1 \input texinfo @c -*-texinfo-*-
3 @setfilename gfortran.info
4 @set copyrights-gfortran 1999-2005
6 @include gcc-common.texi
8 @settitle The GNU Fortran 95 Compiler
10 @c Create a separate index for command line options
12 @c Merge the standard indexes into a single one.
21 @c Use with @@smallbook.
23 @c %** start of document
25 @c Cause even numbered pages to be printed on the left hand side of
26 @c the page and odd numbered pages to be printed on the right hand
27 @c side of the page. Using this, you can print on both sides of a
28 @c sheet of paper and have the text on the same part of the sheet.
30 @c The text on right hand pages is pushed towards the right hand
31 @c margin and the text on left hand pages is pushed toward the left
33 @c (To provide the reverse effect, set bindingoffset to -0.75in.)
36 @c \global\bindingoffset=0.75in
37 @c \global\normaloffset =0.75in
41 Copyright @copyright{} @value{copyrights-gfortran} Free Software Foundation, Inc.
43 Permission is granted to copy, distribute and/or modify this document
44 under the terms of the GNU Free Documentation License, Version 1.1 or
45 any later version published by the Free Software Foundation; with the
46 Invariant Sections being ``GNU General Public License'' and ``Funding
47 Free Software'', the Front-Cover
48 texts being (a) (see below), and with the Back-Cover Texts being (b)
49 (see below). A copy of the license is included in the section entitled
50 ``GNU Free Documentation License''.
52 (a) The FSF's Front-Cover Text is:
56 (b) The FSF's Back-Cover Text is:
58 You have freedom to copy and modify this GNU Manual, like GNU
59 software. Copies published by the Free Software Foundation raise
60 funds for GNU development.
64 @dircategory Programming
66 * gfortran: (gfortran). The GNU Fortran 95 Compiler.
68 This file documents the use and the internals of
69 the GNU Fortran 95 compiler, (@command{gfortran}).
71 Published by the Free Software Foundation
72 51 Franklin Street, Fifth Floor
73 Boston, MA 02110-1301 USA
79 @setchapternewpage odd
81 @title Using GNU Fortran 95
83 @center The gfortran team
85 @vskip 0pt plus 1filll
86 For the @value{version-GCC} Version*
88 Published by the Free Software Foundation @*
89 51 Franklin Street, Fifth Floor@*
90 Boston, MA 02110-1301, USA@*
91 @c Last printed ??ber, 19??.@*
92 @c Printed copies are available for $? each.@*
105 This manual documents the use of @command{gfortran},
106 the GNU Fortran 95 compiler. You can find in this manual how to invoke
107 @command{gfortran}, as well as its features and incompatibilities.
110 @emph{Warning:} This document, and the compiler it describes, are still
111 under development. While efforts are made to keep it up-to-date, it might
112 not accurately reflect the status of the most recent @command{gfortran}.
116 @comment When you add a new menu item, please keep the right hand
117 @comment aligned to the same column. Do not use tabs. This provides
118 @comment better formatting.
121 * Getting Started:: What you should know about @command{gfortran}.
122 * GFORTRAN and GCC:: You can compile Fortran, C, or other programs.
123 * GFORTRAN and G77:: Why we chose to start from scratch.
124 * Invoking GFORTRAN:: Command options supported by @command{gfortran}.
125 * Project Status:: Status of @command{gfortran}, roadmap, proposed extensions.
126 * Contributing:: How you can help.
127 * Standards:: Standards supported by @command{gfortran}
128 * Extensions:: Language extensions implemented by @command{gfortran}
129 * Intrinsic Procedures:: Intrinsic procedures supported by @command{gfortran}
130 * Copying:: GNU General Public License says
131 how you can copy and share GNU Fortran.
132 * GNU Free Documentation License::
133 How you can copy and share this manual.
134 * Funding:: How to help assure continued work for free software.
135 * Index:: Index of this documentation.
140 @c ---------------------------------------------------------------------
142 @c ---------------------------------------------------------------------
144 @node Getting Started
145 @chapter Getting Started
147 Gfortran is the GNU Fortran 95 compiler front end,
148 designed initially as a free replacement for,
149 or alternative to, the unix @command{f95} command;
150 @command{gfortran} is the command you'll use to invoke the compiler.
152 Gfortran is still in an early state of development.
153 @command{gfortran} can generate code for most constructs and expressions,
154 but much work remains to be done.
156 When @command{gfortran} is finished,
157 it will do everything you expect from any decent compiler:
161 Read a user's program,
162 stored in a file and containing instructions written
163 in Fortran 77, Fortran 90 or Fortran 95.
164 This file contains @dfn{source code}.
167 Translate the user's program into instructions a computer
168 can carry out more quickly than it takes to translate the
169 instructions in the first
170 place. The result after compilation of a program is
172 code designed to be efficiently translated and processed
173 by a machine such as your computer.
174 Humans usually aren't as good writing machine code
175 as they are at writing Fortran (or C++, Ada, or Java),
176 because is easy to make tiny mistakes writing machine code.
179 Provide the user with information about the reasons why
180 the compiler is unable to create a binary from the source code.
181 Usually this will be the case if the source code is flawed.
182 When writing Fortran, it is easy to make big mistakes.
183 The Fortran 90 requires that the compiler can point out
184 mistakes to the user.
185 An incorrect usage of the language causes an @dfn{error message}.
187 The compiler will also attempt to diagnose cases where the
188 user's program contains a correct usage of the language,
189 but instructs the computer to do something questionable.
190 This kind of diagnostics message is called a @dfn{warning message}.
193 Provide optional information about the translation passes
194 from the source code to machine code.
195 This can help a user of the compiler to find the cause of
196 certain bugs which may not be obvious in the source code,
197 but may be more easily found at a lower level compiler output.
198 It also helps developers to find bugs in the compiler itself.
201 Provide information in the generated machine code that can
202 make it easier to find bugs in the program (using a debugging tool,
203 called a @dfn{debugger}, such as the GNU Debugger @command{gdb}).
206 Locate and gather machine code already generated to
207 perform actions requested by statements in the user's program.
208 This machine code is organized into @dfn{modules} and is located
209 and @dfn{linked} to the user program.
212 Gfortran consists of several components:
216 A version of the @command{gcc} command
217 (which also might be installed as the system's @command{cc} command)
218 that also understands and accepts Fortran source code.
219 The @command{gcc} command is the @dfn{driver} program for
220 all the languages in the GNU Compiler Collection (GCC);
222 you can compile the source code of any language for
223 which a front end is available in GCC.
226 The @command{gfortran} command itself,
227 which also might be installed as the
228 system's @command{f95} command.
229 @command{gfortran} is just another driver program,
230 but specifically for the Fortran 95 compiler only.
231 The difference with @command{gcc} is that @command{gfortran}
232 will automatically link the correct libraries to your program.
235 A collection of run-time libraries.
236 These libraries contain the machine code needed to support
237 capabilities of the Fortran language that are not directly
238 provided by the machine code generated by the
239 @command{gfortran} compilation phase,
240 such as intrinsic functions and subroutines,
241 and routines for interaction with files and the operating system.
242 @c and mechanisms to spawn,
243 @c unleash and pause threads in parallelized code.
246 The Fortran compiler itself, (@command{f951}).
247 This is the gfortran parser and code generator,
248 linked to and interfaced with the GCC backend library.
249 @command{f951} ``translates'' the source code to
250 assembler code. You would typically not use this
252 instead, the @command{gcc} or @command{gfortran} driver
253 programs will call it for you.
258 @c ---------------------------------------------------------------------
260 @c ---------------------------------------------------------------------
262 @node GFORTRAN and GCC
263 @chapter GFORTRAN and GCC
264 @cindex GNU Compiler Collection
266 GCC used to be the GNU ``C'' Compiler,
267 but is now known as the @dfn{GNU Compiler Collection}.
268 GCC provides the GNU system with a very versatile
269 compiler middle end (shared optimization passes),
270 and back ends (code generators) for many different
271 computer architectures and operating systems.
272 The code of the middle end and back end are shared by all
273 compiler front ends that are in the GNU Compiler Collection.
275 A GCC front end is essentially a source code parser
276 and an intermediate code generator. The code generator translates the
277 semantics of the source code into a language independent form called
280 The parser takes a source file written in a
281 particular computer language, reads and parses it,
282 and tries to make sure that the source code conforms to
284 Once the correctness of a program has been established,
285 the compiler will build a data structure known as the
286 @dfn{Abstract Syntax tree},
287 or just @dfn{AST} or ``tree'' for short.
288 This data structure represents the whole program
289 or a subroutine or a function.
290 The ``tree'' is passed to the GCC middle end,
291 which will perform optimization passes on it. The optimized AST is then
292 handed off too the back end which assembles the program unit.
294 Different phases in this translation process can be,
295 and in fact @emph{are} merged in many compiler front ends.
296 GNU Fortran 95 has a strict separation between the
297 parser and code generator.
299 The goal of the gfortran project is to build a new front end for GCC.
300 Specifically, a Fortran 95 front end.
301 In a non-gfortran installation,
302 @command{gcc} will not be able to compile Fortran 95 source code
303 (only the ``C'' front end has to be compiled if you want to build GCC,
304 all other languages are optional).
305 If you build GCC with gfortran, @command{gcc} will recognize
306 @file{.f/.f90/.f95} source files and accepts Fortran 95 specific
307 command line options.
311 @c ---------------------------------------------------------------------
313 @c ---------------------------------------------------------------------
315 @node GFORTRAN and G77
316 @chapter GFORTRAN and G77
320 Why do we write a compiler front end from scratch?
321 There's a fine Fortran 77 compiler in the
322 GNU Compiler Collection that accepts some features
323 of the Fortran 90 standard as extensions.
324 Why not start from there and revamp it?
326 One of the reasons is that Craig Burley, the author of G77,
327 has decided to stop working on the G77 front end.
328 On @uref{http://world.std.com/~burley/g77-why.html,
329 Craig explains the reasons for his decision to stop working on G77}
330 in one of the pages in his homepage.
331 Among the reasons is a lack of interest in improvements to
333 Users appear to be quite satisfied with @command{g77} as it is.
334 While @command{g77} is still being maintained (by Toon Moene),
335 it is unlikely that sufficient people will be willing
336 to completely rewrite the existing code.
338 But there are other reasons to start from scratch.
339 Many people, including Craig Burley,
340 no longer agreed with certain design decisions in the G77 front end.
341 Also, the interface of @command{g77} to the back end is written in
342 a style which is confusing and not up to date on recommended practice.
343 In fact, a full rewrite had already been planned for GCC 3.0.
345 When Craig decided to stop,
346 it just seemed to be a better idea to start a new project from scratch,
347 because it was expected to be easier to maintain code we
348 develop ourselves than to do a major overhaul of @command{g77} first,
349 and then build a Fortran 95 compiler out of it.
353 @c ---------------------------------------------------------------------
355 @c ---------------------------------------------------------------------
358 @chapter Project Status
361 As soon as gfortran can parse all of the statements correctly,
362 it will be in the ``larva'' state.
363 When we generate code, the ``puppa'' state.
364 When gfortran is done,
365 we'll see if it will be a beautiful butterfly,
366 or just a big bug....
368 --Andy Vaught, April 2000
371 The start of the GNU Fortran 95 project was announced on
372 the GCC homepage in March 18, 2000
373 (even though Andy had already been working on it for a while,
376 Gfortran is currently reaching the stage where is is able to compile real
377 world programs. However it is still under development and has many rough
383 * Proposed Extensions::
386 @node Compiler Status
387 @section Compiler Status
391 This is the part of gfortran which parses a source file, verifies that it
392 is valid Fortran 95, performs compile time replacement of constants
393 (PARAMETER variables) and reads and generate module files. This is
394 almost complete. Every Fortran 95 source should be accepted, and most
395 none-Fortran 95 source should be rejected. If you find a source file where
396 this is not true, please tell us. You can use the -fsyntax-only switch to
397 make gfortran quit after running the front end, effectively reducing it to
400 @item Middle end interface
401 These are the parts of gfortran that take the parse tree generated by the
402 front end and translate it to the GENERIC form required by the GCC back
403 end. Work is ongoing in these parts of gfortran, but a large part has
404 already been completed.
408 @section Library Status
410 Some intrinsic functions map directly to library functions, and in most
411 cases the name of the library function used depends on the type of the
412 arguments. For some intrinsics we generate inline code, and for others,
413 such as sin, cos and sqrt, we rely on the backend to use special
414 instructions in the floating point unit of the CPU if available, or to
415 fall back to a call to libm if these are not available.
417 Implementation of some non-elemental intrinsic functions (eg. DOT_PRODUCT,
418 AVERAGE) is not yet optimal. This is hard because we have to make decisions
419 whether to use inline code (good for small arrays as no function call
420 overhead occurs) or generate function calls (good for large arrays as it
421 allows use of hand-optimized assembly routines, SIMD instructions, etc.)
423 The IO library is in a mostly usable state. Unformatted I/O for
424 @code{REAL(KIND=10)} variables is currently not recommended.
426 Array intrinsics mostly work.
428 @node Proposed Extensions
429 @section Proposed Extensions
431 Here's a list of proposed extensions for @command{gfortran}, in no particular
432 order. Most of these are necessary to be fully compatible with
433 existing Fortran compilers, but they are not part of the official
434 J3 Fortran 95 standard.
436 @subsection Compiler extensions:
439 Flag for defining the kind number for default logicals.
442 User-specified alignment rules for structures.
444 Flag to generate @code{Makefile} info.
447 Automatically extend single precision constants to double.
450 Compile code that conserves memory by dynamically allocating common and
451 module storage either on stack or heap.
454 Flag to cause the compiler to distinguish between upper and lower case
455 names. The Fortran 95 standard does not distinguish them.
458 Compile flag to generate code for array conformance checking (suggest -CC).
461 User control of symbol names (underscores, etc).
464 Compile setting for maximum size of stack frame size before spilling
465 parts to static or heap.
468 Flag to force local variables into static space.
471 Flag to force local variables onto stack.
474 Flag to compile lines beginning with ``D''.
477 Flag to ignore lines beginning with ``D''.
480 Flag for maximum errors before ending compile.
483 Generate code to check for null pointer dereferences -- prints locus of
484 dereference instead of segfaulting. There was some discussion about this
485 option in the g95 development mailing list.
488 Allow setting the default unit number.
491 Option to initialize otherwise uninitialized integer and floating
495 Support for OpenMP directives. This also requires support from the runtime
496 library and the rest of the compiler.
499 Support for Fortran 200x. This includes several new features including
500 floating point exceptions, extended use of allocatable arrays, C
501 interoperability, Parameterizer data types and function pointers.
505 @subsection Environment Options
508 Pluggable library modules for random numbers, linear algebra.
509 LA should use BLAS calling conventions.
512 Environment variables controlling actions on arithmetic exceptions like
513 overflow, underflow, precision loss -- Generate NaN, abort, default.
517 Set precision for fp units that support it (i387).
520 Variable for setting fp rounding mode.
523 Variable to fill uninitialized variables with a user-defined bit
527 Environment variable controlling filename that is opened for that unit
531 Environment variable to clear/trash memory being freed.
534 Environment variable to control tracing of allocations and frees.
537 Environment variable to display allocated memory at normal program end.
540 Environment variable for filename for * IO-unit.
543 Environment variable for temporary file directory.
546 Environment variable forcing standard output to be line buffered (unix).
549 Variable for swapping endianness during unformatted read.
552 Variable for swapping Endianness during unformatted write.
555 @c ---------------------------------------------------------------------
557 @c ---------------------------------------------------------------------
559 @c Maybe this chapter should be merged with the 'Standards' section,
560 @c whenever that is written :-)
566 @command{gfortran} implements a number of extensions over standard
567 Fortran. This chapter contains information on their syntax and
568 meaning. There are currently two categories of @command{gfortran}
569 extensions, those that provide functionality beyond that provided
570 by any standard, and those that are supported by @command{gfortran}
571 purely for backward compatibility with legacy compilers. By default,
572 @option{-std=gnu} allows the compiler to accept both types of
573 extensions, but to warn about the use of the latter. Specifying
574 either @option{-std=f95} or @option{-std=f2003} disables both types
575 of extensions, and @option{-std=legacy} allows both without warning.
578 * Old-style kind specifications::
579 * Old-style variable initialization::
580 * Extensions to namelist::
581 * X format descriptor::
582 * Commas in FORMAT specifications::
584 * Hexadecimal constants::
585 * Real array indices::
587 * Implicitly interconvert LOGICAL and INTEGER::
588 * Hollerith constants support::
592 @node Old-style kind specifications
593 @section Old-style kind specifications
594 @cindex Kind specifications
596 @command{gfortran} allows old-style kind specifications in
597 declarations. These look like:
601 where @code{TYPESPEC} is a basic type, and where @code{k} is a valid kind
602 number for that type. The statement then declares @code{x}, @code{y}
603 and @code{z} to be of type @code{TYPESPEC} with kind @code{k}. In
604 other words, it is equivalent to the standard conforming declaration
609 @node Old-style variable initialization
610 @section Old-style variable initialization
611 @cindex Initialization
613 @command{gfortran} allows old-style initialization of variables of the
617 REAL*8 x(2,2) /3*0.,1./
619 These are only allowed in declarations without double colons
620 (@code{::}), as these were introduced in Fortran 90 which also
621 introduced a new syntax for variable initializations. The syntax for
622 the individual initializers is as for the @code{DATA} statement, but
623 unlike in a @code{DATA} statement, an initializer only applies to the
624 variable immediately preceding. In other words, something like
625 @code{INTEGER I,J/2,3/} is not valid.
627 Examples of standard conforming code equivalent to the above example, are:
630 INTEGER(4) :: i = 1, j = 2
631 REAL(8) :: x(2,2) = RESHAPE((/0.,0.,0.,1./),SHAPE(x))
634 DOUBLE PRECISION x(2,2)
635 DATA i,j,x /1,2,3*0.,1./
638 @node Extensions to namelist
639 @section Extensions to namelist
642 @command{gfortran} fully supports the Fortran 95 standard for namelist I/O
643 including array qualifiers, substrings and fully qualified derived types.
644 The output from a namelist write is compatible with namelist read. The
645 output has all names in upper case and indentation to column 1 after the
646 namelist name. Two extensions are permitted:
648 Old-style use of $ instead of &
651 X(:)%Y(2) = 1.0 2.0 3.0
656 It should be noticed that the default terminator is / rather than &END.
658 Querying of the namelist when inputting from stdin. After at least
659 one space, entering ? sends to stdout the namelist name and the names of
660 the variables in the namelist:
671 Entering =? outputs the namelist to stdout, as if WRITE (*,NML = mynml)
677 X(1)%Y= 0.000000 , 1.000000 , 0.000000 ,
678 X(2)%Y= 0.000000 , 2.000000 , 0.000000 ,
679 X(3)%Y= 0.000000 , 3.000000 , 0.000000 ,
683 To aid this dialog, when input is from stdin, errors send their
684 messages to stderr and execution continues, even if IOSTAT is set.
686 PRINT namelist is permitted. This causes an error if -std=f95 is used.
689 REAL, dimension (4) :: x = (/1.0, 2.0, 3.0, 4.0/)
692 END PROGRAM test_print
695 @node X format descriptor
696 @section X format descriptor
697 @cindex X format descriptor
699 To support legacy codes, @command{gfortran} permits the count field
700 of the X edit descriptor in FORMAT statements to be omitted. When
701 omitted, the count is implicitly assumed to be one.
705 10 FORMAT (I1, X, I1)
708 @node Commas in FORMAT specifications
709 @section Commas in FORMAT specifications
710 @cindex Commas in FORMAT specifications
712 To support legacy codes, @command{gfortran} allows the comma separator
713 to be omitted immediately before and after character string edit
714 descriptors in FORMAT statements.
718 10 FORMAT ('FOO='I1' BAR='I2)
722 @section I/O item lists
723 @cindex I/O item lists
725 To support legacy codes, @command{gfortran} allows the input item list
726 of the READ statement, and the output item lists of the WRITE and PRINT
727 statements to start with a comma.
729 @node Hexadecimal constants
730 @section Hexadecimal constants
731 @cindex Hexadecimal constants
733 As a GNU extension, @command{gfortran} allows hexadecimal constants to
734 be specified using the X prefix, in addition to the standard Z prefix.
736 @node Real array indices
737 @section Real array indices
738 @cindex Real array indices
740 As a GNU extension, @command{gfortran} allows arrays to be indexed using
741 real types, whose values are implicitly converted to integers.
743 @node Unary operators
744 @section Unary operators
745 @cindex Unary operators
747 As a GNU extension, @command{gfortran} allows unary plus and unary
748 minus operators to appear as the second operand of binary arithmetic
749 operators without the need for parenthesis.
755 @node Implicitly interconvert LOGICAL and INTEGER
756 @section Implicitly interconvert LOGICAL and INTEGER
757 @cindex Implicitly interconvert LOGICAL and INTEGER
759 As a GNU extension for backwards compatibility with other compilers,
760 @command{gfortran} allows the implicit conversion of LOGICALs to INTEGERs
761 and vice versa. When converting from a LOGICAL to an INTEGER, the numeric
762 value of @code{.FALSE.} is zero, and that of @code{.TRUE.} is one. When
763 converting from INTEGER to LOGICAL, the value zero is interpreted as
764 @code{.FALSE.} and any nonzero value is interpreted as @code{.TRUE.}.
771 @node Hollerith constants support
772 @section Hollerith constants support
773 @cindex Hollerith constants
775 A Hollerith constant is a string of characters preceded by the letter @samp{H}
776 or @samp{h}, and there must be an literal, unsigned, nonzero default integer
777 constant indicating the number of characters in the string. Hollerith constants
778 are stored as byte strings, one character per byte.
780 @command{gfortran} supports Hollerith constants. They can be used as the right
781 hands in the @code{DATA} statement and @code{ASSIGN} statement, also as the
782 arguments. The left hands can be of Integer, Real, Complex and Logical type.
783 The constant will be padded or trancated to fit the size of left hand.
785 Valid Hollerith constants examples:
788 data x /16Habcdefghijklmnop, 16Hqrstuvwxyz012345/
790 x(1) = 16Habcdefghijklmnop
793 Invalid Hollerith constants examples:
796 a = 8H12345678 ! The Hollerith constant is too long. It will be truncated.
797 a = 0H ! At least one character needed.
801 @section Cray pointers
802 @cindex Cray pointers
804 Cray pointers are part of a non-standard extension that provides a
805 C-like pointer in Fortran. This is accomplished through a pair of
806 variables: an integer "pointer" that holds a memory address, and a
807 "pointee" that is used to dereference the pointer.
809 Pointer/pointee pairs are declared in statements of the form:
811 pointer ( <pointer> , <pointee> )
815 pointer ( <pointer1> , <pointee1> ), ( <pointer2> , <pointee2> ), ...
817 The pointer is an integer that is intended to hold a memory address.
818 The pointee may be an array or scalar. A pointee can be an assumed
819 size array -- that is, the last dimension may be left unspecified by
820 using a '*' in place of a value -- but a pointee cannot be an assumed
821 shape array. No space is allocated for the pointee.
823 The pointee may have its type declared before or after the pointer
824 statement, and its array specification (if any) may be declared
825 before, during, or after the pointer statement. The pointer may be
826 declared as an integer prior to the pointer statement. However, some
827 machines have default integer sizes that are different than the size
828 of a pointer, and so the following code is not portable:
833 If a pointer is declared with a kind that is too small, the compiler
834 will issue a warning; the resulting binary will probably not work
835 correctly, because the memory addresses stored in the pointers may be
836 truncated. It is safer to omit the first line of the above example;
837 if explicit declaration of ipt's type is omitted, then the compiler
838 will ensure that ipt is an integer variable large enough to hold a
841 Pointer arithmetic is valid with Cray pointers, but it is not the same
842 as C pointer arithmetic. Cray pointers are just ordinary integers, so
843 the user is responsible for determining how many bytes to add to a
844 pointer in order to increment it. Consider the following example:
848 pointer (ipt, pointee)
852 The last statement does not set ipt to the address of
853 @code{target(1)}, as one familiar with C pointer arithmetic might
854 expect. Adding 1 to ipt just adds one byte to the address stored in
857 Any expression involving the pointee will be translated to use the
858 value stored in the pointer as the base address.
860 To get the address of elements, this extension provides an intrinsic
861 function loc(), loc() is essentially the C '&' operator, except the
862 address is cast to an integer type:
865 pointer(ipt, arpte(10))
867 ipt = loc(ar) ! Makes arpte is an alias for ar
868 arpte(1) = 1.0 ! Sets ar(1) to 1.0
870 The pointer can also be set by a call to a malloc-type
871 function. There is no malloc intrinsic implemented as part of the
872 Cray pointer extension, but it might be a useful future addition to
873 @command{gfortran}. Even without an intrinsic malloc function,
874 dynamic memory allocation can be combined with Cray pointers by
875 calling a short C function:
879 void mymalloc_(void **ptr, int *nbytes)
881 *ptr = malloc(*nbytes);
890 pointer (ipdata, data(1024))
891 call mymalloc(ipdata,4*1024)
894 Cray pointees often are used to alias an existing variable. For
902 As long as ipt remains unchanged, iarr is now an alias for target.
903 The optimizer, however, will not detect this aliasing, so it is unsafe
904 to use iarr and target simultaneously. Using a pointee in any way
905 that violates the Fortran aliasing rules or assumptions is illegal.
906 It is the user's responsibility to avoid doing this; the compiler
907 works under the assumption that no such aliasing occurs.
909 Cray pointers will work correctly when there is no aliasing (i.e.,
910 when they're used to access a dynamically allocated block of memory),
911 and also in any routine where a pointee is used, but any variable with
912 which it shares storage is not used. Code that violates these rules
913 may not run as the user intends. This is not a bug in the optimizer;
914 any code that violates the aliasing rules is illegal. (Note that this
915 is not unique to gfortran; any Fortran compiler that supports Cray
916 pointers will ``incorrectly'' optimize code with illegal aliasing.)
918 There are a number of restrictions on the attributes that can be
919 applied to Cray pointers and pointees. Pointees may not have the
920 attributes ALLOCATABLE, INTENT, OPTIONAL, DUMMY, TARGET, EXTERNAL,
921 INTRINSIC, or POINTER. Pointers may not have the attributes
922 DIMENSION, POINTER, TARGET, ALLOCATABLE, EXTERNAL, or INTRINSIC.
923 Pointees may not occur in more than one pointer statement. A pointee
924 cannot be a pointer. Pointees cannot occur in equivalence, common, or
927 A pointer may be modified during the course of a program, and this
928 will change the location to which the pointee refers. However, when
929 pointees are passed as arguments, they are treated as ordinary
930 variables in the invoked function. Subsequent changes to the pointer
931 will not change the base address of the array that was passed.
933 @c ---------------------------------------------------------------------
934 @include intrinsic.texi
935 @c ---------------------------------------------------------------------
937 @c ---------------------------------------------------------------------
939 @c ---------------------------------------------------------------------
942 @chapter Contributing
945 Free software is only possible if people contribute to efforts
947 We're always in need of more people helping out with ideas
948 and comments, writing documentation and contributing code.
950 If you want to contribute to GNU Fortran 95,
951 have a look at the long lists of projects you can take on.
952 Some of these projects are small,
953 some of them are large;
954 some are completely orthogonal to the rest of what is
955 happening on @command{gfortran},
956 but others are ``mainstream'' projects in need of enthusiastic hackers.
957 All of these projects are important!
958 We'll eventually get around to the things here,
959 but they are also things doable by someone who is willing and able.
968 @section Contributors to GNU Fortran 95
973 Most of the parser was hand-crafted by @emph{Andy Vaught}, who is
974 also the initiator of the whole project. Thanks Andy!
975 Most of the interface with GCC was written by @emph{Paul Brook}.
977 The following individuals have contributed code and/or
978 ideas and significant help to the gfortran project
979 (in no particular order):
983 @item Katherine Holcomb
984 @item Tobias Schlüter
985 @item Steven Bosscher
988 @item Niels Kristian Bech Jensen
994 @item François-Xavier Coudert
997 @item Janne Blomqvist
1003 The following people have contributed bug reports,
1004 smaller or larger patches,
1005 and much needed feedback and encouragement for the
1006 @command{gfortran} project:
1009 @item Erik Schnetter
1014 Many other individuals have helped debug,
1015 test and improve @command{gfortran} over the past few years,
1016 and we welcome you to do the same!
1017 If you already have done so,
1018 and you would like to see your name listed in the
1019 list above, please contact us.
1027 @item Help build the test suite
1028 Solicit more code for donation to the test suite.
1029 We can keep code private on request.
1031 @item Bug hunting/squishing
1032 Find bugs and write more test cases!
1033 Test cases are especially very welcome,
1034 because it allows us to concentrate on fixing bugs
1035 instead of isolating them.
1037 @item Smaller projects (``bug'' fixes):
1039 @item Allow init exprs to be numbers raised to integer powers.
1040 @item Implement correct rounding.
1041 @item Implement F restrictions on Fortran 95 syntax.
1042 @item See about making Emacs-parsable error messages.
1046 If you wish to work on the runtime libraries,
1047 please contact a project maintainer.
1051 @c ---------------------------------------------------------------------
1053 @c ---------------------------------------------------------------------
1059 The GNU Fortran 95 Compiler aims to be a conforming implementation of
1060 ISO/IEC 1539:1997 (Fortran 95).
1062 In the future it may also support other variants of and extensions to
1063 the Fortran language. These include ANSI Fortran 77, ISO Fortran 90,
1064 ISO Fortran 2003 and OpenMP.
1067 * Fortran 2003 status::
1070 @node Fortran 2003 status
1071 @section Fortran 2003 status
1073 Although @command{gfortran} focuses on implementing the Fortran 95
1074 standard for the time being, a few Fortran 2003 features are currently
1079 Intrinsics @code{command_argument_count}, @code{get_command},
1080 @code{get_command_argument}, and @code{get_environment_variable}.
1083 @cindex Array constructors
1084 @cindex @code{[...]}
1085 Array constructors using square brackets. That is, @code{[...]} rather
1086 than @code{(/.../)}.
1089 @cindex @code{FLUSH} statement
1090 @code{FLUSH} statement.
1093 @cindex @code{IOMSG=} specifier
1094 @code{IOMSG=} specifier for I/O statements.
1097 @cindex @code{ENUM} statement
1098 @cindex @code{ENUMERATOR} statement
1099 @cindex @command{-fshort-enums}
1100 Support for the declaration of enumeration constants via the
1101 @code{ENUM} and @code{ENUMERATOR} statements. Interoperability with
1102 @command{gcc} is guaranteed also for the case where the
1103 @command{-fshort-enums} command line option is given.
1108 @c ---------------------------------------------------------------------
1109 @c GNU General Public License
1110 @c ---------------------------------------------------------------------
1116 @c ---------------------------------------------------------------------
1117 @c GNU Free Documentation License
1118 @c ---------------------------------------------------------------------
1124 @c ---------------------------------------------------------------------
1125 @c Funding Free Software
1126 @c ---------------------------------------------------------------------
1128 @include funding.texi
1130 @c ---------------------------------------------------------------------
1132 @c ---------------------------------------------------------------------