libsim Versione 7.2.6
|
◆ fp_d
double precision floating point (8 byte IEEE) Definizione alla linea 259 del file kinds.F90. 259! Copyright (C) 2010 ARPA-SIM <urpsim@smr.arpa.emr.it>
260! authors:
261! Davide Cesari <dcesari@arpa.emr.it>
262! Paolo Patruno <ppatruno@arpa.emr.it>
263
264! This program is free software; you can redistribute it and/or
265! modify it under the terms of the GNU General Public License as
266! published by the Free Software Foundation; either version 2 of
267! the License, or (at your option) any later version.
268
269! This program is distributed in the hope that it will be useful,
270! but WITHOUT ANY WARRANTY; without even the implied warranty of
271! MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
272! GNU General Public License for more details.
273
274! You should have received a copy of the GNU General Public License
275! along with this program. If not, see <http://www.gnu.org/licenses/>.
276#include "config.h"
277!> \defgroup base Libsim package, base library.
278!! The libsim base library defines modules and classes of general
279!! use for scientifical applications in Fortran 90. In order to
280!! compile and link programs using this library, you have to insert
281!! the required \c USE statements in the program units involved,
282!! specify the location of module files when compiling (tipically \c
283!! -I/usr/lib/gfortran/modules or \c -I/usr/lib64/gfortran/modules or
284!! \c -I/usr/include) and indicate the library name \c -lsim_base when
285!! linking, assuming that the library has been installed in a default
286!! location.
287
288!> Definition of constants to be used for declaring variables of a
289!! desired type. This module defines constants that can be portably
290!! used when declaring variables (through the \c KIND attribute) and
291!! when defining constants (through the underscore character \c _ ) in
292!! order to be sure that the desired type is used.
293!!
294!! There is a subtle difference between platform-default single and
295!! double precision real, obtained by declaring a variable of type \c
296!! REAL or \c DOUBLE \c PRECISION respectively, and single and double
297!! precision IEEE (4 and 8 bytes respectively) which are the standard
298!! IEEE data types and which are declared through \c REAL(kind=fp_s)
299!! and \c REAL(kind=fp_d) respectively: these two pairs of types
300!! usually coincide, but it may be not the case on some platforms, so
301!! you should choose one or the other approach depending on situation.
302!!
303!! Example of typical use:
304!! \code
305!! USE kinds
306!! ...
307!! INTEGER(kind=int_b) :: ab, bb
308!! REAL(kind=fp_d) :: dd
309!!
310!! ab = 13_int_b
311!! dd = REAL(ab, kind=fp_d)
312!! ...
313!! \endcode
314!! \ingroup base
316IMPLICIT NONE
317
318INTEGER, PARAMETER :: int_b = selected_int_kind(1) !< 1-byte integer (byte)
319INTEGER, PARAMETER :: int_s = selected_int_kind(4) !< 2-byte integer (short)
320INTEGER, PARAMETER :: int_l = selected_int_kind(8) !< 4-byte integer (long)
321INTEGER, PARAMETER, PRIVATE :: &
322 int_ll_t = selected_int_kind(16)
323!> 8-byte integer (long long) if supported, otherwise 4-byte integer
324INTEGER, PARAMETER :: int_ll = &
325 ( ( ( 1 + sign( 1, int_ll_t ) ) / 2 ) * int_ll_t ) + &
326 ( ( ( 1 - sign( 1, int_ll_t ) ) / 2 ) * int_l )
327
328INTEGER, PARAMETER :: fp_s = selected_real_kind(6) !< single precision floating point (4 byte IEEE)
329INTEGER, PARAMETER :: fp_d = selected_real_kind(15) !< double precision floating point (8 byte IEEE)
330INTEGER, PARAMETER, PRIVATE :: fp_q_t = selected_real_kind(20)
331!> quad precision floating point (16 byte IEEE) if supported, otherwise double precision floating point
332INTEGER, PARAMETER :: fp_q = &
333 ( ( ( 1 + sign( 1, fp_q_t ) ) / 2 ) * fp_q_t ) + &
334 ( ( ( 1 - sign( 1, fp_q_t ) ) / 2 ) * fp_d )
335
336INTEGER, PARAMETER :: ptr_c = sizeof_ptr_c !< kind for an integer having the same size of a C pointer
337
Definition of constants to be used for declaring variables of a desired type. Definition kinds.F90:245 |