This appendix contains the following subsections:
This appendix describes the intrinsic functions provided with Fortran. Fortran intrinsic functions are identified by two categories of names: specific and generic. An IMPLICIT statement does not change the data type of an intrinsic function.
A generic name is the name given to a class of objects. Intrinsic functions that perform the same mathematical function, such as square root, are given a single name. For example, the generic name of the square root function is SQRT; this function has four specific names for different data types: SQRT, DSQRT, CSQRT, and ZSRT (see Table A-1). However, you can use the generic name SQRT regardless of the data type of the arguments.
An intrinsic function preceded by the letters CD is equivalent to the generic function with the same base name, except that the arguments must be of type DOUBLE COMPLEX.
Intrinsic functions starting with II are equivalent to generic functions with the same base name, except that the arguments must of type INTEGER*2. Similarly, arguments to intrinsic functions starting with JI must be type INTEGER*4: for example, IIAND, IIQINT, IIQNNT, JIQINT, JIQNNT.
When a generic name is referenced, the processor substitutes a function call for a specific name, depending on the data type of the arguments. In this way, the same name can be used for different types of arguments.
When an intrinsic function is to be used as the actual argument to another function, you must always use the specific name, never the generic name.
If a generic name is referenced, the type of the result is the same as the type of the argument, except for functions performing type conversion, nearest integer, and absolute value with a complex argument. Some intrinsic functions allow more than one argument, in which case all the arguments must be of the same type so that the function can decide which specific name function it should use.
If the specific name or generic name appears as the dummy argument of a function or subroutine, that symbolic name cannot identify an intrinsic function in that program unit.
A name in an INTRINSIC statement must be the specific name or generic name of an intrinsic function, as given in Table A-1.
Reference an intrinsic function in the form
fun (a[,a]...) |
where
| fun | is the generic or specific name of the intrinsic function. | |
| a | is an actual argument. |
The actual arguments (a) constitute the argument list and must agree in order, number, and type with the specification described in this appendix and with each other. Each argument can be any expression. The expression cannot contain an concatenation in which one or more of the operand lengths are specified with an asterisk.
A function reference can be used as a primary in an expression. The following example involves referencing an intrinsic function:
X = SQRT(B**2-4*A*C) |
The result of a function becomes undefined when its arguments are not mathematically defined or exceed the numeric range of the processor.
For most intrinsic functions, the data type of the result of the intrinsic function is the same as the arguments. If two or more arguments are required or permitted, then all arguments must be of the same type. An IMPLICIT statement does not change the data type of a specific or generic name of an intrinsic function.
If an intrinsic function name is used as an actual argument in an external procedure reference, the name must be one of the specific names and must appear in an INTRINSIC statement. However, names of intrinsic functions for type conversion, for lexical relationship, and for choosing the smallest or largest value cannot be used as actual arguments.
Table A-1 lists the available intrinsic functions. Operational conventions and restrictions (other than those already given) are listed at the end of the table.
 | Note: REAL*16 intrinsic functions are not supported. The compiler issues a warning message when the name of a REAL*16 intrinsic function is encountered; the equivalent double precision (REAL*8) function is used instead. |
Table A-1. Intrinsic Functions
Function | Number of Arguments | Generic Name | Specific Name | Type of Argument | Type of Result |
|---|---|---|---|---|---|
Conversion to INTEGER | 1 | INT[a] |
IINT JINT KINT IIDINT JIDINT KIDINT | INTEGER*1 INTEGER*1 INTEGER*1 INTEGER*2 INTEGER*2 INTEGER*4 INTEGER*4 INTEGER*8 REAL*4 REAL*4 REAL*4 REAL*8 REAL*8 REAL*8 COMPLEX*8 COMPLEX*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 COMPLEX*16 | INTEGER*2 INTEGER*4 INTEGER*8 INTEGER*4 INTEGER*8 INTEGER*4 INTEGER*8 INTEGER*8 INTEGER*2 INTEGER*4 INTEGER*8 INTEGER*2 INTEGER*4 INTEGER*8 INTEGER*2 INTEGER*4 INTEGER*8 INTEGER*2 INTEGER*4 INTEGER*8 |
|
| SHORT |
| INTEGER*1 INTEGER*2 INTEGER*4 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 | INTEGER*2 INTEGER*2 INTEGER*2 INTEGER*2 INTEGER*2 INTEGER*2 INTEGER*2 |
|
| LONG |
| INTEGER*1 INTEGER*2 INTEGER*4 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 | INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*4 |
| 1 | IFIX | IIFIX JIFIX KIFIX | REAL*4 REAL*4 REAL*4 | INTEGER*2 INTEGER*4 INTEGER*8 |
|
| IDINT | IIDINT JIDINT KIDINT | REAL*8 REAL*8 REAL*8 | INTEGER*2 INTEGER*4 INTEGER*8 |
Truncation |
| AINT | AINT DINT | REAL*4 REAL*8 | REAL*4 REAL*8 |
Conversion to REAL | 1 | REAL |
FLOATI FLOATJ FLOATK
SNGL
| INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 | REAL*4 REAL*4 REAL*4 REAL*4 REAL*4 REAL*4 REAL*4 REAL*4 |
| 1 | FLOAT |
FLOATI FLOATJ FLOATK | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | REAL*4 REAL*4 REAL*4 REAL*4 |
| 1 | SNGL |
FLOATI FLOATJ FLOATK REAL | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 | REAL*4 REAL*4 REAL*4 REAL*4 REAL*4 REAL*4 |
Conversion to DOUBLE PRECISION | 1 | DBLE |
DBLE
| INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 | REAL*8 REAL*8 REAL*8 REAL*8 REAL*8 REAL*8 REAL*8 REAK*8 |
|
| DFLOAT |
DFLOTI DFLOTJ DFLOTK DFLOATK | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 INTEGER*8 | REAL*8 REAL*8 REAL*8 REAL*8 REAL*8 |
Conversion to COMPLEX | 1, 2 1,2 1,2 1,2 1,2 1,2 1 1 | CMPLX |
| INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 | COMPLEX*8 COMPLEX*8 COMPLEX*8 COMPLEX*8 COMPLEX*8 COMPLEX*8 COMPLEX*8 COMPLEX*8 |
Complex Conjugate | 1 | CONJG | CONJG DCONJG | COMPLEX*8 COMPLEX*8 | COMPLEX*8 COMPLEX*16 |
Conversion to DOUBLE COMPLEX | 1, 2 1,2 1,2 1,2 1,2 1,2 1 1 | DCMPLX |
| INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 | COMPLEX*16 COMPLEX*16 COMPLEX*16 COMPLEX*16 COMPLEX*16 COMPLEX*16 COMPLEX*16 COMPLEX*16 |
Conversion to character | 1 |
| CHAR | LOGICAL*1 INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | CHARACTER CHARACTER CHARACTER CHARACTER CHARACTER |
Maximum value | 2 or more | MAX |
IMAX0 JMAX0 KMAX0 AMAX1 DMAX1 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 |
|
| MAX0 |
IMAX0 JMAX0 KMAX0 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
|
| MAX1 | IMAX1 JMAX1 KMAX1 | REAL*4 REAL*4 REAL*4 | INTEGER*2 INTEGER*4 INTEGER*8 |
|
| AMAX0 |
AIMAX0 AJMAX0 AKMAX0 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | REAL*4 REAL*4 REAL*4 REAL*4 |
Minimum value | 2 or more | MIN |
IMIN0 JMIN0 KMIN0 AMIN1 DMIN1 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 |
|
| MIN0 |
IMIN0 JMIN0 KMIN0 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
|
| MIN1 | IMIN1 JMIN1 KMIN1 | REAL*4 REAL*4 REAL*4 | INTEGER*2 INTEGER*4 INTEGER*8 |
|
| AMIN0 |
AIMIN0 AJMIN0 AKMIN0 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | REAL*4 REAL*4 REAL*4 REAL*4 |
Nearest integer | 1 | NINT[b] | ININT JNINT KNINT IIDNNT JIDNNT KIDNNT | REAL*4 REAL*4 REAL*4 REAL*8 REAL*8 REAL*8 | INTEGER*2 INTEGER*4 INTEGER*8 INTEGER*2 INTEGER*4 INTEGER*8 |
|
| ANINT | ANINT DNINT | REAL*4 REAL*8 | REAL*4 REAL*8 |
|
| IDNINT | IIDNNT JIDNNT KIDNNT | REAL*8 REAL*8 REAL*8 | INTEGER*2 INTEGER*4 INTEGER*8 |
Zero-Extend functions | 1 | ZEXT | IZEXT
JZEXT
KZEXT | LOGICAL*1 LOGICAL*2 INTEGER*1 INTEGER*2 LOGICAL*1 LOGICAL*2 LOGICAL*4 INTEGER*1 INTEGER*2 INTEGER*4 LOGICAL*1 LOGICAL*2 LOGICAL*4 LOGICAL*8 INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*2 INTEGER*2 INTEGER*2 INTEGER*2 INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*4 INTEGER*8 INTEGER*8 INTEGER*8 INTEGER*8 INTEGER*8 INTEGER*8 INTEGER*8 INTEGER*8 |
Absolute value | 1 | ABS |
IIABS JIABS KIABS ABS DABS CABS CDABS ZABS | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 |
|
| IABS[c] |
IIABS JIABS KIABS | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Remaindering | 2 | MOD[d] |
IMOD JMOD KMOD AMOD DMOD | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 |
Transfer of sign | 2 | SIGN |
IISIGN JISIGN KISIGN SIGN DSIGN | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 |
|
| ISIGN[e] |
IISIGN JISIGN KISIGN | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Positive difference | 2 | DIM |
IIDIM JIDIM KIDIM DIM DDIM | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 REAL*4 REAL*8 |
|
| IDIM |
IIDIM JIDIM KIDIM | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
DOUBLE PRECISION product of REALs | 2 |
| DPROD | REAL*4 | REAL*4 |
Length of character entry | 1 |
| LEN | CHARACTER | INTEGER*4 |
Index of a substring | 2 |
| INDEX[f] | CHARACTER | INTEGER*4 |
Character (ASCII value of 1-byte character argument) | 1 |
| ICHAR | CHARACTER CHARACTER CHARACTER | INTEGER*2 INTEGER*4 INTEGER*8 |
Logically greater than or equal | 2 |
| LGE | CHARACTER | LOGICAL*4 |
Logically greater than | 2 |
| LGT | CHARACTER | LOGICAL*4 |
Logically less than or equal | 2 |
| LLE | CHARACTER | LOGICAL*4 |
Logically less than | 2 |
| LLT[g] | CHARACTER | LOGICAL*4 |
Imaginary part of complex number | 1 | IMAG | AIMAG DIMAG | COMPLEX*8 COMPLEX*16 | REAL*4 REAL*8 |
Real part of complex number | 1 |
| REAL DREAL | COMPLEX*8 COMPLEX*16 | REAL*4 REAL*8 |
Square root | 1 | SQRT | SQRT[h] DSQRT CSQRT CDSQRT ZSQRT | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 |
Exponential | 1 | EXP | EXP DEXP CEXP CDEXP ZEXP | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 |
Natural logarithm | 1 | LOG | ALOG[i] DLOG CLOG CDLOG ZLOG | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 |
Common logarithm | 1 | LOG10 | ALOG10 DLOG10 | REAL*4 REAL*8 | REAL*4 REAL*8 |
Sine | 1 | SIN | SIN DSIN CSIN CDSIN ZSIN | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 |
Sine (degree) | 1 | SIND[j] | SIND DSIND | REAL*4 REAL*8 | REAL*4 REAL*8 |
Cosine | 1 | COS | COS DCOS CCOS CDCOS ZCOS | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 | REAL*4 REAL*8 COMPLEX*8 COMPLEX*16 COMPLEX*16 |
Cosine (degree) | 1 | COSD | COSD DCOSD | REAL*4 REAL*8 | REAL*4 REAL*8 |
Tangent | 1 | TAN | TAN DTAN | REAL*4 REAL*8 | REAL*4 REAL*8 |
Tangent (degree) | 1 | TAND | TAND DTAND | REAL*4 REAL*8 | REAL*4 REAL*8 |
Arcsine | 1 | ASIN DASIN | REAL*4 REAL*8 | REAL*4 REAL*8 | |
Arcsine (degree) | 1 | ASIND[n] | ASIND DASIND | REAL*4 REAL*8 | REAL*4 REAL*8 |
Arccosine | 1 | ACOS | ACOS DACOS | REAL*4 REAL*8 | REAL*4 REAL*8 |
Arccsine (degree) | 1 | ACOSD | ACOSD DACOSD | REAL*4 REAL*8 | REAL*4 REAL*8 |
Arctangent | 1 | ATAN[o] | ATAN DATAN | REAL*4 REAL*8 | REAL*4 REAL*8 |
Arctangent (degree) | 1 | ATAND[p] | ATAND DATAND | REAL*4 REAL*8 | REAL*4 REAL*8 |
Arctangent | 2 | ATAN2 DATAN2 | REAL*4 REAL*8 | REAL*4 REAL*8 | |
Arctangent (degree) | 2 | ATAN2D | ATAN2D DATAN2D | REAL*4 REAL*8 | REAL*4 REAL*8 |
Hyperbolic sine | 1 | SINH | SINH DSINH | REAL*4 REAL*8 | REAL*4 REAL*8 |
Hyperbolic cosine | 1 | COSH | COSH DCOSH | REAL*4 REAL*8 | REAL*4 REAL*8 |
Hyperbolic tangent | 1 | TANH | TANH DTANH | REAL*4 REAL*8 | REAL*4 REAL*8 |
Bitwise AND | 2 | IAND1 |
IIAND JIAND KIAND | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bitwise inclusive OR | 2 | IOR1 |
IIOR JIOR KIOR | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bitwise complement | 1 | NOT1 | INOT JNOT KNOT | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bitwise exclusive OR | 2 | IEOR1 |
IIEOR JIEOR KIEOR | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bitwise logical shift | 2 | ISHFT |
IISHFT JISHFT KISHFT | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bitwise circular shift | 2 | ISHFTC |
IISHFTC JISHFTC KISHFTC | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bit extraction | 3 | IBITS |
IIBITS JIBITS KIBITS | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bit set | 2 | IBITSET |
IIBSET JIBSET KIBSET | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
Bit test | 2 | BTEST |
BITEST BJTEST BKTEST | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | LOGICAL*4 LOGICAL*2 LOGICAL*4 LOGICAL*8 |
Bit clear | 2 | IBCLR |
IIBCLR JIBCLR KIBCLR | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 | INTEGER*1 INTEGER*2 INTEGER*4 INTEGER*8 |
[a] INT and IFIX return type INTEGER*2 if the -i2 compile option is in effect; otherwise, the result type is [b] When NINT or IDNINT is specified as an argument in a subroutine call or function reference, the compiler
supplies either an INTEGER*2 or an INTEGER*4 function depending on the -i2 command line option [c] The IABS, ISIGN, IDIM, and integer MOD intrinsics accept either INTEGER*2 arguments or INTEGER*4 arguments, and the result is the same type. [d] The result for MOD, AMOD, and DMOD is undefined when the value of the second argument is zero. [e] If the value of the first argument of ISIGN, SIGN, or DSIGN is zero, the result is zero. [f] The result of INDEX is an integer value indicating the position in the first argument of the first substring
which is identical to the second argument. The result of INDEX('ABCDEF', 'CD'), for example, would be
3. If no substring of the first argument matches the second argument, the result is zero. INDEX and ICHAR
return the result type INTEGER*2 if the -i2 compile option is in effect; otherwise, the result type is [g] The character relational intrinsics (LLT, LGT, LEE, and LGE) return result type LOGICAL*2 if the $log2 (see Chapter 11) compile option is in effect; otherwise, the result type is LOGICAL*4. [h] The value of the argument of SQRT and DSQRT must be greater than or equal to zero. The result of CSQRT
is the principal value with the real part greater than or equal to zero. When the real part is zero, the [i] The argument of ALOG and DLOG must be greater than zero. The argument of CLOG must not be (0.,0.). The range of the imaginary part of the result of CLOG is: -p <imaginary part <p. [j] The argument for SIND, COSD, or TAND must be in degrees and is treated as modulo 360. [k] The absolute value of the arguments of ASIN, DASIN, ASIND, DASIND, ACOS, DACOS, ACOSD, and DACSOD must be less than or equal to 1. [l] The range of the result for ASIN and DASIN is -π/2 <result < π/2; the range of the result for DASIN is 0 < result < π; and the range of the result of ACOS and DACOS is less than or equal to one. [m] The result of ASIN, DASIN, ACOS, and DACOS is in radians. [n] The result of ASIND, DASIND, ACOS, DACOSD is in degrees. [o] The result of ATAN, DATAN, ATAN2, and DTAN2 is in radians. [p] The result of ATAND, DATAND, ATAN2D, and DATAN2D is in degrees. [q] If the value of the first argument of ATAN2 or DATAN2 is positive, the result is positive. When the value
of the first argument is zero, the result is zero if the second argumemt is positive and P if the second [r] Note 3 on this page also applies to ATAN2 and DTAN2D, except for the range of the result, which is: | |||||