code{COTAN(X)}   @tab @code{REAL(4) X}  @tab @code{REAL(4)}  @tab GNU extension
4437@item @code{DCOTAN(X)}  @tab @code{REAL(8) X}  @tab @code{REAL(8)}  @tab GNU extension
4438@end multitable
4439
4440@item @emph{See also}:
4441Converse function: @gol
4442@ref{TAN} @gol
4443Degrees function: @gol
4444@ref{COTAND}
4445@end table
4446
4447
4448
4449@node COTAND
4450@section @code{COTAND} --- Cotangent function, degrees
4451@fnindex COTAND
4452@fnindex DCOTAND
4453@cindex trigonometric function, cotangent, degrees
4454@cindex cotangent, degrees
4455
4456@table @asis
4457@item @emph{Description}:
4458@code{COTAND(X)} computes the cotangent of @var{X} in degrees.  Equivalent to
4459@code{COSD(x)} divided by @code{SIND(x)}, or @code{1 / TAND(x)}.
4460
4461@item @emph{Standard}:
4462GNU extension, enabled with @option{-fdec-math}.
4463
4464This function is for compatibility only and should be avoided in favor of
4465standard constructs wherever possible.
4466
4467@item @emph{Class}:
4468Elemental function
4469
4470@item @emph{Syntax}:
4471@code{RESULT = COTAND(X)}
4472
4473@item @emph{Arguments}:
4474@multitable @columnfractions .15 .70
4475@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
4476@end multitable
4477
4478@item @emph{Return value}:
4479The return value has same type and kind as @var{X}, and its value is in degrees.
4480
4481@item @emph{Example}:
4482@smallexample
4483program test_cotand
4484  real(8) :: x = 0.165_8
4485  x = cotand(x)
4486end program test_cotand
4487@end smallexample
4488
4489@item @emph{Specific names}:
4490@multitable @columnfractions .20 .20 .20 .25
4491@item Name            @tab Argument          @tab Return type     @tab Standard
4492@item @code{COTAND(X)}   @tab @code{REAL(4) X}  @tab @code{REAL(4)}  @tab GNU extension
4493@item @code{DCOTAND(X)}  @tab @code{REAL(8) X}  @tab @code{REAL(8)}  @tab GNU extension
4494@end multitable
4495
4496@item @emph{See also}:
4497Converse function: @gol
4498@ref{TAND} @gol
4499Radians function: @gol
4500@ref{COTAN}
4501@end table
4502
4503
4504
4505@node COUNT
4506@section @code{COUNT} --- Count function
4507@fnindex COUNT
4508@cindex array, conditionally count elements
4509@cindex array, element counting
4510@cindex array, number of elements
4511
4512@table @asis
4513@item @emph{Description}:
4514
4515Counts the number of @code{.TRUE.} elements in a logical @var{MASK},
4516or, if the @var{DIM} argument is supplied, counts the number of
4517elements along each row of the array in the @var{DIM} direction.
4518If the array has zero size, or all of the elements of @var{MASK} are
4519@code{.FALSE.}, then the result is @code{0}.
4520
4521@item @emph{Standard}:
4522Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
4523
4524@item @emph{Class}:
4525Transformational function
4526
4527@item @emph{Syntax}:
4528@code{RESULT = COUNT(MASK [, DIM, KIND])}
4529
4530@item @emph{Arguments}:
4531@multitable @columnfractions .15 .70
4532@item @var{MASK} @tab The type shall be @code{LOGICAL}.
4533@item @var{DIM}  @tab (Optional) The type shall be @code{INTEGER}.
4534@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
4535expression indicating the kind parameter of the result.
4536@end multitable
4537
4538@item @emph{Return value}:
4539The return value is of type @code{INTEGER} and of kind @var{KIND}. If
4540@var{KIND} is absent, the return value is of default integer kind.
4541If @var{DIM} is present, the result is an array with a rank one less
4542than the rank of @var{ARRAY}, and a size corresponding to the shape
4543of @var{ARRAY} with the @var{DIM} dimension removed.
4544
4545@item @emph{Example}:
4546@smallexample
4547program test_count
4548    integer, dimension(2,3) :: a, b
4549    logical, dimension(2,3) :: mask
4550    a = reshape( (/ 1, 2, 3, 4, 5, 6 /), (/ 2, 3 /))
4551    b = reshape( (/ 0, 7, 3, 4, 5, 8 /), (/ 2, 3 /))
4552    print '(3i3)', a(1,:)
4553    print '(3i3)', a(2,:)
4554    print *
4555    print '(3i3)', b(1,:)
4556    print '(3i3)', b(2,:)
4557    print *
4558    mask = a.ne.b
4559    print '(3l3)', mask(1,:)
4560    print '(3l3)', mask(2,:)
4561    print *
4562    print '(3i3)', count(mask)
4563    print *
4564    print '(3i3)', count(mask, 1)
4565    print *
4566    print '(3i3)', count(mask, 2)
4567end program test_count
4568@end smallexample
4569@end table
4570
4571
4572
4573@node CPU_TIME
4574@section @code{CPU_TIME} --- CPU elapsed time in seconds
4575@fnindex CPU_TIME
4576@cindex time, elapsed
4577
4578@table @asis
4579@item @emph{Description}:
4580Returns a @code{REAL} value representing the elapsed CPU time in
4581seconds.  This is useful for testing segments of code to determine
4582execution time.
4583
4584If a time source is available, time will be reported with microsecond
4585resolution. If no time source is available, @var{TIME} is set to
4586@code{-1.0}.
4587
4588Note that @var{TIME} may contain a, system dependent, arbitrary offset
4589and may not start with @code{0.0}. For @code{CPU_TIME}, the absolute
4590value is meaningless, only differences between subsequent calls to
4591this subroutine, as shown in the example below, should be used.
4592
4593
4594@item @emph{Standard}:
4595Fortran 95 and later
4596
4597@item @emph{Class}:
4598Subroutine
4599
4600@item @emph{Syntax}:
4601@code{CALL CPU_TIME(TIME)}
4602
4603@item @emph{Arguments}:
4604@multitable @columnfractions .15 .70
4605@item @var{TIME} @tab The type shall be @code{REAL} with @code{INTENT(OUT)}.
4606@end multitable
4607
4608@item @emph{Return value}:
4609None
4610
4611@item @emph{Example}:
4612@smallexample
4613program test_cpu_time
4614    real :: start, finish
4615    call cpu_time(start)
4616        ! put code to test here
4617    call cpu_time(finish)
4618    print '("Time = ",f6.3," seconds.")',finish-start
4619end program test_cpu_time
4620@end smallexample
4621
4622@item @emph{See also}:
4623@ref{SYSTEM_CLOCK}, @gol
4624@ref{DATE_AND_TIME}
4625@end table
4626
4627
4628
4629@node CSHIFT
4630@section @code{CSHIFT} --- Circular shift elements of an array
4631@fnindex CSHIFT
4632@cindex array, shift circularly
4633@cindex array, permutation
4634@cindex array, rotate
4635
4636@table @asis
4637@item @emph{Description}:
4638@code{CSHIFT(ARRAY, SHIFT [, DIM])} performs a circular shift on elements of
4639@var{ARRAY} along the dimension of @var{DIM}.  If @var{DIM} is omitted it is
4640taken to be @code{1}.  @var{DIM} is a scalar of type @code{INTEGER} in the
4641range of @math{1 \leq DIM \leq n)} where @math{n} is the rank of @var{ARRAY}.
4642If the rank of @var{ARRAY} is one, then all elements of @var{ARRAY} are shifted
4643by @var{SHIFT} places.  If rank is greater than one, then all complete rank one
4644sections of @var{ARRAY} along the given dimension are shifted.  Elements
4645shifted out one end of each rank one section are shifted back in the other end.
4646
4647@item @emph{Standard}:
4648Fortran 90 and later
4649
4650@item @emph{Class}:
4651Transformational function
4652
4653@item @emph{Syntax}:
4654@code{RESULT = CSHIFT(ARRAY, SHIFT [, DIM])}
4655
4656@item @emph{Arguments}:
4657@multitable @columnfractions .15 .70
4658@item @var{ARRAY}  @tab Shall be an array of any type.
4659@item @var{SHIFT}  @tab The type shall be @code{INTEGER}.
4660@item @var{DIM}    @tab The type shall be @code{INTEGER}.
4661@end multitable
4662
4663@item @emph{Return value}:
4664Returns an array of same type and rank as the @var{ARRAY} argument.
4665
4666@item @emph{Example}:
4667@smallexample
4668program test_cshift
4669    integer, dimension(3,3) :: a
4670    a = reshape( (/ 1, 2, 3, 4, 5, 6, 7, 8, 9 /), (/ 3, 3 /))
4671    print '(3i3)', a(1,:)
4672    print '(3i3)', a(2,:)
4673    print '(3i3)', a(3,:)
4674    a = cshift(a, SHIFT=(/1, 2, -1/), DIM=2)
4675    print *
4676    print '(3i3)', a(1,:)
4677    print '(3i3)', a(2,:)
4678    print '(3i3)', a(3,:)
4679end program test_cshift
4680@end smallexample
4681@end table
4682
4683
4684
4685@node CTIME
4686@section @code{CTIME} --- Convert a time into a string
4687@fnindex CTIME
4688@cindex time, conversion to string
4689@cindex conversion, to string
4690
4691@table @asis
4692@item @emph{Description}:
4693@code{CTIME} converts a system time value, such as returned by
4694@ref{TIME8}, to a string. The output will be of the form @samp{Sat
4695Aug 19 18:13:14 1995}.
4696
4697This intrinsic is provided in both subroutine and function forms; however,
4698only one form can be used in any given program unit.
4699
4700@item @emph{Standard}:
4701GNU extension
4702
4703@item @emph{Class}:
4704Subroutine, function
4705
4706@item @emph{Syntax}:
4707@multitable @columnfractions .80
4708@item @code{CALL CTIME(TIME, RESULT)}.
4709@item @code{RESULT = CTIME(TIME)}.
4710@end multitable
4711
4712@item @emph{Arguments}:
4713@multitable @columnfractions .15 .70
4714@item @var{TIME}    @tab The type shall be of type @code{INTEGER}.
4715@item @var{RESULT}  @tab The type shall be of type @code{CHARACTER} and
4716of default kind. It is an @code{INTENT(OUT)} argument. If the length
4717of this variable is too short for the time and date string to fit
4718completely, it will be blank on procedure return.
4719@end multitable
4720
4721@item @emph{Return value}:
4722The converted date and time as a string.
4723
4724@item @emph{Example}:
4725@smallexample
4726program test_ctime
4727    integer(8) :: i
4728    character(len=30) :: date
4729    i = time8()
4730
4731    ! Do something, main part of the program
4732
4733    call ctime(i,date)
4734    print *, 'Program was started on ', date
4735end program test_ctime
4736@end smallexample
4737
4738@item @emph{See Also}:
4739@ref{DATE_AND_TIME}, @gol
4740@ref{GMTIME}, @gol
4741@ref{LTIME}, @gol
4742@ref{TIME}, @gol
4743@ref{TIME8}
4744@end table
4745
4746
4747
4748@node DATE_AND_TIME
4749@section @code{DATE_AND_TIME} --- Date and time subroutine
4750@fnindex DATE_AND_TIME
4751@cindex date, current
4752@cindex current date
4753@cindex time, current
4754@cindex current time
4755
4756@table @asis
4757@item @emph{Description}:
4758@code{DATE_AND_TIME(DATE, TIME, ZONE, VALUES)} gets the corresponding date and
4759time information from the real-time system clock.  @var{DATE} is
4760@code{INTENT(OUT)} and has form ccyymmdd.  @var{TIME} is @code{INTENT(OUT)} and
4761has form hhmmss.sss.  @var{ZONE} is @code{INTENT(OUT)} and has form (+-)hhmm,
4762representing the difference with respect to Coordinated Universal Time (UTC).
4763Unavailable time and date parameters return blanks.
4764
4765@var{VALUES} is @code{INTENT(OUT)} and provides the following:
4766
4767@multitable @columnfractions .15 .30 .40
4768@item @tab @code{VALUE(1)}: @tab The year
4769@item @tab @code{VALUE(2)}: @tab The month
4770@item @tab @code{VALUE(3)}: @tab The day of the month
4771@item @tab @code{VALUE(4)}: @tab Time difference with UTC in minutes
4772@item @tab @code{VALUE(5)}: @tab The hour of the day
4773@item @tab @code{VALUE(6)}: @tab The minutes of the hour
4774@item @tab @code{VALUE(7)}: @tab The seconds of the minute
4775@item @tab @code{VALUE(8)}: @tab The milliseconds of the second
4776@end multitable
4777
4778@item @emph{Standard}:
4779Fortran 90 and later
4780
4781@item @emph{Class}:
4782Subroutine
4783
4784@item @emph{Syntax}:
4785@code{CALL DATE_AND_TIME([DATE, TIME, ZONE, VALUES])}
4786
4787@item @emph{Arguments}:
4788@multitable @columnfractions .15 .70
4789@item @var{DATE}  @tab (Optional) The type shall be @code{CHARACTER(LEN=8)}
4790or larger, and of default kind.
4791@item @var{TIME}  @tab (Optional) The type shall be @code{CHARACTER(LEN=10)}
4792or larger, and of default kind.
4793@item @var{ZONE}  @tab (Optional) The type shall be @code{CHARACTER(LEN=5)}
4794or larger, and of default kind.
4795@item @var{VALUES}@tab (Optional) The type shall be @code{INTEGER(8)}.
4796@end multitable
4797
4798@item @emph{Return value}:
4799None
4800
4801@item @emph{Example}:
4802@smallexample
4803program test_time_and_date
4804    character(8)  :: date
4805    character(10) :: time
4806    character(5)  :: zone
4807    integer,dimension(8) :: values
4808    ! using keyword arguments
4809    call date_and_time(date,time,zone,values)
4810    call date_and_time(DATE=date,ZONE=zone)
4811    call date_and_time(TIME=time)
4812    call date_and_time(VALUES=values)
4813    print '(a,2x,a,2x,a)', date, time, zone
4814    print '(8i5)', values
4815end program test_time_and_date
4816@end smallexample
4817
4818@item @emph{See also}:
4819@ref{CPU_TIME}, @gol
4820@ref{SYSTEM_CLOCK}
4821@end table
4822
4823
4824
4825@node DBLE
4826@section @code{DBLE} --- Double conversion function
4827@fnindex DBLE
4828@cindex conversion, to real
4829
4830@table @asis
4831@item @emph{Description}:
4832@code{DBLE(A)} Converts @var{A} to double precision real type.
4833
4834@item @emph{Standard}:
4835Fortran 77 and later
4836
4837@item @emph{Class}:
4838Elemental function
4839
4840@item @emph{Syntax}:
4841@code{RESULT = DBLE(A)}
4842
4843@item @emph{Arguments}:
4844@multitable @columnfractions .15 .70
4845@item @var{A} @tab The type shall be @code{INTEGER}, @code{REAL},
4846or @code{COMPLEX}.
4847@end multitable
4848
4849@item @emph{Return value}:
4850The return value is of type double precision real.
4851
4852@item @emph{Example}:
4853@smallexample
4854program test_dble
4855    real    :: x = 2.18
4856    integer :: i = 5
4857    complex :: z = (2.3,1.14)
4858    print *, dble(x), dble(i), dble(z)
4859end program test_dble
4860@end smallexample
4861
4862@item @emph{See also}:
4863@ref{REAL}
4864@end table
4865
4866
4867
4868@node DCMPLX
4869@section @code{DCMPLX} --- Double complex conversion function
4870@fnindex DCMPLX
4871@cindex complex numbers, conversion to
4872@cindex conversion, to complex
4873
4874@table @asis
4875@item @emph{Description}:
4876@code{DCMPLX(X [,Y])} returns a double complex number where @var{X} is
4877converted to the real component.  If @var{Y} is present it is converted to the
4878imaginary component.  If @var{Y} is not present then the imaginary component is
4879set to 0.0.  If @var{X} is complex then @var{Y} must not be present.
4880
4881@item @emph{Standard}:
4882GNU extension
4883
4884@item @emph{Class}:
4885Elemental function
4886
4887@item @emph{Syntax}:
4888@code{RESULT = DCMPLX(X [, Y])}
4889
4890@item @emph{Arguments}:
4891@multitable @columnfractions .15 .70
4892@item @var{X} @tab The type may be @code{INTEGER}, @code{REAL},
4893or @code{COMPLEX}.
4894@item @var{Y} @tab (Optional if @var{X} is not @code{COMPLEX}.) May be
4895@code{INTEGER} or @code{REAL}.
4896@end multitable
4897
4898@item @emph{Return value}:
4899The return value is of type @code{COMPLEX(8)}
4900
4901@item @emph{Example}:
4902@smallexample
4903program test_dcmplx
4904    integer :: i = 42
4905    real :: x = 3.14
4906    complex :: z
4907    z = cmplx(i, x)
4908    print *, dcmplx(i)
4909    print *, dcmplx(x)
4910    print *, dcmplx(z)
4911    print *, dcmplx(x,i)
4912end program test_dcmplx
4913@end smallexample
4914@end table
4915
4916
4917@node DIGITS
4918@section @code{DIGITS} --- Significant binary digits function
4919@fnindex DIGITS
4920@cindex model representation, significant digits
4921
4922@table @asis
4923@item @emph{Description}:
4924@code{DIGITS(X)} returns the number of significant binary digits of the internal
4925model representation of @var{X}.  For example, on a system using a 32-bit
4926floating point representation, a default real number would likely return 24.
4927
4928@item @emph{Standard}:
4929Fortran 90 and later
4930
4931@item @emph{Class}:
4932Inquiry function
4933
4934@item @emph{Syntax}:
4935@code{RESULT = DIGITS(X)}
4936
4937@item @emph{Arguments}:
4938@multitable @columnfractions .15 .70
4939@item @var{X} @tab The type may be @code{INTEGER} or @code{REAL}.
4940@end multitable
4941
4942@item @emph{Return value}:
4943The return value is of type @code{INTEGER}.
4944
4945@item @emph{Example}:
4946@smallexample
4947program test_digits
4948    integer :: i = 12345
4949    real :: x = 3.143
4950    real(8) :: y = 2.33
4951    print *, digits(i)
4952    print *, digits(x)
4953    print *, digits(y)
4954end program test_digits
4955@end smallexample
4956@end table
4957
4958
4959
4960@node DIM
4961@section @code{DIM} --- Positive difference
4962@fnindex DIM
4963@fnindex IDIM
4964@fnindex DDIM
4965@cindex positive difference
4966
4967@table @asis
4968@item @emph{Description}:
4969@code{DIM(X,Y)} returns the difference @code{X-Y} if the result is positive;
4970otherwise returns zero.
4971
4972@item @emph{Standard}:
4973Fortran 77 and later
4974
4975@item @emph{Class}:
4976Elemental function
4977
4978@item @emph{Syntax}:
4979@code{RESULT = DIM(X, Y)}
4980
4981@item @emph{Arguments}:
4982@multitable @columnfractions .15 .70
4983@item @var{X} @tab The type shall be @code{INTEGER} or @code{REAL}
4984@item @var{Y} @tab The type shall be the same type and kind as @var{X}.  (As
4985a GNU extension, arguments of different kinds are permitted.)
4986@end multitable
4987
4988@item @emph{Return value}:
4989The return value is of type @code{INTEGER} or @code{REAL}.  (As a GNU
4990extension, kind is the largest kind of the actual arguments.)
4991
4992@item @emph{Example}:
4993@smallexample
4994program test_dim
4995    integer :: i
4996    real(8) :: x
4997    i = dim(4, 15)
4998    x = dim(4.345_8, 2.111_8)
4999    print *, i
5000    print *, x
5001end program test_dim
5002@end smallexample
5003
5004@item @emph{Specific names}:
5005@multitable @columnfractions .20 .20 .20 .25
5006@item Name             @tab Argument               @tab Return type       @tab Standard
5007@item @code{DIM(X,Y)}  @tab @code{REAL(4) X, Y}    @tab @code{REAL(4)}    @tab Fortran 77 and later
5008@item @code{IDIM(X,Y)} @tab @code{INTEGER(4) X, Y} @tab @code{INTEGER(4)} @tab Fortran 77 and later
5009@item @code{DDIM(X,Y)} @tab @code{REAL(8) X, Y}    @tab @code{REAL(8)}    @tab Fortran 77 and later
5010@end multitable
5011@end table
5012
5013
5014
5015@node DOT_PRODUCT
5016@section @code{DOT_PRODUCT} --- Dot product function
5017@fnindex DOT_PRODUCT
5018@cindex dot product
5019@cindex vector product
5020@cindex product, vector
5021
5022@table @asis
5023@item @emph{Description}:
5024@code{DOT_PRODUCT(VECTOR_A, VECTOR_B)} computes the dot product multiplication
5025of two vectors @var{VECTOR_A} and @var{VECTOR_B}.  The two vectors may be
5026either numeric or logical and must be arrays of rank one and of equal size. If
5027the vectors are @code{INTEGER} or @code{REAL}, the result is
5028@code{SUM(VECTOR_A*VECTOR_B)}. If the vectors are @code{COMPLEX}, the result
5029is @code{SUM(CONJG(VECTOR_A)*VECTOR_B)}. If the vectors are @code{LOGICAL},
5030the result is @code{ANY(VECTOR_A .AND. VECTOR_B)}.
5031
5032@item @emph{Standard}:
5033Fortran 90 and later
5034
5035@item @emph{Class}:
5036Transformational function
5037
5038@item @emph{Syntax}:
5039@code{RESULT = DOT_PRODUCT(VECTOR_A, VECTOR_B)}
5040
5041@item @emph{Arguments}:
5042@multitable @columnfractions .15 .70
5043@item @var{VECTOR_A} @tab The type shall be numeric or @code{LOGICAL}, rank 1.
5044@item @var{VECTOR_B} @tab The type shall be numeric if @var{VECTOR_A} is of numeric type or @code{LOGICAL} if @var{VECTOR_A} is of type @code{LOGICAL}. @var{VECTOR_B} shall be a rank-one array.
5045@end multitable
5046
5047@item @emph{Return value}:
5048If the arguments are numeric, the return value is a scalar of numeric type,
5049@code{INTEGER}, @code{REAL}, or @code{COMPLEX}.  If the arguments are
5050@code{LOGICAL}, the return value is @code{.TRUE.} or @code{.FALSE.}.
5051
5052@item @emph{Example}:
5053@smallexample
5054program test_dot_prod
5055    integer, dimension(3) :: a, b
5056    a = (/ 1, 2, 3 /)
5057    b = (/ 4, 5, 6 /)
5058    print '(3i3)', a
5059    print *
5060    print '(3i3)', b
5061    print *
5062    print *, dot_product(a,b)
5063end program test_dot_prod
5064@end smallexample
5065@end table
5066
5067
5068
5069@node DPROD
5070@section @code{DPROD} --- Double product function
5071@fnindex DPROD
5072@cindex product, double-precision
5073
5074@table @asis
5075@item @emph{Description}:
5076@code{DPROD(X,Y)} returns the product @code{X*Y}.
5077
5078@item @emph{Standard}:
5079Fortran 77 and later
5080
5081@item @emph{Class}:
5082Elemental function
5083
5084@item @emph{Syntax}:
5085@code{RESULT = DPROD(X, Y)}
5086
5087@item @emph{Arguments}:
5088@multitable @columnfractions .15 .70
5089@item @var{X} @tab The type shall be @code{REAL}.
5090@item @var{Y} @tab The type shall be @code{REAL}.
5091@end multitable
5092
5093@item @emph{Return value}:
5094The return value is of type @code{REAL(8)}.
5095
5096@item @emph{Example}:
5097@smallexample
5098program test_dprod
5099    real :: x = 5.2
5100    real :: y = 2.3
5101    real(8) :: d
5102    d = dprod(x,y)
5103    print *, d
5104end program test_dprod
5105@end smallexample
5106
5107@item @emph{Specific names}:
5108@multitable @columnfractions .20 .20 .20 .25
5109@item Name              @tab Argument               @tab Return type       @tab Standard
5110@item @code{DPROD(X,Y)} @tab @code{REAL(4) X, Y}    @tab @code{REAL(8)}    @tab Fortran 77 and later
5111@end multitable
5112
5113@end table
5114
5115
5116@node DREAL
5117@section @code{DREAL} --- Double real part function
5118@fnindex DREAL
5119@cindex complex numbers, real part
5120
5121@table @asis
5122@item @emph{Description}:
5123@code{DREAL(Z)} returns the real part of complex variable @var{Z}.
5124
5125@item @emph{Standard}:
5126GNU extension
5127
5128@item @emph{Class}:
5129Elemental function
5130
5131@item @emph{Syntax}:
5132@code{RESULT = DREAL(A)}
5133
5134@item @emph{Arguments}:
5135@multitable @columnfractions .15 .70
5136@item @var{A} @tab The type shall be @code{COMPLEX(8)}.
5137@end multitable
5138
5139@item @emph{Return value}:
5140The return value is of type @code{REAL(8)}.
5141
5142@item @emph{Example}:
5143@smallexample
5144program test_dreal
5145    complex(8) :: z = (1.3_8,7.2_8)
5146    print *, dreal(z)
5147end program test_dreal
5148@end smallexample
5149
5150@item @emph{See also}:
5151@ref{AIMAG}
5152
5153@end table
5154
5155
5156
5157@node DSHIFTL
5158@section @code{DSHIFTL} --- Combined left shift
5159@fnindex DSHIFTL
5160@cindex left shift, combined
5161@cindex shift, left
5162
5163@table @asis
5164@item @emph{Description}:
5165@code{DSHIFTL(I, J, SHIFT)} combines bits of @var{I} and @var{J}. The
5166rightmost @var{SHIFT} bits of the result are the leftmost @var{SHIFT}
5167bits of @var{J}, and the remaining bits are the rightmost bits of
5168@var{I}.
5169
5170@item @emph{Standard}:
5171Fortran 2008 and later
5172
5173@item @emph{Class}:
5174Elemental function
5175
5176@item @emph{Syntax}:
5177@code{RESULT = DSHIFTL(I, J, SHIFT)}
5178
5179@item @emph{Arguments}:
5180@multitable @columnfractions .15 .70
5181@item @var{I} @tab Shall be of type @code{INTEGER} or a BOZ constant.
5182@item @var{J} @tab Shall be of type @code{INTEGER} or a BOZ constant.
5183If both @var{I} and @var{J} have integer type, then they shall have
5184the same kind type parameter. @var{I} and @var{J} shall not both be
5185BOZ constants.
5186@item @var{SHIFT} @tab Shall be of type @code{INTEGER}. It shall
5187be nonnegative.  If @var{I} is not a BOZ constant, then @var{SHIFT}
5188shall be less than or equal to @code{BIT_SIZE(I)}; otherwise,
5189@var{SHIFT} shall be less than or equal to @code{BIT_SIZE(J)}.
5190@end multitable
5191
5192@item @emph{Return value}:
5193If either @var{I} or @var{J} is a BOZ constant, it is first converted
5194as if by the intrinsic function @code{INT} to an integer type with the
5195kind type parameter of the other.
5196
5197@item @emph{See also}:
5198@ref{DSHIFTR}
5199@end table
5200
5201
5202@node DSHIFTR
5203@section @code{DSHIFTR} --- Combined right shift
5204@fnindex DSHIFTR
5205@cindex right shift, combined
5206@cindex shift, right
5207
5208@table @asis
5209@item @emph{Description}:
5210@code{DSHIFTR(I, J, SHIFT)} combines bits of @var{I} and @var{J}. The
5211leftmost @var{SHIFT} bits of the result are the rightmost @var{SHIFT}
5212bits of @var{I}, and the remaining bits are the leftmost bits of
5213@var{J}.
5214
5215@item @emph{Standard}:
5216Fortran 2008 and later
5217
5218@item @emph{Class}:
5219Elemental function
5220
5221@item @emph{Syntax}:
5222@code{RESULT = DSHIFTR(I, J, SHIFT)}
5223
5224@item @emph{Arguments}:
5225@multitable @columnfractions .15 .70
5226@item @var{I} @tab Shall be of type @code{INTEGER} or a BOZ constant.
5227@item @var{J} @tab Shall be of type @code{INTEGER} or a BOZ constant.
5228If both @var{I} and @var{J} have integer type, then they shall have
5229the same kind type parameter. @var{I} and @var{J} shall not both be
5230BOZ constants.
5231@item @var{SHIFT} @tab Shall be of type @code{INTEGER}. It shall
5232be nonnegative.  If @var{I} is not a BOZ constant, then @var{SHIFT}
5233shall be less than or equal to @code{BIT_SIZE(I)}; otherwise,
5234@var{SHIFT} shall be less than or equal to @code{BIT_SIZE(J)}.
5235@end multitable
5236
5237@item @emph{Return value}:
5238If either @var{I} or @var{J} is a BOZ constant, it is first converted
5239as if by the intrinsic function @code{INT} to an integer type with the
5240kind type parameter of the other.
5241
5242@item @emph{See also}:
5243@ref{DSHIFTL}
5244@end table
5245
5246
5247@node DTIME
5248@section @code{DTIME} --- Execution time subroutine (or function)
5249@fnindex DTIME
5250@cindex time, elapsed
5251@cindex elapsed time
5252
5253@table @asis
5254@item @emph{Description}:
5255@code{DTIME(VALUES, TIME)} initially returns the number of seconds of runtime
5256since the start of the process's execution in @var{TIME}.  @var{VALUES}
5257returns the user and system components of this time in @code{VALUES(1)} and
5258@code{VALUES(2)} respectively. @var{TIME} is equal to @code{VALUES(1) +
5259VALUES(2)}.
5260
5261Subsequent invocations of @code{DTIME} return values accumulated since the
5262previous invocation.
5263
5264On some systems, the underlying timings are represented using types with
5265sufficiently small limits that overflows (wrap around) are possible, such as
526632-bit types. Therefore, the values returned by this intrinsic might be, or
5267become, negative, or numerically less than previous values, during a single
5268run of the compiled program.
5269
5270Please note, that this implementation is thread safe if used within OpenMP
5271directives, i.e., its state will be consistent while called from multiple
5272threads. However, if @code{DTIME} is called from multiple threads, the result
5273is still the time since the last invocation. This may not give the intended
5274results. If possible, use @code{CPU_TIME} instead.
5275
5276This intrinsic is provided in both subroutine and function forms; however,
5277only one form can be used in any given program unit.
5278
5279@var{VALUES} and @var{TIME} are @code{INTENT(OUT)} and provide the following:
5280
5281@multitable @columnfractions .15 .30 .40
5282@item @tab @code{VALUES(1)}: @tab User time in seconds.
5283@item @tab @code{VALUES(2)}: @tab System time in seconds.
5284@item @tab @code{TIME}: @tab Run time since start in seconds.
5285@end multitable
5286
5287@item @emph{Standard}:
5288GNU extension
5289
5290@item @emph{Class}:
5291Subroutine, function
5292
5293@item @emph{Syntax}:
5294@multitable @columnfractions .80
5295@item @code{CALL DTIME(VALUES, TIME)}.
5296@item @code{TIME = DTIME(VALUES)}, (not recommended).
5297@end multitable
5298
5299@item @emph{Arguments}:
5300@multitable @columnfractions .15 .70
5301@item @var{VALUES}@tab The type shall be @code{REAL(4), DIMENSION(2)}.
5302@item @var{TIME}@tab The type shall be @code{REAL(4)}.
5303@end multitable
5304
5305@item @emph{Return value}:
5306Elapsed time in seconds since the last invocation or since the start of program
5307execution if not called before.
5308
5309@item @emph{Example}:
5310@smallexample
5311program test_dtime
5312    integer(8) :: i, j
5313    real, dimension(2) :: tarray
5314    real :: result
5315    call dtime(tarray, result)
5316    print *, result
5317    print *, tarray(1)
5318    print *, tarray(2)
5319    do i=1,100000000    ! Just a delay
5320        j = i * i - i
5321    end do
5322    call dtime(tarray, result)
5323    print *, result
5324    print *, tarray(1)
5325    print *, tarray(2)
5326end program test_dtime
5327@end smallexample
5328
5329@item @emph{See also}:
5330@ref{CPU_TIME}
5331
5332@end table
5333
5334
5335
5336@node EOSHIFT
5337@section @code{EOSHIFT} --- End-off shift elements of an array
5338@fnindex EOSHIFT
5339@cindex array, shift
5340
5341@table @asis
5342@item @emph{Description}:
5343@code{EOSHIFT(ARRAY, SHIFT[, BOUNDARY, DIM])} performs an end-off shift on
5344elements of @var{ARRAY} along the dimension of @var{DIM}.  If @var{DIM} is
5345omitted it is taken to be @code{1}.  @var{DIM} is a scalar of type
5346@code{INTEGER} in the range of @math{1 \leq DIM \leq n)} where @math{n} is the
5347rank of @var{ARRAY}.  If the rank of @var{ARRAY} is one, then all elements of
5348@var{ARRAY} are shifted by @var{SHIFT} places.  If rank is greater than one,
5349then all complete rank one sections of @var{ARRAY} along the given dimension are
5350shifted.  Elements shifted out one end of each rank one section are dropped.  If
5351@var{BOUNDARY} is present then the corresponding value of from @var{BOUNDARY}
5352is copied back in the other end.  If @var{BOUNDARY} is not present then the
5353following are copied in depending on the type of @var{ARRAY}.
5354
5355@multitable @columnfractions .15 .80
5356@item @emph{Array Type} @tab @emph{Boundary Value}
5357@item Numeric  @tab 0 of the type and kind of @var{ARRAY}.
5358@item Logical  @tab @code{.FALSE.}.
5359@item Character(@var{len}) @tab @var{len} blanks.
5360@end multitable
5361
5362@item @emph{Standard}:
5363Fortran 90 and later
5364
5365@item @emph{Class}:
5366Transformational function
5367
5368@item @emph{Syntax}:
5369@code{RESULT = EOSHIFT(ARRAY, SHIFT [, BOUNDARY, DIM])}
5370
5371@item @emph{Arguments}:
5372@multitable @columnfractions .15 .70
5373@item @var{ARRAY}  @tab May be any type, not scalar.
5374@item @var{SHIFT}  @tab The type shall be @code{INTEGER}.
5375@item @var{BOUNDARY} @tab Same type as @var{ARRAY}.
5376@item @var{DIM}    @tab The type shall be @code{INTEGER}.
5377@end multitable
5378
5379@item @emph{Return value}:
5380Returns an array of same type and rank as the @var{ARRAY} argument.
5381
5382@item @emph{Example}:
5383@smallexample
5384program test_eoshift
5385    integer, dimension(3,3) :: a
5386    a = reshape( (/ 1, 2, 3, 4, 5, 6, 7, 8, 9 /), (/ 3, 3 /))
5387    print '(3i3)', a(1,:)
5388    print '(3i3)', a(2,:)
5389    print '(3i3)', a(3,:)
5390    a = EOSHIFT(a, SHIFT=(/1, 2, 1/), BOUNDARY=-5, DIM=2)
5391    print *
5392    print '(3i3)', a(1,:)
5393    print '(3i3)', a(2,:)
5394    print '(3i3)', a(3,:)
5395end program test_eoshift
5396@end smallexample
5397@end table
5398
5399
5400
5401@node EPSILON
5402@section @code{EPSILON} --- Epsilon function
5403@fnindex EPSILON
5404@cindex model representation, epsilon
5405
5406@table @asis
5407@item @emph{Description}:
5408@code{EPSILON(X)} returns the smallest number @var{E} of the same kind
5409as @var{X} such that @math{1 + E > 1}.
5410
5411@item @emph{Standard}:
5412Fortran 90 and later
5413
5414@item @emph{Class}:
5415Inquiry function
5416
5417@item @emph{Syntax}:
5418@code{RESULT = EPSILON(X)}
5419
5420@item @emph{Arguments}:
5421@multitable @columnfractions .15 .70
5422@item @var{X} @tab The type shall be @code{REAL}.
5423@end multitable
5424
5425@item @emph{Return value}:
5426The return value is of same type as the argument.
5427
5428@item @emph{Example}:
5429@smallexample
5430program test_epsilon
5431    real :: x = 3.143
5432    real(8) :: y = 2.33
5433    print *, EPSILON(x)
5434    print *, EPSILON(y)
5435end program test_epsilon
5436@end smallexample
5437@end table
5438
5439
5440
5441@node ERF
5442@section @code{ERF} --- Error function
5443@fnindex ERF
5444@cindex error function
5445
5446@table @asis
5447@item @emph{Description}:
5448@code{ERF(X)} computes the error function of @var{X}.
5449
5450@item @emph{Standard}:
5451Fortran 2008 and later
5452
5453@item @emph{Class}:
5454Elemental function
5455
5456@item @emph{Syntax}:
5457@code{RESULT = ERF(X)}
5458
5459@item @emph{Arguments}:
5460@multitable @columnfractions .15 .70
5461@item @var{X} @tab The type shall be @code{REAL}.
5462@end multitable
5463
5464@item @emph{Return value}:
5465The return value is of type @code{REAL}, of the same kind as
5466@var{X} and lies in the range @math{-1 \leq erf (x) \leq 1 }.
5467
5468@item @emph{Example}:
5469@smallexample
5470program test_erf
5471  real(8) :: x = 0.17_8
5472  x = erf(x)
5473end program test_erf
5474@end smallexample
5475
5476@item @emph{Specific names}:
5477@multitable @columnfractions .20 .20 .20 .25
5478@item Name            @tab Argument          @tab Return type       @tab Standard
5479@item @code{DERF(X)}  @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
5480@end multitable
5481@end table
5482
5483
5484
5485@node ERFC
5486@section @code{ERFC} --- Error function
5487@fnindex ERFC
5488@cindex error function, complementary
5489
5490@table @asis
5491@item @emph{Description}:
5492@code{ERFC(X)} computes the complementary error function of @var{X}.
5493
5494@item @emph{Standard}:
5495Fortran 2008 and later
5496
5497@item @emph{Class}:
5498Elemental function
5499
5500@item @emph{Syntax}:
5501@code{RESULT = ERFC(X)}
5502
5503@item @emph{Arguments}:
5504@multitable @columnfractions .15 .70
5505@item @var{X} @tab The type shall be @code{REAL}.
5506@end multitable
5507
5508@item @emph{Return value}:
5509The return value is of type @code{REAL} and of the same kind as @var{X}.
5510It lies in the range @math{ 0 \leq erfc (x) \leq 2 }.
5511
5512@item @emph{Example}:
5513@smallexample
5514program test_erfc
5515  real(8) :: x = 0.17_8
5516  x = erfc(x)
5517end program test_erfc
5518@end smallexample
5519
5520@item @emph{Specific names}:
5521@multitable @columnfractions .20 .20 .20 .25
5522@item Name            @tab Argument          @tab Return type       @tab Standard
5523@item @code{DERFC(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab GNU extension
5524@end multitable
5525@end table
5526
5527
5528
5529@node ERFC_SCALED
5530@section @code{ERFC_SCALED} --- Error function
5531@fnindex ERFC_SCALED
5532@cindex error function, complementary, exponentially-scaled
5533
5534@table @asis
5535@item @emph{Description}:
5536@code{ERFC_SCALED(X)} computes the exponentially-scaled complementary
5537error function of @var{X}.
5538
5539@item @emph{Standard}:
5540Fortran 2008 and later
5541
5542@item @emph{Class}:
5543Elemental function
5544
5545@item @emph{Syntax}:
5546@code{RESULT = ERFC_SCALED(X)}
5547
5548@item @emph{Arguments}:
5549@multitable @columnfractions .15 .70
5550@item @var{X} @tab The type shall be @code{REAL}.
5551@end multitable
5552
5553@item @emph{Return value}:
5554The return value is of type @code{REAL} and of the same kind as @var{X}.
5555
5556@item @emph{Example}:
5557@smallexample
5558program test_erfc_scaled
5559  real(8) :: x = 0.17_8
5560  x = erfc_scaled(x)
5561end program test_erfc_scaled
5562@end smallexample
5563@end table
5564
5565
5566
5567@node ETIME
5568@section @code{ETIME} --- Execution time subroutine (or function)
5569@fnindex ETIME
5570@cindex time, elapsed
5571
5572@table @asis
5573@item @emph{Description}:
5574@code{ETIME(VALUES, TIME)} returns the number of seconds of runtime
5575since the start of the process's execution in @var{TIME}.  @var{VALUES}
5576returns the user and system components of this time in @code{VALUES(1)} and
5577@code{VALUES(2)} respectively. @var{TIME} is equal to @code{VALUES(1) + VALUES(2)}.
5578
5579On some systems, the underlying timings are represented using types with
5580sufficiently small limits that overflows (wrap around) are possible, such as
558132-bit types. Therefore, the values returned by this intrinsic might be, or
5582become, negative, or numerically less than previous values, during a single
5583run of the compiled program.
5584
5585This intrinsic is provided in both subroutine and function forms; however,
5586only one form can be used in any given program unit.
5587
5588@var{VALUES} and @var{TIME} are @code{INTENT(OUT)} and provide the following:
5589
5590@multitable @columnfractions .15 .30 .60
5591@item @tab @code{VALUES(1)}: @tab User time in seconds.
5592@item @tab @code{VALUES(2)}: @tab System time in seconds.
5593@item @tab @code{TIME}: @tab Run time since start in seconds.
5594@end multitable
5595
5596@item @emph{Standard}:
5597GNU extension
5598
5599@item @emph{Class}:
5600Subroutine, function
5601
5602@item @emph{Syntax}:
5603@multitable @columnfractions .80
5604@item @code{CALL ETIME(VALUES, TIME)}.
5605@item @code{TIME = ETIME(VALUES)}, (not recommended).
5606@end multitable
5607
5608@item @emph{Arguments}:
5609@multitable @columnfractions .15 .70
5610@item @var{VALUES}@tab The type shall be @code{REAL(4), DIMENSION(2)}.
5611@item @var{TIME}@tab The type shall be @code{REAL(4)}.
5612@end multitable
5613
5614@item @emph{Return value}:
5615Elapsed time in seconds since the start of program execution.
5616
5617@item @emph{Example}:
5618@smallexample
5619program test_etime
5620    integer(8) :: i, j
5621    real, dimension(2) :: tarray
5622    real :: result
5623    call ETIME(tarray, result)
5624    print *, result
5625    print *, tarray(1)
5626    print *, tarray(2)
5627    do i=1,100000000    ! Just a delay
5628        j = i * i - i
5629    end do
5630    call ETIME(tarray, result)
5631    print *, result
5632    print *, tarray(1)
5633    print *, tarray(2)
5634end program test_etime
5635@end smallexample
5636
5637@item @emph{See also}:
5638@ref{CPU_TIME}
5639
5640@end table
5641
5642
5643
5644@node EVENT_QUERY
5645@section @code{EVENT_QUERY} --- Query whether a coarray event has occurred
5646@fnindex EVENT_QUERY
5647@cindex Events, EVENT_QUERY
5648
5649@table @asis
5650@item @emph{Description}:
5651@code{EVENT_QUERY} assignes the number of events to @var{COUNT} which have been
5652posted to the @var{EVENT} variable and not yet been removed by calling
5653@code{EVENT WAIT}. When @var{STAT} is present and the invocation was successful,
5654it is assigned the value 0. If it is present and the invocation has failed,
5655it is assigned a positive value and @var{COUNT} is assigned the value @math{-1}.
5656
5657@item @emph{Standard}:
5658TS 18508 or later
5659
5660@item @emph{Class}:
5661 subroutine
5662
5663@item @emph{Syntax}:
5664@code{CALL EVENT_QUERY (EVENT, COUNT [, STAT])}
5665
5666@item @emph{Arguments}:
5667@multitable @columnfractions .15 .70
5668@item @var{EVENT}  @tab (intent(IN)) Scalar of type @code{EVENT_TYPE},
5669defined in @code{ISO_FORTRAN_ENV}; shall not be coindexed.
5670@item @var{COUNT}  @tab (intent(out))Scalar integer with at least the
5671precision of default integer.
5672@item @var{STAT}   @tab (optional) Scalar default-kind integer variable.
5673@end multitable
5674
5675@item @emph{Example}:
5676@smallexample
5677program atomic
5678  use iso_fortran_env
5679  implicit none
5680  type(event_type) :: event_value_has_been_set[*]
5681  integer :: cnt
5682  if (this_image() == 1) then
5683    call event_query (event_value_has_been_set, cnt)
5684    if (cnt > 0) write(*,*) "Value has been set"
5685  elseif (this_image() == 2) then
5686    event post (event_value_has_been_set[1])
5687  end if
5688end program atomic
5689@end smallexample
5690
5691@end table
5692
5693
5694
5695@node EXECUTE_COMMAND_LINE
5696@section @code{EXECUTE_COMMAND_LINE} --- Execute a shell command
5697@fnindex EXECUTE_COMMAND_LINE
5698@cindex system, system call
5699@cindex command line
5700
5701@table @asis
5702@item @emph{Description}:
5703@code{EXECUTE_COMMAND_LINE} runs a shell command, synchronously or
5704asynchronously.
5705
5706The @code{COMMAND} argument is passed to the shell and executed (The
5707shell is @code{sh} on Unix systems, and @code{cmd.exe} on Windows.).
5708If @code{WAIT} is present and has the value false, the execution of
5709the command is asynchronous if the system supports it; otherwise, the
5710command is executed synchronously using the C library's @code{system}
5711call.
5712
5713The three last arguments allow the user to get status information.  After
5714synchronous execution, @code{EXITSTAT} contains the integer exit code of
5715the command, as returned by @code{system}.  @code{CMDSTAT} is set to zero
5716if the command line was executed (whatever its exit status was).
5717@code{CMDMSG} is assigned an error message if an error has occurred.
5718
5719Note that the @code{system} function need not be thread-safe. It is
5720the responsibility of the user to ensure that @code{system} is not
5721called concurrently.
5722
5723For asynchronous execution on supported targets, the POSIX
5724@code{posix_spawn} or @code{fork} functions are used.  Also, a signal
5725handler for the @code{SIGCHLD} signal is installed.
5726
5727@item @emph{Standard}:
5728Fortran 2008 and later
5729
5730@item @emph{Class}:
5731Subroutine
5732
5733@item @emph{Syntax}:
5734@code{CALL EXECUTE_COMMAND_LINE(COMMAND [, WAIT, EXITSTAT, CMDSTAT, CMDMSG ])}
5735
5736@item @emph{Arguments}:
5737@multitable @columnfractions .15 .70
5738@item @var{COMMAND} @tab Shall be a default @code{CHARACTER} scalar.
5739@item @var{WAIT} @tab (Optional) Shall be a default @code{LOGICAL} scalar.
5740@item @var{EXITSTAT} @tab (Optional) Shall be an @code{INTEGER} of the
5741default kind.
5742@item @var{CMDSTAT} @tab (Optional) Shall be an @code{INTEGER} of the
5743default kind.
5744@item @var{CMDMSG} @tab (Optional) Shall be an @code{CHARACTER} scalar of the
5745default kind.
5746@end multitable
5747
5748@item @emph{Example}:
5749@smallexample
5750program test_exec
5751  integer :: i
5752
5753  call execute_command_line ("external_prog.exe", exitstat=i)
5754  print *, "Exit status of external_prog.exe was ", i
5755
5756  call execute_command_line ("reindex_files.exe", wait=.false.)
5757  print *, "Now reindexing files in the background"
5758
5759end program test_exec
5760@end smallexample
5761
5762
5763@item @emph{Note}:
5764
5765Because this intrinsic is implemented in terms of the @code{system}
5766function call, its behavior with respect to signaling is processor
5767dependent. In particular, on POSIX-compliant systems, the SIGINT and
5768SIGQUIT signals will be ignored, and the SIGCHLD will be blocked. As
5769such, if the parent process is terminated, the child process might not be
5770terminated alongside.
5771
5772
5773@item @emph{See also}:
5774@ref{SYSTEM}
5775@end table
5776
5777
5778
5779@node EXIT
5780@section @code{EXIT} --- Exit the program with status.
5781@fnindex EXIT
5782@cindex program termination
5783@cindex terminate program
5784
5785@table @asis
5786@item @emph{Description}:
5787@code{EXIT} causes immediate termination of the program with status.  If status
5788is omitted it returns the canonical @emph{success} for the system.  All Fortran
5789I/O units are closed.
5790
5791@item @emph{Standard}:
5792GNU extension
5793
5794@item @emph{Class}:
5795Subroutine
5796
5797@item @emph{Syntax}:
5798@code{CALL EXIT([STATUS])}
5799
5800@item @emph{Arguments}:
5801@multitable @columnfractions .15 .70
5802@item @var{STATUS} @tab Shall be an @code{INTEGER} of the default kind.
5803@end multitable
5804
5805@item @emph{Return value}:
5806@code{STATUS} is passed to the parent process on exit.
5807
5808@item @emph{Example}:
5809@smallexample
5810program test_exit
5811  integer :: STATUS = 0
5812  print *, 'This program is going to exit.'
5813  call EXIT(STATUS)
5814end program test_exit
5815@end smallexample
5816
5817@item @emph{See also}:
5818@ref{ABORT}, @gol
5819@ref{KILL}
5820@end table
5821
5822
5823
5824@node EXP
5825@section @code{EXP} --- Exponential function
5826@fnindex EXP
5827@fnindex DEXP
5828@fnindex CEXP
5829@fnindex ZEXP
5830@fnindex CDEXP
5831@cindex exponential function
5832@cindex logarithm function, inverse
5833
5834@table @asis
5835@item @emph{Description}:
5836@code{EXP(X)} computes the base @math{e} exponential of @var{X}.
5837
5838@item @emph{Standard}:
5839Fortran 77 and later, has overloads that are GNU extensions
5840
5841@item @emph{Class}:
5842Elemental function
5843
5844@item @emph{Syntax}:
5845@code{RESULT = EXP(X)}
5846
5847@item @emph{Arguments}:
5848@multitable @columnfractions .15 .70
5849@item @var{X} @tab The type shall be @code{REAL} or
5850@code{COMPLEX}.
5851@end multitable
5852
5853@item @emph{Return value}:
5854The return value has same type and kind as @var{X}.
5855
5856@item @emph{Example}:
5857@smallexample
5858program test_exp
5859  real :: x = 1.0
5860  x = exp(x)
5861end program test_exp
5862@end smallexample
5863
5864@item @emph{Specific names}:
5865@multitable @columnfractions .20 .20 .20 .25
5866@item Name            @tab Argument             @tab Return type         @tab Standard
5867@item @code{EXP(X)}   @tab @code{REAL(4) X}     @tab @code{REAL(4)}      @tab Fortran 77 and later
5868@item @code{DEXP(X)}  @tab @code{REAL(8) X}     @tab @code{REAL(8)}      @tab Fortran 77 and later
5869@item @code{CEXP(X)}  @tab @code{COMPLEX(4) X}  @tab @code{COMPLEX(4)}   @tab Fortran 77 and later
5870@item @code{ZEXP(X)}  @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)}   @tab GNU extension
5871@item @code{CDEXP(X)} @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)}   @tab GNU extension
5872@end multitable
5873@end table
5874
5875
5876
5877@node EXPONENT
5878@section @code{EXPONENT} --- Exponent function
5879@fnindex EXPONENT
5880@cindex real number, exponent
5881@cindex floating point, exponent
5882
5883@table @asis
5884@item @emph{Description}:
5885@code{EXPONENT(X)} returns the value of the exponent part of @var{X}. If @var{X}
5886is zero the value returned is zero.
5887
5888@item @emph{Standard}:
5889Fortran 90 and later
5890
5891@item @emph{Class}:
5892Elemental function
5893
5894@item @emph{Syntax}:
5895@code{RESULT = EXPONENT(X)}
5896
5897@item @emph{Arguments}:
5898@multitable @columnfractions .15 .70
5899@item @var{X} @tab The type shall be @code{REAL}.
5900@end multitable
5901
5902@item @emph{Return value}:
5903The return value is of type default @code{INTEGER}.
5904
5905@item @emph{Example}:
5906@smallexample
5907program test_exponent
5908  real :: x = 1.0
5909  integer :: i
5910  i = exponent(x)
5911  print *, i
5912  print *, exponent(0.0)
5913end program test_exponent
5914@end smallexample
5915@end table
5916
5917
5918
5919@node EXTENDS_TYPE_OF
5920@section @code{EXTENDS_TYPE_OF} ---  Query dynamic type for extension
5921@fnindex EXTENDS_TYPE_OF
5922
5923@table @asis
5924@item @emph{Description}:
5925Query dynamic type for extension.
5926
5927@item @emph{Standard}:
5928Fortran 2003 and later
5929
5930@item @emph{Class}:
5931Inquiry function
5932
5933@item @emph{Syntax}:
5934@code{RESULT = EXTENDS_TYPE_OF(A, MOLD)}
5935
5936@item @emph{Arguments}:
5937@multitable @columnfractions .15 .70
5938@item @var{A} @tab Shall be an object of extensible declared type or
5939unlimited polymorphic.
5940@item @var{MOLD} @tab Shall be an object of extensible declared type or
5941unlimited polymorphic.
5942@end multitable
5943
5944@item @emph{Return value}:
5945The return value is a scalar of type default logical. It is true if and only if
5946the dynamic type of A is an extension type of the dynamic type of MOLD.
5947
5948
5949@item @emph{See also}:
5950@ref{SAME_TYPE_AS}
5951@end table
5952
5953
5954
5955@node FDATE
5956@section @code{FDATE} --- Get the current time as a string
5957@fnindex FDATE
5958@cindex time, current
5959@cindex current time
5960@cindex date, current
5961@cindex current date
5962
5963@table @asis
5964@item @emph{Description}:
5965@code{FDATE(DATE)} returns the current date (using the same format as
5966@ref{CTIME}) in @var{DATE}. It is equivalent to @code{CALL CTIME(DATE,
5967TIME())}.
5968
5969This intrinsic is provided in both subroutine and function forms; however,
5970only one form can be used in any given program unit.
5971
5972@item @emph{Standard}:
5973GNU extension
5974
5975@item @emph{Class}:
5976Subroutine, function
5977
5978@item @emph{Syntax}:
5979@multitable @columnfractions .80
5980@item @code{CALL FDATE(DATE)}.
5981@item @code{DATE = FDATE()}.
5982@end multitable
5983
5984@item @emph{Arguments}:
5985@multitable @columnfractions .15 .70
5986@item @var{DATE}@tab The type shall be of type @code{CHARACTER} of the
5987default kind. It is an @code{INTENT(OUT)} argument.  If the length of
5988this variable is too short for the date and time string to fit
5989completely, it will be blank on procedure return.
5990@end multitable
5991
5992@item @emph{Return value}:
5993The current date and time as a string.
5994
5995@item @emph{Example}:
5996@smallexample
5997program test_fdate
5998    integer(8) :: i, j
5999    character(len=30) :: date
6000    call fdate(date)
6001    print *, 'Program started on ', date
6002    do i = 1, 100000000 ! Just a delay
6003        j = i * i - i
6004    end do
6005    call fdate(date)
6006    print *, 'Program ended on ', date
6007end program test_fdate
6008@end smallexample
6009
6010@item @emph{See also}:
6011@ref{DATE_AND_TIME}, @gol
6012@ref{CTIME}
6013@end table
6014
6015
6016@node FGET
6017@section @code{FGET} --- Read a single character in stream mode from stdin
6018@fnindex FGET
6019@cindex read character, stream mode
6020@cindex stream mode, read character
6021@cindex file operation, read character
6022
6023@table @asis
6024@item @emph{Description}:
6025Read a single character in stream mode from stdin by bypassing normal
6026formatted output. Stream I/O should not be mixed with normal record-oriented
6027(formatted or unformatted) I/O on the same unit; the results are unpredictable.
6028
6029This intrinsic is provided in both subroutine and function forms; however,
6030only one form can be used in any given program unit.
6031
6032Note that the @code{FGET} intrinsic is provided for backwards compatibility with
6033@command{g77}.  GNU Fortran provides the Fortran 2003 Stream facility.
6034Programmers should consider the use of new stream IO feature in new code
6035for future portability. See also @ref{Fortran 2003 status}.
6036
6037@item @emph{Standard}:
6038GNU extension
6039
6040@item @emph{Class}:
6041Subroutine, function
6042
6043@item @emph{Syntax}:
6044@multitable @columnfractions .80
6045@item @code{CALL FGET(C [, STATUS])}
6046@item @code{STATUS = FGET(C)}
6047@end multitable
6048
6049@item @emph{Arguments}:
6050@multitable @columnfractions .15 .70
6051@item @var{C}      @tab The type shall be @code{CHARACTER} and of default
6052kind.
6053@item @var{STATUS} @tab (Optional) status flag of type @code{INTEGER}.
6054Returns 0 on success, -1 on end-of-file, and a system specific positive
6055error code otherwise.
6056@end multitable
6057
6058@item @emph{Example}:
6059@smallexample
6060PROGRAM test_fget
6061  INTEGER, PARAMETER :: strlen = 100
6062  INTEGER :: status, i = 1
6063  CHARACTER(len=strlen) :: str = ""
6064
6065  WRITE (*,*) 'Enter text:'
6066  DO
6067    CALL fget(str(i:i), status)
6068    if (status /= 0 .OR. i > strlen) exit
6069    i = i + 1
6070  END DO
6071  WRITE (*,*) TRIM(str)
6072END PROGRAM
6073@end smallexample
6074
6075@item @emph{See also}:
6076@ref{FGETC}, @gol
6077@ref{FPUT}, @gol
6078@ref{FPUTC}
6079@end table
6080
6081
6082
6083@node FGETC
6084@section @code{FGETC} --- Read a single character in stream mode
6085@fnindex FGETC
6086@cindex read character, stream mode
6087@cindex stream mode, read character
6088@cindex file operation, read character
6089
6090@table @asis
6091@item @emph{Description}:
6092Read a single character in stream mode by bypassing normal formatted output.
6093Stream I/O should not be mixed with normal record-oriented (formatted or
6094unformatted) I/O on the same unit; the results are unpredictable.
6095
6096This intrinsic is provided in both subroutine and function forms; however,
6097only one form can be used in any given program unit.
6098
6099Note that the @code{FGET} intrinsic is provided for backwards compatibility
6100with @command{g77}.  GNU Fortran provides the Fortran 2003 Stream facility.
6101Programmers should consider the use of new stream IO feature in new code
6102for future portability. See also @ref{Fortran 2003 status}.
6103
6104@item @emph{Standard}:
6105GNU extension
6106
6107@item @emph{Class}:
6108Subroutine, function
6109
6110@item @emph{Syntax}:
6111@multitable @columnfractions .80
6112@item @code{CALL FGETC(UNIT, C [, STATUS])}
6113@item @code{STATUS = FGETC(UNIT, C)}
6114@end multitable
6115
6116@item @emph{Arguments}:
6117@multitable @columnfractions .15 .70
6118@item @var{UNIT}   @tab The type shall be @code{INTEGER}.
6119@item @var{C}      @tab The type shall be @code{CHARACTER} and of default
6120kind.
6121@item @var{STATUS} @tab (Optional) status flag of type @code{INTEGER}.
6122Returns 0 on success, -1 on end-of-file and a system specific positive
6123error code otherwise.
6124@end multitable
6125
6126@item @emph{Example}:
6127@smallexample
6128PROGRAM test_fgetc
6129  INTEGER :: fd = 42, status
6130  CHARACTER :: c
6131
6132  OPEN(UNIT=fd, FILE="/etc/passwd", ACTION="READ", STATUS = "OLD")
6133  DO
6134    CALL fgetc(fd, c, status)
6135    IF (status /= 0) EXIT
6136    call fput(c)
6137  END DO
6138  CLOSE(UNIT=fd)
6139END PROGRAM
6140@end smallexample
6141
6142@item @emph{See also}:
6143@ref{FGET}, @gol
6144@ref{FPUT}, @gol
6145@ref{FPUTC}
6146@end table
6147
6148@node FINDLOC
6149@section @code{FINDLOC} --- Search an array for a value
6150@fnindex FINDLOC
6151@cindex findloc
6152
6153@table @asis
6154@item @emph{Description}:
6155Determines the location of the element in the array with the value
6156given in the @var{VALUE} argument, or, if the @var{DIM} argument is
6157supplied, determines the locations of the elements equal to the
6158@var{VALUE} argument element along each
6159row of the array in the @var{DIM} direction.  If @var{MASK} is
6160present, only the elements for which @var{MASK} is @code{.TRUE.} are
6161considered.  If more than one element in the array has the value
6162@var{VALUE}, the location returned is that of the first such element
6163in array element order if the @var{BACK} is not present or if it is
6164@code{.FALSE.}. If @var{BACK} is true, the location returned is that
6165of the last such element. If the array has zero size, or all of the
6166elements of @var{MASK} are @code{.FALSE.}, then the result is an array
6167of zeroes.  Similarly, if @var{DIM} is supplied and all of the
6168elements of @var{MASK} along a given row are zero, the result value
6169for that row is zero.
6170
6171@item @emph{Standard}:
6172Fortran 2008 and later.
6173
6174@item @emph{Class}:
6175Transformational function
6176
6177@item @emph{Syntax}:
6178@multitable @columnfractions .80
6179@item @code{RESULT = FINDLOC(ARRAY, VALUE, DIM [, MASK] [,KIND] [,BACK])}
6180@item @code{RESULT = FINDLOC(ARRAY, VALUE, [, MASK] [,KIND] [,BACK])}
6181@end multitable
6182
6183@item @emph{Arguments}:
6184@multitable @columnfractions .15 .70
6185@item @var{ARRAY} @tab Shall be an array of intrinsic type.
6186@item @var{VALUE} @tab A scalar of intrinsic type which is in type
6187conformance with @var{ARRAY}.
6188@item @var{DIM} @tab (Optional) Shall be a scalar of type
6189@code{INTEGER}, with a value between one and the rank of @var{ARRAY},
6190inclusive.  It may not be an optional dummy argument.
6191@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
6192expression indicating the kind parameter of the result.
6193@item @var{BACK} @tab (Optional) A scalar of type @code{LOGICAL}.
6194@end multitable
6195
6196@item @emph{Return value}:
6197If @var{DIM} is absent, the result is a rank-one array with a length
6198equal to the rank of @var{ARRAY}.  If @var{DIM} is present, the result
6199is an array with a rank one less than the rank of @var{ARRAY}, and a
6200size corresponding to the size of @var{ARRAY} with the @var{DIM}
6201dimension removed.  If @var{DIM} is present and @var{ARRAY} has a rank
6202of one, the result is a scalar.  If the optional argument @var{KIND}
6203is present, the result is an integer of kind @var{KIND}, otherwise it
6204is of default kind.
6205
6206@item @emph{See also}:
6207@ref{MAXLOC}, @gol
6208@ref{MINLOC}
6209
6210@end table
6211
6212@node FLOOR
6213@section @code{FLOOR} --- Integer floor function
6214@fnindex FLOOR
6215@cindex floor
6216@cindex rounding, floor
6217
6218@table @asis
6219@item @emph{Description}:
6220@code{FLOOR(A)} returns the greatest integer less than or equal to @var{X}.
6221
6222@item @emph{Standard}:
6223Fortran 95 and later
6224
6225@item @emph{Class}:
6226Elemental function
6227
6228@item @emph{Syntax}:
6229@code{RESULT = FLOOR(A [, KIND])}
6230
6231@item @emph{Arguments}:
6232@multitable @columnfractions .15 .70
6233@item @var{A} @tab The type shall be @code{REAL}.
6234@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
6235expression indicating the kind parameter of the result.
6236@end multitable
6237
6238@item @emph{Return value}:
6239The return value is of type @code{INTEGER(KIND)} if @var{KIND} is present
6240and of default-kind @code{INTEGER} otherwise.
6241
6242@item @emph{Example}:
6243@smallexample
6244program test_floor
6245    real :: x = 63.29
6246    real :: y = -63.59
6247    print *, floor(x) ! returns 63
6248    print *, floor(y) ! returns -64
6249end program test_floor
6250@end smallexample
6251
6252@item @emph{See also}:
6253@ref{CEILING}, @gol
6254@ref{NINT}
6255@end table
6256
6257
6258
6259@node FLUSH
6260@section @code{FLUSH} --- Flush I/O unit(s)
6261@fnindex FLUSH
6262@cindex file operation, flush
6263
6264@table @asis
6265@item @emph{Description}:
6266Flushes Fortran unit(s) currently open for output. Without the optional
6267argument, all units are flushed, otherwise just the unit specified.
6268
6269@item @emph{Standard}:
6270GNU extension
6271
6272@item @emph{Class}:
6273Subroutine
6274
6275@item @emph{Syntax}:
6276@code{CALL FLUSH(UNIT)}
6277
6278@item @emph{Arguments}:
6279@multitable @columnfractions .15 .70
6280@item @var{UNIT} @tab (Optional) The type shall be @code{INTEGER}.
6281@end multitable
6282
6283@item @emph{Note}:
6284Beginning with the Fortran 2003 standard, there is a @code{FLUSH}
6285statement that should be preferred over the @code{FLUSH} intrinsic.
6286
6287The @code{FLUSH} intrinsic and the Fortran 2003 @code{FLUSH} statement
6288have identical effect: they flush the runtime library's I/O buffer so
6289that the data becomes visible to other processes. This does not guarantee
6290that the data is committed to disk.
6291
6292On POSIX systems, you can request that all data is transferred  to  the
6293storage device by calling the @code{fsync} function, with the POSIX file
6294descriptor of the I/O unit as argument (retrieved with GNU intrinsic
6295@code{FNUM}). The following example shows how:
6296
6297@smallexample
6298  ! Declare the interface for POSIX fsync function
6299  interface
6300    function fsync (fd) bind(c,name="fsync")
6301    use iso_c_binding, only: c_int
6302      integer(c_int), value :: fd
6303      integer(c_int) :: fsync
6304    end function fsync
6305  end interface
6306
6307  ! Variable declaration
6308  integer :: ret
6309
6310  ! Opening unit 10
6311  open (10,file="foo")
6312
6313  ! ...
6314  ! Perform I/O on unit 10
6315  ! ...
6316
6317  ! Flush and sync
6318  flush(10)
6319  ret = fsync(fnum(10))
6320
6321  ! Handle possible error
6322  if (ret /= 0) stop "Error calling FSYNC"
6323@end smallexample
6324
6325@end table
6326
6327
6328
6329@node FNUM
6330@section @code{FNUM} --- File number function
6331@fnindex FNUM
6332@cindex file operation, file number
6333
6334@table @asis
6335@item @emph{Description}:
6336@code{FNUM(UNIT)} returns the POSIX file descriptor number corresponding to the
6337open Fortran I/O unit @code{UNIT}.
6338
6339@item @emph{Standard}:
6340GNU extension
6341
6342@item @emph{Class}:
6343Function
6344
6345@item @emph{Syntax}:
6346@code{RESULT = FNUM(UNIT)}
6347
6348@item @emph{Arguments}:
6349@multitable @columnfractions .15 .70
6350@item @var{UNIT} @tab The type shall be @code{INTEGER}.
6351@end multitable
6352
6353@item @emph{Return value}:
6354The return value is of type @code{INTEGER}
6355
6356@item @emph{Example}:
6357@smallexample
6358program test_fnum
6359  integer :: i
6360  open (unit=10, status = "scratch")
6361  i = fnum(10)
6362  print *, i
6363  close (10)
6364end program test_fnum
6365@end smallexample
6366@end table
6367
6368
6369
6370@node FPUT
6371@section @code{FPUT} --- Write a single character in stream mode to stdout
6372@fnindex FPUT
6373@cindex write character, stream mode
6374@cindex stream mode, write character
6375@cindex file operation, write character
6376
6377@table @asis
6378@item @emph{Description}:
6379Write a single character in stream mode to stdout by bypassing normal
6380formatted output. Stream I/O should not be mixed with normal record-oriented
6381(formatted or unformatted) I/O on the same unit; the results are unpredictable.
6382
6383This intrinsic is provided in both subroutine and function forms; however,
6384only one form can be used in any given program unit.
6385
6386Note that the @code{FGET} intrinsic is provided for backwards compatibility with
6387@command{g77}.  GNU Fortran provides the Fortran 2003 Stream facility.
6388Programmers should consider the use of new stream IO feature in new code
6389for future portability. See also @ref{Fortran 2003 status}.
6390
6391@item @emph{Standard}:
6392GNU extension
6393
6394@item @emph{Class}:
6395Subroutine, function
6396
6397@item @emph{Syntax}:
6398@multitable @columnfractions .80
6399@item @code{CALL FPUT(C [, STATUS])}
6400@item @code{STATUS = FPUT(C)}
6401@end multitable
6402
6403@item @emph{Arguments}:
6404@multitable @columnfractions .15 .70
6405@item @var{C}      @tab The type shall be @code{CHARACTER} and of default
6406kind.
6407@item @var{STATUS} @tab (Optional) status flag of type @code{INTEGER}.
6408Returns 0 on success, -1 on end-of-file and a system specific positive
6409error code otherwise.
6410@end multitable
6411
6412@item @emph{Example}:
6413@smallexample
6414PROGRAM test_fput
6415  CHARACTER(len=10) :: str = "gfortran"
6416  INTEGER :: i
6417  DO i = 1, len_trim(str)
6418    CALL fput(str(i:i))
6419  END DO
6420END PROGRAM
6421@end smallexample
6422
6423@item @emph{See also}:
6424@ref{FPUTC}, @gol
6425@ref{FGET}, @gol
6426@ref{FGETC}
6427@end table
6428
6429
6430
6431@node FPUTC
6432@section @code{FPUTC} --- Write a single character in stream mode
6433@fnindex FPUTC
6434@cindex write character, stream mode
6435@cindex stream mode, write character
6436@cindex file operation, write character
6437
6438@table @asis
6439@item @emph{Description}:
6440Write a single character in stream mode by bypassing normal formatted
6441output. Stream I/O should not be mixed with normal record-oriented
6442(formatted or unformatted) I/O on the same unit; the results are unpredictable.
6443
6444This intrinsic is provided in both subroutine and function forms; however,
6445only one form can be used in any given program unit.
6446
6447Note that the @code{FGET} intrinsic is provided for backwards compatibility with
6448@command{g77}.  GNU Fortran provides the Fortran 2003 Stream facility.
6449Programmers should consider the use of new stream IO feature in new code
6450for future portability. See also @ref{Fortran 2003 status}.
6451
6452@item @emph{Standard}:
6453GNU extension
6454
6455@item @emph{Class}:
6456Subroutine, function
6457
6458@item @emph{Syntax}:
6459@multitable @columnfractions .80
6460@item @code{CALL FPUTC(UNIT, C [, STATUS])}
6461@item @code{STATUS = FPUTC(UNIT, C)}
6462@end multitable
6463
6464@item @emph{Arguments}:
6465@multitable @columnfractions .15 .70
6466@item @var{UNIT}   @tab The type shall be @code{INTEGER}.
6467@item @var{C}      @tab The type shall be @code{CHARACTER} and of default
6468kind.
6469@item @var{STATUS} @tab (Optional) status flag of type @code{INTEGER}.
6470Returns 0 on success, -1 on end-of-file and a system specific positive
6471error code otherwise.
6472@end multitable
6473
6474@item @emph{Example}:
6475@smallexample
6476PROGRAM test_fputc
6477  CHARACTER(len=10) :: str = "gfortran"
6478  INTEGER :: fd = 42, i
6479
6480  OPEN(UNIT = fd, FILE = "out", ACTION = "WRITE", STATUS="NEW")
6481  DO i = 1, len_trim(str)
6482    CALL fputc(fd, str(i:i))
6483  END DO
6484  CLOSE(fd)
6485END PROGRAM
6486@end smallexample
6487
6488@item @emph{See also}:
6489@ref{FPUT}, @gol
6490@ref{FGET}, @gol
6491@ref{FGETC}
6492@end table
6493
6494
6495
6496@node FRACTION
6497@section @code{FRACTION} --- Fractional part of the model representation
6498@fnindex FRACTION
6499@cindex real number, fraction
6500@cindex floating point, fraction
6501
6502@table @asis
6503@item @emph{Description}:
6504@code{FRACTION(X)} returns the fractional part of the model
6505representation of @code{X}.
6506
6507@item @emph{Standard}:
6508Fortran 90 and later
6509
6510@item @emph{Class}:
6511Elemental function
6512
6513@item @emph{Syntax}:
6514@code{Y = FRACTION(X)}
6515
6516@item @emph{Arguments}:
6517@multitable @columnfractions .15 .70
6518@item @var{X} @tab The type of the argument shall be a @code{REAL}.
6519@end multitable
6520
6521@item @emph{Return value}:
6522The return value is of the same type and kind as the argument.
6523The fractional part of the model representation of @code{X} is returned;
6524it is @code{X * RADIX(X)**(-EXPONENT(X))}.
6525
6526@item @emph{Example}:
6527@smallexample
6528program test_fraction
6529  real :: x
6530  x = 178.1387e-4
6531  print *, fraction(x), x * radix(x)**(-exponent(x))
6532end program test_fraction
6533@end smallexample
6534
6535@end table
6536
6537
6538
6539@node FREE
6540@section @code{FREE} --- Frees memory
6541@fnindex FREE
6542@cindex pointer, cray
6543
6544@table @asis
6545@item @emph{Description}:
6546Frees memory previously allocated by @code{MALLOC}. The @code{FREE}
6547intrinsic is an extension intended to be used with Cray pointers, and is
6548provided in GNU Fortran to allow user to compile legacy code. For
6549new code using Fortran 95 pointers, the memory de-allocation intrinsic is
6550@code{DEALLOCATE}.
6551
6552@item @emph{Standard}:
6553GNU extension
6554
6555@item @emph{Class}:
6556Subroutine
6557
6558@item @emph{Syntax}:
6559@code{CALL FREE(PTR)}
6560
6561@item @emph{Arguments}:
6562@multitable @columnfractions .15 .70
6563@item @var{PTR} @tab The type shall be @code{INTEGER}. It represents the
6564location of the memory that should be de-allocated.
6565@end multitable
6566
6567@item @emph{Return value}:
6568None
6569
6570@item @emph{Example}:
6571See @code{MALLOC} for an example.
6572
6573@item @emph{See also}:
6574@ref{MALLOC}
6575@end table
6576
6577
6578
6579@node FSEEK
6580@section @code{FSEEK} --- Low level file positioning subroutine
6581@fnindex FSEEK
6582@cindex file operation, seek
6583@cindex file operation, position
6584
6585@table @asis
6586@item @emph{Description}:
6587Moves @var{UNIT} to the specified @var{OFFSET}. If @var{WHENCE}
6588is set to 0, the @var{OFFSET} is taken as an absolute value @code{SEEK_SET},
6589if set to 1, @var{OFFSET} is taken to be relative to the current position
6590@code{SEEK_CUR}, and if set to 2 relative to the end of the file @code{SEEK_END}.
6591On error, @var{STATUS} is set to a nonzero value. If @var{STATUS} the seek
6592fails silently.
6593
6594This intrinsic routine is not fully backwards compatible with @command{g77}.
6595In @command{g77}, the @code{FSEEK} takes a statement label instead of a
6596@var{STATUS} variable. If FSEEK is used in old code, change
6597@smallexample
6598  CALL FSEEK(UNIT, OFFSET, WHENCE, *label)
6599@end smallexample
6600to
6601@smallexample
6602  INTEGER :: status
6603  CALL FSEEK(UNIT, OFFSET, WHENCE, status)
6604  IF (status /= 0) GOTO label
6605@end smallexample
6606
6607Please note that GNU Fortran provides the Fortran 2003 Stream facility.
6608Programmers should consider the use of new stream IO feature in new code
6609for future portability. See also @ref{Fortran 2003 status}.
6610
6611@item @emph{Standard}:
6612GNU extension
6613
6614@item @emph{Class}:
6615Subroutine
6616
6617@item @emph{Syntax}:
6618@code{CALL FSEEK(UNIT, OFFSET, WHENCE[, STATUS])}
6619
6620@item @emph{Arguments}:
6621@multitable @columnfractions .15 .70
6622@item @var{UNIT}   @tab Shall be a scalar of type @code{INTEGER}.
6623@item @var{OFFSET} @tab Shall be a scalar of type @code{INTEGER}.
6624@item @var{WHENCE} @tab Shall be a scalar of type @code{INTEGER}.
6625Its value shall be either 0, 1 or 2.
6626@item @var{STATUS} @tab (Optional) shall be a scalar of type
6627@code{INTEGER(4)}.
6628@end multitable
6629
6630@item @emph{Example}:
6631@smallexample
6632PROGRAM test_fseek
6633  INTEGER, PARAMETER :: SEEK_SET = 0, SEEK_CUR = 1, SEEK_END = 2
6634  INTEGER :: fd, offset, ierr
6635
6636  ierr   = 0
6637  offset = 5
6638  fd     = 10
6639
6640  OPEN(UNIT=fd, FILE="fseek.test")
6641  CALL FSEEK(fd, offset, SEEK_SET, ierr)  ! move to OFFSET
6642  print *, FTELL(fd), ierr
6643
6644  CALL FSEEK(fd, 0, SEEK_END, ierr)       ! move to end
6645  print *, FTELL(fd), ierr
6646
6647  CALL FSEEK(fd, 0, SEEK_SET, ierr)       ! move to beginning
6648  print *, FTELL(fd), ierr
6649
6650  CLOSE(UNIT=fd)
6651END PROGRAM
6652@end smallexample
6653
6654@item @emph{See also}:
6655@ref{FTELL}
6656@end table
6657
6658
6659
6660@node FSTAT
6661@section @code{FSTAT} --- Get file status
6662@fnindex FSTAT
6663@cindex file system, file status
6664
6665@table @asis
6666@item @emph{Description}:
6667@code{FSTAT} is identical to @ref{STAT}, except that information about an
6668already opened file is obtained.
6669
6670The elements in @code{VALUES} are the same as described by @ref{STAT}.
6671
6672This intrinsic is provided in both subroutine and function forms; however,
6673only one form can be used in any given program unit.
6674
6675@item @emph{Standard}:
6676GNU extension
6677
6678@item @emph{Class}:
6679Subroutine, function
6680
6681@item @emph{Syntax}:
6682@multitable @columnfractions .80
6683@item @code{CALL FSTAT(UNIT, VALUES [, STATUS])}
6684@item @code{STATUS = FSTAT(UNIT, VALUES)}
6685@end multitable
6686
6687@item @emph{Arguments}:
6688@multitable @columnfractions .15 .70
6689@item @var{UNIT}   @tab An open I/O unit number of type @code{INTEGER}.
6690@item @var{VALUES} @tab The type shall be @code{INTEGER(4), DIMENSION(13)}.
6691@item @var{STATUS} @tab (Optional) status flag of type @code{INTEGER(4)}. Returns 0
6692on success and a system specific error code otherwise.
6693@end multitable
6694
6695@item @emph{Example}:
6696See @ref{STAT} for an example.
6697
6698@item @emph{See also}:
6699To stat a link: @gol
6700@ref{LSTAT} @gol
6701To stat a file: @gol
6702@ref{STAT}
6703@end table
6704
6705
6706
6707@node FTELL
6708@section @code{FTELL} --- Current stream position
6709@fnindex FTELL
6710@cindex file operation, position
6711
6712@table @asis
6713@item @emph{Description}:
6714Retrieves the current position within an open file.
6715
6716This intrinsic is provided in both subroutine and function forms; however,
6717only one form can be used in any given program unit.
6718
6719@item @emph{Standard}:
6720GNU extension
6721
6722@item @emph{Class}:
6723Subroutine, function
6724
6725@item @emph{Syntax}:
6726@multitable @columnfractions .80
6727@item @code{CALL FTELL(UNIT, OFFSET)}
6728@item @code{OFFSET = FTELL(UNIT)}
6729@end multitable
6730
6731@item @emph{Arguments}:
6732@multitable @columnfractions .15 .70
6733@item @var{OFFSET}  @tab Shall of type @code{INTEGER}.
6734@item @var{UNIT}    @tab Shall of type @code{INTEGER}.
6735@end multitable
6736
6737@item @emph{Return value}:
6738In either syntax, @var{OFFSET} is set to the current offset of unit
6739number @var{UNIT}, or to @math{-1} if the unit is not currently open.
6740
6741@item @emph{Example}:
6742@smallexample
6743PROGRAM test_ftell
6744  INTEGER :: i
6745  OPEN(10, FILE="temp.dat")
6746  CALL ftell(10,i)
6747  WRITE(*,*) i
6748END PROGRAM
6749@end smallexample
6750
6751@item @emph{See also}:
6752@ref{FSEEK}
6753@end table
6754
6755
6756
6757@node GAMMA
6758@section @code{GAMMA} --- Gamma function
6759@fnindex GAMMA
6760@fnindex DGAMMA
6761@cindex Gamma function
6762@cindex Factorial function
6763
6764@table @asis
6765@item @emph{Description}:
6766@code{GAMMA(X)} computes Gamma (@math{\Gamma}) of @var{X}. For positive,
6767integer values of @var{X} the Gamma function simplifies to the factorial
6768function @math{\Gamma(x)=(x-1)!}.
6769
6770@tex
6771$$
6772\Gamma(x) = \int_0^\infty t^{x-1}{\rm e}^{-t}\,{\rm d}t
6773$$
6774@end tex
6775
6776@item @emph{Standard}:
6777Fortran 2008 and later
6778
6779@item @emph{Class}:
6780Elemental function
6781
6782@item @emph{Syntax}:
6783@code{X = GAMMA(X)}
6784
6785@item @emph{Arguments}:
6786@multitable @columnfractions .15 .70
6787@item @var{X} @tab Shall be of type @code{REAL} and neither zero
6788nor a negative integer.
6789@end multitable
6790
6791@item @emph{Return value}:
6792The return value is of type @code{REAL} of the same kind as @var{X}.
6793
6794@item @emph{Example}:
6795@smallexample
6796program test_gamma
6797  real :: x = 1.0
6798  x = gamma(x) ! returns 1.0
6799end program test_gamma
6800@end smallexample
6801
6802@item @emph{Specific names}:
6803@multitable @columnfractions .20 .20 .20 .25
6804@item Name             @tab Argument         @tab Return type       @tab Standard
6805@item @code{DGAMMA(X)} @tab @code{REAL(8) X} @tab @code{REAL(8)}    @tab GNU extension
6806@end multitable
6807
6808@item @emph{See also}:
6809Logarithm of the Gamma function: @gol
6810@ref{LOG_GAMMA}
6811@end table
6812
6813
6814
6815@node GERROR
6816@section @code{GERROR} --- Get last system error message
6817@fnindex GERROR
6818@cindex system, error handling
6819
6820@table @asis
6821@item @emph{Description}:
6822Returns the system error message corresponding to the last system error.
6823This resembles the functionality of @code{strerror(3)} in C.
6824
6825@item @emph{Standard}:
6826GNU extension
6827
6828@item @emph{Class}:
6829Subroutine
6830
6831@item @emph{Syntax}:
6832@code{CALL GERROR(RESULT)}
6833
6834@item @emph{Arguments}:
6835@multitable @columnfractions .15 .70
6836@item @var{RESULT}  @tab Shall of type @code{CHARACTER} and of default
6837@end multitable
6838
6839@item @emph{Example}:
6840@smallexample
6841PROGRAM test_gerror
6842  CHARACTER(len=100) :: msg
6843  CALL gerror(msg)
6844  WRITE(*,*) msg
6845END PROGRAM
6846@end smallexample
6847
6848@item @emph{See also}:
6849@ref{IERRNO}, @gol
6850@ref{PERROR}
6851@end table
6852
6853
6854
6855@node GETARG
6856@section @code{GETARG} --- Get command line arguments
6857@fnindex GETARG
6858@cindex command-line arguments
6859@cindex arguments, to program
6860
6861@table @asis
6862@item @emph{Description}:
6863Retrieve the @var{POS}-th argument that was passed on the
6864command line when the containing program was invoked.
6865
6866This intrinsic routine is provided for backwards compatibility with
6867GNU Fortran 77.  In new code, programmers should consider the use of
6868the @ref{GET_COMMAND_ARGUMENT} intrinsic defined by the Fortran 2003
6869standard.
6870
6871@item @emph{Standard}:
6872GNU extension
6873
6874@item @emph{Class}:
6875Subroutine
6876
6877@item @emph{Syntax}:
6878@code{CALL GETARG(POS, VALUE)}
6879
6880@item @emph{Arguments}:
6881@multitable @columnfractions .15 .70
6882@item @var{POS}   @tab Shall be of type @code{INTEGER} and not wider than
6883the default integer kind; @math{@var{POS} \geq 0}
6884@item @var{VALUE} @tab Shall be of type @code{CHARACTER} and of default
6885kind.
6886@item @var{VALUE} @tab Shall be of type @code{CHARACTER}.
6887@end multitable
6888
6889@item @emph{Return value}:
6890After @code{GETARG} returns, the @var{VALUE} argument holds the
6891@var{POS}th command line argument. If @var{VALUE} cannot hold the
6892argument, it is truncated to fit the length of @var{VALUE}. If there are
6893less than @var{POS} arguments specified at the command line, @var{VALUE}
6894will be filled with blanks. If @math{@var{POS} = 0}, @var{VALUE} is set
6895to the name of the program (on systems that support this feature).
6896
6897@item @emph{Example}:
6898@smallexample
6899PROGRAM test_getarg
6900  INTEGER :: i
6901  CHARACTER(len=32) :: arg
6902
6903  DO i = 1, iargc()
6904    CALL getarg(i, arg)
6905    WRITE (*,*) arg
6906  END DO
6907END PROGRAM
6908@end smallexample
6909
6910@item @emph{See also}:
6911GNU Fortran 77 compatibility function: @gol
6912@ref{IARGC} @gol
6913Fortran 2003 functions and subroutines: @gol
6914@ref{GET_COMMAND}, @gol
6915@ref{GET_COMMAND_ARGUMENT}, @gol
6916@ref{COMMAND_ARGUMENT_COUNT}
6917@end table
6918
6919
6920
6921@node GET_COMMAND
6922@section @code{GET_COMMAND} --- Get the entire command line
6923@fnindex GET_COMMAND
6924@cindex command-line arguments
6925@cindex arguments, to program
6926
6927@table @asis
6928@item @emph{Description}:
6929Retrieve the entire command line that was used to invoke the program.
6930
6931@item @emph{Standard}:
6932Fortran 2003 and later
6933
6934@item @emph{Class}:
6935Subroutine
6936
6937@item @emph{Syntax}:
6938@code{CALL GET_COMMAND([COMMAND, LENGTH, STATUS])}
6939
6940@item @emph{Arguments}:
6941@multitable @columnfractions .15 .70
6942@item @var{COMMAND} @tab (Optional) shall be of type @code{CHARACTER} and
6943of default kind.
6944@item @var{LENGTH} @tab (Optional) Shall be of type @code{INTEGER} and of
6945default kind.
6946@item @var{STATUS} @tab (Optional) Shall be of type @code{INTEGER} and of
6947default kind.
6948@end multitable
6949
6950@item @emph{Return value}:
6951If @var{COMMAND} is present, stores the entire command line that was used
6952to invoke the program in @var{COMMAND}. If @var{LENGTH} is present, it is
6953assigned the length of the command line. If @var{STATUS} is present, it
6954is assigned 0 upon success of the command, -1 if @var{COMMAND} is too
6955short to store the command line, or a positive value in case of an error.
6956
6957@item @emph{Example}:
6958@smallexample
6959PROGRAM test_get_command
6960  CHARACTER(len=255) :: cmd
6961  CALL get_command(cmd)
6962  WRITE (*,*) TRIM(cmd)
6963END PROGRAM
6964@end smallexample
6965
6966@item @emph{See also}:
6967@ref{GET_COMMAND_ARGUMENT}, @gol
6968@ref{COMMAND_ARGUMENT_COUNT}
6969@end table
6970
6971
6972
6973@node GET_COMMAND_ARGUMENT
6974@section @code{GET_COMMAND_ARGUMENT} --- Get command line arguments
6975@fnindex GET_COMMAND_ARGUMENT
6976@cindex command-line arguments
6977@cindex arguments, to program
6978
6979@table @asis
6980@item @emph{Description}:
6981Retrieve the @var{NUMBER}-th argument that was passed on the
6982command line when the containing program was invoked.
6983
6984@item @emph{Standard}:
6985Fortran 2003 and later
6986
6987@item @emph{Class}:
6988Subroutine
6989
6990@item @emph{Syntax}:
6991@code{CALL GET_COMMAND_ARGUMENT(NUMBER [, VALUE, LENGTH, STATUS])}
6992
6993@item @emph{Arguments}:
6994@multitable @columnfractions .15 .70
6995@item @var{NUMBER} @tab Shall be a scalar of type @code{INTEGER} and of
6996default kind, @math{@var{NUMBER} \geq 0}
6997@item @var{VALUE}  @tab (Optional) Shall be a scalar of type @code{CHARACTER}
6998and of default kind.
6999@item @var{LENGTH} @tab (Optional) Shall be a scalar of type @code{INTEGER}
7000and of default kind.
7001@item @var{STATUS} @tab (Optional) Shall be a scalar of type @code{INTEGER}
7002and of default kind.
7003@end multitable
7004
7005@item @emph{Return value}:
7006After @code{GET_COMMAND_ARGUMENT} returns, the @var{VALUE} argument holds the
7007@var{NUMBER}-th command line argument. If @var{VALUE} cannot hold the argument, it is
7008truncated to fit the length of @var{VALUE}. If there are less than @var{NUMBER}
7009arguments specified at the command line, @var{VALUE} will be filled with blanks.
7010If @math{@var{NUMBER} = 0}, @var{VALUE} is set to the name of the program (on
7011systems that support this feature). The @var{LENGTH} argument contains the
7012length of the @var{NUMBER}-th command line argument. If the argument retrieval
7013fails, @var{STATUS} is a positive number; if @var{VALUE} contains a truncated
7014command line argument, @var{STATUS} is -1; and otherwise the @var{STATUS} is
7015zero.
7016
7017@item @emph{Example}:
7018@smallexample
7019PROGRAM test_get_command_argument
7020  INTEGER :: i
7021  CHARACTER(len=32) :: arg
7022
7023  i = 0
7024  DO
7025    CALL get_command_argument(i, arg)
7026    IF (LEN_TRIM(arg) == 0) EXIT
7027
7028    WRITE (*,*) TRIM(arg)
7029    i = i+1
7030  END DO
7031END PROGRAM
7032@end smallexample
7033
7034@item @emph{See also}:
7035@ref{GET_COMMAND}, @gol
7036@ref{COMMAND_ARGUMENT_COUNT}
7037@end table
7038
7039
7040
7041@node GETCWD
7042@section @code{GETCWD} --- Get current working directory
7043@fnindex GETCWD
7044@cindex system, working directory
7045
7046@table @asis
7047@item @emph{Description}:
7048Get current working directory.
7049
7050This intrinsic is provided in both subroutine and function forms; however,
7051only one form can be used in any given program unit.
7052
7053@item @emph{Standard}:
7054GNU extension
7055
7056@item @emph{Class}:
7057Subroutine, function
7058
7059@item @emph{Syntax}:
7060@multitable @columnfractions .80
7061@item @code{CALL GETCWD(C [, STATUS])}
7062@item @code{STATUS = GETCWD(C)}
7063@end multitable
7064
7065@item @emph{Arguments}:
7066@multitable @columnfractions .15 .70
7067@item @var{C} @tab The type shall be @code{CHARACTER} and of default kind.
7068@item @var{STATUS} @tab (Optional) status flag. Returns 0 on success,
7069a system specific and nonzero error code otherwise.
7070@end multitable
7071
7072@item @emph{Example}:
7073@smallexample
7074PROGRAM test_getcwd
7075  CHARACTER(len=255) :: cwd
7076  CALL getcwd(cwd)
7077  WRITE(*,*) TRIM(cwd)
7078END PROGRAM
7079@end smallexample
7080
7081@item @emph{See also}:
7082@ref{CHDIR}
7083@end table
7084
7085
7086
7087@node GETENV
7088@section @code{GETENV} --- Get an environmental variable
7089@fnindex GETENV
7090@cindex environment variable
7091
7092@table @asis
7093@item @emph{Description}:
7094Get the @var{VALUE} of the environmental variable @var{NAME}.
7095
7096This intrinsic routine is provided for backwards compatibility with
7097GNU Fortran 77.  In new code, programmers should consider the use of
7098the @ref{GET_ENVIRONMENT_VARIABLE} intrinsic defined by the Fortran
70992003 standard.
7100
7101Note that @code{GETENV} need not be thread-safe. It is the
7102responsibility of the user to ensure that the environment is not being
7103updated concurrently with a call to the @code{GETENV} intrinsic.
7104
7105@item @emph{Standard}:
7106GNU extension
7107
7108@item @emph{Class}:
7109Subroutine
7110
7111@item @emph{Syntax}:
7112@code{CALL GETENV(NAME, VALUE)}
7113
7114@item @emph{Arguments}:
7115@multitable @columnfractions .15 .70
7116@item @var{NAME}  @tab Shall be of type @code{CHARACTER} and of default kind.
7117@item @var{VALUE} @tab Shall be of type @code{CHARACTER} and of default kind.
7118@end multitable
7119
7120@item @emph{Return value}:
7121Stores the value of @var{NAME} in @var{VALUE}. If @var{VALUE} is
7122not large enough to hold the data, it is truncated. If @var{NAME}
7123is not set, @var{VALUE} will be filled with blanks.
7124
7125@item @emph{Example}:
7126@smallexample
7127PROGRAM test_getenv
7128  CHARACTER(len=255) :: homedir
7129  CALL getenv("HOME", homedir)
7130  WRITE (*,*) TRIM(homedir)
7131END PROGRAM
7132@end smallexample
7133
7134@item @emph{See also}:
7135@ref{GET_ENVIRONMENT_VARIABLE}
7136@end table
7137
7138
7139
7140@node GET_ENVIRONMENT_VARIABLE
7141@section @code{GET_ENVIRONMENT_VARIABLE} --- Get an environmental variable
7142@fnindex GET_ENVIRONMENT_VARIABLE
7143@cindex environment variable
7144
7145@table @asis
7146@item @emph{Description}:
7147Get the @var{VALUE} of the environmental variable @var{NAME}.
7148
7149Note that @code{GET_ENVIRONMENT_VARIABLE} need not be thread-safe. It
7150is the responsibility of the user to ensure that the environment is
7151not being updated concurrently with a call to the
7152@code{GET_ENVIRONMENT_VARIABLE} intrinsic.
7153
7154@item @emph{Standard}:
7155Fortran 2003 and later
7156
7157@item @emph{Class}:
7158Subroutine
7159
7160@item @emph{Syntax}:
7161@code{CALL GET_ENVIRONMENT_VARIABLE(NAME[, VALUE, LENGTH, STATUS, TRIM_NAME)}
7162
7163@item @emph{Arguments}:
7164@multitable @columnfractions .15 .70
7165@item @var{NAME}      @tab Shall be a scalar of type @code{CHARACTER}
7166and of default kind.
7167@item @var{VALUE}     @tab (Optional) Shall be a scalar of type @code{CHARACTER}
7168and of default kind.
7169@item @var{LENGTH}    @tab (Optional) Shall be a scalar of type @code{INTEGER}
7170and of default kind.
7171@item @var{STATUS}    @tab (Optional) Shall be a scalar of type @code{INTEGER}
7172and of default kind.
7173@item @var{TRIM_NAME} @tab (Optional) Shall be a scalar of type @code{LOGICAL}
7174and of default kind.
7175@end multitable
7176
7177@item @emph{Return value}:
7178Stores the value of @var{NAME} in @var{VALUE}. If @var{VALUE} is
7179not large enough to hold the data, it is truncated. If @var{NAME}
7180is not set, @var{VALUE} will be filled with blanks. Argument @var{LENGTH}
7181contains the length needed for storing the environment variable @var{NAME}
7182or zero if it is not present. @var{STATUS} is -1 if @var{VALUE} is present
7183but too short for the environment variable; it is 1 if the environment
7184variable does not exist and 2 if the processor does not support environment
7185variables; in all other cases @var{STATUS} is zero. If @var{TRIM_NAME} is
7186present with the value @code{.FALSE.}, the trailing blanks in @var{NAME}
7187are significant; otherwise they are not part of the environment variable
7188name.
7189
7190@item @emph{Example}:
7191@smallexample
7192PROGRAM test_getenv
7193  CHARACTER(len=255) :: homedir
7194  CALL get_environment_variable("HOME", homedir)
7195  WRITE (*,*) TRIM(homedir)
7196END PROGRAM
7197@end smallexample
7198@end table
7199
7200
7201
7202@node GETGID
7203@section @code{GETGID} --- Group ID function
7204@fnindex GETGID
7205@cindex system, group ID
7206
7207@table @asis
7208@item @emph{Description}:
7209Returns the numerical group ID of the current process.
7210
7211@item @emph{Standard}:
7212GNU extension
7213
7214@item @emph{Class}:
7215Function
7216
7217@item @emph{Syntax}:
7218@code{RESULT = GETGID()}
7219
7220@item @emph{Return value}:
7221The return value of @code{GETGID} is an @code{INTEGER} of the default
7222kind.
7223
7224
7225@item @emph{Example}:
7226See @code{GETPID} for an example.
7227
7228@item @emph{See also}:
7229@ref{GETPID}, @gol
7230@ref{GETUID}
7231@end table
7232
7233
7234
7235@node GETLOG
7236@section @code{GETLOG} --- Get login name
7237@fnindex GETLOG
7238@cindex system, login name
7239@cindex login name
7240
7241@table @asis
7242@item @emph{Description}:
7243Gets the username under which the program is running.
7244
7245@item @emph{Standard}:
7246GNU extension
7247
7248@item @emph{Class}:
7249Subroutine
7250
7251@item @emph{Syntax}:
7252@code{CALL GETLOG(C)}
7253
7254@item @emph{Arguments}:
7255@multitable @columnfractions .15 .70
7256@item @var{C} @tab Shall be of type @code{CHARACTER} and of default kind.
7257@end multitable
7258
7259@item @emph{Return value}:
7260Stores the current user name in @var{LOGIN}.  (On systems where POSIX
7261functions @code{geteuid} and @code{getpwuid} are not available, and
7262the @code{getlogin} function is not implemented either, this will
7263return a blank string.)
7264
7265@item @emph{Example}:
7266@smallexample
7267PROGRAM TEST_GETLOG
7268  CHARACTER(32) :: login
7269  CALL GETLOG(login)
7270  WRITE(*,*) login
7271END PROGRAM
7272@end smallexample
7273
7274@item @emph{See also}:
7275@ref{GETUID}
7276@end table
7277
7278
7279
7280@node GETPID
7281@section @code{GETPID} --- Process ID function
7282@fnindex GETPID
7283@cindex system, process ID
7284@cindex process ID
7285
7286@table @asis
7287@item @emph{Description}:
7288Returns the numerical process identifier of the current process.
7289
7290@item @emph{Standard}:
7291GNU extension
7292
7293@item @emph{Class}:
7294Function
7295
7296@item @emph{Syntax}:
7297@code{RESULT = GETPID()}
7298
7299@item @emph{Return value}:
7300The return value of @code{GETPID} is an @code{INTEGER} of the default
7301kind.
7302
7303
7304@item @emph{Example}:
7305@smallexample
7306program info
7307  print *, "The current process ID is ", getpid()
7308  print *, "Your numerical user ID is ", getuid()
7309  print *, "Your numerical group ID is ", getgid()
7310end program info
7311@end smallexample
7312
7313@item @emph{See also}:
7314@ref{GETGID}, @gol
7315@ref{GETUID}
7316@end table
7317
7318
7319
7320@node GETUID
7321@section @code{GETUID} --- User ID function
7322@fnindex GETUID
7323@cindex system, user ID
7324@cindex user id
7325
7326@table @asis
7327@item @emph{Description}:
7328Returns the numerical user ID of the current process.
7329
7330@item @emph{Standard}:
7331GNU extension
7332
7333@item @emph{Class}:
7334Function
7335
7336@item @emph{Syntax}:
7337@code{RESULT = GETUID()}
7338
7339@item @emph{Return value}:
7340The return value of @code{GETUID} is an @code{INTEGER} of the default
7341kind.
7342
7343
7344@item @emph{Example}:
7345See @code{GETPID} for an example.
7346
7347@item @emph{See also}:
7348@ref{GETPID}, @gol
7349@ref{GETLOG}
7350@end table
7351
7352
7353
7354@node GMTIME
7355@section @code{GMTIME} --- Convert time to GMT info
7356@fnindex GMTIME
7357@cindex time, conversion to GMT info
7358
7359@table @asis
7360@item @emph{Description}:
7361Given a system time value @var{TIME} (as provided by the @ref{TIME}
7362intrinsic), fills @var{VALUES} with values extracted from it appropriate
7363to the UTC time zone (Universal Coordinated Time, also known in some
7364countries as GMT, Greenwich Mean Time), using @code{gmtime(3)}.
7365
7366This intrinsic routine is provided for backwards compatibility with
7367GNU Fortran 77.  In new code, programmers should consider the use of
7368the @ref{DATE_AND_TIME} intrinsic defined by the Fortran 95
7369standard.
7370
7371@item @emph{Standard}:
7372GNU extension
7373
7374@item @emph{Class}:
7375Subroutine
7376
7377@item @emph{Syntax}:
7378@code{CALL GMTIME(TIME, VALUES)}
7379
7380@item @emph{Arguments}:
7381@multitable @columnfractions .15 .70
7382@item @var{TIME}   @tab An @code{INTEGER} scalar expression
7383corresponding to a system time, with @code{INTENT(IN)}.
7384@item @var{VALUES} @tab A default @code{INTEGER} array with 9 elements,
7385with @code{INTENT(OUT)}.
7386@end multitable
7387
7388@item @emph{Return value}:
7389The elements of @var{VALUES} are assigned as follows:
7390@enumerate
7391@item Seconds after the minute, range 0--59 or 0--61 to allow for leap
7392seconds
7393@item Minutes after the hour, range 0--59
7394@item Hours past midnight, range 0--23
7395@item Day of month, range 1--31
7396@item Number of months since January, range 0--11
7397@item Years since 1900
7398@item Number of days since Sunday, range 0--6
7399@item Days since January 1, range 0--365
7400@item Daylight savings indicator: positive if daylight savings is in
7401effect, zero if not, and negative if the information is not available.
7402@end enumerate
7403
7404@item @emph{See also}:
7405@ref{DATE_AND_TIME}, @gol
7406@ref{CTIME}, @gol
7407@ref{LTIME}, @gol
7408@ref{TIME}, @gol
7409@ref{TIME8}
7410@end table
7411
7412
7413
7414@node HOSTNM
7415@section @code{HOSTNM} --- Get system host name
7416@fnindex HOSTNM
7417@cindex system, host name
7418
7419@table @asis
7420@item @emph{Description}:
7421Retrieves the host name of the system on which the program is running.
7422
7423This intrinsic is provided in both subroutine and function forms; however,
7424only one form can be used in any given program unit.
7425
7426@item @emph{Standard}:
7427GNU extension
7428
7429@item @emph{Class}:
7430Subroutine, function
7431
7432@item @emph{Syntax}:
7433@multitable @columnfractions .80
7434@item @code{CALL HOSTNM(C [, STATUS])}
7435@item @code{STATUS = HOSTNM(NAME)}
7436@end multitable
7437
7438@item @emph{Arguments}:
7439@multitable @columnfractions .15 .70
7440@item @var{C}    @tab Shall of type @code{CHARACTER} and of default kind.
7441@item @var{STATUS}  @tab (Optional) status flag of type @code{INTEGER}.
7442Returns 0 on success, or a system specific error code otherwise.
7443@end multitable
7444
7445@item @emph{Return value}:
7446In either syntax, @var{NAME} is set to the current hostname if it can
7447be obtained, or to a blank string otherwise.
7448
7449@end table
7450
7451
7452
7453@node HUGE
7454@section @code{HUGE} --- Largest number of a kind
7455@fnindex HUGE
7456@cindex limits, largest number
7457@cindex model representation, largest number
7458
7459@table @asis
7460@item @emph{Description}:
7461@code{HUGE(X)} returns the largest number that is not an infinity in
7462the model of the type of @code{X}.
7463
7464@item @emph{Standard}:
7465Fortran 90 and later
7466
7467@item @emph{Class}:
7468Inquiry function
7469
7470@item @emph{Syntax}:
7471@code{RESULT = HUGE(X)}
7472
7473@item @emph{Arguments}:
7474@multitable @columnfractions .15 .70
7475@item @var{X} @tab Shall be of type @code{REAL} or @code{INTEGER}.
7476@end multitable
7477
7478@item @emph{Return value}:
7479The return value is of the same type and kind as @var{X}
7480
7481@item @emph{Example}:
7482@smallexample
7483program test_huge_tiny
7484  print *, huge(0), huge(0.0), huge(0.0d0)
7485  print *, tiny(0.0), tiny(0.0d0)
7486end program test_huge_tiny
7487@end smallexample
7488@end table
7489
7490
7491
7492@node HYPOT
7493@section @code{HYPOT} --- Euclidean distance function
7494@fnindex HYPOT
7495@cindex Euclidean distance
7496
7497@table @asis
7498@item @emph{Description}:
7499@code{HYPOT(X,Y)} is the Euclidean distance function. It is equal to
7500@math{\sqrt{X^2 + Y^2}}, without undue underflow or overflow.
7501
7502@item @emph{Standard}:
7503Fortran 2008 and later
7504
7505@item @emph{Class}:
7506Elemental function
7507
7508@item @emph{Syntax}:
7509@code{RESULT = HYPOT(X, Y)}
7510
7511@item @emph{Arguments}:
7512@multitable @columnfractions .15 .70
7513@item @var{X} @tab The type shall be @code{REAL}.
7514@item @var{Y} @tab The type and kind type parameter shall be the same as
7515@var{X}.
7516@end multitable
7517
7518@item @emph{Return value}:
7519The return value has the same type and kind type parameter as @var{X}.
7520
7521@item @emph{Example}:
7522@smallexample
7523program test_hypot
7524  real(4) :: x = 1.e0_4, y = 0.5e0_4
7525  x = hypot(x,y)
7526end program test_hypot
7527@end smallexample
7528@end table
7529
7530
7531
7532@node IACHAR
7533@section @code{IACHAR} --- Code in @acronym{ASCII} collating sequence
7534@fnindex IACHAR
7535@cindex @acronym{ASCII} collating sequence
7536@cindex collating sequence, @acronym{ASCII}
7537@cindex conversion, to integer
7538
7539@table @asis
7540@item @emph{Description}:
7541@code{IACHAR(C)} returns the code for the @acronym{ASCII} character
7542in the first character position of @code{C}.
7543
7544@item @emph{Standard}:
7545Fortran 95 and later, with @var{KIND} argument Fortran 2003 and later
7546
7547@item @emph{Class}:
7548Elemental function
7549
7550@item @emph{Syntax}:
7551@code{RESULT = IACHAR(C [, KIND])}
7552
7553@item @emph{Arguments}:
7554@multitable @columnfractions .15 .70
7555@item @var{C}    @tab Shall be a scalar @code{CHARACTER}, with @code{INTENT(IN)}
7556@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
7557expression indicating the kind parameter of the result.
7558@end multitable
7559
7560@item @emph{Return value}:
7561The return value is of type @code{INTEGER} and of kind @var{KIND}. If
7562@var{KIND} is absent, the return value is of default integer kind.
7563
7564@item @emph{Example}:
7565@smallexample
7566program test_iachar
7567  integer i
7568  i = iachar(' ')
7569end program test_iachar
7570@end smallexample
7571
7572@item @emph{Note}:
7573See @ref{ICHAR} for a discussion of converting between numerical values
7574and formatted string representations.
7575
7576@item @emph{See also}:
7577@ref{ACHAR}, @gol
7578@ref{CHAR}, @gol
7579@ref{ICHAR}
7580@end table
7581
7582
7583
7584@node IALL
7585@section @code{IALL} --- Bitwise AND of array elements
7586@fnindex IALL
7587@cindex array, AND
7588@cindex bits, AND of array elements
7589
7590@table @asis
7591@item @emph{Description}:
7592Reduces with bitwise AND the elements of @var{ARRAY} along dimension @var{DIM}
7593if the corresponding element in @var{MASK} is @code{TRUE}.
7594
7595@item @emph{Standard}:
7596Fortran 2008 and later
7597
7598@item @emph{Class}:
7599Transformational function
7600
7601@item @emph{Syntax}:
7602@multitable @columnfractions .80
7603@item @code{RESULT = IALL(ARRAY[, MASK])}
7604@item @code{RESULT = IALL(ARRAY, DIM[, MASK])}
7605@end multitable
7606
7607@item @emph{Arguments}:
7608@multitable @columnfractions .15 .70
7609@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER}
7610@item @var{DIM}   @tab (Optional) shall be a scalar of type
7611@code{INTEGER} with a value in the range from 1 to n, where n
7612equals the rank of @var{ARRAY}.
7613@item @var{MASK}  @tab (Optional) shall be of type @code{LOGICAL}
7614and either be a scalar or an array of the same shape as @var{ARRAY}.
7615@end multitable
7616
7617@item @emph{Return value}:
7618The result is of the same type as @var{ARRAY}.
7619
7620If @var{DIM} is absent, a scalar with the bitwise ALL of all elements in
7621@var{ARRAY} is returned. Otherwise, an array of rank n-1, where n equals
7622the rank of @var{ARRAY}, and a shape similar to that of @var{ARRAY} with
7623dimension @var{DIM} dropped is returned.
7624
7625@item @emph{Example}:
7626@smallexample
7627PROGRAM test_iall
7628  INTEGER(1) :: a(2)
7629
7630  a(1) = b'00100100'
7631  a(2) = b'01101010'
7632
7633  ! prints 00100000
7634  PRINT '(b8.8)', IALL(a)
7635END PROGRAM
7636@end smallexample
7637
7638@item @emph{See also}:
7639@ref{IANY}, @gol
7640@ref{IPARITY}, @gol
7641@ref{IAND}
7642@end table
7643
7644
7645
7646@node IAND
7647@section @code{IAND} --- Bitwise logical and
7648@fnindex IAND
7649@fnindex BIAND
7650@fnindex IIAND
7651@fnindex JIAND
7652@fnindex KIAND
7653@cindex bitwise logical and
7654@cindex logical and, bitwise
7655
7656@table @asis
7657@item @emph{Description}:
7658Bitwise logical @code{AND}.
7659
7660@item @emph{Standard}:
7661Fortran 90 and later, with boz-literal-constant Fortran 2008 and later, has overloads that are GNU extensions
7662
7663@item @emph{Class}:
7664Elemental function
7665
7666@item @emph{Syntax}:
7667@code{RESULT = IAND(I, J)}
7668
7669@item @emph{Arguments}:
7670@multitable @columnfractions .15 .70
7671@item @var{I} @tab The type shall be @code{INTEGER} or a boz-literal-constant.
7672@item @var{J} @tab The type shall be @code{INTEGER} with the same
7673kind type parameter as @var{I} or a boz-literal-constant.
7674@var{I} and @var{J} shall not both be boz-literal-constants.
7675@end multitable
7676
7677@item @emph{Return value}:
7678The return type is @code{INTEGER} with the kind type parameter of the
7679arguments.
7680A boz-literal-constant is converted to an @code{INTEGER} with the kind
7681type parameter of the other argument as-if a call to @ref{INT} occurred.
7682
7683@item @emph{Example}:
7684@smallexample
7685PROGRAM test_iand
7686  INTEGER :: a, b
7687  DATA a / Z'F' /, b / Z'3' /
7688  WRITE (*,*) IAND(a, b)
7689END PROGRAM
7690@end smallexample
7691
7692@item @emph{Specific names}:
7693@multitable @columnfractions .20 .20 .20 .25
7694@item Name            @tab Argument            @tab Return type       @tab Standard
7695@item @code{IAND(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
7696@item @code{BIAND(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
7697@item @code{IIAND(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
7698@item @code{JIAND(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
7699@item @code{KIAND(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
7700@end multitable
7701
7702@item @emph{See also}:
7703@ref{IOR}, @gol
7704@ref{IEOR}, @gol
7705@ref{IBITS}, @gol
7706@ref{IBSET}, @gol
7707@ref{IBCLR}, @gol
7708@ref{NOT}
7709@end table
7710
7711
7712
7713@node IANY
7714@section @code{IANY} --- Bitwise OR of array elements
7715@fnindex IANY
7716@cindex array, OR
7717@cindex bits, OR of array elements
7718
7719@table @asis
7720@item @emph{Description}:
7721Reduces with bitwise OR (inclusive or) the elements of @var{ARRAY} along
7722dimension @var{DIM} if the corresponding element in @var{MASK} is @code{TRUE}.
7723
7724@item @emph{Standard}:
7725Fortran 2008 and later
7726
7727@item @emph{Class}:
7728Transformational function
7729
7730@item @emph{Syntax}:
7731@multitable @columnfractions .80
7732@item @code{RESULT = IANY(ARRAY[, MASK])}
7733@item @code{RESULT = IANY(ARRAY, DIM[, MASK])}
7734@end multitable
7735
7736@item @emph{Arguments}:
7737@multitable @columnfractions .15 .70
7738@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER}
7739@item @var{DIM}   @tab (Optional) shall be a scalar of type
7740@code{INTEGER} with a value in the range from 1 to n, where n
7741equals the rank of @var{ARRAY}.
7742@item @var{MASK}  @tab (Optional) shall be of type @code{LOGICAL}
7743and either be a scalar or an array of the same shape as @var{ARRAY}.
7744@end multitable
7745
7746@item @emph{Return value}:
7747The result is of the same type as @var{ARRAY}.
7748
7749If @var{DIM} is absent, a scalar with the bitwise OR of all elements in
7750@var{ARRAY} is returned. Otherwise, an array of rank n-1, where n equals
7751the rank of @var{ARRAY}, and a shape similar to that of @var{ARRAY} with
7752dimension @var{DIM} dropped is returned.
7753
7754@item @emph{Example}:
7755@smallexample
7756PROGRAM test_iany
7757  INTEGER(1) :: a(2)
7758
7759  a(1) = b'00100100'
7760  a(2) = b'01101010'
7761
7762  ! prints 01101110
7763  PRINT '(b8.8)', IANY(a)
7764END PROGRAM
7765@end smallexample
7766
7767@item @emph{See also}:
7768@ref{IPARITY}, @gol
7769@ref{IALL}, @gol
7770@ref{IOR}
7771@end table
7772
7773
7774
7775@node IARGC
7776@section @code{IARGC} --- Get the number of command line arguments
7777@fnindex IARGC
7778@cindex command-line arguments
7779@cindex command-line arguments, number of
7780@cindex arguments, to program
7781
7782@table @asis
7783@item @emph{Description}:
7784@code{IARGC} returns the number of arguments passed on the
7785command line when the containing program was invoked.
7786
7787This intrinsic routine is provided for backwards compatibility with
7788GNU Fortran 77.  In new code, programmers should consider the use of
7789the @ref{COMMAND_ARGUMENT_COUNT} intrinsic defined by the Fortran 2003
7790standard.
7791
7792@item @emph{Standard}:
7793GNU extension
7794
7795@item @emph{Class}:
7796Function
7797
7798@item @emph{Syntax}:
7799@code{RESULT = IARGC()}
7800
7801@item @emph{Arguments}:
7802None
7803
7804@item @emph{Return value}:
7805The number of command line arguments, type @code{INTEGER(4)}.
7806
7807@item @emph{Example}:
7808See @ref{GETARG}
7809
7810@item @emph{See also}:
7811GNU Fortran 77 compatibility subroutine: @gol
7812@ref{GETARG} @gol
7813Fortran 2003 functions and subroutines: @gol
7814@ref{GET_COMMAND}, @gol
7815@ref{GET_COMMAND_ARGUMENT}, @gol
7816@ref{COMMAND_ARGUMENT_COUNT}
7817@end table
7818
7819
7820
7821@node IBCLR
7822@section @code{IBCLR} --- Clear bit
7823@fnindex IBCLR
7824@fnindex BBCLR
7825@fnindex IIBCLR
7826@fnindex JIBCLR
7827@fnindex KIBCLR
7828@cindex bits, unset
7829@cindex bits, clear
7830
7831@table @asis
7832@item @emph{Description}:
7833@code{IBCLR} returns the value of @var{I} with the bit at position
7834@var{POS} set to zero.
7835
7836@item @emph{Standard}:
7837Fortran 90 and later, has overloads that are GNU extensions
7838
7839@item @emph{Class}:
7840Elemental function
7841
7842@item @emph{Syntax}:
7843@code{RESULT = IBCLR(I, POS)}
7844
7845@item @emph{Arguments}:
7846@multitable @columnfractions .15 .70
7847@item @var{I} @tab The type shall be @code{INTEGER}.
7848@item @var{POS} @tab The type shall be @code{INTEGER}.
7849@end multitable
7850
7851@item @emph{Return value}:
7852The return value is of type @code{INTEGER} and of the same kind as
7853@var{I}.
7854
7855@item @emph{Specific names}:
7856@multitable @columnfractions .20 .20 .20 .25
7857@item Name            @tab Argument            @tab Return type       @tab Standard
7858@item @code{IBCLR(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
7859@item @code{BBCLR(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
7860@item @code{IIBCLR(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
7861@item @code{JIBCLR(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
7862@item @code{KIBCLR(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
7863@end multitable
7864
7865@item @emph{See also}:
7866@ref{IBITS}, @gol
7867@ref{IBSET}, @gol
7868@ref{IAND}, @gol
7869@ref{IOR}, @gol
7870@ref{IEOR}, @gol
7871@ref{MVBITS}
7872@end table
7873
7874
7875
7876@node IBITS
7877@section @code{IBITS} --- Bit extraction
7878@fnindex IBITS
7879@fnindex BBITS
7880@fnindex IIBITS
7881@fnindex JIBITS
7882@fnindex KIBITS
7883@cindex bits, get
7884@cindex bits, extract
7885
7886@table @asis
7887@item @emph{Description}:
7888@code{IBITS} extracts a field of length @var{LEN} from @var{I},
7889starting from bit position @var{POS} and extending left for @var{LEN}
7890bits.  The result is right-justified and the remaining bits are
7891zeroed.  The value of @code{POS+LEN} must be less than or equal to the
7892value @code{BIT_SIZE(I)}.
7893
7894@item @emph{Standard}:
7895Fortran 90 and later, has overloads that are GNU extensions
7896
7897@item @emph{Class}:
7898Elemental function
7899
7900@item @emph{Syntax}:
7901@code{RESULT = IBITS(I, POS, LEN)}
7902
7903@item @emph{Arguments}:
7904@multitable @columnfractions .15 .70
7905@item @var{I}   @tab The type shall be @code{INTEGER}.
7906@item @var{POS} @tab The type shall be @code{INTEGER}.
7907@item @var{LEN} @tab The type shall be @code{INTEGER}.
7908@end multitable
7909
7910@item @emph{Return value}:
7911The return value is of type @code{INTEGER} and of the same kind as
7912@var{I}.
7913
7914@item @emph{Specific names}:
7915@multitable @columnfractions .20 .20 .20 .25
7916@item Name            @tab Argument            @tab Return type       @tab Standard
7917@item @code{IBITS(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
7918@item @code{BBITS(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
7919@item @code{IIBITS(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
7920@item @code{JIBITS(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
7921@item @code{KIBITS(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
7922@end multitable
7923
7924@item @emph{See also}:
7925@ref{BIT_SIZE}, @gol
7926@ref{IBCLR}, @gol
7927@ref{IBSET}, @gol
7928@ref{IAND}, @gol
7929@ref{IOR}, @gol
7930@ref{IEOR}
7931@end table
7932
7933
7934
7935@node IBSET
7936@section @code{IBSET} --- Set bit
7937@fnindex IBSET
7938@fnindex BBSET
7939@fnindex IIBSET
7940@fnindex JIBSET
7941@fnindex KIBSET
7942@cindex bits, set
7943
7944@table @asis
7945@item @emph{Description}:
7946@code{IBSET} returns the value of @var{I} with the bit at position
7947@var{POS} set to one.
7948
7949@item @emph{Standard}:
7950Fortran 90 and later, has overloads that are GNU extensions
7951
7952@item @emph{Class}:
7953Elemental function
7954
7955@item @emph{Syntax}:
7956@code{RESULT = IBSET(I, POS)}
7957
7958@item @emph{Arguments}:
7959@multitable @columnfractions .15 .70
7960@item @var{I} @tab The type shall be @code{INTEGER}.
7961@item @var{POS} @tab The type shall be @code{INTEGER}.
7962@end multitable
7963
7964@item @emph{Return value}:
7965The return value is of type @code{INTEGER} and of the same kind as
7966@var{I}.
7967
7968@item @emph{Specific names}:
7969@multitable @columnfractions .20 .20 .20 .25
7970@item Name            @tab Argument            @tab Return type       @tab Standard
7971@item @code{IBSET(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
7972@item @code{BBSET(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
7973@item @code{IIBSET(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
7974@item @code{JIBSET(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
7975@item @code{KIBSET(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
7976@end multitable
7977
7978@item @emph{See also}:
7979@ref{IBCLR}, @gol
7980@ref{IBITS}, @gol
7981@ref{IAND}, @gol
7982@ref{IOR}, @gol
7983@ref{IEOR}, @gol
7984@ref{MVBITS}
7985@end table
7986
7987
7988
7989@node ICHAR
7990@section @code{ICHAR} --- Character-to-integer conversion function
7991@fnindex ICHAR
7992@cindex conversion, to integer
7993
7994@table @asis
7995@item @emph{Description}:
7996@code{ICHAR(C)} returns the code for the character in the first character
7997position of @code{C} in the system's native character set.
7998The correspondence between characters and their codes is not necessarily
7999the same across different GNU Fortran implementations.
8000
8001@item @emph{Standard}:
8002Fortran 77 and later, with @var{KIND} argument Fortran 2003 and later
8003
8004@item @emph{Class}:
8005Elemental function
8006
8007@item @emph{Syntax}:
8008@code{RESULT = ICHAR(C [, KIND])}
8009
8010@item @emph{Arguments}:
8011@multitable @columnfractions .15 .70
8012@item @var{C}    @tab Shall be a scalar @code{CHARACTER}, with @code{INTENT(IN)}
8013@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
8014expression indicating the kind parameter of the result.
8015@end multitable
8016
8017@item @emph{Return value}:
8018The return value is of type @code{INTEGER} and of kind @var{KIND}. If
8019@var{KIND} is absent, the return value is of default integer kind.
8020
8021@item @emph{Example}:
8022@smallexample
8023program test_ichar
8024  integer i
8025  i = ichar(' ')
8026end program test_ichar
8027@end smallexample
8028
8029@item @emph{Specific names}:
8030@multitable @columnfractions .20 .20 .20 .25
8031@item Name             @tab Argument             @tab Return type       @tab Standard
8032@item @code{ICHAR(C)}  @tab @code{CHARACTER C}   @tab @code{INTEGER(4)}    @tab Fortran 77 and later
8033@end multitable
8034
8035@item @emph{Note}:
8036No intrinsic exists to convert between a numeric value and a formatted
8037character string representation -- for instance, given the
8038@code{CHARACTER} value @code{'154'}, obtaining an @code{INTEGER} or
8039@code{REAL} value with the value 154, or vice versa. Instead, this
8040functionality is provided by internal-file I/O, as in the following
8041example:
8042@smallexample
8043program read_val
8044  integer value
8045  character(len=10) string, string2
8046  string = '154'
8047
8048  ! Convert a string to a numeric value
8049  read (string,'(I10)') value
8050  print *, value
8051
8052  ! Convert a value to a formatted string
8053  write (string2,'(I10)') value
8054  print *, string2
8055end program read_val
8056@end smallexample
8057
8058@item @emph{See also}:
8059@ref{ACHAR}, @gol
8060@ref{CHAR}, @gol
8061@ref{IACHAR}
8062@end table
8063
8064
8065
8066@node IDATE
8067@section @code{IDATE} --- Get current local time subroutine (day/month/year)
8068@fnindex IDATE
8069@cindex date, current
8070@cindex current date
8071
8072@table @asis
8073@item @emph{Description}:
8074@code{IDATE(VALUES)} Fills @var{VALUES} with the numerical values at the
8075current local time. The day (in the range 1-31), month (in the range 1-12),
8076and year appear in elements 1, 2, and 3 of @var{VALUES}, respectively.
8077The year has four significant digits.
8078
8079This intrinsic routine is provided for backwards compatibility with
8080GNU Fortran 77.  In new code, programmers should consider the use of
8081the @ref{DATE_AND_TIME} intrinsic defined by the Fortran 95
8082standard.
8083
8084@item @emph{Standard}:
8085GNU extension
8086
8087@item @emph{Class}:
8088Subroutine
8089
8090@item @emph{Syntax}:
8091@code{CALL IDATE(VALUES)}
8092
8093@item @emph{Arguments}:
8094@multitable @columnfractions .15 .70
8095@item @var{VALUES} @tab The type shall be @code{INTEGER, DIMENSION(3)} and
8096the kind shall be the default integer kind.
8097@end multitable
8098
8099@item @emph{Return value}:
8100Does not return anything.
8101
8102@item @emph{Example}:
8103@smallexample
8104program test_idate
8105  integer, dimension(3) :: tarray
8106  call idate(tarray)
8107  print *, tarray(1)
8108  print *, tarray(2)
8109  print *, tarray(3)
8110end program test_idate
8111@end smallexample
8112
8113@item @emph{See also}:
8114@ref{DATE_AND_TIME}
8115@end table
8116
8117
8118@node IEOR
8119@section @code{IEOR} --- Bitwise logical exclusive or
8120@fnindex IEOR
8121@fnindex BIEOR
8122@fnindex IIEOR
8123@fnindex JIEOR
8124@fnindex KIEOR
8125@cindex bitwise logical exclusive or
8126@cindex logical exclusive or, bitwise
8127
8128@table @asis
8129@item @emph{Description}:
8130@code{IEOR} returns the bitwise Boolean exclusive-OR of @var{I} and
8131@var{J}.
8132
8133@item @emph{Standard}:
8134Fortran 90 and later, with boz-literal-constant Fortran 2008 and later, has overloads that are GNU extensions
8135
8136@item @emph{Class}:
8137Elemental function
8138
8139@item @emph{Syntax}:
8140@code{RESULT = IEOR(I, J)}
8141
8142@item @emph{Arguments}:
8143@multitable @columnfractions .15 .70
8144@item @var{I} @tab The type shall be @code{INTEGER} or a boz-literal-constant.
8145@item @var{J} @tab The type shall be @code{INTEGER} with the same
8146kind type parameter as @var{I} or a boz-literal-constant.
8147@var{I} and @var{J} shall not both be boz-literal-constants.
8148@end multitable
8149
8150@item @emph{Return value}:
8151The return type is @code{INTEGER} with the kind type parameter of the
8152arguments.
8153A boz-literal-constant is converted to an @code{INTEGER} with the kind
8154type parameter of the other argument as-if a call to @ref{INT} occurred.
8155
8156@item @emph{Specific names}:
8157@multitable @columnfractions .20 .20 .20 .25
8158@item Name            @tab Argument            @tab Return type       @tab Standard
8159@item @code{IEOR(A)}  @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
8160@item @code{BIEOR(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
8161@item @code{IIEOR(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
8162@item @code{JIEOR(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
8163@item @code{KIEOR(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
8164@end multitable
8165
8166@item @emph{See also}:
8167@ref{IOR}, @gol
8168@ref{IAND}, @gol
8169@ref{IBITS}, @gol
8170@ref{IBSET}, @gol
8171@ref{IBCLR}, @gol
8172@ref{NOT}
8173@end table
8174
8175
8176
8177@node IERRNO
8178@section @code{IERRNO} --- Get the last system error number
8179@fnindex IERRNO
8180@cindex system, error handling
8181
8182@table @asis
8183@item @emph{Description}:
8184Returns the last system error number, as given by the C @code{errno}
8185variable.
8186
8187@item @emph{Standard}:
8188GNU extension
8189
8190@item @emph{Class}:
8191Function
8192
8193@item @emph{Syntax}:
8194@code{RESULT = IERRNO()}
8195
8196@item @emph{Arguments}:
8197None
8198
8199@item @emph{Return value}:
8200The return value is of type @code{INTEGER} and of the default integer
8201kind.
8202
8203@item @emph{See also}:
8204@ref{PERROR}
8205@end table
8206
8207
8208
8209@node IMAGE_INDEX
8210@section @code{IMAGE_INDEX} --- Function that converts a cosubscript to an image index
8211@fnindex IMAGE_INDEX
8212@cindex coarray, @code{IMAGE_INDEX}
8213@cindex images, cosubscript to image index conversion
8214
8215@table @asis
8216@item @emph{Description}:
8217Returns the image index belonging to a cosubscript.
8218
8219@item @emph{Standard}:
8220Fortran 2008 and later
8221
8222@item @emph{Class}:
8223Inquiry function.
8224
8225@item @emph{Syntax}:
8226@code{RESULT = IMAGE_INDEX(COARRAY, SUB)}
8227
8228@item @emph{Arguments}:
8229@multitable @columnfractions .15 .70
8230@item @var{COARRAY} @tab Coarray of any type.
8231@item @var{SUB}     @tab default integer rank-1 array of a size equal to
8232the corank of @var{COARRAY}.
8233@end multitable
8234
8235
8236@item @emph{Return value}:
8237Scalar default integer with the value of the image index which corresponds
8238to the cosubscripts. For invalid cosubscripts the result is zero.
8239
8240@item @emph{Example}:
8241@smallexample
8242INTEGER :: array[2,-1:4,8,*]
8243! Writes  28 (or 0 if there are fewer than 28 images)
8244WRITE (*,*) IMAGE_INDEX (array, [2,0,3,1])
8245@end smallexample
8246
8247@item @emph{See also}:
8248@ref{THIS_IMAGE}, @gol
8249@ref{NUM_IMAGES}
8250@end table
8251
8252
8253
8254@node INDEX intrinsic
8255@section @code{INDEX} --- Position of a substring within a string
8256@fnindex INDEX
8257@cindex substring position
8258@cindex string, find substring
8259
8260@table @asis
8261@item @emph{Description}:
8262Returns the position of the start of the first occurrence of string
8263@var{SUBSTRING} as a substring in @var{STRING}, counting from one.  If
8264@var{SUBSTRING} is not present in @var{STRING}, zero is returned.  If
8265the @var{BACK} argument is present and true, the return value is the
8266start of the last occurrence rather than the first.
8267
8268@item @emph{Standard}:
8269Fortran 77 and later, with @var{KIND} argument Fortran 2003 and later
8270
8271@item @emph{Class}:
8272Elemental function
8273
8274@item @emph{Syntax}:
8275@code{RESULT = INDEX(STRING, SUBSTRING [, BACK [, KIND]])}
8276
8277@item @emph{Arguments}:
8278@multitable @columnfractions .15 .70
8279@item @var{STRING} @tab Shall be a scalar @code{CHARACTER}, with
8280@code{INTENT(IN)}
8281@item @var{SUBSTRING} @tab Shall be a scalar @code{CHARACTER}, with
8282@code{INTENT(IN)}
8283@item @var{BACK} @tab (Optional) Shall be a scalar @code{LOGICAL}, with
8284@code{INTENT(IN)}
8285@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
8286expression indicating the kind parameter of the result.
8287@end multitable
8288
8289@item @emph{Return value}:
8290The return value is of type @code{INTEGER} and of kind @var{KIND}. If
8291@var{KIND} is absent, the return value is of default integer kind.
8292
8293@item @emph{Specific names}:
8294@multitable @columnfractions .20 .20 .20 .25
8295@item Name                            @tab Argument           @tab Return type       @tab Standard
8296@item @code{INDEX(STRING, SUBSTRING)} @tab @code{CHARACTER}   @tab @code{INTEGER(4)} @tab Fortran 77 and later
8297@end multitable
8298
8299@item @emph{See also}:
8300@ref{SCAN}, @gol
8301@ref{VERIFY}
8302@end table
8303
8304
8305
8306@node INT
8307@section @code{INT} --- Convert to integer type
8308@fnindex INT
8309@fnindex IFIX
8310@fnindex IDINT
8311@cindex conversion, to integer
8312
8313@table @asis
8314@item @emph{Description}:
8315Convert to integer type
8316
8317@item @emph{Standard}:
8318Fortran 77 and later, with boz-literal-constant Fortran 2008 and later.
8319
8320@item @emph{Class}:
8321Elemental function
8322
8323@item @emph{Syntax}:
8324@code{RESULT = INT(A [, KIND))}
8325
8326@item @emph{Arguments}:
8327@multitable @columnfractions .15 .70
8328@item @var{A}    @tab Shall be of type @code{INTEGER},
8329@code{REAL}, or @code{COMPLEX} or or a boz-literal-constant.
8330@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
8331expression indicating the kind parameter of the result.
8332@end multitable
8333
8334@item @emph{Return value}:
8335These functions return a @code{INTEGER} variable or array under
8336the following rules:
8337
8338@table @asis
8339@item (A)
8340If @var{A} is of type @code{INTEGER}, @code{INT(A) = A}
8341@item (B)
8342If @var{A} is of type @code{REAL} and @math{|A| < 1}, @code{INT(A)}
8343equals @code{0}. If @math{|A| \geq 1}, then @code{INT(A)} is the integer
8344whose magnitude is the largest integer that does not exceed the magnitude
8345of @var{A} and whose sign is the same as the sign of @var{A}.
8346@item (C)
8347If @var{A} is of type @code{COMPLEX}, rule B is applied to the real part of @var{A}.
8348@end table
8349
8350@item @emph{Example}:
8351@smallexample
8352program test_int
8353  integer :: i = 42
8354  complex :: z = (-3.7, 1.0)
8355  print *, int(i)
8356  print *, int(z), int(z,8)
8357end program
8358@end smallexample
8359
8360@item @emph{Specific names}:
8361@multitable @columnfractions .20 .20 .20 .25
8362@item Name            @tab Argument          @tab Return type       @tab Standard
8363@item @code{INT(A)}   @tab @code{REAL(4) A}  @tab @code{INTEGER}    @tab Fortran 77 and later
8364@item @code{IFIX(A)}  @tab @code{REAL(4) A}  @tab @code{INTEGER}    @tab Fortran 77 and later
8365@item @code{IDINT(A)} @tab @code{REAL(8) A}  @tab @code{INTEGER}    @tab Fortran 77 and later
8366@end multitable
8367
8368@end table
8369
8370
8371@node INT2
8372@section @code{INT2} --- Convert to 16-bit integer type
8373@fnindex INT2
8374@fnindex SHORT
8375@cindex conversion, to integer
8376
8377@table @asis
8378@item @emph{Description}:
8379Convert to a @code{KIND=2} integer type. This is equivalent to the
8380standard @code{INT} intrinsic with an optional argument of
8381@code{KIND=2}, and is only included for backwards compatibility.
8382
8383The @code{SHORT} intrinsic is equivalent to @code{INT2}.
8384
8385@item @emph{Standard}:
8386GNU extension
8387
8388@item @emph{Class}:
8389Elemental function
8390
8391@item @emph{Syntax}:
8392@code{RESULT = INT2(A)}
8393
8394@item @emph{Arguments}:
8395@multitable @columnfractions .15 .70
8396@item @var{A}    @tab Shall be of type @code{INTEGER},
8397@code{REAL}, or @code{COMPLEX}.
8398@end multitable
8399
8400@item @emph{Return value}:
8401The return value is a @code{INTEGER(2)} variable.
8402
8403@item @emph{See also}:
8404@ref{INT}, @gol
8405@ref{INT8}, @gol
8406@ref{LONG}
8407@end table
8408
8409
8410
8411@node INT8
8412@section @code{INT8} --- Convert to 64-bit integer type
8413@fnindex INT8
8414@cindex conversion, to integer
8415
8416@table @asis
8417@item @emph{Description}:
8418Convert to a @code{KIND=8} integer type. This is equivalent to the
8419standard @code{INT} intrinsic with an optional argument of
8420@code{KIND=8}, and is only included for backwards compatibility.
8421
8422@item @emph{Standard}:
8423GNU extension
8424
8425@item @emph{Class}:
8426Elemental function
8427
8428@item @emph{Syntax}:
8429@code{RESULT = INT8(A)}
8430
8431@item @emph{Arguments}:
8432@multitable @columnfractions .15 .70
8433@item @var{A}    @tab Shall be of type @code{INTEGER},
8434@code{REAL}, or @code{COMPLEX}.
8435@end multitable
8436
8437@item @emph{Return value}:
8438The return value is a @code{INTEGER(8)} variable.
8439
8440@item @emph{See also}:
8441@ref{INT}, @gol
8442@ref{INT2}, @gol
8443@ref{LONG}
8444@end table
8445
8446
8447
8448@node IOR
8449@section @code{IOR} --- Bitwise logical or
8450@fnindex IOR
8451@fnindex BIOR
8452@fnindex IIOR
8453@fnindex JIOR
8454@fnindex KIOR
8455@cindex bitwise logical or
8456@cindex logical or, bitwise
8457
8458@table @asis
8459@item @emph{Description}:
8460@code{IOR} returns the bitwise Boolean inclusive-OR of @var{I} and
8461@var{J}.
8462
8463@item @emph{Standard}:
8464Fortran 90 and later, with boz-literal-constant Fortran 2008 and later, has overloads that are GNU extensions
8465
8466@item @emph{Class}:
8467Elemental function
8468
8469@item @emph{Syntax}:
8470@code{RESULT = IOR(I, J)}
8471
8472@item @emph{Arguments}:
8473@multitable @columnfractions .15 .70
8474@item @var{I} @tab The type shall be @code{INTEGER} or a boz-literal-constant.
8475@item @var{J} @tab The type shall be @code{INTEGER} with the same
8476kind type parameter as @var{I} or a boz-literal-constant.
8477@var{I} and @var{J} shall not both be boz-literal-constants.
8478@end multitable
8479
8480@item @emph{Return value}:
8481The return type is @code{INTEGER} with the kind type parameter of the
8482arguments.
8483A boz-literal-constant is converted to an @code{INTEGER} with the kind
8484type parameter of the other argument as-if a call to @ref{INT} occurred.
8485
8486@item @emph{Specific names}:
8487@multitable @columnfractions .20 .20 .20 .25
8488@item Name            @tab Argument            @tab Return type       @tab Standard
8489@item @code{IOR(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
8490@item @code{BIOR(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
8491@item @code{IIOR(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
8492@item @code{JIOR(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
8493@item @code{KIOR(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
8494@end multitable
8495
8496@item @emph{See also}:
8497@ref{IEOR}, @gol
8498@ref{IAND}, @gol
8499@ref{IBITS}, @gol
8500@ref{IBSET}, @gol
8501@ref{IBCLR}, @gol
8502@ref{NOT}
8503@end table
8504
8505
8506
8507@node IPARITY
8508@section @code{IPARITY} --- Bitwise XOR of array elements
8509@fnindex IPARITY
8510@cindex array, parity
8511@cindex array, XOR
8512@cindex bits, XOR of array elements
8513
8514@table @asis
8515@item @emph{Description}:
8516Reduces with bitwise XOR (exclusive or) the elements of @var{ARRAY} along
8517dimension @var{DIM} if the corresponding element in @var{MASK} is @code{TRUE}.
8518
8519@item @emph{Standard}:
8520Fortran 2008 and later
8521
8522@item @emph{Class}:
8523Transformational function
8524
8525@item @emph{Syntax}:
8526@multitable @columnfractions .80
8527@item @code{RESULT = IPARITY(ARRAY[, MASK])}
8528@item @code{RESULT = IPARITY(ARRAY, DIM[, MASK])}
8529@end multitable
8530
8531@item @emph{Arguments}:
8532@multitable @columnfractions .15 .70
8533@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER}
8534@item @var{DIM}   @tab (Optional) shall be a scalar of type
8535@code{INTEGER} with a value in the range from 1 to n, where n
8536equals the rank of @var{ARRAY}.
8537@item @var{MASK}  @tab (Optional) shall be of type @code{LOGICAL}
8538and either be a scalar or an array of the same shape as @var{ARRAY}.
8539@end multitable
8540
8541@item @emph{Return value}:
8542The result is of the same type as @var{ARRAY}.
8543
8544If @var{DIM} is absent, a scalar with the bitwise XOR of all elements in
8545@var{ARRAY} is returned. Otherwise, an array of rank n-1, where n equals
8546the rank of @var{ARRAY}, and a shape similar to that of @var{ARRAY} with
8547dimension @var{DIM} dropped is returned.
8548
8549@item @emph{Example}:
8550@smallexample
8551PROGRAM test_iparity
8552  INTEGER(1) :: a(2)
8553
8554  a(1) = int(b'00100100', 1)
8555  a(2) = int(b'01101010', 1)
8556
8557  ! prints 01001110
8558  PRINT '(b8.8)', IPARITY(a)
8559END PROGRAM
8560@end smallexample
8561
8562@item @emph{See also}:
8563@ref{IANY}, @gol
8564@ref{IALL}, @gol
8565@ref{IEOR}, @gol
8566@ref{PARITY}
8567@end table
8568
8569
8570
8571@node IRAND
8572@section @code{IRAND} --- Integer pseudo-random number
8573@fnindex IRAND
8574@cindex random number generation
8575
8576@table @asis
8577@item @emph{Description}:
8578@code{IRAND(FLAG)} returns a pseudo-random number from a uniform
8579distribution between 0 and a system-dependent limit (which is in most
8580cases 2147483647). If @var{FLAG} is 0, the next number
8581in the current sequence is returned; if @var{FLAG} is 1, the generator
8582is restarted by @code{CALL SRAND(0)}; if @var{FLAG} has any other value,
8583it is used as a new seed with @code{SRAND}.
8584
8585This intrinsic routine is provided for backwards compatibility with
8586GNU Fortran 77. It implements a simple modulo generator as provided
8587by @command{g77}. For new code, one should consider the use of
8588@ref{RANDOM_NUMBER} as it implements a superior algorithm.
8589
8590@item @emph{Standard}:
8591GNU extension
8592
8593@item @emph{Class}:
8594Function
8595
8596@item @emph{Syntax}:
8597@code{RESULT = IRAND(I)}
8598
8599@item @emph{Arguments}:
8600@multitable @columnfractions .15 .70
8601@item @var{I} @tab Shall be a scalar @code{INTEGER} of kind 4.
8602@end multitable
8603
8604@item @emph{Return value}:
8605The return value is of @code{INTEGER(kind=4)} type.
8606
8607@item @emph{Example}:
8608@smallexample
8609program test_irand
8610  integer,parameter :: seed = 86456
8611
8612  call srand(seed)
8613  print *, irand(), irand(), irand(), irand()
8614  print *, irand(seed), irand(), irand(), irand()
8615end program test_irand
8616@end smallexample
8617
8618@end table
8619
8620
8621
8622@node IS_CONTIGUOUS
8623@section @code{IS_CONTIGUOUS} --- Test whether an array is contiguous
8624@fnindex IS_IOSTAT_EOR
8625@cindex array, contiguity
8626
8627@table @asis
8628@item @emph{Description}:
8629@code{IS_CONTIGUOUS} tests whether an array is contiguous.
8630
8631@item @emph{Standard}:
8632Fortran 2008 and later
8633
8634@item @emph{Class}:
8635Inquiry function
8636
8637@item @emph{Syntax}:
8638@code{RESULT = IS_CONTIGUOUS(ARRAY)}
8639
8640@item @emph{Arguments}:
8641@multitable @columnfractions .15 .70
8642@item @var{ARRAY} @tab Shall be an array of any type.
8643@end multitable
8644
8645@item @emph{Return value}:
8646Returns a @code{LOGICAL} of the default kind, which @code{.TRUE.} if
8647@var{ARRAY} is contiguous and false otherwise.
8648
8649@item @emph{Example}:
8650@smallexample
8651program test
8652  integer :: a(10)
8653  a = [1,2,3,4,5,6,7,8,9,10]
8654  call sub (a)      ! every element, is contiguous
8655  call sub (a(::2)) ! every other element, is noncontiguous
8656contains
8657  subroutine sub (x)
8658    integer :: x(:)
8659    if (is_contiguous (x)) then
8660      write (*,*) 'X is contiguous'
8661    else
8662      write (*,*) 'X is not contiguous'
8663    end if
8664  end subroutine sub
8665end program test
8666@end smallexample
8667@end table
8668
8669
8670
8671@node IS_IOSTAT_END
8672@section @code{IS_IOSTAT_END} --- Test for end-of-file value
8673@fnindex IS_IOSTAT_END
8674@cindex @code{IOSTAT}, end of file
8675
8676@table @asis
8677@item @emph{Description}:
8678@code{IS_IOSTAT_END} tests whether an variable has the value of the I/O
8679status ``end of file''. The function is equivalent to comparing the variable
8680with the @code{IOSTAT_END} parameter of the intrinsic module
8681@code{ISO_FORTRAN_ENV}.
8682
8683@item @emph{Standard}:
8684Fortran 2003 and later
8685
8686@item @emph{Class}:
8687Elemental function
8688
8689@item @emph{Syntax}:
8690@code{RESULT = IS_IOSTAT_END(I)}
8691
8692@item @emph{Arguments}:
8693@multitable @columnfractions .15 .70
8694@item @var{I} @tab Shall be of the type @code{INTEGER}.
8695@end multitable
8696
8697@item @emph{Return value}:
8698Returns a @code{LOGICAL} of the default kind, which @code{.TRUE.} if
8699@var{I} has the value which indicates an end of file condition for
8700@code{IOSTAT=} specifiers, and is @code{.FALSE.} otherwise.
8701
8702@item @emph{Example}:
8703@smallexample
8704PROGRAM iostat
8705  IMPLICIT NONE
8706  INTEGER :: stat, i
8707  OPEN(88, FILE='test.dat')
8708  READ(88, *, IOSTAT=stat) i
8709  IF(IS_IOSTAT_END(stat)) STOP 'END OF FILE'
8710END PROGRAM
8711@end smallexample
8712@end table
8713
8714
8715
8716@node IS_IOSTAT_EOR
8717@section @code{IS_IOSTAT_EOR} --- Test for end-of-record value
8718@fnindex IS_IOSTAT_EOR
8719@cindex @code{IOSTAT}, end of record
8720
8721@table @asis
8722@item @emph{Description}:
8723@code{IS_IOSTAT_EOR} tests whether an variable has the value of the I/O
8724status ``end of record''. The function is equivalent to comparing the
8725variable with the @code{IOSTAT_EOR} parameter of the intrinsic module
8726@code{ISO_FORTRAN_ENV}.
8727
8728@item @emph{Standard}:
8729Fortran 2003 and later
8730
8731@item @emph{Class}:
8732Elemental function
8733
8734@item @emph{Syntax}:
8735@code{RESULT = IS_IOSTAT_EOR(I)}
8736
8737@item @emph{Arguments}:
8738@multitable @columnfractions .15 .70
8739@item @var{I} @tab Shall be of the type @code{INTEGER}.
8740@end multitable
8741
8742@item @emph{Return value}:
8743Returns a @code{LOGICAL} of the default kind, which @code{.TRUE.} if
8744@var{I} has the value which indicates an end of file condition for
8745@code{IOSTAT=} specifiers, and is @code{.FALSE.} otherwise.
8746
8747@item @emph{Example}:
8748@smallexample
8749PROGRAM iostat
8750  IMPLICIT NONE
8751  INTEGER :: stat, i(50)
8752  OPEN(88, FILE='test.dat', FORM='UNFORMATTED')
8753  READ(88, IOSTAT=stat) i
8754  IF(IS_IOSTAT_EOR(stat)) STOP 'END OF RECORD'
8755END PROGRAM
8756@end smallexample
8757@end table
8758
8759
8760@node ISATTY
8761@section @code{ISATTY} --- Whether a unit is a terminal device.
8762@fnindex ISATTY
8763@cindex system, terminal
8764
8765@table @asis
8766@item @emph{Description}:
8767Determine whether a unit is connected to a terminal device.
8768
8769@item @emph{Standard}:
8770GNU extension
8771
8772@item @emph{Class}:
8773Function
8774
8775@item @emph{Syntax}:
8776@code{RESULT = ISATTY(UNIT)}
8777
8778@item @emph{Arguments}:
8779@multitable @columnfractions .15 .70
8780@item @var{UNIT} @tab Shall be a scalar @code{INTEGER}.
8781@end multitable
8782
8783@item @emph{Return value}:
8784Returns @code{.TRUE.} if the @var{UNIT} is connected to a terminal
8785device, @code{.FALSE.} otherwise.
8786
8787@item @emph{Example}:
8788@smallexample
8789PROGRAM test_isatty
8790  INTEGER(kind=1) :: unit
8791  DO unit = 1, 10
8792    write(*,*) isatty(unit=unit)
8793  END DO
8794END PROGRAM
8795@end smallexample
8796@item @emph{See also}:
8797@ref{TTYNAM}
8798@end table
8799
8800
8801
8802@node ISHFT
8803@section @code{ISHFT} --- Shift bits
8804@fnindex ISHFT
8805@fnindex BSHFT
8806@fnindex IISHFT
8807@fnindex JISHFT
8808@fnindex KISHFT
8809@cindex bits, shift
8810
8811@table @asis
8812@item @emph{Description}:
8813@code{ISHFT} returns a value corresponding to @var{I} with all of the
8814bits shifted @var{SHIFT} places.  A value of @var{SHIFT} greater than
8815zero corresponds to a left shift, a value of zero corresponds to no
8816shift, and a value less than zero corresponds to a right shift.  If the
8817absolute value of @var{SHIFT} is greater than @code{BIT_SIZE(I)}, the
8818value is undefined.  Bits shifted out from the left end or right end are
8819lost; zeros are shifted in from the opposite end.
8820
8821@item @emph{Standard}:
8822Fortran 90 and later, has overloads that are GNU extensions
8823
8824@item @emph{Class}:
8825Elemental function
8826
8827@item @emph{Syntax}:
8828@code{RESULT = ISHFT(I, SHIFT)}
8829
8830@item @emph{Arguments}:
8831@multitable @columnfractions .15 .70
8832@item @var{I} @tab The type shall be @code{INTEGER}.
8833@item @var{SHIFT} @tab The type shall be @code{INTEGER}.
8834@end multitable
8835
8836@item @emph{Return value}:
8837The return value is of type @code{INTEGER} and of the same kind as
8838@var{I}.
8839
8840@item @emph{Specific names}:
8841@multitable @columnfractions .20 .20 .20 .25
8842@item Name            @tab Argument            @tab Return type       @tab Standard
8843@item @code{ISHFT(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
8844@item @code{BSHFT(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
8845@item @code{IISHFT(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
8846@item @code{JISHFT(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
8847@item @code{KISHFT(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
8848@end multitable
8849
8850@item @emph{See also}:
8851@ref{ISHFTC}
8852@end table
8853
8854
8855
8856@node ISHFTC
8857@section @code{ISHFTC} --- Shift bits circularly
8858@fnindex ISHFTC
8859@fnindex BSHFTC
8860@fnindex IISHFTC
8861@fnindex JISHFTC
8862@fnindex KISHFTC
8863@cindex bits, shift circular
8864
8865@table @asis
8866@item @emph{Description}:
8867@code{ISHFTC} returns a value corresponding to @var{I} with the
8868rightmost @var{SIZE} bits shifted circularly @var{SHIFT} places; that
8869is, bits shifted out one end are shifted into the opposite end.  A value
8870of @var{SHIFT} greater than zero corresponds to a left shift, a value of
8871zero corresponds to no shift, and a value less than zero corresponds to
8872a right shift.  The absolute value of @var{SHIFT} must be less than
8873@var{SIZE}.  If the @var{SIZE} argument is omitted, it is taken to be
8874equivalent to @code{BIT_SIZE(I)}.
8875
8876@item @emph{Standard}:
8877Fortran 90 and later, has overloads that are GNU extensions
8878
8879@item @emph{Class}:
8880Elemental function
8881
8882@item @emph{Syntax}:
8883@code{RESULT = ISHFTC(I, SHIFT [, SIZE])}
8884
8885@item @emph{Arguments}:
8886@multitable @columnfractions .15 .70
8887@item @var{I} @tab The type shall be @code{INTEGER}.
8888@item @var{SHIFT} @tab The type shall be @code{INTEGER}.
8889@item @var{SIZE} @tab (Optional) The type shall be @code{INTEGER};
8890the value must be greater than zero and less than or equal to
8891@code{BIT_SIZE(I)}.
8892@end multitable
8893
8894@item @emph{Return value}:
8895The return value is of type @code{INTEGER} and of the same kind as
8896@var{I}.
8897
8898@item @emph{Specific names}:
8899@multitable @columnfractions .20 .20 .20 .25
8900@item Name            @tab Argument            @tab Return type       @tab Standard
8901@item @code{ISHFTC(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
8902@item @code{BSHFTC(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
8903@item @code{IISHFTC(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
8904@item @code{JISHFTC(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
8905@item @code{KISHFTC(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
8906@end multitable
8907
8908@item @emph{See also}:
8909@ref{ISHFT}
8910@end table
8911
8912
8913
8914@node ISNAN
8915@section @code{ISNAN} --- Test for a NaN
8916@fnindex ISNAN
8917@cindex IEEE, ISNAN
8918
8919@table @asis
8920@item @emph{Description}:
8921@code{ISNAN} tests whether a floating-point value is an IEEE
8922Not-a-Number (NaN).
8923@item @emph{Standard}:
8924GNU extension
8925
8926@item @emph{Class}:
8927Elemental function
8928
8929@item @emph{Syntax}:
8930@code{ISNAN(X)}
8931
8932@item @emph{Arguments}:
8933@multitable @columnfractions .15 .70
8934@item @var{X} @tab Variable of the type @code{REAL}.
8935
8936@end multitable
8937
8938@item @emph{Return value}:
8939Returns a default-kind @code{LOGICAL}. The returned value is @code{TRUE}
8940if @var{X} is a NaN and @code{FALSE} otherwise.
8941
8942@item @emph{Example}:
8943@smallexample
8944program test_nan
8945  implicit none
8946  real :: x
8947  x = -1.0
8948  x = sqrt(x)
8949  if (isnan(x)) stop '"x" is a NaN'
8950end program test_nan
8951@end smallexample
8952@end table
8953
8954
8955
8956@node ITIME
8957@section @code{ITIME} --- Get current local time subroutine (hour/minutes/seconds)
8958@fnindex ITIME
8959@cindex time, current
8960@cindex current time
8961
8962@table @asis
8963@item @emph{Description}:
8964@code{ITIME(VALUES)} Fills @var{VALUES} with the numerical values at the
8965current local time. The hour (in the range 1-24), minute (in the range 1-60),
8966and seconds (in the range 1-60) appear in elements 1, 2, and 3 of @var{VALUES},
8967respectively.
8968
8969This intrinsic routine is provided for backwards compatibility with
8970GNU Fortran 77.  In new code, programmers should consider the use of
8971the @ref{DATE_AND_TIME} intrinsic defined by the Fortran 95
8972standard.
8973
8974@item @emph{Standard}:
8975GNU extension
8976
8977@item @emph{Class}:
8978Subroutine
8979
8980@item @emph{Syntax}:
8981@code{CALL ITIME(VALUES)}
8982
8983@item @emph{Arguments}:
8984@multitable @columnfractions .15 .70
8985@item @var{VALUES} @tab The type shall be @code{INTEGER, DIMENSION(3)}
8986and the kind shall be the default integer kind.
8987@end multitable
8988
8989@item @emph{Return value}:
8990Does not return anything.
8991
8992
8993@item @emph{Example}:
8994@smallexample
8995program test_itime
8996  integer, dimension(3) :: tarray
8997  call itime(tarray)
8998  print *, tarray(1)
8999  print *, tarray(2)
9000  print *, tarray(3)
9001end program test_itime
9002@end smallexample
9003
9004@item @emph{See also}:
9005@ref{DATE_AND_TIME}
9006@end table
9007
9008
9009
9010@node KILL
9011@section @code{KILL} --- Send a signal to a process
9012@fnindex KILL
9013
9014@table @asis
9015@item @emph{Description}:
9016Sends the signal specified by @var{SIG} to the process @var{PID}.
9017See @code{kill(2)}.
9018
9019This intrinsic is provided in both subroutine and function forms;
9020however, only one form can be used in any given program unit.
9021@item @emph{Standard}:
9022GNU extension
9023
9024@item @emph{Standard}:
9025GNU extension
9026
9027@item @emph{Class}:
9028Subroutine, function
9029
9030@item @emph{Syntax}:
9031@multitable @columnfractions .80
9032@item @code{CALL KILL(PID, SIG [, STATUS])}
9033@item @code{STATUS = KILL(PID, SIG)}
9034@end multitable
9035
9036@item @emph{Arguments}:
9037@multitable @columnfractions .15 .70
9038@item @var{PID} @tab Shall be a scalar @code{INTEGER} with @code{INTENT(IN)}.
9039@item @var{SIG} @tab Shall be a scalar @code{INTEGER} with @code{INTENT(IN)}.
9040@item @var{STATUS} @tab [Subroutine](Optional)
9041Shall be a scalar @code{INTEGER}.
9042Returns 0 on success; otherwise a system-specific error code is returned.
9043@item @var{STATUS} @tab [Function] The kind type parameter is that of
9044@code{pid}.
9045Returns 0 on success; otherwise a system-specific error code is returned.
9046@end multitable
9047
9048@item @emph{See also}:
9049@ref{ABORT}, @gol
9050@ref{EXIT}
9051@end table
9052
9053
9054@node KIND
9055@section @code{KIND} --- Kind of an entity
9056@fnindex KIND
9057@cindex kind
9058
9059@table @asis
9060@item @emph{Description}:
9061@code{KIND(X)} returns the kind value of the entity @var{X}.
9062
9063@item @emph{Standard}:
9064Fortran 95 and later
9065
9066@item @emph{Class}:
9067Inquiry function
9068
9069@item @emph{Syntax}:
9070@code{K = KIND(X)}
9071
9072@item @emph{Arguments}:
9073@multitable @columnfractions .15 .70
9074@item @var{X} @tab Shall be of type @code{LOGICAL}, @code{INTEGER},
9075@code{REAL}, @code{COMPLEX} or @code{CHARACTER}.  It may be scalar or
9076array valued.
9077@end multitable
9078
9079@item @emph{Return value}:
9080The return value is a scalar of type @code{INTEGER} and of the default
9081integer kind.
9082
9083@item @emph{Example}:
9084@smallexample
9085program test_kind
9086  integer,parameter :: kc = kind(' ')
9087  integer,parameter :: kl = kind(.true.)
9088
9089  print *, "The default character kind is ", kc
9090  print *, "The default logical kind is ", kl
9091end program test_kind
9092@end smallexample
9093
9094@end table
9095
9096
9097
9098@node LBOUND
9099@section @code{LBOUND} --- Lower dimension bounds of an array
9100@fnindex LBOUND
9101@cindex array, lower bound
9102
9103@table @asis
9104@item @emph{Description}:
9105Returns the lower bounds of an array, or a single lower bound
9106along the @var{DIM} dimension.
9107@item @emph{Standard}:
9108Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
9109
9110@item @emph{Class}:
9111Inquiry function
9112
9113@item @emph{Syntax}:
9114@code{RESULT = LBOUND(ARRAY [, DIM [, KIND]])}
9115
9116@item @emph{Arguments}:
9117@multitable @columnfractions .15 .70
9118@item @var{ARRAY} @tab Shall be an array, of any type.
9119@item @var{DIM} @tab (Optional) Shall be a scalar @code{INTEGER}.
9120@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
9121expression indicating the kind parameter of the result.
9122@end multitable
9123
9124@item @emph{Return value}:
9125The return value is of type @code{INTEGER} and of kind @var{KIND}. If
9126@var{KIND} is absent, the return value is of default integer kind.
9127If @var{DIM} is absent, the result is an array of the lower bounds of
9128@var{ARRAY}.  If @var{DIM} is present, the result is a scalar
9129corresponding to the lower bound of the array along that dimension.  If
9130@var{ARRAY} is an expression rather than a whole array or array
9131structure component, or if it has a zero extent along the relevant
9132dimension, the lower bound is taken to be 1.
9133
9134@item @emph{See also}:
9135@ref{UBOUND}, @gol
9136@ref{LCOBOUND}
9137@end table
9138
9139
9140
9141@node LCOBOUND
9142@section @code{LCOBOUND} --- Lower codimension bounds of an array
9143@fnindex LCOBOUND
9144@cindex coarray, lower bound
9145
9146@table @asis
9147@item @emph{Description}:
9148Returns the lower bounds of a coarray, or a single lower cobound
9149along the @var{DIM} codimension.
9150@item @emph{Standard}:
9151Fortran 2008 and later
9152
9153@item @emph{Class}:
9154Inquiry function
9155
9156@item @emph{Syntax}:
9157@code{RESULT = LCOBOUND(COARRAY [, DIM [, KIND]])}
9158
9159@item @emph{Arguments}:
9160@multitable @columnfractions .15 .70
9161@item @var{ARRAY} @tab Shall be an coarray, of any type.
9162@item @var{DIM} @tab (Optional) Shall be a scalar @code{INTEGER}.
9163@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
9164expression indicating the kind parameter of the result.
9165@end multitable
9166
9167@item @emph{Return value}:
9168The return value is of type @code{INTEGER} and of kind @var{KIND}. If
9169@var{KIND} is absent, the return value is of default integer kind.
9170If @var{DIM} is absent, the result is an array of the lower cobounds of
9171@var{COARRAY}.  If @var{DIM} is present, the result is a scalar
9172corresponding to the lower cobound of the array along that codimension.
9173
9174@item @emph{See also}:
9175@ref{UCOBOUND}, @gol
9176@ref{LBOUND}
9177@end table
9178
9179
9180
9181@node LEADZ
9182@section @code{LEADZ} --- Number of leading zero bits of an integer
9183@fnindex LEADZ
9184@cindex zero bits
9185
9186@table @asis
9187@item @emph{Description}:
9188@code{LEADZ} returns the number of leading zero bits of an integer.
9189
9190@item @emph{Standard}:
9191Fortran 2008 and later
9192
9193@item @emph{Class}:
9194Elemental function
9195
9196@item @emph{Syntax}:
9197@code{RESULT = LEADZ(I)}
9198
9199@item @emph{Arguments}:
9200@multitable @columnfractions .15 .70
9201@item @var{I} @tab Shall be of type @code{INTEGER}.
9202@end multitable
9203
9204@item @emph{Return value}:
9205The type of the return value is the default @code{INTEGER}.
9206If all the bits of @code{I} are zero, the result value is @code{BIT_SIZE(I)}.
9207
9208@item @emph{Example}:
9209@smallexample
9210PROGRAM test_leadz
9211  WRITE (*,*) BIT_SIZE(1)  ! prints 32
9212  WRITE (*,*) LEADZ(1)     ! prints 31
9213END PROGRAM
9214@end smallexample
9215
9216@item @emph{See also}:
9217@ref{BIT_SIZE}, @gol
9218@ref{TRAILZ}, @gol
9219@ref{POPCNT}, @gol
9220@ref{POPPAR}
9221@end table
9222
9223
9224
9225@node LEN
9226@section @code{LEN} --- Length of a character entity
9227@fnindex LEN
9228@cindex string, length
9229
9230@table @asis
9231@item @emph{Description}:
9232Returns the length of a character string.  If @var{STRING} is an array,
9233the length of an element of @var{STRING} is returned.  Note that
9234@var{STRING} need not be defined when this intrinsic is invoked, since
9235only the length, not the content, of @var{STRING} is needed.
9236
9237@item @emph{Standard}:
9238Fortran 77 and later, with @var{KIND} argument Fortran 2003 and later
9239
9240@item @emph{Class}:
9241Inquiry function
9242
9243@item @emph{Syntax}:
9244@code{L = LEN(STRING [, KIND])}
9245
9246@item @emph{Arguments}:
9247@multitable @columnfractions .15 .70
9248@item @var{STRING} @tab Shall be a scalar or array of type
9249@code{CHARACTER}, with @code{INTENT(IN)}
9250@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
9251expression indicating the kind parameter of the result.
9252@end multitable
9253
9254@item @emph{Return value}:
9255The return value is of type @code{INTEGER} and of kind @var{KIND}. If
9256@var{KIND} is absent, the return value is of default integer kind.
9257
9258
9259@item @emph{Specific names}:
9260@multitable @columnfractions .20 .20 .20 .25
9261@item Name               @tab Argument          @tab Return type       @tab Standard
9262@item @code{LEN(STRING)} @tab @code{CHARACTER}  @tab @code{INTEGER}    @tab Fortran 77 and later
9263@end multitable
9264
9265
9266@item @emph{See also}:
9267@ref{LEN_TRIM}, @gol
9268@ref{ADJUSTL}, @gol
9269@ref{ADJUSTR}
9270@end table
9271
9272
9273
9274@node LEN_TRIM
9275@section @code{LEN_TRIM} --- Length of a character entity without trailing blank characters
9276@fnindex LEN_TRIM
9277@cindex string, length, without trailing whitespace
9278
9279@table @asis
9280@item @emph{Description}:
9281Returns the length of a character string, ignoring any trailing blanks.
9282
9283@item @emph{Standard}:
9284Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
9285
9286@item @emph{Class}:
9287Elemental function
9288
9289@item @emph{Syntax}:
9290@code{RESULT = LEN_TRIM(STRING [, KIND])}
9291
9292@item @emph{Arguments}:
9293@multitable @columnfractions .15 .70
9294@item @var{STRING} @tab Shall be a scalar of type @code{CHARACTER},
9295with @code{INTENT(IN)}
9296@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
9297expression indicating the kind parameter of the result.
9298@end multitable
9299
9300@item @emph{Return value}:
9301The return value is of type @code{INTEGER} and of kind @var{KIND}. If
9302@var{KIND} is absent, the return value is of default integer kind.
9303
9304@item @emph{See also}:
9305@ref{LEN}, @gol
9306@ref{ADJUSTL}, @gol
9307@ref{ADJUSTR}
9308@end table
9309
9310
9311
9312@node LGE
9313@section @code{LGE} --- Lexical greater than or equal
9314@fnindex LGE
9315@cindex lexical comparison of strings
9316@cindex string, comparison
9317
9318@table @asis
9319@item @emph{Description}:
9320Determines whether one string is lexically greater than or equal to
9321another string, where the two strings are interpreted as containing
9322ASCII character codes.  If the String A and String B are not the same
9323length, the shorter is compared as if spaces were appended to it to form
9324a value that has the same length as the longer.
9325
9326In general, the lexical comparison intrinsics @code{LGE}, @code{LGT},
9327@code{LLE}, and @code{LLT} differ from the corresponding intrinsic
9328operators @code{.GE.}, @code{.GT.}, @code{.LE.}, and @code{.LT.}, in
9329that the latter use the processor's character ordering (which is not
9330ASCII on some targets), whereas the former always use the ASCII
9331ordering.
9332
9333@item @emph{Standard}:
9334Fortran 77 and later
9335
9336@item @emph{Class}:
9337Elemental function
9338
9339@item @emph{Syntax}:
9340@code{RESULT = LGE(STRING_A, STRING_B)}
9341
9342@item @emph{Arguments}:
9343@multitable @columnfractions .15 .70
9344@item @var{STRING_A} @tab Shall be of default @code{CHARACTER} type.
9345@item @var{STRING_B} @tab Shall be of default @code{CHARACTER} type.
9346@end multitable
9347
9348@item @emph{Return value}:
9349Returns @code{.TRUE.} if @code{STRING_A >= STRING_B}, and @code{.FALSE.}
9350otherwise, based on the ASCII ordering.
9351
9352@item @emph{Specific names}:
9353@multitable @columnfractions .20 .20 .20 .25
9354@item Name                           @tab Argument          @tab Return type       @tab Standard
9355@item @code{LGE(STRING_A, STRING_B)} @tab @code{CHARACTER}  @tab @code{LOGICAL}    @tab Fortran 77 and later
9356@end multitable
9357
9358@item @emph{See also}:
9359@ref{LGT}, @gol
9360@ref{LLE}, @gol
9361@ref{LLT}
9362@end table
9363
9364
9365
9366@node LGT
9367@section @code{LGT} --- Lexical greater than
9368@fnindex LGT
9369@cindex lexical comparison of strings
9370@cindex string, comparison
9371
9372@table @asis
9373@item @emph{Description}:
9374Determines whether one string is lexically greater than another string,
9375where the two strings are interpreted as containing ASCII character
9376codes.  If the String A and String B are not the same length, the
9377shorter is compared as if spaces were appended to it to form a value
9378that has the same length as the longer.
9379
9380In general, the lexical comparison intrinsics @code{LGE}, @code{LGT},
9381@code{LLE}, and @code{LLT} differ from the corresponding intrinsic
9382operators @code{.GE.}, @code{.GT.}, @code{.LE.}, and @code{.LT.}, in
9383that the latter use the processor's character ordering (which is not
9384ASCII on some targets), whereas the former always use the ASCII
9385ordering.
9386
9387@item @emph{Standard}:
9388Fortran 77 and later
9389
9390@item @emph{Class}:
9391Elemental function
9392
9393@item @emph{Syntax}:
9394@code{RESULT = LGT(STRING_A, STRING_B)}
9395
9396@item @emph{Arguments}:
9397@multitable @columnfractions .15 .70
9398@item @var{STRING_A} @tab Shall be of default @code{CHARACTER} type.
9399@item @var{STRING_B} @tab Shall be of default @code{CHARACTER} type.
9400@end multitable
9401
9402@item @emph{Return value}:
9403Returns @code{.TRUE.} if @code{STRING_A > STRING_B}, and @code{.FALSE.}
9404otherwise, based on the ASCII ordering.
9405
9406@item @emph{Specific names}:
9407@multitable @columnfractions .20 .20 .20 .25
9408@item Name                           @tab Argument          @tab Return type       @tab Standard
9409@item @code{LGT(STRING_A, STRING_B)} @tab @code{CHARACTER}  @tab @code{LOGICAL}    @tab Fortran 77 and later
9410@end multitable
9411
9412@item @emph{See also}:
9413@ref{LGE}, @gol
9414@ref{LLE}, @gol
9415@ref{LLT}
9416@end table
9417
9418
9419
9420@node LINK
9421@section @code{LINK} --- Create a hard link
9422@fnindex LINK
9423@cindex file system, create link
9424@cindex file system, hard link
9425
9426@table @asis
9427@item @emph{Description}:
9428Makes a (hard) link from file @var{PATH1} to @var{PATH2}. A null
9429character (@code{CHAR(0)}) can be used to mark the end of the names in
9430@var{PATH1} and @var{PATH2}; otherwise, trailing blanks in the file
9431names are ignored.  If the @var{STATUS} argument is supplied, it
9432contains 0 on success or a nonzero error code upon return; see
9433@code{link(2)}.
9434
9435This intrinsic is provided in both subroutine and function forms;
9436however, only one form can be used in any given program unit.
9437
9438@item @emph{Standard}:
9439GNU extension
9440
9441@item @emph{Class}:
9442Subroutine, function
9443
9444@item @emph{Syntax}:
9445@multitable @columnfractions .80
9446@item @code{CALL LINK(PATH1, PATH2 [, STATUS])}
9447@item @code{STATUS = LINK(PATH1, PATH2)}
9448@end multitable
9449
9450@item @emph{Arguments}:
9451@multitable @columnfractions .15 .70
9452@item @var{PATH1} @tab Shall be of default @code{CHARACTER} type.
9453@item @var{PATH2} @tab Shall be of default @code{CHARACTER} type.
9454@item @var{STATUS} @tab (Optional) Shall be of default @code{INTEGER} type.
9455@end multitable
9456
9457@item @emph{See also}:
9458@ref{SYMLNK}, @gol
9459@ref{UNLINK}
9460@end table
9461
9462
9463
9464@node LLE
9465@section @code{LLE} --- Lexical less than or equal
9466@fnindex LLE
9467@cindex lexical comparison of strings
9468@cindex string, comparison
9469
9470@table @asis
9471@item @emph{Description}:
9472Determines whether one string is lexically less than or equal to another
9473string, where the two strings are interpreted as containing ASCII
9474character codes.  If the String A and String B are not the same length,
9475the shorter is compared as if spaces were appended to it to form a value
9476that has the same length as the longer.
9477
9478In general, the lexical comparison intrinsics @code{LGE}, @code{LGT},
9479@code{LLE}, and @code{LLT} differ from the corresponding intrinsic
9480operators @code{.GE.}, @code{.GT.}, @code{.LE.}, and @code{.LT.}, in
9481that the latter use the processor's character ordering (which is not
9482ASCII on some targets), whereas the former always use the ASCII
9483ordering.
9484
9485@item @emph{Standard}:
9486Fortran 77 and later
9487
9488@item @emph{Class}:
9489Elemental function
9490
9491@item @emph{Syntax}:
9492@code{RESULT = LLE(STRING_A, STRING_B)}
9493
9494@item @emph{Arguments}:
9495@multitable @columnfractions .15 .70
9496@item @var{STRING_A} @tab Shall be of default @code{CHARACTER} type.
9497@item @var{STRING_B} @tab Shall be of default @code{CHARACTER} type.
9498@end multitable
9499
9500@item @emph{Return value}:
9501Returns @code{.TRUE.} if @code{STRING_A <= STRING_B}, and @code{.FALSE.}
9502otherwise, based on the ASCII ordering.
9503
9504@item @emph{Specific names}:
9505@multitable @columnfractions .20 .20 .20 .25
9506@item Name                           @tab Argument          @tab Return type       @tab Standard
9507@item @code{LLE(STRING_A, STRING_B)} @tab @code{CHARACTER}  @tab @code{LOGICAL}    @tab Fortran 77 and later
9508@end multitable
9509
9510@item @emph{See also}:
9511@ref{LGE}, @gol
9512@ref{LGT}, @gol
9513@ref{LLT}
9514@end table
9515
9516
9517
9518@node LLT
9519@section @code{LLT} --- Lexical less than
9520@fnindex LLT
9521@cindex lexical comparison of strings
9522@cindex string, comparison
9523
9524@table @asis
9525@item @emph{Description}:
9526Determines whether one string is lexically less than another string,
9527where the two strings are interpreted as containing ASCII character
9528codes.  If the String A and String B are not the same length, the
9529shorter is compared as if spaces were appended to it to form a value
9530that has the same length as the longer.
9531
9532In general, the lexical comparison intrinsics @code{LGE}, @code{LGT},
9533@code{LLE}, and @code{LLT} differ from the corresponding intrinsic
9534operators @code{.GE.}, @code{.GT.}, @code{.LE.}, and @code{.LT.}, in
9535that the latter use the processor's character ordering (which is not
9536ASCII on some targets), whereas the former always use the ASCII
9537ordering.
9538
9539@item @emph{Standard}:
9540Fortran 77 and later
9541
9542@item @emph{Class}:
9543Elemental function
9544
9545@item @emph{Syntax}:
9546@code{RESULT = LLT(STRING_A, STRING_B)}
9547
9548@item @emph{Arguments}:
9549@multitable @columnfractions .15 .70
9550@item @var{STRING_A} @tab Shall be of default @code{CHARACTER} type.
9551@item @var{STRING_B} @tab Shall be of default @code{CHARACTER} type.
9552@end multitable
9553
9554@item @emph{Return value}:
9555Returns @code{.TRUE.} if @code{STRING_A < STRING_B}, and @code{.FALSE.}
9556otherwise, based on the ASCII ordering.
9557
9558@item @emph{Specific names}:
9559@multitable @columnfractions .20 .20 .20 .25
9560@item Name                           @tab Argument          @tab Return type       @tab Standard
9561@item @code{LLT(STRING_A, STRING_B)} @tab @code{CHARACTER}  @tab @code{LOGICAL}    @tab Fortran 77 and later
9562@end multitable
9563
9564@item @emph{See also}:
9565@ref{LGE}, @gol
9566@ref{LGT}, @gol
9567@ref{LLE}
9568@end table
9569
9570
9571
9572@node LNBLNK
9573@section @code{LNBLNK} --- Index of the last non-blank character in a string
9574@fnindex LNBLNK
9575@cindex string, find non-blank character
9576
9577@table @asis
9578@item @emph{Description}:
9579Returns the length of a character string, ignoring any trailing blanks.
9580This is identical to the standard @code{LEN_TRIM} intrinsic, and is only
9581included for backwards compatibility.
9582
9583@item @emph{Standard}:
9584GNU extension
9585
9586@item @emph{Class}:
9587Elemental function
9588
9589@item @emph{Syntax}:
9590@code{RESULT = LNBLNK(STRING)}
9591
9592@item @emph{Arguments}:
9593@multitable @columnfractions .15 .70
9594@item @var{STRING} @tab Shall be a scalar of type @code{CHARACTER},
9595with @code{INTENT(IN)}
9596@end multitable
9597
9598@item @emph{Return value}:
9599The return value is of @code{INTEGER(kind=4)} type.
9600
9601@item @emph{See also}:
9602@ref{INDEX intrinsic}, @gol
9603@ref{LEN_TRIM}
9604@end table
9605
9606
9607
9608@node LOC
9609@section @code{LOC} --- Returns the address of a variable
9610@fnindex LOC
9611@cindex location of a variable in memory
9612
9613@table @asis
9614@item @emph{Description}:
9615@code{LOC(X)} returns the address of @var{X} as an integer.
9616
9617@item @emph{Standard}:
9618GNU extension
9619
9620@item @emph{Class}:
9621Inquiry function
9622
9623@item @emph{Syntax}:
9624@code{RESULT = LOC(X)}
9625
9626@item @emph{Arguments}:
9627@multitable @columnfractions .15 .70
9628@item @var{X} @tab Variable of any type.
9629@end multitable
9630
9631@item @emph{Return value}:
9632The return value is of type @code{INTEGER}, with a @code{KIND}
9633corresponding to the size (in bytes) of a memory address on the target
9634machine.
9635
9636@item @emph{Example}:
9637@smallexample
9638program test_loc
9639  integer :: i
9640  real :: r
9641  i = loc(r)
9642  print *, i
9643end program test_loc
9644@end smallexample
9645@end table
9646
9647
9648
9649@node LOG
9650@section @code{LOG} --- Natural logarithm function
9651@fnindex LOG
9652@fnindex ALOG
9653@fnindex DLOG
9654@fnindex CLOG
9655@fnindex ZLOG
9656@fnindex CDLOG
9657@cindex exponential function, inverse
9658@cindex logarithm function
9659@cindex natural logarithm function
9660
9661@table @asis
9662@item @emph{Description}:
9663@code{LOG(X)} computes the natural logarithm of @var{X}, i.e. the
9664logarithm to the base @math{e}.
9665
9666@item @emph{Standard}:
9667Fortran 77 and later, has GNU extensions
9668
9669@item @emph{Class}:
9670Elemental function
9671
9672@item @emph{Syntax}:
9673@code{RESULT = LOG(X)}
9674
9675@item @emph{Arguments}:
9676@multitable @columnfractions .15 .70
9677@item @var{X} @tab The type shall be @code{REAL} or
9678@code{COMPLEX}.
9679@end multitable
9680
9681@item @emph{Return value}:
9682The return value is of type @code{REAL} or @code{COMPLEX}.
9683The kind type parameter is the same as @var{X}.
9684If @var{X} is @code{COMPLEX}, the imaginary part @math{\omega} is in the range
9685@math{-\pi < \omega \leq \pi}.
9686
9687@item @emph{Example}:
9688@smallexample
9689program test_log
9690  real(8) :: x = 2.7182818284590451_8
9691  complex :: z = (1.0, 2.0)
9692  x = log(x)    ! will yield (approximately) 1
9693  z = log(z)
9694end program test_log
9695@end smallexample
9696
9697@item @emph{Specific names}:
9698@multitable @columnfractions .20 .20 .20 .25
9699@item Name            @tab Argument          @tab Return type       @tab Standard
9700@item @code{ALOG(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab Fortran 77 or later
9701@item @code{DLOG(X)}  @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab Fortran 77 or later
9702@item @code{CLOG(X)}  @tab @code{COMPLEX(4) X}  @tab @code{COMPLEX(4)}    @tab Fortran 77 or later
9703@item @code{ZLOG(X)}  @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)}    @tab GNU extension
9704@item @code{CDLOG(X)} @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)}    @tab GNU extension
9705@end multitable
9706@end table
9707
9708
9709
9710@node LOG10
9711@section @code{LOG10} --- Base 10 logarithm function
9712@fnindex LOG10
9713@fnindex ALOG10
9714@fnindex DLOG10
9715@cindex exponential function, inverse
9716@cindex logarithm function with base 10
9717@cindex base 10 logarithm function
9718
9719@table @asis
9720@item @emph{Description}:
9721@code{LOG10(X)} computes the base 10 logarithm of @var{X}.
9722
9723@item @emph{Standard}:
9724Fortran 77 and later
9725
9726@item @emph{Class}:
9727Elemental function
9728
9729@item @emph{Syntax}:
9730@code{RESULT = LOG10(X)}
9731
9732@item @emph{Arguments}:
9733@multitable @columnfractions .15 .70
9734@item @var{X} @tab The type shall be @code{REAL}.
9735@end multitable
9736
9737@item @emph{Return value}:
9738The return value is of type @code{REAL} or @code{COMPLEX}.
9739The kind type parameter is the same as @var{X}.
9740
9741@item @emph{Example}:
9742@smallexample
9743program test_log10
9744  real(8) :: x = 10.0_8
9745  x = log10(x)
9746end program test_log10
9747@end smallexample
9748
9749@item @emph{Specific names}:
9750@multitable @columnfractions .20 .20 .20 .25
9751@item Name            @tab Argument          @tab Return type       @tab Standard
9752@item @code{ALOG10(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab Fortran 77 and later
9753@item @code{DLOG10(X)}  @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab Fortran 77 and later
9754@end multitable
9755@end table
9756
9757
9758
9759@node LOG_GAMMA
9760@section @code{LOG_GAMMA} --- Logarithm of the Gamma function
9761@fnindex LOG_GAMMA
9762@fnindex LGAMMA
9763@fnindex ALGAMA
9764@fnindex DLGAMA
9765@cindex Gamma function, logarithm of
9766
9767@table @asis
9768@item @emph{Description}:
9769@code{LOG_GAMMA(X)} computes the natural logarithm of the absolute value
9770of the Gamma (@math{\Gamma}) function.
9771
9772@item @emph{Standard}:
9773Fortran 2008 and later
9774
9775@item @emph{Class}:
9776Elemental function
9777
9778@item @emph{Syntax}:
9779@code{X = LOG_GAMMA(X)}
9780
9781@item @emph{Arguments}:
9782@multitable @columnfractions .15 .70
9783@item @var{X} @tab Shall be of type @code{REAL} and neither zero
9784nor a negative integer.
9785@end multitable
9786
9787@item @emph{Return value}:
9788The return value is of type @code{REAL} of the same kind as @var{X}.
9789
9790@item @emph{Example}:
9791@smallexample
9792program test_log_gamma
9793  real :: x = 1.0
9794  x = lgamma(x) ! returns 0.0
9795end program test_log_gamma
9796@end smallexample
9797
9798@item @emph{Specific names}:
9799@multitable @columnfractions .20 .20 .20 .25
9800@item Name             @tab Argument         @tab Return type       @tab Standard
9801@item @code{LGAMMA(X)} @tab @code{REAL(4) X} @tab @code{REAL(4)}    @tab GNU extension
9802@item @code{ALGAMA(X)} @tab @code{REAL(4) X} @tab @code{REAL(4)}    @tab GNU extension
9803@item @code{DLGAMA(X)} @tab @code{REAL(8) X} @tab @code{REAL(8)}    @tab GNU extension
9804@end multitable
9805
9806@item @emph{See also}:
9807Gamma function: @gol
9808@ref{GAMMA}
9809@end table
9810
9811
9812
9813@node LOGICAL
9814@section @code{LOGICAL} --- Convert to logical type
9815@fnindex LOGICAL
9816@cindex conversion, to logical
9817
9818@table @asis
9819@item @emph{Description}:
9820Converts one kind of @code{LOGICAL} variable to another.
9821
9822@item @emph{Standard}:
9823Fortran 90 and later
9824
9825@item @emph{Class}:
9826Elemental function
9827
9828@item @emph{Syntax}:
9829@code{RESULT = LOGICAL(L [, KIND])}
9830
9831@item @emph{Arguments}:
9832@multitable @columnfractions .15 .70
9833@item @var{L}    @tab The type shall be @code{LOGICAL}.
9834@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
9835expression indicating the kind parameter of the result.
9836@end multitable
9837
9838@item @emph{Return value}:
9839The return value is a @code{LOGICAL} value equal to @var{L}, with a
9840kind corresponding to @var{KIND}, or of the default logical kind if
9841@var{KIND} is not given.
9842
9843@item @emph{See also}:
9844@ref{INT}, @gol
9845@ref{REAL}, @gol
9846@ref{CMPLX}
9847@end table
9848
9849
9850
9851@node LONG
9852@section @code{LONG} --- Convert to integer type
9853@fnindex LONG
9854@cindex conversion, to integer
9855
9856@table @asis
9857@item @emph{Description}:
9858Convert to a @code{KIND=4} integer type, which is the same size as a C
9859@code{long} integer.  This is equivalent to the standard @code{INT}
9860intrinsic with an optional argument of @code{KIND=4}, and is only
9861included for backwards compatibility.
9862
9863@item @emph{Standard}:
9864GNU extension
9865
9866@item @emph{Class}:
9867Elemental function
9868
9869@item @emph{Syntax}:
9870@code{RESULT = LONG(A)}
9871
9872@item @emph{Arguments}:
9873@multitable @columnfractions .15 .70
9874@item @var{A}    @tab Shall be of type @code{INTEGER},
9875@code{REAL}, or @code{COMPLEX}.
9876@end multitable
9877
9878@item @emph{Return value}:
9879The return value is a @code{INTEGER(4)} variable.
9880
9881@item @emph{See also}:
9882@ref{INT}, @gol
9883@ref{INT2}, @gol
9884@ref{INT8}
9885@end table
9886
9887
9888
9889@node LSHIFT
9890@section @code{LSHIFT} --- Left shift bits
9891@fnindex LSHIFT
9892@cindex bits, shift left
9893
9894@table @asis
9895@item @emph{Description}:
9896@code{LSHIFT} returns a value corresponding to @var{I} with all of the
9897bits shifted left by @var{SHIFT} places.  @var{SHIFT} shall be
9898nonnegative and less than or equal to @code{BIT_SIZE(I)}, otherwise
9899the result value is undefined.  Bits shifted out from the left end are
9900lost; zeros are shifted in from the opposite end.
9901
9902This function has been superseded by the @code{ISHFT} intrinsic, which
9903is standard in Fortran 95 and later, and the @code{SHIFTL} intrinsic,
9904which is standard in Fortran 2008 and later.
9905
9906@item @emph{Standard}:
9907GNU extension
9908
9909@item @emph{Class}:
9910Elemental function
9911
9912@item @emph{Syntax}:
9913@code{RESULT = LSHIFT(I, SHIFT)}
9914
9915@item @emph{Arguments}:
9916@multitable @columnfractions .15 .70
9917@item @var{I} @tab The type shall be @code{INTEGER}.
9918@item @var{SHIFT} @tab The type shall be @code{INTEGER}.
9919@end multitable
9920
9921@item @emph{Return value}:
9922The return value is of type @code{INTEGER} and of the same kind as
9923@var{I}.
9924
9925@item @emph{See also}:
9926@ref{ISHFT}, @gol
9927@ref{ISHFTC}, @gol
9928@ref{RSHIFT}, @gol
9929@ref{SHIFTA}, @gol
9930@ref{SHIFTL}, @gol
9931@ref{SHIFTR}
9932@end table
9933
9934
9935
9936@node LSTAT
9937@section @code{LSTAT} --- Get file status
9938@fnindex LSTAT
9939@cindex file system, file status
9940
9941@table @asis
9942@item @emph{Description}:
9943@code{LSTAT} is identical to @ref{STAT}, except that if path is a
9944symbolic link, then the link itself is statted, not the file that it
9945refers to.
9946
9947The elements in @code{VALUES} are the same as described by @ref{STAT}.
9948
9949This intrinsic is provided in both subroutine and function forms;
9950however, only one form can be used in any given program unit.
9951
9952@item @emph{Standard}:
9953GNU extension
9954
9955@item @emph{Class}:
9956Subroutine, function
9957
9958@item @emph{Syntax}:
9959@multitable @columnfractions .80
9960@item @code{CALL LSTAT(NAME, VALUES [, STATUS])}
9961@item @code{STATUS = LSTAT(NAME, VALUES)}
9962@end multitable
9963
9964@item @emph{Arguments}:
9965@multitable @columnfractions .15 .70
9966@item @var{NAME}   @tab The type shall be @code{CHARACTER} of the default
9967kind, a valid path within the file system.
9968@item @var{VALUES} @tab The type shall be @code{INTEGER(4), DIMENSION(13)}.
9969@item @var{STATUS} @tab (Optional) status flag of type @code{INTEGER(4)}.
9970Returns 0 on success and a system specific error code otherwise.
9971@end multitable
9972
9973@item @emph{Example}:
9974See @ref{STAT} for an example.
9975
9976@item @emph{See also}:
9977To stat an open file: @gol
9978@ref{FSTAT} @gol
9979To stat a file: @gol
9980@ref{STAT}
9981@end table
9982
9983
9984
9985@node LTIME
9986@section @code{LTIME} --- Convert time to local time info
9987@fnindex LTIME
9988@cindex time, conversion to local time info
9989
9990@table @asis
9991@item @emph{Description}:
9992Given a system time value @var{TIME} (as provided by the @ref{TIME}
9993intrinsic), fills @var{VALUES} with values extracted from it appropriate
9994to the local time zone using @code{localtime(3)}.
9995
9996This intrinsic routine is provided for backwards compatibility with
9997GNU Fortran 77.  In new code, programmers should consider the use of
9998the @ref{DATE_AND_TIME} intrinsic defined by the Fortran 95
9999standard.
10000
10001@item @emph{Standard}:
10002GNU extension
10003
10004@item @emph{Class}:
10005Subroutine
10006
10007@item @emph{Syntax}:
10008@code{CALL LTIME(TIME, VALUES)}
10009
10010@item @emph{Arguments}:
10011@multitable @columnfractions .15 .70
10012@item @var{TIME}  @tab An @code{INTEGER} scalar expression
10013corresponding to a system time, with @code{INTENT(IN)}.
10014@item @var{VALUES} @tab A default @code{INTEGER} array with 9 elements,
10015with @code{INTENT(OUT)}.
10016@end multitable
10017
10018@item @emph{Return value}:
10019The elements of @var{VALUES} are assigned as follows:
10020@enumerate
10021@item Seconds after the minute, range 0--59 or 0--61 to allow for leap
10022seconds
10023@item Minutes after the hour, range 0--59
10024@item Hours past midnight, range 0--23
10025@item Day of month, range 1--31
10026@item Number of months since January, range 0--11
10027@item Years since 1900
10028@item Number of days since Sunday, range 0--6
10029@item Days since January 1, range 0--365
10030@item Daylight savings indicator: positive if daylight savings is in
10031effect, zero if not, and negative if the information is not available.
10032@end enumerate
10033
10034@item @emph{See also}:
10035@ref{DATE_AND_TIME}, @gol
10036@ref{CTIME}, @gol
10037@ref{GMTIME}, @gol
10038@ref{TIME}, @gol
10039@ref{TIME8}
10040@end table
10041
10042
10043
10044@node MALLOC
10045@section @code{MALLOC} --- Allocate dynamic memory
10046@fnindex MALLOC
10047@cindex pointer, cray
10048
10049@table @asis
10050@item @emph{Description}:
10051@code{MALLOC(SIZE)} allocates @var{SIZE} bytes of dynamic memory and
10052returns the address of the allocated memory. The @code{MALLOC} intrinsic
10053is an extension intended to be used with Cray pointers, and is provided
10054in GNU Fortran to allow the user to compile legacy code. For new code
10055using Fortran 95 pointers, the memory allocation intrinsic is
10056@code{ALLOCATE}.
10057
10058@item @emph{Standard}:
10059GNU extension
10060
10061@item @emph{Class}:
10062Function
10063
10064@item @emph{Syntax}:
10065@code{PTR = MALLOC(SIZE)}
10066
10067@item @emph{Arguments}:
10068@multitable @columnfractions .15 .70
10069@item @var{SIZE} @tab The type shall be @code{INTEGER}.
10070@end multitable
10071
10072@item @emph{Return value}:
10073The return value is of type @code{INTEGER(K)}, with @var{K} such that
10074variables of type @code{INTEGER(K)} have the same size as
10075C pointers (@code{sizeof(void *)}).
10076
10077@item @emph{Example}:
10078The following example demonstrates the use of @code{MALLOC} and
10079@code{FREE} with Cray pointers.
10080
10081@smallexample
10082program test_malloc
10083  implicit none
10084  integer i
10085  real*8 x(*), z
10086  pointer(ptr_x,x)
10087
10088  ptr_x = malloc(20*8)
10089  do i = 1, 20
10090    x(i) = sqrt(1.0d0 / i)
10091  end do
10092  z = 0
10093  do i = 1, 20
10094    z = z + x(i)
10095    print *, z
10096  end do
10097  call free(ptr_x)
10098end program test_malloc
10099@end smallexample
10100
10101@item @emph{See also}:
10102@ref{FREE}
10103@end table
10104
10105
10106
10107@node MASKL
10108@section @code{MASKL} --- Left justified mask
10109@fnindex MASKL
10110@cindex mask, left justified
10111
10112@table @asis
10113@item @emph{Description}:
10114@code{MASKL(I[, KIND])} has its leftmost @var{I} bits set to 1, and the
10115remaining bits set to 0.
10116
10117@item @emph{Standard}:
10118Fortran 2008 and later
10119
10120@item @emph{Class}:
10121Elemental function
10122
10123@item @emph{Syntax}:
10124@code{RESULT = MASKL(I[, KIND])}
10125
10126@item @emph{Arguments}:
10127@multitable @columnfractions .15 .70
10128@item @var{I} @tab Shall be of type @code{INTEGER}.
10129@item @var{KIND} @tab Shall be a scalar constant expression of type
10130@code{INTEGER}.
10131@end multitable
10132
10133@item @emph{Return value}:
10134The return value is of type @code{INTEGER}. If @var{KIND} is present, it
10135specifies the kind value of the return type; otherwise, it is of the
10136default integer kind.
10137
10138@item @emph{See also}:
10139@ref{MASKR}
10140@end table
10141
10142
10143
10144@node MASKR
10145@section @code{MASKR} --- Right justified mask
10146@fnindex MASKR
10147@cindex mask, right justified
10148
10149@table @asis
10150@item @emph{Description}:
10151@code{MASKL(I[, KIND])} has its rightmost @var{I} bits set to 1, and the
10152remaining bits set to 0.
10153
10154@item @emph{Standard}:
10155Fortran 2008 and later
10156
10157@item @emph{Class}:
10158Elemental function
10159
10160@item @emph{Syntax}:
10161@code{RESULT = MASKR(I[, KIND])}
10162
10163@item @emph{Arguments}:
10164@multitable @columnfractions .15 .70
10165@item @var{I} @tab Shall be of type @code{INTEGER}.
10166@item @var{KIND} @tab Shall be a scalar constant expression of type
10167@code{INTEGER}.
10168@end multitable
10169
10170@item @emph{Return value}:
10171The return value is of type @code{INTEGER}. If @var{KIND} is present, it
10172specifies the kind value of the return type; otherwise, it is of the
10173default integer kind.
10174
10175@item @emph{See also}:
10176@ref{MASKL}
10177@end table
10178
10179
10180
10181@node MATMUL
10182@section @code{MATMUL} --- matrix multiplication
10183@fnindex MATMUL
10184@cindex matrix multiplication
10185@cindex product, matrix
10186
10187@table @asis
10188@item @emph{Description}:
10189Performs a matrix multiplication on numeric or logical arguments.
10190
10191@item @emph{Standard}:
10192Fortran 90 and later
10193
10194@item @emph{Class}:
10195Transformational function
10196
10197@item @emph{Syntax}:
10198@code{RESULT = MATMUL(MATRIX_A, MATRIX_B)}
10199
10200@item @emph{Arguments}:
10201@multitable @columnfractions .15 .70
10202@item @var{MATRIX_A} @tab An array of @code{INTEGER},
10203@code{REAL}, @code{COMPLEX}, or @code{LOGICAL} type, with a rank of
10204one or two.
10205@item @var{MATRIX_B} @tab An array of @code{INTEGER},
10206@code{REAL}, or @code{COMPLEX} type if @var{MATRIX_A} is of a numeric
10207type; otherwise, an array of @code{LOGICAL} type. The rank shall be one
10208or two, and the first (or only) dimension of @var{MATRIX_B} shall be
10209equal to the last (or only) dimension of @var{MATRIX_A}.
10210@var{MATRIX_A} and @var{MATRIX_B} shall not both be rank one arrays.
10211@end multitable
10212
10213@item @emph{Return value}:
10214The matrix product of @var{MATRIX_A} and @var{MATRIX_B}.  The type and
10215kind of the result follow the usual type and kind promotion rules, as
10216for the @code{*} or @code{.AND.} operators.
10217@end table
10218
10219
10220
10221@node MAX
10222@section @code{MAX} --- Maximum value of an argument list
10223@fnindex MAX
10224@fnindex MAX0
10225@fnindex AMAX0
10226@fnindex MAX1
10227@fnindex AMAX1
10228@fnindex DMAX1
10229@cindex maximum value
10230
10231@table @asis
10232@item @emph{Description}:
10233Returns the argument with the largest (most positive) value.
10234
10235@item @emph{Standard}:
10236Fortran 77 and later
10237
10238@item @emph{Class}:
10239Elemental function
10240
10241@item @emph{Syntax}:
10242@code{RESULT = MAX(A1, A2 [, A3 [, ...]])}
10243
10244@item @emph{Arguments}:
10245@multitable @columnfractions .15 .70
10246@item @var{A1}          @tab The type shall be @code{INTEGER} or
10247@code{REAL}.
10248@item @var{A2}, @var{A3}, ... @tab An expression of the same type and kind
10249as @var{A1}.  (As a GNU extension, arguments of different kinds are
10250permitted.)
10251@end multitable
10252
10253@item @emph{Return value}:
10254The return value corresponds to the maximum value among the arguments,
10255and has the same type and kind as the first argument.
10256
10257@item @emph{Specific names}:
10258@multitable @columnfractions .20 .20 .20 .25
10259@item Name             @tab Argument             @tab Return type         @tab Standard
10260@item @code{MAX0(A1)}  @tab @code{INTEGER(4) A1} @tab @code{INTEGER(4)}   @tab Fortran 77 and later
10261@item @code{AMAX0(A1)} @tab @code{INTEGER(4) A1} @tab @code{REAL(MAX(X))} @tab Fortran 77 and later
10262@item @code{MAX1(A1)}  @tab @code{REAL A1}       @tab @code{INT(MAX(X))}  @tab Fortran 77 and later
10263@item @code{AMAX1(A1)} @tab @code{REAL(4) A1}    @tab @code{REAL(4)}      @tab Fortran 77 and later
10264@item @code{DMAX1(A1)} @tab @code{REAL(8) A1}    @tab @code{REAL(8)}      @tab Fortran 77 and later
10265@end multitable
10266
10267@item @emph{See also}:
10268@ref{MAXLOC} @gol
10269@ref{MAXVAL}, @gol
10270@ref{MIN}
10271@end table
10272
10273
10274
10275@node MAXEXPONENT
10276@section @code{MAXEXPONENT} --- Maximum exponent of a real kind
10277@fnindex MAXEXPONENT
10278@cindex model representation, maximum exponent
10279
10280@table @asis
10281@item @emph{Description}:
10282@code{MAXEXPONENT(X)} returns the maximum exponent in the model of the
10283type of @code{X}.
10284
10285@item @emph{Standard}:
10286Fortran 90 and later
10287
10288@item @emph{Class}:
10289Inquiry function
10290
10291@item @emph{Syntax}:
10292@code{RESULT = MAXEXPONENT(X)}
10293
10294@item @emph{Arguments}:
10295@multitable @columnfractions .15 .70
10296@item @var{X} @tab Shall be of type @code{REAL}.
10297@end multitable
10298
10299@item @emph{Return value}:
10300The return value is of type @code{INTEGER} and of the default integer
10301kind.
10302
10303@item @emph{Example}:
10304@smallexample
10305program exponents
10306  real(kind=4) :: x
10307  real(kind=8) :: y
10308
10309  print *, minexponent(x), maxexponent(x)
10310  print *, minexponent(y), maxexponent(y)
10311end program exponents
10312@end smallexample
10313@end table
10314
10315
10316
10317@node MAXLOC
10318@section @code{MAXLOC} --- Location of the maximum value within an array
10319@fnindex MAXLOC
10320@cindex array, location of maximum element
10321
10322@table @asis
10323@item @emph{Description}:
10324Determines the location of the element in the array with the maximum
10325value, or, if the @var{DIM} argument is supplied, determines the
10326locations of the maximum element along each row of the array in the
10327@var{DIM} direction.  If @var{MASK} is present, only the elements for
10328which @var{MASK} is @code{.TRUE.} are considered.  If more than one
10329element in the array has the maximum value, the location returned is
10330that of the first such element in array element order if the
10331@var{BACK} is not present, or is false; if @var{BACK} is true, the location
10332returned is that of the last such element. If the array has zero
10333size, or all of the elements of @var{MASK} are @code{.FALSE.}, then
10334the result is an array of zeroes.  Similarly, if @var{DIM} is supplied
10335and all of the elements of @var{MASK} along a given row are zero, the
10336result value for that row is zero.
10337
10338@item @emph{Standard}:
10339Fortran 95 and later; @var{ARRAY} of @code{CHARACTER} and the
10340@var{KIND} argument are available in Fortran 2003 and later.
10341The @var{BACK} argument is available in Fortran 2008 and later.
10342
10343@item @emph{Class}:
10344Transformational function
10345
10346@item @emph{Syntax}:
10347@multitable @columnfractions .80
10348@item @code{RESULT = MAXLOC(ARRAY, DIM [, MASK] [,KIND] [,BACK])}
10349@item @code{RESULT = MAXLOC(ARRAY [, MASK] [,KIND] [,BACK])}
10350@end multitable
10351
10352@item @emph{Arguments}:
10353@multitable @columnfractions .15 .70
10354@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER} or
10355@code{REAL}.
10356@item @var{DIM}   @tab (Optional) Shall be a scalar of type
10357@code{INTEGER}, with a value between one and the rank of @var{ARRAY},
10358inclusive.  It may not be an optional dummy argument.
10359@item @var{MASK}  @tab Shall be an array of type @code{LOGICAL},
10360and conformable with @var{ARRAY}.
10361@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
10362expression indicating the kind parameter of the result.
10363@item @var{BACK} @tab (Optional) A scalar of type @code{LOGICAL}.
10364@end multitable
10365
10366@item @emph{Return value}:
10367If @var{DIM} is absent, the result is a rank-one array with a length
10368equal to the rank of @var{ARRAY}.  If @var{DIM} is present, the result
10369is an array with a rank one less than the rank of @var{ARRAY}, and a
10370size corresponding to the size of @var{ARRAY} with the @var{DIM}
10371dimension removed.  If @var{DIM} is present and @var{ARRAY} has a rank
10372of one, the result is a scalar.   If the optional argument @var{KIND}
10373is present, the result is an integer of kind @var{KIND}, otherwise it
10374is of default kind.
10375
10376@item @emph{See also}:
10377@ref{FINDLOC}, @gol
10378@ref{MAX}, @gol
10379@ref{MAXVAL}
10380@end table
10381
10382
10383
10384@node MAXVAL
10385@section @code{MAXVAL} --- Maximum value of an array
10386@fnindex MAXVAL
10387@cindex array, maximum value
10388@cindex maximum value
10389
10390@table @asis
10391@item @emph{Description}:
10392Determines the maximum value of the elements in an array value, or, if
10393the @var{DIM} argument is supplied, determines the maximum value along
10394each row of the array in the @var{DIM} direction.  If @var{MASK} is
10395present, only the elements for which @var{MASK} is @code{.TRUE.} are
10396considered.  If the array has zero size, or all of the elements of
10397@var{MASK} are @code{.FALSE.}, then the result is @code{-HUGE(ARRAY)}
10398if @var{ARRAY} is numeric, or a string of nulls if @var{ARRAY} is of character
10399type.
10400
10401@item @emph{Standard}:
10402Fortran 90 and later
10403
10404@item @emph{Class}:
10405Transformational function
10406
10407@item @emph{Syntax}:
10408@multitable @columnfractions .80
10409@item @code{RESULT = MAXVAL(ARRAY, DIM [, MASK])}
10410@item @code{RESULT = MAXVAL(ARRAY [, MASK])}
10411@end multitable
10412
10413@item @emph{Arguments}:
10414@multitable @columnfractions .15 .70
10415@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER} or
10416@code{REAL}.
10417@item @var{DIM}   @tab (Optional) Shall be a scalar of type
10418@code{INTEGER}, with a value between one and the rank of @var{ARRAY},
10419inclusive.  It may not be an optional dummy argument.
10420@item @var{MASK}  @tab (Opional) Shall be an array of type @code{LOGICAL},
10421and conformable with @var{ARRAY}.
10422@end multitable
10423
10424@item @emph{Return value}:
10425If @var{DIM} is absent, or if @var{ARRAY} has a rank of one, the result
10426is a scalar.  If @var{DIM} is present, the result is an array with a
10427rank one less than the rank of @var{ARRAY}, and a size corresponding to
10428the size of @var{ARRAY} with the @var{DIM} dimension removed.  In all
10429cases, the result is of the same type and kind as @var{ARRAY}.
10430
10431@item @emph{See also}:
10432@ref{MAX}, @gol
10433@ref{MAXLOC}
10434@end table
10435
10436
10437
10438@node MCLOCK
10439@section @code{MCLOCK} --- Time function
10440@fnindex MCLOCK
10441@cindex time, clock ticks
10442@cindex clock ticks
10443
10444@table @asis
10445@item @emph{Description}:
10446Returns the number of clock ticks since the start of the process, based
10447on the function @code{clock(3)} in the C standard library.
10448
10449This intrinsic is not fully portable, such as to systems with 32-bit
10450@code{INTEGER} types but supporting times wider than 32 bits. Therefore,
10451the values returned by this intrinsic might be, or become, negative, or
10452numerically less than previous values, during a single run of the
10453compiled program.
10454
10455@item @emph{Standard}:
10456GNU extension
10457
10458@item @emph{Class}:
10459Function
10460
10461@item @emph{Syntax}:
10462@code{RESULT = MCLOCK()}
10463
10464@item @emph{Return value}:
10465The return value is a scalar of type @code{INTEGER(4)}, equal to the
10466number of clock ticks since the start of the process, or @code{-1} if
10467the system does not support @code{clock(3)}.
10468
10469@item @emph{See also}:
10470@ref{CTIME}, @gol
10471@ref{GMTIME}, @gol
10472@ref{LTIME}, @gol
10473@ref{MCLOCK}, @gol
10474@ref{TIME}
10475@end table
10476
10477
10478
10479@node MCLOCK8
10480@section @code{MCLOCK8} --- Time function (64-bit)
10481@fnindex MCLOCK8
10482@cindex time, clock ticks
10483@cindex clock ticks
10484
10485@table @asis
10486@item @emph{Description}:
10487Returns the number of clock ticks since the start of the process, based
10488on the function @code{clock(3)} in the C standard library.
10489
10490@emph{Warning:} this intrinsic does not increase the range of the timing
10491values over that returned by @code{clock(3)}. On a system with a 32-bit
10492@code{clock(3)}, @code{MCLOCK8} will return a 32-bit value, even though
10493it is converted to a 64-bit @code{INTEGER(8)} value. That means
10494overflows of the 32-bit value can still occur. Therefore, the values
10495returned by this intrinsic might be or become negative or numerically
10496less than previous values during a single run of the compiled program.
10497
10498@item @emph{Standard}:
10499GNU extension
10500
10501@item @emph{Class}:
10502Function
10503
10504@item @emph{Syntax}:
10505@code{RESULT = MCLOCK8()}
10506
10507@item @emph{Return value}:
10508The return value is a scalar of type @code{INTEGER(8)}, equal to the
10509number of clock ticks since the start of the process, or @code{-1} if
10510the system does not support @code{clock(3)}.
10511
10512@item @emph{See also}:
10513@ref{CTIME}, @gol
10514@ref{GMTIME}, @gol
10515@ref{LTIME}, @gol
10516@ref{MCLOCK}, @gol
10517@ref{TIME8}
10518@end table
10519
10520
10521
10522@node MERGE
10523@section @code{MERGE} --- Merge variables
10524@fnindex MERGE
10525@cindex array, merge arrays
10526@cindex array, combine arrays
10527
10528@table @asis
10529@item @emph{Description}:
10530Select values from two arrays according to a logical mask.  The result
10531is equal to @var{TSOURCE} if @var{MASK} is @code{.TRUE.}, or equal to
10532@var{FSOURCE} if it is @code{.FALSE.}.
10533
10534@item @emph{Standard}:
10535Fortran 90 and later
10536
10537@item @emph{Class}:
10538Elemental function
10539
10540@item @emph{Syntax}:
10541@code{RESULT = MERGE(TSOURCE, FSOURCE, MASK)}
10542
10543@item @emph{Arguments}:
10544@multitable @columnfractions .15 .70
10545@item @var{TSOURCE} @tab May be of any type.
10546@item @var{FSOURCE} @tab Shall be of the same type and type parameters
10547as @var{TSOURCE}.
10548@item @var{MASK}    @tab Shall be of type @code{LOGICAL}.
10549@end multitable
10550
10551@item @emph{Return value}:
10552The result is of the same type and type parameters as @var{TSOURCE}.
10553
10554@end table
10555
10556
10557
10558@node MERGE_BITS
10559@section @code{MERGE_BITS} --- Merge of bits under mask
10560@fnindex MERGE_BITS
10561@cindex bits, merge
10562
10563@table @asis
10564@item @emph{Description}:
10565@code{MERGE_BITS(I, J, MASK)} merges the bits of @var{I} and @var{J}
10566as determined by the mask.  The i-th bit of the result is equal to the
10567i-th bit of @var{I} if the i-th bit of @var{MASK} is 1; it is equal to
10568the i-th bit of @var{J} otherwise.
10569
10570@item @emph{Standard}:
10571Fortran 2008 and later
10572
10573@item @emph{Class}:
10574Elemental function
10575
10576@item @emph{Syntax}:
10577@code{RESULT = MERGE_BITS(I, J, MASK)}
10578
10579@item @emph{Arguments}:
10580@multitable @columnfractions .15 .70
10581@item @var{I} @tab Shall be of type @code{INTEGER} or a boz-literal-constant.
10582@item @var{J} @tab Shall be of type @code{INTEGER} with the same
10583kind type parameter as @var{I} or a boz-literal-constant.
10584@var{I} and @var{J} shall not both be boz-literal-constants.
10585@item @var{MASK} @tab Shall be of type @code{INTEGER} or a boz-literal-constant
10586and of the same kind as @var{I}.
10587@end multitable
10588
10589@item @emph{Return value}:
10590The result is of the same type and kind as @var{I}.
10591
10592@end table
10593
10594
10595
10596@node MIN
10597@section @code{MIN} --- Minimum value of an argument list
10598@fnindex MIN
10599@fnindex MIN0
10600@fnindex AMIN0
10601@fnindex MIN1
10602@fnindex AMIN1
10603@fnindex DMIN1
10604@cindex minimum value
10605
10606@table @asis
10607@item @emph{Description}:
10608Returns the argument with the smallest (most negative) value.
10609
10610@item @emph{Standard}:
10611Fortran 77 and later
10612
10613@item @emph{Class}:
10614Elemental function
10615
10616@item @emph{Syntax}:
10617@code{RESULT = MIN(A1, A2 [, A3, ...])}
10618
10619@item @emph{Arguments}:
10620@multitable @columnfractions .15 .70
10621@item @var{A1}          @tab The type shall be @code{INTEGER} or
10622@code{REAL}.
10623@item @var{A2}, @var{A3}, ... @tab An expression of the same type and kind
10624as @var{A1}.  (As a GNU extension, arguments of different kinds are
10625permitted.)
10626@end multitable
10627
10628@item @emph{Return value}:
10629The return value corresponds to the maximum value among the arguments,
10630and has the same type and kind as the first argument.
10631
10632@item @emph{Specific names}:
10633@multitable @columnfractions .20 .20 .20 .25
10634@item Name              @tab Argument             @tab Return type        @tab Standard
10635@item @code{MIN0(A1)}   @tab @code{INTEGER(4) A1} @tab @code{INTEGER(4)}  @tab Fortran 77 and later
10636@item @code{AMIN0(A1)}  @tab @code{INTEGER(4) A1} @tab @code{REAL(4)}     @tab Fortran 77 and later
10637@item @code{MIN1(A1)}   @tab @code{REAL A1}       @tab @code{INTEGER(4)}  @tab Fortran 77 and later
10638@item @code{AMIN1(A1)}  @tab @code{REAL(4) A1}    @tab @code{REAL(4)}     @tab Fortran 77 and later
10639@item @code{DMIN1(A1)}  @tab @code{REAL(8) A1}    @tab @code{REAL(8)}     @tab Fortran 77 and later
10640@end multitable
10641
10642@item @emph{See also}:
10643@ref{MAX}, @gol
10644@ref{MINLOC}, @gol
10645@ref{MINVAL}
10646@end table
10647
10648
10649
10650@node MINEXPONENT
10651@section @code{MINEXPONENT} --- Minimum exponent of a real kind
10652@fnindex MINEXPONENT
10653@cindex model representation, minimum exponent
10654
10655@table @asis
10656@item @emph{Description}:
10657@code{MINEXPONENT(X)} returns the minimum exponent in the model of the
10658type of @code{X}.
10659
10660@item @emph{Standard}:
10661Fortran 90 and later
10662
10663@item @emph{Class}:
10664Inquiry function
10665
10666@item @emph{Syntax}:
10667@code{RESULT = MINEXPONENT(X)}
10668
10669@item @emph{Arguments}:
10670@multitable @columnfractions .15 .70
10671@item @var{X} @tab Shall be of type @code{REAL}.
10672@end multitable
10673
10674@item @emph{Return value}:
10675The return value is of type @code{INTEGER} and of the default integer
10676kind.
10677
10678@item @emph{Example}:
10679See @code{MAXEXPONENT} for an example.
10680@end table
10681
10682
10683
10684@node MINLOC
10685@section @code{MINLOC} --- Location of the minimum value within an array
10686@fnindex MINLOC
10687@cindex array, location of minimum element
10688
10689@table @asis
10690@item @emph{Description}:
10691Determines the location of the element in the array with the minimum
10692value, or, if the @var{DIM} argument is supplied, determines the
10693locations of the minimum element along each row of the array in the
10694@var{DIM} direction.  If @var{MASK} is present, only the elements for
10695which @var{MASK} is @code{.TRUE.} are considered.  If more than one
10696element in the array has the minimum value, the location returned is
10697that of the first such element in array element order if the
10698@var{BACK} is not present, or is false; if @var{BACK} is true, the location
10699returned is that of the last such element.  If the array has
10700zero size, or all of the elements of @var{MASK} are @code{.FALSE.}, then
10701the result is an array of zeroes.  Similarly, if @var{DIM} is supplied
10702and all of the elements of @var{MASK} along a given row are zero, the
10703result value for that row is zero.
10704
10705@item @emph{Standard}:
10706Fortran 90 and later; @var{ARRAY} of @code{CHARACTER} and the
10707@var{KIND} argument are available in Fortran 2003 and later.
10708The @var{BACK} argument is available in Fortran 2008 and later.
10709
10710@item @emph{Class}:
10711Transformational function
10712
10713@item @emph{Syntax}:
10714@multitable @columnfractions .80
10715@item @code{RESULT = MINLOC(ARRAY, DIM [, MASK] [,KIND] [,BACK])}
10716@item @code{RESULT = MINLOC(ARRAY [, MASK], [,KIND] [,BACK])}
10717@end multitable
10718
10719@item @emph{Arguments}:
10720@multitable @columnfractions .15 .70
10721@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER},
10722@code{REAL} or @code{CHARACTER}.
10723@item @var{DIM}   @tab (Optional) Shall be a scalar of type
10724@code{INTEGER}, with a value between one and the rank of @var{ARRAY},
10725inclusive.  It may not be an optional dummy argument.
10726@item @var{MASK}  @tab Shall be an array of type @code{LOGICAL},
10727and conformable with @var{ARRAY}.
10728@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
10729expression indicating the kind parameter of the result.
10730@item @var{BACK} @tab (Optional) A scalar of type @code{LOGICAL}.
10731@end multitable
10732
10733@item @emph{Return value}:
10734If @var{DIM} is absent, the result is a rank-one array with a length
10735equal to the rank of @var{ARRAY}.  If @var{DIM} is present, the result
10736is an array with a rank one less than the rank of @var{ARRAY}, and a
10737size corresponding to the size of @var{ARRAY} with the @var{DIM}
10738dimension removed.  If @var{DIM} is present and @var{ARRAY} has a rank
10739of one, the result is a scalar.  If the optional argument @var{KIND}
10740is present, the result is an integer of kind @var{KIND}, otherwise it
10741is of default kind.
10742
10743@item @emph{See also}:
10744@ref{FINDLOC}, @gol
10745@ref{MIN}, @gol
10746@ref{MINVAL}
10747@end table
10748
10749
10750
10751@node MINVAL
10752@section @code{MINVAL} --- Minimum value of an array
10753@fnindex MINVAL
10754@cindex array, minimum value
10755@cindex minimum value
10756
10757@table @asis
10758@item @emph{Description}:
10759Determines the minimum value of the elements in an array value, or, if
10760the @var{DIM} argument is supplied, determines the minimum value along
10761each row of the array in the @var{DIM} direction.  If @var{MASK} is
10762present, only the elements for which @var{MASK} is @code{.TRUE.} are
10763considered.  If the array has zero size, or all of the elements of
10764@var{MASK} are @code{.FALSE.}, then the result is @code{HUGE(ARRAY)} if
10765@var{ARRAY} is numeric, or a string of @code{CHAR(255)} characters if
10766@var{ARRAY} is of character type.
10767
10768@item @emph{Standard}:
10769Fortran 90 and later
10770
10771@item @emph{Class}:
10772Transformational function
10773
10774@item @emph{Syntax}:
10775@multitable @columnfractions .80
10776@item @code{RESULT = MINVAL(ARRAY, DIM [, MASK])}
10777@item @code{RESULT = MINVAL(ARRAY [, MASK])}
10778@end multitable
10779
10780@item @emph{Arguments}:
10781@multitable @columnfractions .15 .70
10782@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER} or
10783@code{REAL}.
10784@item @var{DIM}   @tab (Optional) Shall be a scalar of type
10785@code{INTEGER}, with a value between one and the rank of @var{ARRAY},
10786inclusive.  It may not be an optional dummy argument.
10787@item @var{MASK}  @tab Shall be an array of type @code{LOGICAL},
10788and conformable with @var{ARRAY}.
10789@end multitable
10790
10791@item @emph{Return value}:
10792If @var{DIM} is absent, or if @var{ARRAY} has a rank of one, the result
10793is a scalar.  If @var{DIM} is present, the result is an array with a
10794rank one less than the rank of @var{ARRAY}, and a size corresponding to
10795the size of @var{ARRAY} with the @var{DIM} dimension removed.  In all
10796cases, the result is of the same type and kind as @var{ARRAY}.
10797
10798@item @emph{See also}:
10799@ref{MIN}, @gol
10800@ref{MINLOC}
10801@end table
10802
10803
10804
10805@node MOD
10806@section @code{MOD} --- Remainder function
10807@fnindex MOD
10808@fnindex AMOD
10809@fnindex DMOD
10810@fnindex BMOD
10811@fnindex IMOD
10812@fnindex JMOD
10813@fnindex KMOD
10814@cindex remainder
10815@cindex division, remainder
10816
10817@table @asis
10818@item @emph{Description}:
10819@code{MOD(A,P)} computes the remainder of the division of A by P@.
10820
10821@item @emph{Standard}:
10822Fortran 77 and later, has overloads that are GNU extensions
10823
10824@item @emph{Class}:
10825Elemental function
10826
10827@item @emph{Syntax}:
10828@code{RESULT = MOD(A, P)}
10829
10830@item @emph{Arguments}:
10831@multitable @columnfractions .15 .70
10832@item @var{A} @tab Shall be a scalar of type @code{INTEGER} or @code{REAL}.
10833@item @var{P} @tab Shall be a scalar of the same type and kind as @var{A}
10834and not equal to zero.  (As a GNU extension, arguments of different kinds are
10835permitted.)
10836@end multitable
10837
10838@item @emph{Return value}:
10839The return value is the result of @code{A - (INT(A/P) * P)}. The type
10840and kind of the return value is the same as that of the arguments. The
10841returned value has the same sign as A and a magnitude less than the
10842magnitude of P.  (As a GNU extension, kind is the largest kind of the actual
10843arguments.)
10844
10845@item @emph{Example}:
10846@smallexample
10847program test_mod
10848  print *, mod(17,3)
10849  print *, mod(17.5,5.5)
10850  print *, mod(17.5d0,5.5)
10851  print *, mod(17.5,5.5d0)
10852
10853  print *, mod(-17,3)
10854  print *, mod(-17.5,5.5)
10855  print *, mod(-17.5d0,5.5)
10856  print *, mod(-17.5,5.5d0)
10857
10858  print *, mod(17,-3)
10859  print *, mod(17.5,-5.5)
10860  print *, mod(17.5d0,-5.5)
10861  print *, mod(17.5,-5.5d0)
10862end program test_mod
10863@end smallexample
10864
10865@item @emph{Specific names}:
10866@multitable @columnfractions .20 .20 .20 .25
10867@item Name             @tab Arguments          @tab Return type    @tab Standard
10868@item @code{MOD(A,P)}  @tab @code{INTEGER A,P} @tab @code{INTEGER} @tab Fortran 77 and later
10869@item @code{AMOD(A,P)} @tab @code{REAL(4) A,P} @tab @code{REAL(4)} @tab Fortran 77 and later
10870@item @code{DMOD(A,P)} @tab @code{REAL(8) A,P} @tab @code{REAL(8)} @tab Fortran 77 and later
10871@item @code{BMOD(A,P)}  @tab @code{INTEGER(1) A,P} @tab @code{INTEGER(1)} @tab GNU extension
10872@item @code{IMOD(A,P)}  @tab @code{INTEGER(2) A,P} @tab @code{INTEGER(2)} @tab GNU extension
10873@item @code{JMOD(A,P)}  @tab @code{INTEGER(4) A,P} @tab @code{INTEGER(4)} @tab GNU extension
10874@item @code{KMOD(A,P)}  @tab @code{INTEGER(8) A,P} @tab @code{INTEGER(8)} @tab GNU extension
10875@end multitable
10876
10877@item @emph{See also}:
10878@ref{MODULO}
10879
10880@end table
10881
10882
10883
10884@node MODULO
10885@section @code{MODULO} --- Modulo function
10886@fnindex MODULO
10887@cindex modulo
10888@cindex division, modulo
10889
10890@table @asis
10891@item @emph{Description}:
10892@code{MODULO(A,P)} computes the @var{A} modulo @var{P}.
10893
10894@item @emph{Standard}:
10895Fortran 95 and later
10896
10897@item @emph{Class}:
10898Elemental function
10899
10900@item @emph{Syntax}:
10901@code{RESULT = MODULO(A, P)}
10902
10903@item @emph{Arguments}:
10904@multitable @columnfractions .15 .70
10905@item @var{A} @tab Shall be a scalar of type @code{INTEGER} or @code{REAL}.
10906@item @var{P} @tab Shall be a scalar of the same type and kind as @var{A}.
10907It shall not be zero.  (As a GNU extension, arguments of different kinds are
10908permitted.)
10909@end multitable
10910
10911@item @emph{Return value}:
10912The type and kind of the result are those of the arguments.  (As a GNU
10913extension, kind is the largest kind of the actual arguments.)
10914@table @asis
10915@item If @var{A} and @var{P} are of type @code{INTEGER}:
10916@code{MODULO(A,P)} has the value @var{R} such that @code{A=Q*P+R}, where
10917@var{Q} is an integer and @var{R} is between 0 (inclusive) and @var{P}
10918(exclusive).
10919@item If @var{A} and @var{P} are of type @code{REAL}:
10920@code{MODULO(A,P)} has the value of @code{A - FLOOR (A / P) * P}.
10921@end table
10922The returned value has the same sign as P and a magnitude less than
10923the magnitude of P.
10924
10925@item @emph{Example}:
10926@smallexample
10927program test_modulo
10928  print *, modulo(17,3)
10929  print *, modulo(17.5,5.5)
10930
10931  print *, modulo(-17,3)
10932  print *, modulo(-17.5,5.5)
10933
10934  print *, modulo(17,-3)
10935  print *, modulo(17.5,-5.5)
10936end program
10937@end smallexample
10938
10939@item @emph{See also}:
10940@ref{MOD}
10941
10942@end table
10943
10944
10945
10946@node MOVE_ALLOC
10947@section @code{MOVE_ALLOC} --- Move allocation from one object to another
10948@fnindex MOVE_ALLOC
10949@cindex moving allocation
10950@cindex allocation, moving
10951
10952@table @asis
10953@item @emph{Description}:
10954@code{MOVE_ALLOC(FROM, TO)} moves the allocation from @var{FROM} to
10955@var{TO}.  @var{FROM} will become deallocated in the process.
10956
10957@item @emph{Standard}:
10958Fortran 2003 and later
10959
10960@item @emph{Class}:
10961Pure subroutine
10962
10963@item @emph{Syntax}:
10964@code{CALL MOVE_ALLOC(FROM, TO)}
10965
10966@item @emph{Arguments}:
10967@multitable @columnfractions .15 .70
10968@item @var{FROM}  @tab @code{ALLOCATABLE}, @code{INTENT(INOUT)}, may be
10969of any type and kind.
10970@item @var{TO} @tab @code{ALLOCATABLE}, @code{INTENT(OUT)}, shall be
10971of the same type, kind and rank as @var{FROM}.
10972@end multitable
10973
10974@item @emph{Return value}:
10975None
10976
10977@item @emph{Example}:
10978@smallexample
10979program test_move_alloc
10980    integer, allocatable :: a(:), b(:)
10981
10982    allocate(a(3))
10983    a = [ 1, 2, 3 ]
10984    call move_alloc(a, b)
10985    print *, allocated(a), allocated(b)
10986    print *, b
10987end program test_move_alloc
10988@end smallexample
10989@end table
10990
10991
10992
10993@node MVBITS
10994@section @code{MVBITS} --- Move bits from one integer to another
10995@fnindex MVBITS
10996@fnindex BMVBITS
10997@fnindex IMVBITS
10998@fnindex JMVBITS
10999@fnindex KMVBITS
11000@cindex bits, move
11001
11002@table @asis
11003@item @emph{Description}:
11004Moves @var{LEN} bits from positions @var{FROMPOS} through
11005@code{FROMPOS+LEN-1} of @var{FROM} to positions @var{TOPOS} through
11006@code{TOPOS+LEN-1} of @var{TO}. The portion of argument @var{TO} not
11007affected by the movement of bits is unchanged. The values of
11008@code{FROMPOS+LEN-1} and @code{TOPOS+LEN-1} must be less than
11009@code{BIT_SIZE(FROM)}.
11010
11011@item @emph{Standard}:
11012Fortran 90 and later, has overloads that are GNU extensions
11013
11014@item @emph{Class}:
11015Elemental subroutine
11016
11017@item @emph{Syntax}:
11018@code{CALL MVBITS(FROM, FROMPOS, LEN, TO, TOPOS)}
11019
11020@item @emph{Arguments}:
11021@multitable @columnfractions .15 .70
11022@item @var{FROM}    @tab The type shall be @code{INTEGER}.
11023@item @var{FROMPOS} @tab The type shall be @code{INTEGER}.
11024@item @var{LEN}     @tab The type shall be @code{INTEGER}.
11025@item @var{TO}      @tab The type shall be @code{INTEGER}, of the
11026same kind as @var{FROM}.
11027@item @var{TOPOS}   @tab The type shall be @code{INTEGER}.
11028@end multitable
11029
11030@item @emph{Specific names}:
11031@multitable @columnfractions .20 .20 .20 .25
11032@item Name            @tab Argument            @tab Return type       @tab Standard
11033@item @code{MVBITS(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 90 and later
11034@item @code{BMVBITS(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
11035@item @code{IMVBITS(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
11036@item @code{JMVBITS(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
11037@item @code{KMVBITS(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
11038@end multitable
11039
11040@item @emph{See also}:
11041@ref{IBCLR}, @gol
11042@ref{IBSET}, @gol
11043@ref{IBITS}, @gol
11044@ref{IAND}, @gol
11045@ref{IOR}, @gol
11046@ref{IEOR}
11047@end table
11048
11049
11050
11051@node NEAREST
11052@section @code{NEAREST} --- Nearest representable number
11053@fnindex NEAREST
11054@cindex real number, nearest different
11055@cindex floating point, nearest different
11056
11057@table @asis
11058@item @emph{Description}:
11059@code{NEAREST(X, S)} returns the processor-representable number nearest
11060to @code{X} in the direction indicated by the sign of @code{S}.
11061
11062@item @emph{Standard}:
11063Fortran 90 and later
11064
11065@item @emph{Class}:
11066Elemental function
11067
11068@item @emph{Syntax}:
11069@code{RESULT = NEAREST(X, S)}
11070
11071@item @emph{Arguments}:
11072@multitable @columnfractions .15 .70
11073@item @var{X} @tab Shall be of type @code{REAL}.
11074@item @var{S} @tab Shall be of type @code{REAL} and
11075not equal to zero.
11076@end multitable
11077
11078@item @emph{Return value}:
11079The return value is of the same type as @code{X}. If @code{S} is
11080positive, @code{NEAREST} returns the processor-representable number
11081greater than @code{X} and nearest to it. If @code{S} is negative,
11082@code{NEAREST} returns the processor-representable number smaller than
11083@code{X} and nearest to it.
11084
11085@item @emph{Example}:
11086@smallexample
11087program test_nearest
11088  real :: x, y
11089  x = nearest(42.0, 1.0)
11090  y = nearest(42.0, -1.0)
11091  write (*,"(3(G20.15))") x, y, x - y
11092end program test_nearest
11093@end smallexample
11094@end table
11095
11096
11097
11098@node NEW_LINE
11099@section @code{NEW_LINE} --- New line character
11100@fnindex NEW_LINE
11101@cindex newline
11102@cindex output, newline
11103
11104@table @asis
11105@item @emph{Description}:
11106@code{NEW_LINE(C)} returns the new-line character.
11107
11108@item @emph{Standard}:
11109Fortran 2003 and later
11110
11111@item @emph{Class}:
11112Inquiry function
11113
11114@item @emph{Syntax}:
11115@code{RESULT = NEW_LINE(C)}
11116
11117@item @emph{Arguments}:
11118@multitable @columnfractions .15 .70
11119@item @var{C}    @tab The argument shall be a scalar or array of the
11120type @code{CHARACTER}.
11121@end multitable
11122
11123@item @emph{Return value}:
11124Returns a @var{CHARACTER} scalar of length one with the new-line character of
11125the same kind as parameter @var{C}.
11126
11127@item @emph{Example}:
11128@smallexample
11129program newline
11130  implicit none
11131  write(*,'(A)') 'This is record 1.'//NEW_LINE('A')//'This is record 2.'
11132end program newline
11133@end smallexample
11134@end table
11135
11136
11137
11138@node NINT
11139@section @code{NINT} --- Nearest whole number
11140@fnindex NINT
11141@fnindex IDNINT
11142@cindex rounding, nearest whole number
11143
11144@table @asis
11145@item @emph{Description}:
11146@code{NINT(A)} rounds its argument to the nearest whole number.
11147
11148@item @emph{Standard}:
11149Fortran 77 and later, with @var{KIND} argument Fortran 90 and later
11150
11151@item @emph{Class}:
11152Elemental function
11153
11154@item @emph{Syntax}:
11155@code{RESULT = NINT(A [, KIND])}
11156
11157@item @emph{Arguments}:
11158@multitable @columnfractions .15 .70
11159@item @var{A}    @tab The type of the argument shall be @code{REAL}.
11160@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
11161expression indicating the kind parameter of the result.
11162@end multitable
11163
11164@item @emph{Return value}:
11165Returns @var{A} with the fractional portion of its magnitude eliminated by
11166rounding to the nearest whole number and with its sign preserved,
11167converted to an @code{INTEGER} of the default kind.
11168
11169@item @emph{Example}:
11170@smallexample
11171program test_nint
11172  real(4) x4
11173  real(8) x8
11174  x4 = 1.234E0_4
11175  x8 = 4.321_8
11176  print *, nint(x4), idnint(x8)
11177end program test_nint
11178@end smallexample
11179
11180@item @emph{Specific names}:
11181@multitable @columnfractions .20 .20 .20 .25
11182@item Name             @tab Argument           @tab Return Type     @tab Standard
11183@item @code{NINT(A)}   @tab @code{REAL(4) A}   @tab  @code{INTEGER} @tab Fortran 77 and later
11184@item @code{IDNINT(A)} @tab @code{REAL(8) A}   @tab  @code{INTEGER} @tab Fortran 77 and later
11185@end multitable
11186
11187@item @emph{See also}:
11188@ref{CEILING}, @gol
11189@ref{FLOOR}
11190@end table
11191
11192
11193
11194@node NORM2
11195@section @code{NORM2} --- Euclidean vector norms
11196@fnindex NORM2
11197@cindex Euclidean vector norm
11198@cindex L2 vector norm
11199@cindex norm, Euclidean
11200
11201@table @asis
11202@item @emph{Description}:
11203Calculates the Euclidean vector norm (@math{L_2} norm) of
11204of @var{ARRAY} along dimension @var{DIM}.
11205
11206@item @emph{Standard}:
11207Fortran 2008 and later
11208
11209@item @emph{Class}:
11210Transformational function
11211
11212@item @emph{Syntax}:
11213@multitable @columnfractions .80
11214@item @code{RESULT = NORM2(ARRAY[, DIM])}
11215@end multitable
11216
11217@item @emph{Arguments}:
11218@multitable @columnfractions .15 .70
11219@item @var{ARRAY} @tab Shall be an array of type @code{REAL}
11220@item @var{DIM}   @tab (Optional) shall be a scalar of type
11221@code{INTEGER} with a value in the range from 1 to n, where n
11222equals the rank of @var{ARRAY}.
11223@end multitable
11224
11225@item @emph{Return value}:
11226The result is of the same type as @var{ARRAY}.
11227
11228If @var{DIM} is absent, a scalar with the square root of the sum of all
11229elements in @var{ARRAY} squared  is returned. Otherwise, an array of
11230rank @math{n-1}, where @math{n} equals the rank of @var{ARRAY}, and a
11231shape similar to that of @var{ARRAY} with dimension @var{DIM} dropped
11232is returned.
11233
11234@item @emph{Example}:
11235@smallexample
11236PROGRAM test_sum
11237  REAL :: x(5) = [ real :: 1, 2, 3, 4, 5 ]
11238  print *, NORM2(x)  ! = sqrt(55.) ~ 7.416
11239END PROGRAM
11240@end smallexample
11241@end table
11242
11243
11244
11245@node NOT
11246@section @code{NOT} --- Logical negation
11247@fnindex NOT
11248@fnindex BNOT
11249@fnindex INOT
11250@fnindex JNOT
11251@fnindex KNOT
11252@cindex bits, negate
11253@cindex bitwise logical not
11254@cindex logical not, bitwise
11255
11256@table @asis
11257@item @emph{Description}:
11258@code{NOT} returns the bitwise Boolean inverse of @var{I}.
11259
11260@item @emph{Standard}:
11261Fortran 90 and later, has overloads that are GNU extensions
11262
11263@item @emph{Class}:
11264Elemental function
11265
11266@item @emph{Syntax}:
11267@code{RESULT = NOT(I)}
11268
11269@item @emph{Arguments}:
11270@multitable @columnfractions .15 .70
11271@item @var{I} @tab The type shall be @code{INTEGER}.
11272@end multitable
11273
11274@item @emph{Return value}:
11275The return type is @code{INTEGER}, of the same kind as the
11276argument.
11277
11278@item @emph{Specific names}:
11279@multitable @columnfractions .20 .20 .20 .25
11280@item Name            @tab Argument            @tab Return type       @tab Standard
11281@item @code{NOT(A)}   @tab @code{INTEGER A}    @tab @code{INTEGER}    @tab Fortran 95 and later
11282@item @code{BNOT(A)} @tab @code{INTEGER(1) A} @tab @code{INTEGER(1)} @tab GNU extension
11283@item @code{INOT(A)} @tab @code{INTEGER(2) A} @tab @code{INTEGER(2)} @tab GNU extension
11284@item @code{JNOT(A)} @tab @code{INTEGER(4) A} @tab @code{INTEGER(4)} @tab GNU extension
11285@item @code{KNOT(A)} @tab @code{INTEGER(8) A} @tab @code{INTEGER(8)} @tab GNU extension
11286@end multitable
11287
11288@item @emph{See also}:
11289@ref{IAND}, @gol
11290@ref{IEOR}, @gol
11291@ref{IOR}, @gol
11292@ref{IBITS}, @gol
11293@ref{IBSET}, @gol
11294@ref{IBCLR}
11295@end table
11296
11297
11298
11299@node NULL
11300@section @code{NULL} --- Function that returns an disassociated pointer
11301@fnindex NULL
11302@cindex pointer, status
11303@cindex pointer, disassociated
11304
11305@table @asis
11306@item @emph{Description}:
11307Returns a disassociated pointer.
11308
11309If @var{MOLD} is present, a disassociated pointer of the same type is
11310returned, otherwise the type is determined by context.
11311
11312In Fortran 95, @var{MOLD} is optional. Please note that Fortran 2003
11313includes cases where it is required.
11314
11315@item @emph{Standard}:
11316Fortran 95 and later
11317
11318@item @emph{Class}:
11319Transformational function
11320
11321@item @emph{Syntax}:
11322@code{PTR => NULL([MOLD])}
11323
11324@item @emph{Arguments}:
11325@multitable @columnfractions .15 .70
11326@item @var{MOLD} @tab (Optional) shall be a pointer of any association
11327status and of any type.
11328@end multitable
11329
11330@item @emph{Return value}:
11331A disassociated pointer.
11332
11333@item @emph{Example}:
11334@smallexample
11335REAL, POINTER, DIMENSION(:) :: VEC => NULL ()
11336@end smallexample
11337
11338@item @emph{See also}:
11339@ref{ASSOCIATED}
11340@end table
11341
11342
11343
11344@node NUM_IMAGES
11345@section @code{NUM_IMAGES} --- Function that returns the number of images
11346@fnindex NUM_IMAGES
11347@cindex coarray, @code{NUM_IMAGES}
11348@cindex images, number of
11349
11350@table @asis
11351@item @emph{Description}:
11352Returns the number of images.
11353
11354@item @emph{Standard}:
11355Fortran 2008 and later. With @var{DISTANCE} or @var{FAILED} argument,
11356Technical Specification (TS) 18508 or later
11357
11358
11359@item @emph{Class}:
11360Transformational function
11361
11362@item @emph{Syntax}:
11363@code{RESULT = NUM_IMAGES(DISTANCE, FAILED)}
11364
11365@item @emph{Arguments}:
11366@multitable @columnfractions .15 .70
11367@item @var{DISTANCE} @tab (optional, intent(in)) Nonnegative scalar integer
11368@item @var{FAILED}   @tab (optional, intent(in)) Scalar logical expression
11369@end multitable
11370
11371@item @emph{Return value}:
11372Scalar default-kind integer.  If @var{DISTANCE} is not present or has value 0,
11373the number of images in the current team is returned. For values smaller or
11374equal distance to the initial team, it returns the number of images index
11375on the ancestor team which has a distance of @var{DISTANCE} from the invoking
11376team. If @var{DISTANCE} is larger than the distance to the initial team, the
11377number of images of the initial team is returned. If @var{FAILED} is not present
11378the total number of images is returned; if it has the value @code{.TRUE.},
11379the number of failed images is returned, otherwise, the number of images which
11380do have not the failed status.
11381
11382@item @emph{Example}:
11383@smallexample
11384INTEGER :: value[*]
11385INTEGER :: i
11386value = THIS_IMAGE()
11387SYNC ALL
11388IF (THIS_IMAGE() == 1) THEN
11389  DO i = 1, NUM_IMAGES()
11390    WRITE(*,'(2(a,i0))') 'value[', i, '] is ', value[i]
11391  END DO
11392END IF
11393@end smallexample
11394
11395@item @emph{See also}:
11396@ref{THIS_IMAGE}, @gol
11397@ref{IMAGE_INDEX}
11398@end table
11399
11400
11401
11402@node OR
11403@section @code{OR} --- Bitwise logical OR
11404@fnindex OR
11405@cindex bitwise logical or
11406@cindex logical or, bitwise
11407
11408@table @asis
11409@item @emph{Description}:
11410Bitwise logical @code{OR}.
11411
11412This intrinsic routine is provided for backwards compatibility with
11413GNU Fortran 77.  For integer arguments, programmers should consider
11414the use of the @ref{IOR} intrinsic defined by the Fortran standard.
11415
11416@item @emph{Standard}:
11417GNU extension
11418
11419@item @emph{Class}:
11420Function
11421
11422@item @emph{Syntax}:
11423@code{RESULT = OR(I, J)}
11424
11425@item @emph{Arguments}:
11426@multitable @columnfractions .15 .70
11427@item @var{I} @tab The type shall be either a scalar @code{INTEGER}
11428type or a scalar @code{LOGICAL} type or a boz-literal-constant.
11429@item @var{J} @tab The type shall be the same as the type of @var{I} or
11430a boz-literal-constant. @var{I} and @var{J} shall not both be
11431boz-literal-constants.  If either @var{I} and @var{J} is a
11432boz-literal-constant, then the other argument must be a scalar @code{INTEGER}.
11433@end multitable
11434
11435@item @emph{Return value}:
11436The return type is either a scalar @code{INTEGER} or a scalar
11437@code{LOGICAL}.  If the kind type parameters differ, then the
11438smaller kind type is implicitly converted to larger kind, and the
11439return has the larger kind.  A boz-literal-constant is
11440converted to an @code{INTEGER} with the kind type parameter of
11441the other argument as-if a call to @ref{INT} occurred.
11442
11443@item @emph{Example}:
11444@smallexample
11445PROGRAM test_or
11446  LOGICAL :: T = .TRUE., F = .FALSE.
11447  INTEGER :: a, b
11448  DATA a / Z'F' /, b / Z'3' /
11449
11450  WRITE (*,*) OR(T, T), OR(T, F), OR(F, T), OR(F, F)
11451  WRITE (*,*) OR(a, b)
11452END PROGRAM
11453@end smallexample
11454
11455@item @emph{See also}:
11456Fortran 95 elemental function: @gol
11457@ref{IOR}
11458@end table
11459
11460
11461
11462@node PACK
11463@section @code{PACK} --- Pack an array into an array of rank one
11464@fnindex PACK
11465@cindex array, packing
11466@cindex array, reduce dimension
11467@cindex array, gather elements
11468
11469@table @asis
11470@item @emph{Description}:
11471Stores the elements of @var{ARRAY} in an array of rank one.
11472
11473The beginning of the resulting array is made up of elements whose @var{MASK}
11474equals @code{TRUE}. Afterwards, positions are filled with elements taken from
11475@var{VECTOR}.
11476
11477@item @emph{Standard}:
11478Fortran 90 and later
11479
11480@item @emph{Class}:
11481Transformational function
11482
11483@item @emph{Syntax}:
11484@code{RESULT = PACK(ARRAY, MASK[,VECTOR])}
11485
11486@item @emph{Arguments}:
11487@multitable @columnfractions .15 .70
11488@item @var{ARRAY}  @tab Shall be an array of any type.
11489@item @var{MASK}   @tab Shall be an array of type @code{LOGICAL} and
11490of the same size as @var{ARRAY}. Alternatively, it may be a @code{LOGICAL}
11491scalar.
11492@item @var{VECTOR} @tab (Optional) shall be an array of the same type
11493as @var{ARRAY} and of rank one. If present, the number of elements in
11494@var{VECTOR} shall be equal to or greater than the number of true elements
11495in @var{MASK}. If @var{MASK} is scalar, the number of elements in
11496@var{VECTOR} shall be equal to or greater than the number of elements in
11497@var{ARRAY}.
11498@end multitable
11499
11500@item @emph{Return value}:
11501The result is an array of rank one and the same type as that of @var{ARRAY}.
11502If @var{VECTOR} is present, the result size is that of @var{VECTOR}, the
11503number of @code{TRUE} values in @var{MASK} otherwise.
11504
11505@item @emph{Example}:
11506Gathering nonzero elements from an array:
11507@smallexample
11508PROGRAM test_pack_1
11509  INTEGER :: m(6)
11510  m = (/ 1, 0, 0, 0, 5, 0 /)
11511  WRITE(*, FMT="(6(I0, ' '))") pack(m, m /= 0)  ! "1 5"
11512END PROGRAM
11513@end smallexample
11514
11515Gathering nonzero elements from an array and appending elements from @var{VECTOR}:
11516@smallexample
11517PROGRAM test_pack_2
11518  INTEGER :: m(4)
11519  m = (/ 1, 0, 0, 2 /)
11520  ! The following results in "1 2 3 4"
11521  WRITE(*, FMT="(4(I0, ' '))") pack(m, m /= 0, (/ 0, 0, 3, 4 /))
11522END PROGRAM
11523@end smallexample
11524
11525@item @emph{See also}:
11526@ref{UNPACK}
11527@end table
11528
11529
11530
11531@node PARITY
11532@section @code{PARITY} --- Reduction with exclusive OR
11533@fnindex PARITY
11534@cindex Parity
11535@cindex Reduction, XOR
11536@cindex XOR reduction
11537
11538@table @asis
11539@item @emph{Description}:
11540Calculates the parity, i.e. the reduction using @code{.XOR.},
11541of @var{MASK} along dimension @var{DIM}.
11542
11543@item @emph{Standard}:
11544Fortran 2008 and later
11545
11546@item @emph{Class}:
11547Transformational function
11548
11549@item @emph{Syntax}:
11550@multitable @columnfractions .80
11551@item @code{RESULT = PARITY(MASK[, DIM])}
11552@end multitable
11553
11554@item @emph{Arguments}:
11555@multitable @columnfractions .15 .70
11556@item @var{LOGICAL} @tab Shall be an array of type @code{LOGICAL}
11557@item @var{DIM}   @tab (Optional) shall be a scalar of type
11558@code{INTEGER} with a value in the range from 1 to n, where n
11559equals the rank of @var{MASK}.
11560@end multitable
11561
11562@item @emph{Return value}:
11563The result is of the same type as @var{MASK}.
11564
11565If @var{DIM} is absent, a scalar with the parity of all elements in
11566@var{MASK} is returned, i.e. true if an odd number of elements is
11567@code{.true.} and false otherwise.  If @var{DIM} is present, an array
11568of rank @math{n-1}, where @math{n} equals the rank of @var{ARRAY},
11569and a shape similar to that of @var{MASK} with dimension @var{DIM}
11570dropped is returned.
11571
11572@item @emph{Example}:
11573@smallexample
11574PROGRAM test_sum
11575  LOGICAL :: x(2) = [ .true., .false. ]
11576  print *, PARITY(x) ! prints "T" (true).
11577END PROGRAM
11578@end smallexample
11579@end table
11580
11581
11582
11583@node PERROR
11584@section @code{PERROR} --- Print system error message
11585@fnindex PERROR
11586@cindex system, error handling
11587
11588@table @asis
11589@item @emph{Description}:
11590Prints (on the C @code{stderr} stream) a newline-terminated error
11591message corresponding to the last system error. This is prefixed by
11592@var{STRING}, a colon and a space. See @code{perror(3)}.
11593
11594@item @emph{Standard}:
11595GNU extension
11596
11597@item @emph{Class}:
11598Subroutine
11599
11600@item @emph{Syntax}:
11601@code{CALL PERROR(STRING)}
11602
11603@item @emph{Arguments}:
11604@multitable @columnfractions .15 .70
11605@item @var{STRING} @tab A scalar of type @code{CHARACTER} and of the
11606default kind.
11607@end multitable
11608
11609@item @emph{See also}:
11610@ref{IERRNO}
11611@end table
11612
11613
11614
11615@node POPCNT
11616@section @code{POPCNT} --- Number of bits set
11617@fnindex POPCNT
11618@cindex binary representation
11619@cindex bits set
11620
11621@table @asis
11622@item @emph{Description}:
11623@code{POPCNT(I)} returns the number of bits set ('1' bits) in the binary
11624representation of @code{I}.
11625
11626@item @emph{Standard}:
11627Fortran 2008 and later
11628
11629@item @emph{Class}:
11630Elemental function
11631
11632@item @emph{Syntax}:
11633@code{RESULT = POPCNT(I)}
11634
11635@item @emph{Arguments}:
11636@multitable @columnfractions .15 .70
11637@item @var{I} @tab Shall be of type @code{INTEGER}.
11638@end multitable
11639
11640@item @emph{Return value}:
11641The return value is of type @code{INTEGER} and of the default integer
11642kind.
11643
11644@item @emph{Example}:
11645@smallexample
11646program test_population
11647  print *, popcnt(127),       poppar(127)
11648  print *, popcnt(huge(0_4)), poppar(huge(0_4))
11649  print *, popcnt(huge(0_8)), poppar(huge(0_8))
11650end program test_population
11651@end smallexample
11652@item @emph{See also}:
11653@ref{POPPAR}, @gol
11654@ref{LEADZ}, @gol
11655@ref{TRAILZ}
11656@end table
11657
11658
11659
11660@node POPPAR
11661@section @code{POPPAR} --- Parity of the number of bits set
11662@fnindex POPPAR
11663@cindex binary representation
11664@cindex parity
11665
11666@table @asis
11667@item @emph{Description}:
11668@code{POPPAR(I)} returns parity of the integer @code{I}, i.e. the parity
11669of the number of bits set ('1' bits) in the binary representation of
11670@code{I}. It is equal to 0 if @code{I} has an even number of bits set,
11671and 1 for an odd number of '1' bits.
11672
11673@item @emph{Standard}:
11674Fortran 2008 and later
11675
11676@item @emph{Class}:
11677Elemental function
11678
11679@item @emph{Syntax}:
11680@code{RESULT = POPPAR(I)}
11681
11682@item @emph{Arguments}:
11683@multitable @columnfractions .15 .70
11684@item @var{I} @tab Shall be of type @code{INTEGER}.
11685@end multitable
11686
11687@item @emph{Return value}:
11688The return value is of type @code{INTEGER} and of the default integer
11689kind.
11690
11691@item @emph{Example}:
11692@smallexample
11693program test_population
11694  print *, popcnt(127),       poppar(127)
11695  print *, popcnt(huge(0_4)), poppar(huge(0_4))
11696  print *, popcnt(huge(0_8)), poppar(huge(0_8))
11697end program test_population
11698@end smallexample
11699@item @emph{See also}:
11700@ref{POPCNT}, @gol
11701@ref{LEADZ}, @gol
11702@ref{TRAILZ}
11703@end table
11704
11705
11706
11707@node PRECISION
11708@section @code{PRECISION} --- Decimal precision of a real kind
11709@fnindex PRECISION
11710@cindex model representation, precision
11711
11712@table @asis
11713@item @emph{Description}:
11714@code{PRECISION(X)} returns the decimal precision in the model of the
11715type of @code{X}.
11716
11717@item @emph{Standard}:
11718Fortran 90 and later
11719
11720@item @emph{Class}:
11721Inquiry function
11722
11723@item @emph{Syntax}:
11724@code{RESULT = PRECISION(X)}
11725
11726@item @emph{Arguments}:
11727@multitable @columnfractions .15 .70
11728@item @var{X} @tab Shall be of type @code{REAL} or @code{COMPLEX}. It may
11729be scalar or valued.
11730@end multitable
11731
11732@item @emph{Return value}:
11733The return value is of type @code{INTEGER} and of the default integer
11734kind.
11735
11736@item @emph{Example}:
11737@smallexample
11738program prec_and_range
11739  real(kind=4) :: x(2)
11740  complex(kind=8) :: y
11741
11742  print *, precision(x), range(x)
11743  print *, precision(y), range(y)
11744end program prec_and_range
11745@end smallexample
11746@item @emph{See also}:
11747@ref{SELECTED_REAL_KIND}, @gol
11748@ref{RANGE}
11749@end table
11750
11751
11752
11753@node PRESENT
11754@section @code{PRESENT} --- Determine whether an optional dummy argument is specified
11755@fnindex PRESENT
11756
11757@table @asis
11758@item @emph{Description}:
11759Determines whether an optional dummy argument is present.
11760
11761@item @emph{Standard}:
11762Fortran 90 and later
11763
11764@item @emph{Class}:
11765Inquiry function
11766
11767@item @emph{Syntax}:
11768@code{RESULT = PRESENT(A)}
11769
11770@item @emph{Arguments}:
11771@multitable @columnfractions .15 .70
11772@item @var{A} @tab May be of any type and may be a pointer, scalar or array
11773value, or a dummy procedure. It shall be the name of an optional dummy argument
11774accessible within the current subroutine or function.
11775@end multitable
11776
11777@item @emph{Return value}:
11778Returns either @code{TRUE} if the optional argument @var{A} is present, or
11779@code{FALSE} otherwise.
11780
11781@item @emph{Example}:
11782@smallexample
11783PROGRAM test_present
11784  WRITE(*,*) f(), f(42)      ! "F T"
11785CONTAINS
11786  LOGICAL FUNCTION f(x)
11787    INTEGER, INTENT(IN), OPTIONAL :: x
11788    f = PRESENT(x)
11789  END FUNCTION
11790END PROGRAM
11791@end smallexample
11792@end table
11793
11794
11795
11796@node PRODUCT
11797@section @code{PRODUCT} --- Product of array elements
11798@fnindex PRODUCT
11799@cindex array, product
11800@cindex array, multiply elements
11801@cindex array, conditionally multiply elements
11802@cindex multiply array elements
11803
11804@table @asis
11805@item @emph{Description}:
11806Multiplies the elements of @var{ARRAY} along dimension @var{DIM} if
11807the corresponding element in @var{MASK} is @code{TRUE}.
11808
11809@item @emph{Standard}:
11810Fortran 90 and later
11811
11812@item @emph{Class}:
11813Transformational function
11814
11815@item @emph{Syntax}:
11816@multitable @columnfractions .80
11817@item @code{RESULT = PRODUCT(ARRAY[, MASK])}
11818@item @code{RESULT = PRODUCT(ARRAY, DIM[, MASK])}
11819@end multitable
11820
11821@item @emph{Arguments}:
11822@multitable @columnfractions .15 .70
11823@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER},
11824@code{REAL} or @code{COMPLEX}.
11825@item @var{DIM}   @tab (Optional) shall be a scalar of type
11826@code{INTEGER} with a value in the range from 1 to n, where n
11827equals the rank of @var{ARRAY}.
11828@item @var{MASK}  @tab (Optional) shall be of type @code{LOGICAL}
11829and either be a scalar or an array of the same shape as @var{ARRAY}.
11830@end multitable
11831
11832@item @emph{Return value}:
11833The result is of the same type as @var{ARRAY}.
11834
11835If @var{DIM} is absent, a scalar with the product of all elements in
11836@var{ARRAY} is returned. Otherwise, an array of rank n-1, where n equals
11837the rank of @var{ARRAY}, and a shape similar to that of @var{ARRAY} with
11838dimension @var{DIM} dropped is returned.
11839
11840
11841@item @emph{Example}:
11842@smallexample
11843PROGRAM test_product
11844  INTEGER :: x(5) = (/ 1, 2, 3, 4 ,5 /)
11845  print *, PRODUCT(x)                    ! all elements, product = 120
11846  print *, PRODUCT(x, MASK=MOD(x, 2)==1) ! odd elements, product = 15
11847END PROGRAM
11848@end smallexample
11849
11850@item @emph{See also}:
11851@ref{SUM}
11852@end table
11853
11854
11855
11856@node RADIX
11857@section @code{RADIX} --- Base of a model number
11858@fnindex RADIX
11859@cindex model representation, base
11860@cindex model representation, radix
11861
11862@table @asis
11863@item @emph{Description}:
11864@code{RADIX(X)} returns the base of the model representing the entity @var{X}.
11865
11866@item @emph{Standard}:
11867Fortran 90 and later
11868
11869@item @emph{Class}:
11870Inquiry function
11871
11872@item @emph{Syntax}:
11873@code{RESULT = RADIX(X)}
11874
11875@item @emph{Arguments}:
11876@multitable @columnfractions .15 .70
11877@item @var{X} @tab Shall be of type @code{INTEGER} or @code{REAL}
11878@end multitable
11879
11880@item @emph{Return value}:
11881The return value is a scalar of type @code{INTEGER} and of the default
11882integer kind.
11883
11884@item @emph{Example}:
11885@smallexample
11886program test_radix
11887  print *, "The radix for the default integer kind is", radix(0)
11888  print *, "The radix for the default real kind is", radix(0.0)
11889end program test_radix
11890@end smallexample
11891@item @emph{See also}:
11892@ref{SELECTED_REAL_KIND}
11893@end table
11894
11895
11896
11897@node RAN
11898@section @code{RAN} --- Real pseudo-random number
11899@fnindex RAN
11900@cindex random number generation
11901
11902@table @asis
11903@item @emph{Description}:
11904For compatibility with HP FORTRAN 77/iX, the @code{RAN} intrinsic is
11905provided as an alias for @code{RAND}.  See @ref{RAND} for complete
11906documentation.
11907
11908@item @emph{Standard}:
11909GNU extension
11910
11911@item @emph{Class}:
11912Function
11913
11914@item @emph{See also}:
11915@ref{RAND}, @gol
11916@ref{RANDOM_NUMBER}
11917@end table
11918
11919
11920
11921@node RAND
11922@section @code{RAND} --- Real pseudo-random number
11923@fnindex RAND
11924@cindex random number generation
11925
11926@table @asis
11927@item @emph{Description}:
11928@code{RAND(FLAG)} returns a pseudo-random number from a uniform
11929distribution between 0 and 1. If @var{FLAG} is 0, the next number
11930in the current sequence is returned; if @var{FLAG} is 1, the generator
11931is restarted by @code{CALL SRAND(0)}; if @var{FLAG} has any other value,
11932it is used as a new seed with @code{SRAND}.
11933
11934This intrinsic routine is provided for backwards compatibility with
11935GNU Fortran 77. It implements a simple modulo generator as provided
11936by @command{g77}. For new code, one should consider the use of
11937@ref{RANDOM_NUMBER} as it implements a superior algorithm.
11938
11939@item @emph{Standard}:
11940GNU extension
11941
11942@item @emph{Class}:
11943Function
11944
11945@item @emph{Syntax}:
11946@code{RESULT = RAND(I)}
11947
11948@item @emph{Arguments}:
11949@multitable @columnfractions .15 .70
11950@item @var{I} @tab Shall be a scalar @code{INTEGER} of kind 4.
11951@end multitable
11952
11953@item @emph{Return value}:
11954The return value is of @code{REAL} type and the default kind.
11955
11956@item @emph{Example}:
11957@smallexample
11958program test_rand
11959  integer,parameter :: seed = 86456
11960
11961  call srand(seed)
11962  print *, rand(), rand(), rand(), rand()
11963  print *, rand(seed), rand(), rand(), rand()
11964end program test_rand
11965@end smallexample
11966
11967@item @emph{See also}:
11968@ref{SRAND}, @gol
11969@ref{RANDOM_NUMBER}
11970
11971@end table
11972
11973
11974@node RANDOM_INIT
11975@section @code{RANDOM_INIT} --- Initialize a pseudo-random number generator
11976@fnindex RANDOM_INIT
11977@cindex random number generation, initialization
11978
11979@table @asis
11980@item @emph{Description}:
11981Initializes the state of the pseudorandom number generator used by
11982@code{RANDOM_NUMBER}.
11983
11984@item @emph{Standard}:
11985Fortran 2018
11986
11987@item @emph{Class}:
11988Subroutine
11989
11990@item @emph{Syntax}:
11991@code{CALL RANDOM_INIT(REPEATABLE, IMAGE_DISTINCT)}
11992
11993@item @emph{Arguments}:
11994@multitable @columnfractions .25 .70
11995@item @var{REPEATABLE} @tab Shall be a scalar with a @code{LOGICAL} type,
11996and it is @code{INTENT(IN)}.  If it is @code{.true.}, the seed is set to
11997a processor-dependent value that is the same each time @code{RANDOM_INIT}
11998is called from the same image.  The term ``same image'' means a single
11999instance of program execution.  The sequence of random numbers is different
12000for repeated execution of the program.  If it is @code{.false.}, the seed
12001is set to a processor-dependent value.
12002@item @var{IMAGE_DISTINCT} @tab Shall be a scalar with a
12003@code{LOGICAL} type, and it is @code{INTENT(IN)}.  If it is @code{.true.},
12004the seed is set to a processor-dependent value that is distinct from th
12005seed set by a call to @code{RANDOM_INIT} in another image.  If it is
12006@code{.false.}, the seed is set value that does depend which image called
12007@code{RANDOM_INIT}.
12008@end multitable
12009
12010@item @emph{Example}:
12011@smallexample
12012program test_random_seed
12013  implicit none
12014  real x(3), y(3)
12015  call random_init(.true., .true.)
12016  call random_number(x)
12017  call random_init(.true., .true.)
12018  call random_number(y)
12019  ! x and y are the same sequence
12020  if (any(x /= y)) call abort
12021end program test_random_seed
12022@end smallexample
12023
12024@item @emph{See also}:
12025@ref{RANDOM_NUMBER}, @gol
12026@ref{RANDOM_SEED}
12027@end table
12028
12029
12030@node RANDOM_NUMBER
12031@section @code{RANDOM_NUMBER} --- Pseudo-random number
12032@fnindex RANDOM_NUMBER
12033@cindex random number generation
12034
12035@table @asis
12036@item @emph{Description}:
12037Returns a single pseudorandom number or an array of pseudorandom numbers
12038from the uniform distribution over the range @math{ 0 \leq x < 1}.
12039
12040The runtime-library implements the xoshiro256** pseudorandom number
12041generator (PRNG). This generator has a period of @math{2^{256} - 1},
12042and when using multiple threads up to @math{2^{128}} threads can each
12043generate @math{2^{128}} random numbers before any aliasing occurs.
12044
12045Note that in a multi-threaded program (e.g. using OpenMP directives),
12046each thread will have its own random number state. For details of the
12047seeding procedure, see the documentation for the @code{RANDOM_SEED}
12048intrinsic.
12049
12050
12051@item @emph{Standard}:
12052Fortran 90 and later
12053
12054@item @emph{Class}:
12055Subroutine
12056
12057@item @emph{Syntax}:
12058@code{RANDOM_NUMBER(HARVEST)}
12059
12060@item @emph{Arguments}:
12061@multitable @columnfractions .15 .70
12062@item @var{HARVEST} @tab Shall be a scalar or an array of type @code{REAL}.
12063@end multitable
12064
12065@item @emph{Example}:
12066@smallexample
12067program test_random_number
12068  REAL :: r(5,5)
12069  CALL RANDOM_NUMBER(r)
12070end program
12071@end smallexample
12072
12073@item @emph{See also}:
12074@ref{RANDOM_SEED}, @gol
12075@ref{RANDOM_INIT}
12076@end table
12077
12078
12079
12080@node RANDOM_SEED
12081@section @code{RANDOM_SEED} --- Initialize a pseudo-random number sequence
12082@fnindex RANDOM_SEED
12083@cindex random number generation, seeding
12084@cindex seeding a random number generator
12085
12086@table @asis
12087@item @emph{Description}:
12088Restarts or queries the state of the pseudorandom number generator used by
12089@code{RANDOM_NUMBER}.
12090
12091If @code{RANDOM_SEED} is called without arguments, it is seeded with
12092random data retrieved from the operating system.
12093
12094As an extension to the Fortran standard, the GFortran
12095@code{RANDOM_NUMBER} supports multiple threads. Each thread in a
12096multi-threaded program has its own seed.  When @code{RANDOM_SEED} is
12097called either without arguments or with the @var{PUT} argument, the
12098given seed is copied into a master seed as well as the seed of the
12099current thread. When a new thread uses @code{RANDOM_NUMBER} for the
12100first time, the seed is copied from the master seed, and forwarded
12101@math{N * 2^{128}} steps to guarantee that the random stream does not
12102alias any other stream in the system, where @var{N} is the number of
12103threads that have used @code{RANDOM_NUMBER} so far during the program
12104execution.
12105
12106@item @emph{Standard}:
12107Fortran 90 and later
12108
12109@item @emph{Class}:
12110Subroutine
12111
12112@item @emph{Syntax}:
12113@code{CALL RANDOM_SEED([SIZE, PUT, GET])}
12114
12115@item @emph{Arguments}:
12116@multitable @columnfractions .15 .70
12117@item @var{SIZE} @tab (Optional) Shall be a scalar and of type default
12118@code{INTEGER}, with @code{INTENT(OUT)}. It specifies the minimum size
12119of the arrays used with the @var{PUT} and @var{GET} arguments.
12120@item @var{PUT}  @tab (Optional) Shall be an array of type default
12121@code{INTEGER} and rank one. It is @code{INTENT(IN)} and the size of
12122the array must be larger than or equal to the number returned by the
12123@var{SIZE} argument.
12124@item @var{GET}  @tab (Optional) Shall be an array of type default
12125@code{INTEGER} and rank one. It is @code{INTENT(OUT)} and the size
12126of the array must be larger than or equal to the number returned by
12127the @var{SIZE} argument.
12128@end multitable
12129
12130@item @emph{Example}:
12131@smallexample
12132program test_random_seed
12133  implicit none
12134  integer, allocatable :: seed(:)
12135  integer :: n
12136
12137  call random_seed(size = n)
12138  allocate(seed(n))
12139  call random_seed(get=seed)
12140  write (*, *) seed
12141end program test_random_seed
12142@end smallexample
12143
12144@item @emph{See also}:
12145@ref{RANDOM_NUMBER}, @gol
12146@ref{RANDOM_INIT}
12147@end table
12148
12149
12150
12151@node RANGE
12152@section @code{RANGE} --- Decimal exponent range
12153@fnindex RANGE
12154@cindex model representation, range
12155
12156@table @asis
12157@item @emph{Description}:
12158@code{RANGE(X)} returns the decimal exponent range in the model of the
12159type of @code{X}.
12160
12161@item @emph{Standard}:
12162Fortran 90 and later
12163
12164@item @emph{Class}:
12165Inquiry function
12166
12167@item @emph{Syntax}:
12168@code{RESULT = RANGE(X)}
12169
12170@item @emph{Arguments}:
12171@multitable @columnfractions .15 .70
12172@item @var{X} @tab Shall be of type @code{INTEGER}, @code{REAL}
12173or @code{COMPLEX}.
12174@end multitable
12175
12176@item @emph{Return value}:
12177The return value is of type @code{INTEGER} and of the default integer
12178kind.
12179
12180@item @emph{Example}:
12181See @code{PRECISION} for an example.
12182@item @emph{See also}:
12183@ref{SELECTED_REAL_KIND}, @gol
12184@ref{PRECISION}
12185@end table
12186
12187
12188
12189@node RANK
12190@section @code{RANK} --- Rank of a data object
12191@fnindex RANK
12192@cindex rank
12193
12194@table @asis
12195@item @emph{Description}:
12196@code{RANK(A)} returns the rank of a scalar or array data object.
12197
12198@item @emph{Standard}:
12199Technical Specification (TS) 29113
12200
12201@item @emph{Class}:
12202Inquiry function
12203
12204@item @emph{Syntax}:
12205@code{RESULT = RANK(A)}
12206
12207@item @emph{Arguments}:
12208@multitable @columnfractions .15 .70
12209@item @var{A} @tab can be of any type
12210@end multitable
12211
12212@item @emph{Return value}:
12213The return value is of type @code{INTEGER} and of the default integer
12214kind. For arrays, their rank is returned; for scalars zero is returned.
12215
12216@item @emph{Example}:
12217@smallexample
12218program test_rank
12219  integer :: a
12220  real, allocatable :: b(:,:)
12221
12222  print *, rank(a), rank(b) ! Prints:  0  2
12223end program test_rank
12224@end smallexample
12225
12226@end table
12227
12228
12229
12230@node REAL
12231@section @code{REAL} --- Convert to real type
12232@fnindex REAL
12233@fnindex REALPART
12234@fnindex FLOAT
12235@fnindex DFLOAT
12236@fnindex FLOATI
12237@fnindex FLOATJ
12238@fnindex FLOATK
12239@fnindex SNGL
12240@cindex conversion, to real
12241@cindex complex numbers, real part
12242
12243@table @asis
12244@item @emph{Description}:
12245@code{REAL(A [, KIND])} converts its argument @var{A} to a real type.  The
12246@code{REALPART} function is provided for compatibility with @command{g77},
12247and its use is strongly discouraged.
12248
12249@item @emph{Standard}:
12250Fortran 77 and later, with @var{KIND} argument Fortran 90 and later, has GNU extensions
12251
12252@item @emph{Class}:
12253Elemental function
12254
12255@item @emph{Syntax}:
12256@multitable @columnfractions .80
12257@item @code{RESULT = REAL(A [, KIND])}
12258@item @code{RESULT = REALPART(Z)}
12259@end multitable
12260
12261@item @emph{Arguments}:
12262@multitable @columnfractions .15 .70
12263@item @var{A}    @tab Shall be @code{INTEGER}, @code{REAL}, or
12264@code{COMPLEX}.
12265@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
12266expression indicating the kind parameter of the result.
12267@end multitable
12268
12269@item @emph{Return value}:
12270These functions return a @code{REAL} variable or array under
12271the following rules:
12272
12273@table @asis
12274@item (A)
12275@code{REAL(A)} is converted to a default real type if @var{A} is an
12276integer or real variable.
12277@item (B)
12278@code{REAL(A)} is converted to a real type with the kind type parameter
12279of @var{A} if @var{A} is a complex variable.
12280@item (C)
12281@code{REAL(A, KIND)} is converted to a real type with kind type
12282parameter @var{KIND} if @var{A} is a complex, integer, or real
12283variable.
12284@end table
12285
12286@item @emph{Example}:
12287@smallexample
12288program test_real
12289  complex :: x = (1.0, 2.0)
12290  print *, real(x), real(x,8), realpart(x)
12291end program test_real
12292@end smallexample
12293
12294@item @emph{Specific names}:
12295@multitable @columnfractions .20 .20 .20 .25
12296@item Name             @tab Argument           @tab Return type     @tab Standard
12297@item @code{FLOAT(A)}  @tab @code{INTEGER(4)}  @tab @code{REAL(4)}  @tab GNU extension
12298@item @code{DFLOAT(A)} @tab @code{INTEGER(4)}  @tab @code{REAL(8)}  @tab GNU extension
12299@item @code{FLOATI(A)} @tab @code{INTEGER(2)}  @tab @code{REAL(4)}  @tab GNU extension
12300@item @code{FLOATJ(A)} @tab @code{INTEGER(4)}  @tab @code{REAL(4)}  @tab GNU extension
12301@item @code{FLOATK(A)} @tab @code{INTEGER(8)}  @tab @code{REAL(4)}  @tab GNU extension
12302@item @code{SNGL(A)}   @tab @code{INTEGER(8)}  @tab @code{REAL(4)}  @tab GNU extension
12303@end multitable
12304
12305
12306@item @emph{See also}:
12307@ref{DBLE}
12308
12309@end table
12310
12311
12312
12313@node RENAME
12314@section @code{RENAME} --- Rename a file
12315@fnindex RENAME
12316@cindex file system, rename file
12317
12318@table @asis
12319@item @emph{Description}:
12320Renames a file from file @var{PATH1} to @var{PATH2}. A null
12321character (@code{CHAR(0)}) can be used to mark the end of the names in
12322@var{PATH1} and @var{PATH2}; otherwise, trailing blanks in the file
12323names are ignored.  If the @var{STATUS} argument is supplied, it
12324contains 0 on success or a nonzero error code upon return; see
12325@code{rename(2)}.
12326
12327This intrinsic is provided in both subroutine and function forms;
12328however, only one form can be used in any given program unit.
12329
12330@item @emph{Standard}:
12331GNU extension
12332
12333@item @emph{Class}:
12334Subroutine, function
12335
12336@item @emph{Syntax}:
12337@multitable @columnfractions .80
12338@item @code{CALL RENAME(PATH1, PATH2 [, STATUS])}
12339@item @code{STATUS = RENAME(PATH1, PATH2)}
12340@end multitable
12341
12342@item @emph{Arguments}:
12343@multitable @columnfractions .15 .70
12344@item @var{PATH1} @tab Shall be of default @code{CHARACTER} type.
12345@item @var{PATH2} @tab Shall be of default @code{CHARACTER} type.
12346@item @var{STATUS} @tab (Optional) Shall be of default @code{INTEGER} type.
12347@end multitable
12348
12349@item @emph{See also}:
12350@ref{LINK}
12351
12352@end table
12353
12354
12355
12356@node REPEAT
12357@section @code{REPEAT} --- Repeated string concatenation
12358@fnindex REPEAT
12359@cindex string, repeat
12360@cindex string, concatenate
12361
12362@table @asis
12363@item @emph{Description}:
12364Concatenates @var{NCOPIES} copies of a string.
12365
12366@item @emph{Standard}:
12367Fortran 90 and later
12368
12369@item @emph{Class}:
12370Transformational function
12371
12372@item @emph{Syntax}:
12373@code{RESULT = REPEAT(STRING, NCOPIES)}
12374
12375@item @emph{Arguments}:
12376@multitable @columnfractions .15 .70
12377@item @var{STRING}  @tab Shall be scalar and of type @code{CHARACTER}.
12378@item @var{NCOPIES} @tab Shall be scalar and of type @code{INTEGER}.
12379@end multitable
12380
12381@item @emph{Return value}:
12382A new scalar of type @code{CHARACTER} built up from @var{NCOPIES} copies
12383of @var{STRING}.
12384
12385@item @emph{Example}:
12386@smallexample
12387program test_repeat
12388  write(*,*) repeat("x", 5)   ! "xxxxx"
12389end program
12390@end smallexample
12391@end table
12392
12393
12394
12395@node RESHAPE
12396@section @code{RESHAPE} --- Function to reshape an array
12397@fnindex RESHAPE
12398@cindex array, change dimensions
12399@cindex array, transmogrify
12400
12401@table @asis
12402@item @emph{Description}:
12403Reshapes @var{SOURCE} to correspond to @var{SHAPE}. If necessary,
12404the new array may be padded with elements from @var{PAD} or permuted
12405as defined by @var{ORDER}.
12406
12407@item @emph{Standard}:
12408Fortran 90 and later
12409
12410@item @emph{Class}:
12411Transformational function
12412
12413@item @emph{Syntax}:
12414@code{RESULT = RESHAPE(SOURCE, SHAPE[, PAD, ORDER])}
12415
12416@item @emph{Arguments}:
12417@multitable @columnfractions .15 .70
12418@item @var{SOURCE} @tab Shall be an array of any type.
12419@item @var{SHAPE}  @tab Shall be of type @code{INTEGER} and an
12420array of rank one. Its values must be positive or zero.
12421@item @var{PAD}    @tab (Optional) shall be an array of the same
12422type as @var{SOURCE}.
12423@item @var{ORDER}  @tab (Optional) shall be of type @code{INTEGER}
12424and an array of the same shape as @var{SHAPE}. Its values shall
12425be a permutation of the numbers from 1 to n, where n is the size of
12426@var{SHAPE}. If @var{ORDER} is absent, the natural ordering shall
12427be assumed.
12428@end multitable
12429
12430@item @emph{Return value}:
12431The result is an array of shape @var{SHAPE} with the same type as
12432@var{SOURCE}.
12433
12434@item @emph{Example}:
12435@smallexample
12436PROGRAM test_reshape
12437  INTEGER, DIMENSION(4) :: x
12438  WRITE(*,*) SHAPE(x)                       ! prints "4"
12439  WRITE(*,*) SHAPE(RESHAPE(x, (/2, 2/)))    ! prints "2 2"
12440END PROGRAM
12441@end smallexample
12442
12443@item @emph{See also}:
12444@ref{SHAPE}
12445@end table
12446
12447
12448
12449@node RRSPACING
12450@section @code{RRSPACING} --- Reciprocal of the relative spacing
12451@fnindex RRSPACING
12452@cindex real number, relative spacing
12453@cindex floating point, relative spacing
12454
12455
12456@table @asis
12457@item @emph{Description}:
12458@code{RRSPACING(X)} returns the  reciprocal of the relative spacing of
12459model numbers near @var{X}.
12460
12461@item @emph{Standard}:
12462Fortran 90 and later
12463
12464@item @emph{Class}:
12465Elemental function
12466
12467@item @emph{Syntax}:
12468@code{RESULT = RRSPACING(X)}
12469
12470@item @emph{Arguments}:
12471@multitable @columnfractions .15 .70
12472@item @var{X} @tab Shall be of type @code{REAL}.
12473@end multitable
12474
12475@item @emph{Return value}:
12476The return value is of the same type and kind as @var{X}.
12477The value returned is equal to
12478@code{ABS(FRACTION(X)) * FLOAT(RADIX(X))**DIGITS(X)}.
12479
12480@item @emph{See also}:
12481@ref{SPACING}
12482@end table
12483
12484
12485
12486@node RSHIFT
12487@section @code{RSHIFT} --- Right shift bits
12488@fnindex RSHIFT
12489@cindex bits, shift right
12490
12491@table @asis
12492@item @emph{Description}:
12493@code{RSHIFT} returns a value corresponding to @var{I} with all of the
12494bits shifted right by @var{SHIFT} places.  @var{SHIFT} shall be
12495nonnegative and less than or equal to @code{BIT_SIZE(I)}, otherwise
12496the result value is undefined.  Bits shifted out from the right end
12497are lost. The fill is arithmetic: the bits shifted in from the left
12498end are equal to the leftmost bit, which in two's complement
12499representation is the sign bit.
12500
12501This function has been superseded by the @code{SHIFTA} intrinsic, which
12502is standard in Fortran 2008 and later.
12503
12504@item @emph{Standard}:
12505GNU extension
12506
12507@item @emph{Class}:
12508Elemental function
12509
12510@item @emph{Syntax}:
12511@code{RESULT = RSHIFT(I, SHIFT)}
12512
12513@item @emph{Arguments}:
12514@multitable @columnfractions .15 .70
12515@item @var{I} @tab The type shall be @code{INTEGER}.
12516@item @var{SHIFT} @tab The type shall be @code{INTEGER}.
12517@end multitable
12518
12519@item @emph{Return value}:
12520The return value is of type @code{INTEGER} and of the same kind as
12521@var{I}.
12522
12523@item @emph{See also}:
12524@ref{ISHFT}, @gol
12525@ref{ISHFTC}, @gol
12526@ref{LSHIFT}, @gol
12527@ref{SHIFTA}, @gol
12528@ref{SHIFTR}, @gol
12529@ref{SHIFTL}
12530
12531@end table
12532
12533
12534
12535@node SAME_TYPE_AS
12536@section @code{SAME_TYPE_AS} ---  Query dynamic types for equality
12537@fnindex SAME_TYPE_AS
12538
12539@table @asis
12540@item @emph{Description}:
12541Query dynamic types for equality.
12542
12543@item @emph{Standard}:
12544Fortran 2003 and later
12545
12546@item @emph{Class}:
12547Inquiry function
12548
12549@item @emph{Syntax}:
12550@code{RESULT = SAME_TYPE_AS(A, B)}
12551
12552@item @emph{Arguments}:
12553@multitable @columnfractions .15 .70
12554@item @var{A} @tab Shall be an object of extensible declared type or
12555unlimited polymorphic.
12556@item @var{B} @tab Shall be an object of extensible declared type or
12557unlimited polymorphic.
12558@end multitable
12559
12560@item @emph{Return value}:
12561The return value is a scalar of type default logical. It is true if and
12562only if the dynamic type of A is the same as the dynamic type of B.
12563
12564@item @emph{See also}:
12565@ref{EXTENDS_TYPE_OF}
12566
12567@end table
12568
12569
12570
12571@node SCALE
12572@section @code{SCALE} --- Scale a real value
12573@fnindex SCALE
12574@cindex real number, scale
12575@cindex floating point, scale
12576
12577@table @asis
12578@item @emph{Description}:
12579@code{SCALE(X,I)} returns @code{X * RADIX(X)**I}.
12580
12581@item @emph{Standard}:
12582Fortran 90 and later
12583
12584@item @emph{Class}:
12585Elemental function
12586
12587@item @emph{Syntax}:
12588@code{RESULT = SCALE(X, I)}
12589
12590@item @emph{Arguments}:
12591@multitable @columnfractions .15 .70
12592@item @var{X} @tab The type of the argument shall be a @code{REAL}.
12593@item @var{I} @tab The type of the argument shall be a @code{INTEGER}.
12594@end multitable
12595
12596@item @emph{Return value}:
12597The return value is of the same type and kind as @var{X}.
12598Its value is @code{X * RADIX(X)**I}.
12599
12600@item @emph{Example}:
12601@smallexample
12602program test_scale
12603  real :: x = 178.1387e-4
12604  integer :: i = 5
12605  print *, scale(x,i), x*radix(x)**i
12606end program test_scale
12607@end smallexample
12608
12609@end table
12610
12611
12612
12613@node SCAN
12614@section @code{SCAN} --- Scan a string for the presence of a set of characters
12615@fnindex SCAN
12616@cindex string, find subset
12617
12618@table @asis
12619@item @emph{Description}:
12620Scans a @var{STRING} for any of the characters in a @var{SET}
12621of characters.
12622
12623If @var{BACK} is either absent or equals @code{FALSE}, this function
12624returns the position of the leftmost character of @var{STRING} that is
12625in @var{SET}. If @var{BACK} equals @code{TRUE}, the rightmost position
12626is returned. If no character of @var{SET} is found in @var{STRING}, the
12627result is zero.
12628
12629@item @emph{Standard}:
12630Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
12631
12632@item @emph{Class}:
12633Elemental function
12634
12635@item @emph{Syntax}:
12636@code{RESULT = SCAN(STRING, SET[, BACK [, KIND]])}
12637
12638@item @emph{Arguments}:
12639@multitable @columnfractions .15 .70
12640@item @var{STRING} @tab Shall be of type @code{CHARACTER}.
12641@item @var{SET}    @tab Shall be of type @code{CHARACTER}.
12642@item @var{BACK}   @tab (Optional) shall be of type @code{LOGICAL}.
12643@item @var{KIND}   @tab (Optional) An @code{INTEGER} initialization
12644expression indicating the kind parameter of the result.
12645@end multitable
12646
12647@item @emph{Return value}:
12648The return value is of type @code{INTEGER} and of kind @var{KIND}. If
12649@var{KIND} is absent, the return value is of default integer kind.
12650
12651@item @emph{Example}:
12652@smallexample
12653PROGRAM test_scan
12654  WRITE(*,*) SCAN("FORTRAN", "AO")          ! 2, found 'O'
12655  WRITE(*,*) SCAN("FORTRAN", "AO", .TRUE.)  ! 6, found 'A'
12656  WRITE(*,*) SCAN("FORTRAN", "C++")         ! 0, found none
12657END PROGRAM
12658@end smallexample
12659
12660@item @emph{See also}:
12661@ref{INDEX intrinsic}, @gol
12662@ref{VERIFY}
12663@end table
12664
12665
12666
12667@node SECNDS
12668@section @code{SECNDS} --- Time function
12669@fnindex SECNDS
12670@cindex time, elapsed
12671@cindex elapsed time
12672
12673@table @asis
12674@item @emph{Description}:
12675@code{SECNDS(X)} gets the time in seconds from the real-time system clock.
12676@var{X} is a reference time, also in seconds. If this is zero, the time in
12677seconds from midnight is returned. This function is non-standard and its
12678use is discouraged.
12679
12680@item @emph{Standard}:
12681GNU extension
12682
12683@item @emph{Class}:
12684Function
12685
12686@item @emph{Syntax}:
12687@code{RESULT = SECNDS (X)}
12688
12689@item @emph{Arguments}:
12690@multitable @columnfractions .15 .70
12691@item @var{T}     @tab Shall be of type @code{REAL(4)}.
12692@item @var{X}     @tab Shall be of type @code{REAL(4)}.
12693@end multitable
12694
12695@item @emph{Return value}:
12696None
12697
12698@item @emph{Example}:
12699@smallexample
12700program test_secnds
12701    integer :: i
12702    real(4) :: t1, t2
12703    print *, secnds (0.0)   ! seconds since midnight
12704    t1 = secnds (0.0)       ! reference time
12705    do i = 1, 10000000      ! do something
12706    end do
12707    t2 = secnds (t1)        ! elapsed time
12708    print *, "Something took ", t2, " seconds."
12709end program test_secnds
12710@end smallexample
12711@end table
12712
12713
12714
12715@node SECOND
12716@section @code{SECOND} --- CPU time function
12717@fnindex SECOND
12718@cindex time, elapsed
12719@cindex elapsed time
12720
12721@table @asis
12722@item @emph{Description}:
12723Returns a @code{REAL(4)} value representing the elapsed CPU time in
12724seconds.  This provides the same functionality as the standard
12725@code{CPU_TIME} intrinsic, and is only included for backwards
12726compatibility.
12727
12728This intrinsic is provided in both subroutine and function forms;
12729however, only one form can be used in any given program unit.
12730
12731@item @emph{Standard}:
12732GNU extension
12733
12734@item @emph{Class}:
12735Subroutine, function
12736
12737@item @emph{Syntax}:
12738@multitable @columnfractions .80
12739@item @code{CALL SECOND(TIME)}
12740@item @code{TIME = SECOND()}
12741@end multitable
12742
12743@item @emph{Arguments}:
12744@multitable @columnfractions .15 .70
12745@item @var{TIME}  @tab Shall be of type @code{REAL(4)}.
12746@end multitable
12747
12748@item @emph{Return value}:
12749In either syntax, @var{TIME} is set to the process's current runtime in
12750seconds.
12751
12752@item @emph{See also}:
12753@ref{CPU_TIME}
12754
12755@end table
12756
12757
12758
12759@node SELECTED_CHAR_KIND
12760@section @code{SELECTED_CHAR_KIND} --- Choose character kind
12761@fnindex SELECTED_CHAR_KIND
12762@cindex character kind
12763@cindex kind, character
12764
12765@table @asis
12766@item @emph{Description}:
12767
12768@code{SELECTED_CHAR_KIND(NAME)} returns the kind value for the character
12769set named @var{NAME}, if a character set with such a name is supported,
12770or @math{-1} otherwise. Currently, supported character sets include
12771``ASCII'' and ``DEFAULT'', which are equivalent, and ``ISO_10646''
12772(Universal Character Set, UCS-4) which is commonly known as Unicode.
12773
12774@item @emph{Standard}:
12775Fortran 2003 and later
12776
12777@item @emph{Class}:
12778Transformational function
12779
12780@item @emph{Syntax}:
12781@code{RESULT = SELECTED_CHAR_KIND(NAME)}
12782
12783@item @emph{Arguments}:
12784@multitable @columnfractions .15 .70
12785@item @var{NAME} @tab Shall be a scalar and of the default character type.
12786@end multitable
12787
12788@item @emph{Example}:
12789@smallexample
12790program character_kind
12791  use iso_fortran_env
12792  implicit none
12793  integer, parameter :: ascii = selected_char_kind ("ascii")
12794  integer, parameter :: ucs4  = selected_char_kind ('ISO_10646')
12795
12796  character(kind=ascii, len=26) :: alphabet
12797  character(kind=ucs4,  len=30) :: hello_world
12798
12799  alphabet = ascii_"abcdefghijklmnopqrstuvwxyz"
12800  hello_world = ucs4_'Hello World and Ni Hao -- ' &
12801                // char (int (z'4F60'), ucs4)     &
12802                // char (int (z'597D'), ucs4)
12803
12804  write (*,*) alphabet
12805
12806  open (output_unit, encoding='UTF-8')
12807  write (*,*) trim (hello_world)
12808end program character_kind
12809@end smallexample
12810@end table
12811
12812
12813
12814@node SELECTED_INT_KIND
12815@section @code{SELECTED_INT_KIND} --- Choose integer kind
12816@fnindex SELECTED_INT_KIND
12817@cindex integer kind
12818@cindex kind, integer
12819
12820@table @asis
12821@item @emph{Description}:
12822@code{SELECTED_INT_KIND(R)} return the kind value of the smallest integer
12823type that can represent all values ranging from @math{-10^R} (exclusive)
12824to @math{10^R} (exclusive). If there is no integer kind that accommodates
12825this range, @code{SELECTED_INT_KIND} returns @math{-1}.
12826
12827@item @emph{Standard}:
12828Fortran 90 and later
12829
12830@item @emph{Class}:
12831Transformational function
12832
12833@item @emph{Syntax}:
12834@code{RESULT = SELECTED_INT_KIND(R)}
12835
12836@item @emph{Arguments}:
12837@multitable @columnfractions .15 .70
12838@item @var{R} @tab Shall be a scalar and of type @code{INTEGER}.
12839@end multitable
12840
12841@item @emph{Example}:
12842@smallexample
12843program large_integers
12844  integer,parameter :: k5 = selected_int_kind(5)
12845  integer,parameter :: k15 = selected_int_kind(15)
12846  integer(kind=k5) :: i5
12847  integer(kind=k15) :: i15
12848
12849  print *, huge(i5), huge(i15)
12850
12851  ! The following inequalities are always true
12852  print *, huge(i5) >= 10_k5**5-1
12853  print *, huge(i15) >= 10_k15**15-1
12854end program large_integers
12855@end smallexample
12856@end table
12857
12858
12859
12860@node SELECTED_REAL_KIND
12861@section @code{SELECTED_REAL_KIND} --- Choose real kind
12862@fnindex SELECTED_REAL_KIND
12863@cindex real kind
12864@cindex kind, real
12865@cindex radix, real
12866
12867@table @asis
12868@item @emph{Description}:
12869@code{SELECTED_REAL_KIND(P,R)} returns the kind value of a real data type
12870with decimal precision of at least @code{P} digits, exponent range of
12871at least @code{R}, and with a radix of @code{RADIX}.
12872
12873@item @emph{Standard}:
12874Fortran 90 and later, with @code{RADIX} Fortran 2008 or later
12875
12876@item @emph{Class}:
12877Transformational function
12878
12879@item @emph{Syntax}:
12880@code{RESULT = SELECTED_REAL_KIND([P, R, RADIX])}
12881
12882@item @emph{Arguments}:
12883@multitable @columnfractions .15 .70
12884@item @var{P} @tab (Optional) shall be a scalar and of type @code{INTEGER}.
12885@item @var{R} @tab (Optional) shall be a scalar and of type @code{INTEGER}.
12886@item @var{RADIX} @tab (Optional) shall be a scalar and of type @code{INTEGER}.
12887@end multitable
12888Before Fortran 2008, at least one of the arguments @var{R} or @var{P} shall
12889be present; since Fortran 2008, they are assumed to be zero if absent.
12890
12891@item @emph{Return value}:
12892
12893@code{SELECTED_REAL_KIND} returns the value of the kind type parameter of
12894a real data type with decimal precision of at least @code{P} digits, a
12895decimal exponent range of at least @code{R}, and with the requested
12896@code{RADIX}. If the @code{RADIX} parameter is absent, real kinds with
12897any radix can be returned. If more than one real data type meet the
12898criteria, the kind of the data type with the smallest decimal precision
12899is returned. If no real data type matches the criteria, the result is
12900@table @asis
12901@item -1 if the processor does not support a real data type with a
12902precision greater than or equal to @code{P}, but the @code{R} and
12903@code{RADIX} requirements can be fulfilled
12904@item -2 if the processor does not support a real type with an exponent
12905range greater than or equal to @code{R}, but @code{P} and @code{RADIX}
12906are fulfillable
12907@item -3 if @code{RADIX} but not @code{P} and @code{R} requirements
12908are fulfillable
12909@item -4 if @code{RADIX} and either @code{P} or @code{R} requirements
12910are fulfillable
12911@item -5 if there is no real type with the given @code{RADIX}
12912@end table
12913
12914@item @emph{Example}:
12915@smallexample
12916program real_kinds
12917  integer,parameter :: p6 = selected_real_kind(6)
12918  integer,parameter :: p10r100 = selected_real_kind(10,100)
12919  integer,parameter :: r400 = selected_real_kind(r=400)
12920  real(kind=p6) :: x
12921  real(kind=p10r100) :: y
12922  real(kind=r400) :: z
12923
12924  print *, precision(x), range(x)
12925  print *, precision(y), range(y)
12926  print *, precision(z), range(z)
12927end program real_kinds
12928@end smallexample
12929@item @emph{See also}:
12930@ref{PRECISION}, @gol
12931@ref{RANGE}, @gol
12932@ref{RADIX}
12933@end table
12934
12935
12936
12937@node SET_EXPONENT
12938@section @code{SET_EXPONENT} --- Set the exponent of the model
12939@fnindex SET_EXPONENT
12940@cindex real number, set exponent
12941@cindex floating point, set exponent
12942
12943@table @asis
12944@item @emph{Description}:
12945@code{SET_EXPONENT(X, I)} returns the real number whose fractional part
12946is that that of @var{X} and whose exponent part is @var{I}.
12947
12948@item @emph{Standard}:
12949Fortran 90 and later
12950
12951@item @emph{Class}:
12952Elemental function
12953
12954@item @emph{Syntax}:
12955@code{RESULT = SET_EXPONENT(X, I)}
12956
12957@item @emph{Arguments}:
12958@multitable @columnfractions .15 .70
12959@item @var{X} @tab Shall be of type @code{REAL}.
12960@item @var{I} @tab Shall be of type @code{INTEGER}.
12961@end multitable
12962
12963@item @emph{Return value}:
12964The return value is of the same type and kind as @var{X}.
12965The real number whose fractional part
12966is that that of @var{X} and whose exponent part if @var{I} is returned;
12967it is @code{FRACTION(X) * RADIX(X)**I}.
12968
12969@item @emph{Example}:
12970@smallexample
12971PROGRAM test_setexp
12972  REAL :: x = 178.1387e-4
12973  INTEGER :: i = 17
12974  PRINT *, SET_EXPONENT(x, i), FRACTION(x) * RADIX(x)**i
12975END PROGRAM
12976@end smallexample
12977
12978@end table
12979
12980
12981
12982@node SHAPE
12983@section @code{SHAPE} --- Determine the shape of an array
12984@fnindex SHAPE
12985@cindex array, shape
12986
12987@table @asis
12988@item @emph{Description}:
12989Determines the shape of an array.
12990
12991@item @emph{Standard}:
12992Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
12993
12994@item @emph{Class}:
12995Inquiry function
12996
12997@item @emph{Syntax}:
12998@code{RESULT = SHAPE(SOURCE [, KIND])}
12999
13000@item @emph{Arguments}:
13001@multitable @columnfractions .15 .70
13002@item @var{SOURCE} @tab Shall be an array or scalar of any type.
13003If @var{SOURCE} is a pointer it must be associated and allocatable
13004arrays must be allocated.
13005@item @var{KIND}   @tab (Optional) An @code{INTEGER} initialization
13006expression indicating the kind parameter of the result.
13007@end multitable
13008
13009@item @emph{Return value}:
13010An @code{INTEGER} array of rank one with as many elements as @var{SOURCE}
13011has dimensions. The elements of the resulting array correspond to the extend
13012of @var{SOURCE} along the respective dimensions. If @var{SOURCE} is a scalar,
13013the result is the rank one array of size zero. If @var{KIND} is absent, the
13014return value has the default integer kind otherwise the specified kind.
13015
13016@item @emph{Example}:
13017@smallexample
13018PROGRAM test_shape
13019  INTEGER, DIMENSION(-1:1, -1:2) :: A
13020  WRITE(*,*) SHAPE(A)             ! (/ 3, 4 /)
13021  WRITE(*,*) SIZE(SHAPE(42))      ! (/ /)
13022END PROGRAM
13023@end smallexample
13024
13025@item @emph{See also}:
13026@ref{RESHAPE}, @gol
13027@ref{SIZE}
13028@end table
13029
13030
13031
13032@node SHIFTA
13033@section @code{SHIFTA} --- Right shift with fill
13034@fnindex SHIFTA
13035@cindex bits, shift right
13036@cindex shift, right with fill
13037
13038@table @asis
13039@item @emph{Description}:
13040@code{SHIFTA} returns a value corresponding to @var{I} with all of the
13041bits shifted right by @var{SHIFT} places.  @var{SHIFT} that be
13042nonnegative and less than or equal to @code{BIT_SIZE(I)}, otherwise
13043the result value is undefined.  Bits shifted out from the right end
13044are lost. The fill is arithmetic: the bits shifted in from the left
13045end are equal to the leftmost bit, which in two's complement
13046representation is the sign bit.
13047
13048@item @emph{Standard}:
13049Fortran 2008 and later
13050
13051@item @emph{Class}:
13052Elemental function
13053
13054@item @emph{Syntax}:
13055@code{RESULT = SHIFTA(I, SHIFT)}
13056
13057@item @emph{Arguments}:
13058@multitable @columnfractions .15 .70
13059@item @var{I} @tab The type shall be @code{INTEGER}.
13060@item @var{SHIFT} @tab The type shall be @code{INTEGER}.
13061@end multitable
13062
13063@item @emph{Return value}:
13064The return value is of type @code{INTEGER} and of the same kind as
13065@var{I}.
13066
13067@item @emph{See also}:
13068@ref{SHIFTL}, @gol
13069@ref{SHIFTR}
13070@end table
13071
13072
13073
13074@node SHIFTL
13075@section @code{SHIFTL} --- Left shift
13076@fnindex SHIFTL
13077@cindex bits, shift left
13078@cindex shift, left
13079
13080@table @asis
13081@item @emph{Description}:
13082@code{SHIFTL} returns a value corresponding to @var{I} with all of the
13083bits shifted left by @var{SHIFT} places.  @var{SHIFT} shall be
13084nonnegative and less than or equal to @code{BIT_SIZE(I)}, otherwise
13085the result value is undefined.  Bits shifted out from the left end are
13086lost, and bits shifted in from the right end are set to 0.
13087
13088@item @emph{Standard}:
13089Fortran 2008 and later
13090
13091@item @emph{Class}:
13092Elemental function
13093
13094@item @emph{Syntax}:
13095@code{RESULT = SHIFTL(I, SHIFT)}
13096
13097@item @emph{Arguments}:
13098@multitable @columnfractions .15 .70
13099@item @var{I} @tab The type shall be @code{INTEGER}.
13100@item @var{SHIFT} @tab The type shall be @code{INTEGER}.
13101@end multitable
13102
13103@item @emph{Return value}:
13104The return value is of type @code{INTEGER} and of the same kind as
13105@var{I}.
13106
13107@item @emph{See also}:
13108@ref{SHIFTA}, @gol
13109@ref{SHIFTR}
13110@end table
13111
13112
13113
13114@node SHIFTR
13115@section @code{SHIFTR} --- Right shift
13116@fnindex SHIFTR
13117@cindex bits, shift right
13118@cindex shift, right
13119
13120@table @asis
13121@item @emph{Description}:
13122@code{SHIFTR} returns a value corresponding to @var{I} with all of the
13123bits shifted right by @var{SHIFT} places.  @var{SHIFT} shall be
13124nonnegative and less than or equal to @code{BIT_SIZE(I)}, otherwise
13125the result value is undefined.  Bits shifted out from the right end
13126are lost, and bits shifted in from the left end are set to 0.
13127
13128@item @emph{Standard}:
13129Fortran 2008 and later
13130
13131@item @emph{Class}:
13132Elemental function
13133
13134@item @emph{Syntax}:
13135@code{RESULT = SHIFTR(I, SHIFT)}
13136
13137@item @emph{Arguments}:
13138@multitable @columnfractions .15 .70
13139@item @var{I} @tab The type shall be @code{INTEGER}.
13140@item @var{SHIFT} @tab The type shall be @code{INTEGER}.
13141@end multitable
13142
13143@item @emph{Return value}:
13144The return value is of type @code{INTEGER} and of the same kind as
13145@var{I}.
13146
13147@item @emph{See also}:
13148@ref{SHIFTA}, @gol
13149@ref{SHIFTL}
13150@end table
13151
13152
13153
13154@node SIGN
13155@section @code{SIGN} --- Sign copying function
13156@fnindex SIGN
13157@fnindex ISIGN
13158@fnindex DSIGN
13159@cindex sign copying
13160
13161@table @asis
13162@item @emph{Description}:
13163@code{SIGN(A,B)} returns the value of @var{A} with the sign of @var{B}.
13164
13165@item @emph{Standard}:
13166Fortran 77 and later
13167
13168@item @emph{Class}:
13169Elemental function
13170
13171@item @emph{Syntax}:
13172@code{RESULT = SIGN(A, B)}
13173
13174@item @emph{Arguments}:
13175@multitable @columnfractions .15 .70
13176@item @var{A} @tab Shall be of type @code{INTEGER} or @code{REAL}
13177@item @var{B} @tab Shall be of the same type and kind as @var{A}.
13178@end multitable
13179
13180@item @emph{Return value}:
13181The kind of the return value is that of @var{A} and @var{B}.
13182If @math{B\ge 0} then the result is @code{ABS(A)}, else
13183it is @code{-ABS(A)}.
13184
13185@item @emph{Example}:
13186@smallexample
13187program test_sign
13188  print *, sign(-12,1)
13189  print *, sign(-12,0)
13190  print *, sign(-12,-1)
13191
13192  print *, sign(-12.,1.)
13193  print *, sign(-12.,0.)
13194  print *, sign(-12.,-1.)
13195end program test_sign
13196@end smallexample
13197
13198@item @emph{Specific names}:
13199@multitable @columnfractions .20 .20 .20 .25
13200@item Name              @tab Arguments              @tab Return type       @tab Standard
13201@item @code{SIGN(A,B)}  @tab @code{REAL(4) A, B}    @tab @code{REAL(4)}    @tab Fortran 77 and later
13202@item @code{ISIGN(A,B)} @tab @code{INTEGER(4) A, B} @tab @code{INTEGER(4)} @tab Fortran 77 and later
13203@item @code{DSIGN(A,B)} @tab @code{REAL(8) A, B}    @tab @code{REAL(8)}    @tab Fortran 77 and later
13204@end multitable
13205@end table
13206
13207
13208
13209@node SIGNAL
13210@section @code{SIGNAL} --- Signal handling subroutine (or function)
13211@fnindex SIGNAL
13212@cindex system, signal handling
13213
13214@table @asis
13215@item @emph{Description}:
13216@code{SIGNAL(NUMBER, HANDLER [, STATUS])} causes external subroutine
13217@var{HANDLER} to be executed with a single integer argument when signal
13218@var{NUMBER} occurs.  If @var{HANDLER} is an integer, it can be used to
13219turn off handling of signal @var{NUMBER} or revert to its default
13220action.  See @code{signal(2)}.
13221
13222If @code{SIGNAL} is called as a subroutine and the @var{STATUS} argument
13223is supplied, it is set to the value returned by @code{signal(2)}.
13224
13225@item @emph{Standard}:
13226GNU extension
13227
13228@item @emph{Class}:
13229Subroutine, function
13230
13231@item @emph{Syntax}:
13232@multitable @columnfractions .80
13233@item @code{CALL SIGNAL(NUMBER, HANDLER [, STATUS])}
13234@item @code{STATUS = SIGNAL(NUMBER, HANDLER)}
13235@end multitable
13236
13237@item @emph{Arguments}:
13238@multitable @columnfractions .15 .70
13239@item @var{NUMBER} @tab Shall be a scalar integer, with @code{INTENT(IN)}
13240@item @var{HANDLER}@tab Signal handler (@code{INTEGER FUNCTION} or
13241@code{SUBROUTINE}) or dummy/global @code{INTEGER} scalar.
13242@code{INTEGER}. It is @code{INTENT(IN)}.
13243@item @var{STATUS} @tab (Optional) @var{STATUS} shall be a scalar
13244integer. It has @code{INTENT(OUT)}.
13245@end multitable
13246@c TODO: What should the interface of the handler be?  Does it take arguments?
13247
13248@item @emph{Return value}:
13249The @code{SIGNAL} function returns the value returned by @code{signal(2)}.
13250
13251@item @emph{Example}:
13252@smallexample
13253program test_signal
13254  intrinsic signal
13255  external handler_print
13256
13257  call signal (12, handler_print)
13258  call signal (10, 1)
13259
13260  call sleep (30)
13261end program test_signal
13262@end smallexample
13263@end table
13264
13265
13266
13267@node SIN
13268@section @code{SIN} --- Sine function
13269@fnindex SIN
13270@fnindex DSIN
13271@fnindex CSIN
13272@fnindex ZSIN
13273@fnindex CDSIN
13274@cindex trigonometric function, sine
13275@cindex sine
13276
13277@table @asis
13278@item @emph{Description}:
13279@code{SIN(X)} computes the sine of @var{X}.
13280
13281@item @emph{Standard}:
13282Fortran 77 and later
13283
13284@item @emph{Class}:
13285Elemental function
13286
13287@item @emph{Syntax}:
13288@code{RESULT = SIN(X)}
13289
13290@item @emph{Arguments}:
13291@multitable @columnfractions .15 .70
13292@item @var{X} @tab The type shall be @code{REAL} or
13293@code{COMPLEX}.
13294@end multitable
13295
13296@item @emph{Return value}:
13297The return value has same type and kind as @var{X}.
13298
13299@item @emph{Example}:
13300@smallexample
13301program test_sin
13302  real :: x = 0.0
13303  x = sin(x)
13304end program test_sin
13305@end smallexample
13306
13307@item @emph{Specific names}:
13308@multitable @columnfractions .20 .20 .20 .25
13309@item Name            @tab Argument             @tab Return type       @tab Standard
13310@item @code{SIN(X)}   @tab @code{REAL(4) X}     @tab @code{REAL(4)}    @tab Fortran 77 and later
13311@item @code{DSIN(X)}  @tab @code{REAL(8) X}     @tab @code{REAL(8)}    @tab Fortran 77 and later
13312@item @code{CSIN(X)}  @tab @code{COMPLEX(4) X}  @tab @code{COMPLEX(4)} @tab Fortran 77 and later
13313@item @code{ZSIN(X)}  @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)} @tab GNU extension
13314@item @code{CDSIN(X)} @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)} @tab GNU extension
13315@end multitable
13316
13317@item @emph{See also}:
13318Inverse function: @gol
13319@ref{ASIN} @gol
13320Degrees function: @gol
13321@ref{SIND}
13322@end table
13323
13324
13325
13326@node SIND
13327@section @code{SIND} --- Sine function, degrees
13328@fnindex SIND
13329@fnindex DSIND
13330@fnindex CSIND
13331@fnindex ZSIND
13332@fnindex CDSIND
13333@cindex trigonometric function, sine, degrees
13334@cindex sine, degrees
13335
13336@table @asis
13337@item @emph{Description}:
13338@code{SIND(X)} computes the sine of @var{X} in degrees.
13339
13340This function is for compatibility only and should be avoided in favor of
13341standard constructs wherever possible.
13342
13343@item @emph{Standard}:
13344GNU extension, enabled with @option{-fdec-math}.
13345
13346@item @emph{Class}:
13347Elemental function
13348
13349@item @emph{Syntax}:
13350@code{RESULT = SIND(X)}
13351
13352@item @emph{Arguments}:
13353@multitable @columnfractions .15 .70
13354@item @var{X} @tab The type shall be @code{REAL} or
13355@code{COMPLEX}.
13356@end multitable
13357
13358@item @emph{Return value}:
13359The return value has same type and kind as @var{X}, and its value is in degrees.
13360
13361@item @emph{Example}:
13362@smallexample
13363program test_sind
13364  real :: x = 0.0
13365  x = sind(x)
13366end program test_sind
13367@end smallexample
13368
13369@item @emph{Specific names}:
13370@multitable @columnfractions .20 .20 .20 .25
13371@item Name            @tab Argument             @tab Return type       @tab Standard
13372@item @code{SIND(X)}   @tab @code{REAL(4) X}     @tab @code{REAL(4)}    @tab GNU extension
13373@item @code{DSIND(X)}  @tab @code{REAL(8) X}     @tab @code{REAL(8)}    @tab GNU extension
13374@item @code{CSIND(X)}  @tab @code{COMPLEX(4) X}  @tab @code{COMPLEX(4)} @tab GNU extension
13375@item @code{ZSIND(X)}  @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)} @tab GNU extension
13376@item @code{CDSIND(X)} @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)} @tab GNU extension
13377@end multitable
13378
13379@item @emph{See also}:
13380Inverse function: @gol
13381@ref{ASIND} @gol
13382Radians function: @gol
13383@ref{SIN} @gol
13384@end table
13385
13386
13387
13388@node SINH
13389@section @code{SINH} --- Hyperbolic sine function
13390@fnindex SINH
13391@fnindex DSINH
13392@cindex hyperbolic sine
13393@cindex hyperbolic function, sine
13394@cindex sine, hyperbolic
13395
13396@table @asis
13397@item @emph{Description}:
13398@code{SINH(X)} computes the hyperbolic sine of @var{X}.
13399
13400@item @emph{Standard}:
13401Fortran 90 and later, for a complex argument Fortran 2008 or later, has
13402a GNU extension
13403
13404@item @emph{Class}:
13405Elemental function
13406
13407@item @emph{Syntax}:
13408@code{RESULT = SINH(X)}
13409
13410@item @emph{Arguments}:
13411@multitable @columnfractions .15 .70
13412@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
13413@end multitable
13414
13415@item @emph{Return value}:
13416The return value has same type and kind as @var{X}.
13417
13418@item @emph{Example}:
13419@smallexample
13420program test_sinh
13421  real(8) :: x = - 1.0_8
13422  x = sinh(x)
13423end program test_sinh
13424@end smallexample
13425
13426@item @emph{Specific names}:
13427@multitable @columnfractions .20 .20 .20 .25
13428@item Name            @tab Argument          @tab Return type       @tab Standard
13429@item @code{DSINH(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab Fortran 90 and later
13430@end multitable
13431
13432@item @emph{See also}:
13433@ref{ASINH}
13434@end table
13435
13436
13437
13438@node SIZE
13439@section @code{SIZE} --- Determine the size of an array
13440@fnindex SIZE
13441@cindex array, size
13442@cindex array, number of elements
13443@cindex array, count elements
13444
13445@table @asis
13446@item @emph{Description}:
13447Determine the extent of @var{ARRAY} along a specified dimension @var{DIM},
13448or the total number of elements in @var{ARRAY} if @var{DIM} is absent.
13449
13450@item @emph{Standard}:
13451Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
13452
13453@item @emph{Class}:
13454Inquiry function
13455
13456@item @emph{Syntax}:
13457@code{RESULT = SIZE(ARRAY[, DIM [, KIND]])}
13458
13459@item @emph{Arguments}:
13460@multitable @columnfractions .15 .70
13461@item @var{ARRAY} @tab Shall be an array of any type. If @var{ARRAY} is
13462a pointer it must be associated and allocatable arrays must be allocated.
13463@item @var{DIM}   @tab (Optional) shall be a scalar of type @code{INTEGER}
13464and its value shall be in the range from 1 to n, where n equals the rank
13465of @var{ARRAY}.
13466@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
13467expression indicating the kind parameter of the result.
13468@end multitable
13469
13470@item @emph{Return value}:
13471The return value is of type @code{INTEGER} and of kind @var{KIND}. If
13472@var{KIND} is absent, the return value is of default integer kind.
13473
13474@item @emph{Example}:
13475@smallexample
13476PROGRAM test_size
13477  WRITE(*,*) SIZE((/ 1, 2 /))    ! 2
13478END PROGRAM
13479@end smallexample
13480
13481@item @emph{See also}:
13482@ref{SHAPE}, @gol
13483@ref{RESHAPE}
13484@end table
13485
13486
13487@node SIZEOF
13488@section @code{SIZEOF} --- Size in bytes of an expression
13489@fnindex SIZEOF
13490@cindex expression size
13491@cindex size of an expression
13492
13493@table @asis
13494@item @emph{Description}:
13495@code{SIZEOF(X)} calculates the number of bytes of storage the
13496expression @code{X} occupies.
13497
13498@item @emph{Standard}:
13499GNU extension
13500
13501@item @emph{Class}:
13502Inquiry function
13503
13504@item @emph{Syntax}:
13505@code{N = SIZEOF(X)}
13506
13507@item @emph{Arguments}:
13508@multitable @columnfractions .15 .70
13509@item @var{X} @tab The argument shall be of any type, rank or shape.
13510@end multitable
13511
13512@item @emph{Return value}:
13513The return value is of type integer and of the system-dependent kind
13514@var{C_SIZE_T} (from the @var{ISO_C_BINDING} module). Its value is the
13515number of bytes occupied by the argument.  If the argument has the
13516@code{POINTER} attribute, the number of bytes of the storage area pointed
13517to is returned.  If the argument is of a derived type with @code{POINTER}
13518or @code{ALLOCATABLE} components, the return value does not account for
13519the sizes of the data pointed to by these components. If the argument is
13520polymorphic, the size according to the dynamic type is returned. The argument
13521may not be a procedure or procedure pointer. Note that the code assumes for
13522arrays that those are contiguous; for contiguous arrays, it returns the
13523storage or an array element multiplied by the size of the array.
13524
13525@item @emph{Example}:
13526@smallexample
13527   integer :: i
13528   real :: r, s(5)
13529   print *, (sizeof(s)/sizeof(r) == 5)
13530   end
13531@end smallexample
13532The example will print @code{.TRUE.} unless you are using a platform
13533where default @code{REAL} variables are unusually padded.
13534
13535@item @emph{See also}:
13536@ref{C_SIZEOF}, @gol
13537@ref{STORAGE_SIZE}
13538@end table
13539
13540
13541@node SLEEP
13542@section @code{SLEEP} --- Sleep for the specified number of seconds
13543@fnindex SLEEP
13544@cindex delayed execution
13545
13546@table @asis
13547@item @emph{Description}:
13548Calling this subroutine causes the process to pause for @var{SECONDS} seconds.
13549
13550@item @emph{Standard}:
13551GNU extension
13552
13553@item @emph{Class}:
13554Subroutine
13555
13556@item @emph{Syntax}:
13557@code{CALL SLEEP(SECONDS)}
13558
13559@item @emph{Arguments}:
13560@multitable @columnfractions .15 .70
13561@item @var{SECONDS} @tab The type shall be of default @code{INTEGER}.
13562@end multitable
13563
13564@item @emph{Example}:
13565@smallexample
13566program test_sleep
13567  call sleep(5)
13568end
13569@end smallexample
13570@end table
13571
13572
13573
13574@node SPACING
13575@section @code{SPACING} --- Smallest distance between two numbers of a given type
13576@fnindex SPACING
13577@cindex real number, relative spacing
13578@cindex floating point, relative spacing
13579
13580@table @asis
13581@item @emph{Description}:
13582Determines the distance between the argument @var{X} and the nearest
13583adjacent number of the same type.
13584
13585@item @emph{Standard}:
13586Fortran 90 and later
13587
13588@item @emph{Class}:
13589Elemental function
13590
13591@item @emph{Syntax}:
13592@code{RESULT = SPACING(X)}
13593
13594@item @emph{Arguments}:
13595@multitable @columnfractions .15 .70
13596@item @var{X} @tab Shall be of type @code{REAL}.
13597@end multitable
13598
13599@item @emph{Return value}:
13600The result is of the same type as the input argument @var{X}.
13601
13602@item @emph{Example}:
13603@smallexample
13604PROGRAM test_spacing
13605  INTEGER, PARAMETER :: SGL = SELECTED_REAL_KIND(p=6, r=37)
13606  INTEGER, PARAMETER :: DBL = SELECTED_REAL_KIND(p=13, r=200)
13607
13608  WRITE(*,*) spacing(1.0_SGL)      ! "1.1920929E-07"          on i686
13609  WRITE(*,*) spacing(1.0_DBL)      ! "2.220446049250313E-016" on i686
13610END PROGRAM
13611@end smallexample
13612
13613@item @emph{See also}:
13614@ref{RRSPACING}
13615@end table
13616
13617
13618
13619@node SPREAD
13620@section @code{SPREAD} --- Add a dimension to an array
13621@fnindex SPREAD
13622@cindex array, increase dimension
13623@cindex array, duplicate elements
13624@cindex array, duplicate dimensions
13625
13626@table @asis
13627@item @emph{Description}:
13628Replicates a @var{SOURCE} array @var{NCOPIES} times along a specified
13629dimension @var{DIM}.
13630
13631@item @emph{Standard}:
13632Fortran 90 and later
13633
13634@item @emph{Class}:
13635Transformational function
13636
13637@item @emph{Syntax}:
13638@code{RESULT = SPREAD(SOURCE, DIM, NCOPIES)}
13639
13640@item @emph{Arguments}:
13641@multitable @columnfractions .15 .70
13642@item @var{SOURCE}  @tab Shall be a scalar or an array of any type and
13643a rank less than seven.
13644@item @var{DIM}     @tab Shall be a scalar of type @code{INTEGER} with a
13645value in the range from 1 to n+1, where n equals the rank of @var{SOURCE}.
13646@item @var{NCOPIES} @tab Shall be a scalar of type @code{INTEGER}.
13647@end multitable
13648
13649@item @emph{Return value}:
13650The result is an array of the same type as @var{SOURCE} and has rank n+1
13651where n equals the rank of @var{SOURCE}.
13652
13653@item @emph{Example}:
13654@smallexample
13655PROGRAM test_spread
13656  INTEGER :: a = 1, b(2) = (/ 1, 2 /)
13657  WRITE(*,*) SPREAD(A, 1, 2)            ! "1 1"
13658  WRITE(*,*) SPREAD(B, 1, 2)            ! "1 1 2 2"
13659END PROGRAM
13660@end smallexample
13661
13662@item @emph{See also}:
13663@ref{UNPACK}
13664@end table
13665
13666
13667
13668@node SQRT
13669@section @code{SQRT} --- Square-root function
13670@fnindex SQRT
13671@fnindex DSQRT
13672@fnindex CSQRT
13673@fnindex ZSQRT
13674@fnindex CDSQRT
13675@cindex root
13676@cindex square-root
13677
13678@table @asis
13679@item @emph{Description}:
13680@code{SQRT(X)} computes the square root of @var{X}.
13681
13682@item @emph{Standard}:
13683Fortran 77 and later
13684
13685@item @emph{Class}:
13686Elemental function
13687
13688@item @emph{Syntax}:
13689@code{RESULT = SQRT(X)}
13690
13691@item @emph{Arguments}:
13692@multitable @columnfractions .15 .70
13693@item @var{X} @tab The type shall be @code{REAL} or
13694@code{COMPLEX}.
13695@end multitable
13696
13697@item @emph{Return value}:
13698The return value is of type @code{REAL} or @code{COMPLEX}.
13699The kind type parameter is the same as @var{X}.
13700
13701@item @emph{Example}:
13702@smallexample
13703program test_sqrt
13704  real(8) :: x = 2.0_8
13705  complex :: z = (1.0, 2.0)
13706  x = sqrt(x)
13707  z = sqrt(z)
13708end program test_sqrt
13709@end smallexample
13710
13711@item @emph{Specific names}:
13712@multitable @columnfractions .20 .20 .20 .25
13713@item Name             @tab Argument             @tab Return type          @tab Standard
13714@item @code{SQRT(X)}   @tab @code{REAL(4) X}     @tab @code{REAL(4)}       @tab Fortran 77 and later
13715@item @code{DSQRT(X)}  @tab @code{REAL(8) X}     @tab @code{REAL(8)}       @tab Fortran 77 and later
13716@item @code{CSQRT(X)}  @tab @code{COMPLEX(4) X}  @tab @code{COMPLEX(4)}    @tab Fortran 77 and later
13717@item @code{ZSQRT(X)}  @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)}    @tab GNU extension
13718@item @code{CDSQRT(X)} @tab @code{COMPLEX(8) X}  @tab @code{COMPLEX(8)}    @tab GNU extension
13719@end multitable
13720@end table
13721
13722
13723
13724@node SRAND
13725@section @code{SRAND} --- Reinitialize the random number generator
13726@fnindex SRAND
13727@cindex random number generation, seeding
13728@cindex seeding a random number generator
13729
13730@table @asis
13731@item @emph{Description}:
13732@code{SRAND} reinitializes the pseudo-random number generator
13733called by @code{RAND} and @code{IRAND}. The new seed used by the
13734generator is specified by the required argument @var{SEED}.
13735
13736@item @emph{Standard}:
13737GNU extension
13738
13739@item @emph{Class}:
13740Subroutine
13741
13742@item @emph{Syntax}:
13743@code{CALL SRAND(SEED)}
13744
13745@item @emph{Arguments}:
13746@multitable @columnfractions .15 .70
13747@item @var{SEED} @tab Shall be a scalar @code{INTEGER(kind=4)}.
13748@end multitable
13749
13750@item @emph{Return value}:
13751Does not return anything.
13752
13753@item @emph{Example}:
13754See @code{RAND} and @code{IRAND} for examples.
13755
13756@item @emph{Notes}:
13757The Fortran standard specifies the intrinsic subroutines
13758@code{RANDOM_SEED} to initialize the pseudo-random number
13759generator and @code{RANDOM_NUMBER} to generate pseudo-random numbers.
13760These subroutines should be used in new codes.
13761
13762Please note that in GNU Fortran, these two sets of intrinsics (@code{RAND},
13763@code{IRAND} and @code{SRAND} on the one hand, @code{RANDOM_NUMBER} and
13764@code{RANDOM_SEED} on the other hand) access two independent
13765pseudo-random number generators.
13766
13767@item @emph{See also}:
13768@ref{RAND}, @gol
13769@ref{RANDOM_SEED}, @gol
13770@ref{RANDOM_NUMBER}
13771@end table
13772
13773
13774
13775@node STAT
13776@section @code{STAT} --- Get file status
13777@fnindex STAT
13778@cindex file system, file status
13779
13780@table @asis
13781@item @emph{Description}:
13782This function returns information about a file. No permissions are required on
13783the file itself, but execute (search) permission is required on all of the
13784directories in path that lead to the file.
13785
13786The elements that are obtained and stored in the array @code{VALUES}:
13787@multitable @columnfractions .15 .70
13788@item @code{VALUES(1)}   @tab  Device ID
13789@item @code{VALUES(2)}   @tab  Inode number
13790@item @code{VALUES(3)}   @tab  File mode
13791@item @code{VALUES(4)}   @tab  Number of links
13792@item @code{VALUES(5)}   @tab  Owner's uid
13793@item @code{VALUES(6)}   @tab  Owner's gid
13794@item @code{VALUES(7)}   @tab  ID of device containing directory entry for file (0 if not available)
13795@item @code{VALUES(8)}   @tab  File size (bytes)
13796@item @code{VALUES(9)}   @tab  Last access time
13797@item @code{VALUES(10)}  @tab  Last modification time
13798@item @code{VALUES(11)}  @tab  Last file status change time
13799@item @code{VALUES(12)}  @tab  Preferred I/O block size (-1 if not available)
13800@item @code{VALUES(13)}  @tab  Number of blocks allocated (-1 if not available)
13801@end multitable
13802
13803Not all these elements are relevant on all systems.
13804If an element is not relevant, it is returned as 0.
13805
13806This intrinsic is provided in both subroutine and function forms; however,
13807only one form can be used in any given program unit.
13808
13809@item @emph{Standard}:
13810GNU extension
13811
13812@item @emph{Class}:
13813Subroutine, function
13814
13815@item @emph{Syntax}:
13816@multitable @columnfractions .80
13817@item @code{CALL STAT(NAME, VALUES [, STATUS])}
13818@item @code{STATUS = STAT(NAME, VALUES)}
13819@end multitable
13820
13821@item @emph{Arguments}:
13822@multitable @columnfractions .15 .70
13823@item @var{NAME}   @tab The type shall be @code{CHARACTER}, of the
13824default kind and a valid path within the file system.
13825@item @var{VALUES} @tab The type shall be @code{INTEGER(4), DIMENSION(13)}.
13826@item @var{STATUS} @tab (Optional) status flag of type @code{INTEGER(4)}. Returns 0
13827on success and a system specific error code otherwise.
13828@end multitable
13829
13830@item @emph{Example}:
13831@smallexample
13832PROGRAM test_stat
13833  INTEGER, DIMENSION(13) :: buff
13834  INTEGER :: status
13835
13836  CALL STAT("/etc/passwd", buff, status)
13837
13838  IF (status == 0) THEN
13839    WRITE (*, FMT="('Device ID:',               T30, I19)") buff(1)
13840    WRITE (*, FMT="('Inode number:',            T30, I19)") buff(2)
13841    WRITE (*, FMT="('File mode (octal):',       T30, O19)") buff(3)
13842    WRITE (*, FMT="('Number of links:',         T30, I19)") buff(4)
13843    WRITE (*, FMT="('Owner''s uid:',            T30, I19)") buff(5)
13844    WRITE (*, FMT="('Owner''s gid:',            T30, I19)") buff(6)
13845    WRITE (*, FMT="('Device where located:',    T30, I19)") buff(7)
13846    WRITE (*, FMT="('File size:',               T30, I19)") buff(8)
13847    WRITE (*, FMT="('Last access time:',        T30, A19)") CTIME(buff(9))
13848    WRITE (*, FMT="('Last modification time',   T30, A19)") CTIME(buff(10))
13849    WRITE (*, FMT="('Last status change time:', T30, A19)") CTIME(buff(11))
13850    WRITE (*, FMT="('Preferred block size:',    T30, I19)") buff(12)
13851    WRITE (*, FMT="('No. of blocks allocated:', T30, I19)") buff(13)
13852  END IF
13853END PROGRAM
13854@end smallexample
13855
13856@item @emph{See also}:
13857To stat an open file: @gol
13858@ref{FSTAT} @gol
13859To stat a link: @gol
13860@ref{LSTAT}
13861@end table
13862
13863
13864
13865@node STORAGE_SIZE
13866@section @code{STORAGE_SIZE} --- Storage size in bits
13867@fnindex STORAGE_SIZE
13868@cindex storage size
13869
13870@table @asis
13871@item @emph{Description}:
13872Returns the storage size of argument @var{A} in bits.
13873@item @emph{Standard}:
13874Fortran 2008 and later
13875@item @emph{Class}:
13876Inquiry function
13877@item @emph{Syntax}:
13878@code{RESULT = STORAGE_SIZE(A [, KIND])}
13879
13880@item @emph{Arguments}:
13881@multitable @columnfractions .15 .70
13882@item @var{A} @tab Shall be a scalar or array of any type.
13883@item @var{KIND} @tab (Optional) shall be a scalar integer constant expression.
13884@end multitable
13885
13886@item @emph{Return Value}:
13887The result is a scalar integer with the kind type parameter specified by KIND
13888(or default integer type if KIND is missing). The result value is the size
13889expressed in bits for an element of an array that has the dynamic type and type
13890parameters of A.
13891
13892@item @emph{See also}:
13893@ref{C_SIZEOF}, @gol
13894@ref{SIZEOF}
13895@end table
13896
13897
13898
13899@node SUM
13900@section @code{SUM} --- Sum of array elements
13901@fnindex SUM
13902@cindex array, sum
13903@cindex array, add elements
13904@cindex array, conditionally add elements
13905@cindex sum array elements
13906
13907@table @asis
13908@item @emph{Description}:
13909Adds the elements of @var{ARRAY} along dimension @var{DIM} if
13910the corresponding element in @var{MASK} is @code{TRUE}.
13911
13912@item @emph{Standard}:
13913Fortran 90 and later
13914
13915@item @emph{Class}:
13916Transformational function
13917
13918@item @emph{Syntax}:
13919@multitable @columnfractions .80
13920@item @code{RESULT = SUM(ARRAY[, MASK])}
13921@item @code{RESULT = SUM(ARRAY, DIM[, MASK])}
13922@end multitable
13923
13924@item @emph{Arguments}:
13925@multitable @columnfractions .15 .70
13926@item @var{ARRAY} @tab Shall be an array of type @code{INTEGER},
13927@code{REAL} or @code{COMPLEX}.
13928@item @var{DIM}   @tab (Optional) shall be a scalar of type
13929@code{INTEGER} with a value in the range from 1 to n, where n
13930equals the rank of @var{ARRAY}.
13931@item @var{MASK}  @tab (Optional) shall be of type @code{LOGICAL}
13932and either be a scalar or an array of the same shape as @var{ARRAY}.
13933@end multitable
13934
13935@item @emph{Return value}:
13936The result is of the same type as @var{ARRAY}.
13937
13938If @var{DIM} is absent, a scalar with the sum of all elements in @var{ARRAY}
13939is returned. Otherwise, an array of rank n-1, where n equals the rank of
13940@var{ARRAY}, and a shape similar to that of @var{ARRAY} with dimension @var{DIM}
13941dropped is returned.
13942
13943@item @emph{Example}:
13944@smallexample
13945PROGRAM test_sum
13946  INTEGER :: x(5) = (/ 1, 2, 3, 4 ,5 /)
13947  print *, SUM(x)                        ! all elements, sum = 15
13948  print *, SUM(x, MASK=MOD(x, 2)==1)     ! odd elements, sum = 9
13949END PROGRAM
13950@end smallexample
13951
13952@item @emph{See also}:
13953@ref{PRODUCT}
13954@end table
13955
13956
13957
13958@node SYMLNK
13959@section @code{SYMLNK} --- Create a symbolic link
13960@fnindex SYMLNK
13961@cindex file system, create link
13962@cindex file system, soft link
13963
13964@table @asis
13965@item @emph{Description}:
13966Makes a symbolic link from file @var{PATH1} to @var{PATH2}. A null
13967character (@code{CHAR(0)}) can be used to mark the end of the names in
13968@var{PATH1} and @var{PATH2}; otherwise, trailing blanks in the file
13969names are ignored.  If the @var{STATUS} argument is supplied, it
13970contains 0 on success or a nonzero error code upon return; see
13971@code{symlink(2)}.  If the system does not supply @code{symlink(2)},
13972@code{ENOSYS} is returned.
13973
13974This intrinsic is provided in both subroutine and function forms;
13975however, only one form can be used in any given program unit.
13976
13977@item @emph{Standard}:
13978GNU extension
13979
13980@item @emph{Class}:
13981Subroutine, function
13982
13983@item @emph{Syntax}:
13984@multitable @columnfractions .80
13985@item @code{CALL SYMLNK(PATH1, PATH2 [, STATUS])}
13986@item @code{STATUS = SYMLNK(PATH1, PATH2)}
13987@end multitable
13988
13989@item @emph{Arguments}:
13990@multitable @columnfractions .15 .70
13991@item @var{PATH1} @tab Shall be of default @code{CHARACTER} type.
13992@item @var{PATH2} @tab Shall be of default @code{CHARACTER} type.
13993@item @var{STATUS} @tab (Optional) Shall be of default @code{INTEGER} type.
13994@end multitable
13995
13996@item @emph{See also}:
13997@ref{LINK}, @gol
13998@ref{UNLINK}
13999@end table
14000
14001
14002
14003@node SYSTEM
14004@section @code{SYSTEM} --- Execute a shell command
14005@fnindex SYSTEM
14006@cindex system, system call
14007
14008@table @asis
14009@item @emph{Description}:
14010Passes the command @var{COMMAND} to a shell (see @code{system(3)}). If
14011argument @var{STATUS} is present, it contains the value returned by
14012@code{system(3)}, which is presumably 0 if the shell command succeeded.
14013Note that which shell is used to invoke the command is system-dependent
14014and environment-dependent.
14015
14016This intrinsic is provided in both subroutine and function forms;
14017however, only one form can be used in any given program unit.
14018
14019Note that the @code{system} function need not be thread-safe. It is
14020the responsibility of the user to ensure that @code{system} is not
14021called concurrently.
14022
14023@item @emph{Standard}:
14024GNU extension
14025
14026@item @emph{Class}:
14027Subroutine, function
14028
14029@item @emph{Syntax}:
14030@multitable @columnfractions .80
14031@item @code{CALL SYSTEM(COMMAND [, STATUS])}
14032@item @code{STATUS = SYSTEM(COMMAND)}
14033@end multitable
14034
14035@item @emph{Arguments}:
14036@multitable @columnfractions .15 .70
14037@item @var{COMMAND} @tab Shall be of default @code{CHARACTER} type.
14038@item @var{STATUS}  @tab (Optional) Shall be of default @code{INTEGER} type.
14039@end multitable
14040
14041@item @emph{See also}:
14042@ref{EXECUTE_COMMAND_LINE}, which is part of the Fortran 2008 standard
14043and should considered in new code for future portability.
14044@end table
14045
14046
14047
14048@node SYSTEM_CLOCK
14049@section @code{SYSTEM_CLOCK} --- Time function
14050@fnindex SYSTEM_CLOCK
14051@cindex time, clock ticks
14052@cindex clock ticks
14053
14054@table @asis
14055@item @emph{Description}:
14056Determines the @var{COUNT} of a processor clock since an unspecified
14057time in the past modulo @var{COUNT_MAX}, @var{COUNT_RATE} determines
14058the number of clock ticks per second.  If the platform supports a
14059monotonic clock, that clock is used and can, depending on the platform
14060clock implementation, provide up to nanosecond resolution.  If a
14061monotonic clock is not available, the implementation falls back to a
14062realtime clock.
14063
14064@var{COUNT_RATE} is system dependent and can vary depending on the kind of
14065the arguments. For @var{kind=4} arguments (and smaller integer kinds),
14066@var{COUNT} represents milliseconds, while for @var{kind=8} arguments (and
14067larger integer kinds), @var{COUNT} typically represents micro- or
14068nanoseconds depending on resolution of the underlying platform clock.
14069@var{COUNT_MAX} usually equals @code{HUGE(COUNT_MAX)}. Note that the
14070millisecond resolution of the @var{kind=4} version implies that the
14071@var{COUNT} will wrap around in roughly 25 days. In order to avoid issues
14072with the wrap around and for more precise timing, please use the
14073@var{kind=8} version.
14074
14075If there is no clock, or querying the clock fails, @var{COUNT} is set
14076to @code{-HUGE(COUNT)}, and @var{COUNT_RATE} and @var{COUNT_MAX} are
14077set to zero.
14078
14079When running on a platform using the GNU C library (glibc) version
140802.16 or older, or a derivative thereof, the high resolution monotonic
14081clock is available only when linking with the @var{rt} library.  This
14082can be done explicitly by adding the @code{-lrt} flag when linking the
14083application, but is also done implicitly when using OpenMP.
14084
14085On the Windows platform, the version with @var{kind=4} arguments uses
14086the @code{GetTickCount} function, whereas the @var{kind=8} version
14087uses @code{QueryPerformanceCounter} and
14088@code{QueryPerformanceCounterFrequency}. For more information, and
14089potential caveats, please see the platform documentation.
14090
14091@item @emph{Standard}:
14092Fortran 90 and later
14093
14094@item @emph{Class}:
14095Subroutine
14096
14097@item @emph{Syntax}:
14098@code{CALL SYSTEM_CLOCK([COUNT, COUNT_RATE, COUNT_MAX])}
14099
14100@item @emph{Arguments}:
14101@multitable @columnfractions .20 .65
14102@item @var{COUNT}      @tab (Optional) shall be a scalar of type
14103@code{INTEGER} with @code{INTENT(OUT)}.
14104@item @var{COUNT_RATE} @tab (Optional) shall be a scalar of type
14105@code{INTEGER} or @code{REAL}, with @code{INTENT(OUT)}.
14106@item @var{COUNT_MAX}  @tab (Optional) shall be a scalar of type
14107@code{INTEGER} with @code{INTENT(OUT)}.
14108@end multitable
14109
14110@item @emph{Example}:
14111@smallexample
14112PROGRAM test_system_clock
14113  INTEGER :: count, count_rate, count_max
14114  CALL SYSTEM_CLOCK(count, count_rate, count_max)
14115  WRITE(*,*) count, count_rate, count_max
14116END PROGRAM
14117@end smallexample
14118
14119@item @emph{See also}:
14120@ref{DATE_AND_TIME}, @gol
14121@ref{CPU_TIME}
14122@end table
14123
14124
14125
14126@node TAN
14127@section @code{TAN} --- Tangent function
14128@fnindex TAN
14129@fnindex DTAN
14130@cindex trigonometric function, tangent
14131@cindex tangent
14132
14133@table @asis
14134@item @emph{Description}:
14135@code{TAN(X)} computes the tangent of @var{X}.
14136
14137@item @emph{Standard}:
14138Fortran 77 and later, for a complex argument Fortran 2008 or later
14139
14140@item @emph{Class}:
14141Elemental function
14142
14143@item @emph{Syntax}:
14144@code{RESULT = TAN(X)}
14145
14146@item @emph{Arguments}:
14147@multitable @columnfractions .15 .70
14148@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
14149@end multitable
14150
14151@item @emph{Return value}:
14152The return value has same type and kind as @var{X}, and its value is in radians.
14153
14154@item @emph{Example}:
14155@smallexample
14156program test_tan
14157  real(8) :: x = 0.165_8
14158  x = tan(x)
14159end program test_tan
14160@end smallexample
14161
14162@item @emph{Specific names}:
14163@multitable @columnfractions .20 .20 .20 .25
14164@item Name            @tab Argument          @tab Return type     @tab Standard
14165@item @code{TAN(X)}   @tab @code{REAL(4) X}  @tab @code{REAL(4)}  @tab Fortran 77 and later
14166@item @code{DTAN(X)}  @tab @code{REAL(8) X}  @tab @code{REAL(8)}  @tab Fortran 77 and later
14167@end multitable
14168
14169@item @emph{See also}:
14170Inverse function: @gol
14171@ref{ATAN} @gol
14172Degrees function: @gol
14173@ref{TAND}
14174@end table
14175
14176
14177
14178@node TAND
14179@section @code{TAND} --- Tangent function, degrees
14180@fnindex TAND
14181@fnindex DTAND
14182@cindex trigonometric function, tangent, degrees
14183@cindex tangent, degrees
14184
14185@table @asis
14186@item @emph{Description}:
14187@code{TAND(X)} computes the tangent of @var{X} in degrees.
14188
14189This function is for compatibility only and should be avoided in favor of
14190standard constructs wherever possible.
14191
14192@item @emph{Standard}:
14193GNU extension, enabled with @option{-fdec-math}.
14194
14195@item @emph{Class}:
14196Elemental function
14197
14198@item @emph{Syntax}:
14199@code{RESULT = TAND(X)}
14200
14201@item @emph{Arguments}:
14202@multitable @columnfractions .15 .70
14203@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
14204@end multitable
14205
14206@item @emph{Return value}:
14207The return value has same type and kind as @var{X}, and its value is in degrees.
14208
14209@item @emph{Example}:
14210@smallexample
14211program test_tand
14212  real(8) :: x = 0.165_8
14213  x = tand(x)
14214end program test_tand
14215@end smallexample
14216
14217@item @emph{Specific names}:
14218@multitable @columnfractions .20 .20 .20 .25
14219@item Name            @tab Argument          @tab Return type     @tab Standard
14220@item @code{TAND(X)}   @tab @code{REAL(4) X}  @tab @code{REAL(4)}  @tab GNU extension
14221@item @code{DTAND(X)}  @tab @code{REAL(8) X}  @tab @code{REAL(8)}  @tab GNU extension
14222@end multitable
14223
14224@item @emph{See also}:
14225Inverse function: @gol
14226@ref{ATAND} @gol
14227Radians function: @gol
14228@ref{TAN}
14229@end table
14230
14231
14232
14233@node TANH
14234@section @code{TANH} --- Hyperbolic tangent function
14235@fnindex TANH
14236@fnindex DTANH
14237@cindex hyperbolic tangent
14238@cindex hyperbolic function, tangent
14239@cindex tangent, hyperbolic
14240
14241@table @asis
14242@item @emph{Description}:
14243@code{TANH(X)} computes the hyperbolic tangent of @var{X}.
14244
14245@item @emph{Standard}:
14246Fortran 77 and later, for a complex argument Fortran 2008 or later
14247
14248@item @emph{Class}:
14249Elemental function
14250
14251@item @emph{Syntax}:
14252@code{X = TANH(X)}
14253
14254@item @emph{Arguments}:
14255@multitable @columnfractions .15 .70
14256@item @var{X} @tab The type shall be @code{REAL} or @code{COMPLEX}.
14257@end multitable
14258
14259@item @emph{Return value}:
14260The return value has same type and kind as @var{X}. If @var{X} is
14261complex, the imaginary part of the result is in radians. If @var{X}
14262is @code{REAL}, the return value lies in the range
14263@math{ - 1 \leq tanh(x) \leq 1 }.
14264
14265@item @emph{Example}:
14266@smallexample
14267program test_tanh
14268  real(8) :: x = 2.1_8
14269  x = tanh(x)
14270end program test_tanh
14271@end smallexample
14272
14273@item @emph{Specific names}:
14274@multitable @columnfractions .20 .20 .20 .25
14275@item Name            @tab Argument          @tab Return type       @tab Standard
14276@item @code{TANH(X)}  @tab @code{REAL(4) X}  @tab @code{REAL(4)}    @tab Fortran 77 and later
14277@item @code{DTANH(X)} @tab @code{REAL(8) X}  @tab @code{REAL(8)}    @tab Fortran 77 and later
14278@end multitable
14279
14280@item @emph{See also}:
14281@ref{ATANH}
14282@end table
14283
14284
14285
14286@node THIS_IMAGE
14287@section @code{THIS_IMAGE} --- Function that returns the cosubscript index of this image
14288@fnindex THIS_IMAGE
14289@cindex coarray, @code{THIS_IMAGE}
14290@cindex images, index of this image
14291
14292@table @asis
14293@item @emph{Description}:
14294Returns the cosubscript for this image.
14295
14296@item @emph{Standard}:
14297Fortran 2008 and later. With @var{DISTANCE} argument,
14298Technical Specification (TS) 18508 or later
14299
14300@item @emph{Class}:
14301Transformational function
14302
14303@item @emph{Syntax}:
14304@multitable @columnfractions .80
14305@item @code{RESULT = THIS_IMAGE()}
14306@item @code{RESULT = THIS_IMAGE(DISTANCE)}
14307@item @code{RESULT = THIS_IMAGE(COARRAY [, DIM])}
14308@end multitable
14309
14310@item @emph{Arguments}:
14311@multitable @columnfractions .15 .70
14312@item @var{DISTANCE} @tab (optional, intent(in)) Nonnegative scalar integer
14313(not permitted together with @var{COARRAY}).
14314@item @var{COARRAY} @tab Coarray of any type  (optional; if @var{DIM}
14315present, required).
14316@item @var{DIM}     @tab default integer scalar (optional). If present,
14317@var{DIM} shall be between one and the corank of @var{COARRAY}.
14318@end multitable
14319
14320
14321@item @emph{Return value}:
14322Default integer. If @var{COARRAY} is not present, it is scalar; if
14323@var{DISTANCE} is not present or has value 0, its value is the image index on
14324the invoking image for the current team, for values smaller or equal
14325distance to the initial team, it returns the image index on the ancestor team
14326which has a distance of @var{DISTANCE} from the invoking team. If
14327@var{DISTANCE} is larger than the distance to the initial team, the image
14328index of the initial team is returned. Otherwise when the @var{COARRAY} is
14329present, if @var{DIM} is not present, a rank-1 array with corank elements is
14330returned, containing the cosubscripts for @var{COARRAY} specifying the invoking
14331image. If @var{DIM} is present, a scalar is returned, with the value of
14332the @var{DIM} element of @code{THIS_IMAGE(COARRAY)}.
14333
14334@item @emph{Example}:
14335@smallexample
14336INTEGER :: value[*]
14337INTEGER :: i
14338value = THIS_IMAGE()
14339SYNC ALL
14340IF (THIS_IMAGE() == 1) THEN
14341  DO i = 1, NUM_IMAGES()
14342    WRITE(*,'(2(a,i0))') 'value[', i, '] is ', value[i]
14343  END DO
14344END IF
14345
14346! Check whether the current image is the initial image
14347IF (THIS_IMAGE(HUGE(1)) /= THIS_IMAGE())
14348  error stop "something is rotten here"
14349@end smallexample
14350
14351@item @emph{See also}:
14352@ref{NUM_IMAGES}, @gol
14353@ref{IMAGE_INDEX}
14354@end table
14355
14356
14357
14358@node TIME
14359@section @code{TIME} --- Time function
14360@fnindex TIME
14361@cindex time, current
14362@cindex current time
14363
14364@table @asis
14365@item @emph{Description}:
14366Returns the current time encoded as an integer (in the manner of the
14367function @code{time(3)} in the C standard library). This value is
14368suitable for passing to @ref{CTIME}, @ref{GMTIME}, and @ref{LTIME}.
14369
14370This intrinsic is not fully portable, such as to systems with 32-bit
14371@code{INTEGER} types but supporting times wider than 32 bits. Therefore,
14372the values returned by this intrinsic might be, or become, negative, or
14373numerically less than previous values, during a single run of the
14374compiled program.
14375
14376See @ref{TIME8}, for information on a similar intrinsic that might be
14377portable to more GNU Fortran implementations, though to fewer Fortran
14378compilers.
14379
14380@item @emph{Standard}:
14381GNU extension
14382
14383@item @emph{Class}:
14384Function
14385
14386@item @emph{Syntax}:
14387@code{RESULT = TIME()}
14388
14389@item @emph{Return value}:
14390The return value is a scalar of type @code{INTEGER(4)}.
14391
14392@item @emph{See also}:
14393@ref{DATE_AND_TIME}, @gol
14394@ref{CTIME}, @gol
14395@ref{GMTIME}, @gol
14396@ref{LTIME}, @gol
14397@ref{MCLOCK}, @gol
14398@ref{TIME8}
14399@end table
14400
14401
14402
14403@node TIME8
14404@section @code{TIME8} --- Time function (64-bit)
14405@fnindex TIME8
14406@cindex time, current
14407@cindex current time
14408
14409@table @asis
14410@item @emph{Description}:
14411Returns the current time encoded as an integer (in the manner of the
14412function @code{time(3)} in the C standard library). This value is
14413suitable for passing to @ref{CTIME}, @ref{GMTIME}, and @ref{LTIME}.
14414
14415@emph{Warning:} this intrinsic does not increase the range of the timing
14416values over that returned by @code{time(3)}. On a system with a 32-bit
14417@code{time(3)}, @code{TIME8} will return a 32-bit value, even though
14418it is converted to a 64-bit @code{INTEGER(8)} value. That means
14419overflows of the 32-bit value can still occur. Therefore, the values
14420returned by this intrinsic might be or become negative or numerically
14421less than previous values during a single run of the compiled program.
14422
14423@item @emph{Standard}:
14424GNU extension
14425
14426@item @emph{Class}:
14427Function
14428
14429@item @emph{Syntax}:
14430@code{RESULT = TIME8()}
14431
14432@item @emph{Return value}:
14433The return value is a scalar of type @code{INTEGER(8)}.
14434
14435@item @emph{See also}:
14436@ref{DATE_AND_TIME}, @gol
14437@ref{CTIME}, @gol
14438@ref{GMTIME}, @gol
14439@ref{LTIME}, @gol
14440@ref{MCLOCK8}, @gol
14441@ref{TIME}
14442@end table
14443
14444
14445
14446@node TINY
14447@section @code{TINY} --- Smallest positive number of a real kind
14448@fnindex TINY
14449@cindex limits, smallest number
14450@cindex model representation, smallest number
14451
14452@table @asis
14453@item @emph{Description}:
14454@code{TINY(X)} returns the smallest positive (non zero) number
14455in the model of the type of @code{X}.
14456
14457@item @emph{Standard}:
14458Fortran 90 and later
14459
14460@item @emph{Class}:
14461Inquiry function
14462
14463@item @emph{Syntax}:
14464@code{RESULT = TINY(X)}
14465
14466@item @emph{Arguments}:
14467@multitable @columnfractions .15 .70
14468@item @var{X} @tab Shall be of type @code{REAL}.
14469@end multitable
14470
14471@item @emph{Return value}:
14472The return value is of the same type and kind as @var{X}
14473
14474@item @emph{Example}:
14475See @code{HUGE} for an example.
14476@end table
14477
14478
14479
14480@node TRAILZ
14481@section @code{TRAILZ} --- Number of trailing zero bits of an integer
14482@fnindex TRAILZ
14483@cindex zero bits
14484
14485@table @asis
14486@item @emph{Description}:
14487@code{TRAILZ} returns the number of trailing zero bits of an integer.
14488
14489@item @emph{Standard}:
14490Fortran 2008 and later
14491
14492@item @emph{Class}:
14493Elemental function
14494
14495@item @emph{Syntax}:
14496@code{RESULT = TRAILZ(I)}
14497
14498@item @emph{Arguments}:
14499@multitable @columnfractions .15 .70
14500@item @var{I} @tab Shall be of type @code{INTEGER}.
14501@end multitable
14502
14503@item @emph{Return value}:
14504The type of the return value is the default @code{INTEGER}.
14505If all the bits of @code{I} are zero, the result value is @code{BIT_SIZE(I)}.
14506
14507@item @emph{Example}:
14508@smallexample
14509PROGRAM test_trailz
14510  WRITE (*,*) TRAILZ(8)  ! prints 3
14511END PROGRAM
14512@end smallexample
14513
14514@item @emph{See also}:
14515@ref{BIT_SIZE}, @gol
14516@ref{LEADZ}, @gol
14517@ref{POPPAR}, @gol
14518@ref{POPCNT}
14519@end table
14520
14521
14522
14523@node TRANSFER
14524@section @code{TRANSFER} --- Transfer bit patterns
14525@fnindex TRANSFER
14526@cindex bits, move
14527@cindex type cast
14528
14529@table @asis
14530@item @emph{Description}:
14531Interprets the bitwise representation of @var{SOURCE} in memory as if it
14532is the representation of a variable or array of the same type and type
14533parameters as @var{MOLD}.
14534
14535This is approximately equivalent to the C concept of @emph{casting} one
14536type to another.
14537
14538@item @emph{Standard}:
14539Fortran 90 and later
14540
14541@item @emph{Class}:
14542Transformational function
14543
14544@item @emph{Syntax}:
14545@code{RESULT = TRANSFER(SOURCE, MOLD[, SIZE])}
14546
14547@item @emph{Arguments}:
14548@multitable @columnfractions .15 .70
14549@item @var{SOURCE} @tab Shall be a scalar or an array of any type.
14550@item @var{MOLD}   @tab Shall be a scalar or an array of any type.
14551@item @var{SIZE}   @tab (Optional) shall be a scalar of type
14552@code{INTEGER}.
14553@end multitable
14554
14555@item @emph{Return value}:
14556The result has the same type as @var{MOLD}, with the bit level
14557representation of @var{SOURCE}.  If @var{SIZE} is present, the result is
14558a one-dimensional array of length @var{SIZE}.  If @var{SIZE} is absent
14559but @var{MOLD} is an array (of any size or shape), the result is a one-
14560dimensional array of the minimum length needed to contain the entirety
14561of the bitwise representation of @var{SOURCE}.   If @var{SIZE} is absent
14562and @var{MOLD} is a scalar, the result is a scalar.
14563
14564If the bitwise representation of the result is longer than that of
14565@var{SOURCE}, then the leading bits of the result correspond to those of
14566@var{SOURCE} and any trailing bits are filled arbitrarily.
14567
14568When the resulting bit representation does not correspond to a valid
14569representation of a variable of the same type as @var{MOLD}, the results
14570are undefined, and subsequent operations on the result cannot be
14571guaranteed to produce sensible behavior.  For example, it is possible to
14572create @code{LOGICAL} variables for which @code{@var{VAR}} and
14573@code{.NOT.@var{VAR}} both appear to be true.
14574
14575@item @emph{Example}:
14576@smallexample
14577PROGRAM test_transfer
14578  integer :: x = 2143289344
14579  print *, transfer(x, 1.0)    ! prints "NaN" on i686
14580END PROGRAM
14581@end smallexample
14582@end table
14583
14584
14585
14586@node TRANSPOSE
14587@section @code{TRANSPOSE} --- Transpose an array of rank two
14588@fnindex TRANSPOSE
14589@cindex array, transpose
14590@cindex matrix, transpose
14591@cindex transpose
14592
14593@table @asis
14594@item @emph{Description}:
14595Transpose an array of rank two. Element (i, j) of the result has the value
14596@code{MATRIX(j, i)}, for all i, j.
14597
14598@item @emph{Standard}:
14599Fortran 90 and later
14600
14601@item @emph{Class}:
14602Transformational function
14603
14604@item @emph{Syntax}:
14605@code{RESULT = TRANSPOSE(MATRIX)}
14606
14607@item @emph{Arguments}:
14608@multitable @columnfractions .15 .70
14609@item @var{MATRIX} @tab Shall be an array of any type and have a rank of two.
14610@end multitable
14611
14612@item @emph{Return value}:
14613The result has the same type as @var{MATRIX}, and has shape
14614@code{(/ m, n /)} if @var{MATRIX} has shape @code{(/ n, m /)}.
14615@end table
14616
14617
14618
14619@node TRIM
14620@section @code{TRIM} --- Remove trailing blank characters of a string
14621@fnindex TRIM
14622@cindex string, remove trailing whitespace
14623
14624@table @asis
14625@item @emph{Description}:
14626Removes trailing blank characters of a string.
14627
14628@item @emph{Standard}:
14629Fortran 90 and later
14630
14631@item @emph{Class}:
14632Transformational function
14633
14634@item @emph{Syntax}:
14635@code{RESULT = TRIM(STRING)}
14636
14637@item @emph{Arguments}:
14638@multitable @columnfractions .15 .70
14639@item @var{STRING} @tab Shall be a scalar of type @code{CHARACTER}.
14640@end multitable
14641
14642@item @emph{Return value}:
14643A scalar of type @code{CHARACTER} which length is that of @var{STRING}
14644less the number of trailing blanks.
14645
14646@item @emph{Example}:
14647@smallexample
14648PROGRAM test_trim
14649  CHARACTER(len=10), PARAMETER :: s = "GFORTRAN  "
14650  WRITE(*,*) LEN(s), LEN(TRIM(s))  ! "10 8", with/without trailing blanks
14651END PROGRAM
14652@end smallexample
14653
14654@item @emph{See also}:
14655@ref{ADJUSTL}, @gol
14656@ref{ADJUSTR}
14657@end table
14658
14659
14660
14661@node TTYNAM
14662@section @code{TTYNAM} --- Get the name of a terminal device.
14663@fnindex TTYNAM
14664@cindex system, terminal
14665
14666@table @asis
14667@item @emph{Description}:
14668Get the name of a terminal device. For more information,
14669see @code{ttyname(3)}.
14670
14671This intrinsic is provided in both subroutine and function forms;
14672however, only one form can be used in any given program unit.
14673
14674@item @emph{Standard}:
14675GNU extension
14676
14677@item @emph{Class}:
14678Subroutine, function
14679
14680@item @emph{Syntax}:
14681@multitable @columnfractions .80
14682@item @code{CALL TTYNAM(UNIT, NAME)}
14683@item @code{NAME = TTYNAM(UNIT)}
14684@end multitable
14685
14686@item @emph{Arguments}:
14687@multitable @columnfractions .15 .70
14688@item @var{UNIT} @tab Shall be a scalar @code{INTEGER}.
14689@item @var{NAME} @tab Shall be of type @code{CHARACTER}.
14690@end multitable
14691
14692@item @emph{Example}:
14693@smallexample
14694PROGRAM test_ttynam
14695  INTEGER :: unit
14696  DO unit = 1, 10
14697    IF (isatty(unit=unit)) write(*,*) ttynam(unit)
14698  END DO
14699END PROGRAM
14700@end smallexample
14701
14702@item @emph{See also}:
14703@ref{ISATTY}
14704@end table
14705
14706
14707
14708@node UBOUND
14709@section @code{UBOUND} --- Upper dimension bounds of an array
14710@fnindex UBOUND
14711@cindex array, upper bound
14712
14713@table @asis
14714@item @emph{Description}:
14715Returns the upper bounds of an array, or a single upper bound
14716along the @var{DIM} dimension.
14717@item @emph{Standard}:
14718Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
14719
14720@item @emph{Class}:
14721Inquiry function
14722
14723@item @emph{Syntax}:
14724@code{RESULT = UBOUND(ARRAY [, DIM [, KIND]])}
14725
14726@item @emph{Arguments}:
14727@multitable @columnfractions .15 .70
14728@item @var{ARRAY} @tab Shall be an array, of any type.
14729@item @var{DIM} @tab (Optional) Shall be a scalar @code{INTEGER}.
14730@item @var{KIND}@tab (Optional) An @code{INTEGER} initialization
14731expression indicating the kind parameter of the result.
14732@end multitable
14733
14734@item @emph{Return value}:
14735The return value is of type @code{INTEGER} and of kind @var{KIND}. If
14736@var{KIND} is absent, the return value is of default integer kind.
14737If @var{DIM} is absent, the result is an array of the upper bounds of
14738@var{ARRAY}.  If @var{DIM} is present, the result is a scalar
14739corresponding to the upper bound of the array along that dimension.  If
14740@var{ARRAY} is an expression rather than a whole array or array
14741structure component, or if it has a zero extent along the relevant
14742dimension, the upper bound is taken to be the number of elements along
14743the relevant dimension.
14744
14745@item @emph{See also}:
14746@ref{LBOUND}, @gol
14747@ref{LCOBOUND}
14748@end table
14749
14750
14751
14752@node UCOBOUND
14753@section @code{UCOBOUND} --- Upper codimension bounds of an array
14754@fnindex UCOBOUND
14755@cindex coarray, upper bound
14756
14757@table @asis
14758@item @emph{Description}:
14759Returns the upper cobounds of a coarray, or a single upper cobound
14760along the @var{DIM} codimension.
14761@item @emph{Standard}:
14762Fortran 2008 and later
14763
14764@item @emph{Class}:
14765Inquiry function
14766
14767@item @emph{Syntax}:
14768@code{RESULT = UCOBOUND(COARRAY [, DIM [, KIND]])}
14769
14770@item @emph{Arguments}:
14771@multitable @columnfractions .15 .70
14772@item @var{ARRAY} @tab Shall be an coarray, of any type.
14773@item @var{DIM} @tab (Optional) Shall be a scalar @code{INTEGER}.
14774@item @var{KIND} @tab (Optional) An @code{INTEGER} initialization
14775expression indicating the kind parameter of the result.
14776@end multitable
14777
14778@item @emph{Return value}:
14779The return value is of type @code{INTEGER} and of kind @var{KIND}. If
14780@var{KIND} is absent, the return value is of default integer kind.
14781If @var{DIM} is absent, the result is an array of the lower cobounds of
14782@var{COARRAY}.  If @var{DIM} is present, the result is a scalar
14783corresponding to the lower cobound of the array along that codimension.
14784
14785@item @emph{See also}:
14786@ref{LCOBOUND}, @gol
14787@ref{LBOUND}
14788@end table
14789
14790
14791
14792@node UMASK
14793@section @code{UMASK} --- Set the file creation mask
14794@fnindex UMASK
14795@cindex file system, file creation mask
14796
14797@table @asis
14798@item @emph{Description}:
14799Sets the file creation mask to @var{MASK}. If called as a function, it
14800returns the old value. If called as a subroutine and argument @var{OLD}
14801if it is supplied, it is set to the old value. See @code{umask(2)}.
14802
14803@item @emph{Standard}:
14804GNU extension
14805
14806@item @emph{Class}:
14807Subroutine, function
14808
14809@item @emph{Syntax}:
14810@multitable @columnfractions .80
14811@item @code{CALL UMASK(MASK [, OLD])}
14812@item @code{OLD = UMASK(MASK)}
14813@end multitable
14814
14815@item @emph{Arguments}:
14816@multitable @columnfractions .15 .70
14817@item @var{MASK} @tab Shall be a scalar of type @code{INTEGER}.
14818@item @var{OLD} @tab (Optional) Shall be a scalar of type
14819@code{INTEGER}.
14820@end multitable
14821
14822@end table
14823
14824
14825
14826@node UNLINK
14827@section @code{UNLINK} --- Remove a file from the file system
14828@fnindex UNLINK
14829@cindex file system, remove file
14830
14831@table @asis
14832@item @emph{Description}:
14833Unlinks the file @var{PATH}. A null character (@code{CHAR(0)}) can be
14834used to mark the end of the name in @var{PATH}; otherwise, trailing
14835blanks in the file name are ignored.  If the @var{STATUS} argument is
14836supplied, it contains 0 on success or a nonzero error code upon return;
14837see @code{unlink(2)}.
14838
14839This intrinsic is provided in both subroutine and function forms;
14840however, only one form can be used in any given program unit.
14841
14842@item @emph{Standard}:
14843GNU extension
14844
14845@item @emph{Class}:
14846Subroutine, function
14847
14848@item @emph{Syntax}:
14849@multitable @columnfractions .80
14850@item @code{CALL UNLINK(PATH [, STATUS])}
14851@item @code{STATUS = UNLINK(PATH)}
14852@end multitable
14853
14854@item @emph{Arguments}:
14855@multitable @columnfractions .15 .70
14856@item @var{PATH} @tab Shall be of default @code{CHARACTER} type.
14857@item @var{STATUS} @tab (Optional) Shall be of default @code{INTEGER} type.
14858@end multitable
14859
14860@item @emph{See also}:
14861@ref{LINK}, @gol
14862@ref{SYMLNK}
14863@end table
14864
14865
14866
14867@node UNPACK
14868@section @code{UNPACK} --- Unpack an array of rank one into an array
14869@fnindex UNPACK
14870@cindex array, unpacking
14871@cindex array, increase dimension
14872@cindex array, scatter elements
14873
14874@table @asis
14875@item @emph{Description}:
14876Store the elements of @var{VECTOR} in an array of higher rank.
14877
14878@item @emph{Standard}:
14879Fortran 90 and later
14880
14881@item @emph{Class}:
14882Transformational function
14883
14884@item @emph{Syntax}:
14885@code{RESULT = UNPACK(VECTOR, MASK, FIELD)}
14886
14887@item @emph{Arguments}:
14888@multitable @columnfractions .15 .70
14889@item @var{VECTOR} @tab Shall be an array of any type and rank one. It
14890shall have at least as many elements as @var{MASK} has @code{TRUE} values.
14891@item @var{MASK}   @tab Shall be an array of type @code{LOGICAL}.
14892@item @var{FIELD}  @tab Shall be of the same type as @var{VECTOR} and have
14893the same shape as @var{MASK}.
14894@end multitable
14895
14896@item @emph{Return value}:
14897The resulting array corresponds to @var{FIELD} with @code{TRUE} elements
14898of @var{MASK} replaced by values from @var{VECTOR} in array element order.
14899
14900@item @emph{Example}:
14901@smallexample
14902PROGRAM test_unpack
14903  integer :: vector(2)  = (/1,1/)
14904  logical :: mask(4)  = (/ .TRUE., .FALSE., .FALSE., .TRUE. /)
14905  integer :: field(2,2) = 0, unity(2,2)
14906
14907  ! result: unity matrix
14908  unity = unpack(vector, reshape(mask, (/2,2/)), field)
14909END PROGRAM
14910@end smallexample
14911
14912@item @emph{See also}:
14913@ref{PACK}, @gol
14914@ref{SPREAD}
14915@end table
14916
14917
14918
14919@node VERIFY
14920@section @code{VERIFY} --- Scan a string for characters not a given set
14921@fnindex VERIFY
14922@cindex string, find missing set
14923
14924@table @asis
14925@item @emph{Description}:
14926Verifies that all the characters in @var{STRING} belong to the set of
14927characters in @var{SET}.
14928
14929If @var{BACK} is either absent or equals @code{FALSE}, this function
14930returns the position of the leftmost character of @var{STRING} that is
14931not in @var{SET}. If @var{BACK} equals @code{TRUE}, the rightmost
14932position is returned. If all characters of @var{STRING} are found in
14933@var{SET}, the result is zero.
14934
14935@item @emph{Standard}:
14936Fortran 90 and later, with @var{KIND} argument Fortran 2003 and later
14937
14938@item @emph{Class}:
14939Elemental function
14940
14941@item @emph{Syntax}:
14942@code{RESULT = VERIFY(STRING, SET[, BACK [, KIND]])}
14943
14944@item @emph{Arguments}:
14945@multitable @columnfractions .15 .70
14946@item @var{STRING} @tab Shall be of type @code{CHARACTER}.
14947@item @var{SET}    @tab Shall be of type @code{CHARACTER}.
14948@item @var{BACK}   @tab (Optional) shall be of type @code{LOGICAL}.
14949@item @var{KIND}   @tab (Optional) An @code{INTEGER} initialization
14950expression indicating the kind parameter of the result.
14951@end multitable
14952
14953@item @emph{Return value}:
14954The return value is of type @code{INTEGER} and of kind @var{KIND}. If
14955@var{KIND} is absent, the return value is of default integer kind.
14956
14957@item @emph{Example}:
14958@smallexample
14959PROGRAM test_verify
14960  WRITE(*,*) VERIFY("FORTRAN", "AO")           ! 1, found 'F'
14961  WRITE(*,*) VERIFY("FORTRAN", "FOO")          ! 3, found 'R'
14962  WRITE(*,*) VERIFY("FORTRAN", "C++")          ! 1, found 'F'
14963  WRITE(*,*) VERIFY("FORTRAN", "C++", .TRUE.)  ! 7, found 'N'
14964  WRITE(*,*) VERIFY("FORTRAN", "FORTRAN")      ! 0' found none
14965END PROGRAM
14966@end smallexample
14967
14968@item @emph{See also}:
14969@ref{SCAN}, @gol
14970@ref{INDEX intrinsic}
14971@end table
14972
14973
14974
14975@node XOR
14976@section @code{XOR} --- Bitwise logical exclusive OR
14977@fnindex XOR
14978@cindex bitwise logical exclusive or
14979@cindex logical exclusive or, bitwise
14980
14981@table @asis
14982@item @emph{Description}:
14983Bitwise logical exclusive or.
14984
14985This intrinsic routine is provided for backwards compatibility with
14986GNU Fortran 77.  For integer arguments, programmers should consider
14987the use of the @ref{IEOR} intrinsic and for logical arguments the
14988@code{.NEQV.} operator, which are both defined by the Fortran standard.
14989
14990@item @emph{Standard}:
14991GNU extension
14992
14993@item @emph{Class}:
14994Function
14995
14996@item @emph{Syntax}:
14997@code{RESULT = XOR(I, J)}
14998
14999@item @emph{Arguments}:
15000@multitable @columnfractions .15 .70
15001@item @var{I} @tab The type shall be either a scalar @code{INTEGER}
15002type or a scalar @code{LOGICAL} type or a boz-literal-constant.
15003@item @var{J} @tab The type shall be the same as the type of @var{I} or
15004a boz-literal-constant. @var{I} and @var{J} shall not both be
15005boz-literal-constants.  If either @var{I} and @var{J} is a
15006boz-literal-constant, then the other argument must be a scalar @code{INTEGER}.
15007@end multitable
15008
15009@item @emph{Return value}:
15010The return type is either a scalar @code{INTEGER} or a scalar
15011@code{LOGICAL}.  If the kind type parameters differ, then the
15012smaller kind type is implicitly converted to larger kind, and the
15013return has the larger kind.  A boz-literal-constant is
15014converted to an @code{INTEGER} with the kind type parameter of
15015the other argument as-if a call to @ref{INT} occurred.
15016
15017@item @emph{Example}:
15018@smallexample
15019PROGRAM test_xor
15020  LOGICAL :: T = .TRUE., F = .FALSE.
15021  INTEGER :: a, b
15022  DATA a / Z'F' /, b / Z'3' /
15023
15024  WRITE (*,*) XOR(T, T), XOR(T, F), XOR(F, T), XOR(F, F)
15025  WRITE (*,*) XOR(a, b)
15026END PROGRAM
15027@end smallexample
15028
15029@item @emph{See also}:
15030Fortran 95 elemental function: @gol
15031@ref{IEOR}
15032@end table
15033
15034
15035
15036@node Intrinsic Modules
15037@chapter Intrinsic Modules
15038@cindex intrinsic Modules
15039
15040@menu
15041* ISO_FORTRAN_ENV::
15042* ISO_C_BINDING::
15043* IEEE modules::
15044* OpenMP Modules OMP_LIB and OMP_LIB_KINDS::
15045* OpenACC Module OPENACC::
15046@end menu
15047
15048@node ISO_FORTRAN_ENV
15049@section @code{ISO_FORTRAN_ENV}
15050@table @asis
15051@item @emph{Standard}:
15052Fortran 2003 and later, except when otherwise noted
15053@end table
15054
15055The @code{ISO_FORTRAN_ENV} module provides the following scalar default-integer
15056named constants:
15057
15058@table @asis
15059@item @code{ATOMIC_INT_KIND}:
15060Default-kind integer constant to be used as kind parameter when defining
15061integer variables used in atomic operations. (Fortran 2008 or later.)
15062
15063@item @code{ATOMIC_LOGICAL_KIND}:
15064Default-kind integer constant to be used as kind parameter when defining
15065logical variables used in atomic operations. (Fortran 2008 or later.)
15066
15067@item @code{CHARACTER_KINDS}:
15068Default-kind integer constant array of rank one containing the supported kind
15069parameters of the @code{CHARACTER} type. (Fortran 2008 or later.)
15070
15071@item @code{CHARACTER_STORAGE_SIZE}:
15072Size in bits of the character storage unit.
15073
15074@item @code{ERROR_UNIT}:
15075Identifies the preconnected unit used for error reporting.
15076
15077@item @code{FILE_STORAGE_SIZE}:
15078Size in bits of the file-storage unit.
15079
15080@item @code{INPUT_UNIT}:
15081Identifies the preconnected unit identified by the asterisk
15082(@code{*}) in @code{READ} statement.
15083
15084@item @code{INT8}, @code{INT16}, @code{INT32}, @code{INT64}:
15085Kind type parameters to specify an INTEGER type with a storage
15086size of 16, 32, and 64 bits. It is negative if a target platform
15087does not support the particular kind. (Fortran 2008 or later.)
15088
15089@item @code{INTEGER_KINDS}:
15090Default-kind integer constant array of rank one containing the supported kind
15091parameters of the @code{INTEGER} type. (Fortran 2008 or later.)
15092
15093@item @code{IOSTAT_END}:
15094The value assigned to the variable passed to the @code{IOSTAT=} specifier of
15095an input/output statement if an end-of-file condition occurred.
15096
15097@item @code{IOSTAT_EOR}:
15098The value assigned to the variable passed to the @code{IOSTAT=} specifier of
15099an input/output statement if an end-of-record condition occurred.
15100
15101@item @code{IOSTAT_INQUIRE_INTERNAL_UNIT}:
15102Scalar default-integer constant, used by @code{INQUIRE} for the
15103@code{IOSTAT=} specifier to denote an that a unit number identifies an
15104internal unit. (Fortran 2008 or later.)
15105
15106@item @code{NUMERIC_STORAGE_SIZE}:
15107The size in bits of the numeric storage unit.
15108
15109@item @code{LOGICAL_KINDS}:
15110Default-kind integer constant array of rank one containing the supported kind
15111parameters of the @code{LOGICAL} type. (Fortran 2008 or later.)
15112
15113@item @code{OUTPUT_UNIT}:
15114Identifies the preconnected unit identified by the asterisk
15115(@code{*}) in @code{WRITE} statement.
15116
15117@item @code{REAL32}, @code{REAL64}, @code{REAL128}:
15118Kind type parameters to specify a REAL type with a storage
15119size of 32, 64, and 128 bits. It is negative if a target platform
15120does not support the particular kind. (Fortran 2008 or later.)
15121
15122@item @code{REAL_KINDS}:
15123Default-kind integer constant array of rank one containing the supported kind
15124parameters of the @code{REAL} type. (Fortran 2008 or later.)
15125
15126@item @code{STAT_LOCKED}:
15127Scalar default-integer constant used as STAT= return value by @code{LOCK} to
15128denote that the lock variable is locked by the executing image. (Fortran 2008
15129or later.)
15130
15131@item @code{STAT_LOCKED_OTHER_IMAGE}:
15132Scalar default-integer constant used as STAT= return value by @code{UNLOCK} to
15133denote that the lock variable is locked by another image. (Fortran 2008 or
15134later.)
15135
15136@item @code{STAT_STOPPED_IMAGE}:
15137Positive, scalar default-integer constant used as STAT= return value if the
15138argument in the statement requires synchronisation with an image, which has
15139initiated the termination of the execution. (Fortran 2008 or later.)
15140
15141@item @code{STAT_FAILED_IMAGE}:
15142Positive, scalar default-integer constant used as STAT= return value if the
15143argument in the statement requires communication with an image, which has
15144is in the failed state. (TS 18508 or later.)
15145
15146@item @code{STAT_UNLOCKED}:
15147Scalar default-integer constant used as STAT= return value by @code{UNLOCK} to
15148denote that the lock variable is unlocked. (Fortran 2008 or later.)
15149@end table
15150
15151The module provides the following derived type:
15152
15153@table @asis
15154@item @code{LOCK_TYPE}:
15155Derived type with private components to be use with the @code{LOCK} and
15156@code{UNLOCK} statement. A variable of its type has to be always declared
15157as coarray and may not appear in a variable-definition context.
15158(Fortran 2008 or later.)
15159@end table
15160
15161The module also provides the following intrinsic procedures:
15162@ref{COMPILER_OPTIONS} and @ref{COMPILER_VERSION}.
15163
15164
15165
15166@node ISO_C_BINDING
15167@section @code{ISO_C_BINDING}
15168@table @asis
15169@item @emph{Standard}:
15170Fortran 2003 and later, GNU extensions
15171@end table
15172
15173The following intrinsic procedures are provided by the module; their
15174definition can be found in the section Intrinsic Procedures of this
15175manual.
15176
15177@table @asis
15178@item @code{C_ASSOCIATED}
15179@item @code{C_F_POINTER}
15180@item @code{C_F_PROCPOINTER}
15181@item @code{C_FUNLOC}
15182@item @code{C_LOC}
15183@item @code{C_SIZEOF}
15184@end table
15185@c TODO: Vertical spacing between C_FUNLOC and C_LOC wrong in PDF,
15186@c don't really know why.
15187
15188The @code{ISO_C_BINDING} module provides the following named constants of
15189type default integer, which can be used as KIND type parameters.
15190
15191In addition to the integer named constants required by the Fortran 2003
15192standard and @code{C_PTRDIFF_T} of TS 29113, GNU Fortran provides as an
15193extension named constants for the 128-bit integer types supported by the
15194C compiler: @code{C_INT128_T, C_INT_LEAST128_T, C_INT_FAST128_T}.
15195Furthermore, if @code{__float128} is supported in C, the named constants
15196@code{C_FLOAT128, C_FLOAT128_COMPLEX} are defined.
15197
15198@multitable @columnfractions .15 .35 .35 .35
15199@item Fortran Type  @tab Named constant         @tab C type                                @tab Extension
15200@item @code{INTEGER}@tab @code{C_INT}           @tab @code{int}
15201@item @code{INTEGER}@tab @code{C_SHORT}         @tab @code{short int}
15202@item @code{INTEGER}@tab @code{C_LONG}          @tab @code{long int}
15203@item @code{INTEGER}@tab @code{C_LONG_LONG}     @tab @code{long long int}
15204@item @code{INTEGER}@tab @code{C_SIGNED_CHAR}   @tab @code{signed char}/@code{unsigned char}
15205@item @code{INTEGER}@tab @code{C_SIZE_T}        @tab @code{size_t}
15206@item @code{INTEGER}@tab @code{C_INT8_T}        @tab @code{int8_t}
15207@item @code{INTEGER}@tab @code{C_INT16_T}       @tab @code{int16_t}
15208@item @code{INTEGER}@tab @code{C_INT32_T}       @tab @code{int32_t}
15209@item @code{INTEGER}@tab @code{C_INT64_T}       @tab @code{int64_t}
15210@item @code{INTEGER}@tab @code{C_INT128_T}      @tab @code{int128_t}                      @tab Ext.
15211@item @code{INTEGER}@tab @code{C_INT_LEAST8_T}  @tab @code{int_least8_t}
15212@item @code{INTEGER}@tab @code{C_INT_LEAST16_T} @tab @code{int_least16_t}
15213@item @code{INTEGER}@tab @code{C_INT_LEAST32_T} @tab @code{int_least32_t}
15214@item @code{INTEGER}@tab @code{C_INT_LEAST64_T} @tab @code{int_least64_t}
15215@item @code{INTEGER}@tab @code{C_INT_LEAST128_T}@tab @code{int_least128_t}                @tab Ext.
15216@item @code{INTEGER}@tab @code{C_INT_FAST8_T}   @tab @code{int_fast8_t}
15217@item @code{INTEGER}@tab @code{C_INT_FAST16_T}  @tab @code{int_fast16_t}
15218@item @code{INTEGER}@tab @code{C_INT_FAST32_T}  @tab @code{int_fast32_t}
15219@item @code{INTEGER}@tab @code{C_INT_FAST64_T}  @tab @code{int_fast64_t}
15220@item @code{INTEGER}@tab @code{C_INT_FAST128_T} @tab @code{int_fast128_t}                 @tab Ext.
15221@item @code{INTEGER}@tab @code{C_INTMAX_T}      @tab @code{intmax_t}
15222@item @code{INTEGER}@tab @code{C_INTPTR_T}      @tab @code{intptr_t}
15223@item @code{INTEGER}@tab @code{C_PTRDIFF_T}     @tab @code{ptrdiff_t}                     @tab TS 29113
15224@item @code{REAL}   @tab @code{C_FLOAT}         @tab @code{float}
15225@item @code{REAL}   @tab @code{C_DOUBLE}        @tab @code{double}
15226@item @code{REAL}   @tab @code{C_LONG_DOUBLE}   @tab @code{long double}
15227@item @code{REAL}   @tab @code{C_FLOAT128}      @tab @code{__float128}                    @tab Ext.
15228@item @code{COMPLEX}@tab @code{C_FLOAT_COMPLEX} @tab @code{float _Complex}
15229@item @code{COMPLEX}@tab @code{C_DOUBLE_COMPLEX}@tab @code{double _Complex}
15230@item @code{COMPLEX}@tab @code{C_LONG_DOUBLE_COMPLEX}@tab @code{long double _Complex}
15231@item @code{REAL}   @tab @code{C_FLOAT128_COMPLEX}   @tab @code{__float128 _Complex}      @tab Ext.
15232@item @code{LOGICAL}@tab @code{C_BOOL}          @tab @code{_Bool}
15233@item @code{CHARACTER}@tab @code{C_CHAR}        @tab @code{char}
15234@end multitable
15235
15236Additionally, the following parameters of type @code{CHARACTER(KIND=C_CHAR)}
15237are defined.
15238
15239@multitable @columnfractions .20 .45 .15
15240@item Name                     @tab C definition    @tab Value
15241@item @code{C_NULL_CHAR}       @tab null character  @tab @code{'\0'}
15242@item @code{C_ALERT}           @tab alert           @tab @code{'\a'}
15243@item @code{C_BACKSPACE}       @tab backspace       @tab @code{'\b'}
15244@item @code{C_FORM_FEED}       @tab form feed       @tab @code{'\f'}
15245@item @code{C_NEW_LINE}        @tab new line        @tab @code{'\n'}
15246@item @code{C_CARRIAGE_RETURN} @tab carriage return @tab @code{'\r'}
15247@item @code{C_HORIZONTAL_TAB}  @tab horizontal tab  @tab @code{'\t'}
15248@item @code{C_VERTICAL_TAB}    @tab vertical tab    @tab @code{'\v'}
15249@end multitable
15250
15251Moreover, the following two named constants are defined:
15252
15253@multitable @columnfractions .20 .80
15254@item Name                 @tab Type
15255@item @code{C_NULL_PTR}    @tab @code{C_PTR}
15256@item @code{C_NULL_FUNPTR} @tab @code{C_FUNPTR}
15257@end multitable
15258
15259Both are equivalent to the value @code{NULL} in C.
15260
15261
15262
15263@node IEEE modules
15264@section IEEE modules: @code{IEEE_EXCEPTIONS}, @code{IEEE_ARITHMETIC}, and @code{IEEE_FEATURES}
15265@table @asis
15266@item @emph{Standard}:
15267Fortran 2003 and later
15268@end table
15269
15270The @code{IEEE_EXCEPTIONS}, @code{IEEE_ARITHMETIC}, and @code{IEEE_FEATURES}
15271intrinsic modules provide support for exceptions and IEEE arithmetic, as
15272defined in Fortran 2003 and later standards, and the IEC 60559:1989 standard
15273(@emph{Binary floating-point arithmetic for microprocessor systems}). These
15274modules are only provided on the following supported platforms:
15275
15276@itemize @bullet
15277@item i386 and x86_64 processors
15278@item platforms which use the GNU C Library (glibc)
15279@item platforms with support for SysV/386 routines for floating point
15280interface (including Solaris and BSDs)
15281@item platforms with the AIX OS
15282@end itemize
15283
15284For full compliance with the Fortran standards, code using the
15285@code{IEEE_EXCEPTIONS} or @code{IEEE_ARITHMETIC} modules should be compiled
15286with the following options: @code{-fno-unsafe-math-optimizations
15287-frounding-math -fsignaling-nans}.
15288
15289
15290
15291@node OpenMP Modules OMP_LIB and OMP_LIB_KINDS
15292@section OpenMP Modules @code{OMP_LIB} and @code{OMP_LIB_KINDS}
15293@table @asis
15294@item @emph{Standard}:
15295OpenMP Application Program Interface v4.5
15296@end table
15297
15298
15299The OpenMP Fortran runtime library routines are provided both in
15300a form of two Fortran 90 modules, named @code{OMP_LIB} and
15301@code{OMP_LIB_KINDS}, and in a form of a Fortran @code{include} file named
15302@file{omp_lib.h}. The procedures provided by @code{OMP_LIB} can be found
15303in the @ref{Top,,Introduction,libgomp,GNU Offloading and Multi
15304Processing Runtime Library} manual,
15305the named constants defined in the modules are listed
15306below.
15307
15308For details refer to the actual
15309@uref{http://www.openmp.org/wp-content/uploads/openmp-4.5.pdf,
15310OpenMP Application Program Interface v4.5}.
15311And for the @code{pause}-related constants to the OpenMP 5.0 specification.
15312
15313@code{OMP_LIB_KINDS} provides the following scalar default-integer
15314named constants:
15315
15316@table @asis
15317@item @code{omp_lock_kind}
15318@item @code{omp_lock_hint_kind}
15319@item @code{omp_nest_lock_kind}
15320@item @code{omp_pause_resource_kind}
15321@item @code{omp_proc_bind_kind}
15322@item @code{omp_sched_kind}
15323@end table
15324
15325@code{OMP_LIB} provides the scalar default-integer
15326named constant @code{openmp_version} with a value of the form
15327@var{yyyymm}, where @code{yyyy} is the year and @var{mm} the month
15328of the OpenMP version; for OpenMP v4.5 the value is @code{201511}.
15329
15330The following scalar integer named constants of the
15331kind @code{omp_sched_kind}:
15332
15333@table @asis
15334@item @code{omp_sched_static}
15335@item @code{omp_sched_dynamic}
15336@item @code{omp_sched_guided}
15337@item @code{omp_sched_auto}
15338@end table
15339
15340And the following scalar integer named constants of the
15341kind @code{omp_proc_bind_kind}:
15342
15343@table @asis
15344@item @code{omp_proc_bind_false}
15345@item @code{omp_proc_bind_true}
15346@item @code{omp_proc_bind_master}
15347@item @code{omp_proc_bind_close}
15348@item @code{omp_proc_bind_spread}
15349@end table
15350
15351The following scalar integer named constants are of the
15352kind @code{omp_lock_hint_kind}:
15353
15354@table @asis
15355@item @code{omp_lock_hint_none}
15356@item @code{omp_lock_hint_uncontended}
15357@item @code{omp_lock_hint_contended}
15358@item @code{omp_lock_hint_nonspeculative}
15359@item @code{omp_lock_hint_speculative}
15360@end table
15361
15362And the following two scalar integer named constants are of the
15363kind @code{omp_pause_resource_kind}:
15364
15365@table @asis
15366@item @code{omp_pause_soft}
15367@item @code{omp_pause_hard}
15368@end table
15369
15370
15371@node OpenACC Module OPENACC
15372@section OpenACC Module @code{OPENACC}
15373@table @asis
15374@item @emph{Standard}:
15375OpenACC Application Programming Interface v2.6
15376@end table
15377
15378
15379The OpenACC Fortran runtime library routines are provided both in a
15380form of a Fortran 90 module, named @code{OPENACC}, and in form of a
15381Fortran @code{include} file named @file{openacc_lib.h}.  The
15382procedures provided by @code{OPENACC} can be found in the
15383@ref{Top,,Introduction,libgomp,GNU Offloading and Multi Processing
15384Runtime Library} manual, the named constants defined in the modules
15385are listed below.
15386
15387For details refer to the actual
15388@uref{http://www.openacc.org/,
15389OpenACC Application Programming Interface v2.6}.
15390
15391@code{OPENACC} provides the scalar default-integer
15392named constant @code{openacc_version} with a value of the form
15393@var{yyyymm}, where @code{yyyy} is the year and @var{mm} the month
15394of the OpenACC version; for OpenACC v2.6 the value is @code{201711}.
15395