1 \input texinfo @c -*-texinfo-*-
4 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
6 @c GNAT DOCUMENTATION o
10 @c Copyright (C) 1992-2010, AdaCore o
12 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
14 @setfilename gnat_ugn.info
17 Copyright @copyright{} 1995-2009 Free Software Foundation,
20 Permission is granted to copy, distribute and/or modify this document
21 under the terms of the GNU Free Documentation License, Version 1.3 or
22 any later version published by the Free Software Foundation; with no
23 Invariant Sections, with no Front-Cover Texts and with no Back-Cover
24 Texts. A copy of the license is included in the section entitled
25 ``GNU Free Documentation License''.
28 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
30 @c GNAT_UGN Style Guide
32 @c 1. Always put a @noindent on the line before the first paragraph
33 @c after any of these commands:
45 @c 2. DO NOT use @example. Use @smallexample instead.
46 @c a) DO NOT use highlighting commands (@b{}, @i{}) inside an @smallexample
47 @c context. These can interfere with the readability of the texi
48 @c source file. Instead, use one of the following annotated
49 @c @smallexample commands, and preprocess the texi file with the
50 @c ada2texi tool (which generates appropriate highlighting):
51 @c @smallexample @c ada
52 @c @smallexample @c adanocomment
53 @c @smallexample @c projectfile
54 @c b) The "@c ada" markup will result in boldface for reserved words
55 @c and italics for comments
56 @c c) The "@c adanocomment" markup will result only in boldface for
57 @c reserved words (comments are left alone)
58 @c d) The "@c projectfile" markup is like "@c ada" except that the set
59 @c of reserved words include the new reserved words for project files
61 @c 3. Each @chapter, @section, @subsection, @subsubsection, etc.
62 @c command must be preceded by two empty lines
64 @c 4. The @item command should be on a line of its own if it is in an
65 @c @itemize or @enumerate command.
67 @c 5. When talking about ALI files use "ALI" (all uppercase), not "Ali"
70 @c 6. DO NOT put trailing spaces at the end of a line. Such spaces will
71 @c cause the document build to fail.
73 @c 7. DO NOT use @cartouche for examples that are longer than around 10 lines.
74 @c This command inhibits page breaks, so long examples in a @cartouche can
75 @c lead to large, ugly patches of empty space on a page.
77 @c NOTE: This file should be submitted to xgnatugn with either the vms flag
78 @c or the unw flag set. The unw flag covers topics for both Unix and
81 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
84 @c This flag is used where the text refers to conditions that exist when the
85 @c text was entered into the document but which may change over time.
86 @c Update the setting for the flag, and (if necessary) the text surrounding,
87 @c the references to the flag, on future doc revisions:
88 @c search for @value{NOW}.
92 @set DEFAULTLANGUAGEVERSION Ada 2005
93 @set NONDEFAULTLANGUAGEVERSION Ada 95
100 @set PLATFORM OpenVMS
105 @c The ARG is an optional argument. To be used for macro arguments in
106 @c their documentation (@defmac).
108 @r{[}@var{\varname\}@r{]}@c
110 @c Status as of November 2009:
111 @c Unfortunately texi2pdf and texi2html treat the trailing "@c"
112 @c differently, and faulty output is produced by one or the other
113 @c depending on whether the "@c" is present or absent.
114 @c As a result, the @ovar macro is not used, and all invocations
115 @c of the @ovar macro have been expanded inline.
118 @settitle @value{EDITION} User's Guide @value{PLATFORM}
119 @dircategory GNU Ada tools
121 * @value{EDITION} User's Guide: (gnat_ugn). @value{PLATFORM}
124 @include gcc-common.texi
126 @setchapternewpage odd
131 @title @value{EDITION} User's Guide
135 @titlefont{@i{@value{PLATFORM}}}
141 @subtitle GNAT, The GNU Ada Compiler
146 @vskip 0pt plus 1filll
153 @node Top, About This Guide, (dir), (dir)
154 @top @value{EDITION} User's Guide
157 @value{EDITION} User's Guide @value{PLATFORM}
160 GNAT, The GNU Ada Compiler@*
161 GCC version @value{version-GCC}@*
168 * Getting Started with GNAT::
169 * The GNAT Compilation Model::
170 * Compiling Using gcc::
171 * Binding Using gnatbind::
172 * Linking Using gnatlink::
173 * The GNAT Make Program gnatmake::
174 * Improving Performance::
175 * Renaming Files Using gnatchop::
176 * Configuration Pragmas::
177 * Handling Arbitrary File Naming Conventions Using gnatname::
178 * GNAT Project Manager::
179 * Tools Supporting Project Files::
180 * The Cross-Referencing Tools gnatxref and gnatfind::
181 * The GNAT Pretty-Printer gnatpp::
182 * The GNAT Metric Tool gnatmetric::
183 * File Name Krunching Using gnatkr::
184 * Preprocessing Using gnatprep::
185 * The GNAT Library Browser gnatls::
186 * Cleaning Up Using gnatclean::
188 * GNAT and Libraries::
189 * Using the GNU make Utility::
191 * Memory Management Issues::
192 * Stack Related Facilities::
193 * Verifying Properties Using gnatcheck::
194 * Creating Sample Bodies Using gnatstub::
195 * Generating Ada Bindings for C and C++ headers::
196 * Other Utility Programs::
197 * Running and Debugging Ada Programs::
199 * Code Coverage and Profiling::
202 * Compatibility with HP Ada::
204 * Platform-Specific Information for the Run-Time Libraries::
205 * Example of Binder Output File::
206 * Elaboration Order Handling in GNAT::
207 * Conditional Compilation::
209 * Compatibility and Porting Guide::
211 * Microsoft Windows Topics::
213 * GNU Free Documentation License::
216 --- The Detailed Node Listing ---
220 * What This Guide Contains::
221 * What You Should Know before Reading This Guide::
222 * Related Information::
225 Getting Started with GNAT
228 * Running a Simple Ada Program::
229 * Running a Program with Multiple Units::
230 * Using the gnatmake Utility::
232 * Editing with Emacs::
235 * Introduction to GPS::
238 The GNAT Compilation Model
240 * Source Representation::
241 * Foreign Language Representation::
242 * File Naming Rules::
243 * Using Other File Names::
244 * Alternative File Naming Schemes::
245 * Generating Object Files::
246 * Source Dependencies::
247 * The Ada Library Information Files::
248 * Binding an Ada Program::
249 * Mixed Language Programming::
251 * Building Mixed Ada & C++ Programs::
252 * Comparison between GNAT and C/C++ Compilation Models::
254 * Comparison between GNAT and Conventional Ada Library Models::
256 * Placement of temporary files::
259 Foreign Language Representation
262 * Other 8-Bit Codes::
263 * Wide Character Encodings::
265 Compiling Ada Programs With gcc
267 * Compiling Programs::
269 * Search Paths and the Run-Time Library (RTL)::
270 * Order of Compilation Issues::
275 * Output and Error Message Control::
276 * Warning Message Control::
277 * Debugging and Assertion Control::
278 * Validity Checking::
281 * Using gcc for Syntax Checking::
282 * Using gcc for Semantic Checking::
283 * Compiling Different Versions of Ada::
284 * Character Set Control::
285 * File Naming Control::
286 * Subprogram Inlining Control::
287 * Auxiliary Output Control::
288 * Debugging Control::
289 * Exception Handling Control::
290 * Units to Sources Mapping Files::
291 * Integrated Preprocessing::
296 Binding Ada Programs With gnatbind
299 * Switches for gnatbind::
300 * Command-Line Access::
301 * Search Paths for gnatbind::
302 * Examples of gnatbind Usage::
304 Switches for gnatbind
306 * Consistency-Checking Modes::
307 * Binder Error Message Control::
308 * Elaboration Control::
310 * Binding with Non-Ada Main Programs::
311 * Binding Programs with No Main Subprogram::
313 Linking Using gnatlink
316 * Switches for gnatlink::
318 The GNAT Make Program gnatmake
321 * Switches for gnatmake::
322 * Mode Switches for gnatmake::
323 * Notes on the Command Line::
324 * How gnatmake Works::
325 * Examples of gnatmake Usage::
327 Improving Performance
328 * Performance Considerations::
329 * Text_IO Suggestions::
330 * Reducing Size of Ada Executables with gnatelim::
331 * Reducing Size of Executables with unused subprogram/data elimination::
333 Performance Considerations
334 * Controlling Run-Time Checks::
335 * Use of Restrictions::
336 * Optimization Levels::
337 * Debugging Optimized Code::
338 * Inlining of Subprograms::
339 * Other Optimization Switches::
340 * Optimization and Strict Aliasing::
342 * Coverage Analysis::
345 Reducing Size of Ada Executables with gnatelim
348 * Processing Precompiled Libraries::
349 * Correcting the List of Eliminate Pragmas::
350 * Making Your Executables Smaller::
351 * Summary of the gnatelim Usage Cycle::
353 Reducing Size of Executables with unused subprogram/data elimination
354 * About unused subprogram/data elimination::
355 * Compilation options::
357 Renaming Files Using gnatchop
359 * Handling Files with Multiple Units::
360 * Operating gnatchop in Compilation Mode::
361 * Command Line for gnatchop::
362 * Switches for gnatchop::
363 * Examples of gnatchop Usage::
365 Configuration Pragmas
367 * Handling of Configuration Pragmas::
368 * The Configuration Pragmas Files::
370 Handling Arbitrary File Naming Conventions Using gnatname
372 * Arbitrary File Naming Conventions::
374 * Switches for gnatname::
375 * Examples of gnatname Usage::
377 The Cross-Referencing Tools gnatxref and gnatfind
379 * Switches for gnatxref::
380 * Switches for gnatfind::
381 * Project Files for gnatxref and gnatfind::
382 * Regular Expressions in gnatfind and gnatxref::
383 * Examples of gnatxref Usage::
384 * Examples of gnatfind Usage::
386 The GNAT Pretty-Printer gnatpp
388 * Switches for gnatpp::
391 The GNAT Metrics Tool gnatmetric
393 * Switches for gnatmetric::
395 File Name Krunching Using gnatkr
400 * Examples of gnatkr Usage::
402 Preprocessing Using gnatprep
403 * Preprocessing Symbols::
405 * Switches for gnatprep::
406 * Form of Definitions File::
407 * Form of Input Text for gnatprep::
409 The GNAT Library Browser gnatls
412 * Switches for gnatls::
413 * Examples of gnatls Usage::
415 Cleaning Up Using gnatclean
417 * Running gnatclean::
418 * Switches for gnatclean::
419 @c * Examples of gnatclean Usage::
425 * Introduction to Libraries in GNAT::
426 * General Ada Libraries::
427 * Stand-alone Ada Libraries::
428 * Rebuilding the GNAT Run-Time Library::
430 Using the GNU make Utility
432 * Using gnatmake in a Makefile::
433 * Automatically Creating a List of Directories::
434 * Generating the Command Line Switches::
435 * Overcoming Command Line Length Limits::
438 Memory Management Issues
440 * Some Useful Memory Pools::
441 * The GNAT Debug Pool Facility::
446 Stack Related Facilities
448 * Stack Overflow Checking::
449 * Static Stack Usage Analysis::
450 * Dynamic Stack Usage Analysis::
452 Some Useful Memory Pools
454 The GNAT Debug Pool Facility
460 * Switches for gnatmem::
461 * Example of gnatmem Usage::
464 Verifying Properties Using gnatcheck
466 Sample Bodies Using gnatstub
469 * Switches for gnatstub::
471 Other Utility Programs
473 * Using Other Utility Programs with GNAT::
474 * The External Symbol Naming Scheme of GNAT::
475 * Converting Ada Files to html with gnathtml::
478 Code Coverage and Profiling
480 * Code Coverage of Ada Programs using gcov::
481 * Profiling an Ada Program using gprof::
484 Running and Debugging Ada Programs
486 * The GNAT Debugger GDB::
488 * Introduction to GDB Commands::
489 * Using Ada Expressions::
490 * Calling User-Defined Subprograms::
491 * Using the Next Command in a Function::
494 * Debugging Generic Units::
495 * Remote Debugging using gdbserver::
496 * GNAT Abnormal Termination or Failure to Terminate::
497 * Naming Conventions for GNAT Source Files::
498 * Getting Internal Debugging Information::
506 Compatibility with HP Ada
508 * Ada Language Compatibility::
509 * Differences in the Definition of Package System::
510 * Language-Related Features::
511 * The Package STANDARD::
512 * The Package SYSTEM::
513 * Tasking and Task-Related Features::
514 * Pragmas and Pragma-Related Features::
515 * Library of Predefined Units::
517 * Main Program Definition::
518 * Implementation-Defined Attributes::
519 * Compiler and Run-Time Interfacing::
520 * Program Compilation and Library Management::
522 * Implementation Limits::
523 * Tools and Utilities::
525 Language-Related Features
527 * Integer Types and Representations::
528 * Floating-Point Types and Representations::
529 * Pragmas Float_Representation and Long_Float::
530 * Fixed-Point Types and Representations::
531 * Record and Array Component Alignment::
533 * Other Representation Clauses::
535 Tasking and Task-Related Features
537 * Implementation of Tasks in HP Ada for OpenVMS Alpha Systems::
538 * Assigning Task IDs::
539 * Task IDs and Delays::
540 * Task-Related Pragmas::
541 * Scheduling and Task Priority::
543 * External Interrupts::
545 Pragmas and Pragma-Related Features
547 * Restrictions on the Pragma INLINE::
548 * Restrictions on the Pragma INTERFACE::
549 * Restrictions on the Pragma SYSTEM_NAME::
551 Library of Predefined Units
553 * Changes to DECLIB::
557 * Shared Libraries and Options Files::
561 Platform-Specific Information for the Run-Time Libraries
563 * Summary of Run-Time Configurations::
564 * Specifying a Run-Time Library::
565 * Choosing the Scheduling Policy::
566 * Solaris-Specific Considerations::
567 * Linux-Specific Considerations::
568 * AIX-Specific Considerations::
569 * Irix-Specific Considerations::
570 * RTX-Specific Considerations::
571 * HP-UX-Specific Considerations::
573 Example of Binder Output File
575 Elaboration Order Handling in GNAT
578 * Checking the Elaboration Order::
579 * Controlling the Elaboration Order::
580 * Controlling Elaboration in GNAT - Internal Calls::
581 * Controlling Elaboration in GNAT - External Calls::
582 * Default Behavior in GNAT - Ensuring Safety::
583 * Treatment of Pragma Elaborate::
584 * Elaboration Issues for Library Tasks::
585 * Mixing Elaboration Models::
586 * What to Do If the Default Elaboration Behavior Fails::
587 * Elaboration for Access-to-Subprogram Values::
588 * Summary of Procedures for Elaboration Control::
589 * Other Elaboration Order Considerations::
591 Conditional Compilation
592 * Use of Boolean Constants::
593 * Debugging - A Special Case::
594 * Conditionalizing Declarations::
595 * Use of Alternative Implementations::
600 * Basic Assembler Syntax::
601 * A Simple Example of Inline Assembler::
602 * Output Variables in Inline Assembler::
603 * Input Variables in Inline Assembler::
604 * Inlining Inline Assembler Code::
605 * Other Asm Functionality::
607 Compatibility and Porting Guide
609 * Compatibility with Ada 83::
610 * Compatibility between Ada 95 and Ada 2005::
611 * Implementation-dependent characteristics::
613 @c This brief section is only in the non-VMS version
614 @c The complete chapter on HP Ada issues is in the VMS version
615 * Compatibility with HP Ada 83::
617 * Compatibility with Other Ada Systems::
618 * Representation Clauses::
620 * Transitioning to 64-Bit GNAT for OpenVMS::
624 Microsoft Windows Topics
626 * Using GNAT on Windows::
627 * CONSOLE and WINDOWS subsystems::
629 * Mixed-Language Programming on Windows::
630 * Windows Calling Conventions::
631 * Introduction to Dynamic Link Libraries (DLLs)::
632 * Using DLLs with GNAT::
633 * Building DLLs with GNAT::
634 * GNAT and Windows Resources::
636 * Setting Stack Size from gnatlink::
637 * Setting Heap Size from gnatlink::
644 @node About This Guide
645 @unnumbered About This Guide
649 This guide describes the use of @value{EDITION},
650 a compiler and software development toolset for the full Ada
651 programming language, implemented on OpenVMS for HP's Alpha and
652 Integrity server (I64) platforms.
655 This guide describes the use of @value{EDITION},
656 a compiler and software development
657 toolset for the full Ada programming language.
659 It documents the features of the compiler and tools, and explains
660 how to use them to build Ada applications.
662 @value{EDITION} implements Ada 95 and Ada 2005, and it may also be invoked in
663 Ada 83 compatibility mode.
664 By default, @value{EDITION} assumes @value{DEFAULTLANGUAGEVERSION},
665 but you can override with a compiler switch
666 (@pxref{Compiling Different Versions of Ada})
667 to explicitly specify the language version.
668 Throughout this manual, references to ``Ada'' without a year suffix
669 apply to both the Ada 95 and Ada 2005 versions of the language.
673 For ease of exposition, ``@value{EDITION}'' will be referred to simply as
674 ``GNAT'' in the remainder of this document.
681 * What This Guide Contains::
682 * What You Should Know before Reading This Guide::
683 * Related Information::
687 @node What This Guide Contains
688 @unnumberedsec What This Guide Contains
691 This guide contains the following chapters:
695 @ref{Getting Started with GNAT}, describes how to get started compiling
696 and running Ada programs with the GNAT Ada programming environment.
698 @ref{The GNAT Compilation Model}, describes the compilation model used
702 @ref{Compiling Using gcc}, describes how to compile
703 Ada programs with @command{gcc}, the Ada compiler.
706 @ref{Binding Using gnatbind}, describes how to
707 perform binding of Ada programs with @code{gnatbind}, the GNAT binding
711 @ref{Linking Using gnatlink},
712 describes @command{gnatlink}, a
713 program that provides for linking using the GNAT run-time library to
714 construct a program. @command{gnatlink} can also incorporate foreign language
715 object units into the executable.
718 @ref{The GNAT Make Program gnatmake}, describes @command{gnatmake}, a
719 utility that automatically determines the set of sources
720 needed by an Ada compilation unit, and executes the necessary compilations
724 @ref{Improving Performance}, shows various techniques for making your
725 Ada program run faster or take less space.
726 It discusses the effect of the compiler's optimization switch and
727 also describes the @command{gnatelim} tool and unused subprogram/data
731 @ref{Renaming Files Using gnatchop}, describes
732 @code{gnatchop}, a utility that allows you to preprocess a file that
733 contains Ada source code, and split it into one or more new files, one
734 for each compilation unit.
737 @ref{Configuration Pragmas}, describes the configuration pragmas
741 @ref{Handling Arbitrary File Naming Conventions Using gnatname},
742 shows how to override the default GNAT file naming conventions,
743 either for an individual unit or globally.
746 @ref{GNAT Project Manager}, describes how to use project files
747 to organize large projects.
750 @ref{The Cross-Referencing Tools gnatxref and gnatfind}, discusses
751 @code{gnatxref} and @code{gnatfind}, two tools that provide an easy
752 way to navigate through sources.
755 @ref{The GNAT Pretty-Printer gnatpp}, shows how to produce a reformatted
756 version of an Ada source file with control over casing, indentation,
757 comment placement, and other elements of program presentation style.
760 @ref{The GNAT Metric Tool gnatmetric}, shows how to compute various
761 metrics for an Ada source file, such as the number of types and subprograms,
762 and assorted complexity measures.
765 @ref{File Name Krunching Using gnatkr}, describes the @code{gnatkr}
766 file name krunching utility, used to handle shortened
767 file names on operating systems with a limit on the length of names.
770 @ref{Preprocessing Using gnatprep}, describes @code{gnatprep}, a
771 preprocessor utility that allows a single source file to be used to
772 generate multiple or parameterized source files by means of macro
776 @ref{The GNAT Library Browser gnatls}, describes @code{gnatls}, a
777 utility that displays information about compiled units, including dependences
778 on the corresponding sources files, and consistency of compilations.
781 @ref{Cleaning Up Using gnatclean}, describes @code{gnatclean}, a utility
782 to delete files that are produced by the compiler, binder and linker.
786 @ref{GNAT and Libraries}, describes the process of creating and using
787 Libraries with GNAT. It also describes how to recompile the GNAT run-time
791 @ref{Using the GNU make Utility}, describes some techniques for using
792 the GNAT toolset in Makefiles.
796 @ref{Memory Management Issues}, describes some useful predefined storage pools
797 and in particular the GNAT Debug Pool facility, which helps detect incorrect
800 It also describes @command{gnatmem}, a utility that monitors dynamic
801 allocation and deallocation and helps detect ``memory leaks''.
805 @ref{Stack Related Facilities}, describes some useful tools associated with
806 stack checking and analysis.
809 @ref{Verifying Properties Using gnatcheck}, discusses @code{gnatcheck},
810 a utility that checks Ada code against a set of rules.
813 @ref{Creating Sample Bodies Using gnatstub}, discusses @code{gnatstub},
814 a utility that generates empty but compilable bodies for library units.
817 @ref{Generating Ada Bindings for C and C++ headers}, describes how to
818 generate automatically Ada bindings from C and C++ headers.
821 @ref{Other Utility Programs}, discusses several other GNAT utilities,
822 including @code{gnathtml}.
826 @ref{Code Coverage and Profiling}, describes how to perform a structural
827 coverage and profile the execution of Ada programs.
831 @ref{Running and Debugging Ada Programs}, describes how to run and debug
836 @ref{Compatibility with HP Ada}, details the compatibility of GNAT with
837 HP Ada 83 @footnote{``HP Ada'' refers to the legacy product originally
838 developed by Digital Equipment Corporation and currently supported by HP.}
839 for OpenVMS Alpha. This product was formerly known as DEC Ada,
842 historical compatibility reasons, the relevant libraries still use the
847 @ref{Platform-Specific Information for the Run-Time Libraries},
848 describes the various run-time
849 libraries supported by GNAT on various platforms and explains how to
850 choose a particular library.
853 @ref{Example of Binder Output File}, shows the source code for the binder
854 output file for a sample program.
857 @ref{Elaboration Order Handling in GNAT}, describes how GNAT helps
858 you deal with elaboration order issues.
861 @ref{Conditional Compilation}, describes how to model conditional compilation,
862 both with Ada in general and with GNAT facilities in particular.
865 @ref{Inline Assembler}, shows how to use the inline assembly facility
869 @ref{Compatibility and Porting Guide}, contains sections on compatibility
870 of GNAT with other Ada development environments (including Ada 83 systems),
871 to assist in porting code from those environments.
875 @ref{Microsoft Windows Topics}, presents information relevant to the
876 Microsoft Windows platform.
880 @c *************************************************
881 @node What You Should Know before Reading This Guide
882 @c *************************************************
883 @unnumberedsec What You Should Know before Reading This Guide
885 @cindex Ada 95 Language Reference Manual
886 @cindex Ada 2005 Language Reference Manual
888 This guide assumes a basic familiarity with the Ada 95 language, as
889 described in the International Standard ANSI/ISO/IEC-8652:1995, January
891 It does not require knowledge of the new features introduced by Ada 2005,
892 (officially known as ISO/IEC 8652:1995 with Technical Corrigendum 1
894 Both reference manuals are included in the GNAT documentation
897 @node Related Information
898 @unnumberedsec Related Information
901 For further information about related tools, refer to the following
906 @xref{Top, GNAT Reference Manual, About This Guide, gnat_rm, GNAT
907 Reference Manual}, which contains all reference material for the GNAT
908 implementation of Ada.
912 @cite{Using the GNAT Programming Studio}, which describes the GPS
913 Integrated Development Environment.
916 @cite{GNAT Programming Studio Tutorial}, which introduces the
917 main GPS features through examples.
921 @cite{Ada 95 Reference Manual}, which contains reference
922 material for the Ada 95 programming language.
925 @cite{Ada 2005 Reference Manual}, which contains reference
926 material for the Ada 2005 programming language.
929 @xref{Top,, Debugging with GDB, gdb, Debugging with GDB},
931 in the GNU:[DOCS] directory,
933 for all details on the use of the GNU source-level debugger.
936 @xref{Top,, The extensible self-documenting text editor, emacs,
939 located in the GNU:[DOCS] directory if the EMACS kit is installed,
941 for full information on the extensible editor and programming
948 @unnumberedsec Conventions
950 @cindex Typographical conventions
953 Following are examples of the typographical and graphic conventions used
958 @code{Functions}, @command{utility program names}, @code{standard names},
962 @option{Option flags}
965 @file{File names}, @samp{button names}, and @samp{field names}.
968 @code{Variables}, @env{environment variables}, and @var{metasyntactic
975 @r{[}optional information or parameters@r{]}
978 Examples are described by text
980 and then shown this way.
985 Commands that are entered by the user are preceded in this manual by the
986 characters @w{``@code{$ }''} (dollar sign followed by space). If your system
987 uses this sequence as a prompt, then the commands will appear exactly as
988 you see them in the manual. If your system uses some other prompt, then
989 the command will appear with the @code{$} replaced by whatever prompt
990 character you are using.
993 Full file names are shown with the ``@code{/}'' character
994 as the directory separator; e.g., @file{parent-dir/subdir/myfile.adb}.
995 If you are using GNAT on a Windows platform, please note that
996 the ``@code{\}'' character should be used instead.
999 @c ****************************
1000 @node Getting Started with GNAT
1001 @chapter Getting Started with GNAT
1004 This chapter describes some simple ways of using GNAT to build
1005 executable Ada programs.
1007 @ref{Running GNAT}, through @ref{Using the gnatmake Utility},
1008 show how to use the command line environment.
1009 @ref{Introduction to GPS}, provides a brief
1010 introduction to the GNAT Programming Studio, a visually-oriented
1011 Integrated Development Environment for GNAT.
1012 GPS offers a graphical ``look and feel'', support for development in
1013 other programming languages, comprehensive browsing features, and
1014 many other capabilities.
1015 For information on GPS please refer to
1016 @cite{Using the GNAT Programming Studio}.
1021 * Running a Simple Ada Program::
1022 * Running a Program with Multiple Units::
1023 * Using the gnatmake Utility::
1025 * Editing with Emacs::
1028 * Introduction to GPS::
1033 @section Running GNAT
1036 Three steps are needed to create an executable file from an Ada source
1041 The source file(s) must be compiled.
1043 The file(s) must be bound using the GNAT binder.
1045 All appropriate object files must be linked to produce an executable.
1049 All three steps are most commonly handled by using the @command{gnatmake}
1050 utility program that, given the name of the main program, automatically
1051 performs the necessary compilation, binding and linking steps.
1053 @node Running a Simple Ada Program
1054 @section Running a Simple Ada Program
1057 Any text editor may be used to prepare an Ada program.
1059 used, the optional Ada mode may be helpful in laying out the program.)
1061 program text is a normal text file. We will assume in our initial
1062 example that you have used your editor to prepare the following
1063 standard format text file:
1065 @smallexample @c ada
1067 with Ada.Text_IO; use Ada.Text_IO;
1070 Put_Line ("Hello WORLD!");
1076 This file should be named @file{hello.adb}.
1077 With the normal default file naming conventions, GNAT requires
1079 contain a single compilation unit whose file name is the
1081 with periods replaced by hyphens; the
1082 extension is @file{ads} for a
1083 spec and @file{adb} for a body.
1084 You can override this default file naming convention by use of the
1085 special pragma @code{Source_File_Name} (@pxref{Using Other File Names}).
1086 Alternatively, if you want to rename your files according to this default
1087 convention, which is probably more convenient if you will be using GNAT
1088 for all your compilations, then the @code{gnatchop} utility
1089 can be used to generate correctly-named source files
1090 (@pxref{Renaming Files Using gnatchop}).
1092 You can compile the program using the following command (@code{$} is used
1093 as the command prompt in the examples in this document):
1100 @command{gcc} is the command used to run the compiler. This compiler is
1101 capable of compiling programs in several languages, including Ada and
1102 C. It assumes that you have given it an Ada program if the file extension is
1103 either @file{.ads} or @file{.adb}, and it will then call
1104 the GNAT compiler to compile the specified file.
1107 The @option{-c} switch is required. It tells @command{gcc} to only do a
1108 compilation. (For C programs, @command{gcc} can also do linking, but this
1109 capability is not used directly for Ada programs, so the @option{-c}
1110 switch must always be present.)
1113 This compile command generates a file
1114 @file{hello.o}, which is the object
1115 file corresponding to your Ada program. It also generates
1116 an ``Ada Library Information'' file @file{hello.ali},
1117 which contains additional information used to check
1118 that an Ada program is consistent.
1119 To build an executable file,
1120 use @code{gnatbind} to bind the program
1121 and @command{gnatlink} to link it. The
1122 argument to both @code{gnatbind} and @command{gnatlink} is the name of the
1123 @file{ALI} file, but the default extension of @file{.ali} can
1124 be omitted. This means that in the most common case, the argument
1125 is simply the name of the main program:
1133 A simpler method of carrying out these steps is to use
1135 a master program that invokes all the required
1136 compilation, binding and linking tools in the correct order. In particular,
1137 @command{gnatmake} automatically recompiles any sources that have been
1138 modified since they were last compiled, or sources that depend
1139 on such modified sources, so that ``version skew'' is avoided.
1140 @cindex Version skew (avoided by @command{gnatmake})
1143 $ gnatmake hello.adb
1147 The result is an executable program called @file{hello}, which can be
1155 assuming that the current directory is on the search path
1156 for executable programs.
1159 and, if all has gone well, you will see
1166 appear in response to this command.
1168 @c ****************************************
1169 @node Running a Program with Multiple Units
1170 @section Running a Program with Multiple Units
1173 Consider a slightly more complicated example that has three files: a
1174 main program, and the spec and body of a package:
1176 @smallexample @c ada
1179 package Greetings is
1184 with Ada.Text_IO; use Ada.Text_IO;
1185 package body Greetings is
1188 Put_Line ("Hello WORLD!");
1191 procedure Goodbye is
1193 Put_Line ("Goodbye WORLD!");
1210 Following the one-unit-per-file rule, place this program in the
1211 following three separate files:
1215 spec of package @code{Greetings}
1218 body of package @code{Greetings}
1221 body of main program
1225 To build an executable version of
1226 this program, we could use four separate steps to compile, bind, and link
1227 the program, as follows:
1231 $ gcc -c greetings.adb
1237 Note that there is no required order of compilation when using GNAT.
1238 In particular it is perfectly fine to compile the main program first.
1239 Also, it is not necessary to compile package specs in the case where
1240 there is an accompanying body; you only need to compile the body. If you want
1241 to submit these files to the compiler for semantic checking and not code
1242 generation, then use the
1243 @option{-gnatc} switch:
1246 $ gcc -c greetings.ads -gnatc
1250 Although the compilation can be done in separate steps as in the
1251 above example, in practice it is almost always more convenient
1252 to use the @command{gnatmake} tool. All you need to know in this case
1253 is the name of the main program's source file. The effect of the above four
1254 commands can be achieved with a single one:
1257 $ gnatmake gmain.adb
1261 In the next section we discuss the advantages of using @command{gnatmake} in
1264 @c *****************************
1265 @node Using the gnatmake Utility
1266 @section Using the @command{gnatmake} Utility
1269 If you work on a program by compiling single components at a time using
1270 @command{gcc}, you typically keep track of the units you modify. In order to
1271 build a consistent system, you compile not only these units, but also any
1272 units that depend on the units you have modified.
1273 For example, in the preceding case,
1274 if you edit @file{gmain.adb}, you only need to recompile that file. But if
1275 you edit @file{greetings.ads}, you must recompile both
1276 @file{greetings.adb} and @file{gmain.adb}, because both files contain
1277 units that depend on @file{greetings.ads}.
1279 @code{gnatbind} will warn you if you forget one of these compilation
1280 steps, so that it is impossible to generate an inconsistent program as a
1281 result of forgetting to do a compilation. Nevertheless it is tedious and
1282 error-prone to keep track of dependencies among units.
1283 One approach to handle the dependency-bookkeeping is to use a
1284 makefile. However, makefiles present maintenance problems of their own:
1285 if the dependencies change as you change the program, you must make
1286 sure that the makefile is kept up-to-date manually, which is also an
1287 error-prone process.
1289 The @command{gnatmake} utility takes care of these details automatically.
1290 Invoke it using either one of the following forms:
1293 $ gnatmake gmain.adb
1294 $ gnatmake ^gmain^GMAIN^
1298 The argument is the name of the file containing the main program;
1299 you may omit the extension. @command{gnatmake}
1300 examines the environment, automatically recompiles any files that need
1301 recompiling, and binds and links the resulting set of object files,
1302 generating the executable file, @file{^gmain^GMAIN.EXE^}.
1303 In a large program, it
1304 can be extremely helpful to use @command{gnatmake}, because working out by hand
1305 what needs to be recompiled can be difficult.
1307 Note that @command{gnatmake}
1308 takes into account all the Ada rules that
1309 establish dependencies among units. These include dependencies that result
1310 from inlining subprogram bodies, and from
1311 generic instantiation. Unlike some other
1312 Ada make tools, @command{gnatmake} does not rely on the dependencies that were
1313 found by the compiler on a previous compilation, which may possibly
1314 be wrong when sources change. @command{gnatmake} determines the exact set of
1315 dependencies from scratch each time it is run.
1318 @node Editing with Emacs
1319 @section Editing with Emacs
1323 Emacs is an extensible self-documenting text editor that is available in a
1324 separate VMSINSTAL kit.
1326 Invoke Emacs by typing @kbd{Emacs} at the command prompt. To get started,
1327 click on the Emacs Help menu and run the Emacs Tutorial.
1328 In a character cell terminal, Emacs help is invoked with @kbd{Ctrl-h} (also
1329 written as @kbd{C-h}), and the tutorial by @kbd{C-h t}.
1331 Documentation on Emacs and other tools is available in Emacs under the
1332 pull-down menu button: @code{Help - Info}. After selecting @code{Info},
1333 use the middle mouse button to select a topic (e.g.@: Emacs).
1335 In a character cell terminal, do @kbd{C-h i} to invoke info, and then @kbd{m}
1336 (stands for menu) followed by the menu item desired, as in @kbd{m Emacs}, to
1337 get to the Emacs manual.
1338 Help on Emacs is also available by typing @kbd{HELP EMACS} at the DCL command
1341 The tutorial is highly recommended in order to learn the intricacies of Emacs,
1342 which is sufficiently extensible to provide for a complete programming
1343 environment and shell for the sophisticated user.
1347 @node Introduction to GPS
1348 @section Introduction to GPS
1349 @cindex GPS (GNAT Programming Studio)
1350 @cindex GNAT Programming Studio (GPS)
1352 Although the command line interface (@command{gnatmake}, etc.) alone
1353 is sufficient, a graphical Interactive Development
1354 Environment can make it easier for you to compose, navigate, and debug
1355 programs. This section describes the main features of GPS
1356 (``GNAT Programming Studio''), the GNAT graphical IDE.
1357 You will see how to use GPS to build and debug an executable, and
1358 you will also learn some of the basics of the GNAT ``project'' facility.
1360 GPS enables you to do much more than is presented here;
1361 e.g., you can produce a call graph, interface to a third-party
1362 Version Control System, and inspect the generated assembly language
1364 Indeed, GPS also supports languages other than Ada.
1365 Such additional information, and an explanation of all of the GPS menu
1366 items. may be found in the on-line help, which includes
1367 a user's guide and a tutorial (these are also accessible from the GNAT
1371 * Building a New Program with GPS::
1372 * Simple Debugging with GPS::
1375 @node Building a New Program with GPS
1376 @subsection Building a New Program with GPS
1378 GPS invokes the GNAT compilation tools using information
1379 contained in a @emph{project} (also known as a @emph{project file}):
1380 a collection of properties such
1381 as source directories, identities of main subprograms, tool switches, etc.,
1382 and their associated values.
1383 See @ref{GNAT Project Manager} for details.
1384 In order to run GPS, you will need to either create a new project
1385 or else open an existing one.
1387 This section will explain how you can use GPS to create a project,
1388 to associate Ada source files with a project, and to build and run
1392 @item @emph{Creating a project}
1394 Invoke GPS, either from the command line or the platform's IDE.
1395 After it starts, GPS will display a ``Welcome'' screen with three
1400 @code{Start with default project in directory}
1403 @code{Create new project with wizard}
1406 @code{Open existing project}
1410 Select @code{Create new project with wizard} and press @code{OK}.
1411 A new window will appear. In the text box labeled with
1412 @code{Enter the name of the project to create}, type @file{sample}
1413 as the project name.
1414 In the next box, browse to choose the directory in which you
1415 would like to create the project file.
1416 After selecting an appropriate directory, press @code{Forward}.
1418 A window will appear with the title
1419 @code{Version Control System Configuration}.
1420 Simply press @code{Forward}.
1422 A window will appear with the title
1423 @code{Please select the source directories for this project}.
1424 The directory that you specified for the project file will be selected
1425 by default as the one to use for sources; simply press @code{Forward}.
1427 A window will appear with the title
1428 @code{Please select the build directory for this project}.
1429 The directory that you specified for the project file will be selected
1430 by default for object files and executables;
1431 simply press @code{Forward}.
1433 A window will appear with the title
1434 @code{Please select the main units for this project}.
1435 You will supply this information later, after creating the source file.
1436 Simply press @code{Forward} for now.
1438 A window will appear with the title
1439 @code{Please select the switches to build the project}.
1440 Press @code{Apply}. This will create a project file named
1441 @file{sample.prj} in the directory that you had specified.
1443 @item @emph{Creating and saving the source file}
1445 After you create the new project, a GPS window will appear, which is
1446 partitioned into two main sections:
1450 A @emph{Workspace area}, initially greyed out, which you will use for
1451 creating and editing source files
1454 Directly below, a @emph{Messages area}, which initially displays a
1455 ``Welcome'' message.
1456 (If the Messages area is not visible, drag its border upward to expand it.)
1460 Select @code{File} on the menu bar, and then the @code{New} command.
1461 The Workspace area will become white, and you can now
1462 enter the source program explicitly.
1463 Type the following text
1465 @smallexample @c ada
1467 with Ada.Text_IO; use Ada.Text_IO;
1470 Put_Line("Hello from GPS!");
1476 Select @code{File}, then @code{Save As}, and enter the source file name
1478 The file will be saved in the same directory you specified as the
1479 location of the default project file.
1481 @item @emph{Updating the project file}
1483 You need to add the new source file to the project.
1485 the @code{Project} menu and then @code{Edit project properties}.
1486 Click the @code{Main files} tab on the left, and then the
1488 Choose @file{hello.adb} from the list, and press @code{Open}.
1489 The project settings window will reflect this action.
1492 @item @emph{Building and running the program}
1494 In the main GPS window, now choose the @code{Build} menu, then @code{Make},
1495 and select @file{hello.adb}.
1496 The Messages window will display the resulting invocations of @command{gcc},
1497 @command{gnatbind}, and @command{gnatlink}
1498 (reflecting the default switch settings from the
1499 project file that you created) and then a ``successful compilation/build''
1502 To run the program, choose the @code{Build} menu, then @code{Run}, and
1503 select @command{hello}.
1504 An @emph{Arguments Selection} window will appear.
1505 There are no command line arguments, so just click @code{OK}.
1507 The Messages window will now display the program's output (the string
1508 @code{Hello from GPS}), and at the bottom of the GPS window a status
1509 update is displayed (@code{Run: hello}).
1510 Close the GPS window (or select @code{File}, then @code{Exit}) to
1511 terminate this GPS session.
1514 @node Simple Debugging with GPS
1515 @subsection Simple Debugging with GPS
1517 This section illustrates basic debugging techniques (setting breakpoints,
1518 examining/modifying variables, single stepping).
1521 @item @emph{Opening a project}
1523 Start GPS and select @code{Open existing project}; browse to
1524 specify the project file @file{sample.prj} that you had created in the
1527 @item @emph{Creating a source file}
1529 Select @code{File}, then @code{New}, and type in the following program:
1531 @smallexample @c ada
1533 with Ada.Text_IO; use Ada.Text_IO;
1534 procedure Example is
1535 Line : String (1..80);
1538 Put_Line("Type a line of text at each prompt; an empty line to exit");
1542 Put_Line (Line (1..N) );
1550 Select @code{File}, then @code{Save as}, and enter the file name
1553 @item @emph{Updating the project file}
1555 Add @code{Example} as a new main unit for the project:
1558 Select @code{Project}, then @code{Edit Project Properties}.
1561 Select the @code{Main files} tab, click @code{Add}, then
1562 select the file @file{example.adb} from the list, and
1564 You will see the file name appear in the list of main units
1570 @item @emph{Building/running the executable}
1572 To build the executable
1573 select @code{Build}, then @code{Make}, and then choose @file{example.adb}.
1575 Run the program to see its effect (in the Messages area).
1576 Each line that you enter is displayed; an empty line will
1577 cause the loop to exit and the program to terminate.
1579 @item @emph{Debugging the program}
1581 Note that the @option{-g} switches to @command{gcc} and @command{gnatlink},
1582 which are required for debugging, are on by default when you create
1584 Thus unless you intentionally remove these settings, you will be able
1585 to debug any program that you develop using GPS.
1588 @item @emph{Initializing}
1590 Select @code{Debug}, then @code{Initialize}, then @file{example}
1592 @item @emph{Setting a breakpoint}
1594 After performing the initialization step, you will observe a small
1595 icon to the right of each line number.
1596 This serves as a toggle for breakpoints; clicking the icon will
1597 set a breakpoint at the corresponding line (the icon will change to
1598 a red circle with an ``x''), and clicking it again
1599 will remove the breakpoint / reset the icon.
1601 For purposes of this example, set a breakpoint at line 10 (the
1602 statement @code{Put_Line@ (Line@ (1..N));}
1604 @item @emph{Starting program execution}
1606 Select @code{Debug}, then @code{Run}. When the
1607 @code{Program Arguments} window appears, click @code{OK}.
1608 A console window will appear; enter some line of text,
1609 e.g.@: @code{abcde}, at the prompt.
1610 The program will pause execution when it gets to the
1611 breakpoint, and the corresponding line is highlighted.
1613 @item @emph{Examining a variable}
1615 Move the mouse over one of the occurrences of the variable @code{N}.
1616 You will see the value (5) displayed, in ``tool tip'' fashion.
1617 Right click on @code{N}, select @code{Debug}, then select @code{Display N}.
1618 You will see information about @code{N} appear in the @code{Debugger Data}
1619 pane, showing the value as 5.
1621 @item @emph{Assigning a new value to a variable}
1623 Right click on the @code{N} in the @code{Debugger Data} pane, and
1624 select @code{Set value of N}.
1625 When the input window appears, enter the value @code{4} and click
1627 This value does not automatically appear in the @code{Debugger Data}
1628 pane; to see it, right click again on the @code{N} in the
1629 @code{Debugger Data} pane and select @code{Update value}.
1630 The new value, 4, will appear in red.
1632 @item @emph{Single stepping}
1634 Select @code{Debug}, then @code{Next}.
1635 This will cause the next statement to be executed, in this case the
1636 call of @code{Put_Line} with the string slice.
1637 Notice in the console window that the displayed string is simply
1638 @code{abcd} and not @code{abcde} which you had entered.
1639 This is because the upper bound of the slice is now 4 rather than 5.
1641 @item @emph{Removing a breakpoint}
1643 Toggle the breakpoint icon at line 10.
1645 @item @emph{Resuming execution from a breakpoint}
1647 Select @code{Debug}, then @code{Continue}.
1648 The program will reach the next iteration of the loop, and
1649 wait for input after displaying the prompt.
1650 This time, just hit the @kbd{Enter} key.
1651 The value of @code{N} will be 0, and the program will terminate.
1652 The console window will disappear.
1657 @node The GNAT Compilation Model
1658 @chapter The GNAT Compilation Model
1659 @cindex GNAT compilation model
1660 @cindex Compilation model
1663 * Source Representation::
1664 * Foreign Language Representation::
1665 * File Naming Rules::
1666 * Using Other File Names::
1667 * Alternative File Naming Schemes::
1668 * Generating Object Files::
1669 * Source Dependencies::
1670 * The Ada Library Information Files::
1671 * Binding an Ada Program::
1672 * Mixed Language Programming::
1674 * Building Mixed Ada & C++ Programs::
1675 * Comparison between GNAT and C/C++ Compilation Models::
1677 * Comparison between GNAT and Conventional Ada Library Models::
1679 * Placement of temporary files::
1684 This chapter describes the compilation model used by GNAT. Although
1685 similar to that used by other languages, such as C and C++, this model
1686 is substantially different from the traditional Ada compilation models,
1687 which are based on a library. The model is initially described without
1688 reference to the library-based model. If you have not previously used an
1689 Ada compiler, you need only read the first part of this chapter. The
1690 last section describes and discusses the differences between the GNAT
1691 model and the traditional Ada compiler models. If you have used other
1692 Ada compilers, this section will help you to understand those
1693 differences, and the advantages of the GNAT model.
1695 @node Source Representation
1696 @section Source Representation
1700 Ada source programs are represented in standard text files, using
1701 Latin-1 coding. Latin-1 is an 8-bit code that includes the familiar
1702 7-bit ASCII set, plus additional characters used for
1703 representing foreign languages (@pxref{Foreign Language Representation}
1704 for support of non-USA character sets). The format effector characters
1705 are represented using their standard ASCII encodings, as follows:
1710 Vertical tab, @code{16#0B#}
1714 Horizontal tab, @code{16#09#}
1718 Carriage return, @code{16#0D#}
1722 Line feed, @code{16#0A#}
1726 Form feed, @code{16#0C#}
1730 Source files are in standard text file format. In addition, GNAT will
1731 recognize a wide variety of stream formats, in which the end of
1732 physical lines is marked by any of the following sequences:
1733 @code{LF}, @code{CR}, @code{CR-LF}, or @code{LF-CR}. This is useful
1734 in accommodating files that are imported from other operating systems.
1736 @cindex End of source file
1737 @cindex Source file, end
1739 The end of a source file is normally represented by the physical end of
1740 file. However, the control character @code{16#1A#} (@code{SUB}) is also
1741 recognized as signalling the end of the source file. Again, this is
1742 provided for compatibility with other operating systems where this
1743 code is used to represent the end of file.
1745 Each file contains a single Ada compilation unit, including any pragmas
1746 associated with the unit. For example, this means you must place a
1747 package declaration (a package @dfn{spec}) and the corresponding body in
1748 separate files. An Ada @dfn{compilation} (which is a sequence of
1749 compilation units) is represented using a sequence of files. Similarly,
1750 you will place each subunit or child unit in a separate file.
1752 @node Foreign Language Representation
1753 @section Foreign Language Representation
1756 GNAT supports the standard character sets defined in Ada as well as
1757 several other non-standard character sets for use in localized versions
1758 of the compiler (@pxref{Character Set Control}).
1761 * Other 8-Bit Codes::
1762 * Wide Character Encodings::
1770 The basic character set is Latin-1. This character set is defined by ISO
1771 standard 8859, part 1. The lower half (character codes @code{16#00#}
1772 @dots{} @code{16#7F#)} is identical to standard ASCII coding, but the upper
1773 half is used to represent additional characters. These include extended letters
1774 used by European languages, such as French accents, the vowels with umlauts
1775 used in German, and the extra letter A-ring used in Swedish.
1777 @findex Ada.Characters.Latin_1
1778 For a complete list of Latin-1 codes and their encodings, see the source
1779 file of library unit @code{Ada.Characters.Latin_1} in file
1780 @file{a-chlat1.ads}.
1781 You may use any of these extended characters freely in character or
1782 string literals. In addition, the extended characters that represent
1783 letters can be used in identifiers.
1785 @node Other 8-Bit Codes
1786 @subsection Other 8-Bit Codes
1789 GNAT also supports several other 8-bit coding schemes:
1792 @item ISO 8859-2 (Latin-2)
1795 Latin-2 letters allowed in identifiers, with uppercase and lowercase
1798 @item ISO 8859-3 (Latin-3)
1801 Latin-3 letters allowed in identifiers, with uppercase and lowercase
1804 @item ISO 8859-4 (Latin-4)
1807 Latin-4 letters allowed in identifiers, with uppercase and lowercase
1810 @item ISO 8859-5 (Cyrillic)
1813 ISO 8859-5 letters (Cyrillic) allowed in identifiers, with uppercase and
1814 lowercase equivalence.
1816 @item ISO 8859-15 (Latin-9)
1819 ISO 8859-15 (Latin-9) letters allowed in identifiers, with uppercase and
1820 lowercase equivalence
1822 @item IBM PC (code page 437)
1823 @cindex code page 437
1824 This code page is the normal default for PCs in the U.S. It corresponds
1825 to the original IBM PC character set. This set has some, but not all, of
1826 the extended Latin-1 letters, but these letters do not have the same
1827 encoding as Latin-1. In this mode, these letters are allowed in
1828 identifiers with uppercase and lowercase equivalence.
1830 @item IBM PC (code page 850)
1831 @cindex code page 850
1832 This code page is a modification of 437 extended to include all the
1833 Latin-1 letters, but still not with the usual Latin-1 encoding. In this
1834 mode, all these letters are allowed in identifiers with uppercase and
1835 lowercase equivalence.
1837 @item Full Upper 8-bit
1838 Any character in the range 80-FF allowed in identifiers, and all are
1839 considered distinct. In other words, there are no uppercase and lowercase
1840 equivalences in this range. This is useful in conjunction with
1841 certain encoding schemes used for some foreign character sets (e.g.,
1842 the typical method of representing Chinese characters on the PC).
1845 No upper-half characters in the range 80-FF are allowed in identifiers.
1846 This gives Ada 83 compatibility for identifier names.
1850 For precise data on the encodings permitted, and the uppercase and lowercase
1851 equivalences that are recognized, see the file @file{csets.adb} in
1852 the GNAT compiler sources. You will need to obtain a full source release
1853 of GNAT to obtain this file.
1855 @node Wide Character Encodings
1856 @subsection Wide Character Encodings
1859 GNAT allows wide character codes to appear in character and string
1860 literals, and also optionally in identifiers, by means of the following
1861 possible encoding schemes:
1866 In this encoding, a wide character is represented by the following five
1874 Where @code{a}, @code{b}, @code{c}, @code{d} are the four hexadecimal
1875 characters (using uppercase letters) of the wide character code. For
1876 example, ESC A345 is used to represent the wide character with code
1878 This scheme is compatible with use of the full Wide_Character set.
1880 @item Upper-Half Coding
1881 @cindex Upper-Half Coding
1882 The wide character with encoding @code{16#abcd#} where the upper bit is on
1883 (in other words, ``a'' is in the range 8-F) is represented as two bytes,
1884 @code{16#ab#} and @code{16#cd#}. The second byte cannot be a format control
1885 character, but is not required to be in the upper half. This method can
1886 be also used for shift-JIS or EUC, where the internal coding matches the
1889 @item Shift JIS Coding
1890 @cindex Shift JIS Coding
1891 A wide character is represented by a two-character sequence,
1893 @code{16#cd#}, with the restrictions described for upper-half encoding as
1894 described above. The internal character code is the corresponding JIS
1895 character according to the standard algorithm for Shift-JIS
1896 conversion. Only characters defined in the JIS code set table can be
1897 used with this encoding method.
1901 A wide character is represented by a two-character sequence
1903 @code{16#cd#}, with both characters being in the upper half. The internal
1904 character code is the corresponding JIS character according to the EUC
1905 encoding algorithm. Only characters defined in the JIS code set table
1906 can be used with this encoding method.
1909 A wide character is represented using
1910 UCS Transformation Format 8 (UTF-8) as defined in Annex R of ISO
1911 10646-1/Am.2. Depending on the character value, the representation
1912 is a one, two, or three byte sequence:
1917 16#0000#-16#007f#: 2#0@var{xxxxxxx}#
1918 16#0080#-16#07ff#: 2#110@var{xxxxx}# 2#10@var{xxxxxx}#
1919 16#0800#-16#ffff#: 2#1110@var{xxxx}# 2#10@var{xxxxxx}# 2#10@var{xxxxxx}#
1924 where the @var{xxx} bits correspond to the left-padded bits of the
1925 16-bit character value. Note that all lower half ASCII characters
1926 are represented as ASCII bytes and all upper half characters and
1927 other wide characters are represented as sequences of upper-half
1928 (The full UTF-8 scheme allows for encoding 31-bit characters as
1929 6-byte sequences, but in this implementation, all UTF-8 sequences
1930 of four or more bytes length will be treated as illegal).
1931 @item Brackets Coding
1932 In this encoding, a wide character is represented by the following eight
1940 Where @code{a}, @code{b}, @code{c}, @code{d} are the four hexadecimal
1941 characters (using uppercase letters) of the wide character code. For
1942 example, [``A345''] is used to represent the wide character with code
1943 @code{16#A345#}. It is also possible (though not required) to use the
1944 Brackets coding for upper half characters. For example, the code
1945 @code{16#A3#} can be represented as @code{[``A3'']}.
1947 This scheme is compatible with use of the full Wide_Character set,
1948 and is also the method used for wide character encoding in the standard
1949 ACVC (Ada Compiler Validation Capability) test suite distributions.
1954 Note: Some of these coding schemes do not permit the full use of the
1955 Ada character set. For example, neither Shift JIS, nor EUC allow the
1956 use of the upper half of the Latin-1 set.
1958 @node File Naming Rules
1959 @section File Naming Rules
1962 The default file name is determined by the name of the unit that the
1963 file contains. The name is formed by taking the full expanded name of
1964 the unit and replacing the separating dots with hyphens and using
1965 ^lowercase^uppercase^ for all letters.
1967 An exception arises if the file name generated by the above rules starts
1968 with one of the characters
1970 @samp{A}, @samp{G}, @samp{I}, or @samp{S},
1973 @samp{a}, @samp{g}, @samp{i}, or @samp{s},
1975 and the second character is a
1976 minus. In this case, the character ^tilde^dollar sign^ is used in place
1977 of the minus. The reason for this special rule is to avoid clashes with
1978 the standard names for child units of the packages System, Ada,
1979 Interfaces, and GNAT, which use the prefixes
1981 @samp{S-}, @samp{A-}, @samp{I-}, and @samp{G-},
1984 @samp{s-}, @samp{a-}, @samp{i-}, and @samp{g-},
1988 The file extension is @file{.ads} for a spec and
1989 @file{.adb} for a body. The following list shows some
1990 examples of these rules.
1997 @item arith_functions.ads
1998 Arith_Functions (package spec)
1999 @item arith_functions.adb
2000 Arith_Functions (package body)
2002 Func.Spec (child package spec)
2004 Func.Spec (child package body)
2006 Sub (subunit of Main)
2007 @item ^a~bad.adb^A$BAD.ADB^
2008 A.Bad (child package body)
2012 Following these rules can result in excessively long
2013 file names if corresponding
2014 unit names are long (for example, if child units or subunits are
2015 heavily nested). An option is available to shorten such long file names
2016 (called file name ``krunching''). This may be particularly useful when
2017 programs being developed with GNAT are to be used on operating systems
2018 with limited file name lengths. @xref{Using gnatkr}.
2020 Of course, no file shortening algorithm can guarantee uniqueness over
2021 all possible unit names; if file name krunching is used, it is your
2022 responsibility to ensure no name clashes occur. Alternatively you
2023 can specify the exact file names that you want used, as described
2024 in the next section. Finally, if your Ada programs are migrating from a
2025 compiler with a different naming convention, you can use the gnatchop
2026 utility to produce source files that follow the GNAT naming conventions.
2027 (For details @pxref{Renaming Files Using gnatchop}.)
2029 Note: in the case of @code{Windows NT/XP} or @code{OpenVMS} operating
2030 systems, case is not significant. So for example on @code{Windows XP}
2031 if the canonical name is @code{main-sub.adb}, you can use the file name
2032 @code{Main-Sub.adb} instead. However, case is significant for other
2033 operating systems, so for example, if you want to use other than
2034 canonically cased file names on a Unix system, you need to follow
2035 the procedures described in the next section.
2037 @node Using Other File Names
2038 @section Using Other File Names
2042 In the previous section, we have described the default rules used by
2043 GNAT to determine the file name in which a given unit resides. It is
2044 often convenient to follow these default rules, and if you follow them,
2045 the compiler knows without being explicitly told where to find all
2048 However, in some cases, particularly when a program is imported from
2049 another Ada compiler environment, it may be more convenient for the
2050 programmer to specify which file names contain which units. GNAT allows
2051 arbitrary file names to be used by means of the Source_File_Name pragma.
2052 The form of this pragma is as shown in the following examples:
2053 @cindex Source_File_Name pragma
2055 @smallexample @c ada
2057 pragma Source_File_Name (My_Utilities.Stacks,
2058 Spec_File_Name => "myutilst_a.ada");
2059 pragma Source_File_name (My_Utilities.Stacks,
2060 Body_File_Name => "myutilst.ada");
2065 As shown in this example, the first argument for the pragma is the unit
2066 name (in this example a child unit). The second argument has the form
2067 of a named association. The identifier
2068 indicates whether the file name is for a spec or a body;
2069 the file name itself is given by a string literal.
2071 The source file name pragma is a configuration pragma, which means that
2072 normally it will be placed in the @file{gnat.adc}
2073 file used to hold configuration
2074 pragmas that apply to a complete compilation environment.
2075 For more details on how the @file{gnat.adc} file is created and used
2076 see @ref{Handling of Configuration Pragmas}.
2077 @cindex @file{gnat.adc}
2080 GNAT allows completely arbitrary file names to be specified using the
2081 source file name pragma. However, if the file name specified has an
2082 extension other than @file{.ads} or @file{.adb} it is necessary to use
2083 a special syntax when compiling the file. The name in this case must be
2084 preceded by the special sequence @option{-x} followed by a space and the name
2085 of the language, here @code{ada}, as in:
2088 $ gcc -c -x ada peculiar_file_name.sim
2093 @command{gnatmake} handles non-standard file names in the usual manner (the
2094 non-standard file name for the main program is simply used as the
2095 argument to gnatmake). Note that if the extension is also non-standard,
2096 then it must be included in the @command{gnatmake} command, it may not
2099 @node Alternative File Naming Schemes
2100 @section Alternative File Naming Schemes
2101 @cindex File naming schemes, alternative
2104 In the previous section, we described the use of the @code{Source_File_Name}
2105 pragma to allow arbitrary names to be assigned to individual source files.
2106 However, this approach requires one pragma for each file, and especially in
2107 large systems can result in very long @file{gnat.adc} files, and also create
2108 a maintenance problem.
2110 GNAT also provides a facility for specifying systematic file naming schemes
2111 other than the standard default naming scheme previously described. An
2112 alternative scheme for naming is specified by the use of
2113 @code{Source_File_Name} pragmas having the following format:
2114 @cindex Source_File_Name pragma
2116 @smallexample @c ada
2117 pragma Source_File_Name (
2118 Spec_File_Name => FILE_NAME_PATTERN
2119 @r{[},Casing => CASING_SPEC@r{]}
2120 @r{[},Dot_Replacement => STRING_LITERAL@r{]});
2122 pragma Source_File_Name (
2123 Body_File_Name => FILE_NAME_PATTERN
2124 @r{[},Casing => CASING_SPEC@r{]}
2125 @r{[},Dot_Replacement => STRING_LITERAL@r{]});
2127 pragma Source_File_Name (
2128 Subunit_File_Name => FILE_NAME_PATTERN
2129 @r{[},Casing => CASING_SPEC@r{]}
2130 @r{[},Dot_Replacement => STRING_LITERAL@r{]});
2132 FILE_NAME_PATTERN ::= STRING_LITERAL
2133 CASING_SPEC ::= Lowercase | Uppercase | Mixedcase
2137 The @code{FILE_NAME_PATTERN} string shows how the file name is constructed.
2138 It contains a single asterisk character, and the unit name is substituted
2139 systematically for this asterisk. The optional parameter
2140 @code{Casing} indicates
2141 whether the unit name is to be all upper-case letters, all lower-case letters,
2142 or mixed-case. If no
2143 @code{Casing} parameter is used, then the default is all
2144 ^lower-case^upper-case^.
2146 The optional @code{Dot_Replacement} string is used to replace any periods
2147 that occur in subunit or child unit names. If no @code{Dot_Replacement}
2148 argument is used then separating dots appear unchanged in the resulting
2150 Although the above syntax indicates that the
2151 @code{Casing} argument must appear
2152 before the @code{Dot_Replacement} argument, but it
2153 is also permissible to write these arguments in the opposite order.
2155 As indicated, it is possible to specify different naming schemes for
2156 bodies, specs, and subunits. Quite often the rule for subunits is the
2157 same as the rule for bodies, in which case, there is no need to give
2158 a separate @code{Subunit_File_Name} rule, and in this case the
2159 @code{Body_File_name} rule is used for subunits as well.
2161 The separate rule for subunits can also be used to implement the rather
2162 unusual case of a compilation environment (e.g.@: a single directory) which
2163 contains a subunit and a child unit with the same unit name. Although
2164 both units cannot appear in the same partition, the Ada Reference Manual
2165 allows (but does not require) the possibility of the two units coexisting
2166 in the same environment.
2168 The file name translation works in the following steps:
2173 If there is a specific @code{Source_File_Name} pragma for the given unit,
2174 then this is always used, and any general pattern rules are ignored.
2177 If there is a pattern type @code{Source_File_Name} pragma that applies to
2178 the unit, then the resulting file name will be used if the file exists. If
2179 more than one pattern matches, the latest one will be tried first, and the
2180 first attempt resulting in a reference to a file that exists will be used.
2183 If no pattern type @code{Source_File_Name} pragma that applies to the unit
2184 for which the corresponding file exists, then the standard GNAT default
2185 naming rules are used.
2190 As an example of the use of this mechanism, consider a commonly used scheme
2191 in which file names are all lower case, with separating periods copied
2192 unchanged to the resulting file name, and specs end with @file{.1.ada}, and
2193 bodies end with @file{.2.ada}. GNAT will follow this scheme if the following
2196 @smallexample @c ada
2197 pragma Source_File_Name
2198 (Spec_File_Name => "*.1.ada");
2199 pragma Source_File_Name
2200 (Body_File_Name => "*.2.ada");
2204 The default GNAT scheme is actually implemented by providing the following
2205 default pragmas internally:
2207 @smallexample @c ada
2208 pragma Source_File_Name
2209 (Spec_File_Name => "*.ads", Dot_Replacement => "-");
2210 pragma Source_File_Name
2211 (Body_File_Name => "*.adb", Dot_Replacement => "-");
2215 Our final example implements a scheme typically used with one of the
2216 Ada 83 compilers, where the separator character for subunits was ``__''
2217 (two underscores), specs were identified by adding @file{_.ADA}, bodies
2218 by adding @file{.ADA}, and subunits by
2219 adding @file{.SEP}. All file names were
2220 upper case. Child units were not present of course since this was an
2221 Ada 83 compiler, but it seems reasonable to extend this scheme to use
2222 the same double underscore separator for child units.
2224 @smallexample @c ada
2225 pragma Source_File_Name
2226 (Spec_File_Name => "*_.ADA",
2227 Dot_Replacement => "__",
2228 Casing = Uppercase);
2229 pragma Source_File_Name
2230 (Body_File_Name => "*.ADA",
2231 Dot_Replacement => "__",
2232 Casing = Uppercase);
2233 pragma Source_File_Name
2234 (Subunit_File_Name => "*.SEP",
2235 Dot_Replacement => "__",
2236 Casing = Uppercase);
2239 @node Generating Object Files
2240 @section Generating Object Files
2243 An Ada program consists of a set of source files, and the first step in
2244 compiling the program is to generate the corresponding object files.
2245 These are generated by compiling a subset of these source files.
2246 The files you need to compile are the following:
2250 If a package spec has no body, compile the package spec to produce the
2251 object file for the package.
2254 If a package has both a spec and a body, compile the body to produce the
2255 object file for the package. The source file for the package spec need
2256 not be compiled in this case because there is only one object file, which
2257 contains the code for both the spec and body of the package.
2260 For a subprogram, compile the subprogram body to produce the object file
2261 for the subprogram. The spec, if one is present, is as usual in a
2262 separate file, and need not be compiled.
2266 In the case of subunits, only compile the parent unit. A single object
2267 file is generated for the entire subunit tree, which includes all the
2271 Compile child units independently of their parent units
2272 (though, of course, the spec of all the ancestor unit must be present in order
2273 to compile a child unit).
2277 Compile generic units in the same manner as any other units. The object
2278 files in this case are small dummy files that contain at most the
2279 flag used for elaboration checking. This is because GNAT always handles generic
2280 instantiation by means of macro expansion. However, it is still necessary to
2281 compile generic units, for dependency checking and elaboration purposes.
2285 The preceding rules describe the set of files that must be compiled to
2286 generate the object files for a program. Each object file has the same
2287 name as the corresponding source file, except that the extension is
2290 You may wish to compile other files for the purpose of checking their
2291 syntactic and semantic correctness. For example, in the case where a
2292 package has a separate spec and body, you would not normally compile the
2293 spec. However, it is convenient in practice to compile the spec to make
2294 sure it is error-free before compiling clients of this spec, because such
2295 compilations will fail if there is an error in the spec.
2297 GNAT provides an option for compiling such files purely for the
2298 purposes of checking correctness; such compilations are not required as
2299 part of the process of building a program. To compile a file in this
2300 checking mode, use the @option{-gnatc} switch.
2302 @node Source Dependencies
2303 @section Source Dependencies
2306 A given object file clearly depends on the source file which is compiled
2307 to produce it. Here we are using @dfn{depends} in the sense of a typical
2308 @code{make} utility; in other words, an object file depends on a source
2309 file if changes to the source file require the object file to be
2311 In addition to this basic dependency, a given object may depend on
2312 additional source files as follows:
2316 If a file being compiled @code{with}'s a unit @var{X}, the object file
2317 depends on the file containing the spec of unit @var{X}. This includes
2318 files that are @code{with}'ed implicitly either because they are parents
2319 of @code{with}'ed child units or they are run-time units required by the
2320 language constructs used in a particular unit.
2323 If a file being compiled instantiates a library level generic unit, the
2324 object file depends on both the spec and body files for this generic
2328 If a file being compiled instantiates a generic unit defined within a
2329 package, the object file depends on the body file for the package as
2330 well as the spec file.
2334 @cindex @option{-gnatn} switch
2335 If a file being compiled contains a call to a subprogram for which
2336 pragma @code{Inline} applies and inlining is activated with the
2337 @option{-gnatn} switch, the object file depends on the file containing the
2338 body of this subprogram as well as on the file containing the spec. Note
2339 that for inlining to actually occur as a result of the use of this switch,
2340 it is necessary to compile in optimizing mode.
2342 @cindex @option{-gnatN} switch
2343 The use of @option{-gnatN} activates inlining optimization
2344 that is performed by the front end of the compiler. This inlining does
2345 not require that the code generation be optimized. Like @option{-gnatn},
2346 the use of this switch generates additional dependencies.
2348 When using a gcc-based back end (in practice this means using any version
2349 of GNAT other than the JGNAT, .NET or GNAAMP versions), then the use of
2350 @option{-gnatN} is deprecated, and the use of @option{-gnatn} is preferred.
2351 Historically front end inlining was more extensive than the gcc back end
2352 inlining, but that is no longer the case.
2355 If an object file @file{O} depends on the proper body of a subunit through
2356 inlining or instantiation, it depends on the parent unit of the subunit.
2357 This means that any modification of the parent unit or one of its subunits
2358 affects the compilation of @file{O}.
2361 The object file for a parent unit depends on all its subunit body files.
2364 The previous two rules meant that for purposes of computing dependencies and
2365 recompilation, a body and all its subunits are treated as an indivisible whole.
2368 These rules are applied transitively: if unit @code{A} @code{with}'s
2369 unit @code{B}, whose elaboration calls an inlined procedure in package
2370 @code{C}, the object file for unit @code{A} will depend on the body of
2371 @code{C}, in file @file{c.adb}.
2373 The set of dependent files described by these rules includes all the
2374 files on which the unit is semantically dependent, as dictated by the
2375 Ada language standard. However, it is a superset of what the
2376 standard describes, because it includes generic, inline, and subunit
2379 An object file must be recreated by recompiling the corresponding source
2380 file if any of the source files on which it depends are modified. For
2381 example, if the @code{make} utility is used to control compilation,
2382 the rule for an Ada object file must mention all the source files on
2383 which the object file depends, according to the above definition.
2384 The determination of the necessary
2385 recompilations is done automatically when one uses @command{gnatmake}.
2388 @node The Ada Library Information Files
2389 @section The Ada Library Information Files
2390 @cindex Ada Library Information files
2391 @cindex @file{ALI} files
2394 Each compilation actually generates two output files. The first of these
2395 is the normal object file that has a @file{.o} extension. The second is a
2396 text file containing full dependency information. It has the same
2397 name as the source file, but an @file{.ali} extension.
2398 This file is known as the Ada Library Information (@file{ALI}) file.
2399 The following information is contained in the @file{ALI} file.
2403 Version information (indicates which version of GNAT was used to compile
2404 the unit(s) in question)
2407 Main program information (including priority and time slice settings,
2408 as well as the wide character encoding used during compilation).
2411 List of arguments used in the @command{gcc} command for the compilation
2414 Attributes of the unit, including configuration pragmas used, an indication
2415 of whether the compilation was successful, exception model used etc.
2418 A list of relevant restrictions applying to the unit (used for consistency)
2422 Categorization information (e.g.@: use of pragma @code{Pure}).
2425 Information on all @code{with}'ed units, including presence of
2426 @code{Elaborate} or @code{Elaborate_All} pragmas.
2429 Information from any @code{Linker_Options} pragmas used in the unit
2432 Information on the use of @code{Body_Version} or @code{Version}
2433 attributes in the unit.
2436 Dependency information. This is a list of files, together with
2437 time stamp and checksum information. These are files on which
2438 the unit depends in the sense that recompilation is required
2439 if any of these units are modified.
2442 Cross-reference data. Contains information on all entities referenced
2443 in the unit. Used by tools like @code{gnatxref} and @code{gnatfind} to
2444 provide cross-reference information.
2449 For a full detailed description of the format of the @file{ALI} file,
2450 see the source of the body of unit @code{Lib.Writ}, contained in file
2451 @file{lib-writ.adb} in the GNAT compiler sources.
2453 @node Binding an Ada Program
2454 @section Binding an Ada Program
2457 When using languages such as C and C++, once the source files have been
2458 compiled the only remaining step in building an executable program
2459 is linking the object modules together. This means that it is possible to
2460 link an inconsistent version of a program, in which two units have
2461 included different versions of the same header.
2463 The rules of Ada do not permit such an inconsistent program to be built.
2464 For example, if two clients have different versions of the same package,
2465 it is illegal to build a program containing these two clients.
2466 These rules are enforced by the GNAT binder, which also determines an
2467 elaboration order consistent with the Ada rules.
2469 The GNAT binder is run after all the object files for a program have
2470 been created. It is given the name of the main program unit, and from
2471 this it determines the set of units required by the program, by reading the
2472 corresponding ALI files. It generates error messages if the program is
2473 inconsistent or if no valid order of elaboration exists.
2475 If no errors are detected, the binder produces a main program, in Ada by
2476 default, that contains calls to the elaboration procedures of those
2477 compilation unit that require them, followed by
2478 a call to the main program. This Ada program is compiled to generate the
2479 object file for the main program. The name of
2480 the Ada file is @file{b~@var{xxx}.adb} (with the corresponding spec
2481 @file{b~@var{xxx}.ads}) where @var{xxx} is the name of the
2484 Finally, the linker is used to build the resulting executable program,
2485 using the object from the main program from the bind step as well as the
2486 object files for the Ada units of the program.
2488 @node Mixed Language Programming
2489 @section Mixed Language Programming
2490 @cindex Mixed Language Programming
2493 This section describes how to develop a mixed-language program,
2494 specifically one that comprises units in both Ada and C.
2497 * Interfacing to C::
2498 * Calling Conventions::
2501 @node Interfacing to C
2502 @subsection Interfacing to C
2504 Interfacing Ada with a foreign language such as C involves using
2505 compiler directives to import and/or export entity definitions in each
2506 language---using @code{extern} statements in C, for instance, and the
2507 @code{Import}, @code{Export}, and @code{Convention} pragmas in Ada.
2508 A full treatment of these topics is provided in Appendix B, section 1
2509 of the Ada Reference Manual.
2511 There are two ways to build a program using GNAT that contains some Ada
2512 sources and some foreign language sources, depending on whether or not
2513 the main subprogram is written in Ada. Here is a source example with
2514 the main subprogram in Ada:
2520 void print_num (int num)
2522 printf ("num is %d.\n", num);
2528 /* num_from_Ada is declared in my_main.adb */
2529 extern int num_from_Ada;
2533 return num_from_Ada;
2537 @smallexample @c ada
2539 procedure My_Main is
2541 -- Declare then export an Integer entity called num_from_Ada
2542 My_Num : Integer := 10;
2543 pragma Export (C, My_Num, "num_from_Ada");
2545 -- Declare an Ada function spec for Get_Num, then use
2546 -- C function get_num for the implementation.
2547 function Get_Num return Integer;
2548 pragma Import (C, Get_Num, "get_num");
2550 -- Declare an Ada procedure spec for Print_Num, then use
2551 -- C function print_num for the implementation.
2552 procedure Print_Num (Num : Integer);
2553 pragma Import (C, Print_Num, "print_num");
2556 Print_Num (Get_Num);
2562 To build this example, first compile the foreign language files to
2563 generate object files:
2565 ^gcc -c file1.c^gcc -c FILE1.C^
2566 ^gcc -c file2.c^gcc -c FILE2.C^
2570 Then, compile the Ada units to produce a set of object files and ALI
2573 gnatmake ^-c^/ACTIONS=COMPILE^ my_main.adb
2577 Run the Ada binder on the Ada main program:
2579 gnatbind my_main.ali
2583 Link the Ada main program, the Ada objects and the other language
2586 gnatlink my_main.ali file1.o file2.o
2590 The last three steps can be grouped in a single command:
2592 gnatmake my_main.adb -largs file1.o file2.o
2595 @cindex Binder output file
2597 If the main program is in a language other than Ada, then you may have
2598 more than one entry point into the Ada subsystem. You must use a special
2599 binder option to generate callable routines that initialize and
2600 finalize the Ada units (@pxref{Binding with Non-Ada Main Programs}).
2601 Calls to the initialization and finalization routines must be inserted
2602 in the main program, or some other appropriate point in the code. The
2603 call to initialize the Ada units must occur before the first Ada
2604 subprogram is called, and the call to finalize the Ada units must occur
2605 after the last Ada subprogram returns. The binder will place the
2606 initialization and finalization subprograms into the
2607 @file{b~@var{xxx}.adb} file where they can be accessed by your C
2608 sources. To illustrate, we have the following example:
2612 extern void adainit (void);
2613 extern void adafinal (void);
2614 extern int add (int, int);
2615 extern int sub (int, int);
2617 int main (int argc, char *argv[])
2623 /* Should print "21 + 7 = 28" */
2624 printf ("%d + %d = %d\n", a, b, add (a, b));
2625 /* Should print "21 - 7 = 14" */
2626 printf ("%d - %d = %d\n", a, b, sub (a, b));
2632 @smallexample @c ada
2635 function Add (A, B : Integer) return Integer;
2636 pragma Export (C, Add, "add");
2640 package body Unit1 is
2641 function Add (A, B : Integer) return Integer is
2649 function Sub (A, B : Integer) return Integer;
2650 pragma Export (C, Sub, "sub");
2654 package body Unit2 is
2655 function Sub (A, B : Integer) return Integer is
2664 The build procedure for this application is similar to the last
2665 example's. First, compile the foreign language files to generate object
2668 ^gcc -c main.c^gcc -c main.c^
2672 Next, compile the Ada units to produce a set of object files and ALI
2675 gnatmake ^-c^/ACTIONS=COMPILE^ unit1.adb
2676 gnatmake ^-c^/ACTIONS=COMPILE^ unit2.adb
2680 Run the Ada binder on every generated ALI file. Make sure to use the
2681 @option{-n} option to specify a foreign main program:
2683 gnatbind ^-n^/NOMAIN^ unit1.ali unit2.ali
2687 Link the Ada main program, the Ada objects and the foreign language
2688 objects. You need only list the last ALI file here:
2690 gnatlink unit2.ali main.o -o exec_file
2693 This procedure yields a binary executable called @file{exec_file}.
2697 Depending on the circumstances (for example when your non-Ada main object
2698 does not provide symbol @code{main}), you may also need to instruct the
2699 GNAT linker not to include the standard startup objects by passing the
2700 @option{^-nostartfiles^/NOSTART_FILES^} switch to @command{gnatlink}.
2702 @node Calling Conventions
2703 @subsection Calling Conventions
2704 @cindex Foreign Languages
2705 @cindex Calling Conventions
2706 GNAT follows standard calling sequence conventions and will thus interface
2707 to any other language that also follows these conventions. The following
2708 Convention identifiers are recognized by GNAT:
2711 @cindex Interfacing to Ada
2712 @cindex Other Ada compilers
2713 @cindex Convention Ada
2715 This indicates that the standard Ada calling sequence will be
2716 used and all Ada data items may be passed without any limitations in the
2717 case where GNAT is used to generate both the caller and callee. It is also
2718 possible to mix GNAT generated code and code generated by another Ada
2719 compiler. In this case, the data types should be restricted to simple
2720 cases, including primitive types. Whether complex data types can be passed
2721 depends on the situation. Probably it is safe to pass simple arrays, such
2722 as arrays of integers or floats. Records may or may not work, depending
2723 on whether both compilers lay them out identically. Complex structures
2724 involving variant records, access parameters, tasks, or protected types,
2725 are unlikely to be able to be passed.
2727 Note that in the case of GNAT running
2728 on a platform that supports HP Ada 83, a higher degree of compatibility
2729 can be guaranteed, and in particular records are layed out in an identical
2730 manner in the two compilers. Note also that if output from two different
2731 compilers is mixed, the program is responsible for dealing with elaboration
2732 issues. Probably the safest approach is to write the main program in the
2733 version of Ada other than GNAT, so that it takes care of its own elaboration
2734 requirements, and then call the GNAT-generated adainit procedure to ensure
2735 elaboration of the GNAT components. Consult the documentation of the other
2736 Ada compiler for further details on elaboration.
2738 However, it is not possible to mix the tasking run time of GNAT and
2739 HP Ada 83, All the tasking operations must either be entirely within
2740 GNAT compiled sections of the program, or entirely within HP Ada 83
2741 compiled sections of the program.
2743 @cindex Interfacing to Assembly
2744 @cindex Convention Assembler
2746 Specifies assembler as the convention. In practice this has the
2747 same effect as convention Ada (but is not equivalent in the sense of being
2748 considered the same convention).
2750 @cindex Convention Asm
2753 Equivalent to Assembler.
2755 @cindex Interfacing to COBOL
2756 @cindex Convention COBOL
2759 Data will be passed according to the conventions described
2760 in section B.4 of the Ada Reference Manual.
2763 @cindex Interfacing to C
2764 @cindex Convention C
2766 Data will be passed according to the conventions described
2767 in section B.3 of the Ada Reference Manual.
2769 A note on interfacing to a C ``varargs'' function:
2770 @findex C varargs function
2771 @cindex Interfacing to C varargs function
2772 @cindex varargs function interfaces
2776 In C, @code{varargs} allows a function to take a variable number of
2777 arguments. There is no direct equivalent in this to Ada. One
2778 approach that can be used is to create a C wrapper for each
2779 different profile and then interface to this C wrapper. For
2780 example, to print an @code{int} value using @code{printf},
2781 create a C function @code{printfi} that takes two arguments, a
2782 pointer to a string and an int, and calls @code{printf}.
2783 Then in the Ada program, use pragma @code{Import} to
2784 interface to @code{printfi}.
2787 It may work on some platforms to directly interface to
2788 a @code{varargs} function by providing a specific Ada profile
2789 for a particular call. However, this does not work on
2790 all platforms, since there is no guarantee that the
2791 calling sequence for a two argument normal C function
2792 is the same as for calling a @code{varargs} C function with
2793 the same two arguments.
2796 @cindex Convention Default
2801 @cindex Convention External
2808 @cindex Interfacing to C++
2809 @cindex Convention C++
2810 @item C_Plus_Plus (or CPP)
2811 This stands for C++. For most purposes this is identical to C.
2812 See the separate description of the specialized GNAT pragmas relating to
2813 C++ interfacing for further details.
2817 @cindex Interfacing to Fortran
2818 @cindex Convention Fortran
2820 Data will be passed according to the conventions described
2821 in section B.5 of the Ada Reference Manual.
2824 This applies to an intrinsic operation, as defined in the Ada
2825 Reference Manual. If a pragma Import (Intrinsic) applies to a subprogram,
2826 this means that the body of the subprogram is provided by the compiler itself,
2827 usually by means of an efficient code sequence, and that the user does not
2828 supply an explicit body for it. In an application program, the pragma may
2829 be applied to the following sets of names:
2833 Rotate_Left, Rotate_Right, Shift_Left, Shift_Right,
2834 Shift_Right_Arithmetic. The corresponding subprogram declaration must have
2835 two formal parameters. The
2836 first one must be a signed integer type or a modular type with a binary
2837 modulus, and the second parameter must be of type Natural.
2838 The return type must be the same as the type of the first argument. The size
2839 of this type can only be 8, 16, 32, or 64.
2842 Binary arithmetic operators: ``+'', ``-'', ``*'', ``/''
2843 The corresponding operator declaration must have parameters and result type
2844 that have the same root numeric type (for example, all three are long_float
2845 types). This simplifies the definition of operations that use type checking
2846 to perform dimensional checks:
2848 @smallexample @c ada
2849 type Distance is new Long_Float;
2850 type Time is new Long_Float;
2851 type Velocity is new Long_Float;
2852 function "/" (D : Distance; T : Time)
2854 pragma Import (Intrinsic, "/");
2858 This common idiom is often programmed with a generic definition and an
2859 explicit body. The pragma makes it simpler to introduce such declarations.
2860 It incurs no overhead in compilation time or code size, because it is
2861 implemented as a single machine instruction.
2864 General subprogram entities, to bind an Ada subprogram declaration to
2865 a compiler builtin by name with back-ends where such interfaces are
2866 available. A typical example is the set of ``__builtin'' functions
2867 exposed by the GCC back-end, as in the following example:
2869 @smallexample @c ada
2870 function builtin_sqrt (F : Float) return Float;
2871 pragma Import (Intrinsic, builtin_sqrt, "__builtin_sqrtf");
2874 Most of the GCC builtins are accessible this way, and as for other
2875 import conventions (e.g. C), it is the user's responsibility to ensure
2876 that the Ada subprogram profile matches the underlying builtin
2884 @cindex Convention Stdcall
2886 This is relevant only to Windows XP/2000/NT implementations of GNAT,
2887 and specifies that the @code{Stdcall} calling sequence will be used,
2888 as defined by the NT API. Nevertheless, to ease building
2889 cross-platform bindings this convention will be handled as a @code{C} calling
2890 convention on non-Windows platforms.
2893 @cindex Convention DLL
2895 This is equivalent to @code{Stdcall}.
2898 @cindex Convention Win32
2900 This is equivalent to @code{Stdcall}.
2904 @cindex Convention Stubbed
2906 This is a special convention that indicates that the compiler
2907 should provide a stub body that raises @code{Program_Error}.
2911 GNAT additionally provides a useful pragma @code{Convention_Identifier}
2912 that can be used to parameterize conventions and allow additional synonyms
2913 to be specified. For example if you have legacy code in which the convention
2914 identifier Fortran77 was used for Fortran, you can use the configuration
2917 @smallexample @c ada
2918 pragma Convention_Identifier (Fortran77, Fortran);
2922 And from now on the identifier Fortran77 may be used as a convention
2923 identifier (for example in an @code{Import} pragma) with the same
2927 @node Building Mixed Ada & C++ Programs
2928 @section Building Mixed Ada and C++ Programs
2931 A programmer inexperienced with mixed-language development may find that
2932 building an application containing both Ada and C++ code can be a
2933 challenge. This section gives a few
2934 hints that should make this task easier. The first section addresses
2935 the differences between interfacing with C and interfacing with C++.
2937 looks into the delicate problem of linking the complete application from
2938 its Ada and C++ parts. The last section gives some hints on how the GNAT
2939 run-time library can be adapted in order to allow inter-language dispatching
2940 with a new C++ compiler.
2943 * Interfacing to C++::
2944 * Linking a Mixed C++ & Ada Program::
2945 * A Simple Example::
2946 * Interfacing with C++ constructors::
2947 * Interfacing with C++ at the Class Level::
2950 @node Interfacing to C++
2951 @subsection Interfacing to C++
2954 GNAT supports interfacing with the G++ compiler (or any C++ compiler
2955 generating code that is compatible with the G++ Application Binary
2956 Interface ---see http://www.codesourcery.com/archives/cxx-abi).
2959 Interfacing can be done at 3 levels: simple data, subprograms, and
2960 classes. In the first two cases, GNAT offers a specific @code{Convention
2961 C_Plus_Plus} (or @code{CPP}) that behaves exactly like @code{Convention C}.
2962 Usually, C++ mangles the names of subprograms. To generate proper mangled
2963 names automatically, see @ref{Generating Ada Bindings for C and C++ headers}).
2964 This problem can also be addressed manually in two ways:
2968 by modifying the C++ code in order to force a C convention using
2969 the @code{extern "C"} syntax.
2972 by figuring out the mangled name (using e.g. @command{nm}) and using it as the
2973 Link_Name argument of the pragma import.
2977 Interfacing at the class level can be achieved by using the GNAT specific
2978 pragmas such as @code{CPP_Constructor}. @xref{Interfacing to C++,,,
2979 gnat_rm, GNAT Reference Manual}, for additional information.
2981 @node Linking a Mixed C++ & Ada Program
2982 @subsection Linking a Mixed C++ & Ada Program
2985 Usually the linker of the C++ development system must be used to link
2986 mixed applications because most C++ systems will resolve elaboration
2987 issues (such as calling constructors on global class instances)
2988 transparently during the link phase. GNAT has been adapted to ease the
2989 use of a foreign linker for the last phase. Three cases can be
2994 Using GNAT and G++ (GNU C++ compiler) from the same GCC installation:
2995 The C++ linker can simply be called by using the C++ specific driver
2998 Note that if the C++ code uses inline functions, you will need to
2999 compile your C++ code with the @code{-fkeep-inline-functions} switch in
3000 order to provide an existing function implementation that the Ada code can
3004 $ g++ -c -fkeep-inline-functions file1.C
3005 $ g++ -c -fkeep-inline-functions file2.C
3006 $ gnatmake ada_unit -largs file1.o file2.o --LINK=g++
3010 Using GNAT and G++ from two different GCC installations: If both
3011 compilers are on the @env{PATH}, the previous method may be used. It is
3012 important to note that environment variables such as
3013 @env{C_INCLUDE_PATH}, @env{GCC_EXEC_PREFIX}, @env{BINUTILS_ROOT}, and
3014 @env{GCC_ROOT} will affect both compilers
3015 at the same time and may make one of the two compilers operate
3016 improperly if set during invocation of the wrong compiler. It is also
3017 very important that the linker uses the proper @file{libgcc.a} GCC
3018 library -- that is, the one from the C++ compiler installation. The
3019 implicit link command as suggested in the @command{gnatmake} command
3020 from the former example can be replaced by an explicit link command with
3021 the full-verbosity option in order to verify which library is used:
3024 $ gnatlink -v -v ada_unit file1.o file2.o --LINK=c++
3026 If there is a problem due to interfering environment variables, it can
3027 be worked around by using an intermediate script. The following example
3028 shows the proper script to use when GNAT has not been installed at its
3029 default location and g++ has been installed at its default location:
3037 $ gnatlink -v -v ada_unit file1.o file2.o --LINK=./my_script
3041 Using a non-GNU C++ compiler: The commands previously described can be
3042 used to insure that the C++ linker is used. Nonetheless, you need to add
3043 a few more parameters to the link command line, depending on the exception
3046 If the @code{setjmp/longjmp} exception mechanism is used, only the paths
3047 to the libgcc libraries are required:
3052 CC $* `gcc -print-file-name=libgcc.a` `gcc -print-file-name=libgcc_eh.a`
3053 $ gnatlink ada_unit file1.o file2.o --LINK=./my_script
3056 Where CC is the name of the non-GNU C++ compiler.
3058 If the @code{zero cost} exception mechanism is used, and the platform
3059 supports automatic registration of exception tables (e.g.@: Solaris or IRIX),
3060 paths to more objects are required:
3065 CC `gcc -print-file-name=crtbegin.o` $* \
3066 `gcc -print-file-name=libgcc.a` `gcc -print-file-name=libgcc_eh.a` \
3067 `gcc -print-file-name=crtend.o`
3068 $ gnatlink ada_unit file1.o file2.o --LINK=./my_script
3071 If the @code{zero cost} exception mechanism is used, and the platform
3072 doesn't support automatic registration of exception tables (e.g.@: HP-UX,
3073 Tru64 or AIX), the simple approach described above will not work and
3074 a pre-linking phase using GNAT will be necessary.
3078 Another alternative is to use the @command{gprbuild} multi-language builder
3079 which has a large knowledge base and knows how to link Ada and C++ code
3080 together automatically in most cases.
3082 @node A Simple Example
3083 @subsection A Simple Example
3085 The following example, provided as part of the GNAT examples, shows how
3086 to achieve procedural interfacing between Ada and C++ in both
3087 directions. The C++ class A has two methods. The first method is exported
3088 to Ada by the means of an extern C wrapper function. The second method
3089 calls an Ada subprogram. On the Ada side, The C++ calls are modelled by
3090 a limited record with a layout comparable to the C++ class. The Ada
3091 subprogram, in turn, calls the C++ method. So, starting from the C++
3092 main program, the process passes back and forth between the two
3096 Here are the compilation commands:
3098 $ gnatmake -c simple_cpp_interface
3101 $ gnatbind -n simple_cpp_interface
3102 $ gnatlink simple_cpp_interface -o cpp_main --LINK=g++
3103 -lstdc++ ex7.o cpp_main.o
3107 Here are the corresponding sources:
3115 void adainit (void);
3116 void adafinal (void);
3117 void method1 (A *t);
3139 class A : public Origin @{
3141 void method1 (void);
3142 void method2 (int v);
3152 extern "C" @{ void ada_method2 (A *t, int v);@}
3154 void A::method1 (void)
3157 printf ("in A::method1, a_value = %d \n",a_value);
3161 void A::method2 (int v)
3163 ada_method2 (this, v);
3164 printf ("in A::method2, a_value = %d \n",a_value);
3171 printf ("in A::A, a_value = %d \n",a_value);
3175 @smallexample @c ada
3177 package body Simple_Cpp_Interface is
3179 procedure Ada_Method2 (This : in out A; V : Integer) is
3185 end Simple_Cpp_Interface;
3188 package Simple_Cpp_Interface is
3191 Vptr : System.Address;
3195 pragma Convention (C, A);
3197 procedure Method1 (This : in out A);
3198 pragma Import (C, Method1);
3200 procedure Ada_Method2 (This : in out A; V : Integer);
3201 pragma Export (C, Ada_Method2);
3203 end Simple_Cpp_Interface;
3206 @node Interfacing with C++ constructors
3207 @subsection Interfacing with C++ constructors
3210 In order to interface with C++ constructors GNAT provides the
3211 @code{pragma CPP_Constructor} (@xref{Interfacing to C++,,,
3212 gnat_rm, GNAT Reference Manual}, for additional information).
3213 In this section we present some common uses of C++ constructors
3214 in mixed-languages programs in GNAT.
3216 Let us assume that we need to interface with the following
3224 @b{virtual} int Get_Value ();
3225 Root(); // Default constructor
3226 Root(int v); // 1st non-default constructor
3227 Root(int v, int w); // 2nd non-default constructor
3231 For this purpose we can write the following package spec (further
3232 information on how to build this spec is available in
3233 @ref{Interfacing with C++ at the Class Level} and
3234 @ref{Generating Ada Bindings for C and C++ headers}).
3236 @smallexample @c ada
3237 with Interfaces.C; use Interfaces.C;
3239 type Root is tagged limited record
3243 pragma Import (CPP, Root);
3245 function Get_Value (Obj : Root) return int;
3246 pragma Import (CPP, Get_Value);
3248 function Constructor return Root;
3249 pragma Cpp_Constructor (Constructor, "_ZN4RootC1Ev");
3251 function Constructor (v : Integer) return Root;
3252 pragma Cpp_Constructor (Constructor, "_ZN4RootC1Ei");
3254 function Constructor (v, w : Integer) return Root;
3255 pragma Cpp_Constructor (Constructor, "_ZN4RootC1Eii");
3259 On the Ada side the constructor is represented by a function (whose
3260 name is arbitrary) that returns the classwide type corresponding to
3261 the imported C++ class. Although the constructor is described as a
3262 function, it is typically a procedure with an extra implicit argument
3263 (the object being initialized) at the implementation level. GNAT
3264 issues the appropriate call, whatever it is, to get the object
3265 properly initialized.
3267 Constructors can only appear in the following contexts:
3271 On the right side of an initialization of an object of type @var{T}.
3273 On the right side of an initialization of a record component of type @var{T}.
3275 In an Ada 2005 limited aggregate.
3277 In an Ada 2005 nested limited aggregate.
3279 In an Ada 2005 limited aggregate that initializes an object built in
3280 place by an extended return statement.
3284 In a declaration of an object whose type is a class imported from C++,
3285 either the default C++ constructor is implicitly called by GNAT, or
3286 else the required C++ constructor must be explicitly called in the
3287 expression that initializes the object. For example:
3289 @smallexample @c ada
3291 Obj2 : Root := Constructor;
3292 Obj3 : Root := Constructor (v => 10);
3293 Obj4 : Root := Constructor (30, 40);
3296 The first two declarations are equivalent: in both cases the default C++
3297 constructor is invoked (in the former case the call to the constructor is
3298 implicit, and in the latter case the call is explicit in the object
3299 declaration). @code{Obj3} is initialized by the C++ non-default constructor
3300 that takes an integer argument, and @code{Obj4} is initialized by the
3301 non-default C++ constructor that takes two integers.
3303 Let us derive the imported C++ class in the Ada side. For example:
3305 @smallexample @c ada
3306 type DT is new Root with record
3307 C_Value : Natural := 2009;
3311 In this case the components DT inherited from the C++ side must be
3312 initialized by a C++ constructor, and the additional Ada components
3313 of type DT are initialized by GNAT. The initialization of such an
3314 object is done either by default, or by means of a function returning
3315 an aggregate of type DT, or by means of an extension aggregate.
3317 @smallexample @c ada
3319 Obj6 : DT := Function_Returning_DT (50);
3320 Obj7 : DT := (Constructor (30,40) with C_Value => 50);
3323 The declaration of @code{Obj5} invokes the default constructors: the
3324 C++ default constructor of the parent type takes care of the initialization
3325 of the components inherited from Root, and GNAT takes care of the default
3326 initialization of the additional Ada components of type DT (that is,
3327 @code{C_Value} is initialized to value 2009). The order of invocation of
3328 the constructors is consistent with the order of elaboration required by
3329 Ada and C++. That is, the constructor of the parent type is always called
3330 before the constructor of the derived type.
3332 Let us now consider a record that has components whose type is imported
3333 from C++. For example:
3335 @smallexample @c ada
3336 type Rec1 is limited record
3337 Data1 : Root := Constructor (10);
3338 Value : Natural := 1000;
3341 type Rec2 (D : Integer := 20) is limited record
3343 Data2 : Root := Constructor (D, 30);
3347 The initialization of an object of type @code{Rec2} will call the
3348 non-default C++ constructors specified for the imported components.
3351 @smallexample @c ada
3355 Using Ada 2005 we can use limited aggregates to initialize an object
3356 invoking C++ constructors that differ from those specified in the type
3357 declarations. For example:
3359 @smallexample @c ada
3360 Obj9 : Rec2 := (Rec => (Data1 => Constructor (15, 16),
3365 The above declaration uses an Ada 2005 limited aggregate to
3366 initialize @code{Obj9}, and the C++ constructor that has two integer
3367 arguments is invoked to initialize the @code{Data1} component instead
3368 of the constructor specified in the declaration of type @code{Rec1}. In
3369 Ada 2005 the box in the aggregate indicates that unspecified components
3370 are initialized using the expression (if any) available in the component
3371 declaration. That is, in this case discriminant @code{D} is initialized
3372 to value @code{20}, @code{Value} is initialized to value 1000, and the
3373 non-default C++ constructor that handles two integers takes care of
3374 initializing component @code{Data2} with values @code{20,30}.
3376 In Ada 2005 we can use the extended return statement to build the Ada
3377 equivalent to C++ non-default constructors. For example:
3379 @smallexample @c ada
3380 function Constructor (V : Integer) return Rec2 is
3382 return Obj : Rec2 := (Rec => (Data1 => Constructor (V, 20),
3385 -- Further actions required for construction of
3386 -- objects of type Rec2
3392 In this example the extended return statement construct is used to
3393 build in place the returned object whose components are initialized
3394 by means of a limited aggregate. Any further action associated with
3395 the constructor can be placed inside the construct.
3397 @node Interfacing with C++ at the Class Level
3398 @subsection Interfacing with C++ at the Class Level
3400 In this section we demonstrate the GNAT features for interfacing with
3401 C++ by means of an example making use of Ada 2005 abstract interface
3402 types. This example consists of a classification of animals; classes
3403 have been used to model our main classification of animals, and
3404 interfaces provide support for the management of secondary
3405 classifications. We first demonstrate a case in which the types and
3406 constructors are defined on the C++ side and imported from the Ada
3407 side, and latter the reverse case.
3409 The root of our derivation will be the @code{Animal} class, with a
3410 single private attribute (the @code{Age} of the animal) and two public
3411 primitives to set and get the value of this attribute.
3416 @b{virtual} void Set_Age (int New_Age);
3417 @b{virtual} int Age ();
3423 Abstract interface types are defined in C++ by means of classes with pure
3424 virtual functions and no data members. In our example we will use two
3425 interfaces that provide support for the common management of @code{Carnivore}
3426 and @code{Domestic} animals:
3429 @b{class} Carnivore @{
3431 @b{virtual} int Number_Of_Teeth () = 0;
3434 @b{class} Domestic @{
3436 @b{virtual void} Set_Owner (char* Name) = 0;
3440 Using these declarations, we can now say that a @code{Dog} is an animal that is
3441 both Carnivore and Domestic, that is:
3444 @b{class} Dog : Animal, Carnivore, Domestic @{
3446 @b{virtual} int Number_Of_Teeth ();
3447 @b{virtual} void Set_Owner (char* Name);
3449 Dog(); // Constructor
3456 In the following examples we will assume that the previous declarations are
3457 located in a file named @code{animals.h}. The following package demonstrates
3458 how to import these C++ declarations from the Ada side:
3460 @smallexample @c ada
3461 with Interfaces.C.Strings; use Interfaces.C.Strings;
3463 type Carnivore is interface;
3464 pragma Convention (C_Plus_Plus, Carnivore);
3465 function Number_Of_Teeth (X : Carnivore)
3466 return Natural is abstract;
3468 type Domestic is interface;
3469 pragma Convention (C_Plus_Plus, Set_Owner);
3471 (X : in out Domestic;
3472 Name : Chars_Ptr) is abstract;
3474 type Animal is tagged record
3477 pragma Import (C_Plus_Plus, Animal);
3479 procedure Set_Age (X : in out Animal; Age : Integer);
3480 pragma Import (C_Plus_Plus, Set_Age);
3482 function Age (X : Animal) return Integer;
3483 pragma Import (C_Plus_Plus, Age);
3485 type Dog is new Animal and Carnivore and Domestic with record
3486 Tooth_Count : Natural;
3487 Owner : String (1 .. 30);
3489 pragma Import (C_Plus_Plus, Dog);
3491 function Number_Of_Teeth (A : Dog) return Integer;
3492 pragma Import (C_Plus_Plus, Number_Of_Teeth);
3494 procedure Set_Owner (A : in out Dog; Name : Chars_Ptr);
3495 pragma Import (C_Plus_Plus, Set_Owner);
3497 function New_Dog return Dog;
3498 pragma CPP_Constructor (New_Dog);
3499 pragma Import (CPP, New_Dog, "_ZN3DogC2Ev");
3503 Thanks to the compatibility between GNAT run-time structures and the C++ ABI,
3504 interfacing with these C++ classes is easy. The only requirement is that all
3505 the primitives and components must be declared exactly in the same order in
3508 Regarding the abstract interfaces, we must indicate to the GNAT compiler by
3509 means of a @code{pragma Convention (C_Plus_Plus)}, the convention used to pass
3510 the arguments to the called primitives will be the same as for C++. For the
3511 imported classes we use @code{pragma Import} with convention @code{C_Plus_Plus}
3512 to indicate that they have been defined on the C++ side; this is required
3513 because the dispatch table associated with these tagged types will be built
3514 in the C++ side and therefore will not contain the predefined Ada primitives
3515 which Ada would otherwise expect.
3517 As the reader can see there is no need to indicate the C++ mangled names
3518 associated with each subprogram because it is assumed that all the calls to
3519 these primitives will be dispatching calls. The only exception is the
3520 constructor, which must be registered with the compiler by means of
3521 @code{pragma CPP_Constructor} and needs to provide its associated C++
3522 mangled name because the Ada compiler generates direct calls to it.
3524 With the above packages we can now declare objects of type Dog on the Ada side
3525 and dispatch calls to the corresponding subprograms on the C++ side. We can
3526 also extend the tagged type Dog with further fields and primitives, and
3527 override some of its C++ primitives on the Ada side. For example, here we have
3528 a type derivation defined on the Ada side that inherits all the dispatching
3529 primitives of the ancestor from the C++ side.
3532 @b{with} Animals; @b{use} Animals;
3533 @b{package} Vaccinated_Animals @b{is}
3534 @b{type} Vaccinated_Dog @b{is new} Dog @b{with null record};
3535 @b{function} Vaccination_Expired (A : Vaccinated_Dog) @b{return} Boolean;
3536 @b{end} Vaccinated_Animals;
3539 It is important to note that, because of the ABI compatibility, the programmer
3540 does not need to add any further information to indicate either the object
3541 layout or the dispatch table entry associated with each dispatching operation.
3543 Now let us define all the types and constructors on the Ada side and export
3544 them to C++, using the same hierarchy of our previous example:
3546 @smallexample @c ada
3547 with Interfaces.C.Strings;
3548 use Interfaces.C.Strings;
3550 type Carnivore is interface;
3551 pragma Convention (C_Plus_Plus, Carnivore);
3552 function Number_Of_Teeth (X : Carnivore)
3553 return Natural is abstract;
3555 type Domestic is interface;
3556 pragma Convention (C_Plus_Plus, Set_Owner);
3558 (X : in out Domestic;
3559 Name : Chars_Ptr) is abstract;
3561 type Animal is tagged record
3564 pragma Convention (C_Plus_Plus, Animal);
3566 procedure Set_Age (X : in out Animal; Age : Integer);
3567 pragma Export (C_Plus_Plus, Set_Age);
3569 function Age (X : Animal) return Integer;
3570 pragma Export (C_Plus_Plus, Age);
3572 type Dog is new Animal and Carnivore and Domestic with record
3573 Tooth_Count : Natural;
3574 Owner : String (1 .. 30);
3576 pragma Convention (C_Plus_Plus, Dog);
3578 function Number_Of_Teeth (A : Dog) return Integer;
3579 pragma Export (C_Plus_Plus, Number_Of_Teeth);
3581 procedure Set_Owner (A : in out Dog; Name : Chars_Ptr);
3582 pragma Export (C_Plus_Plus, Set_Owner);
3584 function New_Dog return Dog'Class;
3585 pragma Export (C_Plus_Plus, New_Dog);
3589 Compared with our previous example the only difference is the use of
3590 @code{pragma Export} to indicate to the GNAT compiler that the primitives will
3591 be available to C++. Thanks to the ABI compatibility, on the C++ side there is
3592 nothing else to be done; as explained above, the only requirement is that all
3593 the primitives and components are declared in exactly the same order.
3595 For completeness, let us see a brief C++ main program that uses the
3596 declarations available in @code{animals.h} (presented in our first example) to
3597 import and use the declarations from the Ada side, properly initializing and
3598 finalizing the Ada run-time system along the way:
3601 @b{#include} "animals.h"
3602 @b{#include} <iostream>
3603 @b{using namespace} std;
3605 void Check_Carnivore (Carnivore *obj) @{@dots{}@}
3606 void Check_Domestic (Domestic *obj) @{@dots{}@}
3607 void Check_Animal (Animal *obj) @{@dots{}@}
3608 void Check_Dog (Dog *obj) @{@dots{}@}
3611 void adainit (void);
3612 void adafinal (void);
3618 Dog *obj = new_dog(); // Ada constructor
3619 Check_Carnivore (obj); // Check secondary DT
3620 Check_Domestic (obj); // Check secondary DT
3621 Check_Animal (obj); // Check primary DT
3622 Check_Dog (obj); // Check primary DT
3627 adainit (); test(); adafinal ();
3632 @node Comparison between GNAT and C/C++ Compilation Models
3633 @section Comparison between GNAT and C/C++ Compilation Models
3636 The GNAT model of compilation is close to the C and C++ models. You can
3637 think of Ada specs as corresponding to header files in C. As in C, you
3638 don't need to compile specs; they are compiled when they are used. The
3639 Ada @code{with} is similar in effect to the @code{#include} of a C
3642 One notable difference is that, in Ada, you may compile specs separately
3643 to check them for semantic and syntactic accuracy. This is not always
3644 possible with C headers because they are fragments of programs that have
3645 less specific syntactic or semantic rules.
3647 The other major difference is the requirement for running the binder,
3648 which performs two important functions. First, it checks for
3649 consistency. In C or C++, the only defense against assembling
3650 inconsistent programs lies outside the compiler, in a makefile, for
3651 example. The binder satisfies the Ada requirement that it be impossible
3652 to construct an inconsistent program when the compiler is used in normal
3655 @cindex Elaboration order control
3656 The other important function of the binder is to deal with elaboration
3657 issues. There are also elaboration issues in C++ that are handled
3658 automatically. This automatic handling has the advantage of being
3659 simpler to use, but the C++ programmer has no control over elaboration.
3660 Where @code{gnatbind} might complain there was no valid order of
3661 elaboration, a C++ compiler would simply construct a program that
3662 malfunctioned at run time.
3665 @node Comparison between GNAT and Conventional Ada Library Models
3666 @section Comparison between GNAT and Conventional Ada Library Models
3669 This section is intended for Ada programmers who have
3670 used an Ada compiler implementing the traditional Ada library
3671 model, as described in the Ada Reference Manual.
3673 @cindex GNAT library
3674 In GNAT, there is no ``library'' in the normal sense. Instead, the set of
3675 source files themselves acts as the library. Compiling Ada programs does
3676 not generate any centralized information, but rather an object file and
3677 a ALI file, which are of interest only to the binder and linker.
3678 In a traditional system, the compiler reads information not only from
3679 the source file being compiled, but also from the centralized library.
3680 This means that the effect of a compilation depends on what has been
3681 previously compiled. In particular:
3685 When a unit is @code{with}'ed, the unit seen by the compiler corresponds
3686 to the version of the unit most recently compiled into the library.
3689 Inlining is effective only if the necessary body has already been
3690 compiled into the library.
3693 Compiling a unit may obsolete other units in the library.
3697 In GNAT, compiling one unit never affects the compilation of any other
3698 units because the compiler reads only source files. Only changes to source
3699 files can affect the results of a compilation. In particular:
3703 When a unit is @code{with}'ed, the unit seen by the compiler corresponds
3704 to the source version of the unit that is currently accessible to the
3709 Inlining requires the appropriate source files for the package or
3710 subprogram bodies to be available to the compiler. Inlining is always
3711 effective, independent of the order in which units are complied.
3714 Compiling a unit never affects any other compilations. The editing of
3715 sources may cause previous compilations to be out of date if they
3716 depended on the source file being modified.
3720 The most important result of these differences is that order of compilation
3721 is never significant in GNAT. There is no situation in which one is
3722 required to do one compilation before another. What shows up as order of
3723 compilation requirements in the traditional Ada library becomes, in
3724 GNAT, simple source dependencies; in other words, there is only a set
3725 of rules saying what source files must be present when a file is
3729 @node Placement of temporary files
3730 @section Placement of temporary files
3731 @cindex Temporary files (user control over placement)
3734 GNAT creates temporary files in the directory designated by the environment
3735 variable @env{TMPDIR}.
3736 (See the HP @emph{C RTL Reference Manual} on the function @code{getenv()}
3737 for detailed information on how environment variables are resolved.
3738 For most users the easiest way to make use of this feature is to simply
3739 define @env{TMPDIR} as a job level logical name).
3740 For example, if you wish to use a Ramdisk (assuming DECRAM is installed)
3741 for compiler temporary files, then you can include something like the
3742 following command in your @file{LOGIN.COM} file:
3745 $ define/job TMPDIR "/disk$scratchram/000000/temp/"
3749 If @env{TMPDIR} is not defined, then GNAT uses the directory designated by
3750 @env{TMP}; if @env{TMP} is not defined, then GNAT uses the directory
3751 designated by @env{TEMP}.
3752 If none of these environment variables are defined then GNAT uses the
3753 directory designated by the logical name @code{SYS$SCRATCH:}
3754 (by default the user's home directory). If all else fails
3755 GNAT uses the current directory for temporary files.
3758 @c *************************
3759 @node Compiling Using gcc
3760 @chapter Compiling Using @command{gcc}
3763 This chapter discusses how to compile Ada programs using the @command{gcc}
3764 command. It also describes the set of switches
3765 that can be used to control the behavior of the compiler.
3767 * Compiling Programs::
3768 * Switches for gcc::
3769 * Search Paths and the Run-Time Library (RTL)::
3770 * Order of Compilation Issues::
3774 @node Compiling Programs
3775 @section Compiling Programs
3778 The first step in creating an executable program is to compile the units
3779 of the program using the @command{gcc} command. You must compile the
3784 the body file (@file{.adb}) for a library level subprogram or generic
3788 the spec file (@file{.ads}) for a library level package or generic
3789 package that has no body
3792 the body file (@file{.adb}) for a library level package
3793 or generic package that has a body
3798 You need @emph{not} compile the following files
3803 the spec of a library unit which has a body
3810 because they are compiled as part of compiling related units. GNAT
3812 when the corresponding body is compiled, and subunits when the parent is
3815 @cindex cannot generate code
3816 If you attempt to compile any of these files, you will get one of the
3817 following error messages (where @var{fff} is the name of the file you
3821 cannot generate code for file @var{fff} (package spec)
3822 to check package spec, use -gnatc
3824 cannot generate code for file @var{fff} (missing subunits)
3825 to check parent unit, use -gnatc
3827 cannot generate code for file @var{fff} (subprogram spec)
3828 to check subprogram spec, use -gnatc
3830 cannot generate code for file @var{fff} (subunit)
3831 to check subunit, use -gnatc
3835 As indicated by the above error messages, if you want to submit
3836 one of these files to the compiler to check for correct semantics
3837 without generating code, then use the @option{-gnatc} switch.
3839 The basic command for compiling a file containing an Ada unit is
3842 @c $ gcc -c @ovar{switches} @file{file name}
3843 @c Expanding @ovar macro inline (explanation in macro def comments)
3844 $ gcc -c @r{[}@var{switches}@r{]} @file{file name}
3848 where @var{file name} is the name of the Ada file (usually
3850 @file{.ads} for a spec or @file{.adb} for a body).
3853 @option{-c} switch to tell @command{gcc} to compile, but not link, the file.
3855 The result of a successful compilation is an object file, which has the
3856 same name as the source file but an extension of @file{.o} and an Ada
3857 Library Information (ALI) file, which also has the same name as the
3858 source file, but with @file{.ali} as the extension. GNAT creates these
3859 two output files in the current directory, but you may specify a source
3860 file in any directory using an absolute or relative path specification
3861 containing the directory information.
3864 @command{gcc} is actually a driver program that looks at the extensions of
3865 the file arguments and loads the appropriate compiler. For example, the
3866 GNU C compiler is @file{cc1}, and the Ada compiler is @file{gnat1}.
3867 These programs are in directories known to the driver program (in some
3868 configurations via environment variables you set), but need not be in
3869 your path. The @command{gcc} driver also calls the assembler and any other
3870 utilities needed to complete the generation of the required object
3873 It is possible to supply several file names on the same @command{gcc}
3874 command. This causes @command{gcc} to call the appropriate compiler for
3875 each file. For example, the following command lists three separate
3876 files to be compiled:
3879 $ gcc -c x.adb y.adb z.c
3883 calls @code{gnat1} (the Ada compiler) twice to compile @file{x.adb} and
3884 @file{y.adb}, and @code{cc1} (the C compiler) once to compile @file{z.c}.
3885 The compiler generates three object files @file{x.o}, @file{y.o} and
3886 @file{z.o} and the two ALI files @file{x.ali} and @file{y.ali} from the
3887 Ada compilations. Any switches apply to all the files ^listed,^listed.^
3890 @option{-gnat@var{x}} switches, which apply only to Ada compilations.
3893 @node Switches for gcc
3894 @section Switches for @command{gcc}
3897 The @command{gcc} command accepts switches that control the
3898 compilation process. These switches are fully described in this section.
3899 First we briefly list all the switches, in alphabetical order, then we
3900 describe the switches in more detail in functionally grouped sections.
3902 More switches exist for GCC than those documented here, especially
3903 for specific targets. However, their use is not recommended as
3904 they may change code generation in ways that are incompatible with
3905 the Ada run-time library, or can cause inconsistencies between
3909 * Output and Error Message Control::
3910 * Warning Message Control::
3911 * Debugging and Assertion Control::
3912 * Validity Checking::
3915 * Using gcc for Syntax Checking::
3916 * Using gcc for Semantic Checking::
3917 * Compiling Different Versions of Ada::
3918 * Character Set Control::
3919 * File Naming Control::
3920 * Subprogram Inlining Control::
3921 * Auxiliary Output Control::
3922 * Debugging Control::
3923 * Exception Handling Control::
3924 * Units to Sources Mapping Files::
3925 * Integrated Preprocessing::
3926 * Code Generation Control::
3935 @cindex @option{-b} (@command{gcc})
3936 @item -b @var{target}
3937 Compile your program to run on @var{target}, which is the name of a
3938 system configuration. You must have a GNAT cross-compiler built if
3939 @var{target} is not the same as your host system.
3942 @cindex @option{-B} (@command{gcc})
3943 Load compiler executables (for example, @code{gnat1}, the Ada compiler)
3944 from @var{dir} instead of the default location. Only use this switch
3945 when multiple versions of the GNAT compiler are available.
3946 @xref{Directory Options,, Options for Directory Search, gcc, Using the
3947 GNU Compiler Collection (GCC)}, for further details. You would normally
3948 use the @option{-b} or @option{-V} switch instead.
3951 @cindex @option{-c} (@command{gcc})
3952 Compile. Always use this switch when compiling Ada programs.
3954 Note: for some other languages when using @command{gcc}, notably in
3955 the case of C and C++, it is possible to use
3956 use @command{gcc} without a @option{-c} switch to
3957 compile and link in one step. In the case of GNAT, you
3958 cannot use this approach, because the binder must be run
3959 and @command{gcc} cannot be used to run the GNAT binder.
3963 @cindex @option{-fno-inline} (@command{gcc})
3964 Suppresses all back-end inlining, even if other optimization or inlining
3966 This includes suppression of inlining that results
3967 from the use of the pragma @code{Inline_Always}.
3968 Any occurrences of pragma @code{Inline} or @code{Inline_Always}
3969 are ignored, and @option{-gnatn} and @option{-gnatN} have no
3970 effect if this switch is present.
3972 @item -fno-inline-functions
3973 @cindex @option{-fno-inline-functions} (@command{gcc})
3974 Suppresses automatic inlining of subprograms, which is enabled
3975 if @option{-O3} is used.
3977 @item -fno-inline-small-functions
3978 @cindex @option{-fno-inline-small-functions} (@command{gcc})
3979 Suppresses automatic inlining of small subprograms, which is enabled
3980 if @option{-O2} is used.
3982 @item -fno-inline-functions-called-once
3983 @cindex @option{-fno-inline-functions-called-once} (@command{gcc})
3984 Suppresses inlining of subprograms local to the unit and called once
3985 from within it, which is enabled if @option{-O1} is used.
3988 @cindex @option{-fno-ivopts} (@command{gcc})
3989 Suppresses high-level loop induction variable optimizations, which are
3990 enabled if @option{-O1} is used. These optimizations are generally
3991 profitable but, for some specific cases of loops with numerous uses
3992 of the iteration variable that follow a common pattern, they may end
3993 up destroying the regularity that could be exploited at a lower level
3994 and thus producing inferior code.
3996 @item -fno-strict-aliasing
3997 @cindex @option{-fno-strict-aliasing} (@command{gcc})
3998 Causes the compiler to avoid assumptions regarding non-aliasing
3999 of objects of different types. See
4000 @ref{Optimization and Strict Aliasing} for details.
4003 @cindex @option{-fstack-check} (@command{gcc})
4004 Activates stack checking.
4005 See @ref{Stack Overflow Checking} for details.
4008 @cindex @option{-fstack-usage} (@command{gcc})
4009 Makes the compiler output stack usage information for the program, on a
4010 per-function basis. See @ref{Static Stack Usage Analysis} for details.
4012 @item -fcallgraph-info@r{[}=su@r{]}
4013 @cindex @option{-fcallgraph-info} (@command{gcc})
4014 Makes the compiler output callgraph information for the program, on a
4015 per-file basis. The information is generated in the VCG format. It can
4016 be decorated with stack-usage per-node information.
4019 @cindex @option{^-g^/DEBUG^} (@command{gcc})
4020 Generate debugging information. This information is stored in the object
4021 file and copied from there to the final executable file by the linker,
4022 where it can be read by the debugger. You must use the
4023 @option{^-g^/DEBUG^} switch if you plan on using the debugger.
4026 @cindex @option{-gnat83} (@command{gcc})
4027 Enforce Ada 83 restrictions.
4030 @cindex @option{-gnat95} (@command{gcc})
4031 Enforce Ada 95 restrictions.
4034 @cindex @option{-gnat05} (@command{gcc})
4035 Allow full Ada 2005 features.
4038 @cindex @option{-gnat2005} (@command{gcc})
4039 Allow full Ada 2005 features (same as @option{-gnat05})
4042 @cindex @option{-gnat12} (@command{gcc})
4045 @cindex @option{-gnat2012} (@command{gcc})
4046 Allow full Ada 2012 features (same as @option{-gnat12})
4049 @cindex @option{-gnata} (@command{gcc})
4050 Assertions enabled. @code{Pragma Assert} and @code{pragma Debug} to be
4051 activated. Note that these pragmas can also be controlled using the
4052 configuration pragmas @code{Assertion_Policy} and @code{Debug_Policy}.
4053 It also activates pragmas @code{Check}, @code{Precondition}, and
4054 @code{Postcondition}. Note that these pragmas can also be controlled
4055 using the configuration pragma @code{Check_Policy}.
4058 @cindex @option{-gnatA} (@command{gcc})
4059 Avoid processing @file{gnat.adc}. If a @file{gnat.adc} file is present,
4063 @cindex @option{-gnatb} (@command{gcc})
4064 Generate brief messages to @file{stderr} even if verbose mode set.
4067 @cindex @option{-gnatB} (@command{gcc})
4068 Assume no invalid (bad) values except for 'Valid attribute use
4069 (@pxref{Validity Checking}).
4072 @cindex @option{-gnatc} (@command{gcc})
4073 Check syntax and semantics only (no code generation attempted).
4076 @cindex @option{-gnatC} (@command{gcc})
4077 Generate CodePeer information (no code generation attempted).
4078 This switch will generate an intermediate representation suitable for
4079 use by CodePeer (@file{.scil} files). This switch is not compatible with
4080 code generation (it will, among other things, disable some switches such
4081 as -gnatn, and enable others such as -gnata).
4084 @cindex @option{-gnatd} (@command{gcc})
4085 Specify debug options for the compiler. The string of characters after
4086 the @option{-gnatd} specify the specific debug options. The possible
4087 characters are 0-9, a-z, A-Z, optionally preceded by a dot. See
4088 compiler source file @file{debug.adb} for details of the implemented
4089 debug options. Certain debug options are relevant to applications
4090 programmers, and these are documented at appropriate points in this
4095 @cindex @option{-gnatD[nn]} (@command{gcc})
4098 @item /XDEBUG /LXDEBUG=nnn
4100 Create expanded source files for source level debugging. This switch
4101 also suppress generation of cross-reference information
4102 (see @option{-gnatx}).
4104 @item -gnatec=@var{path}
4105 @cindex @option{-gnatec} (@command{gcc})
4106 Specify a configuration pragma file
4108 (the equal sign is optional)
4110 (@pxref{The Configuration Pragmas Files}).
4112 @item ^-gnateD^/DATA_PREPROCESSING=^symbol@r{[}=@var{value}@r{]}
4113 @cindex @option{-gnateD} (@command{gcc})
4114 Defines a symbol, associated with @var{value}, for preprocessing.
4115 (@pxref{Integrated Preprocessing}).
4118 @cindex @option{-gnateE} (@command{gcc})
4119 Generate extra information in exception messages, in particular display
4120 extra column information and the value and range associated with index and
4121 range check failures, and extra column information for access checks.
4124 @cindex @option{-gnatef} (@command{gcc})
4125 Display full source path name in brief error messages.
4128 @cindex @option{-gnateG} (@command{gcc})
4129 Save result of preprocessing in a text file.
4131 @item -gnatem=@var{path}
4132 @cindex @option{-gnatem} (@command{gcc})
4133 Specify a mapping file
4135 (the equal sign is optional)
4137 (@pxref{Units to Sources Mapping Files}).
4139 @item -gnatep=@var{file}
4140 @cindex @option{-gnatep} (@command{gcc})
4141 Specify a preprocessing data file
4143 (the equal sign is optional)
4145 (@pxref{Integrated Preprocessing}).
4148 @cindex @option{-gnateP} (@command{gcc})
4149 Turn categorization dependency errors into warnings.
4150 Ada requires that units that WITH one another have compatible categories, for
4151 example a Pure unit cannto WITH a Preelaborate unit. If this switch is used,
4152 these errors become warnings (which can be ignored, or suppressed in the usual
4153 manner). This can be useful in some specialized circumstances such as the
4154 temporary use of special test software.
4156 @cindex @option{-gnateS} (@command{gcc})
4157 Generate SCO (Source Coverage Obligation) information in the ALI
4158 file. This information is used by advanced coverage tools. See
4159 unit @file{SCOs} in the compiler sources for details in files
4160 @file{scos.ads} and @file{scos.adb}.
4163 @cindex @option{-gnatE} (@command{gcc})
4164 Full dynamic elaboration checks.
4167 @cindex @option{-gnatf} (@command{gcc})
4168 Full errors. Multiple errors per line, all undefined references, do not
4169 attempt to suppress cascaded errors.
4172 @cindex @option{-gnatF} (@command{gcc})
4173 Externals names are folded to all uppercase.
4175 @item ^-gnatg^/GNAT_INTERNAL^
4176 @cindex @option{^-gnatg^/GNAT_INTERNAL^} (@command{gcc})
4177 Internal GNAT implementation mode. This should not be used for
4178 applications programs, it is intended only for use by the compiler
4179 and its run-time library. For documentation, see the GNAT sources.
4180 Note that @option{^-gnatg^/GNAT_INTERNAL^} implies
4181 @option{^-gnatwae^/WARNINGS=ALL,ERRORS^} and
4182 @option{^-gnatyg^/STYLE_CHECKS=GNAT^}
4183 so that all standard warnings and all standard style options are turned on.
4184 All warnings and style messages are treated as errors.
4188 @cindex @option{-gnatG[nn]} (@command{gcc})
4191 @item /EXPAND_SOURCE, /LEXPAND_SOURCE=nnn
4193 List generated expanded code in source form.
4195 @item ^-gnath^/HELP^
4196 @cindex @option{^-gnath^/HELP^} (@command{gcc})
4197 Output usage information. The output is written to @file{stdout}.
4199 @item ^-gnati^/IDENTIFIER_CHARACTER_SET=^@var{c}
4200 @cindex @option{^-gnati^/IDENTIFIER_CHARACTER_SET^} (@command{gcc})
4201 Identifier character set
4203 (@var{c}=1/2/3/4/8/9/p/f/n/w).
4205 For details of the possible selections for @var{c},
4206 see @ref{Character Set Control}.
4208 @item ^-gnatI^/IGNORE_REP_CLAUSES^
4209 @cindex @option{^-gnatI^IGNORE_REP_CLAUSES^} (@command{gcc})
4210 Ignore representation clauses. When this switch is used,
4211 representation clauses are treated as comments. This is useful
4212 when initially porting code where you want to ignore rep clause
4213 problems, and also for compiling foreign code (particularly
4214 for use with ASIS). The representation clauses that are ignored
4215 are: enumeration_representation_clause, record_representation_clause,
4216 and attribute_definition_clause for the following attributes:
4217 Address, Alignment, Bit_Order, Component_Size, Machine_Radix,
4218 Object_Size, Size, Small, Stream_Size, and Value_Size.
4219 Note that this option should be used only for compiling -- the
4220 code is likely to malfunction at run time.
4223 @cindex @option{-gnatjnn} (@command{gcc})
4224 Reformat error messages to fit on nn character lines
4226 @item -gnatk=@var{n}
4227 @cindex @option{-gnatk} (@command{gcc})
4228 Limit file names to @var{n} (1-999) characters ^(@code{k} = krunch)^^.
4231 @cindex @option{-gnatl} (@command{gcc})
4232 Output full source listing with embedded error messages.
4235 @cindex @option{-gnatL} (@command{gcc})
4236 Used in conjunction with -gnatG or -gnatD to intersperse original
4237 source lines (as comment lines with line numbers) in the expanded
4240 @item -gnatm=@var{n}
4241 @cindex @option{-gnatm} (@command{gcc})
4242 Limit number of detected error or warning messages to @var{n}
4243 where @var{n} is in the range 1..999999. The default setting if
4244 no switch is given is 9999. If the number of warnings reaches this
4245 limit, then a message is output and further warnings are suppressed,
4246 but the compilation is continued. If the number of error messages
4247 reaches this limit, then a message is output and the compilation
4248 is abandoned. The equal sign here is optional. A value of zero
4249 means that no limit applies.
4252 @cindex @option{-gnatn} (@command{gcc})
4253 Activate inlining for subprograms for which
4254 pragma @code{Inline} is specified. This inlining is performed
4255 by the GCC back-end.
4258 @cindex @option{-gnatN} (@command{gcc})
4259 Activate front end inlining for subprograms for which
4260 pragma @code{Inline} is specified. This inlining is performed
4261 by the front end and will be visible in the
4262 @option{-gnatG} output.
4264 When using a gcc-based back end (in practice this means using any version
4265 of GNAT other than the JGNAT, .NET or GNAAMP versions), then the use of
4266 @option{-gnatN} is deprecated, and the use of @option{-gnatn} is preferred.
4267 Historically front end inlining was more extensive than the gcc back end
4268 inlining, but that is no longer the case.
4271 @cindex @option{-gnato} (@command{gcc})
4272 Enable numeric overflow checking (which is not normally enabled by
4273 default). Note that division by zero is a separate check that is not
4274 controlled by this switch (division by zero checking is on by default).
4277 @cindex @option{-gnatp} (@command{gcc})
4278 Suppress all checks. See @ref{Run-Time Checks} for details. This switch
4279 has no effect if cancelled by a subsequent @option{-gnat-p} switch.
4282 @cindex @option{-gnat-p} (@command{gcc})
4283 Cancel effect of previous @option{-gnatp} switch.
4286 @cindex @option{-gnatP} (@command{gcc})
4287 Enable polling. This is required on some systems (notably Windows NT) to
4288 obtain asynchronous abort and asynchronous transfer of control capability.
4289 @xref{Pragma Polling,,, gnat_rm, GNAT Reference Manual}, for full
4293 @cindex @option{-gnatq} (@command{gcc})
4294 Don't quit. Try semantics, even if parse errors.
4297 @cindex @option{-gnatQ} (@command{gcc})
4298 Don't quit. Generate @file{ALI} and tree files even if illegalities.
4301 @cindex @option{-gnatr} (@command{gcc})
4302 Treat pragma Restrictions as Restriction_Warnings.
4304 @item ^-gnatR@r{[}0@r{/}1@r{/}2@r{/}3@r{[}s@r{]]}^/REPRESENTATION_INFO^
4305 @cindex @option{-gnatR} (@command{gcc})
4306 Output representation information for declared types and objects.
4309 @cindex @option{-gnats} (@command{gcc})
4313 @cindex @option{-gnatS} (@command{gcc})
4314 Print package Standard.
4317 @cindex @option{-gnatt} (@command{gcc})
4318 Generate tree output file.
4320 @item ^-gnatT^/TABLE_MULTIPLIER=^@var{nnn}
4321 @cindex @option{^-gnatT^/TABLE_MULTIPLIER^} (@command{gcc})
4322 All compiler tables start at @var{nnn} times usual starting size.
4325 @cindex @option{-gnatu} (@command{gcc})
4326 List units for this compilation.
4329 @cindex @option{-gnatU} (@command{gcc})
4330 Tag all error messages with the unique string ``error:''
4333 @cindex @option{-gnatv} (@command{gcc})
4334 Verbose mode. Full error output with source lines to @file{stdout}.
4337 @cindex @option{-gnatV} (@command{gcc})
4338 Control level of validity checking (@pxref{Validity Checking}).
4340 @item ^-gnatw@var{xxx}^/WARNINGS=(@var{option}@r{[},@dots{}@r{]})^
4341 @cindex @option{^-gnatw^/WARNINGS^} (@command{gcc})
4343 ^@var{xxx} is a string of option letters that^the list of options^ denotes
4344 the exact warnings that
4345 are enabled or disabled (@pxref{Warning Message Control}).
4347 @item ^-gnatW^/WIDE_CHARACTER_ENCODING=^@var{e}
4348 @cindex @option{^-gnatW^/WIDE_CHARACTER_ENCODING^} (@command{gcc})
4349 Wide character encoding method
4351 (@var{e}=n/h/u/s/e/8).
4354 (@var{e}=@code{BRACKETS, NONE, HEX, UPPER, SHIFT_JIS, EUC, UTF8})
4358 @cindex @option{-gnatx} (@command{gcc})
4359 Suppress generation of cross-reference information.
4362 @cindex @option{-gnatX} (@command{gcc})
4363 Enable GNAT implementation extensions and latest Ada version.
4365 @item ^-gnaty^/STYLE_CHECKS=(option,option@dots{})^
4366 @cindex @option{^-gnaty^/STYLE_CHECKS^} (@command{gcc})
4367 Enable built-in style checks (@pxref{Style Checking}).
4369 @item ^-gnatz^/DISTRIBUTION_STUBS=^@var{m}
4370 @cindex @option{^-gnatz^/DISTRIBUTION_STUBS^} (@command{gcc})
4371 Distribution stub generation and compilation
4373 (@var{m}=r/c for receiver/caller stubs).
4376 (@var{m}=@code{RECEIVER} or @code{CALLER} to specify the type of stubs
4377 to be generated and compiled).
4380 @item ^-I^/SEARCH=^@var{dir}
4381 @cindex @option{^-I^/SEARCH^} (@command{gcc})
4383 Direct GNAT to search the @var{dir} directory for source files needed by
4384 the current compilation
4385 (@pxref{Search Paths and the Run-Time Library (RTL)}).
4387 @item ^-I-^/NOCURRENT_DIRECTORY^
4388 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@command{gcc})
4390 Except for the source file named in the command line, do not look for source
4391 files in the directory containing the source file named in the command line
4392 (@pxref{Search Paths and the Run-Time Library (RTL)}).
4396 @cindex @option{-mbig-switch} (@command{gcc})
4397 @cindex @code{case} statement (effect of @option{-mbig-switch} option)
4398 This standard gcc switch causes the compiler to use larger offsets in its
4399 jump table representation for @code{case} statements.
4400 This may result in less efficient code, but is sometimes necessary
4401 (for example on HP-UX targets)
4402 @cindex HP-UX and @option{-mbig-switch} option
4403 in order to compile large and/or nested @code{case} statements.
4406 @cindex @option{-o} (@command{gcc})
4407 This switch is used in @command{gcc} to redirect the generated object file
4408 and its associated ALI file. Beware of this switch with GNAT, because it may
4409 cause the object file and ALI file to have different names which in turn
4410 may confuse the binder and the linker.
4414 @cindex @option{-nostdinc} (@command{gcc})
4415 Inhibit the search of the default location for the GNAT Run Time
4416 Library (RTL) source files.
4419 @cindex @option{-nostdlib} (@command{gcc})
4420 Inhibit the search of the default location for the GNAT Run Time
4421 Library (RTL) ALI files.
4425 @c Expanding @ovar macro inline (explanation in macro def comments)
4426 @item -O@r{[}@var{n}@r{]}
4427 @cindex @option{-O} (@command{gcc})
4428 @var{n} controls the optimization level.
4432 No optimization, the default setting if no @option{-O} appears
4435 Normal optimization, the default if you specify @option{-O} without
4436 an operand. A good compromise between code quality and compilation
4440 Extensive optimization, may improve execution time, possibly at the cost of
4441 substantially increased compilation time.
4444 Same as @option{-O2}, and also includes inline expansion for small subprograms
4448 Optimize space usage
4452 See also @ref{Optimization Levels}.
4457 @cindex @option{/NOOPTIMIZE} (@code{GNAT COMPILE})
4458 Equivalent to @option{/OPTIMIZE=NONE}.
4459 This is the default behavior in the absence of an @option{/OPTIMIZE}
4462 @item /OPTIMIZE@r{[}=(keyword@r{[},@dots{}@r{]})@r{]}
4463 @cindex @option{/OPTIMIZE} (@code{GNAT COMPILE})
4464 Selects the level of optimization for your program. The supported
4465 keywords are as follows:
4468 Perform most optimizations, including those that
4470 This is the default if the @option{/OPTIMIZE} qualifier is supplied
4471 without keyword options.
4474 Do not do any optimizations. Same as @code{/NOOPTIMIZE}.
4477 Perform some optimizations, but omit ones that are costly.
4480 Same as @code{SOME}.
4483 Full optimization as in @option{/OPTIMIZE=ALL}, and also attempts
4484 automatic inlining of small subprograms within a unit
4487 Try to unroll loops. This keyword may be specified together with
4488 any keyword above other than @code{NONE}. Loop unrolling
4489 usually, but not always, improves the performance of programs.
4492 Optimize space usage
4496 See also @ref{Optimization Levels}.
4500 @item -pass-exit-codes
4501 @cindex @option{-pass-exit-codes} (@command{gcc})
4502 Catch exit codes from the compiler and use the most meaningful as
4506 @item --RTS=@var{rts-path}
4507 @cindex @option{--RTS} (@command{gcc})
4508 Specifies the default location of the runtime library. Same meaning as the
4509 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
4512 @cindex @option{^-S^/ASM^} (@command{gcc})
4513 ^Used in place of @option{-c} to^Used to^
4514 cause the assembler source file to be
4515 generated, using @file{^.s^.S^} as the extension,
4516 instead of the object file.
4517 This may be useful if you need to examine the generated assembly code.
4519 @item ^-fverbose-asm^/VERBOSE_ASM^
4520 @cindex @option{^-fverbose-asm^/VERBOSE_ASM^} (@command{gcc})
4521 ^Used in conjunction with @option{-S}^Used in place of @option{/ASM}^
4522 to cause the generated assembly code file to be annotated with variable
4523 names, making it significantly easier to follow.
4526 @cindex @option{^-v^/VERBOSE^} (@command{gcc})
4527 Show commands generated by the @command{gcc} driver. Normally used only for
4528 debugging purposes or if you need to be sure what version of the
4529 compiler you are executing.
4533 @cindex @option{-V} (@command{gcc})
4534 Execute @var{ver} version of the compiler. This is the @command{gcc}
4535 version, not the GNAT version.
4538 @item ^-w^/NO_BACK_END_WARNINGS^
4539 @cindex @option{-w} (@command{gcc})
4540 Turn off warnings generated by the back end of the compiler. Use of
4541 this switch also causes the default for front end warnings to be set
4542 to suppress (as though @option{-gnatws} had appeared at the start of
4548 @c Combining qualifiers does not work on VMS
4549 You may combine a sequence of GNAT switches into a single switch. For
4550 example, the combined switch
4552 @cindex Combining GNAT switches
4558 is equivalent to specifying the following sequence of switches:
4561 -gnato -gnatf -gnati3
4566 The following restrictions apply to the combination of switches
4571 The switch @option{-gnatc} if combined with other switches must come
4572 first in the string.
4575 The switch @option{-gnats} if combined with other switches must come
4576 first in the string.
4580 ^^@option{/DISTRIBUTION_STUBS=},^
4581 @option{-gnatzc} and @option{-gnatzr} may not be combined with any other
4582 switches, and only one of them may appear in the command line.
4585 The switch @option{-gnat-p} may not be combined with any other switch.
4589 Once a ``y'' appears in the string (that is a use of the @option{-gnaty}
4590 switch), then all further characters in the switch are interpreted
4591 as style modifiers (see description of @option{-gnaty}).
4594 Once a ``d'' appears in the string (that is a use of the @option{-gnatd}
4595 switch), then all further characters in the switch are interpreted
4596 as debug flags (see description of @option{-gnatd}).
4599 Once a ``w'' appears in the string (that is a use of the @option{-gnatw}
4600 switch), then all further characters in the switch are interpreted
4601 as warning mode modifiers (see description of @option{-gnatw}).
4604 Once a ``V'' appears in the string (that is a use of the @option{-gnatV}
4605 switch), then all further characters in the switch are interpreted
4606 as validity checking options (@pxref{Validity Checking}).
4609 Option ``em'', ``ec'', ``ep'', ``l='' and ``R'' must be the last options in
4610 a combined list of options.
4614 @node Output and Error Message Control
4615 @subsection Output and Error Message Control
4619 The standard default format for error messages is called ``brief format''.
4620 Brief format messages are written to @file{stderr} (the standard error
4621 file) and have the following form:
4624 e.adb:3:04: Incorrect spelling of keyword "function"
4625 e.adb:4:20: ";" should be "is"
4629 The first integer after the file name is the line number in the file,
4630 and the second integer is the column number within the line.
4632 @code{GPS} can parse the error messages
4633 and point to the referenced character.
4635 The following switches provide control over the error message
4641 @cindex @option{-gnatv} (@command{gcc})
4644 The v stands for verbose.
4646 The effect of this setting is to write long-format error
4647 messages to @file{stdout} (the standard output file.
4648 The same program compiled with the
4649 @option{-gnatv} switch would generate:
4653 3. funcion X (Q : Integer)
4655 >>> Incorrect spelling of keyword "function"
4658 >>> ";" should be "is"
4663 The vertical bar indicates the location of the error, and the @samp{>>>}
4664 prefix can be used to search for error messages. When this switch is
4665 used the only source lines output are those with errors.
4668 @cindex @option{-gnatl} (@command{gcc})
4670 The @code{l} stands for list.
4672 This switch causes a full listing of
4673 the file to be generated. In the case where a body is
4674 compiled, the corresponding spec is also listed, along
4675 with any subunits. Typical output from compiling a package
4676 body @file{p.adb} might look like:
4678 @smallexample @c ada
4682 1. package body p is
4684 3. procedure a is separate;
4695 2. pragma Elaborate_Body
4719 When you specify the @option{-gnatv} or @option{-gnatl} switches and
4720 standard output is redirected, a brief summary is written to
4721 @file{stderr} (standard error) giving the number of error messages and
4722 warning messages generated.
4724 @item ^-gnatl^/OUTPUT_FILE^=file
4725 @cindex @option{^-gnatl^/OUTPUT_FILE^=fname} (@command{gcc})
4726 This has the same effect as @option{-gnatl} except that the output is
4727 written to a file instead of to standard output. If the given name
4728 @file{fname} does not start with a period, then it is the full name
4729 of the file to be written. If @file{fname} is an extension, it is
4730 appended to the name of the file being compiled. For example, if
4731 file @file{xyz.adb} is compiled with @option{^-gnatl^/OUTPUT_FILE^=.lst},
4732 then the output is written to file ^xyz.adb.lst^xyz.adb_lst^.
4735 @cindex @option{-gnatU} (@command{gcc})
4736 This switch forces all error messages to be preceded by the unique
4737 string ``error:''. This means that error messages take a few more
4738 characters in space, but allows easy searching for and identification
4742 @cindex @option{-gnatb} (@command{gcc})
4744 The @code{b} stands for brief.
4746 This switch causes GNAT to generate the
4747 brief format error messages to @file{stderr} (the standard error
4748 file) as well as the verbose
4749 format message or full listing (which as usual is written to
4750 @file{stdout} (the standard output file).
4752 @item -gnatm=@var{n}
4753 @cindex @option{-gnatm} (@command{gcc})
4755 The @code{m} stands for maximum.
4757 @var{n} is a decimal integer in the
4758 range of 1 to 999999 and limits the number of error or warning
4759 messages to be generated. For example, using
4760 @option{-gnatm2} might yield
4763 e.adb:3:04: Incorrect spelling of keyword "function"
4764 e.adb:5:35: missing ".."
4765 fatal error: maximum number of errors detected
4766 compilation abandoned
4770 The default setting if
4771 no switch is given is 9999. If the number of warnings reaches this
4772 limit, then a message is output and further warnings are suppressed,
4773 but the compilation is continued. If the number of error messages
4774 reaches this limit, then a message is output and the compilation
4775 is abandoned. A value of zero means that no limit applies.
4778 Note that the equal sign is optional, so the switches
4779 @option{-gnatm2} and @option{-gnatm=2} are equivalent.
4782 @cindex @option{-gnatf} (@command{gcc})
4783 @cindex Error messages, suppressing
4785 The @code{f} stands for full.
4787 Normally, the compiler suppresses error messages that are likely to be
4788 redundant. This switch causes all error
4789 messages to be generated. In particular, in the case of
4790 references to undefined variables. If a given variable is referenced
4791 several times, the normal format of messages is
4793 e.adb:7:07: "V" is undefined (more references follow)
4797 where the parenthetical comment warns that there are additional
4798 references to the variable @code{V}. Compiling the same program with the
4799 @option{-gnatf} switch yields
4802 e.adb:7:07: "V" is undefined
4803 e.adb:8:07: "V" is undefined
4804 e.adb:8:12: "V" is undefined
4805 e.adb:8:16: "V" is undefined
4806 e.adb:9:07: "V" is undefined
4807 e.adb:9:12: "V" is undefined
4811 The @option{-gnatf} switch also generates additional information for
4812 some error messages. Some examples are:
4816 Details on possibly non-portable unchecked conversion
4818 List possible interpretations for ambiguous calls
4820 Additional details on incorrect parameters
4824 @cindex @option{-gnatjnn} (@command{gcc})
4825 In normal operation mode (or if @option{-gnatj0} is used, then error messages
4826 with continuation lines are treated as though the continuation lines were
4827 separate messages (and so a warning with two continuation lines counts as
4828 three warnings, and is listed as three separate messages).
4830 If the @option{-gnatjnn} switch is used with a positive value for nn, then
4831 messages are output in a different manner. A message and all its continuation
4832 lines are treated as a unit, and count as only one warning or message in the
4833 statistics totals. Furthermore, the message is reformatted so that no line
4834 is longer than nn characters.
4837 @cindex @option{-gnatq} (@command{gcc})
4839 The @code{q} stands for quit (really ``don't quit'').
4841 In normal operation mode, the compiler first parses the program and
4842 determines if there are any syntax errors. If there are, appropriate
4843 error messages are generated and compilation is immediately terminated.
4845 GNAT to continue with semantic analysis even if syntax errors have been
4846 found. This may enable the detection of more errors in a single run. On
4847 the other hand, the semantic analyzer is more likely to encounter some
4848 internal fatal error when given a syntactically invalid tree.
4851 @cindex @option{-gnatQ} (@command{gcc})
4852 In normal operation mode, the @file{ALI} file is not generated if any
4853 illegalities are detected in the program. The use of @option{-gnatQ} forces
4854 generation of the @file{ALI} file. This file is marked as being in
4855 error, so it cannot be used for binding purposes, but it does contain
4856 reasonably complete cross-reference information, and thus may be useful
4857 for use by tools (e.g., semantic browsing tools or integrated development
4858 environments) that are driven from the @file{ALI} file. This switch
4859 implies @option{-gnatq}, since the semantic phase must be run to get a
4860 meaningful ALI file.
4862 In addition, if @option{-gnatt} is also specified, then the tree file is
4863 generated even if there are illegalities. It may be useful in this case
4864 to also specify @option{-gnatq} to ensure that full semantic processing
4865 occurs. The resulting tree file can be processed by ASIS, for the purpose
4866 of providing partial information about illegal units, but if the error
4867 causes the tree to be badly malformed, then ASIS may crash during the
4870 When @option{-gnatQ} is used and the generated @file{ALI} file is marked as
4871 being in error, @command{gnatmake} will attempt to recompile the source when it
4872 finds such an @file{ALI} file, including with switch @option{-gnatc}.
4874 Note that @option{-gnatQ} has no effect if @option{-gnats} is specified,
4875 since ALI files are never generated if @option{-gnats} is set.
4879 @node Warning Message Control
4880 @subsection Warning Message Control
4881 @cindex Warning messages
4883 In addition to error messages, which correspond to illegalities as defined
4884 in the Ada Reference Manual, the compiler detects two kinds of warning
4887 First, the compiler considers some constructs suspicious and generates a
4888 warning message to alert you to a possible error. Second, if the
4889 compiler detects a situation that is sure to raise an exception at
4890 run time, it generates a warning message. The following shows an example
4891 of warning messages:
4893 e.adb:4:24: warning: creation of object may raise Storage_Error
4894 e.adb:10:17: warning: static value out of range
4895 e.adb:10:17: warning: "Constraint_Error" will be raised at run time
4899 GNAT considers a large number of situations as appropriate
4900 for the generation of warning messages. As always, warnings are not
4901 definite indications of errors. For example, if you do an out-of-range
4902 assignment with the deliberate intention of raising a
4903 @code{Constraint_Error} exception, then the warning that may be
4904 issued does not indicate an error. Some of the situations for which GNAT
4905 issues warnings (at least some of the time) are given in the following
4906 list. This list is not complete, and new warnings are often added to
4907 subsequent versions of GNAT. The list is intended to give a general idea
4908 of the kinds of warnings that are generated.
4912 Possible infinitely recursive calls
4915 Out-of-range values being assigned
4918 Possible order of elaboration problems
4921 Assertions (pragma Assert) that are sure to fail
4927 Address clauses with possibly unaligned values, or where an attempt is
4928 made to overlay a smaller variable with a larger one.
4931 Fixed-point type declarations with a null range
4934 Direct_IO or Sequential_IO instantiated with a type that has access values
4937 Variables that are never assigned a value
4940 Variables that are referenced before being initialized
4943 Task entries with no corresponding @code{accept} statement
4946 Duplicate accepts for the same task entry in a @code{select}
4949 Objects that take too much storage
4952 Unchecked conversion between types of differing sizes
4955 Missing @code{return} statement along some execution path in a function
4958 Incorrect (unrecognized) pragmas
4961 Incorrect external names
4964 Allocation from empty storage pool
4967 Potentially blocking operation in protected type
4970 Suspicious parenthesization of expressions
4973 Mismatching bounds in an aggregate
4976 Attempt to return local value by reference
4979 Premature instantiation of a generic body
4982 Attempt to pack aliased components
4985 Out of bounds array subscripts
4988 Wrong length on string assignment
4991 Violations of style rules if style checking is enabled
4994 Unused @code{with} clauses
4997 @code{Bit_Order} usage that does not have any effect
5000 @code{Standard.Duration} used to resolve universal fixed expression
5003 Dereference of possibly null value
5006 Declaration that is likely to cause storage error
5009 Internal GNAT unit @code{with}'ed by application unit
5012 Values known to be out of range at compile time
5015 Unreferenced labels and variables
5018 Address overlays that could clobber memory
5021 Unexpected initialization when address clause present
5024 Bad alignment for address clause
5027 Useless type conversions
5030 Redundant assignment statements and other redundant constructs
5033 Useless exception handlers
5036 Accidental hiding of name by child unit
5039 Access before elaboration detected at compile time
5042 A range in a @code{for} loop that is known to be null or might be null
5047 The following section lists compiler switches that are available
5048 to control the handling of warning messages. It is also possible
5049 to exercise much finer control over what warnings are issued and
5050 suppressed using the GNAT pragma Warnings, @xref{Pragma Warnings,,,
5051 gnat_rm, GNAT Reference manual}.
5056 @emph{Activate most optional warnings.}
5057 @cindex @option{-gnatwa} (@command{gcc})
5058 This switch activates most optional warning messages. See the remaining list
5059 in this section for details on optional warning messages that can be
5060 individually controlled. The warnings that are not turned on by this
5062 @option{-gnatwd} (implicit dereferencing),
5063 @option{-gnatwh} (hiding),
5064 @option{-gnatw.h} (holes (gaps) in record layouts)
5065 @option{-gnatwl} (elaboration warnings),
5066 @option{-gnatw.o} (warn on values set by out parameters ignored)
5067 and @option{-gnatwt} (tracking of deleted conditional code).
5068 All other optional warnings are turned on.
5071 @emph{Suppress all optional errors.}
5072 @cindex @option{-gnatwA} (@command{gcc})
5073 This switch suppresses all optional warning messages, see remaining list
5074 in this section for details on optional warning messages that can be
5075 individually controlled.
5078 @emph{Activate warnings on failing assertions.}
5079 @cindex @option{-gnatw.a} (@command{gcc})
5080 @cindex Assert failures
5081 This switch activates warnings for assertions where the compiler can tell at
5082 compile time that the assertion will fail. Note that this warning is given
5083 even if assertions are disabled. The default is that such warnings are
5087 @emph{Suppress warnings on failing assertions.}
5088 @cindex @option{-gnatw.A} (@command{gcc})
5089 @cindex Assert failures
5090 This switch suppresses warnings for assertions where the compiler can tell at
5091 compile time that the assertion will fail.
5094 @emph{Activate warnings on bad fixed values.}
5095 @cindex @option{-gnatwb} (@command{gcc})
5096 @cindex Bad fixed values
5097 @cindex Fixed-point Small value
5099 This switch activates warnings for static fixed-point expressions whose
5100 value is not an exact multiple of Small. Such values are implementation
5101 dependent, since an implementation is free to choose either of the multiples
5102 that surround the value. GNAT always chooses the closer one, but this is not
5103 required behavior, and it is better to specify a value that is an exact
5104 multiple, ensuring predictable execution. The default is that such warnings
5108 @emph{Suppress warnings on bad fixed values.}
5109 @cindex @option{-gnatwB} (@command{gcc})
5110 This switch suppresses warnings for static fixed-point expressions whose
5111 value is not an exact multiple of Small.
5114 @emph{Activate warnings on biased representation.}
5115 @cindex @option{-gnatw.b} (@command{gcc})
5116 @cindex Biased representation
5117 This switch activates warnings when a size clause, value size clause, component
5118 clause, or component size clause forces the use of biased representation for an
5119 integer type (e.g. representing a range of 10..11 in a single bit by using 0/1
5120 to represent 10/11). The default is that such warnings are generated.
5123 @emph{Suppress warnings on biased representation.}
5124 @cindex @option{-gnatwB} (@command{gcc})
5125 This switch suppresses warnings for representation clauses that force the use
5126 of biased representation.
5129 @emph{Activate warnings on conditionals.}
5130 @cindex @option{-gnatwc} (@command{gcc})
5131 @cindex Conditionals, constant
5132 This switch activates warnings for conditional expressions used in
5133 tests that are known to be True or False at compile time. The default
5134 is that such warnings are not generated.
5135 Note that this warning does
5136 not get issued for the use of boolean variables or constants whose
5137 values are known at compile time, since this is a standard technique
5138 for conditional compilation in Ada, and this would generate too many
5139 false positive warnings.
5141 This warning option also activates a special test for comparisons using
5142 the operators ``>='' and`` <=''.
5143 If the compiler can tell that only the equality condition is possible,
5144 then it will warn that the ``>'' or ``<'' part of the test
5145 is useless and that the operator could be replaced by ``=''.
5146 An example would be comparing a @code{Natural} variable <= 0.
5148 This warning option also generates warnings if
5149 one or both tests is optimized away in a membership test for integer
5150 values if the result can be determined at compile time. Range tests on
5151 enumeration types are not included, since it is common for such tests
5152 to include an end point.
5154 This warning can also be turned on using @option{-gnatwa}.
5157 @emph{Suppress warnings on conditionals.}
5158 @cindex @option{-gnatwC} (@command{gcc})
5159 This switch suppresses warnings for conditional expressions used in
5160 tests that are known to be True or False at compile time.
5163 @emph{Activate warnings on missing component clauses.}
5164 @cindex @option{-gnatw.c} (@command{gcc})
5165 @cindex Component clause, missing
5166 This switch activates warnings for record components where a record
5167 representation clause is present and has component clauses for the
5168 majority, but not all, of the components. A warning is given for each
5169 component for which no component clause is present.
5171 This warning can also be turned on using @option{-gnatwa}.
5174 @emph{Suppress warnings on missing component clauses.}
5175 @cindex @option{-gnatwC} (@command{gcc})
5176 This switch suppresses warnings for record components that are
5177 missing a component clause in the situation described above.
5180 @emph{Activate warnings on implicit dereferencing.}
5181 @cindex @option{-gnatwd} (@command{gcc})
5182 If this switch is set, then the use of a prefix of an access type
5183 in an indexed component, slice, or selected component without an
5184 explicit @code{.all} will generate a warning. With this warning
5185 enabled, access checks occur only at points where an explicit
5186 @code{.all} appears in the source code (assuming no warnings are
5187 generated as a result of this switch). The default is that such
5188 warnings are not generated.
5189 Note that @option{-gnatwa} does not affect the setting of
5190 this warning option.
5193 @emph{Suppress warnings on implicit dereferencing.}
5194 @cindex @option{-gnatwD} (@command{gcc})
5195 @cindex Implicit dereferencing
5196 @cindex Dereferencing, implicit
5197 This switch suppresses warnings for implicit dereferences in
5198 indexed components, slices, and selected components.
5201 @emph{Treat warnings and style checks as errors.}
5202 @cindex @option{-gnatwe} (@command{gcc})
5203 @cindex Warnings, treat as error
5204 This switch causes warning messages and style check messages to be
5206 The warning string still appears, but the warning messages are counted
5207 as errors, and prevent the generation of an object file. Note that this
5208 is the only -gnatw switch that affects the handling of style check messages.
5211 @emph{Activate every optional warning}
5212 @cindex @option{-gnatw.e} (@command{gcc})
5213 @cindex Warnings, activate every optional warning
5214 This switch activates all optional warnings, including those which
5215 are not activated by @code{-gnatwa}.
5218 @emph{Activate warnings on unreferenced formals.}
5219 @cindex @option{-gnatwf} (@command{gcc})
5220 @cindex Formals, unreferenced
5221 This switch causes a warning to be generated if a formal parameter
5222 is not referenced in the body of the subprogram. This warning can
5223 also be turned on using @option{-gnatwa} or @option{-gnatwu}. The
5224 default is that these warnings are not generated.
5227 @emph{Suppress warnings on unreferenced formals.}
5228 @cindex @option{-gnatwF} (@command{gcc})
5229 This switch suppresses warnings for unreferenced formal
5230 parameters. Note that the
5231 combination @option{-gnatwu} followed by @option{-gnatwF} has the
5232 effect of warning on unreferenced entities other than subprogram
5236 @emph{Activate warnings on unrecognized pragmas.}
5237 @cindex @option{-gnatwg} (@command{gcc})
5238 @cindex Pragmas, unrecognized
5239 This switch causes a warning to be generated if an unrecognized
5240 pragma is encountered. Apart from issuing this warning, the
5241 pragma is ignored and has no effect. This warning can
5242 also be turned on using @option{-gnatwa}. The default
5243 is that such warnings are issued (satisfying the Ada Reference
5244 Manual requirement that such warnings appear).
5247 @emph{Suppress warnings on unrecognized pragmas.}
5248 @cindex @option{-gnatwG} (@command{gcc})
5249 This switch suppresses warnings for unrecognized pragmas.
5252 @emph{Activate warnings on hiding.}
5253 @cindex @option{-gnatwh} (@command{gcc})
5254 @cindex Hiding of Declarations
5255 This switch activates warnings on hiding declarations.
5256 A declaration is considered hiding
5257 if it is for a non-overloadable entity, and it declares an entity with the
5258 same name as some other entity that is directly or use-visible. The default
5259 is that such warnings are not generated.
5260 Note that @option{-gnatwa} does not affect the setting of this warning option.
5263 @emph{Suppress warnings on hiding.}
5264 @cindex @option{-gnatwH} (@command{gcc})
5265 This switch suppresses warnings on hiding declarations.
5268 @emph{Activate warnings on holes/gaps in records.}
5269 @cindex @option{-gnatw.h} (@command{gcc})
5270 @cindex Record Representation (gaps)
5271 This switch activates warnings on component clauses in record
5272 representation clauses that leave holes (gaps) in the record layout.
5273 If this warning option is active, then record representation clauses
5274 should specify a contiguous layout, adding unused fill fields if needed.
5275 Note that @option{-gnatwa} does not affect the setting of this warning option.
5278 @emph{Suppress warnings on holes/gaps in records.}
5279 @cindex @option{-gnatw.H} (@command{gcc})
5280 This switch suppresses warnings on component clauses in record
5281 representation clauses that leave holes (haps) in the record layout.
5284 @emph{Activate warnings on implementation units.}
5285 @cindex @option{-gnatwi} (@command{gcc})
5286 This switch activates warnings for a @code{with} of an internal GNAT
5287 implementation unit, defined as any unit from the @code{Ada},
5288 @code{Interfaces}, @code{GNAT},
5289 ^^@code{DEC},^ or @code{System}
5290 hierarchies that is not
5291 documented in either the Ada Reference Manual or the GNAT
5292 Programmer's Reference Manual. Such units are intended only
5293 for internal implementation purposes and should not be @code{with}'ed
5294 by user programs. The default is that such warnings are generated
5295 This warning can also be turned on using @option{-gnatwa}.
5298 @emph{Disable warnings on implementation units.}
5299 @cindex @option{-gnatwI} (@command{gcc})
5300 This switch disables warnings for a @code{with} of an internal GNAT
5301 implementation unit.
5304 @emph{Activate warnings on overlapping actuals.}
5305 @cindex @option{-gnatw.i} (@command{gcc})
5306 This switch enables a warning on statically detectable overlapping actuals in
5307 a subprogram call, when one of the actuals is an in-out parameter, and the
5308 types of the actuals are not by-copy types. The warning is off by default,
5309 and is not included under -gnatwa.
5312 @emph{Disable warnings on overlapping actuals.}
5313 @cindex @option{-gnatw.I} (@command{gcc})
5314 This switch disables warnings on overlapping actuals in a call..
5317 @emph{Activate warnings on obsolescent features (Annex J).}
5318 @cindex @option{-gnatwj} (@command{gcc})
5319 @cindex Features, obsolescent
5320 @cindex Obsolescent features
5321 If this warning option is activated, then warnings are generated for
5322 calls to subprograms marked with @code{pragma Obsolescent} and
5323 for use of features in Annex J of the Ada Reference Manual. In the
5324 case of Annex J, not all features are flagged. In particular use
5325 of the renamed packages (like @code{Text_IO}) and use of package
5326 @code{ASCII} are not flagged, since these are very common and
5327 would generate many annoying positive warnings. The default is that
5328 such warnings are not generated. This warning is also turned on by
5329 the use of @option{-gnatwa}.
5331 In addition to the above cases, warnings are also generated for
5332 GNAT features that have been provided in past versions but which
5333 have been superseded (typically by features in the new Ada standard).
5334 For example, @code{pragma Ravenscar} will be flagged since its
5335 function is replaced by @code{pragma Profile(Ravenscar)}.
5337 Note that this warning option functions differently from the
5338 restriction @code{No_Obsolescent_Features} in two respects.
5339 First, the restriction applies only to annex J features.
5340 Second, the restriction does flag uses of package @code{ASCII}.
5343 @emph{Suppress warnings on obsolescent features (Annex J).}
5344 @cindex @option{-gnatwJ} (@command{gcc})
5345 This switch disables warnings on use of obsolescent features.
5348 @emph{Activate warnings on variables that could be constants.}
5349 @cindex @option{-gnatwk} (@command{gcc})
5350 This switch activates warnings for variables that are initialized but
5351 never modified, and then could be declared constants. The default is that
5352 such warnings are not given.
5353 This warning can also be turned on using @option{-gnatwa}.
5356 @emph{Suppress warnings on variables that could be constants.}
5357 @cindex @option{-gnatwK} (@command{gcc})
5358 This switch disables warnings on variables that could be declared constants.
5361 @emph{Activate warnings for elaboration pragmas.}
5362 @cindex @option{-gnatwl} (@command{gcc})
5363 @cindex Elaboration, warnings
5364 This switch activates warnings on missing
5365 @code{Elaborate_All} and @code{Elaborate} pragmas.
5366 See the section in this guide on elaboration checking for details on
5367 when such pragmas should be used. In dynamic elaboration mode, this switch
5368 generations warnings about the need to add elaboration pragmas. Note however,
5369 that if you blindly follow these warnings, and add @code{Elaborate_All}
5370 warnings wherever they are recommended, you basically end up with the
5371 equivalent of the static elaboration model, which may not be what you want for
5372 legacy code for which the static model does not work.
5374 For the static model, the messages generated are labeled "info:" (for
5375 information messages). They are not warnings to add elaboration pragmas,
5376 merely informational messages showing what implicit elaboration pragmas
5377 have been added, for use in analyzing elaboration circularity problems.
5379 Warnings are also generated if you
5380 are using the static mode of elaboration, and a @code{pragma Elaborate}
5381 is encountered. The default is that such warnings
5383 This warning is not automatically turned on by the use of @option{-gnatwa}.
5386 @emph{Suppress warnings for elaboration pragmas.}
5387 @cindex @option{-gnatwL} (@command{gcc})
5388 This switch suppresses warnings on missing Elaborate and Elaborate_All pragmas.
5389 See the section in this guide on elaboration checking for details on
5390 when such pragmas should be used.
5393 @emph{Activate warnings on modified but unreferenced variables.}
5394 @cindex @option{-gnatwm} (@command{gcc})
5395 This switch activates warnings for variables that are assigned (using
5396 an initialization value or with one or more assignment statements) but
5397 whose value is never read. The warning is suppressed for volatile
5398 variables and also for variables that are renamings of other variables
5399 or for which an address clause is given.
5400 This warning can also be turned on using @option{-gnatwa}.
5401 The default is that these warnings are not given.
5404 @emph{Disable warnings on modified but unreferenced variables.}
5405 @cindex @option{-gnatwM} (@command{gcc})
5406 This switch disables warnings for variables that are assigned or
5407 initialized, but never read.
5410 @emph{Activate warnings on suspicious modulus values.}
5411 @cindex @option{-gnatw.m} (@command{gcc})
5412 This switch activates warnings for modulus values that seem suspicious.
5413 The cases caught are where the size is the same as the modulus (e.g.
5414 a modulus of 7 with a size of 7 bits), and modulus values of 32 or 64
5415 with no size clause. The guess in both cases is that 2**x was intended
5416 rather than x. The default is that these warnings are given.
5419 @emph{Disable warnings on suspicious modulus values.}
5420 @cindex @option{-gnatw.M} (@command{gcc})
5421 This switch disables warnings for suspicious modulus values.
5424 @emph{Set normal warnings mode.}
5425 @cindex @option{-gnatwn} (@command{gcc})
5426 This switch sets normal warning mode, in which enabled warnings are
5427 issued and treated as warnings rather than errors. This is the default
5428 mode. the switch @option{-gnatwn} can be used to cancel the effect of
5429 an explicit @option{-gnatws} or
5430 @option{-gnatwe}. It also cancels the effect of the
5431 implicit @option{-gnatwe} that is activated by the
5432 use of @option{-gnatg}.
5435 @emph{Activate warnings on address clause overlays.}
5436 @cindex @option{-gnatwo} (@command{gcc})
5437 @cindex Address Clauses, warnings
5438 This switch activates warnings for possibly unintended initialization
5439 effects of defining address clauses that cause one variable to overlap
5440 another. The default is that such warnings are generated.
5441 This warning can also be turned on using @option{-gnatwa}.
5444 @emph{Suppress warnings on address clause overlays.}
5445 @cindex @option{-gnatwO} (@command{gcc})
5446 This switch suppresses warnings on possibly unintended initialization
5447 effects of defining address clauses that cause one variable to overlap
5451 @emph{Activate warnings on modified but unreferenced out parameters.}
5452 @cindex @option{-gnatw.o} (@command{gcc})
5453 This switch activates warnings for variables that are modified by using
5454 them as actuals for a call to a procedure with an out mode formal, where
5455 the resulting assigned value is never read. It is applicable in the case
5456 where there is more than one out mode formal. If there is only one out
5457 mode formal, the warning is issued by default (controlled by -gnatwu).
5458 The warning is suppressed for volatile
5459 variables and also for variables that are renamings of other variables
5460 or for which an address clause is given.
5461 The default is that these warnings are not given. Note that this warning
5462 is not included in -gnatwa, it must be activated explicitly.
5465 @emph{Disable warnings on modified but unreferenced out parameters.}
5466 @cindex @option{-gnatw.O} (@command{gcc})
5467 This switch suppresses warnings for variables that are modified by using
5468 them as actuals for a call to a procedure with an out mode formal, where
5469 the resulting assigned value is never read.
5472 @emph{Activate warnings on ineffective pragma Inlines.}
5473 @cindex @option{-gnatwp} (@command{gcc})
5474 @cindex Inlining, warnings
5475 This switch activates warnings for failure of front end inlining
5476 (activated by @option{-gnatN}) to inline a particular call. There are
5477 many reasons for not being able to inline a call, including most
5478 commonly that the call is too complex to inline. The default is
5479 that such warnings are not given.
5480 This warning can also be turned on using @option{-gnatwa}.
5481 Warnings on ineffective inlining by the gcc back-end can be activated
5482 separately, using the gcc switch -Winline.
5485 @emph{Suppress warnings on ineffective pragma Inlines.}
5486 @cindex @option{-gnatwP} (@command{gcc})
5487 This switch suppresses warnings on ineffective pragma Inlines. If the
5488 inlining mechanism cannot inline a call, it will simply ignore the
5492 @emph{Activate warnings on parameter ordering.}
5493 @cindex @option{-gnatw.p} (@command{gcc})
5494 @cindex Parameter order, warnings
5495 This switch activates warnings for cases of suspicious parameter
5496 ordering when the list of arguments are all simple identifiers that
5497 match the names of the formals, but are in a different order. The
5498 warning is suppressed if any use of named parameter notation is used,
5499 so this is the appropriate way to suppress a false positive (and
5500 serves to emphasize that the "misordering" is deliberate). The
5502 that such warnings are not given.
5503 This warning can also be turned on using @option{-gnatwa}.
5506 @emph{Suppress warnings on parameter ordering.}
5507 @cindex @option{-gnatw.P} (@command{gcc})
5508 This switch suppresses warnings on cases of suspicious parameter
5512 @emph{Activate warnings on questionable missing parentheses.}
5513 @cindex @option{-gnatwq} (@command{gcc})
5514 @cindex Parentheses, warnings
5515 This switch activates warnings for cases where parentheses are not used and
5516 the result is potential ambiguity from a readers point of view. For example
5517 (not a > b) when a and b are modular means ((not a) > b) and very likely the
5518 programmer intended (not (a > b)). Similarly (-x mod 5) means (-(x mod 5)) and
5519 quite likely ((-x) mod 5) was intended. In such situations it seems best to
5520 follow the rule of always parenthesizing to make the association clear, and
5521 this warning switch warns if such parentheses are not present. The default
5522 is that these warnings are given.
5523 This warning can also be turned on using @option{-gnatwa}.
5526 @emph{Suppress warnings on questionable missing parentheses.}
5527 @cindex @option{-gnatwQ} (@command{gcc})
5528 This switch suppresses warnings for cases where the association is not
5529 clear and the use of parentheses is preferred.
5532 @emph{Activate warnings on redundant constructs.}
5533 @cindex @option{-gnatwr} (@command{gcc})
5534 This switch activates warnings for redundant constructs. The following
5535 is the current list of constructs regarded as redundant:
5539 Assignment of an item to itself.
5541 Type conversion that converts an expression to its own type.
5543 Use of the attribute @code{Base} where @code{typ'Base} is the same
5546 Use of pragma @code{Pack} when all components are placed by a record
5547 representation clause.
5549 Exception handler containing only a reraise statement (raise with no
5550 operand) which has no effect.
5552 Use of the operator abs on an operand that is known at compile time
5555 Comparison of boolean expressions to an explicit True value.
5558 This warning can also be turned on using @option{-gnatwa}.
5559 The default is that warnings for redundant constructs are not given.
5562 @emph{Suppress warnings on redundant constructs.}
5563 @cindex @option{-gnatwR} (@command{gcc})
5564 This switch suppresses warnings for redundant constructs.
5567 @emph{Activate warnings for object renaming function.}
5568 @cindex @option{-gnatw.r} (@command{gcc})
5569 This switch activates warnings for an object renaming that renames a
5570 function call, which is equivalent to a constant declaration (as
5571 opposed to renaming the function itself). The default is that these
5572 warnings are given. This warning can also be turned on using
5576 @emph{Suppress warnings for object renaming function.}
5577 @cindex @option{-gnatwT} (@command{gcc})
5578 This switch suppresses warnings for object renaming function.
5581 @emph{Suppress all warnings.}
5582 @cindex @option{-gnatws} (@command{gcc})
5583 This switch completely suppresses the
5584 output of all warning messages from the GNAT front end.
5585 Note that it does not suppress warnings from the @command{gcc} back end.
5586 To suppress these back end warnings as well, use the switch @option{-w}
5587 in addition to @option{-gnatws}. Also this switch has no effect on the
5588 handling of style check messages.
5591 @emph{Activate warnings on overridden size clauses.}
5592 @cindex @option{-gnatw.s} (@command{gcc})
5593 @cindex Record Representation (component sizes)
5594 This switch activates warnings on component clauses in record
5595 representation clauses where the length given overrides that
5596 specified by an explicit size clause for the component type. A
5597 warning is similarly given in the array case if a specified
5598 component size overrides an explicit size clause for the array
5600 Note that @option{-gnatwa} does not affect the setting of this warning option.
5603 @emph{Suppress warnings on overriddebn size clauses.}
5604 @cindex @option{-gnatw.S} (@command{gcc})
5605 This switch suppresses warnings on component clauses in record
5606 representation clauses that override size clauses, and similar
5607 warnings when an array component size overrides a size clause.
5610 @emph{Activate warnings for tracking of deleted conditional code.}
5611 @cindex @option{-gnatwt} (@command{gcc})
5612 @cindex Deactivated code, warnings
5613 @cindex Deleted code, warnings
5614 This switch activates warnings for tracking of code in conditionals (IF and
5615 CASE statements) that is detected to be dead code which cannot be executed, and
5616 which is removed by the front end. This warning is off by default, and is not
5617 turned on by @option{-gnatwa}, it has to be turned on explicitly. This may be
5618 useful for detecting deactivated code in certified applications.
5621 @emph{Suppress warnings for tracking of deleted conditional code.}
5622 @cindex @option{-gnatwT} (@command{gcc})
5623 This switch suppresses warnings for tracking of deleted conditional code.
5626 @emph{Activate warnings on unused entities.}
5627 @cindex @option{-gnatwu} (@command{gcc})
5628 This switch activates warnings to be generated for entities that
5629 are declared but not referenced, and for units that are @code{with}'ed
5631 referenced. In the case of packages, a warning is also generated if
5632 no entities in the package are referenced. This means that if the package
5633 is referenced but the only references are in @code{use}
5634 clauses or @code{renames}
5635 declarations, a warning is still generated. A warning is also generated
5636 for a generic package that is @code{with}'ed but never instantiated.
5637 In the case where a package or subprogram body is compiled, and there
5638 is a @code{with} on the corresponding spec
5639 that is only referenced in the body,
5640 a warning is also generated, noting that the
5641 @code{with} can be moved to the body. The default is that
5642 such warnings are not generated.
5643 This switch also activates warnings on unreferenced formals
5644 (it includes the effect of @option{-gnatwf}).
5645 This warning can also be turned on using @option{-gnatwa}.
5648 @emph{Suppress warnings on unused entities.}
5649 @cindex @option{-gnatwU} (@command{gcc})
5650 This switch suppresses warnings for unused entities and packages.
5651 It also turns off warnings on unreferenced formals (and thus includes
5652 the effect of @option{-gnatwF}).
5655 @emph{Activate warnings on unordered enumeration types.}
5656 @cindex @option{-gnatw.u} (@command{gcc})
5657 This switch causes enumeration types to be considered as conceptually
5658 unordered, unless an explicit pragma @code{Ordered} is given for the type.
5659 The effect is to generate warnings in clients that use explicit comparisons
5660 or subranges, since these constructs both treat objects of the type as
5661 ordered. (A @emph{client} is defined as a unit that is other than the unit in
5662 which the type is declared, or its body or subunits.) Please refer to
5663 the description of pragma @code{Ordered} in the
5664 @cite{@value{EDITION} Reference Manual} for further details.
5667 @emph{Deactivate warnings on unordered enumeration types.}
5668 @cindex @option{-gnatw.U} (@command{gcc})
5669 This switch causes all enumeration types to be considered as ordered, so
5670 that no warnings are given for comparisons or subranges for any type.
5673 @emph{Activate warnings on unassigned variables.}
5674 @cindex @option{-gnatwv} (@command{gcc})
5675 @cindex Unassigned variable warnings
5676 This switch activates warnings for access to variables which
5677 may not be properly initialized. The default is that
5678 such warnings are generated.
5679 This warning can also be turned on using @option{-gnatwa}.
5682 @emph{Suppress warnings on unassigned variables.}
5683 @cindex @option{-gnatwV} (@command{gcc})
5684 This switch suppresses warnings for access to variables which
5685 may not be properly initialized.
5686 For variables of a composite type, the warning can also be suppressed in
5687 Ada 2005 by using a default initialization with a box. For example, if
5688 Table is an array of records whose components are only partially uninitialized,
5689 then the following code:
5691 @smallexample @c ada
5692 Tab : Table := (others => <>);
5695 will suppress warnings on subsequent statements that access components
5699 @emph{Activate warnings on wrong low bound assumption.}
5700 @cindex @option{-gnatww} (@command{gcc})
5701 @cindex String indexing warnings
5702 This switch activates warnings for indexing an unconstrained string parameter
5703 with a literal or S'Length. This is a case where the code is assuming that the
5704 low bound is one, which is in general not true (for example when a slice is
5705 passed). The default is that such warnings are generated.
5706 This warning can also be turned on using @option{-gnatwa}.
5709 @emph{Suppress warnings on wrong low bound assumption.}
5710 @cindex @option{-gnatwW} (@command{gcc})
5711 This switch suppresses warnings for indexing an unconstrained string parameter
5712 with a literal or S'Length. Note that this warning can also be suppressed
5713 in a particular case by adding an
5714 assertion that the lower bound is 1,
5715 as shown in the following example.
5717 @smallexample @c ada
5718 procedure K (S : String) is
5719 pragma Assert (S'First = 1);
5724 @emph{Activate warnings on unnecessary Warnings Off pragmas}
5725 @cindex @option{-gnatw.w} (@command{gcc})
5726 @cindex Warnings Off control
5727 This switch activates warnings for use of @code{pragma Warnings (Off, entity}
5728 where either the pragma is entirely useless (because it suppresses no
5729 warnings), or it could be replaced by @code{pragma Unreferenced} or
5730 @code{pragma Unmodified}.The default is that these warnings are not given.
5731 Note that this warning is not included in -gnatwa, it must be
5732 activated explicitly.
5735 @emph{Suppress warnings on unnecessary Warnings Off pragmas}
5736 @cindex @option{-gnatw.W} (@command{gcc})
5737 This switch suppresses warnings for use of @code{pragma Warnings (Off, entity}.
5740 @emph{Activate warnings on Export/Import pragmas.}
5741 @cindex @option{-gnatwx} (@command{gcc})
5742 @cindex Export/Import pragma warnings
5743 This switch activates warnings on Export/Import pragmas when
5744 the compiler detects a possible conflict between the Ada and
5745 foreign language calling sequences. For example, the use of
5746 default parameters in a convention C procedure is dubious
5747 because the C compiler cannot supply the proper default, so
5748 a warning is issued. The default is that such warnings are
5750 This warning can also be turned on using @option{-gnatwa}.
5753 @emph{Suppress warnings on Export/Import pragmas.}
5754 @cindex @option{-gnatwX} (@command{gcc})
5755 This switch suppresses warnings on Export/Import pragmas.
5756 The sense of this is that you are telling the compiler that
5757 you know what you are doing in writing the pragma, and it
5758 should not complain at you.
5761 @emph{Activate warnings for No_Exception_Propagation mode.}
5762 @cindex @option{-gnatwm} (@command{gcc})
5763 This switch activates warnings for exception usage when pragma Restrictions
5764 (No_Exception_Propagation) is in effect. Warnings are given for implicit or
5765 explicit exception raises which are not covered by a local handler, and for
5766 exception handlers which do not cover a local raise. The default is that these
5767 warnings are not given.
5770 @emph{Disable warnings for No_Exception_Propagation mode.}
5771 This switch disables warnings for exception usage when pragma Restrictions
5772 (No_Exception_Propagation) is in effect.
5775 @emph{Activate warnings for Ada 2005 compatibility issues.}
5776 @cindex @option{-gnatwy} (@command{gcc})
5777 @cindex Ada 2005 compatibility issues warnings
5778 For the most part Ada 2005 is upwards compatible with Ada 95,
5779 but there are some exceptions (for example the fact that
5780 @code{interface} is now a reserved word in Ada 2005). This
5781 switch activates several warnings to help in identifying
5782 and correcting such incompatibilities. The default is that
5783 these warnings are generated. Note that at one point Ada 2005
5784 was called Ada 0Y, hence the choice of character.
5785 This warning can also be turned on using @option{-gnatwa}.
5788 @emph{Disable warnings for Ada 2005 compatibility issues.}
5789 @cindex @option{-gnatwY} (@command{gcc})
5790 @cindex Ada 2005 compatibility issues warnings
5791 This switch suppresses several warnings intended to help in identifying
5792 incompatibilities between Ada 95 and Ada 2005.
5795 @emph{Activate warnings on unchecked conversions.}
5796 @cindex @option{-gnatwz} (@command{gcc})
5797 @cindex Unchecked_Conversion warnings
5798 This switch activates warnings for unchecked conversions
5799 where the types are known at compile time to have different
5801 is that such warnings are generated. Warnings are also
5802 generated for subprogram pointers with different conventions,
5803 and, on VMS only, for data pointers with different conventions.
5804 This warning can also be turned on using @option{-gnatwa}.
5807 @emph{Suppress warnings on unchecked conversions.}
5808 @cindex @option{-gnatwZ} (@command{gcc})
5809 This switch suppresses warnings for unchecked conversions
5810 where the types are known at compile time to have different
5811 sizes or conventions.
5813 @item ^-Wunused^WARNINGS=UNUSED^
5814 @cindex @option{-Wunused}
5815 The warnings controlled by the @option{-gnatw} switch are generated by
5816 the front end of the compiler. The @option{GCC} back end can provide
5817 additional warnings and they are controlled by the @option{-W} switch.
5818 For example, @option{^-Wunused^WARNINGS=UNUSED^} activates back end
5819 warnings for entities that are declared but not referenced.
5821 @item ^-Wuninitialized^WARNINGS=UNINITIALIZED^
5822 @cindex @option{-Wuninitialized}
5823 Similarly, @option{^-Wuninitialized^WARNINGS=UNINITIALIZED^} activates
5824 the back end warning for uninitialized variables. This switch must be
5825 used in conjunction with an optimization level greater than zero.
5827 @item ^-Wall^/ALL_BACK_END_WARNINGS^
5828 @cindex @option{-Wall}
5829 This switch enables all the above warnings from the @option{GCC} back end.
5830 The code generator detects a number of warning situations that are missed
5831 by the @option{GNAT} front end, and this switch can be used to activate them.
5832 The use of this switch also sets the default front end warning mode to
5833 @option{-gnatwa}, that is, most front end warnings activated as well.
5835 @item ^-w^/NO_BACK_END_WARNINGS^
5837 Conversely, this switch suppresses warnings from the @option{GCC} back end.
5838 The use of this switch also sets the default front end warning mode to
5839 @option{-gnatws}, that is, front end warnings suppressed as well.
5845 A string of warning parameters can be used in the same parameter. For example:
5852 will turn on all optional warnings except for elaboration pragma warnings,
5853 and also specify that warnings should be treated as errors.
5855 When no switch @option{^-gnatw^/WARNINGS^} is used, this is equivalent to:
5880 @node Debugging and Assertion Control
5881 @subsection Debugging and Assertion Control
5885 @cindex @option{-gnata} (@command{gcc})
5891 The pragmas @code{Assert} and @code{Debug} normally have no effect and
5892 are ignored. This switch, where @samp{a} stands for assert, causes
5893 @code{Assert} and @code{Debug} pragmas to be activated.
5895 The pragmas have the form:
5899 @b{pragma} Assert (@var{Boolean-expression} @r{[},
5900 @var{static-string-expression}@r{]})
5901 @b{pragma} Debug (@var{procedure call})
5906 The @code{Assert} pragma causes @var{Boolean-expression} to be tested.
5907 If the result is @code{True}, the pragma has no effect (other than
5908 possible side effects from evaluating the expression). If the result is
5909 @code{False}, the exception @code{Assert_Failure} declared in the package
5910 @code{System.Assertions} is
5911 raised (passing @var{static-string-expression}, if present, as the
5912 message associated with the exception). If no string expression is
5913 given the default is a string giving the file name and line number
5916 The @code{Debug} pragma causes @var{procedure} to be called. Note that
5917 @code{pragma Debug} may appear within a declaration sequence, allowing
5918 debugging procedures to be called between declarations.
5921 @item /DEBUG@r{[}=debug-level@r{]}
5923 Specifies how much debugging information is to be included in
5924 the resulting object file where 'debug-level' is one of the following:
5927 Include both debugger symbol records and traceback
5929 This is the default setting.
5931 Include both debugger symbol records and traceback in
5934 Excludes both debugger symbol records and traceback
5935 the object file. Same as /NODEBUG.
5937 Includes only debugger symbol records in the object
5938 file. Note that this doesn't include traceback information.
5943 @node Validity Checking
5944 @subsection Validity Checking
5945 @findex Validity Checking
5948 The Ada Reference Manual defines the concept of invalid values (see
5949 RM 13.9.1). The primary source of invalid values is uninitialized
5950 variables. A scalar variable that is left uninitialized may contain
5951 an invalid value; the concept of invalid does not apply to access or
5954 It is an error to read an invalid value, but the RM does not require
5955 run-time checks to detect such errors, except for some minimal
5956 checking to prevent erroneous execution (i.e. unpredictable
5957 behavior). This corresponds to the @option{-gnatVd} switch below,
5958 which is the default. For example, by default, if the expression of a
5959 case statement is invalid, it will raise Constraint_Error rather than
5960 causing a wild jump, and if an array index on the left-hand side of an
5961 assignment is invalid, it will raise Constraint_Error rather than
5962 overwriting an arbitrary memory location.
5964 The @option{-gnatVa} may be used to enable additional validity checks,
5965 which are not required by the RM. These checks are often very
5966 expensive (which is why the RM does not require them). These checks
5967 are useful in tracking down uninitialized variables, but they are
5968 not usually recommended for production builds.
5970 The other @option{-gnatV^@var{x}^^} switches below allow finer-grained
5971 control; you can enable whichever validity checks you desire. However,
5972 for most debugging purposes, @option{-gnatVa} is sufficient, and the
5973 default @option{-gnatVd} (i.e. standard Ada behavior) is usually
5974 sufficient for non-debugging use.
5976 The @option{-gnatB} switch tells the compiler to assume that all
5977 values are valid (that is, within their declared subtype range)
5978 except in the context of a use of the Valid attribute. This means
5979 the compiler can generate more efficient code, since the range
5980 of values is better known at compile time. However, an uninitialized
5981 variable can cause wild jumps and memory corruption in this mode.
5983 The @option{-gnatV^@var{x}^^} switch allows control over the validity
5984 checking mode as described below.
5986 The @code{x} argument is a string of letters that
5987 indicate validity checks that are performed or not performed in addition
5988 to the default checks required by Ada as described above.
5991 The options allowed for this qualifier
5992 indicate validity checks that are performed or not performed in addition
5993 to the default checks required by Ada as described above.
5999 @emph{All validity checks.}
6000 @cindex @option{-gnatVa} (@command{gcc})
6001 All validity checks are turned on.
6003 That is, @option{-gnatVa} is
6004 equivalent to @option{gnatVcdfimorst}.
6008 @emph{Validity checks for copies.}
6009 @cindex @option{-gnatVc} (@command{gcc})
6010 The right hand side of assignments, and the initializing values of
6011 object declarations are validity checked.
6014 @emph{Default (RM) validity checks.}
6015 @cindex @option{-gnatVd} (@command{gcc})
6016 Some validity checks are done by default following normal Ada semantics
6018 A check is done in case statements that the expression is within the range
6019 of the subtype. If it is not, Constraint_Error is raised.
6020 For assignments to array components, a check is done that the expression used
6021 as index is within the range. If it is not, Constraint_Error is raised.
6022 Both these validity checks may be turned off using switch @option{-gnatVD}.
6023 They are turned on by default. If @option{-gnatVD} is specified, a subsequent
6024 switch @option{-gnatVd} will leave the checks turned on.
6025 Switch @option{-gnatVD} should be used only if you are sure that all such
6026 expressions have valid values. If you use this switch and invalid values
6027 are present, then the program is erroneous, and wild jumps or memory
6028 overwriting may occur.
6031 @emph{Validity checks for elementary components.}
6032 @cindex @option{-gnatVe} (@command{gcc})
6033 In the absence of this switch, assignments to record or array components are
6034 not validity checked, even if validity checks for assignments generally
6035 (@option{-gnatVc}) are turned on. In Ada, assignment of composite values do not
6036 require valid data, but assignment of individual components does. So for
6037 example, there is a difference between copying the elements of an array with a
6038 slice assignment, compared to assigning element by element in a loop. This
6039 switch allows you to turn off validity checking for components, even when they
6040 are assigned component by component.
6043 @emph{Validity checks for floating-point values.}
6044 @cindex @option{-gnatVf} (@command{gcc})
6045 In the absence of this switch, validity checking occurs only for discrete
6046 values. If @option{-gnatVf} is specified, then validity checking also applies
6047 for floating-point values, and NaNs and infinities are considered invalid,
6048 as well as out of range values for constrained types. Note that this means
6049 that standard IEEE infinity mode is not allowed. The exact contexts
6050 in which floating-point values are checked depends on the setting of other
6051 options. For example,
6052 @option{^-gnatVif^VALIDITY_CHECKING=(IN_PARAMS,FLOATS)^} or
6053 @option{^-gnatVfi^VALIDITY_CHECKING=(FLOATS,IN_PARAMS)^}
6054 (the order does not matter) specifies that floating-point parameters of mode
6055 @code{in} should be validity checked.
6058 @emph{Validity checks for @code{in} mode parameters}
6059 @cindex @option{-gnatVi} (@command{gcc})
6060 Arguments for parameters of mode @code{in} are validity checked in function
6061 and procedure calls at the point of call.
6064 @emph{Validity checks for @code{in out} mode parameters.}
6065 @cindex @option{-gnatVm} (@command{gcc})
6066 Arguments for parameters of mode @code{in out} are validity checked in
6067 procedure calls at the point of call. The @code{'m'} here stands for
6068 modify, since this concerns parameters that can be modified by the call.
6069 Note that there is no specific option to test @code{out} parameters,
6070 but any reference within the subprogram will be tested in the usual
6071 manner, and if an invalid value is copied back, any reference to it
6072 will be subject to validity checking.
6075 @emph{No validity checks.}
6076 @cindex @option{-gnatVn} (@command{gcc})
6077 This switch turns off all validity checking, including the default checking
6078 for case statements and left hand side subscripts. Note that the use of
6079 the switch @option{-gnatp} suppresses all run-time checks, including
6080 validity checks, and thus implies @option{-gnatVn}. When this switch
6081 is used, it cancels any other @option{-gnatV} previously issued.
6084 @emph{Validity checks for operator and attribute operands.}
6085 @cindex @option{-gnatVo} (@command{gcc})
6086 Arguments for predefined operators and attributes are validity checked.
6087 This includes all operators in package @code{Standard},
6088 the shift operators defined as intrinsic in package @code{Interfaces}
6089 and operands for attributes such as @code{Pos}. Checks are also made
6090 on individual component values for composite comparisons, and on the
6091 expressions in type conversions and qualified expressions. Checks are
6092 also made on explicit ranges using @samp{..} (e.g.@: slices, loops etc).
6095 @emph{Validity checks for parameters.}
6096 @cindex @option{-gnatVp} (@command{gcc})
6097 This controls the treatment of parameters within a subprogram (as opposed
6098 to @option{-gnatVi} and @option{-gnatVm} which control validity testing
6099 of parameters on a call. If either of these call options is used, then
6100 normally an assumption is made within a subprogram that the input arguments
6101 have been validity checking at the point of call, and do not need checking
6102 again within a subprogram). If @option{-gnatVp} is set, then this assumption
6103 is not made, and parameters are not assumed to be valid, so their validity
6104 will be checked (or rechecked) within the subprogram.
6107 @emph{Validity checks for function returns.}
6108 @cindex @option{-gnatVr} (@command{gcc})
6109 The expression in @code{return} statements in functions is validity
6113 @emph{Validity checks for subscripts.}
6114 @cindex @option{-gnatVs} (@command{gcc})
6115 All subscripts expressions are checked for validity, whether they appear
6116 on the right side or left side (in default mode only left side subscripts
6117 are validity checked).
6120 @emph{Validity checks for tests.}
6121 @cindex @option{-gnatVt} (@command{gcc})
6122 Expressions used as conditions in @code{if}, @code{while} or @code{exit}
6123 statements are checked, as well as guard expressions in entry calls.
6128 The @option{-gnatV} switch may be followed by
6129 ^a string of letters^a list of options^
6130 to turn on a series of validity checking options.
6132 @option{^-gnatVcr^/VALIDITY_CHECKING=(COPIES, RETURNS)^}
6133 specifies that in addition to the default validity checking, copies and
6134 function return expressions are to be validity checked.
6135 In order to make it easier
6136 to specify the desired combination of effects,
6138 the upper case letters @code{CDFIMORST} may
6139 be used to turn off the corresponding lower case option.
6142 the prefix @code{NO} on an option turns off the corresponding validity
6145 @item @code{NOCOPIES}
6146 @item @code{NODEFAULT}
6147 @item @code{NOFLOATS}
6148 @item @code{NOIN_PARAMS}
6149 @item @code{NOMOD_PARAMS}
6150 @item @code{NOOPERANDS}
6151 @item @code{NORETURNS}
6152 @item @code{NOSUBSCRIPTS}
6153 @item @code{NOTESTS}
6157 @option{^-gnatVaM^/VALIDITY_CHECKING=(ALL, NOMOD_PARAMS)^}
6158 turns on all validity checking options except for
6159 checking of @code{@b{in out}} procedure arguments.
6161 The specification of additional validity checking generates extra code (and
6162 in the case of @option{-gnatVa} the code expansion can be substantial).
6163 However, these additional checks can be very useful in detecting
6164 uninitialized variables, incorrect use of unchecked conversion, and other
6165 errors leading to invalid values. The use of pragma @code{Initialize_Scalars}
6166 is useful in conjunction with the extra validity checking, since this
6167 ensures that wherever possible uninitialized variables have invalid values.
6169 See also the pragma @code{Validity_Checks} which allows modification of
6170 the validity checking mode at the program source level, and also allows for
6171 temporary disabling of validity checks.
6173 @node Style Checking
6174 @subsection Style Checking
6175 @findex Style checking
6178 The @option{-gnaty^x^(option,option,@dots{})^} switch
6179 @cindex @option{-gnaty} (@command{gcc})
6180 causes the compiler to
6181 enforce specified style rules. A limited set of style rules has been used
6182 in writing the GNAT sources themselves. This switch allows user programs
6183 to activate all or some of these checks. If the source program fails a
6184 specified style check, an appropriate message is given, preceded by
6185 the character sequence ``(style)''. This message does not prevent
6186 successful compilation (unless the @option{-gnatwe} switch is used).
6188 Note that this is by no means intended to be a general facility for
6189 checking arbitrary coding standards. It is simply an embedding of the
6190 style rules we have chosen for the GNAT sources. If you are starting
6191 a project which does not have established style standards, you may
6192 find it useful to adopt the entire set of GNAT coding standards, or
6193 some subset of them. If you already have an established set of coding
6194 standards, then it may be that selected style checking options do
6195 indeed correspond to choices you have made, but for general checking
6196 of an existing set of coding rules, you should look to the gnatcheck
6197 tool, which is designed for that purpose.
6200 @code{(option,option,@dots{})} is a sequence of keywords
6203 The string @var{x} is a sequence of letters or digits
6205 indicating the particular style
6206 checks to be performed. The following checks are defined:
6211 @emph{Specify indentation level.}
6212 If a digit from 1-9 appears
6213 ^in the string after @option{-gnaty}^as an option for /STYLE_CHECKS^
6214 then proper indentation is checked, with the digit indicating the
6215 indentation level required. A value of zero turns off this style check.
6216 The general style of required indentation is as specified by
6217 the examples in the Ada Reference Manual. Full line comments must be
6218 aligned with the @code{--} starting on a column that is a multiple of
6219 the alignment level, or they may be aligned the same way as the following
6220 non-blank line (this is useful when full line comments appear in the middle
6224 @emph{Check attribute casing.}
6225 Attribute names, including the case of keywords such as @code{digits}
6226 used as attributes names, must be written in mixed case, that is, the
6227 initial letter and any letter following an underscore must be uppercase.
6228 All other letters must be lowercase.
6230 @item ^A^ARRAY_INDEXES^
6231 @emph{Use of array index numbers in array attributes.}
6232 When using the array attributes First, Last, Range,
6233 or Length, the index number must be omitted for one-dimensional arrays
6234 and is required for multi-dimensional arrays.
6237 @emph{Blanks not allowed at statement end.}
6238 Trailing blanks are not allowed at the end of statements. The purpose of this
6239 rule, together with h (no horizontal tabs), is to enforce a canonical format
6240 for the use of blanks to separate source tokens.
6242 @item ^B^BOOLEAN_OPERATORS^
6243 @emph{Check Boolean operators.}
6244 The use of AND/OR operators is not permitted except in the cases of modular
6245 operands, array operands, and simple stand-alone boolean variables or
6246 boolean constants. In all other cases AND THEN/OR ELSE are required.
6249 @emph{Check comments.}
6250 Comments must meet the following set of rules:
6255 The ``@code{--}'' that starts the column must either start in column one,
6256 or else at least one blank must precede this sequence.
6259 Comments that follow other tokens on a line must have at least one blank
6260 following the ``@code{--}'' at the start of the comment.
6263 Full line comments must have at least two blanks following the
6264 ``@code{--}'' that starts the comment, with the following exceptions.
6267 A line consisting only of the ``@code{--}'' characters, possibly preceded
6268 by blanks is permitted.
6271 A comment starting with ``@code{--x}'' where @code{x} is a special character
6273 This allows proper processing of the output generated by specialized tools
6274 including @command{gnatprep} (where ``@code{--!}'' is used) and the SPARK
6276 language (where ``@code{--#}'' is used). For the purposes of this rule, a
6277 special character is defined as being in one of the ASCII ranges
6278 @code{16#21#@dots{}16#2F#} or @code{16#3A#@dots{}16#3F#}.
6279 Note that this usage is not permitted
6280 in GNAT implementation units (i.e., when @option{-gnatg} is used).
6283 A line consisting entirely of minus signs, possibly preceded by blanks, is
6284 permitted. This allows the construction of box comments where lines of minus
6285 signs are used to form the top and bottom of the box.
6288 A comment that starts and ends with ``@code{--}'' is permitted as long as at
6289 least one blank follows the initial ``@code{--}''. Together with the preceding
6290 rule, this allows the construction of box comments, as shown in the following
6293 ---------------------------
6294 -- This is a box comment --
6295 -- with two text lines. --
6296 ---------------------------
6300 @item ^d^DOS_LINE_ENDINGS^
6301 @emph{Check no DOS line terminators present.}
6302 All lines must be terminated by a single ASCII.LF
6303 character (in particular the DOS line terminator sequence CR/LF is not
6307 @emph{Check end/exit labels.}
6308 Optional labels on @code{end} statements ending subprograms and on
6309 @code{exit} statements exiting named loops, are required to be present.
6312 @emph{No form feeds or vertical tabs.}
6313 Neither form feeds nor vertical tab characters are permitted
6317 @emph{GNAT style mode}
6318 The set of style check switches is set to match that used by the GNAT sources.
6319 This may be useful when developing code that is eventually intended to be
6320 incorporated into GNAT. For further details, see GNAT sources.
6323 @emph{No horizontal tabs.}
6324 Horizontal tab characters are not permitted in the source text.
6325 Together with the b (no blanks at end of line) check, this
6326 enforces a canonical form for the use of blanks to separate
6330 @emph{Check if-then layout.}
6331 The keyword @code{then} must appear either on the same
6332 line as corresponding @code{if}, or on a line on its own, lined
6333 up under the @code{if} with at least one non-blank line in between
6334 containing all or part of the condition to be tested.
6337 @emph{check mode IN keywords}
6338 Mode @code{in} (the default mode) is not
6339 allowed to be given explicitly. @code{in out} is fine,
6340 but not @code{in} on its own.
6343 @emph{Check keyword casing.}
6344 All keywords must be in lower case (with the exception of keywords
6345 such as @code{digits} used as attribute names to which this check
6349 @emph{Check layout.}
6350 Layout of statement and declaration constructs must follow the
6351 recommendations in the Ada Reference Manual, as indicated by the
6352 form of the syntax rules. For example an @code{else} keyword must
6353 be lined up with the corresponding @code{if} keyword.
6355 There are two respects in which the style rule enforced by this check
6356 option are more liberal than those in the Ada Reference Manual. First
6357 in the case of record declarations, it is permissible to put the
6358 @code{record} keyword on the same line as the @code{type} keyword, and
6359 then the @code{end} in @code{end record} must line up under @code{type}.
6360 This is also permitted when the type declaration is split on two lines.
6361 For example, any of the following three layouts is acceptable:
6363 @smallexample @c ada
6386 Second, in the case of a block statement, a permitted alternative
6387 is to put the block label on the same line as the @code{declare} or
6388 @code{begin} keyword, and then line the @code{end} keyword up under
6389 the block label. For example both the following are permitted:
6391 @smallexample @c ada
6409 The same alternative format is allowed for loops. For example, both of
6410 the following are permitted:
6412 @smallexample @c ada
6414 Clear : while J < 10 loop
6425 @item ^Lnnn^MAX_NESTING=nnn^
6426 @emph{Set maximum nesting level}
6427 The maximum level of nesting of constructs (including subprograms, loops,
6428 blocks, packages, and conditionals) may not exceed the given value
6429 @option{nnn}. A value of zero disconnects this style check.
6431 @item ^m^LINE_LENGTH^
6432 @emph{Check maximum line length.}
6433 The length of source lines must not exceed 79 characters, including
6434 any trailing blanks. The value of 79 allows convenient display on an
6435 80 character wide device or window, allowing for possible special
6436 treatment of 80 character lines. Note that this count is of
6437 characters in the source text. This means that a tab character counts
6438 as one character in this count but a wide character sequence counts as
6439 a single character (however many bytes are needed in the encoding).
6441 @item ^Mnnn^MAX_LENGTH=nnn^
6442 @emph{Set maximum line length.}
6443 The length of lines must not exceed the
6444 given value @option{nnn}. The maximum value that can be specified is 32767.
6446 @item ^n^STANDARD_CASING^
6447 @emph{Check casing of entities in Standard.}
6448 Any identifier from Standard must be cased
6449 to match the presentation in the Ada Reference Manual (for example,
6450 @code{Integer} and @code{ASCII.NUL}).
6453 @emph{Turn off all style checks}
6454 All style check options are turned off.
6456 @item ^o^ORDERED_SUBPROGRAMS^
6457 @emph{Check order of subprogram bodies.}
6458 All subprogram bodies in a given scope
6459 (e.g.@: a package body) must be in alphabetical order. The ordering
6460 rule uses normal Ada rules for comparing strings, ignoring casing
6461 of letters, except that if there is a trailing numeric suffix, then
6462 the value of this suffix is used in the ordering (e.g.@: Junk2 comes
6465 @item ^O^OVERRIDING_INDICATORS^
6466 @emph{Check that overriding subprograms are explicitly marked as such.}
6467 The declaration of a primitive operation of a type extension that overrides
6468 an inherited operation must carry an overriding indicator.
6471 @emph{Check pragma casing.}
6472 Pragma names must be written in mixed case, that is, the
6473 initial letter and any letter following an underscore must be uppercase.
6474 All other letters must be lowercase.
6476 @item ^r^REFERENCES^
6477 @emph{Check references.}
6478 All identifier references must be cased in the same way as the
6479 corresponding declaration. No specific casing style is imposed on
6480 identifiers. The only requirement is for consistency of references
6483 @item ^S^STATEMENTS_AFTER_THEN_ELSE^
6484 @emph{Check no statements after THEN/ELSE.}
6485 No statements are allowed
6486 on the same line as a THEN or ELSE keyword following the
6487 keyword in an IF statement. OR ELSE and AND THEN are not affected,
6488 and a special exception allows a pragma to appear after ELSE.
6491 @emph{Check separate specs.}
6492 Separate declarations (``specs'') are required for subprograms (a
6493 body is not allowed to serve as its own declaration). The only
6494 exception is that parameterless library level procedures are
6495 not required to have a separate declaration. This exception covers
6496 the most frequent form of main program procedures.
6499 @emph{Check token spacing.}
6500 The following token spacing rules are enforced:
6505 The keywords @code{@b{abs}} and @code{@b{not}} must be followed by a space.
6508 The token @code{=>} must be surrounded by spaces.
6511 The token @code{<>} must be preceded by a space or a left parenthesis.
6514 Binary operators other than @code{**} must be surrounded by spaces.
6515 There is no restriction on the layout of the @code{**} binary operator.
6518 Colon must be surrounded by spaces.
6521 Colon-equal (assignment, initialization) must be surrounded by spaces.
6524 Comma must be the first non-blank character on the line, or be
6525 immediately preceded by a non-blank character, and must be followed
6529 If the token preceding a left parenthesis ends with a letter or digit, then
6530 a space must separate the two tokens.
6533 if the token following a right parenthesis starts with a letter or digit, then
6534 a space must separate the two tokens.
6537 A right parenthesis must either be the first non-blank character on
6538 a line, or it must be preceded by a non-blank character.
6541 A semicolon must not be preceded by a space, and must not be followed by
6542 a non-blank character.
6545 A unary plus or minus may not be followed by a space.
6548 A vertical bar must be surrounded by spaces.
6551 @item ^u^UNNECESSARY_BLANK_LINES^
6552 @emph{Check unnecessary blank lines.}
6553 Unnecessary blank lines are not allowed. A blank line is considered
6554 unnecessary if it appears at the end of the file, or if more than
6555 one blank line occurs in sequence.
6557 @item ^x^XTRA_PARENS^
6558 @emph{Check extra parentheses.}
6559 Unnecessary extra level of parentheses (C-style) are not allowed
6560 around conditions in @code{if} statements, @code{while} statements and
6561 @code{exit} statements.
6563 @item ^y^ALL_BUILTIN^
6564 @emph{Set all standard style check options}
6565 This is equivalent to @code{gnaty3aAbcefhiklmnprst}, that is all checking
6566 options enabled with the exception of @option{-gnatyo}, @option{-gnatyI},
6567 @option{-gnatyS}, @option{-gnatyLnnn},
6568 @option{-gnatyd}, @option{-gnatyu}, and @option{-gnatyx}.
6572 @emph{Remove style check options}
6573 This causes any subsequent options in the string to act as canceling the
6574 corresponding style check option. To cancel maximum nesting level control,
6575 use @option{L} parameter witout any integer value after that, because any
6576 digit following @option{-} in the parameter string of the @option{-gnaty}
6577 option will be threated as canceling indentation check. The same is true
6578 for @option{M} parameter. @option{y} and @option{N} parameters are not
6579 allowed after @option{-}.
6582 This causes any subsequent options in the string to enable the corresponding
6583 style check option. That is, it cancels the effect of a previous ^-^REMOVE^,
6589 @emph{Removing style check options}
6590 If the name of a style check is preceded by @option{NO} then the corresponding
6591 style check is turned off. For example @option{NOCOMMENTS} turns off style
6592 checking for comments.
6597 In the above rules, appearing in column one is always permitted, that is,
6598 counts as meeting either a requirement for a required preceding space,
6599 or as meeting a requirement for no preceding space.
6601 Appearing at the end of a line is also always permitted, that is, counts
6602 as meeting either a requirement for a following space, or as meeting
6603 a requirement for no following space.
6606 If any of these style rules is violated, a message is generated giving
6607 details on the violation. The initial characters of such messages are
6608 always ``@code{(style)}''. Note that these messages are treated as warning
6609 messages, so they normally do not prevent the generation of an object
6610 file. The @option{-gnatwe} switch can be used to treat warning messages,
6611 including style messages, as fatal errors.
6615 @option{-gnaty} on its own (that is not
6616 followed by any letters or digits), then the effect is equivalent
6617 to the use of @option{-gnatyy}, as described above, that is all
6618 built-in standard style check options are enabled.
6622 /STYLE_CHECKS=ALL_BUILTIN enables all checking options with
6623 the exception of ORDERED_SUBPROGRAMS, UNNECESSARY_BLANK_LINES,
6624 XTRA_PARENS, and DOS_LINE_ENDINGS. In addition
6634 clears any previously set style checks.
6636 @node Run-Time Checks
6637 @subsection Run-Time Checks
6638 @cindex Division by zero
6639 @cindex Access before elaboration
6640 @cindex Checks, division by zero
6641 @cindex Checks, access before elaboration
6642 @cindex Checks, stack overflow checking
6645 By default, the following checks are suppressed: integer overflow
6646 checks, stack overflow checks, and checks for access before
6647 elaboration on subprogram calls. All other checks, including range
6648 checks and array bounds checks, are turned on by default. The
6649 following @command{gcc} switches refine this default behavior.
6654 @cindex @option{-gnatp} (@command{gcc})
6655 @cindex Suppressing checks
6656 @cindex Checks, suppressing
6658 This switch causes the unit to be compiled
6659 as though @code{pragma Suppress (All_checks)}
6660 had been present in the source. Validity checks are also eliminated (in
6661 other words @option{-gnatp} also implies @option{-gnatVn}.
6662 Use this switch to improve the performance
6663 of the code at the expense of safety in the presence of invalid data or
6666 Note that when checks are suppressed, the compiler is allowed, but not
6667 required, to omit the checking code. If the run-time cost of the
6668 checking code is zero or near-zero, the compiler will generate it even
6669 if checks are suppressed. In particular, if the compiler can prove
6670 that a certain check will necessarily fail, it will generate code to
6671 do an unconditional ``raise'', even if checks are suppressed. The
6672 compiler warns in this case. Another case in which checks may not be
6673 eliminated is when they are embedded in certain run time routines such
6674 as math library routines.
6676 Of course, run-time checks are omitted whenever the compiler can prove
6677 that they will not fail, whether or not checks are suppressed.
6679 Note that if you suppress a check that would have failed, program
6680 execution is erroneous, which means the behavior is totally
6681 unpredictable. The program might crash, or print wrong answers, or
6682 do anything else. It might even do exactly what you wanted it to do
6683 (and then it might start failing mysteriously next week or next
6684 year). The compiler will generate code based on the assumption that
6685 the condition being checked is true, which can result in disaster if
6686 that assumption is wrong.
6688 The @option{-gnatp} switch has no effect if a subsequent
6689 @option{-gnat-p} switch appears.
6692 @cindex @option{-gnat-p} (@command{gcc})
6693 @cindex Suppressing checks
6694 @cindex Checks, suppressing
6696 This switch cancels the effect of a previous @option{gnatp} switch.
6699 @cindex @option{-gnato} (@command{gcc})
6700 @cindex Overflow checks
6701 @cindex Check, overflow
6702 Enables overflow checking for integer operations.
6703 This causes GNAT to generate slower and larger executable
6704 programs by adding code to check for overflow (resulting in raising
6705 @code{Constraint_Error} as required by standard Ada
6706 semantics). These overflow checks correspond to situations in which
6707 the true value of the result of an operation may be outside the base
6708 range of the result type. The following example shows the distinction:
6710 @smallexample @c ada
6711 X1 : Integer := "Integer'Last";
6712 X2 : Integer range 1 .. 5 := "5";
6713 X3 : Integer := "Integer'Last";
6714 X4 : Integer range 1 .. 5 := "5";
6715 F : Float := "2.0E+20";
6724 Note that if explicit values are assigned at compile time, the
6725 compiler may be able to detect overflow at compile time, in which case
6726 no actual run-time checking code is required, and Constraint_Error
6727 will be raised unconditionally, with or without
6728 @option{-gnato}. That's why the assigned values in the above fragment
6729 are in quotes, the meaning is "assign a value not known to the
6730 compiler that happens to be equal to ...". The remaining discussion
6731 assumes that the compiler cannot detect the values at compile time.
6733 Here the first addition results in a value that is outside the base range
6734 of Integer, and hence requires an overflow check for detection of the
6735 constraint error. Thus the first assignment to @code{X1} raises a
6736 @code{Constraint_Error} exception only if @option{-gnato} is set.
6738 The second increment operation results in a violation of the explicit
6739 range constraint; such range checks are performed by default, and are
6740 unaffected by @option{-gnato}.
6742 The two conversions of @code{F} both result in values that are outside
6743 the base range of type @code{Integer} and thus will raise
6744 @code{Constraint_Error} exceptions only if @option{-gnato} is used.
6745 The fact that the result of the second conversion is assigned to
6746 variable @code{X4} with a restricted range is irrelevant, since the problem
6747 is in the conversion, not the assignment.
6749 Basically the rule is that in the default mode (@option{-gnato} not
6750 used), the generated code assures that all integer variables stay
6751 within their declared ranges, or within the base range if there is
6752 no declared range. This prevents any serious problems like indexes
6753 out of range for array operations.
6755 What is not checked in default mode is an overflow that results in
6756 an in-range, but incorrect value. In the above example, the assignments
6757 to @code{X1}, @code{X2}, @code{X3} all give results that are within the
6758 range of the target variable, but the result is wrong in the sense that
6759 it is too large to be represented correctly. Typically the assignment
6760 to @code{X1} will result in wrap around to the largest negative number.
6761 The conversions of @code{F} will result in some @code{Integer} value
6762 and if that integer value is out of the @code{X4} range then the
6763 subsequent assignment would generate an exception.
6765 @findex Machine_Overflows
6766 Note that the @option{-gnato} switch does not affect the code generated
6767 for any floating-point operations; it applies only to integer
6769 For floating-point, GNAT has the @code{Machine_Overflows}
6770 attribute set to @code{False} and the normal mode of operation is to
6771 generate IEEE NaN and infinite values on overflow or invalid operations
6772 (such as dividing 0.0 by 0.0).
6774 The reason that we distinguish overflow checking from other kinds of
6775 range constraint checking is that a failure of an overflow check, unlike
6776 for example the failure of a range check, can result in an incorrect
6777 value, but cannot cause random memory destruction (like an out of range
6778 subscript), or a wild jump (from an out of range case value). Overflow
6779 checking is also quite expensive in time and space, since in general it
6780 requires the use of double length arithmetic.
6782 Note again that @option{-gnato} is off by default, so overflow checking is
6783 not performed in default mode. This means that out of the box, with the
6784 default settings, GNAT does not do all the checks expected from the
6785 language description in the Ada Reference Manual. If you want all constraint
6786 checks to be performed, as described in this Manual, then you must
6787 explicitly use the -gnato switch either on the @command{gnatmake} or
6788 @command{gcc} command.
6791 @cindex @option{-gnatE} (@command{gcc})
6792 @cindex Elaboration checks
6793 @cindex Check, elaboration
6794 Enables dynamic checks for access-before-elaboration
6795 on subprogram calls and generic instantiations.
6796 Note that @option{-gnatE} is not necessary for safety, because in the
6797 default mode, GNAT ensures statically that the checks would not fail.
6798 For full details of the effect and use of this switch,
6799 @xref{Compiling Using gcc}.
6802 @cindex @option{-fstack-check} (@command{gcc})
6803 @cindex Stack Overflow Checking
6804 @cindex Checks, stack overflow checking
6805 Activates stack overflow checking. For full details of the effect and use of
6806 this switch see @ref{Stack Overflow Checking}.
6811 The setting of these switches only controls the default setting of the
6812 checks. You may modify them using either @code{Suppress} (to remove
6813 checks) or @code{Unsuppress} (to add back suppressed checks) pragmas in
6816 @node Using gcc for Syntax Checking
6817 @subsection Using @command{gcc} for Syntax Checking
6820 @cindex @option{-gnats} (@command{gcc})
6824 The @code{s} stands for ``syntax''.
6827 Run GNAT in syntax checking only mode. For
6828 example, the command
6831 $ gcc -c -gnats x.adb
6835 compiles file @file{x.adb} in syntax-check-only mode. You can check a
6836 series of files in a single command
6838 , and can use wild cards to specify such a group of files.
6839 Note that you must specify the @option{-c} (compile
6840 only) flag in addition to the @option{-gnats} flag.
6843 You may use other switches in conjunction with @option{-gnats}. In
6844 particular, @option{-gnatl} and @option{-gnatv} are useful to control the
6845 format of any generated error messages.
6847 When the source file is empty or contains only empty lines and/or comments,
6848 the output is a warning:
6851 $ gcc -c -gnats -x ada toto.txt
6852 toto.txt:1:01: warning: empty file, contains no compilation units
6856 Otherwise, the output is simply the error messages, if any. No object file or
6857 ALI file is generated by a syntax-only compilation. Also, no units other
6858 than the one specified are accessed. For example, if a unit @code{X}
6859 @code{with}'s a unit @code{Y}, compiling unit @code{X} in syntax
6860 check only mode does not access the source file containing unit
6863 @cindex Multiple units, syntax checking
6864 Normally, GNAT allows only a single unit in a source file. However, this
6865 restriction does not apply in syntax-check-only mode, and it is possible
6866 to check a file containing multiple compilation units concatenated
6867 together. This is primarily used by the @code{gnatchop} utility
6868 (@pxref{Renaming Files Using gnatchop}).
6871 @node Using gcc for Semantic Checking
6872 @subsection Using @command{gcc} for Semantic Checking
6875 @cindex @option{-gnatc} (@command{gcc})
6879 The @code{c} stands for ``check''.
6881 Causes the compiler to operate in semantic check mode,
6882 with full checking for all illegalities specified in the
6883 Ada Reference Manual, but without generation of any object code
6884 (no object file is generated).
6886 Because dependent files must be accessed, you must follow the GNAT
6887 semantic restrictions on file structuring to operate in this mode:
6891 The needed source files must be accessible
6892 (@pxref{Search Paths and the Run-Time Library (RTL)}).
6895 Each file must contain only one compilation unit.
6898 The file name and unit name must match (@pxref{File Naming Rules}).
6901 The output consists of error messages as appropriate. No object file is
6902 generated. An @file{ALI} file is generated for use in the context of
6903 cross-reference tools, but this file is marked as not being suitable
6904 for binding (since no object file is generated).
6905 The checking corresponds exactly to the notion of
6906 legality in the Ada Reference Manual.
6908 Any unit can be compiled in semantics-checking-only mode, including
6909 units that would not normally be compiled (subunits,
6910 and specifications where a separate body is present).
6913 @node Compiling Different Versions of Ada
6914 @subsection Compiling Different Versions of Ada
6917 The switches described in this section allow you to explicitly specify
6918 the version of the Ada language that your programs are written in.
6919 By default @value{EDITION} assumes @value{DEFAULTLANGUAGEVERSION},
6920 but you can also specify @value{NONDEFAULTLANGUAGEVERSION} or
6921 indicate Ada 83 compatibility mode.
6924 @cindex Compatibility with Ada 83
6926 @item -gnat83 (Ada 83 Compatibility Mode)
6927 @cindex @option{-gnat83} (@command{gcc})
6928 @cindex ACVC, Ada 83 tests
6932 Although GNAT is primarily an Ada 95 / Ada 2005 compiler, this switch
6933 specifies that the program is to be compiled in Ada 83 mode. With
6934 @option{-gnat83}, GNAT rejects most post-Ada 83 extensions and applies Ada 83
6935 semantics where this can be done easily.
6936 It is not possible to guarantee this switch does a perfect
6937 job; some subtle tests, such as are
6938 found in earlier ACVC tests (and that have been removed from the ACATS suite
6939 for Ada 95), might not compile correctly.
6940 Nevertheless, this switch may be useful in some circumstances, for example
6941 where, due to contractual reasons, existing code needs to be maintained
6942 using only Ada 83 features.
6944 With few exceptions (most notably the need to use @code{<>} on
6945 @cindex Generic formal parameters
6946 unconstrained generic formal parameters, the use of the new Ada 95 / Ada 2005
6947 reserved words, and the use of packages
6948 with optional bodies), it is not necessary to specify the
6949 @option{-gnat83} switch when compiling Ada 83 programs, because, with rare
6950 exceptions, Ada 95 and Ada 2005 are upwardly compatible with Ada 83. Thus
6951 a correct Ada 83 program is usually also a correct program
6952 in these later versions of the language standard.
6953 For further information, please refer to @ref{Compatibility and Porting Guide}.
6955 @item -gnat95 (Ada 95 mode)
6956 @cindex @option{-gnat95} (@command{gcc})
6960 This switch directs the compiler to implement the Ada 95 version of the
6962 Since Ada 95 is almost completely upwards
6963 compatible with Ada 83, Ada 83 programs may generally be compiled using
6964 this switch (see the description of the @option{-gnat83} switch for further
6965 information about Ada 83 mode).
6966 If an Ada 2005 program is compiled in Ada 95 mode,
6967 uses of the new Ada 2005 features will cause error
6968 messages or warnings.
6970 This switch also can be used to cancel the effect of a previous
6971 @option{-gnat83}, @option{-gnat05/2005}, or @option{-gnat12/2012}
6972 switch earlier in the command line.
6974 @item -gnat05 or -gnat2005 (Ada 2005 mode)
6975 @cindex @option{-gnat05} (@command{gcc})
6976 @cindex @option{-gnat2005} (@command{gcc})
6977 @cindex Ada 2005 mode
6980 This switch directs the compiler to implement the Ada 2005 version of the
6981 language, as documented in the official Ada standards document.
6982 Since Ada 2005 is almost completely upwards
6983 compatible with Ada 95 (and thus also with Ada 83), Ada 83 and Ada 95 programs
6984 may generally be compiled using this switch (see the description of the
6985 @option{-gnat83} and @option{-gnat95} switches for further
6989 Note that even though Ada 2005 is the current official version of the
6990 language, GNAT still compiles in Ada 95 mode by default, so if you are
6991 using Ada 2005 features in your program, you must use this switch (or
6992 the equivalent Ada_05 or Ada_2005 configuration pragmas).
6995 @item -gnat12 or -gnat2012 (Ada 2012 mode)
6996 @cindex @option{-gnat12} (@command{gcc})
6997 @cindex @option{-gnat2012} (@command{gcc})
6998 @cindex Ada 2012 mode
7001 This switch directs the compiler to implement the Ada 2012 version of the
7003 Since Ada 2012 is almost completely upwards
7004 compatible with Ada 2005 (and thus also with Ada 83, and Ada 95),
7005 Ada 83 and Ada 95 programs
7006 may generally be compiled using this switch (see the description of the
7007 @option{-gnat83}, @option{-gnat95}, and @option{-gnat05/2005} switches
7008 for further information).
7010 For information about the approved ``Ada Issues'' that have been incorporated
7011 into Ada 2012, see @url{http://www.ada-auth.org/cgi-bin/cvsweb.cgi/AIs}.
7012 Included with GNAT releases is a file @file{features-ada12} that describes
7013 the set of implemented Ada 2012 features.
7015 @item -gnatX (Enable GNAT Extensions)
7016 @cindex @option{-gnatX} (@command{gcc})
7017 @cindex Ada language extensions
7018 @cindex GNAT extensions
7021 This switch directs the compiler to implement the latest version of the
7022 language (currently Ada 2012) and also to enable certain GNAT implementation
7023 extensions that are not part of any Ada standard. For a full list of these
7024 extensions, see the GNAT reference manual.
7028 @node Character Set Control
7029 @subsection Character Set Control
7031 @item ^-gnati^/IDENTIFIER_CHARACTER_SET=^@var{c}
7032 @cindex @option{^-gnati^/IDENTIFIER_CHARACTER_SET^} (@command{gcc})
7035 Normally GNAT recognizes the Latin-1 character set in source program
7036 identifiers, as described in the Ada Reference Manual.
7038 GNAT to recognize alternate character sets in identifiers. @var{c} is a
7039 single character ^^or word^ indicating the character set, as follows:
7043 ISO 8859-1 (Latin-1) identifiers
7046 ISO 8859-2 (Latin-2) letters allowed in identifiers
7049 ISO 8859-3 (Latin-3) letters allowed in identifiers
7052 ISO 8859-4 (Latin-4) letters allowed in identifiers
7055 ISO 8859-5 (Cyrillic) letters allowed in identifiers
7058 ISO 8859-15 (Latin-9) letters allowed in identifiers
7061 IBM PC letters (code page 437) allowed in identifiers
7064 IBM PC letters (code page 850) allowed in identifiers
7066 @item ^f^FULL_UPPER^
7067 Full upper-half codes allowed in identifiers
7070 No upper-half codes allowed in identifiers
7073 Wide-character codes (that is, codes greater than 255)
7074 allowed in identifiers
7077 @xref{Foreign Language Representation}, for full details on the
7078 implementation of these character sets.
7080 @item ^-gnatW^/WIDE_CHARACTER_ENCODING=^@var{e}
7081 @cindex @option{^-gnatW^/WIDE_CHARACTER_ENCODING^} (@command{gcc})
7082 Specify the method of encoding for wide characters.
7083 @var{e} is one of the following:
7088 Hex encoding (brackets coding also recognized)
7091 Upper half encoding (brackets encoding also recognized)
7094 Shift/JIS encoding (brackets encoding also recognized)
7097 EUC encoding (brackets encoding also recognized)
7100 UTF-8 encoding (brackets encoding also recognized)
7103 Brackets encoding only (default value)
7105 For full details on these encoding
7106 methods see @ref{Wide Character Encodings}.
7107 Note that brackets coding is always accepted, even if one of the other
7108 options is specified, so for example @option{-gnatW8} specifies that both
7109 brackets and UTF-8 encodings will be recognized. The units that are
7110 with'ed directly or indirectly will be scanned using the specified
7111 representation scheme, and so if one of the non-brackets scheme is
7112 used, it must be used consistently throughout the program. However,
7113 since brackets encoding is always recognized, it may be conveniently
7114 used in standard libraries, allowing these libraries to be used with
7115 any of the available coding schemes.
7118 If no @option{-gnatW?} parameter is present, then the default
7119 representation is normally Brackets encoding only. However, if the
7120 first three characters of the file are 16#EF# 16#BB# 16#BF# (the standard
7121 byte order mark or BOM for UTF-8), then these three characters are
7122 skipped and the default representation for the file is set to UTF-8.
7124 Note that the wide character representation that is specified (explicitly
7125 or by default) for the main program also acts as the default encoding used
7126 for Wide_Text_IO files if not specifically overridden by a WCEM form
7130 @node File Naming Control
7131 @subsection File Naming Control
7134 @item ^-gnatk^/FILE_NAME_MAX_LENGTH=^@var{n}
7135 @cindex @option{-gnatk} (@command{gcc})
7136 Activates file name ``krunching''. @var{n}, a decimal integer in the range
7137 1-999, indicates the maximum allowable length of a file name (not
7138 including the @file{.ads} or @file{.adb} extension). The default is not
7139 to enable file name krunching.
7141 For the source file naming rules, @xref{File Naming Rules}.
7144 @node Subprogram Inlining Control
7145 @subsection Subprogram Inlining Control
7150 @cindex @option{-gnatn} (@command{gcc})
7152 The @code{n} here is intended to suggest the first syllable of the
7155 GNAT recognizes and processes @code{Inline} pragmas. However, for the
7156 inlining to actually occur, optimization must be enabled. To enable
7157 inlining of subprograms specified by pragma @code{Inline},
7158 you must also specify this switch.
7159 In the absence of this switch, GNAT does not attempt
7160 inlining and does not need to access the bodies of
7161 subprograms for which @code{pragma Inline} is specified if they are not
7162 in the current unit.
7164 If you specify this switch the compiler will access these bodies,
7165 creating an extra source dependency for the resulting object file, and
7166 where possible, the call will be inlined.
7167 For further details on when inlining is possible
7168 see @ref{Inlining of Subprograms}.
7171 @cindex @option{-gnatN} (@command{gcc})
7172 This switch activates front-end inlining which also
7173 generates additional dependencies.
7175 When using a gcc-based back end (in practice this means using any version
7176 of GNAT other than the JGNAT, .NET or GNAAMP versions), then the use of
7177 @option{-gnatN} is deprecated, and the use of @option{-gnatn} is preferred.
7178 Historically front end inlining was more extensive than the gcc back end
7179 inlining, but that is no longer the case.
7182 @node Auxiliary Output Control
7183 @subsection Auxiliary Output Control
7187 @cindex @option{-gnatt} (@command{gcc})
7188 @cindex Writing internal trees
7189 @cindex Internal trees, writing to file
7190 Causes GNAT to write the internal tree for a unit to a file (with the
7191 extension @file{.adt}.
7192 This not normally required, but is used by separate analysis tools.
7194 these tools do the necessary compilations automatically, so you should
7195 not have to specify this switch in normal operation.
7196 Note that the combination of switches @option{-gnatct}
7197 generates a tree in the form required by ASIS applications.
7200 @cindex @option{-gnatu} (@command{gcc})
7201 Print a list of units required by this compilation on @file{stdout}.
7202 The listing includes all units on which the unit being compiled depends
7203 either directly or indirectly.
7206 @item -pass-exit-codes
7207 @cindex @option{-pass-exit-codes} (@command{gcc})
7208 If this switch is not used, the exit code returned by @command{gcc} when
7209 compiling multiple files indicates whether all source files have
7210 been successfully used to generate object files or not.
7212 When @option{-pass-exit-codes} is used, @command{gcc} exits with an extended
7213 exit status and allows an integrated development environment to better
7214 react to a compilation failure. Those exit status are:
7218 There was an error in at least one source file.
7220 At least one source file did not generate an object file.
7222 The compiler died unexpectedly (internal error for example).
7224 An object file has been generated for every source file.
7229 @node Debugging Control
7230 @subsection Debugging Control
7234 @cindex Debugging options
7237 @cindex @option{-gnatd} (@command{gcc})
7238 Activate internal debugging switches. @var{x} is a letter or digit, or
7239 string of letters or digits, which specifies the type of debugging
7240 outputs desired. Normally these are used only for internal development
7241 or system debugging purposes. You can find full documentation for these
7242 switches in the body of the @code{Debug} unit in the compiler source
7243 file @file{debug.adb}.
7247 @cindex @option{-gnatG} (@command{gcc})
7248 This switch causes the compiler to generate auxiliary output containing
7249 a pseudo-source listing of the generated expanded code. Like most Ada
7250 compilers, GNAT works by first transforming the high level Ada code into
7251 lower level constructs. For example, tasking operations are transformed
7252 into calls to the tasking run-time routines. A unique capability of GNAT
7253 is to list this expanded code in a form very close to normal Ada source.
7254 This is very useful in understanding the implications of various Ada
7255 usage on the efficiency of the generated code. There are many cases in
7256 Ada (e.g.@: the use of controlled types), where simple Ada statements can
7257 generate a lot of run-time code. By using @option{-gnatG} you can identify
7258 these cases, and consider whether it may be desirable to modify the coding
7259 approach to improve efficiency.
7261 The optional parameter @code{nn} if present after -gnatG specifies an
7262 alternative maximum line length that overrides the normal default of 72.
7263 This value is in the range 40-999999, values less than 40 being silently
7264 reset to 40. The equal sign is optional.
7266 The format of the output is very similar to standard Ada source, and is
7267 easily understood by an Ada programmer. The following special syntactic
7268 additions correspond to low level features used in the generated code that
7269 do not have any exact analogies in pure Ada source form. The following
7270 is a partial list of these special constructions. See the spec
7271 of package @code{Sprint} in file @file{sprint.ads} for a full list.
7273 If the switch @option{-gnatL} is used in conjunction with
7274 @cindex @option{-gnatL} (@command{gcc})
7275 @option{-gnatG}, then the original source lines are interspersed
7276 in the expanded source (as comment lines with the original line number).
7279 @item new @var{xxx} @r{[}storage_pool = @var{yyy}@r{]}
7280 Shows the storage pool being used for an allocator.
7282 @item at end @var{procedure-name};
7283 Shows the finalization (cleanup) procedure for a scope.
7285 @item (if @var{expr} then @var{expr} else @var{expr})
7286 Conditional expression equivalent to the @code{x?y:z} construction in C.
7288 @item @var{target}^^^(@var{source})
7289 A conversion with floating-point truncation instead of rounding.
7291 @item @var{target}?(@var{source})
7292 A conversion that bypasses normal Ada semantic checking. In particular
7293 enumeration types and fixed-point types are treated simply as integers.
7295 @item @var{target}?^^^(@var{source})
7296 Combines the above two cases.
7298 @item @var{x} #/ @var{y}
7299 @itemx @var{x} #mod @var{y}
7300 @itemx @var{x} #* @var{y}
7301 @itemx @var{x} #rem @var{y}
7302 A division or multiplication of fixed-point values which are treated as
7303 integers without any kind of scaling.
7305 @item free @var{expr} @r{[}storage_pool = @var{xxx}@r{]}
7306 Shows the storage pool associated with a @code{free} statement.
7308 @item [subtype or type declaration]
7309 Used to list an equivalent declaration for an internally generated
7310 type that is referenced elsewhere in the listing.
7312 @c @item freeze @var{type-name} @ovar{actions}
7313 @c Expanding @ovar macro inline (explanation in macro def comments)
7314 @item freeze @var{type-name} @r{[}@var{actions}@r{]}
7315 Shows the point at which @var{type-name} is frozen, with possible
7316 associated actions to be performed at the freeze point.
7318 @item reference @var{itype}
7319 Reference (and hence definition) to internal type @var{itype}.
7321 @item @var{function-name}! (@var{arg}, @var{arg}, @var{arg})
7322 Intrinsic function call.
7324 @item @var{label-name} : label
7325 Declaration of label @var{labelname}.
7327 @item #$ @var{subprogram-name}
7328 An implicit call to a run-time support routine
7329 (to meet the requirement of H.3.1(9) in a
7332 @item @var{expr} && @var{expr} && @var{expr} @dots{} && @var{expr}
7333 A multiple concatenation (same effect as @var{expr} & @var{expr} &
7334 @var{expr}, but handled more efficiently).
7336 @item [constraint_error]
7337 Raise the @code{Constraint_Error} exception.
7339 @item @var{expression}'reference
7340 A pointer to the result of evaluating @var{expression}.
7342 @item @var{target-type}!(@var{source-expression})
7343 An unchecked conversion of @var{source-expression} to @var{target-type}.
7345 @item [@var{numerator}/@var{denominator}]
7346 Used to represent internal real literals (that) have no exact
7347 representation in base 2-16 (for example, the result of compile time
7348 evaluation of the expression 1.0/27.0).
7352 @cindex @option{-gnatD} (@command{gcc})
7353 When used in conjunction with @option{-gnatG}, this switch causes
7354 the expanded source, as described above for
7355 @option{-gnatG} to be written to files with names
7356 @file{^xxx.dg^XXX_DG^}, where @file{xxx} is the normal file name,
7357 instead of to the standard output file. For
7358 example, if the source file name is @file{hello.adb}, then a file
7359 @file{^hello.adb.dg^HELLO.ADB_DG^} will be written. The debugging
7360 information generated by the @command{gcc} @option{^-g^/DEBUG^} switch
7361 will refer to the generated @file{^xxx.dg^XXX_DG^} file. This allows
7362 you to do source level debugging using the generated code which is
7363 sometimes useful for complex code, for example to find out exactly
7364 which part of a complex construction raised an exception. This switch
7365 also suppress generation of cross-reference information (see
7366 @option{-gnatx}) since otherwise the cross-reference information
7367 would refer to the @file{^.dg^.DG^} file, which would cause
7368 confusion since this is not the original source file.
7370 Note that @option{-gnatD} actually implies @option{-gnatG}
7371 automatically, so it is not necessary to give both options.
7372 In other words @option{-gnatD} is equivalent to @option{-gnatDG}).
7374 If the switch @option{-gnatL} is used in conjunction with
7375 @cindex @option{-gnatL} (@command{gcc})
7376 @option{-gnatDG}, then the original source lines are interspersed
7377 in the expanded source (as comment lines with the original line number).
7379 The optional parameter @code{nn} if present after -gnatD specifies an
7380 alternative maximum line length that overrides the normal default of 72.
7381 This value is in the range 40-999999, values less than 40 being silently
7382 reset to 40. The equal sign is optional.
7385 @cindex @option{-gnatr} (@command{gcc})
7386 @cindex pragma Restrictions
7387 This switch causes pragma Restrictions to be treated as Restriction_Warnings
7388 so that violation of restrictions causes warnings rather than illegalities.
7389 This is useful during the development process when new restrictions are added
7390 or investigated. The switch also causes pragma Profile to be treated as
7391 Profile_Warnings, and pragma Restricted_Run_Time and pragma Ravenscar set
7392 restriction warnings rather than restrictions.
7395 @item -gnatR@r{[}0@r{|}1@r{|}2@r{|}3@r{[}s@r{]]}
7396 @cindex @option{-gnatR} (@command{gcc})
7397 This switch controls output from the compiler of a listing showing
7398 representation information for declared types and objects. For
7399 @option{-gnatR0}, no information is output (equivalent to omitting
7400 the @option{-gnatR} switch). For @option{-gnatR1} (which is the default,
7401 so @option{-gnatR} with no parameter has the same effect), size and alignment
7402 information is listed for declared array and record types. For
7403 @option{-gnatR2}, size and alignment information is listed for all
7404 declared types and objects. Finally @option{-gnatR3} includes symbolic
7405 expressions for values that are computed at run time for
7406 variant records. These symbolic expressions have a mostly obvious
7407 format with #n being used to represent the value of the n'th
7408 discriminant. See source files @file{repinfo.ads/adb} in the
7409 @code{GNAT} sources for full details on the format of @option{-gnatR3}
7410 output. If the switch is followed by an s (e.g.@: @option{-gnatR2s}), then
7411 the output is to a file with the name @file{^file.rep^file_REP^} where
7412 file is the name of the corresponding source file.
7415 @item /REPRESENTATION_INFO
7416 @cindex @option{/REPRESENTATION_INFO} (@command{gcc})
7417 This qualifier controls output from the compiler of a listing showing
7418 representation information for declared types and objects. For
7419 @option{/REPRESENTATION_INFO=NONE}, no information is output
7420 (equivalent to omitting the @option{/REPRESENTATION_INFO} qualifier).
7421 @option{/REPRESENTATION_INFO} without option is equivalent to
7422 @option{/REPRESENTATION_INFO=ARRAYS}.
7423 For @option{/REPRESENTATION_INFO=ARRAYS}, size and alignment
7424 information is listed for declared array and record types. For
7425 @option{/REPRESENTATION_INFO=OBJECTS}, size and alignment information
7426 is listed for all expression information for values that are computed
7427 at run time for variant records. These symbolic expressions have a mostly
7428 obvious format with #n being used to represent the value of the n'th
7429 discriminant. See source files @file{REPINFO.ADS/ADB} in the
7430 @code{GNAT} sources for full details on the format of
7431 @option{/REPRESENTATION_INFO=SYMBOLIC} output.
7432 If _FILE is added at the end of an option
7433 (e.g.@: @option{/REPRESENTATION_INFO=ARRAYS_FILE}),
7434 then the output is to a file with the name @file{file_REP} where
7435 file is the name of the corresponding source file.
7437 Note that it is possible for record components to have zero size. In
7438 this case, the component clause uses an obvious extension of permitted
7439 Ada syntax, for example @code{at 0 range 0 .. -1}.
7441 Representation information requires that code be generated (since it is the
7442 code generator that lays out complex data structures). If an attempt is made
7443 to output representation information when no code is generated, for example
7444 when a subunit is compiled on its own, then no information can be generated
7445 and the compiler outputs a message to this effect.
7448 @cindex @option{-gnatS} (@command{gcc})
7449 The use of the switch @option{-gnatS} for an
7450 Ada compilation will cause the compiler to output a
7451 representation of package Standard in a form very
7452 close to standard Ada. It is not quite possible to
7453 do this entirely in standard Ada (since new
7454 numeric base types cannot be created in standard
7455 Ada), but the output is easily
7456 readable to any Ada programmer, and is useful to
7457 determine the characteristics of target dependent
7458 types in package Standard.
7461 @cindex @option{-gnatx} (@command{gcc})
7462 Normally the compiler generates full cross-referencing information in
7463 the @file{ALI} file. This information is used by a number of tools,
7464 including @code{gnatfind} and @code{gnatxref}. The @option{-gnatx} switch
7465 suppresses this information. This saves some space and may slightly
7466 speed up compilation, but means that these tools cannot be used.
7469 @node Exception Handling Control
7470 @subsection Exception Handling Control
7473 GNAT uses two methods for handling exceptions at run-time. The
7474 @code{setjmp/longjmp} method saves the context when entering
7475 a frame with an exception handler. Then when an exception is
7476 raised, the context can be restored immediately, without the
7477 need for tracing stack frames. This method provides very fast
7478 exception propagation, but introduces significant overhead for
7479 the use of exception handlers, even if no exception is raised.
7481 The other approach is called ``zero cost'' exception handling.
7482 With this method, the compiler builds static tables to describe
7483 the exception ranges. No dynamic code is required when entering
7484 a frame containing an exception handler. When an exception is
7485 raised, the tables are used to control a back trace of the
7486 subprogram invocation stack to locate the required exception
7487 handler. This method has considerably poorer performance for
7488 the propagation of exceptions, but there is no overhead for
7489 exception handlers if no exception is raised. Note that in this
7490 mode and in the context of mixed Ada and C/C++ programming,
7491 to propagate an exception through a C/C++ code, the C/C++ code
7492 must be compiled with the @option{-funwind-tables} GCC's
7495 The following switches may be used to control which of the
7496 two exception handling methods is used.
7502 @cindex @option{--RTS=sjlj} (@command{gnatmake})
7503 This switch causes the setjmp/longjmp run-time (when available) to be used
7504 for exception handling. If the default
7505 mechanism for the target is zero cost exceptions, then
7506 this switch can be used to modify this default, and must be
7507 used for all units in the partition.
7508 This option is rarely used. One case in which it may be
7509 advantageous is if you have an application where exception
7510 raising is common and the overall performance of the
7511 application is improved by favoring exception propagation.
7514 @cindex @option{--RTS=zcx} (@command{gnatmake})
7515 @cindex Zero Cost Exceptions
7516 This switch causes the zero cost approach to be used
7517 for exception handling. If this is the default mechanism for the
7518 target (see below), then this switch is unneeded. If the default
7519 mechanism for the target is setjmp/longjmp exceptions, then
7520 this switch can be used to modify this default, and must be
7521 used for all units in the partition.
7522 This option can only be used if the zero cost approach
7523 is available for the target in use, otherwise it will generate an error.
7527 The same option @option{--RTS} must be used both for @command{gcc}
7528 and @command{gnatbind}. Passing this option to @command{gnatmake}
7529 (@pxref{Switches for gnatmake}) will ensure the required consistency
7530 through the compilation and binding steps.
7532 @node Units to Sources Mapping Files
7533 @subsection Units to Sources Mapping Files
7537 @item -gnatem=@var{path}
7538 @cindex @option{-gnatem} (@command{gcc})
7539 A mapping file is a way to communicate to the compiler two mappings:
7540 from unit names to file names (without any directory information) and from
7541 file names to path names (with full directory information). These mappings
7542 are used by the compiler to short-circuit the path search.
7544 The use of mapping files is not required for correct operation of the
7545 compiler, but mapping files can improve efficiency, particularly when
7546 sources are read over a slow network connection. In normal operation,
7547 you need not be concerned with the format or use of mapping files,
7548 and the @option{-gnatem} switch is not a switch that you would use
7549 explicitly. It is intended primarily for use by automatic tools such as
7550 @command{gnatmake} running under the project file facility. The
7551 description here of the format of mapping files is provided
7552 for completeness and for possible use by other tools.
7554 A mapping file is a sequence of sets of three lines. In each set, the
7555 first line is the unit name, in lower case, with @code{%s} appended
7556 for specs and @code{%b} appended for bodies; the second line is the
7557 file name; and the third line is the path name.
7563 /gnat/project1/sources/main.2.ada
7566 When the switch @option{-gnatem} is specified, the compiler will
7567 create in memory the two mappings from the specified file. If there is
7568 any problem (nonexistent file, truncated file or duplicate entries),
7569 no mapping will be created.
7571 Several @option{-gnatem} switches may be specified; however, only the
7572 last one on the command line will be taken into account.
7574 When using a project file, @command{gnatmake} creates a temporary
7575 mapping file and communicates it to the compiler using this switch.
7579 @node Integrated Preprocessing
7580 @subsection Integrated Preprocessing
7583 GNAT sources may be preprocessed immediately before compilation.
7584 In this case, the actual
7585 text of the source is not the text of the source file, but is derived from it
7586 through a process called preprocessing. Integrated preprocessing is specified
7587 through switches @option{-gnatep} and/or @option{-gnateD}. @option{-gnatep}
7588 indicates, through a text file, the preprocessing data to be used.
7589 @option{-gnateD} specifies or modifies the values of preprocessing symbol.
7592 Note that when integrated preprocessing is used, the output from the
7593 preprocessor is not written to any external file. Instead it is passed
7594 internally to the compiler. If you need to preserve the result of
7595 preprocessing in a file, then you should use @command{gnatprep}
7596 to perform the desired preprocessing in stand-alone mode.
7599 It is recommended that @command{gnatmake} switch ^-s^/SWITCH_CHECK^ should be
7600 used when Integrated Preprocessing is used. The reason is that preprocessing
7601 with another Preprocessing Data file without changing the sources will
7602 not trigger recompilation without this switch.
7605 Note that @command{gnatmake} switch ^-m^/MINIMAL_RECOMPILATION^ will almost
7606 always trigger recompilation for sources that are preprocessed,
7607 because @command{gnatmake} cannot compute the checksum of the source after
7611 The actual preprocessing function is described in details in section
7612 @ref{Preprocessing Using gnatprep}. This section only describes how integrated
7613 preprocessing is triggered and parameterized.
7617 @item -gnatep=@var{file}
7618 @cindex @option{-gnatep} (@command{gcc})
7619 This switch indicates to the compiler the file name (without directory
7620 information) of the preprocessor data file to use. The preprocessor data file
7621 should be found in the source directories.
7624 A preprocessing data file is a text file with significant lines indicating
7625 how should be preprocessed either a specific source or all sources not
7626 mentioned in other lines. A significant line is a nonempty, non-comment line.
7627 Comments are similar to Ada comments.
7630 Each significant line starts with either a literal string or the character '*'.
7631 A literal string is the file name (without directory information) of the source
7632 to preprocess. A character '*' indicates the preprocessing for all the sources
7633 that are not specified explicitly on other lines (order of the lines is not
7634 significant). It is an error to have two lines with the same file name or two
7635 lines starting with the character '*'.
7638 After the file name or the character '*', another optional literal string
7639 indicating the file name of the definition file to be used for preprocessing
7640 (@pxref{Form of Definitions File}). The definition files are found by the
7641 compiler in one of the source directories. In some cases, when compiling
7642 a source in a directory other than the current directory, if the definition
7643 file is in the current directory, it may be necessary to add the current
7644 directory as a source directory through switch ^-I.^/SEARCH=[]^, otherwise
7645 the compiler would not find the definition file.
7648 Then, optionally, ^switches^switches^ similar to those of @code{gnatprep} may
7649 be found. Those ^switches^switches^ are:
7654 Causes both preprocessor lines and the lines deleted by
7655 preprocessing to be replaced by blank lines, preserving the line number.
7656 This ^switch^switch^ is always implied; however, if specified after @option{-c}
7657 it cancels the effect of @option{-c}.
7660 Causes both preprocessor lines and the lines deleted
7661 by preprocessing to be retained as comments marked
7662 with the special string ``@code{--! }''.
7664 @item -Dsymbol=value
7665 Define or redefine a symbol, associated with value. A symbol is an Ada
7666 identifier, or an Ada reserved word, with the exception of @code{if},
7667 @code{else}, @code{elsif}, @code{end}, @code{and}, @code{or} and @code{then}.
7668 @code{value} is either a literal string, an Ada identifier or any Ada reserved
7669 word. A symbol declared with this ^switch^switch^ replaces a symbol with the
7670 same name defined in a definition file.
7673 Causes a sorted list of symbol names and values to be
7674 listed on the standard output file.
7677 Causes undefined symbols to be treated as having the value @code{FALSE}
7679 of a preprocessor test. In the absence of this option, an undefined symbol in
7680 a @code{#if} or @code{#elsif} test will be treated as an error.
7685 Examples of valid lines in a preprocessor data file:
7688 "toto.adb" "prep.def" -u
7689 -- preprocess "toto.adb", using definition file "prep.def",
7690 -- undefined symbol are False.
7693 -- preprocess all other sources without a definition file;
7694 -- suppressed lined are commented; symbol VERSION has the value V101.
7696 "titi.adb" "prep2.def" -s
7697 -- preprocess "titi.adb", using definition file "prep2.def";
7698 -- list all symbols with their values.
7701 @item ^-gnateD^/DATA_PREPROCESSING=^symbol@r{[}=value@r{]}
7702 @cindex @option{-gnateD} (@command{gcc})
7703 Define or redefine a preprocessing symbol, associated with value. If no value
7704 is given on the command line, then the value of the symbol is @code{True}.
7705 A symbol is an identifier, following normal Ada (case-insensitive)
7706 rules for its syntax, and value is any sequence (including an empty sequence)
7707 of characters from the set (letters, digits, period, underline).
7708 Ada reserved words may be used as symbols, with the exceptions of @code{if},
7709 @code{else}, @code{elsif}, @code{end}, @code{and}, @code{or} and @code{then}.
7712 A symbol declared with this ^switch^switch^ on the command line replaces a
7713 symbol with the same name either in a definition file or specified with a
7714 ^switch^switch^ -D in the preprocessor data file.
7717 This switch is similar to switch @option{^-D^/ASSOCIATE^} of @code{gnatprep}.
7720 When integrated preprocessing is performed and the preprocessor modifies
7721 the source text, write the result of this preprocessing into a file
7722 <source>^.prep^_prep^.
7726 @node Code Generation Control
7727 @subsection Code Generation Control
7731 The GCC technology provides a wide range of target dependent
7732 @option{-m} switches for controlling
7733 details of code generation with respect to different versions of
7734 architectures. This includes variations in instruction sets (e.g.@:
7735 different members of the power pc family), and different requirements
7736 for optimal arrangement of instructions (e.g.@: different members of
7737 the x86 family). The list of available @option{-m} switches may be
7738 found in the GCC documentation.
7740 Use of these @option{-m} switches may in some cases result in improved
7743 The GNAT Pro technology is tested and qualified without any
7744 @option{-m} switches,
7745 so generally the most reliable approach is to avoid the use of these
7746 switches. However, we generally expect most of these switches to work
7747 successfully with GNAT Pro, and many customers have reported successful
7748 use of these options.
7750 Our general advice is to avoid the use of @option{-m} switches unless
7751 special needs lead to requirements in this area. In particular,
7752 there is no point in using @option{-m} switches to improve performance
7753 unless you actually see a performance improvement.
7757 @subsection Return Codes
7758 @cindex Return Codes
7759 @cindex @option{/RETURN_CODES=VMS}
7762 On VMS, GNAT compiled programs return POSIX-style codes by default,
7763 e.g.@: @option{/RETURN_CODES=POSIX}.
7765 To enable VMS style return codes, use GNAT BIND and LINK with the option
7766 @option{/RETURN_CODES=VMS}. For example:
7769 GNAT BIND MYMAIN.ALI /RETURN_CODES=VMS
7770 GNAT LINK MYMAIN.ALI /RETURN_CODES=VMS
7774 Programs built with /RETURN_CODES=VMS are suitable to be called in
7775 VMS DCL scripts. Programs compiled with the default /RETURN_CODES=POSIX
7776 are suitable for spawning with appropriate GNAT RTL routines.
7780 @node Search Paths and the Run-Time Library (RTL)
7781 @section Search Paths and the Run-Time Library (RTL)
7784 With the GNAT source-based library system, the compiler must be able to
7785 find source files for units that are needed by the unit being compiled.
7786 Search paths are used to guide this process.
7788 The compiler compiles one source file whose name must be given
7789 explicitly on the command line. In other words, no searching is done
7790 for this file. To find all other source files that are needed (the most
7791 common being the specs of units), the compiler examines the following
7792 directories, in the following order:
7796 The directory containing the source file of the main unit being compiled
7797 (the file name on the command line).
7800 Each directory named by an @option{^-I^/SOURCE_SEARCH^} switch given on the
7801 @command{gcc} command line, in the order given.
7804 @findex ADA_PRJ_INCLUDE_FILE
7805 Each of the directories listed in the text file whose name is given
7806 by the @env{ADA_PRJ_INCLUDE_FILE} ^environment variable^logical name^.
7809 @env{ADA_PRJ_INCLUDE_FILE} is normally set by gnatmake or by the ^gnat^GNAT^
7810 driver when project files are used. It should not normally be set
7814 @findex ADA_INCLUDE_PATH
7815 Each of the directories listed in the value of the
7816 @env{ADA_INCLUDE_PATH} ^environment variable^logical name^.
7818 Construct this value
7819 exactly as the @env{PATH} environment variable: a list of directory
7820 names separated by colons (semicolons when working with the NT version).
7823 Normally, define this value as a logical name containing a comma separated
7824 list of directory names.
7826 This variable can also be defined by means of an environment string
7827 (an argument to the HP C exec* set of functions).
7831 DEFINE ANOTHER_PATH FOO:[BAG]
7832 DEFINE ADA_INCLUDE_PATH ANOTHER_PATH,FOO:[BAM],FOO:[BAR]
7835 By default, the path includes GNU:[LIB.OPENVMS7_x.2_8_x.DECLIB]
7836 first, followed by the standard Ada
7837 libraries in GNU:[LIB.OPENVMS7_x.2_8_x.ADAINCLUDE].
7838 If this is not redefined, the user will obtain the HP Ada 83 IO packages
7839 (Text_IO, Sequential_IO, etc)
7840 instead of the standard Ada packages. Thus, in order to get the standard Ada
7841 packages by default, ADA_INCLUDE_PATH must be redefined.
7845 The content of the @file{ada_source_path} file which is part of the GNAT
7846 installation tree and is used to store standard libraries such as the
7847 GNAT Run Time Library (RTL) source files.
7849 @ref{Installing a library}
7854 Specifying the switch @option{^-I-^/NOCURRENT_DIRECTORY^}
7855 inhibits the use of the directory
7856 containing the source file named in the command line. You can still
7857 have this directory on your search path, but in this case it must be
7858 explicitly requested with a @option{^-I^/SOURCE_SEARCH^} switch.
7860 Specifying the switch @option{-nostdinc}
7861 inhibits the search of the default location for the GNAT Run Time
7862 Library (RTL) source files.
7864 The compiler outputs its object files and ALI files in the current
7867 Caution: The object file can be redirected with the @option{-o} switch;
7868 however, @command{gcc} and @code{gnat1} have not been coordinated on this
7869 so the @file{ALI} file will not go to the right place. Therefore, you should
7870 avoid using the @option{-o} switch.
7874 The packages @code{Ada}, @code{System}, and @code{Interfaces} and their
7875 children make up the GNAT RTL, together with the simple @code{System.IO}
7876 package used in the @code{"Hello World"} example. The sources for these units
7877 are needed by the compiler and are kept together in one directory. Not
7878 all of the bodies are needed, but all of the sources are kept together
7879 anyway. In a normal installation, you need not specify these directory
7880 names when compiling or binding. Either the environment variables or
7881 the built-in defaults cause these files to be found.
7883 In addition to the language-defined hierarchies (@code{System}, @code{Ada} and
7884 @code{Interfaces}), the GNAT distribution provides a fourth hierarchy,
7885 consisting of child units of @code{GNAT}. This is a collection of generally
7886 useful types, subprograms, etc. @xref{Top, GNAT Reference Manual, About
7887 This Guid, gnat_rm, GNAT Reference Manual}, for further details.
7889 Besides simplifying access to the RTL, a major use of search paths is
7890 in compiling sources from multiple directories. This can make
7891 development environments much more flexible.
7893 @node Order of Compilation Issues
7894 @section Order of Compilation Issues
7897 If, in our earlier example, there was a spec for the @code{hello}
7898 procedure, it would be contained in the file @file{hello.ads}; yet this
7899 file would not have to be explicitly compiled. This is the result of the
7900 model we chose to implement library management. Some of the consequences
7901 of this model are as follows:
7905 There is no point in compiling specs (except for package
7906 specs with no bodies) because these are compiled as needed by clients. If
7907 you attempt a useless compilation, you will receive an error message.
7908 It is also useless to compile subunits because they are compiled as needed
7912 There are no order of compilation requirements: performing a
7913 compilation never obsoletes anything. The only way you can obsolete
7914 something and require recompilations is to modify one of the
7915 source files on which it depends.
7918 There is no library as such, apart from the ALI files
7919 (@pxref{The Ada Library Information Files}, for information on the format
7920 of these files). For now we find it convenient to create separate ALI files,
7921 but eventually the information therein may be incorporated into the object
7925 When you compile a unit, the source files for the specs of all units
7926 that it @code{with}'s, all its subunits, and the bodies of any generics it
7927 instantiates must be available (reachable by the search-paths mechanism
7928 described above), or you will receive a fatal error message.
7935 The following are some typical Ada compilation command line examples:
7938 @item $ gcc -c xyz.adb
7939 Compile body in file @file{xyz.adb} with all default options.
7942 @item $ gcc -c -O2 -gnata xyz-def.adb
7945 @item $ GNAT COMPILE /OPTIMIZE=ALL -gnata xyz-def.adb
7948 Compile the child unit package in file @file{xyz-def.adb} with extensive
7949 optimizations, and pragma @code{Assert}/@code{Debug} statements
7952 @item $ gcc -c -gnatc abc-def.adb
7953 Compile the subunit in file @file{abc-def.adb} in semantic-checking-only
7957 @node Binding Using gnatbind
7958 @chapter Binding Using @code{gnatbind}
7962 * Running gnatbind::
7963 * Switches for gnatbind::
7964 * Command-Line Access::
7965 * Search Paths for gnatbind::
7966 * Examples of gnatbind Usage::
7970 This chapter describes the GNAT binder, @code{gnatbind}, which is used
7971 to bind compiled GNAT objects.
7973 Note: to invoke @code{gnatbind} with a project file, use the @code{gnat}
7974 driver (see @ref{The GNAT Driver and Project Files}).
7976 The @code{gnatbind} program performs four separate functions:
7980 Checks that a program is consistent, in accordance with the rules in
7981 Chapter 10 of the Ada Reference Manual. In particular, error
7982 messages are generated if a program uses inconsistent versions of a
7986 Checks that an acceptable order of elaboration exists for the program
7987 and issues an error message if it cannot find an order of elaboration
7988 that satisfies the rules in Chapter 10 of the Ada Language Manual.
7991 Generates a main program incorporating the given elaboration order.
7992 This program is a small Ada package (body and spec) that
7993 must be subsequently compiled
7994 using the GNAT compiler. The necessary compilation step is usually
7995 performed automatically by @command{gnatlink}. The two most important
7996 functions of this program
7997 are to call the elaboration routines of units in an appropriate order
7998 and to call the main program.
8001 Determines the set of object files required by the given main program.
8002 This information is output in the forms of comments in the generated program,
8003 to be read by the @command{gnatlink} utility used to link the Ada application.
8006 @node Running gnatbind
8007 @section Running @code{gnatbind}
8010 The form of the @code{gnatbind} command is
8013 @c $ gnatbind @ovar{switches} @var{mainprog}@r{[}.ali@r{]} @ovar{switches}
8014 @c Expanding @ovar macro inline (explanation in macro def comments)
8015 $ gnatbind @r{[}@var{switches}@r{]} @var{mainprog}@r{[}.ali@r{]} @r{[}@var{switches}@r{]}
8019 where @file{@var{mainprog}.adb} is the Ada file containing the main program
8020 unit body. @code{gnatbind} constructs an Ada
8021 package in two files whose names are
8022 @file{b~@var{mainprog}.ads}, and @file{b~@var{mainprog}.adb}.
8023 For example, if given the
8024 parameter @file{hello.ali}, for a main program contained in file
8025 @file{hello.adb}, the binder output files would be @file{b~hello.ads}
8026 and @file{b~hello.adb}.
8028 When doing consistency checking, the binder takes into consideration
8029 any source files it can locate. For example, if the binder determines
8030 that the given main program requires the package @code{Pack}, whose
8032 file is @file{pack.ali} and whose corresponding source spec file is
8033 @file{pack.ads}, it attempts to locate the source file @file{pack.ads}
8034 (using the same search path conventions as previously described for the
8035 @command{gcc} command). If it can locate this source file, it checks that
8037 or source checksums of the source and its references to in @file{ALI} files
8038 match. In other words, any @file{ALI} files that mentions this spec must have
8039 resulted from compiling this version of the source file (or in the case
8040 where the source checksums match, a version close enough that the
8041 difference does not matter).
8043 @cindex Source files, use by binder
8044 The effect of this consistency checking, which includes source files, is
8045 that the binder ensures that the program is consistent with the latest
8046 version of the source files that can be located at bind time. Editing a
8047 source file without compiling files that depend on the source file cause
8048 error messages to be generated by the binder.
8050 For example, suppose you have a main program @file{hello.adb} and a
8051 package @code{P}, from file @file{p.ads} and you perform the following
8056 Enter @code{gcc -c hello.adb} to compile the main program.
8059 Enter @code{gcc -c p.ads} to compile package @code{P}.
8062 Edit file @file{p.ads}.
8065 Enter @code{gnatbind hello}.
8069 At this point, the file @file{p.ali} contains an out-of-date time stamp
8070 because the file @file{p.ads} has been edited. The attempt at binding
8071 fails, and the binder generates the following error messages:
8074 error: "hello.adb" must be recompiled ("p.ads" has been modified)
8075 error: "p.ads" has been modified and must be recompiled
8079 Now both files must be recompiled as indicated, and then the bind can
8080 succeed, generating a main program. You need not normally be concerned
8081 with the contents of this file, but for reference purposes a sample
8082 binder output file is given in @ref{Example of Binder Output File}.
8084 In most normal usage, the default mode of @command{gnatbind} which is to
8085 generate the main package in Ada, as described in the previous section.
8086 In particular, this means that any Ada programmer can read and understand
8087 the generated main program. It can also be debugged just like any other
8088 Ada code provided the @option{^-g^/DEBUG^} switch is used for
8089 @command{gnatbind} and @command{gnatlink}.
8091 @node Switches for gnatbind
8092 @section Switches for @command{gnatbind}
8095 The following switches are available with @code{gnatbind}; details will
8096 be presented in subsequent sections.
8099 * Consistency-Checking Modes::
8100 * Binder Error Message Control::
8101 * Elaboration Control::
8103 * Dynamic Allocation Control::
8104 * Binding with Non-Ada Main Programs::
8105 * Binding Programs with No Main Subprogram::
8112 @cindex @option{--version} @command{gnatbind}
8113 Display Copyright and version, then exit disregarding all other options.
8116 @cindex @option{--help} @command{gnatbind}
8117 If @option{--version} was not used, display usage, then exit disregarding
8121 @cindex @option{-a} @command{gnatbind}
8122 Indicates that, if supported by the platform, the adainit procedure should
8123 be treated as an initialisation routine by the linker (a constructor). This
8124 is intended to be used by the Project Manager to automatically initialize
8125 shared Stand-Alone Libraries.
8127 @item ^-aO^/OBJECT_SEARCH^
8128 @cindex @option{^-aO^/OBJECT_SEARCH^} (@command{gnatbind})
8129 Specify directory to be searched for ALI files.
8131 @item ^-aI^/SOURCE_SEARCH^
8132 @cindex @option{^-aI^/SOURCE_SEARCH^} (@command{gnatbind})
8133 Specify directory to be searched for source file.
8135 @item ^-A^/ALI_LIST^@r{[=}@var{filename}@r{]}
8136 @cindex @option{^-A^/ALI_LIST^} (@command{gnatbind})
8137 Output ALI list (to standard output or to the named file).
8139 @item ^-b^/REPORT_ERRORS=BRIEF^
8140 @cindex @option{^-b^/REPORT_ERRORS=BRIEF^} (@command{gnatbind})
8141 Generate brief messages to @file{stderr} even if verbose mode set.
8143 @item ^-c^/NOOUTPUT^
8144 @cindex @option{^-c^/NOOUTPUT^} (@command{gnatbind})
8145 Check only, no generation of binder output file.
8147 @item ^-d^/DEFAULT_STACK_SIZE=^@var{nn}@r{[}k@r{|}m@r{]}
8148 @cindex @option{^-d^/DEFAULT_STACK_SIZE=^@var{nn}@r{[}k@r{|}m@r{]}} (@command{gnatbind})
8149 This switch can be used to change the default task stack size value
8150 to a specified size @var{nn}, which is expressed in bytes by default, or
8151 in kilobytes when suffixed with @var{k} or in megabytes when suffixed
8153 In the absence of a @samp{@r{[}k@r{|}m@r{]}} suffix, this switch is equivalent,
8154 in effect, to completing all task specs with
8155 @smallexample @c ada
8156 pragma Storage_Size (nn);
8158 When they do not already have such a pragma.
8160 @item ^-D^/DEFAULT_SECONDARY_STACK_SIZE=^@var{nn}@r{[}k@r{|}m@r{]}
8161 @cindex @option{^-D^/DEFAULT_SECONDARY_STACK_SIZE=nnnnn^} (@command{gnatbind})
8162 This switch can be used to change the default secondary stack size value
8163 to a specified size @var{nn}, which is expressed in bytes by default, or
8164 in kilobytes when suffixed with @var{k} or in megabytes when suffixed
8167 The secondary stack is used to deal with functions that return a variable
8168 sized result, for example a function returning an unconstrained
8169 String. There are two ways in which this secondary stack is allocated.
8171 For most targets, the secondary stack is growing on demand and is allocated
8172 as a chain of blocks in the heap. The -D option is not very
8173 relevant. It only give some control over the size of the allocated
8174 blocks (whose size is the minimum of the default secondary stack size value,
8175 and the actual size needed for the current allocation request).
8177 For certain targets, notably VxWorks 653,
8178 the secondary stack is allocated by carving off a fixed ratio chunk of the
8179 primary task stack. The -D option is used to define the
8180 size of the environment task's secondary stack.
8182 @item ^-e^/ELABORATION_DEPENDENCIES^
8183 @cindex @option{^-e^/ELABORATION_DEPENDENCIES^} (@command{gnatbind})
8184 Output complete list of elaboration-order dependencies.
8186 @item ^-E^/STORE_TRACEBACKS^
8187 @cindex @option{^-E^/STORE_TRACEBACKS^} (@command{gnatbind})
8188 Store tracebacks in exception occurrences when the target supports it.
8190 @c The following may get moved to an appendix
8191 This option is currently supported on the following targets:
8192 all x86 ports, Solaris, Windows, HP-UX, AIX, PowerPC VxWorks and Alpha VxWorks.
8194 See also the packages @code{GNAT.Traceback} and
8195 @code{GNAT.Traceback.Symbolic} for more information.
8197 Note that on x86 ports, you must not use @option{-fomit-frame-pointer}
8198 @command{gcc} option.
8201 @item ^-F^/FORCE_ELABS_FLAGS^
8202 @cindex @option{^-F^/FORCE_ELABS_FLAGS^} (@command{gnatbind})
8203 Force the checks of elaboration flags. @command{gnatbind} does not normally
8204 generate checks of elaboration flags for the main executable, except when
8205 a Stand-Alone Library is used. However, there are cases when this cannot be
8206 detected by gnatbind. An example is importing an interface of a Stand-Alone
8207 Library through a pragma Import and only specifying through a linker switch
8208 this Stand-Alone Library. This switch is used to guarantee that elaboration
8209 flag checks are generated.
8212 @cindex @option{^-h^/HELP^} (@command{gnatbind})
8213 Output usage (help) information
8215 @item ^-H32^/32_MALLOC^
8216 @cindex @option{^-H32^/32_MALLOC^} (@command{gnatbind})
8217 Use 32-bit allocations for @code{__gnat_malloc} (and thus for access types).
8218 For further details see @ref{Dynamic Allocation Control}.
8220 @item ^-H64^/64_MALLOC^
8221 @cindex @option{^-H32^/32_MALLOC^} (@command{gnatbind})
8222 Use 64-bit allocations for @code{__gnat_malloc} (and thus for access types).
8223 @cindex @code{__gnat_malloc}
8224 For further details see @ref{Dynamic Allocation Control}.
8227 @cindex @option{^-I^/SEARCH^} (@command{gnatbind})
8228 Specify directory to be searched for source and ALI files.
8230 @item ^-I-^/NOCURRENT_DIRECTORY^
8231 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@command{gnatbind})
8232 Do not look for sources in the current directory where @code{gnatbind} was
8233 invoked, and do not look for ALI files in the directory containing the
8234 ALI file named in the @code{gnatbind} command line.
8236 @item ^-l^/ORDER_OF_ELABORATION^
8237 @cindex @option{^-l^/ORDER_OF_ELABORATION^} (@command{gnatbind})
8238 Output chosen elaboration order.
8240 @item ^-L@var{xxx}^/BUILD_LIBRARY=@var{xxx}^
8241 @cindex @option{^-L^/BUILD_LIBRARY^} (@command{gnatbind})
8242 Bind the units for library building. In this case the adainit and
8243 adafinal procedures (@pxref{Binding with Non-Ada Main Programs})
8244 are renamed to ^@var{xxx}init^@var{XXX}INIT^ and
8245 ^@var{xxx}final^@var{XXX}FINAL^.
8246 Implies ^-n^/NOCOMPILE^.
8248 (@xref{GNAT and Libraries}, for more details.)
8251 On OpenVMS, these init and final procedures are exported in uppercase
8252 letters. For example if /BUILD_LIBRARY=toto is used, the exported name of
8253 the init procedure will be "TOTOINIT" and the exported name of the final
8254 procedure will be "TOTOFINAL".
8257 @item ^-Mxyz^/RENAME_MAIN=xyz^
8258 @cindex @option{^-M^/RENAME_MAIN^} (@command{gnatbind})
8259 Rename generated main program from main to xyz. This option is
8260 supported on cross environments only.
8262 @item ^-m^/ERROR_LIMIT=^@var{n}
8263 @cindex @option{^-m^/ERROR_LIMIT^} (@command{gnatbind})
8264 Limit number of detected errors or warnings to @var{n}, where @var{n} is
8265 in the range 1..999999. The default value if no switch is
8266 given is 9999. If the number of warnings reaches this limit, then a
8267 message is output and further warnings are suppressed, the bind
8268 continues in this case. If the number of errors reaches this
8269 limit, then a message is output and the bind is abandoned.
8270 A value of zero means that no limit is enforced. The equal
8274 Furthermore, under Windows, the sources pointed to by the libraries path
8275 set in the registry are not searched for.
8279 @cindex @option{^-n^/NOMAIN^} (@command{gnatbind})
8283 @cindex @option{-nostdinc} (@command{gnatbind})
8284 Do not look for sources in the system default directory.
8287 @cindex @option{-nostdlib} (@command{gnatbind})
8288 Do not look for library files in the system default directory.
8290 @item --RTS=@var{rts-path}
8291 @cindex @option{--RTS} (@code{gnatbind})
8292 Specifies the default location of the runtime library. Same meaning as the
8293 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
8295 @item ^-o ^/OUTPUT=^@var{file}
8296 @cindex @option{^-o ^/OUTPUT^} (@command{gnatbind})
8297 Name the output file @var{file} (default is @file{b~@var{xxx}.adb}).
8298 Note that if this option is used, then linking must be done manually,
8299 gnatlink cannot be used.
8301 @item ^-O^/OBJECT_LIST^@r{[=}@var{filename}@r{]}
8302 @cindex @option{^-O^/OBJECT_LIST^} (@command{gnatbind})
8303 Output object list (to standard output or to the named file).
8305 @item ^-p^/PESSIMISTIC_ELABORATION^
8306 @cindex @option{^-p^/PESSIMISTIC_ELABORATION^} (@command{gnatbind})
8307 Pessimistic (worst-case) elaboration order
8310 @cindex @option{^-R^-R^} (@command{gnatbind})
8311 Output closure source list.
8313 @item ^-s^/READ_SOURCES=ALL^
8314 @cindex @option{^-s^/READ_SOURCES=ALL^} (@command{gnatbind})
8315 Require all source files to be present.
8317 @item ^-S@var{xxx}^/INITIALIZE_SCALARS=@var{xxx}^
8318 @cindex @option{^-S^/INITIALIZE_SCALARS^} (@command{gnatbind})
8319 Specifies the value to be used when detecting uninitialized scalar
8320 objects with pragma Initialize_Scalars.
8321 The @var{xxx} ^string specified with the switch^option^ may be either
8323 @item ``@option{^in^INVALID^}'' requesting an invalid value where possible
8324 @item ``@option{^lo^LOW^}'' for the lowest possible value
8325 @item ``@option{^hi^HIGH^}'' for the highest possible value
8326 @item ``@option{@var{xx}}'' for a value consisting of repeated bytes with the
8327 value @code{16#@var{xx}#} (i.e., @var{xx} is a string of two hexadecimal digits).
8330 In addition, you can specify @option{-Sev} to indicate that the value is
8331 to be set at run time. In this case, the program will look for an environment
8332 @cindex GNAT_INIT_SCALARS
8333 variable of the form @env{GNAT_INIT_SCALARS=@var{xx}}, where @var{xx} is one
8334 of @option{in/lo/hi/@var{xx}} with the same meanings as above.
8335 If no environment variable is found, or if it does not have a valid value,
8336 then the default is @option{in} (invalid values).
8340 @cindex @option{-static} (@code{gnatbind})
8341 Link against a static GNAT run time.
8344 @cindex @option{-shared} (@code{gnatbind})
8345 Link against a shared GNAT run time when available.
8348 @item ^-t^/NOTIME_STAMP_CHECK^
8349 @cindex @option{^-t^/NOTIME_STAMP_CHECK^} (@code{gnatbind})
8350 Tolerate time stamp and other consistency errors
8352 @item ^-T@var{n}^/TIME_SLICE=@var{n}^
8353 @cindex @option{^-T^/TIME_SLICE^} (@code{gnatbind})
8354 Set the time slice value to @var{n} milliseconds. If the system supports
8355 the specification of a specific time slice value, then the indicated value
8356 is used. If the system does not support specific time slice values, but
8357 does support some general notion of round-robin scheduling, then any
8358 nonzero value will activate round-robin scheduling.
8360 A value of zero is treated specially. It turns off time
8361 slicing, and in addition, indicates to the tasking run time that the
8362 semantics should match as closely as possible the Annex D
8363 requirements of the Ada RM, and in particular sets the default
8364 scheduling policy to @code{FIFO_Within_Priorities}.
8366 @item ^-u@var{n}^/DYNAMIC_STACK_USAGE=@var{n}^
8367 @cindex @option{^-u^/DYNAMIC_STACK_USAGE^} (@code{gnatbind})
8368 Enable dynamic stack usage, with @var{n} results stored and displayed
8369 at program termination. A result is generated when a task
8370 terminates. Results that can't be stored are displayed on the fly, at
8371 task termination. This option is currently not supported on Itanium
8372 platforms. (See @ref{Dynamic Stack Usage Analysis} for details.)
8374 @item ^-v^/REPORT_ERRORS=VERBOSE^
8375 @cindex @option{^-v^/REPORT_ERRORS=VERBOSE^} (@code{gnatbind})
8376 Verbose mode. Write error messages, header, summary output to
8381 @cindex @option{-w} (@code{gnatbind})
8382 Warning mode (@var{x}=s/e for suppress/treat as error)
8386 @item /WARNINGS=NORMAL
8387 @cindex @option{/WARNINGS} (@code{gnatbind})
8388 Normal warnings mode. Warnings are issued but ignored
8390 @item /WARNINGS=SUPPRESS
8391 @cindex @option{/WARNINGS} (@code{gnatbind})
8392 All warning messages are suppressed
8394 @item /WARNINGS=ERROR
8395 @cindex @option{/WARNINGS} (@code{gnatbind})
8396 Warning messages are treated as fatal errors
8399 @item ^-Wx^/WIDE_CHARACTER_ENCODING=^@var{e}
8400 @cindex @option{^-Wx^/WIDE_CHARACTER_ENCODING^} (@code{gnatbind})
8401 Override default wide character encoding for standard Text_IO files.
8403 @item ^-x^/READ_SOURCES=NONE^
8404 @cindex @option{^-x^/READ_SOURCES^} (@code{gnatbind})
8405 Exclude source files (check object consistency only).
8408 @item /READ_SOURCES=AVAILABLE
8409 @cindex @option{/READ_SOURCES} (@code{gnatbind})
8410 Default mode, in which sources are checked for consistency only if
8414 @item ^-y^/ENABLE_LEAP_SECONDS^
8415 @cindex @option{^-y^/ENABLE_LEAP_SECONDS^} (@code{gnatbind})
8416 Enable leap seconds support in @code{Ada.Calendar} and its children.
8418 @item ^-z^/ZERO_MAIN^
8419 @cindex @option{^-z^/ZERO_MAIN^} (@code{gnatbind})
8425 You may obtain this listing of switches by running @code{gnatbind} with
8429 @node Consistency-Checking Modes
8430 @subsection Consistency-Checking Modes
8433 As described earlier, by default @code{gnatbind} checks
8434 that object files are consistent with one another and are consistent
8435 with any source files it can locate. The following switches control binder
8440 @item ^-s^/READ_SOURCES=ALL^
8441 @cindex @option{^-s^/READ_SOURCES=ALL^} (@code{gnatbind})
8442 Require source files to be present. In this mode, the binder must be
8443 able to locate all source files that are referenced, in order to check
8444 their consistency. In normal mode, if a source file cannot be located it
8445 is simply ignored. If you specify this switch, a missing source
8448 @item ^-Wx^/WIDE_CHARACTER_ENCODING=^@var{e}
8449 @cindex @option{^-Wx^/WIDE_CHARACTER_ENCODING^} (@code{gnatbind})
8450 Override default wide character encoding for standard Text_IO files.
8451 Normally the default wide character encoding method used for standard
8452 [Wide_[Wide_]]Text_IO files is taken from the encoding specified for
8453 the main source input (see description of switch
8454 @option{^-gnatWx^/WIDE_CHARACTER_ENCODING^} for the compiler). The
8455 use of this switch for the binder (which has the same set of
8456 possible arguments) overrides this default as specified.
8458 @item ^-x^/READ_SOURCES=NONE^
8459 @cindex @option{^-x^/READ_SOURCES=NONE^} (@code{gnatbind})
8460 Exclude source files. In this mode, the binder only checks that ALI
8461 files are consistent with one another. Source files are not accessed.
8462 The binder runs faster in this mode, and there is still a guarantee that
8463 the resulting program is self-consistent.
8464 If a source file has been edited since it was last compiled, and you
8465 specify this switch, the binder will not detect that the object
8466 file is out of date with respect to the source file. Note that this is the
8467 mode that is automatically used by @command{gnatmake} because in this
8468 case the checking against sources has already been performed by
8469 @command{gnatmake} in the course of compilation (i.e.@: before binding).
8472 @item /READ_SOURCES=AVAILABLE
8473 @cindex @code{/READ_SOURCES=AVAILABLE} (@code{gnatbind})
8474 This is the default mode in which source files are checked if they are
8475 available, and ignored if they are not available.
8479 @node Binder Error Message Control
8480 @subsection Binder Error Message Control
8483 The following switches provide control over the generation of error
8484 messages from the binder:
8488 @item ^-v^/REPORT_ERRORS=VERBOSE^
8489 @cindex @option{^-v^/REPORT_ERRORS=VERBOSE^} (@code{gnatbind})
8490 Verbose mode. In the normal mode, brief error messages are generated to
8491 @file{stderr}. If this switch is present, a header is written
8492 to @file{stdout} and any error messages are directed to @file{stdout}.
8493 All that is written to @file{stderr} is a brief summary message.
8495 @item ^-b^/REPORT_ERRORS=BRIEF^
8496 @cindex @option{^-b^/REPORT_ERRORS=BRIEF^} (@code{gnatbind})
8497 Generate brief error messages to @file{stderr} even if verbose mode is
8498 specified. This is relevant only when used with the
8499 @option{^-v^/REPORT_ERRORS=VERBOSE^} switch.
8503 @cindex @option{-m} (@code{gnatbind})
8504 Limits the number of error messages to @var{n}, a decimal integer in the
8505 range 1-999. The binder terminates immediately if this limit is reached.
8508 @cindex @option{-M} (@code{gnatbind})
8509 Renames the generated main program from @code{main} to @code{xxx}.
8510 This is useful in the case of some cross-building environments, where
8511 the actual main program is separate from the one generated
8515 @item ^-ws^/WARNINGS=SUPPRESS^
8516 @cindex @option{^-ws^/WARNINGS=SUPPRESS^} (@code{gnatbind})
8518 Suppress all warning messages.
8520 @item ^-we^/WARNINGS=ERROR^
8521 @cindex @option{^-we^/WARNINGS=ERROR^} (@code{gnatbind})
8522 Treat any warning messages as fatal errors.
8525 @item /WARNINGS=NORMAL
8526 Standard mode with warnings generated, but warnings do not get treated
8530 @item ^-t^/NOTIME_STAMP_CHECK^
8531 @cindex @option{^-t^/NOTIME_STAMP_CHECK^} (@code{gnatbind})
8532 @cindex Time stamp checks, in binder
8533 @cindex Binder consistency checks
8534 @cindex Consistency checks, in binder
8535 The binder performs a number of consistency checks including:
8539 Check that time stamps of a given source unit are consistent
8541 Check that checksums of a given source unit are consistent
8543 Check that consistent versions of @code{GNAT} were used for compilation
8545 Check consistency of configuration pragmas as required
8549 Normally failure of such checks, in accordance with the consistency
8550 requirements of the Ada Reference Manual, causes error messages to be
8551 generated which abort the binder and prevent the output of a binder
8552 file and subsequent link to obtain an executable.
8554 The @option{^-t^/NOTIME_STAMP_CHECK^} switch converts these error messages
8555 into warnings, so that
8556 binding and linking can continue to completion even in the presence of such
8557 errors. The result may be a failed link (due to missing symbols), or a
8558 non-functional executable which has undefined semantics.
8559 @emph{This means that
8560 @option{^-t^/NOTIME_STAMP_CHECK^} should be used only in unusual situations,
8564 @node Elaboration Control
8565 @subsection Elaboration Control
8568 The following switches provide additional control over the elaboration
8569 order. For full details see @ref{Elaboration Order Handling in GNAT}.
8572 @item ^-p^/PESSIMISTIC_ELABORATION^
8573 @cindex @option{^-p^/PESSIMISTIC_ELABORATION^} (@code{gnatbind})
8574 Normally the binder attempts to choose an elaboration order that is
8575 likely to minimize the likelihood of an elaboration order error resulting
8576 in raising a @code{Program_Error} exception. This switch reverses the
8577 action of the binder, and requests that it deliberately choose an order
8578 that is likely to maximize the likelihood of an elaboration error.
8579 This is useful in ensuring portability and avoiding dependence on
8580 accidental fortuitous elaboration ordering.
8582 Normally it only makes sense to use the @option{^-p^/PESSIMISTIC_ELABORATION^}
8584 elaboration checking is used (@option{-gnatE} switch used for compilation).
8585 This is because in the default static elaboration mode, all necessary
8586 @code{Elaborate} and @code{Elaborate_All} pragmas are implicitly inserted.
8587 These implicit pragmas are still respected by the binder in
8588 @option{^-p^/PESSIMISTIC_ELABORATION^} mode, so a
8589 safe elaboration order is assured.
8592 @node Output Control
8593 @subsection Output Control
8596 The following switches allow additional control over the output
8597 generated by the binder.
8602 @item ^-c^/NOOUTPUT^
8603 @cindex @option{^-c^/NOOUTPUT^} (@code{gnatbind})
8604 Check only. Do not generate the binder output file. In this mode the
8605 binder performs all error checks but does not generate an output file.
8607 @item ^-e^/ELABORATION_DEPENDENCIES^
8608 @cindex @option{^-e^/ELABORATION_DEPENDENCIES^} (@code{gnatbind})
8609 Output complete list of elaboration-order dependencies, showing the
8610 reason for each dependency. This output can be rather extensive but may
8611 be useful in diagnosing problems with elaboration order. The output is
8612 written to @file{stdout}.
8615 @cindex @option{^-h^/HELP^} (@code{gnatbind})
8616 Output usage information. The output is written to @file{stdout}.
8618 @item ^-K^/LINKER_OPTION_LIST^
8619 @cindex @option{^-K^/LINKER_OPTION_LIST^} (@code{gnatbind})
8620 Output linker options to @file{stdout}. Includes library search paths,
8621 contents of pragmas Ident and Linker_Options, and libraries added
8624 @item ^-l^/ORDER_OF_ELABORATION^
8625 @cindex @option{^-l^/ORDER_OF_ELABORATION^} (@code{gnatbind})
8626 Output chosen elaboration order. The output is written to @file{stdout}.
8628 @item ^-O^/OBJECT_LIST^
8629 @cindex @option{^-O^/OBJECT_LIST^} (@code{gnatbind})
8630 Output full names of all the object files that must be linked to provide
8631 the Ada component of the program. The output is written to @file{stdout}.
8632 This list includes the files explicitly supplied and referenced by the user
8633 as well as implicitly referenced run-time unit files. The latter are
8634 omitted if the corresponding units reside in shared libraries. The
8635 directory names for the run-time units depend on the system configuration.
8637 @item ^-o ^/OUTPUT=^@var{file}
8638 @cindex @option{^-o^/OUTPUT^} (@code{gnatbind})
8639 Set name of output file to @var{file} instead of the normal
8640 @file{b~@var{mainprog}.adb} default. Note that @var{file} denote the Ada
8641 binder generated body filename.
8642 Note that if this option is used, then linking must be done manually.
8643 It is not possible to use gnatlink in this case, since it cannot locate
8646 @item ^-r^/RESTRICTION_LIST^
8647 @cindex @option{^-r^/RESTRICTION_LIST^} (@code{gnatbind})
8648 Generate list of @code{pragma Restrictions} that could be applied to
8649 the current unit. This is useful for code audit purposes, and also may
8650 be used to improve code generation in some cases.
8654 @node Dynamic Allocation Control
8655 @subsection Dynamic Allocation Control
8658 The heap control switches -- @option{-H32} and @option{-H64} --
8659 determine whether dynamic allocation uses 32-bit or 64-bit memory.
8660 They only affect compiler-generated allocations via @code{__gnat_malloc};
8661 explicit calls to @code{malloc} and related functions from the C
8662 run-time library are unaffected.
8666 Allocate memory on 32-bit heap
8669 Allocate memory on 64-bit heap. This is the default
8670 unless explicitly overridden by a @code{'Size} clause on the access type.
8675 See also @ref{Access types and 32/64-bit allocation}.
8679 These switches are only effective on VMS platforms.
8683 @node Binding with Non-Ada Main Programs
8684 @subsection Binding with Non-Ada Main Programs
8687 In our description so far we have assumed that the main
8688 program is in Ada, and that the task of the binder is to generate a
8689 corresponding function @code{main} that invokes this Ada main
8690 program. GNAT also supports the building of executable programs where
8691 the main program is not in Ada, but some of the called routines are
8692 written in Ada and compiled using GNAT (@pxref{Mixed Language Programming}).
8693 The following switch is used in this situation:
8697 @cindex @option{^-n^/NOMAIN^} (@code{gnatbind})
8698 No main program. The main program is not in Ada.
8702 In this case, most of the functions of the binder are still required,
8703 but instead of generating a main program, the binder generates a file
8704 containing the following callable routines:
8709 You must call this routine to initialize the Ada part of the program by
8710 calling the necessary elaboration routines. A call to @code{adainit} is
8711 required before the first call to an Ada subprogram.
8713 Note that it is assumed that the basic execution environment must be setup
8714 to be appropriate for Ada execution at the point where the first Ada
8715 subprogram is called. In particular, if the Ada code will do any
8716 floating-point operations, then the FPU must be setup in an appropriate
8717 manner. For the case of the x86, for example, full precision mode is
8718 required. The procedure GNAT.Float_Control.Reset may be used to ensure
8719 that the FPU is in the right state.
8723 You must call this routine to perform any library-level finalization
8724 required by the Ada subprograms. A call to @code{adafinal} is required
8725 after the last call to an Ada subprogram, and before the program
8730 If the @option{^-n^/NOMAIN^} switch
8731 @cindex @option{^-n^/NOMAIN^} (@command{gnatbind})
8732 @cindex Binder, multiple input files
8733 is given, more than one ALI file may appear on
8734 the command line for @code{gnatbind}. The normal @dfn{closure}
8735 calculation is performed for each of the specified units. Calculating
8736 the closure means finding out the set of units involved by tracing
8737 @code{with} references. The reason it is necessary to be able to
8738 specify more than one ALI file is that a given program may invoke two or
8739 more quite separate groups of Ada units.
8741 The binder takes the name of its output file from the last specified ALI
8742 file, unless overridden by the use of the @option{^-o file^/OUTPUT=file^}.
8743 @cindex @option{^-o^/OUTPUT^} (@command{gnatbind})
8744 The output is an Ada unit in source form that can be compiled with GNAT.
8745 This compilation occurs automatically as part of the @command{gnatlink}
8748 Currently the GNAT run time requires a FPU using 80 bits mode
8749 precision. Under targets where this is not the default it is required to
8750 call GNAT.Float_Control.Reset before using floating point numbers (this
8751 include float computation, float input and output) in the Ada code. A
8752 side effect is that this could be the wrong mode for the foreign code
8753 where floating point computation could be broken after this call.
8755 @node Binding Programs with No Main Subprogram
8756 @subsection Binding Programs with No Main Subprogram
8759 It is possible to have an Ada program which does not have a main
8760 subprogram. This program will call the elaboration routines of all the
8761 packages, then the finalization routines.
8763 The following switch is used to bind programs organized in this manner:
8766 @item ^-z^/ZERO_MAIN^
8767 @cindex @option{^-z^/ZERO_MAIN^} (@code{gnatbind})
8768 Normally the binder checks that the unit name given on the command line
8769 corresponds to a suitable main subprogram. When this switch is used,
8770 a list of ALI files can be given, and the execution of the program
8771 consists of elaboration of these units in an appropriate order. Note
8772 that the default wide character encoding method for standard Text_IO
8773 files is always set to Brackets if this switch is set (you can use
8775 @option{^-Wx^WIDE_CHARACTER_ENCODING^} to override this default).
8778 @node Command-Line Access
8779 @section Command-Line Access
8782 The package @code{Ada.Command_Line} provides access to the command-line
8783 arguments and program name. In order for this interface to operate
8784 correctly, the two variables
8796 are declared in one of the GNAT library routines. These variables must
8797 be set from the actual @code{argc} and @code{argv} values passed to the
8798 main program. With no @option{^n^/NOMAIN^} present, @code{gnatbind}
8799 generates the C main program to automatically set these variables.
8800 If the @option{^n^/NOMAIN^} switch is used, there is no automatic way to
8801 set these variables. If they are not set, the procedures in
8802 @code{Ada.Command_Line} will not be available, and any attempt to use
8803 them will raise @code{Constraint_Error}. If command line access is
8804 required, your main program must set @code{gnat_argc} and
8805 @code{gnat_argv} from the @code{argc} and @code{argv} values passed to
8808 @node Search Paths for gnatbind
8809 @section Search Paths for @code{gnatbind}
8812 The binder takes the name of an ALI file as its argument and needs to
8813 locate source files as well as other ALI files to verify object consistency.
8815 For source files, it follows exactly the same search rules as @command{gcc}
8816 (@pxref{Search Paths and the Run-Time Library (RTL)}). For ALI files the
8817 directories searched are:
8821 The directory containing the ALI file named in the command line, unless
8822 the switch @option{^-I-^/NOCURRENT_DIRECTORY^} is specified.
8825 All directories specified by @option{^-I^/SEARCH^}
8826 switches on the @code{gnatbind}
8827 command line, in the order given.
8830 @findex ADA_PRJ_OBJECTS_FILE
8831 Each of the directories listed in the text file whose name is given
8832 by the @env{ADA_PRJ_OBJECTS_FILE} ^environment variable^logical name^.
8835 @env{ADA_PRJ_OBJECTS_FILE} is normally set by gnatmake or by the ^gnat^GNAT^
8836 driver when project files are used. It should not normally be set
8840 @findex ADA_OBJECTS_PATH
8841 Each of the directories listed in the value of the
8842 @env{ADA_OBJECTS_PATH} ^environment variable^logical name^.
8844 Construct this value
8845 exactly as the @env{PATH} environment variable: a list of directory
8846 names separated by colons (semicolons when working with the NT version
8850 Normally, define this value as a logical name containing a comma separated
8851 list of directory names.
8853 This variable can also be defined by means of an environment string
8854 (an argument to the HP C exec* set of functions).
8858 DEFINE ANOTHER_PATH FOO:[BAG]
8859 DEFINE ADA_OBJECTS_PATH ANOTHER_PATH,FOO:[BAM],FOO:[BAR]
8862 By default, the path includes GNU:[LIB.OPENVMS7_x.2_8_x.DECLIB]
8863 first, followed by the standard Ada
8864 libraries in GNU:[LIB.OPENVMS7_x.2_8_x.ADALIB].
8865 If this is not redefined, the user will obtain the HP Ada 83 IO packages
8866 (Text_IO, Sequential_IO, etc)
8867 instead of the standard Ada packages. Thus, in order to get the standard Ada
8868 packages by default, ADA_OBJECTS_PATH must be redefined.
8872 The content of the @file{ada_object_path} file which is part of the GNAT
8873 installation tree and is used to store standard libraries such as the
8874 GNAT Run Time Library (RTL) unless the switch @option{-nostdlib} is
8877 @ref{Installing a library}
8882 In the binder the switch @option{^-I^/SEARCH^}
8883 @cindex @option{^-I^/SEARCH^} (@command{gnatbind})
8884 is used to specify both source and
8885 library file paths. Use @option{^-aI^/SOURCE_SEARCH^}
8886 @cindex @option{^-aI^/SOURCE_SEARCH^} (@command{gnatbind})
8887 instead if you want to specify
8888 source paths only, and @option{^-aO^/LIBRARY_SEARCH^}
8889 @cindex @option{^-aO^/LIBRARY_SEARCH^} (@command{gnatbind})
8890 if you want to specify library paths
8891 only. This means that for the binder
8892 @option{^-I^/SEARCH=^}@var{dir} is equivalent to
8893 @option{^-aI^/SOURCE_SEARCH=^}@var{dir}
8894 @option{^-aO^/OBJECT_SEARCH=^}@var{dir}.
8895 The binder generates the bind file (a C language source file) in the
8896 current working directory.
8902 The packages @code{Ada}, @code{System}, and @code{Interfaces} and their
8903 children make up the GNAT Run-Time Library, together with the package
8904 GNAT and its children, which contain a set of useful additional
8905 library functions provided by GNAT. The sources for these units are
8906 needed by the compiler and are kept together in one directory. The ALI
8907 files and object files generated by compiling the RTL are needed by the
8908 binder and the linker and are kept together in one directory, typically
8909 different from the directory containing the sources. In a normal
8910 installation, you need not specify these directory names when compiling
8911 or binding. Either the environment variables or the built-in defaults
8912 cause these files to be found.
8914 Besides simplifying access to the RTL, a major use of search paths is
8915 in compiling sources from multiple directories. This can make
8916 development environments much more flexible.
8918 @node Examples of gnatbind Usage
8919 @section Examples of @code{gnatbind} Usage
8922 This section contains a number of examples of using the GNAT binding
8923 utility @code{gnatbind}.
8926 @item gnatbind hello
8927 The main program @code{Hello} (source program in @file{hello.adb}) is
8928 bound using the standard switch settings. The generated main program is
8929 @file{b~hello.adb}. This is the normal, default use of the binder.
8932 @item gnatbind hello -o mainprog.adb
8935 @item gnatbind HELLO.ALI /OUTPUT=Mainprog.ADB
8937 The main program @code{Hello} (source program in @file{hello.adb}) is
8938 bound using the standard switch settings. The generated main program is
8939 @file{mainprog.adb} with the associated spec in
8940 @file{mainprog.ads}. Note that you must specify the body here not the
8941 spec. Note that if this option is used, then linking must be done manually,
8942 since gnatlink will not be able to find the generated file.
8945 @c ------------------------------------
8946 @node Linking Using gnatlink
8947 @chapter Linking Using @command{gnatlink}
8948 @c ------------------------------------
8952 This chapter discusses @command{gnatlink}, a tool that links
8953 an Ada program and builds an executable file. This utility
8954 invokes the system linker ^(via the @command{gcc} command)^^
8955 with a correct list of object files and library references.
8956 @command{gnatlink} automatically determines the list of files and
8957 references for the Ada part of a program. It uses the binder file
8958 generated by the @command{gnatbind} to determine this list.
8960 Note: to invoke @code{gnatlink} with a project file, use the @code{gnat}
8961 driver (see @ref{The GNAT Driver and Project Files}).
8964 * Running gnatlink::
8965 * Switches for gnatlink::
8968 @node Running gnatlink
8969 @section Running @command{gnatlink}
8972 The form of the @command{gnatlink} command is
8975 @c $ gnatlink @ovar{switches} @var{mainprog}@r{[}.ali@r{]}
8976 @c @ovar{non-Ada objects} @ovar{linker options}
8977 @c Expanding @ovar macro inline (explanation in macro def comments)
8978 $ gnatlink @r{[}@var{switches}@r{]} @var{mainprog}@r{[}.ali@r{]}
8979 @r{[}@var{non-Ada objects}@r{]} @r{[}@var{linker options}@r{]}
8984 The arguments of @command{gnatlink} (switches, main @file{ALI} file,
8986 or linker options) may be in any order, provided that no non-Ada object may
8987 be mistaken for a main @file{ALI} file.
8988 Any file name @file{F} without the @file{.ali}
8989 extension will be taken as the main @file{ALI} file if a file exists
8990 whose name is the concatenation of @file{F} and @file{.ali}.
8993 @file{@var{mainprog}.ali} references the ALI file of the main program.
8994 The @file{.ali} extension of this file can be omitted. From this
8995 reference, @command{gnatlink} locates the corresponding binder file
8996 @file{b~@var{mainprog}.adb} and, using the information in this file along
8997 with the list of non-Ada objects and linker options, constructs a
8998 linker command file to create the executable.
9000 The arguments other than the @command{gnatlink} switches and the main
9001 @file{ALI} file are passed to the linker uninterpreted.
9002 They typically include the names of
9003 object files for units written in other languages than Ada and any library
9004 references required to resolve references in any of these foreign language
9005 units, or in @code{Import} pragmas in any Ada units.
9007 @var{linker options} is an optional list of linker specific
9009 The default linker called by gnatlink is @command{gcc} which in
9010 turn calls the appropriate system linker.
9012 One useful option for the linker is @option{-s}: it reduces the size of the
9013 executable by removing all symbol table and relocation information from the
9016 Standard options for the linker such as @option{-lmy_lib} or
9017 @option{-Ldir} can be added as is.
9018 For options that are not recognized by
9019 @command{gcc} as linker options, use the @command{gcc} switches
9020 @option{-Xlinker} or @option{-Wl,}.
9022 Refer to the GCC documentation for
9025 Here is an example showing how to generate a linker map:
9028 $ ^gnatlink my_prog -Wl,-Map,MAPFILE^GNAT LINK my_prog.ali /MAP^
9031 Using @var{linker options} it is possible to set the program stack and
9034 See @ref{Setting Stack Size from gnatlink} and
9035 @ref{Setting Heap Size from gnatlink}.
9038 @command{gnatlink} determines the list of objects required by the Ada
9039 program and prepends them to the list of objects passed to the linker.
9040 @command{gnatlink} also gathers any arguments set by the use of
9041 @code{pragma Linker_Options} and adds them to the list of arguments
9042 presented to the linker.
9045 @command{gnatlink} accepts the following types of extra files on the command
9046 line: objects (@file{.OBJ}), libraries (@file{.OLB}), sharable images
9047 (@file{.EXE}), and options files (@file{.OPT}). These are recognized and
9048 handled according to their extension.
9051 @node Switches for gnatlink
9052 @section Switches for @command{gnatlink}
9055 The following switches are available with the @command{gnatlink} utility:
9061 @cindex @option{--version} @command{gnatlink}
9062 Display Copyright and version, then exit disregarding all other options.
9065 @cindex @option{--help} @command{gnatlink}
9066 If @option{--version} was not used, display usage, then exit disregarding
9069 @item ^-f^/FORCE_OBJECT_FILE_LIST^
9070 @cindex Command line length
9071 @cindex @option{^-f^/FORCE_OBJECT_FILE_LIST^} (@command{gnatlink})
9072 On some targets, the command line length is limited, and @command{gnatlink}
9073 will generate a separate file for the linker if the list of object files
9075 The @option{^-f^/FORCE_OBJECT_FILE_LIST^} switch forces this file
9076 to be generated even if
9077 the limit is not exceeded. This is useful in some cases to deal with
9078 special situations where the command line length is exceeded.
9081 @cindex Debugging information, including
9082 @cindex @option{^-g^/DEBUG^} (@command{gnatlink})
9083 The option to include debugging information causes the Ada bind file (in
9084 other words, @file{b~@var{mainprog}.adb}) to be compiled with
9085 @option{^-g^/DEBUG^}.
9086 In addition, the binder does not delete the @file{b~@var{mainprog}.adb},
9087 @file{b~@var{mainprog}.o} and @file{b~@var{mainprog}.ali} files.
9088 Without @option{^-g^/DEBUG^}, the binder removes these files by
9089 default. The same procedure apply if a C bind file was generated using
9090 @option{^-C^/BIND_FILE=C^} @code{gnatbind} option, in this case the filenames
9091 are @file{b_@var{mainprog}.c} and @file{b_@var{mainprog}.o}.
9093 @item ^-n^/NOCOMPILE^
9094 @cindex @option{^-n^/NOCOMPILE^} (@command{gnatlink})
9095 Do not compile the file generated by the binder. This may be used when
9096 a link is rerun with different options, but there is no need to recompile
9100 @cindex @option{^-v^/VERBOSE^} (@command{gnatlink})
9101 Causes additional information to be output, including a full list of the
9102 included object files. This switch option is most useful when you want
9103 to see what set of object files are being used in the link step.
9105 @item ^-v -v^/VERBOSE/VERBOSE^
9106 @cindex @option{^-v -v^/VERBOSE/VERBOSE^} (@command{gnatlink})
9107 Very verbose mode. Requests that the compiler operate in verbose mode when
9108 it compiles the binder file, and that the system linker run in verbose mode.
9110 @item ^-o ^/EXECUTABLE=^@var{exec-name}
9111 @cindex @option{^-o^/EXECUTABLE^} (@command{gnatlink})
9112 @var{exec-name} specifies an alternate name for the generated
9113 executable program. If this switch is omitted, the executable has the same
9114 name as the main unit. For example, @code{gnatlink try.ali} creates
9115 an executable called @file{^try^TRY.EXE^}.
9118 @item -b @var{target}
9119 @cindex @option{-b} (@command{gnatlink})
9120 Compile your program to run on @var{target}, which is the name of a
9121 system configuration. You must have a GNAT cross-compiler built if
9122 @var{target} is not the same as your host system.
9125 @cindex @option{-B} (@command{gnatlink})
9126 Load compiler executables (for example, @code{gnat1}, the Ada compiler)
9127 from @var{dir} instead of the default location. Only use this switch
9128 when multiple versions of the GNAT compiler are available.
9129 @xref{Directory Options,,, gcc, The GNU Compiler Collection},
9130 for further details. You would normally use the @option{-b} or
9131 @option{-V} switch instead.
9133 @item --GCC=@var{compiler_name}
9134 @cindex @option{--GCC=compiler_name} (@command{gnatlink})
9135 Program used for compiling the binder file. The default is
9136 @command{gcc}. You need to use quotes around @var{compiler_name} if
9137 @code{compiler_name} contains spaces or other separator characters.
9138 As an example @option{--GCC="foo -x -y"} will instruct @command{gnatlink} to
9139 use @code{foo -x -y} as your compiler. Note that switch @option{-c} is always
9140 inserted after your command name. Thus in the above example the compiler
9141 command that will be used by @command{gnatlink} will be @code{foo -c -x -y}.
9142 A limitation of this syntax is that the name and path name of the executable
9143 itself must not include any embedded spaces. If the compiler executable is
9144 different from the default one (gcc or <prefix>-gcc), then the back-end
9145 switches in the ALI file are not used to compile the binder generated source.
9146 For example, this is the case with @option{--GCC="foo -x -y"}. But the back end
9147 switches will be used for @option{--GCC="gcc -gnatv"}. If several
9148 @option{--GCC=compiler_name} are used, only the last @var{compiler_name}
9149 is taken into account. However, all the additional switches are also taken
9151 @option{--GCC="foo -x -y" --GCC="bar -z -t"} is equivalent to
9152 @option{--GCC="bar -x -y -z -t"}.
9154 @item --LINK=@var{name}
9155 @cindex @option{--LINK=} (@command{gnatlink})
9156 @var{name} is the name of the linker to be invoked. This is especially
9157 useful in mixed language programs since languages such as C++ require
9158 their own linker to be used. When this switch is omitted, the default
9159 name for the linker is @command{gcc}. When this switch is used, the
9160 specified linker is called instead of @command{gcc} with exactly the same
9161 parameters that would have been passed to @command{gcc} so if the desired
9162 linker requires different parameters it is necessary to use a wrapper
9163 script that massages the parameters before invoking the real linker. It
9164 may be useful to control the exact invocation by using the verbose
9170 @item /DEBUG=TRACEBACK
9171 @cindex @code{/DEBUG=TRACEBACK} (@command{gnatlink})
9172 This qualifier causes sufficient information to be included in the
9173 executable file to allow a traceback, but does not include the full
9174 symbol information needed by the debugger.
9176 @item /IDENTIFICATION="<string>"
9177 @code{"<string>"} specifies the string to be stored in the image file
9178 identification field in the image header.
9179 It overrides any pragma @code{Ident} specified string.
9181 @item /NOINHIBIT-EXEC
9182 Generate the executable file even if there are linker warnings.
9184 @item /NOSTART_FILES
9185 Don't link in the object file containing the ``main'' transfer address.
9186 Used when linking with a foreign language main program compiled with an
9190 Prefer linking with object libraries over sharable images, even without
9196 @node The GNAT Make Program gnatmake
9197 @chapter The GNAT Make Program @command{gnatmake}
9201 * Running gnatmake::
9202 * Switches for gnatmake::
9203 * Mode Switches for gnatmake::
9204 * Notes on the Command Line::
9205 * How gnatmake Works::
9206 * Examples of gnatmake Usage::
9209 A typical development cycle when working on an Ada program consists of
9210 the following steps:
9214 Edit some sources to fix bugs.
9220 Compile all sources affected.
9230 The third step can be tricky, because not only do the modified files
9231 @cindex Dependency rules
9232 have to be compiled, but any files depending on these files must also be
9233 recompiled. The dependency rules in Ada can be quite complex, especially
9234 in the presence of overloading, @code{use} clauses, generics and inlined
9237 @command{gnatmake} automatically takes care of the third and fourth steps
9238 of this process. It determines which sources need to be compiled,
9239 compiles them, and binds and links the resulting object files.
9241 Unlike some other Ada make programs, the dependencies are always
9242 accurately recomputed from the new sources. The source based approach of
9243 the GNAT compilation model makes this possible. This means that if
9244 changes to the source program cause corresponding changes in
9245 dependencies, they will always be tracked exactly correctly by
9248 @node Running gnatmake
9249 @section Running @command{gnatmake}
9252 The usual form of the @command{gnatmake} command is
9255 @c $ gnatmake @ovar{switches} @var{file_name}
9256 @c @ovar{file_names} @ovar{mode_switches}
9257 @c Expanding @ovar macro inline (explanation in macro def comments)
9258 $ gnatmake @r{[}@var{switches}@r{]} @var{file_name}
9259 @r{[}@var{file_names}@r{]} @r{[}@var{mode_switches}@r{]}
9263 The only required argument is one @var{file_name}, which specifies
9264 a compilation unit that is a main program. Several @var{file_names} can be
9265 specified: this will result in several executables being built.
9266 If @code{switches} are present, they can be placed before the first
9267 @var{file_name}, between @var{file_names} or after the last @var{file_name}.
9268 If @var{mode_switches} are present, they must always be placed after
9269 the last @var{file_name} and all @code{switches}.
9271 If you are using standard file extensions (@file{.adb} and @file{.ads}), then the
9272 extension may be omitted from the @var{file_name} arguments. However, if
9273 you are using non-standard extensions, then it is required that the
9274 extension be given. A relative or absolute directory path can be
9275 specified in a @var{file_name}, in which case, the input source file will
9276 be searched for in the specified directory only. Otherwise, the input
9277 source file will first be searched in the directory where
9278 @command{gnatmake} was invoked and if it is not found, it will be search on
9279 the source path of the compiler as described in
9280 @ref{Search Paths and the Run-Time Library (RTL)}.
9282 All @command{gnatmake} output (except when you specify
9283 @option{^-M^/DEPENDENCIES_LIST^}) is to
9284 @file{stderr}. The output produced by the
9285 @option{^-M^/DEPENDENCIES_LIST^} switch is send to
9288 @node Switches for gnatmake
9289 @section Switches for @command{gnatmake}
9292 You may specify any of the following switches to @command{gnatmake}:
9298 @cindex @option{--version} @command{gnatmake}
9299 Display Copyright and version, then exit disregarding all other options.
9302 @cindex @option{--help} @command{gnatmake}
9303 If @option{--version} was not used, display usage, then exit disregarding
9307 @item --GCC=@var{compiler_name}
9308 @cindex @option{--GCC=compiler_name} (@command{gnatmake})
9309 Program used for compiling. The default is `@command{gcc}'. You need to use
9310 quotes around @var{compiler_name} if @code{compiler_name} contains
9311 spaces or other separator characters. As an example @option{--GCC="foo -x
9312 -y"} will instruct @command{gnatmake} to use @code{foo -x -y} as your
9313 compiler. A limitation of this syntax is that the name and path name of
9314 the executable itself must not include any embedded spaces. Note that
9315 switch @option{-c} is always inserted after your command name. Thus in the
9316 above example the compiler command that will be used by @command{gnatmake}
9317 will be @code{foo -c -x -y}. If several @option{--GCC=compiler_name} are
9318 used, only the last @var{compiler_name} is taken into account. However,
9319 all the additional switches are also taken into account. Thus,
9320 @option{--GCC="foo -x -y" --GCC="bar -z -t"} is equivalent to
9321 @option{--GCC="bar -x -y -z -t"}.
9323 @item --GNATBIND=@var{binder_name}
9324 @cindex @option{--GNATBIND=binder_name} (@command{gnatmake})
9325 Program used for binding. The default is `@code{gnatbind}'. You need to
9326 use quotes around @var{binder_name} if @var{binder_name} contains spaces
9327 or other separator characters. As an example @option{--GNATBIND="bar -x
9328 -y"} will instruct @command{gnatmake} to use @code{bar -x -y} as your
9329 binder. Binder switches that are normally appended by @command{gnatmake}
9330 to `@code{gnatbind}' are now appended to the end of @code{bar -x -y}.
9331 A limitation of this syntax is that the name and path name of the executable
9332 itself must not include any embedded spaces.
9334 @item --GNATLINK=@var{linker_name}
9335 @cindex @option{--GNATLINK=linker_name} (@command{gnatmake})
9336 Program used for linking. The default is `@command{gnatlink}'. You need to
9337 use quotes around @var{linker_name} if @var{linker_name} contains spaces
9338 or other separator characters. As an example @option{--GNATLINK="lan -x
9339 -y"} will instruct @command{gnatmake} to use @code{lan -x -y} as your
9340 linker. Linker switches that are normally appended by @command{gnatmake} to
9341 `@command{gnatlink}' are now appended to the end of @code{lan -x -y}.
9342 A limitation of this syntax is that the name and path name of the executable
9343 itself must not include any embedded spaces.
9347 @item ^--subdirs^/SUBDIRS^=subdir
9348 Actual object directory of each project file is the subdirectory subdir of the
9349 object directory specified or defaulted in the project file.
9351 @item ^--single-compile-per-obj-dir^/SINGLE_COMPILE_PER_OBJ_DIR^
9352 Disallow simultaneous compilations in the same object directory when
9353 project files are used.
9355 @item ^--unchecked-shared-lib-imports^/UNCHECKED_SHARED_LIB_IMPORTS^
9356 By default, shared library projects are not allowed to import static library
9357 projects. When this switch is used on the command line, this restriction is
9360 @item ^--source-info=<source info file>^/SRC_INFO=source-info-file^
9361 Specify a source info file. This switch is active only when project files
9362 are used. If the source info file is specified as a relative path, then it is
9363 relative to the object directory of the main project. If the source info file
9364 does not exist, then after the Project Manager has successfully parsed and
9365 processed the project files and found the sources, it creates the source info
9366 file. If the source info file already exists and can be read successfully,
9367 then the Project Manager will get all the needed information about the sources
9368 from the source info file and will not look for them. This reduces the time
9369 to process the project files, especially when looking for sources that take a
9370 long time. If the source info file exists but cannot be parsed successfully,
9371 the Project Manager will attempt to recreate it. If the Project Manager fails
9372 to create the source info file, a message is issued, but gnatmake does not
9376 @item --create-map-file
9377 When linking an executable, create a map file. The name of the map file
9378 has the same name as the executable with extension ".map".
9380 @item --create-map-file=mapfile
9381 When linking an executable, create a map file. The name of the map file is
9386 @item ^-a^/ALL_FILES^
9387 @cindex @option{^-a^/ALL_FILES^} (@command{gnatmake})
9388 Consider all files in the make process, even the GNAT internal system
9389 files (for example, the predefined Ada library files), as well as any
9390 locked files. Locked files are files whose ALI file is write-protected.
9392 @command{gnatmake} does not check these files,
9393 because the assumption is that the GNAT internal files are properly up
9394 to date, and also that any write protected ALI files have been properly
9395 installed. Note that if there is an installation problem, such that one
9396 of these files is not up to date, it will be properly caught by the
9398 You may have to specify this switch if you are working on GNAT
9399 itself. The switch @option{^-a^/ALL_FILES^} is also useful
9400 in conjunction with @option{^-f^/FORCE_COMPILE^}
9401 if you need to recompile an entire application,
9402 including run-time files, using special configuration pragmas,
9403 such as a @code{Normalize_Scalars} pragma.
9406 @code{gnatmake ^-a^/ALL_FILES^} compiles all GNAT
9409 @code{gcc -c -gnatpg} rather than @code{gcc -c}.
9412 the @code{/CHECKS=SUPPRESS_ALL /STYLE_CHECKS=GNAT} switch.
9415 @item ^-b^/ACTIONS=BIND^
9416 @cindex @option{^-b^/ACTIONS=BIND^} (@command{gnatmake})
9417 Bind only. Can be combined with @option{^-c^/ACTIONS=COMPILE^} to do
9418 compilation and binding, but no link.
9419 Can be combined with @option{^-l^/ACTIONS=LINK^}
9420 to do binding and linking. When not combined with
9421 @option{^-c^/ACTIONS=COMPILE^}
9422 all the units in the closure of the main program must have been previously
9423 compiled and must be up to date. The root unit specified by @var{file_name}
9424 may be given without extension, with the source extension or, if no GNAT
9425 Project File is specified, with the ALI file extension.
9427 @item ^-c^/ACTIONS=COMPILE^
9428 @cindex @option{^-c^/ACTIONS=COMPILE^} (@command{gnatmake})
9429 Compile only. Do not perform binding, except when @option{^-b^/ACTIONS=BIND^}
9430 is also specified. Do not perform linking, except if both
9431 @option{^-b^/ACTIONS=BIND^} and
9432 @option{^-l^/ACTIONS=LINK^} are also specified.
9433 If the root unit specified by @var{file_name} is not a main unit, this is the
9434 default. Otherwise @command{gnatmake} will attempt binding and linking
9435 unless all objects are up to date and the executable is more recent than
9439 @cindex @option{^-C^/MAPPING^} (@command{gnatmake})
9440 Use a temporary mapping file. A mapping file is a way to communicate
9441 to the compiler two mappings: from unit names to file names (without
9442 any directory information) and from file names to path names (with
9443 full directory information). A mapping file can make the compiler's
9444 file searches faster, especially if there are many source directories,
9445 or the sources are read over a slow network connection. If
9446 @option{^-P^/PROJECT_FILE^} is used, a mapping file is always used, so
9447 @option{^-C^/MAPPING^} is unnecessary; in this case the mapping file
9448 is initially populated based on the project file. If
9449 @option{^-C^/MAPPING^} is used without
9450 @option{^-P^/PROJECT_FILE^},
9451 the mapping file is initially empty. Each invocation of the compiler
9452 will add any newly accessed sources to the mapping file.
9454 @item ^-C=^/USE_MAPPING_FILE=^@var{file}
9455 @cindex @option{^-C=^/USE_MAPPING^} (@command{gnatmake})
9456 Use a specific mapping file. The file, specified as a path name (absolute or
9457 relative) by this switch, should already exist, otherwise the switch is
9458 ineffective. The specified mapping file will be communicated to the compiler.
9459 This switch is not compatible with a project file
9460 (^-P^/PROJECT_FILE=^@var{file}) or with multiple compiling processes
9461 (^-j^/PROCESSES=^nnn, when nnn is greater than 1).
9463 @item ^-d^/DISPLAY_PROGRESS^
9464 @cindex @option{^-d^/DISPLAY_PROGRESS^} (@command{gnatmake})
9465 Display progress for each source, up to date or not, as a single line
9468 completed x out of y (zz%)
9471 If the file needs to be compiled this is displayed after the invocation of
9472 the compiler. These lines are displayed even in quiet output mode.
9474 @item ^-D ^/DIRECTORY_OBJECTS=^@var{dir}
9475 @cindex @option{^-D^/DIRECTORY_OBJECTS^} (@command{gnatmake})
9476 Put all object files and ALI file in directory @var{dir}.
9477 If the @option{^-D^/DIRECTORY_OBJECTS^} switch is not used, all object files
9478 and ALI files go in the current working directory.
9480 This switch cannot be used when using a project file.
9484 @cindex @option{-eL} (@command{gnatmake})
9485 @cindex symbolic links
9486 Follow all symbolic links when processing project files.
9487 This should be used if your project uses symbolic links for files or
9488 directories, but is not needed in other cases.
9490 @cindex naming scheme
9491 This also assumes that no directory matches the naming scheme for files (for
9492 instance that you do not have a directory called "sources.ads" when using the
9493 default GNAT naming scheme).
9495 When you do not have to use this switch (ie by default), gnatmake is able to
9496 save a lot of system calls (several per source file and object file), which
9497 can result in a significant speed up to load and manipulate a project file,
9498 especially when using source files from a remote system.
9502 @item ^-eS^/STANDARD_OUTPUT_FOR_COMMANDS^
9503 @cindex @option{^-eS^/STANDARD_OUTPUT_FOR_COMMANDS^} (@command{gnatmake})
9504 Output the commands for the compiler, the binder and the linker
9505 on ^standard output^SYS$OUTPUT^,
9506 instead of ^standard error^SYS$ERROR^.
9508 @item ^-f^/FORCE_COMPILE^
9509 @cindex @option{^-f^/FORCE_COMPILE^} (@command{gnatmake})
9510 Force recompilations. Recompile all sources, even though some object
9511 files may be up to date, but don't recompile predefined or GNAT internal
9512 files or locked files (files with a write-protected ALI file),
9513 unless the @option{^-a^/ALL_FILES^} switch is also specified.
9515 @item ^-F^/FULL_PATH_IN_BRIEF_MESSAGES^
9516 @cindex @option{^-F^/FULL_PATH_IN_BRIEF_MESSAGES^} (@command{gnatmake})
9517 When using project files, if some errors or warnings are detected during
9518 parsing and verbose mode is not in effect (no use of switch
9519 ^-v^/VERBOSE^), then error lines start with the full path name of the project
9520 file, rather than its simple file name.
9523 @cindex @option{^-g^/DEBUG^} (@command{gnatmake})
9524 Enable debugging. This switch is simply passed to the compiler and to the
9527 @item ^-i^/IN_PLACE^
9528 @cindex @option{^-i^/IN_PLACE^} (@command{gnatmake})
9529 In normal mode, @command{gnatmake} compiles all object files and ALI files
9530 into the current directory. If the @option{^-i^/IN_PLACE^} switch is used,
9531 then instead object files and ALI files that already exist are overwritten
9532 in place. This means that once a large project is organized into separate
9533 directories in the desired manner, then @command{gnatmake} will automatically
9534 maintain and update this organization. If no ALI files are found on the
9535 Ada object path (@ref{Search Paths and the Run-Time Library (RTL)}),
9536 the new object and ALI files are created in the
9537 directory containing the source being compiled. If another organization
9538 is desired, where objects and sources are kept in different directories,
9539 a useful technique is to create dummy ALI files in the desired directories.
9540 When detecting such a dummy file, @command{gnatmake} will be forced to
9541 recompile the corresponding source file, and it will be put the resulting
9542 object and ALI files in the directory where it found the dummy file.
9544 @item ^-j^/PROCESSES=^@var{n}
9545 @cindex @option{^-j^/PROCESSES^} (@command{gnatmake})
9546 @cindex Parallel make
9547 Use @var{n} processes to carry out the (re)compilations. On a
9548 multiprocessor machine compilations will occur in parallel. In the
9549 event of compilation errors, messages from various compilations might
9550 get interspersed (but @command{gnatmake} will give you the full ordered
9551 list of failing compiles at the end). If this is problematic, rerun
9552 the make process with n set to 1 to get a clean list of messages.
9554 @item ^-k^/CONTINUE_ON_ERROR^
9555 @cindex @option{^-k^/CONTINUE_ON_ERROR^} (@command{gnatmake})
9556 Keep going. Continue as much as possible after a compilation error. To
9557 ease the programmer's task in case of compilation errors, the list of
9558 sources for which the compile fails is given when @command{gnatmake}
9561 If @command{gnatmake} is invoked with several @file{file_names} and with this
9562 switch, if there are compilation errors when building an executable,
9563 @command{gnatmake} will not attempt to build the following executables.
9565 @item ^-l^/ACTIONS=LINK^
9566 @cindex @option{^-l^/ACTIONS=LINK^} (@command{gnatmake})
9567 Link only. Can be combined with @option{^-b^/ACTIONS=BIND^} to binding
9568 and linking. Linking will not be performed if combined with
9569 @option{^-c^/ACTIONS=COMPILE^}
9570 but not with @option{^-b^/ACTIONS=BIND^}.
9571 When not combined with @option{^-b^/ACTIONS=BIND^}
9572 all the units in the closure of the main program must have been previously
9573 compiled and must be up to date, and the main program needs to have been bound.
9574 The root unit specified by @var{file_name}
9575 may be given without extension, with the source extension or, if no GNAT
9576 Project File is specified, with the ALI file extension.
9578 @item ^-m^/MINIMAL_RECOMPILATION^
9579 @cindex @option{^-m^/MINIMAL_RECOMPILATION^} (@command{gnatmake})
9580 Specify that the minimum necessary amount of recompilations
9581 be performed. In this mode @command{gnatmake} ignores time
9582 stamp differences when the only
9583 modifications to a source file consist in adding/removing comments,
9584 empty lines, spaces or tabs. This means that if you have changed the
9585 comments in a source file or have simply reformatted it, using this
9586 switch will tell @command{gnatmake} not to recompile files that depend on it
9587 (provided other sources on which these files depend have undergone no
9588 semantic modifications). Note that the debugging information may be
9589 out of date with respect to the sources if the @option{-m} switch causes
9590 a compilation to be switched, so the use of this switch represents a
9591 trade-off between compilation time and accurate debugging information.
9593 @item ^-M^/DEPENDENCIES_LIST^
9594 @cindex Dependencies, producing list
9595 @cindex @option{^-M^/DEPENDENCIES_LIST^} (@command{gnatmake})
9596 Check if all objects are up to date. If they are, output the object
9597 dependences to @file{stdout} in a form that can be directly exploited in
9598 a @file{Makefile}. By default, each source file is prefixed with its
9599 (relative or absolute) directory name. This name is whatever you
9600 specified in the various @option{^-aI^/SOURCE_SEARCH^}
9601 and @option{^-I^/SEARCH^} switches. If you use
9602 @code{gnatmake ^-M^/DEPENDENCIES_LIST^}
9603 @option{^-q^/QUIET^}
9604 (see below), only the source file names,
9605 without relative paths, are output. If you just specify the
9606 @option{^-M^/DEPENDENCIES_LIST^}
9607 switch, dependencies of the GNAT internal system files are omitted. This
9608 is typically what you want. If you also specify
9609 the @option{^-a^/ALL_FILES^} switch,
9610 dependencies of the GNAT internal files are also listed. Note that
9611 dependencies of the objects in external Ada libraries (see switch
9612 @option{^-aL^/SKIP_MISSING=^}@var{dir} in the following list)
9615 @item ^-n^/DO_OBJECT_CHECK^
9616 @cindex @option{^-n^/DO_OBJECT_CHECK^} (@command{gnatmake})
9617 Don't compile, bind, or link. Checks if all objects are up to date.
9618 If they are not, the full name of the first file that needs to be
9619 recompiled is printed.
9620 Repeated use of this option, followed by compiling the indicated source
9621 file, will eventually result in recompiling all required units.
9623 @item ^-o ^/EXECUTABLE=^@var{exec_name}
9624 @cindex @option{^-o^/EXECUTABLE^} (@command{gnatmake})
9625 Output executable name. The name of the final executable program will be
9626 @var{exec_name}. If the @option{^-o^/EXECUTABLE^} switch is omitted the default
9627 name for the executable will be the name of the input file in appropriate form
9628 for an executable file on the host system.
9630 This switch cannot be used when invoking @command{gnatmake} with several
9633 @item ^-p or --create-missing-dirs^/CREATE_MISSING_DIRS^
9634 @cindex @option{^-p^/CREATE_MISSING_DIRS^} (@command{gnatmake})
9635 When using project files (^-P^/PROJECT_FILE=^@var{project}), create
9636 automatically missing object directories, library directories and exec
9639 @item ^-P^/PROJECT_FILE=^@var{project}
9640 @cindex @option{^-P^/PROJECT_FILE^} (@command{gnatmake})
9641 Use project file @var{project}. Only one such switch can be used.
9642 @xref{gnatmake and Project Files}.
9645 @cindex @option{^-q^/QUIET^} (@command{gnatmake})
9646 Quiet. When this flag is not set, the commands carried out by
9647 @command{gnatmake} are displayed.
9649 @item ^-s^/SWITCH_CHECK/^
9650 @cindex @option{^-s^/SWITCH_CHECK^} (@command{gnatmake})
9651 Recompile if compiler switches have changed since last compilation.
9652 All compiler switches but -I and -o are taken into account in the
9654 orders between different ``first letter'' switches are ignored, but
9655 orders between same switches are taken into account. For example,
9656 @option{-O -O2} is different than @option{-O2 -O}, but @option{-g -O}
9657 is equivalent to @option{-O -g}.
9659 This switch is recommended when Integrated Preprocessing is used.
9662 @cindex @option{^-u^/UNIQUE^} (@command{gnatmake})
9663 Unique. Recompile at most the main files. It implies -c. Combined with
9664 -f, it is equivalent to calling the compiler directly. Note that using
9665 ^-u^/UNIQUE^ with a project file and no main has a special meaning
9666 (@pxref{Project Files and Main Subprograms}).
9668 @item ^-U^/ALL_PROJECTS^
9669 @cindex @option{^-U^/ALL_PROJECTS^} (@command{gnatmake})
9670 When used without a project file or with one or several mains on the command
9671 line, is equivalent to ^-u^/UNIQUE^. When used with a project file and no main
9672 on the command line, all sources of all project files are checked and compiled
9673 if not up to date, and libraries are rebuilt, if necessary.
9676 @cindex @option{^-v^/REASONS^} (@command{gnatmake})
9677 Verbose. Display the reason for all recompilations @command{gnatmake}
9678 decides are necessary, with the highest verbosity level.
9680 @item ^-vl^/LOW_VERBOSITY^
9681 @cindex @option{^-vl^/LOW_VERBOSITY^} (@command{gnatmake})
9682 Verbosity level Low. Display fewer lines than in verbosity Medium.
9684 @item ^-vm^/MEDIUM_VERBOSITY^
9685 @cindex @option{^-vm^/MEDIUM_VERBOSITY^} (@command{gnatmake})
9686 Verbosity level Medium. Potentially display fewer lines than in verbosity High.
9688 @item ^-vh^/HIGH_VERBOSITY^
9689 @cindex @option{^-vm^/HIGH_VERBOSITY^} (@command{gnatmake})
9690 Verbosity level High. Equivalent to ^-v^/REASONS^.
9692 @item ^-vP^/MESSAGES_PROJECT_FILE=^@emph{x}
9693 Indicate the verbosity of the parsing of GNAT project files.
9694 @xref{Switches Related to Project Files}.
9696 @item ^-x^/NON_PROJECT_UNIT_COMPILATION^
9697 @cindex @option{^-x^/NON_PROJECT_UNIT_COMPILATION^} (@command{gnatmake})
9698 Indicate that sources that are not part of any Project File may be compiled.
9699 Normally, when using Project Files, only sources that are part of a Project
9700 File may be compile. When this switch is used, a source outside of all Project
9701 Files may be compiled. The ALI file and the object file will be put in the
9702 object directory of the main Project. The compilation switches used will only
9703 be those specified on the command line. Even when
9704 @option{^-x^/NON_PROJECT_UNIT_COMPILATION^} is used, mains specified on the
9705 command line need to be sources of a project file.
9707 @item ^-X^/EXTERNAL_REFERENCE=^@var{name=value}
9708 Indicate that external variable @var{name} has the value @var{value}.
9709 The Project Manager will use this value for occurrences of
9710 @code{external(name)} when parsing the project file.
9711 @xref{Switches Related to Project Files}.
9714 @cindex @option{^-z^/NOMAIN^} (@command{gnatmake})
9715 No main subprogram. Bind and link the program even if the unit name
9716 given on the command line is a package name. The resulting executable
9717 will execute the elaboration routines of the package and its closure,
9718 then the finalization routines.
9723 @item @command{gcc} @asis{switches}
9725 Any uppercase or multi-character switch that is not a @command{gnatmake} switch
9726 is passed to @command{gcc} (e.g.@: @option{-O}, @option{-gnato,} etc.)
9729 Any qualifier that cannot be recognized as a qualifier for @code{GNAT MAKE}
9730 but is recognizable as a valid qualifier for @code{GNAT COMPILE} is
9731 automatically treated as a compiler switch, and passed on to all
9732 compilations that are carried out.
9737 Source and library search path switches:
9741 @item ^-aI^/SOURCE_SEARCH=^@var{dir}
9742 @cindex @option{^-aI^/SOURCE_SEARCH^} (@command{gnatmake})
9743 When looking for source files also look in directory @var{dir}.
9744 The order in which source files search is undertaken is
9745 described in @ref{Search Paths and the Run-Time Library (RTL)}.
9747 @item ^-aL^/SKIP_MISSING=^@var{dir}
9748 @cindex @option{^-aL^/SKIP_MISSING^} (@command{gnatmake})
9749 Consider @var{dir} as being an externally provided Ada library.
9750 Instructs @command{gnatmake} to skip compilation units whose @file{.ALI}
9751 files have been located in directory @var{dir}. This allows you to have
9752 missing bodies for the units in @var{dir} and to ignore out of date bodies
9753 for the same units. You still need to specify
9754 the location of the specs for these units by using the switches
9755 @option{^-aI^/SOURCE_SEARCH=^@var{dir}}
9756 or @option{^-I^/SEARCH=^@var{dir}}.
9757 Note: this switch is provided for compatibility with previous versions
9758 of @command{gnatmake}. The easier method of causing standard libraries
9759 to be excluded from consideration is to write-protect the corresponding
9762 @item ^-aO^/OBJECT_SEARCH=^@var{dir}
9763 @cindex @option{^-aO^/OBJECT_SEARCH^} (@command{gnatmake})
9764 When searching for library and object files, look in directory
9765 @var{dir}. The order in which library files are searched is described in
9766 @ref{Search Paths for gnatbind}.
9768 @item ^-A^/CONDITIONAL_SOURCE_SEARCH=^@var{dir}
9769 @cindex Search paths, for @command{gnatmake}
9770 @cindex @option{^-A^/CONDITIONAL_SOURCE_SEARCH^} (@command{gnatmake})
9771 Equivalent to @option{^-aL^/SKIP_MISSING=^@var{dir}
9772 ^-aI^/SOURCE_SEARCH=^@var{dir}}.
9774 @item ^-I^/SEARCH=^@var{dir}
9775 @cindex @option{^-I^/SEARCH^} (@command{gnatmake})
9776 Equivalent to @option{^-aO^/OBJECT_SEARCH=^@var{dir}
9777 ^-aI^/SOURCE_SEARCH=^@var{dir}}.
9779 @item ^-I-^/NOCURRENT_DIRECTORY^
9780 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@command{gnatmake})
9781 @cindex Source files, suppressing search
9782 Do not look for source files in the directory containing the source
9783 file named in the command line.
9784 Do not look for ALI or object files in the directory
9785 where @command{gnatmake} was invoked.
9787 @item ^-L^/LIBRARY_SEARCH=^@var{dir}
9788 @cindex @option{^-L^/LIBRARY_SEARCH^} (@command{gnatmake})
9789 @cindex Linker libraries
9790 Add directory @var{dir} to the list of directories in which the linker
9791 will search for libraries. This is equivalent to
9792 @option{-largs ^-L^/LIBRARY_SEARCH=^}@var{dir}.
9794 Furthermore, under Windows, the sources pointed to by the libraries path
9795 set in the registry are not searched for.
9799 @cindex @option{-nostdinc} (@command{gnatmake})
9800 Do not look for source files in the system default directory.
9803 @cindex @option{-nostdlib} (@command{gnatmake})
9804 Do not look for library files in the system default directory.
9806 @item --RTS=@var{rts-path}
9807 @cindex @option{--RTS} (@command{gnatmake})
9808 Specifies the default location of the runtime library. GNAT looks for the
9810 in the following directories, and stops as soon as a valid runtime is found
9811 (@file{adainclude} or @file{ada_source_path}, and @file{adalib} or
9812 @file{ada_object_path} present):
9815 @item <current directory>/$rts_path
9817 @item <default-search-dir>/$rts_path
9819 @item <default-search-dir>/rts-$rts_path
9823 The selected path is handled like a normal RTS path.
9827 @node Mode Switches for gnatmake
9828 @section Mode Switches for @command{gnatmake}
9831 The mode switches (referred to as @code{mode_switches}) allow the
9832 inclusion of switches that are to be passed to the compiler itself, the
9833 binder or the linker. The effect of a mode switch is to cause all
9834 subsequent switches up to the end of the switch list, or up to the next
9835 mode switch, to be interpreted as switches to be passed on to the
9836 designated component of GNAT.
9840 @item -cargs @var{switches}
9841 @cindex @option{-cargs} (@command{gnatmake})
9842 Compiler switches. Here @var{switches} is a list of switches
9843 that are valid switches for @command{gcc}. They will be passed on to
9844 all compile steps performed by @command{gnatmake}.
9846 @item -bargs @var{switches}
9847 @cindex @option{-bargs} (@command{gnatmake})
9848 Binder switches. Here @var{switches} is a list of switches
9849 that are valid switches for @code{gnatbind}. They will be passed on to
9850 all bind steps performed by @command{gnatmake}.
9852 @item -largs @var{switches}
9853 @cindex @option{-largs} (@command{gnatmake})
9854 Linker switches. Here @var{switches} is a list of switches
9855 that are valid switches for @command{gnatlink}. They will be passed on to
9856 all link steps performed by @command{gnatmake}.
9858 @item -margs @var{switches}
9859 @cindex @option{-margs} (@command{gnatmake})
9860 Make switches. The switches are directly interpreted by @command{gnatmake},
9861 regardless of any previous occurrence of @option{-cargs}, @option{-bargs}
9865 @node Notes on the Command Line
9866 @section Notes on the Command Line
9869 This section contains some additional useful notes on the operation
9870 of the @command{gnatmake} command.
9874 @cindex Recompilation, by @command{gnatmake}
9875 If @command{gnatmake} finds no ALI files, it recompiles the main program
9876 and all other units required by the main program.
9877 This means that @command{gnatmake}
9878 can be used for the initial compile, as well as during subsequent steps of
9879 the development cycle.
9882 If you enter @code{gnatmake @var{file}.adb}, where @file{@var{file}.adb}
9883 is a subunit or body of a generic unit, @command{gnatmake} recompiles
9884 @file{@var{file}.adb} (because it finds no ALI) and stops, issuing a
9888 In @command{gnatmake} the switch @option{^-I^/SEARCH^}
9889 is used to specify both source and
9890 library file paths. Use @option{^-aI^/SOURCE_SEARCH^}
9891 instead if you just want to specify
9892 source paths only and @option{^-aO^/OBJECT_SEARCH^}
9893 if you want to specify library paths
9897 @command{gnatmake} will ignore any files whose ALI file is write-protected.
9898 This may conveniently be used to exclude standard libraries from
9899 consideration and in particular it means that the use of the
9900 @option{^-f^/FORCE_COMPILE^} switch will not recompile these files
9901 unless @option{^-a^/ALL_FILES^} is also specified.
9904 @command{gnatmake} has been designed to make the use of Ada libraries
9905 particularly convenient. Assume you have an Ada library organized
9906 as follows: @i{^obj-dir^[OBJ_DIR]^} contains the objects and ALI files for
9907 of your Ada compilation units,
9908 whereas @i{^include-dir^[INCLUDE_DIR]^} contains the
9909 specs of these units, but no bodies. Then to compile a unit
9910 stored in @code{main.adb}, which uses this Ada library you would just type
9914 $ gnatmake -aI@var{include-dir} -aL@var{obj-dir} main
9917 $ gnatmake /SOURCE_SEARCH=@i{[INCLUDE_DIR]}
9918 /SKIP_MISSING=@i{[OBJ_DIR]} main
9923 Using @command{gnatmake} along with the
9924 @option{^-m (minimal recompilation)^/MINIMAL_RECOMPILATION^}
9925 switch provides a mechanism for avoiding unnecessary recompilations. Using
9927 you can update the comments/format of your
9928 source files without having to recompile everything. Note, however, that
9929 adding or deleting lines in a source files may render its debugging
9930 info obsolete. If the file in question is a spec, the impact is rather
9931 limited, as that debugging info will only be useful during the
9932 elaboration phase of your program. For bodies the impact can be more
9933 significant. In all events, your debugger will warn you if a source file
9934 is more recent than the corresponding object, and alert you to the fact
9935 that the debugging information may be out of date.
9938 @node How gnatmake Works
9939 @section How @command{gnatmake} Works
9942 Generally @command{gnatmake} automatically performs all necessary
9943 recompilations and you don't need to worry about how it works. However,
9944 it may be useful to have some basic understanding of the @command{gnatmake}
9945 approach and in particular to understand how it uses the results of
9946 previous compilations without incorrectly depending on them.
9948 First a definition: an object file is considered @dfn{up to date} if the
9949 corresponding ALI file exists and if all the source files listed in the
9950 dependency section of this ALI file have time stamps matching those in
9951 the ALI file. This means that neither the source file itself nor any
9952 files that it depends on have been modified, and hence there is no need
9953 to recompile this file.
9955 @command{gnatmake} works by first checking if the specified main unit is up
9956 to date. If so, no compilations are required for the main unit. If not,
9957 @command{gnatmake} compiles the main program to build a new ALI file that
9958 reflects the latest sources. Then the ALI file of the main unit is
9959 examined to find all the source files on which the main program depends,
9960 and @command{gnatmake} recursively applies the above procedure on all these
9963 This process ensures that @command{gnatmake} only trusts the dependencies
9964 in an existing ALI file if they are known to be correct. Otherwise it
9965 always recompiles to determine a new, guaranteed accurate set of
9966 dependencies. As a result the program is compiled ``upside down'' from what may
9967 be more familiar as the required order of compilation in some other Ada
9968 systems. In particular, clients are compiled before the units on which
9969 they depend. The ability of GNAT to compile in any order is critical in
9970 allowing an order of compilation to be chosen that guarantees that
9971 @command{gnatmake} will recompute a correct set of new dependencies if
9974 When invoking @command{gnatmake} with several @var{file_names}, if a unit is
9975 imported by several of the executables, it will be recompiled at most once.
9977 Note: when using non-standard naming conventions
9978 (@pxref{Using Other File Names}), changing through a configuration pragmas
9979 file the version of a source and invoking @command{gnatmake} to recompile may
9980 have no effect, if the previous version of the source is still accessible
9981 by @command{gnatmake}. It may be necessary to use the switch
9982 ^-f^/FORCE_COMPILE^.
9984 @node Examples of gnatmake Usage
9985 @section Examples of @command{gnatmake} Usage
9988 @item gnatmake hello.adb
9989 Compile all files necessary to bind and link the main program
9990 @file{hello.adb} (containing unit @code{Hello}) and bind and link the
9991 resulting object files to generate an executable file @file{^hello^HELLO.EXE^}.
9993 @item gnatmake main1 main2 main3
9994 Compile all files necessary to bind and link the main programs
9995 @file{main1.adb} (containing unit @code{Main1}), @file{main2.adb}
9996 (containing unit @code{Main2}) and @file{main3.adb}
9997 (containing unit @code{Main3}) and bind and link the resulting object files
9998 to generate three executable files @file{^main1^MAIN1.EXE^},
9999 @file{^main2^MAIN2.EXE^}
10000 and @file{^main3^MAIN3.EXE^}.
10003 @item gnatmake -q Main_Unit -cargs -O2 -bargs -l
10007 @item gnatmake Main_Unit /QUIET
10008 /COMPILER_QUALIFIERS /OPTIMIZE=ALL
10009 /BINDER_QUALIFIERS /ORDER_OF_ELABORATION
10011 Compile all files necessary to bind and link the main program unit
10012 @code{Main_Unit} (from file @file{main_unit.adb}). All compilations will
10013 be done with optimization level 2 and the order of elaboration will be
10014 listed by the binder. @command{gnatmake} will operate in quiet mode, not
10015 displaying commands it is executing.
10018 @c *************************
10019 @node Improving Performance
10020 @chapter Improving Performance
10021 @cindex Improving performance
10024 This chapter presents several topics related to program performance.
10025 It first describes some of the tradeoffs that need to be considered
10026 and some of the techniques for making your program run faster.
10027 It then documents the @command{gnatelim} tool and unused subprogram/data
10028 elimination feature, which can reduce the size of program executables.
10030 Note: to invoke @command{gnatelim} with a project file, use the @code{gnat}
10031 driver (see @ref{The GNAT Driver and Project Files}).
10035 * Performance Considerations::
10036 * Text_IO Suggestions::
10037 * Reducing Size of Ada Executables with gnatelim::
10038 * Reducing Size of Executables with unused subprogram/data elimination::
10042 @c *****************************
10043 @node Performance Considerations
10044 @section Performance Considerations
10047 The GNAT system provides a number of options that allow a trade-off
10052 performance of the generated code
10055 speed of compilation
10058 minimization of dependences and recompilation
10061 the degree of run-time checking.
10065 The defaults (if no options are selected) aim at improving the speed
10066 of compilation and minimizing dependences, at the expense of performance
10067 of the generated code:
10074 no inlining of subprogram calls
10077 all run-time checks enabled except overflow and elaboration checks
10081 These options are suitable for most program development purposes. This
10082 chapter describes how you can modify these choices, and also provides
10083 some guidelines on debugging optimized code.
10086 * Controlling Run-Time Checks::
10087 * Use of Restrictions::
10088 * Optimization Levels::
10089 * Debugging Optimized Code::
10090 * Inlining of Subprograms::
10091 * Other Optimization Switches::
10092 * Optimization and Strict Aliasing::
10095 * Coverage Analysis::
10099 @node Controlling Run-Time Checks
10100 @subsection Controlling Run-Time Checks
10103 By default, GNAT generates all run-time checks, except integer overflow
10104 checks, stack overflow checks, and checks for access before elaboration on
10105 subprogram calls. The latter are not required in default mode, because all
10106 necessary checking is done at compile time.
10107 @cindex @option{-gnatp} (@command{gcc})
10108 @cindex @option{-gnato} (@command{gcc})
10109 Two gnat switches, @option{-gnatp} and @option{-gnato} allow this default to
10110 be modified. @xref{Run-Time Checks}.
10112 Our experience is that the default is suitable for most development
10115 We treat integer overflow specially because these
10116 are quite expensive and in our experience are not as important as other
10117 run-time checks in the development process. Note that division by zero
10118 is not considered an overflow check, and divide by zero checks are
10119 generated where required by default.
10121 Elaboration checks are off by default, and also not needed by default, since
10122 GNAT uses a static elaboration analysis approach that avoids the need for
10123 run-time checking. This manual contains a full chapter discussing the issue
10124 of elaboration checks, and if the default is not satisfactory for your use,
10125 you should read this chapter.
10127 For validity checks, the minimal checks required by the Ada Reference
10128 Manual (for case statements and assignments to array elements) are on
10129 by default. These can be suppressed by use of the @option{-gnatVn} switch.
10130 Note that in Ada 83, there were no validity checks, so if the Ada 83 mode
10131 is acceptable (or when comparing GNAT performance with an Ada 83 compiler),
10132 it may be reasonable to routinely use @option{-gnatVn}. Validity checks
10133 are also suppressed entirely if @option{-gnatp} is used.
10135 @cindex Overflow checks
10136 @cindex Checks, overflow
10139 @cindex pragma Suppress
10140 @cindex pragma Unsuppress
10141 Note that the setting of the switches controls the default setting of
10142 the checks. They may be modified using either @code{pragma Suppress} (to
10143 remove checks) or @code{pragma Unsuppress} (to add back suppressed
10144 checks) in the program source.
10146 @node Use of Restrictions
10147 @subsection Use of Restrictions
10150 The use of pragma Restrictions allows you to control which features are
10151 permitted in your program. Apart from the obvious point that if you avoid
10152 relatively expensive features like finalization (enforceable by the use
10153 of pragma Restrictions (No_Finalization), the use of this pragma does not
10154 affect the generated code in most cases.
10156 One notable exception to this rule is that the possibility of task abort
10157 results in some distributed overhead, particularly if finalization or
10158 exception handlers are used. The reason is that certain sections of code
10159 have to be marked as non-abortable.
10161 If you use neither the @code{abort} statement, nor asynchronous transfer
10162 of control (@code{select @dots{} then abort}), then this distributed overhead
10163 is removed, which may have a general positive effect in improving
10164 overall performance. Especially code involving frequent use of tasking
10165 constructs and controlled types will show much improved performance.
10166 The relevant restrictions pragmas are
10168 @smallexample @c ada
10169 pragma Restrictions (No_Abort_Statements);
10170 pragma Restrictions (Max_Asynchronous_Select_Nesting => 0);
10174 It is recommended that these restriction pragmas be used if possible. Note
10175 that this also means that you can write code without worrying about the
10176 possibility of an immediate abort at any point.
10178 @node Optimization Levels
10179 @subsection Optimization Levels
10180 @cindex @option{^-O^/OPTIMIZE^} (@command{gcc})
10183 Without any optimization ^option,^qualifier,^
10184 the compiler's goal is to reduce the cost of
10185 compilation and to make debugging produce the expected results.
10186 Statements are independent: if you stop the program with a breakpoint between
10187 statements, you can then assign a new value to any variable or change
10188 the program counter to any other statement in the subprogram and get exactly
10189 the results you would expect from the source code.
10191 Turning on optimization makes the compiler attempt to improve the
10192 performance and/or code size at the expense of compilation time and
10193 possibly the ability to debug the program.
10195 If you use multiple
10196 ^-O options, with or without level numbers,^/OPTIMIZE qualifiers,^
10197 the last such option is the one that is effective.
10200 The default is optimization off. This results in the fastest compile
10201 times, but GNAT makes absolutely no attempt to optimize, and the
10202 generated programs are considerably larger and slower than when
10203 optimization is enabled. You can use the
10205 @option{-O} switch (the permitted forms are @option{-O0}, @option{-O1}
10206 @option{-O2}, @option{-O3}, and @option{-Os})
10209 @code{OPTIMIZE} qualifier
10211 to @command{gcc} to control the optimization level:
10214 @item ^-O0^/OPTIMIZE=NONE^
10215 No optimization (the default);
10216 generates unoptimized code but has
10217 the fastest compilation time.
10219 Note that many other compilers do fairly extensive optimization
10220 even if ``no optimization'' is specified. With gcc, it is
10221 very unusual to use ^-O0^/OPTIMIZE=NONE^ for production if
10222 execution time is of any concern, since ^-O0^/OPTIMIZE=NONE^
10223 really does mean no optimization at all. This difference between
10224 gcc and other compilers should be kept in mind when doing
10225 performance comparisons.
10227 @item ^-O1^/OPTIMIZE=SOME^
10228 Moderate optimization;
10229 optimizes reasonably well but does not
10230 degrade compilation time significantly.
10232 @item ^-O2^/OPTIMIZE=ALL^
10234 @itemx /OPTIMIZE=DEVELOPMENT
10237 generates highly optimized code and has
10238 the slowest compilation time.
10240 @item ^-O3^/OPTIMIZE=INLINING^
10241 Full optimization as in @option{-O2};
10242 also uses more aggressive automatic inlining of subprograms within a unit
10243 (@pxref{Inlining of Subprograms}) and attemps to vectorize loops.
10245 @item ^-Os^/OPTIMIZE=SPACE^
10246 Optimize space usage (code and data) of resulting program.
10250 Higher optimization levels perform more global transformations on the
10251 program and apply more expensive analysis algorithms in order to generate
10252 faster and more compact code. The price in compilation time, and the
10253 resulting improvement in execution time,
10254 both depend on the particular application and the hardware environment.
10255 You should experiment to find the best level for your application.
10257 Since the precise set of optimizations done at each level will vary from
10258 release to release (and sometime from target to target), it is best to think
10259 of the optimization settings in general terms.
10260 @xref{Optimize Options,, Options That Control Optimization, gcc, Using
10261 the GNU Compiler Collection (GCC)}, for details about
10262 ^the @option{-O} settings and a number of @option{-f} options that^how to^
10263 individually enable or disable specific optimizations.
10265 Unlike some other compilation systems, ^@command{gcc}^GNAT^ has
10266 been tested extensively at all optimization levels. There are some bugs
10267 which appear only with optimization turned on, but there have also been
10268 bugs which show up only in @emph{unoptimized} code. Selecting a lower
10269 level of optimization does not improve the reliability of the code
10270 generator, which in practice is highly reliable at all optimization
10273 Note regarding the use of @option{-O3}: The use of this optimization level
10274 is generally discouraged with GNAT, since it often results in larger
10275 executables which may run more slowly. See further discussion of this point
10276 in @ref{Inlining of Subprograms}.
10278 @node Debugging Optimized Code
10279 @subsection Debugging Optimized Code
10280 @cindex Debugging optimized code
10281 @cindex Optimization and debugging
10284 Although it is possible to do a reasonable amount of debugging at
10286 nonzero optimization levels,
10287 the higher the level the more likely that
10290 @option{/OPTIMIZE} settings other than @code{NONE},
10291 such settings will make it more likely that
10293 source-level constructs will have been eliminated by optimization.
10294 For example, if a loop is strength-reduced, the loop
10295 control variable may be completely eliminated and thus cannot be
10296 displayed in the debugger.
10297 This can only happen at @option{-O2} or @option{-O3}.
10298 Explicit temporary variables that you code might be eliminated at
10299 ^level^setting^ @option{-O1} or higher.
10301 The use of the @option{^-g^/DEBUG^} switch,
10302 @cindex @option{^-g^/DEBUG^} (@command{gcc})
10303 which is needed for source-level debugging,
10304 affects the size of the program executable on disk,
10305 and indeed the debugging information can be quite large.
10306 However, it has no effect on the generated code (and thus does not
10307 degrade performance)
10309 Since the compiler generates debugging tables for a compilation unit before
10310 it performs optimizations, the optimizing transformations may invalidate some
10311 of the debugging data. You therefore need to anticipate certain
10312 anomalous situations that may arise while debugging optimized code.
10313 These are the most common cases:
10317 @i{The ``hopping Program Counter'':} Repeated @code{step} or @code{next}
10319 the PC bouncing back and forth in the code. This may result from any of
10320 the following optimizations:
10324 @i{Common subexpression elimination:} using a single instance of code for a
10325 quantity that the source computes several times. As a result you
10326 may not be able to stop on what looks like a statement.
10329 @i{Invariant code motion:} moving an expression that does not change within a
10330 loop, to the beginning of the loop.
10333 @i{Instruction scheduling:} moving instructions so as to
10334 overlap loads and stores (typically) with other code, or in
10335 general to move computations of values closer to their uses. Often
10336 this causes you to pass an assignment statement without the assignment
10337 happening and then later bounce back to the statement when the
10338 value is actually needed. Placing a breakpoint on a line of code
10339 and then stepping over it may, therefore, not always cause all the
10340 expected side-effects.
10344 @i{The ``big leap'':} More commonly known as @emph{cross-jumping}, in which
10345 two identical pieces of code are merged and the program counter suddenly
10346 jumps to a statement that is not supposed to be executed, simply because
10347 it (and the code following) translates to the same thing as the code
10348 that @emph{was} supposed to be executed. This effect is typically seen in
10349 sequences that end in a jump, such as a @code{goto}, a @code{return}, or
10350 a @code{break} in a C @code{^switch^switch^} statement.
10353 @i{The ``roving variable'':} The symptom is an unexpected value in a variable.
10354 There are various reasons for this effect:
10358 In a subprogram prologue, a parameter may not yet have been moved to its
10362 A variable may be dead, and its register re-used. This is
10363 probably the most common cause.
10366 As mentioned above, the assignment of a value to a variable may
10370 A variable may be eliminated entirely by value propagation or
10371 other means. In this case, GCC may incorrectly generate debugging
10372 information for the variable
10376 In general, when an unexpected value appears for a local variable or parameter
10377 you should first ascertain if that value was actually computed by
10378 your program, as opposed to being incorrectly reported by the debugger.
10380 array elements in an object designated by an access value
10381 are generally less of a problem, once you have ascertained that the access
10383 Typically, this means checking variables in the preceding code and in the
10384 calling subprogram to verify that the value observed is explainable from other
10385 values (one must apply the procedure recursively to those
10386 other values); or re-running the code and stopping a little earlier
10387 (perhaps before the call) and stepping to better see how the variable obtained
10388 the value in question; or continuing to step @emph{from} the point of the
10389 strange value to see if code motion had simply moved the variable's
10394 In light of such anomalies, a recommended technique is to use @option{-O0}
10395 early in the software development cycle, when extensive debugging capabilities
10396 are most needed, and then move to @option{-O1} and later @option{-O2} as
10397 the debugger becomes less critical.
10398 Whether to use the @option{^-g^/DEBUG^} switch in the release version is
10399 a release management issue.
10401 Note that if you use @option{-g} you can then use the @command{strip} program
10402 on the resulting executable,
10403 which removes both debugging information and global symbols.
10406 @node Inlining of Subprograms
10407 @subsection Inlining of Subprograms
10410 A call to a subprogram in the current unit is inlined if all the
10411 following conditions are met:
10415 The optimization level is at least @option{-O1}.
10418 The called subprogram is suitable for inlining: It must be small enough
10419 and not contain something that @command{gcc} cannot support in inlined
10423 @cindex pragma Inline
10425 Any one of the following applies: @code{pragma Inline} is applied to the
10426 subprogram and the @option{^-gnatn^/INLINE^} switch is specified; the
10427 subprogram is local to the unit and called once from within it; the
10428 subprogram is small and optimization level @option{-O2} is specified;
10429 optimization level @option{-O3}) is specified.
10433 Calls to subprograms in @code{with}'ed units are normally not inlined.
10434 To achieve actual inlining (that is, replacement of the call by the code
10435 in the body of the subprogram), the following conditions must all be true.
10439 The optimization level is at least @option{-O1}.
10442 The called subprogram is suitable for inlining: It must be small enough
10443 and not contain something that @command{gcc} cannot support in inlined
10447 The call appears in a body (not in a package spec).
10450 There is a @code{pragma Inline} for the subprogram.
10453 The @option{^-gnatn^/INLINE^} switch is used on the command line.
10456 Even if all these conditions are met, it may not be possible for
10457 the compiler to inline the call, due to the length of the body,
10458 or features in the body that make it impossible for the compiler
10459 to do the inlining.
10461 Note that specifying the @option{-gnatn} switch causes additional
10462 compilation dependencies. Consider the following:
10464 @smallexample @c ada
10484 With the default behavior (no @option{-gnatn} switch specified), the
10485 compilation of the @code{Main} procedure depends only on its own source,
10486 @file{main.adb}, and the spec of the package in file @file{r.ads}. This
10487 means that editing the body of @code{R} does not require recompiling
10490 On the other hand, the call @code{R.Q} is not inlined under these
10491 circumstances. If the @option{-gnatn} switch is present when @code{Main}
10492 is compiled, the call will be inlined if the body of @code{Q} is small
10493 enough, but now @code{Main} depends on the body of @code{R} in
10494 @file{r.adb} as well as on the spec. This means that if this body is edited,
10495 the main program must be recompiled. Note that this extra dependency
10496 occurs whether or not the call is in fact inlined by @command{gcc}.
10498 The use of front end inlining with @option{-gnatN} generates similar
10499 additional dependencies.
10501 @cindex @option{^-fno-inline^/INLINE=SUPPRESS^} (@command{gcc})
10502 Note: The @option{^-fno-inline^/INLINE=SUPPRESS^} switch
10503 can be used to prevent
10504 all inlining. This switch overrides all other conditions and ensures
10505 that no inlining occurs. The extra dependences resulting from
10506 @option{-gnatn} will still be active, even if
10507 this switch is used to suppress the resulting inlining actions.
10509 @cindex @option{-fno-inline-functions} (@command{gcc})
10510 Note: The @option{-fno-inline-functions} switch can be used to prevent
10511 automatic inlining of subprograms if @option{-O3} is used.
10513 @cindex @option{-fno-inline-small-functions} (@command{gcc})
10514 Note: The @option{-fno-inline-small-functions} switch can be used to prevent
10515 automatic inlining of small subprograms if @option{-O2} is used.
10517 @cindex @option{-fno-inline-functions-called-once} (@command{gcc})
10518 Note: The @option{-fno-inline-functions-called-once} switch
10519 can be used to prevent inlining of subprograms local to the unit
10520 and called once from within it if @option{-O1} is used.
10522 Note regarding the use of @option{-O3}: There is no difference in inlining
10523 behavior between @option{-O2} and @option{-O3} for subprograms with an explicit
10524 pragma @code{Inline} assuming the use of @option{-gnatn}
10525 or @option{-gnatN} (the switches that activate inlining). If you have used
10526 pragma @code{Inline} in appropriate cases, then it is usually much better
10527 to use @option{-O2} and @option{-gnatn} and avoid the use of @option{-O3} which
10528 in this case only has the effect of inlining subprograms you did not
10529 think should be inlined. We often find that the use of @option{-O3} slows
10530 down code by performing excessive inlining, leading to increased instruction
10531 cache pressure from the increased code size. So the bottom line here is
10532 that you should not automatically assume that @option{-O3} is better than
10533 @option{-O2}, and indeed you should use @option{-O3} only if tests show that
10534 it actually improves performance.
10536 @node Other Optimization Switches
10537 @subsection Other Optimization Switches
10538 @cindex Optimization Switches
10540 Since @code{GNAT} uses the @command{gcc} back end, all the specialized
10541 @command{gcc} optimization switches are potentially usable. These switches
10542 have not been extensively tested with GNAT but can generally be expected
10543 to work. Examples of switches in this category are
10544 @option{-funroll-loops} and
10545 the various target-specific @option{-m} options (in particular, it has been
10546 observed that @option{-march=pentium4} can significantly improve performance
10547 on appropriate machines). For full details of these switches, see
10548 @ref{Submodel Options,, Hardware Models and Configurations, gcc, Using
10549 the GNU Compiler Collection (GCC)}.
10551 @node Optimization and Strict Aliasing
10552 @subsection Optimization and Strict Aliasing
10554 @cindex Strict Aliasing
10555 @cindex No_Strict_Aliasing
10558 The strong typing capabilities of Ada allow an optimizer to generate
10559 efficient code in situations where other languages would be forced to
10560 make worst case assumptions preventing such optimizations. Consider
10561 the following example:
10563 @smallexample @c ada
10566 type Int1 is new Integer;
10567 type Int2 is new Integer;
10568 type Int1A is access Int1;
10569 type Int2A is access Int2;
10576 for J in Data'Range loop
10577 if Data (J) = Int1V.all then
10578 Int2V.all := Int2V.all + 1;
10587 In this example, since the variable @code{Int1V} can only access objects
10588 of type @code{Int1}, and @code{Int2V} can only access objects of type
10589 @code{Int2}, there is no possibility that the assignment to
10590 @code{Int2V.all} affects the value of @code{Int1V.all}. This means that
10591 the compiler optimizer can "know" that the value @code{Int1V.all} is constant
10592 for all iterations of the loop and avoid the extra memory reference
10593 required to dereference it each time through the loop.
10595 This kind of optimization, called strict aliasing analysis, is
10596 triggered by specifying an optimization level of @option{-O2} or
10597 higher or @option{-Os} and allows @code{GNAT} to generate more efficient code
10598 when access values are involved.
10600 However, although this optimization is always correct in terms of
10601 the formal semantics of the Ada Reference Manual, difficulties can
10602 arise if features like @code{Unchecked_Conversion} are used to break
10603 the typing system. Consider the following complete program example:
10605 @smallexample @c ada
10608 type int1 is new integer;
10609 type int2 is new integer;
10610 type a1 is access int1;
10611 type a2 is access int2;
10616 function to_a2 (Input : a1) return a2;
10619 with Unchecked_Conversion;
10621 function to_a2 (Input : a1) return a2 is
10623 new Unchecked_Conversion (a1, a2);
10625 return to_a2u (Input);
10631 with Text_IO; use Text_IO;
10633 v1 : a1 := new int1;
10634 v2 : a2 := to_a2 (v1);
10638 put_line (int1'image (v1.all));
10644 This program prints out 0 in @option{-O0} or @option{-O1}
10645 mode, but it prints out 1 in @option{-O2} mode. That's
10646 because in strict aliasing mode, the compiler can and
10647 does assume that the assignment to @code{v2.all} could not
10648 affect the value of @code{v1.all}, since different types
10651 This behavior is not a case of non-conformance with the standard, since
10652 the Ada RM specifies that an unchecked conversion where the resulting
10653 bit pattern is not a correct value of the target type can result in an
10654 abnormal value and attempting to reference an abnormal value makes the
10655 execution of a program erroneous. That's the case here since the result
10656 does not point to an object of type @code{int2}. This means that the
10657 effect is entirely unpredictable.
10659 However, although that explanation may satisfy a language
10660 lawyer, in practice an applications programmer expects an
10661 unchecked conversion involving pointers to create true
10662 aliases and the behavior of printing 1 seems plain wrong.
10663 In this case, the strict aliasing optimization is unwelcome.
10665 Indeed the compiler recognizes this possibility, and the
10666 unchecked conversion generates a warning:
10669 p2.adb:5:07: warning: possible aliasing problem with type "a2"
10670 p2.adb:5:07: warning: use -fno-strict-aliasing switch for references
10671 p2.adb:5:07: warning: or use "pragma No_Strict_Aliasing (a2);"
10675 Unfortunately the problem is recognized when compiling the body of
10676 package @code{p2}, but the actual "bad" code is generated while
10677 compiling the body of @code{m} and this latter compilation does not see
10678 the suspicious @code{Unchecked_Conversion}.
10680 As implied by the warning message, there are approaches you can use to
10681 avoid the unwanted strict aliasing optimization in a case like this.
10683 One possibility is to simply avoid the use of @option{-O2}, but
10684 that is a bit drastic, since it throws away a number of useful
10685 optimizations that do not involve strict aliasing assumptions.
10687 A less drastic approach is to compile the program using the
10688 option @option{-fno-strict-aliasing}. Actually it is only the
10689 unit containing the dereferencing of the suspicious pointer
10690 that needs to be compiled. So in this case, if we compile
10691 unit @code{m} with this switch, then we get the expected
10692 value of zero printed. Analyzing which units might need
10693 the switch can be painful, so a more reasonable approach
10694 is to compile the entire program with options @option{-O2}
10695 and @option{-fno-strict-aliasing}. If the performance is
10696 satisfactory with this combination of options, then the
10697 advantage is that the entire issue of possible "wrong"
10698 optimization due to strict aliasing is avoided.
10700 To avoid the use of compiler switches, the configuration
10701 pragma @code{No_Strict_Aliasing} with no parameters may be
10702 used to specify that for all access types, the strict
10703 aliasing optimization should be suppressed.
10705 However, these approaches are still overkill, in that they causes
10706 all manipulations of all access values to be deoptimized. A more
10707 refined approach is to concentrate attention on the specific
10708 access type identified as problematic.
10710 First, if a careful analysis of uses of the pointer shows
10711 that there are no possible problematic references, then
10712 the warning can be suppressed by bracketing the
10713 instantiation of @code{Unchecked_Conversion} to turn
10716 @smallexample @c ada
10717 pragma Warnings (Off);
10719 new Unchecked_Conversion (a1, a2);
10720 pragma Warnings (On);
10724 Of course that approach is not appropriate for this particular
10725 example, since indeed there is a problematic reference. In this
10726 case we can take one of two other approaches.
10728 The first possibility is to move the instantiation of unchecked
10729 conversion to the unit in which the type is declared. In
10730 this example, we would move the instantiation of
10731 @code{Unchecked_Conversion} from the body of package
10732 @code{p2} to the spec of package @code{p1}. Now the
10733 warning disappears. That's because any use of the
10734 access type knows there is a suspicious unchecked
10735 conversion, and the strict aliasing optimization
10736 is automatically suppressed for the type.
10738 If it is not practical to move the unchecked conversion to the same unit
10739 in which the destination access type is declared (perhaps because the
10740 source type is not visible in that unit), you may use pragma
10741 @code{No_Strict_Aliasing} for the type. This pragma must occur in the
10742 same declarative sequence as the declaration of the access type:
10744 @smallexample @c ada
10745 type a2 is access int2;
10746 pragma No_Strict_Aliasing (a2);
10750 Here again, the compiler now knows that the strict aliasing optimization
10751 should be suppressed for any reference to type @code{a2} and the
10752 expected behavior is obtained.
10754 Finally, note that although the compiler can generate warnings for
10755 simple cases of unchecked conversions, there are tricker and more
10756 indirect ways of creating type incorrect aliases which the compiler
10757 cannot detect. Examples are the use of address overlays and unchecked
10758 conversions involving composite types containing access types as
10759 components. In such cases, no warnings are generated, but there can
10760 still be aliasing problems. One safe coding practice is to forbid the
10761 use of address clauses for type overlaying, and to allow unchecked
10762 conversion only for primitive types. This is not really a significant
10763 restriction since any possible desired effect can be achieved by
10764 unchecked conversion of access values.
10766 The aliasing analysis done in strict aliasing mode can certainly
10767 have significant benefits. We have seen cases of large scale
10768 application code where the time is increased by up to 5% by turning
10769 this optimization off. If you have code that includes significant
10770 usage of unchecked conversion, you might want to just stick with
10771 @option{-O1} and avoid the entire issue. If you get adequate
10772 performance at this level of optimization level, that's probably
10773 the safest approach. If tests show that you really need higher
10774 levels of optimization, then you can experiment with @option{-O2}
10775 and @option{-O2 -fno-strict-aliasing} to see how much effect this
10776 has on size and speed of the code. If you really need to use
10777 @option{-O2} with strict aliasing in effect, then you should
10778 review any uses of unchecked conversion of access types,
10779 particularly if you are getting the warnings described above.
10782 @node Coverage Analysis
10783 @subsection Coverage Analysis
10786 GNAT supports the HP Performance Coverage Analyzer (PCA), which allows
10787 the user to determine the distribution of execution time across a program,
10788 @pxref{Profiling} for details of usage.
10792 @node Text_IO Suggestions
10793 @section @code{Text_IO} Suggestions
10794 @cindex @code{Text_IO} and performance
10797 The @code{Ada.Text_IO} package has fairly high overheads due in part to
10798 the requirement of maintaining page and line counts. If performance
10799 is critical, a recommendation is to use @code{Stream_IO} instead of
10800 @code{Text_IO} for volume output, since this package has less overhead.
10802 If @code{Text_IO} must be used, note that by default output to the standard
10803 output and standard error files is unbuffered (this provides better
10804 behavior when output statements are used for debugging, or if the
10805 progress of a program is observed by tracking the output, e.g. by
10806 using the Unix @command{tail -f} command to watch redirected output.
10808 If you are generating large volumes of output with @code{Text_IO} and
10809 performance is an important factor, use a designated file instead
10810 of the standard output file, or change the standard output file to
10811 be buffered using @code{Interfaces.C_Streams.setvbuf}.
10815 @node Reducing Size of Ada Executables with gnatelim
10816 @section Reducing Size of Ada Executables with @code{gnatelim}
10820 This section describes @command{gnatelim}, a tool which detects unused
10821 subprograms and helps the compiler to create a smaller executable for your
10826 * Running gnatelim::
10827 * Processing Precompiled Libraries::
10828 * Correcting the List of Eliminate Pragmas::
10829 * Making Your Executables Smaller::
10830 * Summary of the gnatelim Usage Cycle::
10833 @node About gnatelim
10834 @subsection About @code{gnatelim}
10837 When a program shares a set of Ada
10838 packages with other programs, it may happen that this program uses
10839 only a fraction of the subprograms defined in these packages. The code
10840 created for these unused subprograms increases the size of the executable.
10842 @code{gnatelim} tracks unused subprograms in an Ada program and
10843 outputs a list of GNAT-specific pragmas @code{Eliminate} marking all the
10844 subprograms that are declared but never called. By placing the list of
10845 @code{Eliminate} pragmas in the GNAT configuration file @file{gnat.adc} and
10846 recompiling your program, you may decrease the size of its executable,
10847 because the compiler will not generate the code for 'eliminated' subprograms.
10848 @xref{Pragma Eliminate,,, gnat_rm, GNAT Reference Manual}, for more
10849 information about this pragma.
10851 @code{gnatelim} needs as its input data the name of the main subprogram.
10853 If a set of source files is specified as @code{gnatelim} arguments, it
10854 treats these files as a complete set of sources making up a program to
10855 analyse, and analyses only these sources.
10857 After a full successful build of the main subprogram @code{gnatelim} can be
10858 called without specifying sources to analyse, in this case it computes
10859 the source closure of the main unit from the @file{ALI} files.
10861 The following command will create the set of @file{ALI} files needed for
10865 $ gnatmake ^-c Main_Prog^/ACTIONS=COMPILE MAIN_PROG^
10868 Note that @code{gnatelim} does not need object files.
10870 @node Running gnatelim
10871 @subsection Running @code{gnatelim}
10874 @code{gnatelim} has the following command-line interface:
10877 $ gnatelim [@var{switches}] ^-main^?MAIN^=@var{main_unit_name} @{@var{filename}@} @r{[}-cargs @var{gcc_switches}@r{]}
10881 @var{main_unit_name} should be a name of a source file that contains the main
10882 subprogram of a program (partition).
10884 Each @var{filename} is the name (including the extension) of a source
10885 file to process. ``Wildcards'' are allowed, and
10886 the file name may contain path information.
10888 @samp{@var{gcc_switches}} is a list of switches for
10889 @command{gcc}. They will be passed on to all compiler invocations made by
10890 @command{gnatelim} to generate the ASIS trees. Here you can provide
10891 @option{^-I^/INCLUDE_DIRS=^} switches to form the source search path,
10892 use the @option{-gnatec} switch to set the configuration file,
10893 use the @option{-gnat05} switch if sources should be compiled in
10896 @code{gnatelim} has the following switches:
10900 @item ^-files^/FILES^=@var{filename}
10901 @cindex @option{^-files^/FILES^} (@code{gnatelim})
10902 Take the argument source files from the specified file. This file should be an
10903 ordinary text file containing file names separated by spaces or
10904 line breaks. You can use this switch more than once in the same call to
10905 @command{gnatelim}. You also can combine this switch with
10906 an explicit list of files.
10909 @cindex @option{^-log^/LOG^} (@command{gnatelim})
10910 Duplicate all the output sent to @file{stderr} into a log file. The log file
10911 is named @file{gnatelim.log} and is located in the current directory.
10913 @item ^-log^/LOGFILE^=@var{filename}
10914 @cindex @option{^-log^/LOGFILE^} (@command{gnatelim})
10915 Duplicate all the output sent to @file{stderr} into a specified log file.
10917 @cindex @option{^--no-elim-dispatch^/NO_DISPATCH^} (@command{gnatelim})
10918 @item ^--no-elim-dispatch^/NO_DISPATCH^
10919 Do not generate pragmas for dispatching operations.
10921 @item ^--ignore^/IGNORE^=@var{filename}
10922 @cindex @option{^--ignore^/IGNORE^} (@command{gnatelim})
10923 Do not generate pragmas for subprograms declared in the sources
10924 listed in a specified file
10926 @cindex @option{^-o^/OUTPUT^} (@command{gnatelim})
10927 @item ^-o^/OUTPUT^=@var{report_file}
10928 Put @command{gnatelim} output into a specified file. If this file already exists,
10929 it is overridden. If this switch is not used, @command{gnatelim} outputs its results
10933 @cindex @option{^-q^/QUIET^} (@command{gnatelim})
10934 Quiet mode: by default @code{gnatelim} outputs to the standard error
10935 stream the number of program units left to be processed. This option turns
10938 @cindex @option{^-t^/TIME^} (@command{gnatelim})
10940 Print out execution time.
10942 @item ^-v^/VERBOSE^
10943 @cindex @option{^-v^/VERBOSE^} (@command{gnatelim})
10944 Verbose mode: @code{gnatelim} version information is printed as Ada
10945 comments to the standard output stream. Also, in addition to the number of
10946 program units left @code{gnatelim} will output the name of the current unit
10949 @item ^-wq^/WARNINGS=QUIET^
10950 @cindex @option{^-wq^/WARNINGS=QUIET^} (@command{gnatelim})
10951 Quet warning mode - some warnings are suppressed. In particular warnings that
10952 indicate that the analysed set of sources is incomplete to make up a
10953 partition and that some subprogram bodies are missing are not generated.
10956 @node Processing Precompiled Libraries
10957 @subsection Processing Precompiled Libraries
10960 If some program uses a precompiled Ada library, it can be processed by
10961 @code{gnatelim} in a usual way. @code{gnatelim} will newer generate an
10962 Eliminate pragma for a subprogram if the body of this subprogram has not
10963 been analysed, this is a typical case for subprograms from precompiled
10964 libraries. Switch @option{^-wq^/WARNINGS=QUIET^} may be used to suppress
10965 warnings about missing source files and non-analyzed subprogram bodies
10966 that can be generated when processing precompiled Ada libraries.
10968 @node Correcting the List of Eliminate Pragmas
10969 @subsection Correcting the List of Eliminate Pragmas
10972 In some rare cases @code{gnatelim} may try to eliminate
10973 subprograms that are actually called in the program. In this case, the
10974 compiler will generate an error message of the form:
10977 main.adb:4:08: cannot reference subprogram "P" eliminated at elim.out:5
10981 You will need to manually remove the wrong @code{Eliminate} pragmas from
10982 the configuration file indicated in the error message. You should recompile
10983 your program from scratch after that, because you need a consistent
10984 configuration file(s) during the entire compilation.
10986 @node Making Your Executables Smaller
10987 @subsection Making Your Executables Smaller
10990 In order to get a smaller executable for your program you now have to
10991 recompile the program completely with the configuration file containing
10992 pragmas Eliminate generated by gnatelim. If these pragmas are placed in
10993 @file{gnat.adc} file located in your current directory, just do:
10996 $ gnatmake ^-f main_prog^/FORCE_COMPILE MAIN_PROG^
11000 (Use the @option{^-f^/FORCE_COMPILE^} option for @command{gnatmake} to
11001 recompile everything
11002 with the set of pragmas @code{Eliminate} that you have obtained with
11003 @command{gnatelim}).
11005 Be aware that the set of @code{Eliminate} pragmas is specific to each
11006 program. It is not recommended to merge sets of @code{Eliminate}
11007 pragmas created for different programs in one configuration file.
11009 @node Summary of the gnatelim Usage Cycle
11010 @subsection Summary of the @code{gnatelim} Usage Cycle
11013 Here is a quick summary of the steps to be taken in order to reduce
11014 the size of your executables with @code{gnatelim}. You may use
11015 other GNAT options to control the optimization level,
11016 to produce the debugging information, to set search path, etc.
11020 Create a complete set of @file{ALI} files (if the program has not been
11024 $ gnatmake ^-c main_prog^/ACTIONS=COMPILE MAIN_PROG^
11028 Generate a list of @code{Eliminate} pragmas in default configuration file
11029 @file{gnat.adc} in the current directory
11032 $ PIPE GNAT ELIM MAIN_PROG > GNAT.ADC
11035 $ gnatelim main_prog >@r{[}>@r{]} gnat.adc
11040 Recompile the application
11043 $ gnatmake ^-f main_prog^/FORCE_COMPILE MAIN_PROG^
11048 @node Reducing Size of Executables with unused subprogram/data elimination
11049 @section Reducing Size of Executables with Unused Subprogram/Data Elimination
11050 @findex unused subprogram/data elimination
11053 This section describes how you can eliminate unused subprograms and data from
11054 your executable just by setting options at compilation time.
11057 * About unused subprogram/data elimination::
11058 * Compilation options::
11059 * Example of unused subprogram/data elimination::
11062 @node About unused subprogram/data elimination
11063 @subsection About unused subprogram/data elimination
11066 By default, an executable contains all code and data of its composing objects
11067 (directly linked or coming from statically linked libraries), even data or code
11068 never used by this executable.
11070 This feature will allow you to eliminate such unused code from your
11071 executable, making it smaller (in disk and in memory).
11073 This functionality is available on all Linux platforms except for the IA-64
11074 architecture and on all cross platforms using the ELF binary file format.
11075 In both cases GNU binutils version 2.16 or later are required to enable it.
11077 @node Compilation options
11078 @subsection Compilation options
11081 The operation of eliminating the unused code and data from the final executable
11082 is directly performed by the linker.
11084 In order to do this, it has to work with objects compiled with the
11086 @option{-ffunction-sections} @option{-fdata-sections}.
11087 @cindex @option{-ffunction-sections} (@command{gcc})
11088 @cindex @option{-fdata-sections} (@command{gcc})
11089 These options are usable with C and Ada files.
11090 They will place respectively each
11091 function or data in a separate section in the resulting object file.
11093 Once the objects and static libraries are created with these options, the
11094 linker can perform the dead code elimination. You can do this by setting
11095 the @option{-Wl,--gc-sections} option to gcc command or in the
11096 @option{-largs} section of @command{gnatmake}. This will perform a
11097 garbage collection of code and data never referenced.
11099 If the linker performs a partial link (@option{-r} ld linker option), then you
11100 will need to provide one or several entry point using the
11101 @option{-e} / @option{--entry} ld option.
11103 Note that objects compiled without the @option{-ffunction-sections} and
11104 @option{-fdata-sections} options can still be linked with the executable.
11105 However, no dead code elimination will be performed on those objects (they will
11108 The GNAT static library is now compiled with -ffunction-sections and
11109 -fdata-sections on some platforms. This allows you to eliminate the unused code
11110 and data of the GNAT library from your executable.
11112 @node Example of unused subprogram/data elimination
11113 @subsection Example of unused subprogram/data elimination
11116 Here is a simple example:
11118 @smallexample @c ada
11127 Used_Data : Integer;
11128 Unused_Data : Integer;
11130 procedure Used (Data : Integer);
11131 procedure Unused (Data : Integer);
11134 package body Aux is
11135 procedure Used (Data : Integer) is
11140 procedure Unused (Data : Integer) is
11142 Unused_Data := Data;
11148 @code{Unused} and @code{Unused_Data} are never referenced in this code
11149 excerpt, and hence they may be safely removed from the final executable.
11154 $ nm test | grep used
11155 020015f0 T aux__unused
11156 02005d88 B aux__unused_data
11157 020015cc T aux__used
11158 02005d84 B aux__used_data
11160 $ gnatmake test -cargs -fdata-sections -ffunction-sections \
11161 -largs -Wl,--gc-sections
11163 $ nm test | grep used
11164 02005350 T aux__used
11165 0201ffe0 B aux__used_data
11169 It can be observed that the procedure @code{Unused} and the object
11170 @code{Unused_Data} are removed by the linker when using the
11171 appropriate options.
11173 @c ********************************
11174 @node Renaming Files Using gnatchop
11175 @chapter Renaming Files Using @code{gnatchop}
11179 This chapter discusses how to handle files with multiple units by using
11180 the @code{gnatchop} utility. This utility is also useful in renaming
11181 files to meet the standard GNAT default file naming conventions.
11184 * Handling Files with Multiple Units::
11185 * Operating gnatchop in Compilation Mode::
11186 * Command Line for gnatchop::
11187 * Switches for gnatchop::
11188 * Examples of gnatchop Usage::
11191 @node Handling Files with Multiple Units
11192 @section Handling Files with Multiple Units
11195 The basic compilation model of GNAT requires that a file submitted to the
11196 compiler have only one unit and there be a strict correspondence
11197 between the file name and the unit name.
11199 The @code{gnatchop} utility allows both of these rules to be relaxed,
11200 allowing GNAT to process files which contain multiple compilation units
11201 and files with arbitrary file names. @code{gnatchop}
11202 reads the specified file and generates one or more output files,
11203 containing one unit per file. The unit and the file name correspond,
11204 as required by GNAT.
11206 If you want to permanently restructure a set of ``foreign'' files so that
11207 they match the GNAT rules, and do the remaining development using the
11208 GNAT structure, you can simply use @command{gnatchop} once, generate the
11209 new set of files and work with them from that point on.
11211 Alternatively, if you want to keep your files in the ``foreign'' format,
11212 perhaps to maintain compatibility with some other Ada compilation
11213 system, you can set up a procedure where you use @command{gnatchop} each
11214 time you compile, regarding the source files that it writes as temporary
11215 files that you throw away.
11217 Note that if your file containing multiple units starts with a byte order
11218 mark (BOM) specifying UTF-8 encoding, then the files generated by gnatchop
11219 will each start with a copy of this BOM, meaning that they can be compiled
11220 automatically in UTF-8 mode without needing to specify an explicit encoding.
11222 @node Operating gnatchop in Compilation Mode
11223 @section Operating gnatchop in Compilation Mode
11226 The basic function of @code{gnatchop} is to take a file with multiple units
11227 and split it into separate files. The boundary between files is reasonably
11228 clear, except for the issue of comments and pragmas. In default mode, the
11229 rule is that any pragmas between units belong to the previous unit, except
11230 that configuration pragmas always belong to the following unit. Any comments
11231 belong to the following unit. These rules
11232 almost always result in the right choice of
11233 the split point without needing to mark it explicitly and most users will
11234 find this default to be what they want. In this default mode it is incorrect to
11235 submit a file containing only configuration pragmas, or one that ends in
11236 configuration pragmas, to @code{gnatchop}.
11238 However, using a special option to activate ``compilation mode'',
11240 can perform another function, which is to provide exactly the semantics
11241 required by the RM for handling of configuration pragmas in a compilation.
11242 In the absence of configuration pragmas (at the main file level), this
11243 option has no effect, but it causes such configuration pragmas to be handled
11244 in a quite different manner.
11246 First, in compilation mode, if @code{gnatchop} is given a file that consists of
11247 only configuration pragmas, then this file is appended to the
11248 @file{gnat.adc} file in the current directory. This behavior provides
11249 the required behavior described in the RM for the actions to be taken
11250 on submitting such a file to the compiler, namely that these pragmas
11251 should apply to all subsequent compilations in the same compilation
11252 environment. Using GNAT, the current directory, possibly containing a
11253 @file{gnat.adc} file is the representation
11254 of a compilation environment. For more information on the
11255 @file{gnat.adc} file, see @ref{Handling of Configuration Pragmas}.
11257 Second, in compilation mode, if @code{gnatchop}
11258 is given a file that starts with
11259 configuration pragmas, and contains one or more units, then these
11260 configuration pragmas are prepended to each of the chopped files. This
11261 behavior provides the required behavior described in the RM for the
11262 actions to be taken on compiling such a file, namely that the pragmas
11263 apply to all units in the compilation, but not to subsequently compiled
11266 Finally, if configuration pragmas appear between units, they are appended
11267 to the previous unit. This results in the previous unit being illegal,
11268 since the compiler does not accept configuration pragmas that follow
11269 a unit. This provides the required RM behavior that forbids configuration
11270 pragmas other than those preceding the first compilation unit of a
11273 For most purposes, @code{gnatchop} will be used in default mode. The
11274 compilation mode described above is used only if you need exactly
11275 accurate behavior with respect to compilations, and you have files
11276 that contain multiple units and configuration pragmas. In this
11277 circumstance the use of @code{gnatchop} with the compilation mode
11278 switch provides the required behavior, and is for example the mode
11279 in which GNAT processes the ACVC tests.
11281 @node Command Line for gnatchop
11282 @section Command Line for @code{gnatchop}
11285 The @code{gnatchop} command has the form:
11288 @c $ gnatchop switches @var{file name} @r{[}@var{file name} @dots{}@r{]}
11289 @c @ovar{directory}
11290 @c Expanding @ovar macro inline (explanation in macro def comments)
11291 $ gnatchop switches @var{file name} @r{[}@var{file name} @dots{}@r{]}
11292 @r{[}@var{directory}@r{]}
11296 The only required argument is the file name of the file to be chopped.
11297 There are no restrictions on the form of this file name. The file itself
11298 contains one or more Ada units, in normal GNAT format, concatenated
11299 together. As shown, more than one file may be presented to be chopped.
11301 When run in default mode, @code{gnatchop} generates one output file in
11302 the current directory for each unit in each of the files.
11304 @var{directory}, if specified, gives the name of the directory to which
11305 the output files will be written. If it is not specified, all files are
11306 written to the current directory.
11308 For example, given a
11309 file called @file{hellofiles} containing
11311 @smallexample @c ada
11316 with Text_IO; use Text_IO;
11319 Put_Line ("Hello");
11329 $ gnatchop ^hellofiles^HELLOFILES.^
11333 generates two files in the current directory, one called
11334 @file{hello.ads} containing the single line that is the procedure spec,
11335 and the other called @file{hello.adb} containing the remaining text. The
11336 original file is not affected. The generated files can be compiled in
11340 When gnatchop is invoked on a file that is empty or that contains only empty
11341 lines and/or comments, gnatchop will not fail, but will not produce any
11344 For example, given a
11345 file called @file{toto.txt} containing
11347 @smallexample @c ada
11359 $ gnatchop ^toto.txt^TOT.TXT^
11363 will not produce any new file and will result in the following warnings:
11366 toto.txt:1:01: warning: empty file, contains no compilation units
11367 no compilation units found
11368 no source files written
11371 @node Switches for gnatchop
11372 @section Switches for @code{gnatchop}
11375 @command{gnatchop} recognizes the following switches:
11381 @cindex @option{--version} @command{gnatchop}
11382 Display Copyright and version, then exit disregarding all other options.
11385 @cindex @option{--help} @command{gnatchop}
11386 If @option{--version} was not used, display usage, then exit disregarding
11389 @item ^-c^/COMPILATION^
11390 @cindex @option{^-c^/COMPILATION^} (@code{gnatchop})
11391 Causes @code{gnatchop} to operate in compilation mode, in which
11392 configuration pragmas are handled according to strict RM rules. See
11393 previous section for a full description of this mode.
11396 @item -gnat@var{xxx}
11397 This passes the given @option{-gnat@var{xxx}} switch to @code{gnat} which is
11398 used to parse the given file. Not all @var{xxx} options make sense,
11399 but for example, the use of @option{-gnati2} allows @code{gnatchop} to
11400 process a source file that uses Latin-2 coding for identifiers.
11404 Causes @code{gnatchop} to generate a brief help summary to the standard
11405 output file showing usage information.
11407 @item ^-k@var{mm}^/FILE_NAME_MAX_LENGTH=@var{mm}^
11408 @cindex @option{^-k^/FILE_NAME_MAX_LENGTH^} (@code{gnatchop})
11409 Limit generated file names to the specified number @code{mm}
11411 This is useful if the
11412 resulting set of files is required to be interoperable with systems
11413 which limit the length of file names.
11415 If no value is given, or
11416 if no @code{/FILE_NAME_MAX_LENGTH} qualifier is given,
11417 a default of 39, suitable for OpenVMS Alpha
11418 Systems, is assumed
11421 No space is allowed between the @option{-k} and the numeric value. The numeric
11422 value may be omitted in which case a default of @option{-k8},
11424 with DOS-like file systems, is used. If no @option{-k} switch
11426 there is no limit on the length of file names.
11429 @item ^-p^/PRESERVE^
11430 @cindex @option{^-p^/PRESERVE^} (@code{gnatchop})
11431 Causes the file ^modification^creation^ time stamp of the input file to be
11432 preserved and used for the time stamp of the output file(s). This may be
11433 useful for preserving coherency of time stamps in an environment where
11434 @code{gnatchop} is used as part of a standard build process.
11437 @cindex @option{^-q^/QUIET^} (@code{gnatchop})
11438 Causes output of informational messages indicating the set of generated
11439 files to be suppressed. Warnings and error messages are unaffected.
11441 @item ^-r^/REFERENCE^
11442 @cindex @option{^-r^/REFERENCE^} (@code{gnatchop})
11443 @findex Source_Reference
11444 Generate @code{Source_Reference} pragmas. Use this switch if the output
11445 files are regarded as temporary and development is to be done in terms
11446 of the original unchopped file. This switch causes
11447 @code{Source_Reference} pragmas to be inserted into each of the
11448 generated files to refers back to the original file name and line number.
11449 The result is that all error messages refer back to the original
11451 In addition, the debugging information placed into the object file (when
11452 the @option{^-g^/DEBUG^} switch of @command{gcc} or @command{gnatmake} is
11454 also refers back to this original file so that tools like profilers and
11455 debuggers will give information in terms of the original unchopped file.
11457 If the original file to be chopped itself contains
11458 a @code{Source_Reference}
11459 pragma referencing a third file, then gnatchop respects
11460 this pragma, and the generated @code{Source_Reference} pragmas
11461 in the chopped file refer to the original file, with appropriate
11462 line numbers. This is particularly useful when @code{gnatchop}
11463 is used in conjunction with @code{gnatprep} to compile files that
11464 contain preprocessing statements and multiple units.
11466 @item ^-v^/VERBOSE^
11467 @cindex @option{^-v^/VERBOSE^} (@code{gnatchop})
11468 Causes @code{gnatchop} to operate in verbose mode. The version
11469 number and copyright notice are output, as well as exact copies of
11470 the gnat1 commands spawned to obtain the chop control information.
11472 @item ^-w^/OVERWRITE^
11473 @cindex @option{^-w^/OVERWRITE^} (@code{gnatchop})
11474 Overwrite existing file names. Normally @code{gnatchop} regards it as a
11475 fatal error if there is already a file with the same name as a
11476 file it would otherwise output, in other words if the files to be
11477 chopped contain duplicated units. This switch bypasses this
11478 check, and causes all but the last instance of such duplicated
11479 units to be skipped.
11482 @item --GCC=@var{xxxx}
11483 @cindex @option{--GCC=} (@code{gnatchop})
11484 Specify the path of the GNAT parser to be used. When this switch is used,
11485 no attempt is made to add the prefix to the GNAT parser executable.
11489 @node Examples of gnatchop Usage
11490 @section Examples of @code{gnatchop} Usage
11494 @item gnatchop /OVERWRITE HELLO_S.ADA [PRERELEASE.FILES]
11497 @item gnatchop -w hello_s.ada prerelease/files
11500 Chops the source file @file{hello_s.ada}. The output files will be
11501 placed in the directory @file{^prerelease/files^[PRERELEASE.FILES]^},
11503 files with matching names in that directory (no files in the current
11504 directory are modified).
11506 @item gnatchop ^archive^ARCHIVE.^
11507 Chops the source file @file{^archive^ARCHIVE.^}
11508 into the current directory. One
11509 useful application of @code{gnatchop} is in sending sets of sources
11510 around, for example in email messages. The required sources are simply
11511 concatenated (for example, using a ^Unix @code{cat}^VMS @code{APPEND/NEW}^
11513 @command{gnatchop} is used at the other end to reconstitute the original
11516 @item gnatchop file1 file2 file3 direc
11517 Chops all units in files @file{file1}, @file{file2}, @file{file3}, placing
11518 the resulting files in the directory @file{direc}. Note that if any units
11519 occur more than once anywhere within this set of files, an error message
11520 is generated, and no files are written. To override this check, use the
11521 @option{^-w^/OVERWRITE^} switch,
11522 in which case the last occurrence in the last file will
11523 be the one that is output, and earlier duplicate occurrences for a given
11524 unit will be skipped.
11527 @node Configuration Pragmas
11528 @chapter Configuration Pragmas
11529 @cindex Configuration pragmas
11530 @cindex Pragmas, configuration
11533 Configuration pragmas include those pragmas described as
11534 such in the Ada Reference Manual, as well as
11535 implementation-dependent pragmas that are configuration pragmas.
11536 @xref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference Manual},
11537 for details on these additional GNAT-specific configuration pragmas.
11538 Most notably, the pragma @code{Source_File_Name}, which allows
11539 specifying non-default names for source files, is a configuration
11540 pragma. The following is a complete list of configuration pragmas
11541 recognized by GNAT:
11551 Assume_No_Invalid_Values
11556 Compile_Time_Warning
11558 Component_Alignment
11559 Convention_Identifier
11562 Default_Storage_Pool
11568 External_Name_Casing
11571 Float_Representation
11584 Priority_Specific_Dispatching
11587 Propagate_Exceptions
11590 Restricted_Run_Time
11592 Restrictions_Warnings
11594 Short_Circuit_And_Or
11596 Source_File_Name_Project
11599 Suppress_Exception_Locations
11600 Task_Dispatching_Policy
11606 Wide_Character_Encoding
11611 * Handling of Configuration Pragmas::
11612 * The Configuration Pragmas Files::
11615 @node Handling of Configuration Pragmas
11616 @section Handling of Configuration Pragmas
11618 Configuration pragmas may either appear at the start of a compilation
11619 unit, in which case they apply only to that unit, or they may apply to
11620 all compilations performed in a given compilation environment.
11622 GNAT also provides the @code{gnatchop} utility to provide an automatic
11623 way to handle configuration pragmas following the semantics for
11624 compilations (that is, files with multiple units), described in the RM.
11625 See @ref{Operating gnatchop in Compilation Mode} for details.
11626 However, for most purposes, it will be more convenient to edit the
11627 @file{gnat.adc} file that contains configuration pragmas directly,
11628 as described in the following section.
11630 @node The Configuration Pragmas Files
11631 @section The Configuration Pragmas Files
11632 @cindex @file{gnat.adc}
11635 In GNAT a compilation environment is defined by the current
11636 directory at the time that a compile command is given. This current
11637 directory is searched for a file whose name is @file{gnat.adc}. If
11638 this file is present, it is expected to contain one or more
11639 configuration pragmas that will be applied to the current compilation.
11640 However, if the switch @option{-gnatA} is used, @file{gnat.adc} is not
11643 Configuration pragmas may be entered into the @file{gnat.adc} file
11644 either by running @code{gnatchop} on a source file that consists only of
11645 configuration pragmas, or more conveniently by
11646 direct editing of the @file{gnat.adc} file, which is a standard format
11649 In addition to @file{gnat.adc}, additional files containing configuration
11650 pragmas may be applied to the current compilation using the switch
11651 @option{-gnatec}@var{path}. @var{path} must designate an existing file that
11652 contains only configuration pragmas. These configuration pragmas are
11653 in addition to those found in @file{gnat.adc} (provided @file{gnat.adc}
11654 is present and switch @option{-gnatA} is not used).
11656 It is allowed to specify several switches @option{-gnatec}, all of which
11657 will be taken into account.
11659 If you are using project file, a separate mechanism is provided using
11660 project attributes, see @ref{Specifying Configuration Pragmas} for more
11664 Of special interest to GNAT OpenVMS Alpha is the following
11665 configuration pragma:
11667 @smallexample @c ada
11669 pragma Extend_System (Aux_DEC);
11674 In the presence of this pragma, GNAT adds to the definition of the
11675 predefined package SYSTEM all the additional types and subprograms that are
11676 defined in HP Ada. See @ref{Compatibility with HP Ada} for details.
11679 @node Handling Arbitrary File Naming Conventions Using gnatname
11680 @chapter Handling Arbitrary File Naming Conventions Using @code{gnatname}
11681 @cindex Arbitrary File Naming Conventions
11684 * Arbitrary File Naming Conventions::
11685 * Running gnatname::
11686 * Switches for gnatname::
11687 * Examples of gnatname Usage::
11690 @node Arbitrary File Naming Conventions
11691 @section Arbitrary File Naming Conventions
11694 The GNAT compiler must be able to know the source file name of a compilation
11695 unit. When using the standard GNAT default file naming conventions
11696 (@code{.ads} for specs, @code{.adb} for bodies), the GNAT compiler
11697 does not need additional information.
11700 When the source file names do not follow the standard GNAT default file naming
11701 conventions, the GNAT compiler must be given additional information through
11702 a configuration pragmas file (@pxref{Configuration Pragmas})
11704 When the non-standard file naming conventions are well-defined,
11705 a small number of pragmas @code{Source_File_Name} specifying a naming pattern
11706 (@pxref{Alternative File Naming Schemes}) may be sufficient. However,
11707 if the file naming conventions are irregular or arbitrary, a number
11708 of pragma @code{Source_File_Name} for individual compilation units
11710 To help maintain the correspondence between compilation unit names and
11711 source file names within the compiler,
11712 GNAT provides a tool @code{gnatname} to generate the required pragmas for a
11715 @node Running gnatname
11716 @section Running @code{gnatname}
11719 The usual form of the @code{gnatname} command is
11722 @c $ gnatname @ovar{switches} @var{naming_pattern} @ovar{naming_patterns}
11723 @c @r{[}--and @ovar{switches} @var{naming_pattern} @ovar{naming_patterns}@r{]}
11724 @c Expanding @ovar macro inline (explanation in macro def comments)
11725 $ gnatname @r{[}@var{switches}@r{]} @var{naming_pattern} @r{[}@var{naming_patterns}@r{]}
11726 @r{[}--and @r{[}@var{switches}@r{]} @var{naming_pattern} @r{[}@var{naming_patterns}@r{]}@r{]}
11730 All of the arguments are optional. If invoked without any argument,
11731 @code{gnatname} will display its usage.
11734 When used with at least one naming pattern, @code{gnatname} will attempt to
11735 find all the compilation units in files that follow at least one of the
11736 naming patterns. To find these compilation units,
11737 @code{gnatname} will use the GNAT compiler in syntax-check-only mode on all
11741 One or several Naming Patterns may be given as arguments to @code{gnatname}.
11742 Each Naming Pattern is enclosed between double quotes (or single
11743 quotes on Windows).
11744 A Naming Pattern is a regular expression similar to the wildcard patterns
11745 used in file names by the Unix shells or the DOS prompt.
11748 @code{gnatname} may be called with several sections of directories/patterns.
11749 Sections are separated by switch @code{--and}. In each section, there must be
11750 at least one pattern. If no directory is specified in a section, the current
11751 directory (or the project directory is @code{-P} is used) is implied.
11752 The options other that the directory switches and the patterns apply globally
11753 even if they are in different sections.
11756 Examples of Naming Patterns are
11765 For a more complete description of the syntax of Naming Patterns,
11766 see the second kind of regular expressions described in @file{g-regexp.ads}
11767 (the ``Glob'' regular expressions).
11770 When invoked with no switch @code{-P}, @code{gnatname} will create a
11771 configuration pragmas file @file{gnat.adc} in the current working directory,
11772 with pragmas @code{Source_File_Name} for each file that contains a valid Ada
11775 @node Switches for gnatname
11776 @section Switches for @code{gnatname}
11779 Switches for @code{gnatname} must precede any specified Naming Pattern.
11782 You may specify any of the following switches to @code{gnatname}:
11788 @cindex @option{--version} @command{gnatname}
11789 Display Copyright and version, then exit disregarding all other options.
11792 @cindex @option{--help} @command{gnatname}
11793 If @option{--version} was not used, display usage, then exit disregarding
11797 Start another section of directories/patterns.
11799 @item ^-c^/CONFIG_FILE=^@file{file}
11800 @cindex @option{^-c^/CONFIG_FILE^} (@code{gnatname})
11801 Create a configuration pragmas file @file{file} (instead of the default
11804 There may be zero, one or more space between @option{-c} and
11807 @file{file} may include directory information. @file{file} must be
11808 writable. There may be only one switch @option{^-c^/CONFIG_FILE^}.
11809 When a switch @option{^-c^/CONFIG_FILE^} is
11810 specified, no switch @option{^-P^/PROJECT_FILE^} may be specified (see below).
11812 @item ^-d^/SOURCE_DIRS=^@file{dir}
11813 @cindex @option{^-d^/SOURCE_DIRS^} (@code{gnatname})
11814 Look for source files in directory @file{dir}. There may be zero, one or more
11815 spaces between @option{^-d^/SOURCE_DIRS=^} and @file{dir}.
11816 When a switch @option{^-d^/SOURCE_DIRS^}
11817 is specified, the current working directory will not be searched for source
11818 files, unless it is explicitly specified with a @option{^-d^/SOURCE_DIRS^}
11819 or @option{^-D^/DIR_FILES^} switch.
11820 Several switches @option{^-d^/SOURCE_DIRS^} may be specified.
11821 If @file{dir} is a relative path, it is relative to the directory of
11822 the configuration pragmas file specified with switch
11823 @option{^-c^/CONFIG_FILE^},
11824 or to the directory of the project file specified with switch
11825 @option{^-P^/PROJECT_FILE^} or,
11826 if neither switch @option{^-c^/CONFIG_FILE^}
11827 nor switch @option{^-P^/PROJECT_FILE^} are specified, it is relative to the
11828 current working directory. The directory
11829 specified with switch @option{^-d^/SOURCE_DIRS^} must exist and be readable.
11831 @item ^-D^/DIRS_FILE=^@file{file}
11832 @cindex @option{^-D^/DIRS_FILE^} (@code{gnatname})
11833 Look for source files in all directories listed in text file @file{file}.
11834 There may be zero, one or more spaces between @option{^-D^/DIRS_FILE=^}
11836 @file{file} must be an existing, readable text file.
11837 Each nonempty line in @file{file} must be a directory.
11838 Specifying switch @option{^-D^/DIRS_FILE^} is equivalent to specifying as many
11839 switches @option{^-d^/SOURCE_DIRS^} as there are nonempty lines in
11842 @item ^-f^/FOREIGN_PATTERN=^@file{pattern}
11843 @cindex @option{^-f^/FOREIGN_PATTERN^} (@code{gnatname})
11844 Foreign patterns. Using this switch, it is possible to add sources of languages
11845 other than Ada to the list of sources of a project file.
11846 It is only useful if a ^-P^/PROJECT_FILE^ switch is used.
11849 gnatname ^-Pprj -f"*.c"^/PROJECT_FILE=PRJ /FOREIGN_PATTERN=*.C^ "*.ada"
11852 will look for Ada units in all files with the @file{.ada} extension,
11853 and will add to the list of file for project @file{prj.gpr} the C files
11854 with extension @file{.^c^C^}.
11857 @cindex @option{^-h^/HELP^} (@code{gnatname})
11858 Output usage (help) information. The output is written to @file{stdout}.
11860 @item ^-P^/PROJECT_FILE=^@file{proj}
11861 @cindex @option{^-P^/PROJECT_FILE^} (@code{gnatname})
11862 Create or update project file @file{proj}. There may be zero, one or more space
11863 between @option{-P} and @file{proj}. @file{proj} may include directory
11864 information. @file{proj} must be writable.
11865 There may be only one switch @option{^-P^/PROJECT_FILE^}.
11866 When a switch @option{^-P^/PROJECT_FILE^} is specified,
11867 no switch @option{^-c^/CONFIG_FILE^} may be specified.
11869 @item ^-v^/VERBOSE^
11870 @cindex @option{^-v^/VERBOSE^} (@code{gnatname})
11871 Verbose mode. Output detailed explanation of behavior to @file{stdout}.
11872 This includes name of the file written, the name of the directories to search
11873 and, for each file in those directories whose name matches at least one of
11874 the Naming Patterns, an indication of whether the file contains a unit,
11875 and if so the name of the unit.
11877 @item ^-v -v^/VERBOSE /VERBOSE^
11878 @cindex @option{^-v -v^/VERBOSE /VERBOSE^} (@code{gnatname})
11879 Very Verbose mode. In addition to the output produced in verbose mode,
11880 for each file in the searched directories whose name matches none of
11881 the Naming Patterns, an indication is given that there is no match.
11883 @item ^-x^/EXCLUDED_PATTERN=^@file{pattern}
11884 @cindex @option{^-x^/EXCLUDED_PATTERN^} (@code{gnatname})
11885 Excluded patterns. Using this switch, it is possible to exclude some files
11886 that would match the name patterns. For example,
11888 gnatname ^-x "*_nt.ada"^/EXCLUDED_PATTERN=*_nt.ada^ "*.ada"
11891 will look for Ada units in all files with the @file{.ada} extension,
11892 except those whose names end with @file{_nt.ada}.
11896 @node Examples of gnatname Usage
11897 @section Examples of @code{gnatname} Usage
11901 $ gnatname /CONFIG_FILE=[HOME.ME]NAMES.ADC /SOURCE_DIRS=SOURCES "[a-z]*.ada*"
11907 $ gnatname -c /home/me/names.adc -d sources "[a-z]*.ada*"
11912 In this example, the directory @file{^/home/me^[HOME.ME]^} must already exist
11913 and be writable. In addition, the directory
11914 @file{^/home/me/sources^[HOME.ME.SOURCES]^} (specified by
11915 @option{^-d sources^/SOURCE_DIRS=SOURCES^}) must exist and be readable.
11918 Note the optional spaces after @option{-c} and @option{-d}.
11923 $ gnatname -P/home/me/proj -x "*_nt_body.ada"
11924 -dsources -dsources/plus -Dcommon_dirs.txt "body_*" "spec_*"
11927 $ gnatname /PROJECT_FILE=[HOME.ME]PROJ
11928 /EXCLUDED_PATTERN=*_nt_body.ada
11929 /SOURCE_DIRS=(SOURCES,[SOURCES.PLUS])
11930 /DIRS_FILE=COMMON_DIRS.TXT "body_*" "spec_*"
11934 Note that several switches @option{^-d^/SOURCE_DIRS^} may be used,
11935 even in conjunction with one or several switches
11936 @option{^-D^/DIRS_FILE^}. Several Naming Patterns and one excluded pattern
11937 are used in this example.
11939 @c *****************************************
11940 @c * G N A T P r o j e c t M a n a g e r *
11941 @c *****************************************
11943 @c ------ macros for projects.texi
11944 @c These macros are needed when building the gprbuild documentation, but
11945 @c should have no effect in the gnat user's guide
11947 @macro CODESAMPLE{TXT}
11955 @macro PROJECTFILE{TXT}
11959 @c simulates a newline when in a @CODESAMPLE
11970 @macro TIPHTML{TXT}
11974 @macro IMPORTANT{TXT}
11989 @include projects.texi
11991 @c *****************************************
11992 @c * Cross-referencing tools
11993 @c *****************************************
11995 @node The Cross-Referencing Tools gnatxref and gnatfind
11996 @chapter The Cross-Referencing Tools @code{gnatxref} and @code{gnatfind}
12001 The compiler generates cross-referencing information (unless
12002 you set the @samp{-gnatx} switch), which are saved in the @file{.ali} files.
12003 This information indicates where in the source each entity is declared and
12004 referenced. Note that entities in package Standard are not included, but
12005 entities in all other predefined units are included in the output.
12007 Before using any of these two tools, you need to compile successfully your
12008 application, so that GNAT gets a chance to generate the cross-referencing
12011 The two tools @code{gnatxref} and @code{gnatfind} take advantage of this
12012 information to provide the user with the capability to easily locate the
12013 declaration and references to an entity. These tools are quite similar,
12014 the difference being that @code{gnatfind} is intended for locating
12015 definitions and/or references to a specified entity or entities, whereas
12016 @code{gnatxref} is oriented to generating a full report of all
12019 To use these tools, you must not compile your application using the
12020 @option{-gnatx} switch on the @command{gnatmake} command line
12021 (@pxref{The GNAT Make Program gnatmake}). Otherwise, cross-referencing
12022 information will not be generated.
12024 Note: to invoke @code{gnatxref} or @code{gnatfind} with a project file,
12025 use the @code{gnat} driver (see @ref{The GNAT Driver and Project Files}).
12028 * Switches for gnatxref::
12029 * Switches for gnatfind::
12030 * Project Files for gnatxref and gnatfind::
12031 * Regular Expressions in gnatfind and gnatxref::
12032 * Examples of gnatxref Usage::
12033 * Examples of gnatfind Usage::
12036 @node Switches for gnatxref
12037 @section @code{gnatxref} Switches
12040 The command invocation for @code{gnatxref} is:
12042 @c $ gnatxref @ovar{switches} @var{sourcefile1} @r{[}@var{sourcefile2} @dots{}@r{]}
12043 @c Expanding @ovar macro inline (explanation in macro def comments)
12044 $ gnatxref @r{[}@var{switches}@r{]} @var{sourcefile1} @r{[}@var{sourcefile2} @dots{}@r{]}
12053 identifies the source files for which a report is to be generated. The
12054 ``with''ed units will be processed too. You must provide at least one file.
12056 These file names are considered to be regular expressions, so for instance
12057 specifying @file{source*.adb} is the same as giving every file in the current
12058 directory whose name starts with @file{source} and whose extension is
12061 You shouldn't specify any directory name, just base names. @command{gnatxref}
12062 and @command{gnatfind} will be able to locate these files by themselves using
12063 the source path. If you specify directories, no result is produced.
12068 The switches can be:
12072 @cindex @option{--version} @command{gnatxref}
12073 Display Copyright and version, then exit disregarding all other options.
12076 @cindex @option{--help} @command{gnatxref}
12077 If @option{--version} was not used, display usage, then exit disregarding
12080 @item ^-a^/ALL_FILES^
12081 @cindex @option{^-a^/ALL_FILES^} (@command{gnatxref})
12082 If this switch is present, @code{gnatfind} and @code{gnatxref} will parse
12083 the read-only files found in the library search path. Otherwise, these files
12084 will be ignored. This option can be used to protect Gnat sources or your own
12085 libraries from being parsed, thus making @code{gnatfind} and @code{gnatxref}
12086 much faster, and their output much smaller. Read-only here refers to access
12087 or permissions status in the file system for the current user.
12090 @cindex @option{-aIDIR} (@command{gnatxref})
12091 When looking for source files also look in directory DIR. The order in which
12092 source file search is undertaken is the same as for @command{gnatmake}.
12095 @cindex @option{-aODIR} (@command{gnatxref})
12096 When searching for library and object files, look in directory
12097 DIR. The order in which library files are searched is the same as for
12098 @command{gnatmake}.
12101 @cindex @option{-nostdinc} (@command{gnatxref})
12102 Do not look for sources in the system default directory.
12105 @cindex @option{-nostdlib} (@command{gnatxref})
12106 Do not look for library files in the system default directory.
12108 @item --ext=@var{extension}
12109 @cindex @option{--ext} (@command{gnatxref})
12110 Specify an alternate ali file extension. The default is @code{ali} and other
12111 extensions (e.g. @code{sli} for SPARK library files) may be specified via this
12112 switch. Note that if this switch overrides the default, which means that only
12113 the new extension will be considered.
12115 @item --RTS=@var{rts-path}
12116 @cindex @option{--RTS} (@command{gnatxref})
12117 Specifies the default location of the runtime library. Same meaning as the
12118 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
12120 @item ^-d^/DERIVED_TYPES^
12121 @cindex @option{^-d^/DERIVED_TYPES^} (@command{gnatxref})
12122 If this switch is set @code{gnatxref} will output the parent type
12123 reference for each matching derived types.
12125 @item ^-f^/FULL_PATHNAME^
12126 @cindex @option{^-f^/FULL_PATHNAME^} (@command{gnatxref})
12127 If this switch is set, the output file names will be preceded by their
12128 directory (if the file was found in the search path). If this switch is
12129 not set, the directory will not be printed.
12131 @item ^-g^/IGNORE_LOCALS^
12132 @cindex @option{^-g^/IGNORE_LOCALS^} (@command{gnatxref})
12133 If this switch is set, information is output only for library-level
12134 entities, ignoring local entities. The use of this switch may accelerate
12135 @code{gnatfind} and @code{gnatxref}.
12138 @cindex @option{-IDIR} (@command{gnatxref})
12139 Equivalent to @samp{-aODIR -aIDIR}.
12142 @cindex @option{-pFILE} (@command{gnatxref})
12143 Specify a project file to use @xref{GNAT Project Manager}.
12144 If you need to use the @file{.gpr}
12145 project files, you should use gnatxref through the GNAT driver
12146 (@command{gnat xref -Pproject}).
12148 By default, @code{gnatxref} and @code{gnatfind} will try to locate a
12149 project file in the current directory.
12151 If a project file is either specified or found by the tools, then the content
12152 of the source directory and object directory lines are added as if they
12153 had been specified respectively by @samp{^-aI^/SOURCE_SEARCH^}
12154 and @samp{^-aO^OBJECT_SEARCH^}.
12156 Output only unused symbols. This may be really useful if you give your
12157 main compilation unit on the command line, as @code{gnatxref} will then
12158 display every unused entity and 'with'ed package.
12162 Instead of producing the default output, @code{gnatxref} will generate a
12163 @file{tags} file that can be used by vi. For examples how to use this
12164 feature, see @ref{Examples of gnatxref Usage}. The tags file is output
12165 to the standard output, thus you will have to redirect it to a file.
12171 All these switches may be in any order on the command line, and may even
12172 appear after the file names. They need not be separated by spaces, thus
12173 you can say @samp{gnatxref ^-ag^/ALL_FILES/IGNORE_LOCALS^} instead of
12174 @samp{gnatxref ^-a -g^/ALL_FILES /IGNORE_LOCALS^}.
12176 @node Switches for gnatfind
12177 @section @code{gnatfind} Switches
12180 The command line for @code{gnatfind} is:
12183 @c $ gnatfind @ovar{switches} @var{pattern}@r{[}:@var{sourcefile}@r{[}:@var{line}@r{[}:@var{column}@r{]]]}
12184 @c @r{[}@var{file1} @var{file2} @dots{}]
12185 @c Expanding @ovar macro inline (explanation in macro def comments)
12186 $ gnatfind @r{[}@var{switches}@r{]} @var{pattern}@r{[}:@var{sourcefile}@r{[}:@var{line}@r{[}:@var{column}@r{]]]}
12187 @r{[}@var{file1} @var{file2} @dots{}@r{]}
12195 An entity will be output only if it matches the regular expression found
12196 in @var{pattern}, see @ref{Regular Expressions in gnatfind and gnatxref}.
12198 Omitting the pattern is equivalent to specifying @samp{*}, which
12199 will match any entity. Note that if you do not provide a pattern, you
12200 have to provide both a sourcefile and a line.
12202 Entity names are given in Latin-1, with uppercase/lowercase equivalence
12203 for matching purposes. At the current time there is no support for
12204 8-bit codes other than Latin-1, or for wide characters in identifiers.
12207 @code{gnatfind} will look for references, bodies or declarations
12208 of symbols referenced in @file{@var{sourcefile}}, at line @var{line}
12209 and column @var{column}. See @ref{Examples of gnatfind Usage}
12210 for syntax examples.
12213 is a decimal integer identifying the line number containing
12214 the reference to the entity (or entities) to be located.
12217 is a decimal integer identifying the exact location on the
12218 line of the first character of the identifier for the
12219 entity reference. Columns are numbered from 1.
12221 @item file1 file2 @dots{}
12222 The search will be restricted to these source files. If none are given, then
12223 the search will be done for every library file in the search path.
12224 These file must appear only after the pattern or sourcefile.
12226 These file names are considered to be regular expressions, so for instance
12227 specifying @file{source*.adb} is the same as giving every file in the current
12228 directory whose name starts with @file{source} and whose extension is
12231 The location of the spec of the entity will always be displayed, even if it
12232 isn't in one of @file{@var{file1}}, @file{@var{file2}},@enddots{} The
12233 occurrences of the entity in the separate units of the ones given on the
12234 command line will also be displayed.
12236 Note that if you specify at least one file in this part, @code{gnatfind} may
12237 sometimes not be able to find the body of the subprograms.
12242 At least one of 'sourcefile' or 'pattern' has to be present on
12245 The following switches are available:
12249 @cindex @option{--version} @command{gnatfind}
12250 Display Copyright and version, then exit disregarding all other options.
12253 @cindex @option{--help} @command{gnatfind}
12254 If @option{--version} was not used, display usage, then exit disregarding
12257 @item ^-a^/ALL_FILES^
12258 @cindex @option{^-a^/ALL_FILES^} (@command{gnatfind})
12259 If this switch is present, @code{gnatfind} and @code{gnatxref} will parse
12260 the read-only files found in the library search path. Otherwise, these files
12261 will be ignored. This option can be used to protect Gnat sources or your own
12262 libraries from being parsed, thus making @code{gnatfind} and @code{gnatxref}
12263 much faster, and their output much smaller. Read-only here refers to access
12264 or permission status in the file system for the current user.
12267 @cindex @option{-aIDIR} (@command{gnatfind})
12268 When looking for source files also look in directory DIR. The order in which
12269 source file search is undertaken is the same as for @command{gnatmake}.
12272 @cindex @option{-aODIR} (@command{gnatfind})
12273 When searching for library and object files, look in directory
12274 DIR. The order in which library files are searched is the same as for
12275 @command{gnatmake}.
12278 @cindex @option{-nostdinc} (@command{gnatfind})
12279 Do not look for sources in the system default directory.
12282 @cindex @option{-nostdlib} (@command{gnatfind})
12283 Do not look for library files in the system default directory.
12285 @item --ext=@var{extension}
12286 @cindex @option{--ext} (@command{gnatfind})
12287 Specify an alternate ali file extension. The default is @code{ali} and other
12288 extensions (e.g. @code{sli} for SPARK library files) may be specified via this
12289 switch. Note that if this switch overrides the default, which means that only
12290 the new extension will be considered.
12292 @item --RTS=@var{rts-path}
12293 @cindex @option{--RTS} (@command{gnatfind})
12294 Specifies the default location of the runtime library. Same meaning as the
12295 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
12297 @item ^-d^/DERIVED_TYPE_INFORMATION^
12298 @cindex @option{^-d^/DERIVED_TYPE_INFORMATION^} (@code{gnatfind})
12299 If this switch is set, then @code{gnatfind} will output the parent type
12300 reference for each matching derived types.
12302 @item ^-e^/EXPRESSIONS^
12303 @cindex @option{^-e^/EXPRESSIONS^} (@command{gnatfind})
12304 By default, @code{gnatfind} accept the simple regular expression set for
12305 @samp{pattern}. If this switch is set, then the pattern will be
12306 considered as full Unix-style regular expression.
12308 @item ^-f^/FULL_PATHNAME^
12309 @cindex @option{^-f^/FULL_PATHNAME^} (@command{gnatfind})
12310 If this switch is set, the output file names will be preceded by their
12311 directory (if the file was found in the search path). If this switch is
12312 not set, the directory will not be printed.
12314 @item ^-g^/IGNORE_LOCALS^
12315 @cindex @option{^-g^/IGNORE_LOCALS^} (@command{gnatfind})
12316 If this switch is set, information is output only for library-level
12317 entities, ignoring local entities. The use of this switch may accelerate
12318 @code{gnatfind} and @code{gnatxref}.
12321 @cindex @option{-IDIR} (@command{gnatfind})
12322 Equivalent to @samp{-aODIR -aIDIR}.
12325 @cindex @option{-pFILE} (@command{gnatfind})
12326 Specify a project file (@pxref{GNAT Project Manager}) to use.
12327 By default, @code{gnatxref} and @code{gnatfind} will try to locate a
12328 project file in the current directory.
12330 If a project file is either specified or found by the tools, then the content
12331 of the source directory and object directory lines are added as if they
12332 had been specified respectively by @samp{^-aI^/SOURCE_SEARCH^} and
12333 @samp{^-aO^/OBJECT_SEARCH^}.
12335 @item ^-r^/REFERENCES^
12336 @cindex @option{^-r^/REFERENCES^} (@command{gnatfind})
12337 By default, @code{gnatfind} will output only the information about the
12338 declaration, body or type completion of the entities. If this switch is
12339 set, the @code{gnatfind} will locate every reference to the entities in
12340 the files specified on the command line (or in every file in the search
12341 path if no file is given on the command line).
12343 @item ^-s^/PRINT_LINES^
12344 @cindex @option{^-s^/PRINT_LINES^} (@command{gnatfind})
12345 If this switch is set, then @code{gnatfind} will output the content
12346 of the Ada source file lines were the entity was found.
12348 @item ^-t^/TYPE_HIERARCHY^
12349 @cindex @option{^-t^/TYPE_HIERARCHY^} (@command{gnatfind})
12350 If this switch is set, then @code{gnatfind} will output the type hierarchy for
12351 the specified type. It act like -d option but recursively from parent
12352 type to parent type. When this switch is set it is not possible to
12353 specify more than one file.
12358 All these switches may be in any order on the command line, and may even
12359 appear after the file names. They need not be separated by spaces, thus
12360 you can say @samp{gnatxref ^-ag^/ALL_FILES/IGNORE_LOCALS^} instead of
12361 @samp{gnatxref ^-a -g^/ALL_FILES /IGNORE_LOCALS^}.
12363 As stated previously, gnatfind will search in every directory in the
12364 search path. You can force it to look only in the current directory if
12365 you specify @code{*} at the end of the command line.
12367 @node Project Files for gnatxref and gnatfind
12368 @section Project Files for @command{gnatxref} and @command{gnatfind}
12371 Project files allow a programmer to specify how to compile its
12372 application, where to find sources, etc. These files are used
12374 primarily by GPS, but they can also be used
12377 @code{gnatxref} and @code{gnatfind}.
12379 A project file name must end with @file{.gpr}. If a single one is
12380 present in the current directory, then @code{gnatxref} and @code{gnatfind} will
12381 extract the information from it. If multiple project files are found, none of
12382 them is read, and you have to use the @samp{-p} switch to specify the one
12385 The following lines can be included, even though most of them have default
12386 values which can be used in most cases.
12387 The lines can be entered in any order in the file.
12388 Except for @file{src_dir} and @file{obj_dir}, you can only have one instance of
12389 each line. If you have multiple instances, only the last one is taken into
12394 [default: @code{"^./^[]^"}]
12395 specifies a directory where to look for source files. Multiple @code{src_dir}
12396 lines can be specified and they will be searched in the order they
12400 [default: @code{"^./^[]^"}]
12401 specifies a directory where to look for object and library files. Multiple
12402 @code{obj_dir} lines can be specified, and they will be searched in the order
12405 @item comp_opt=SWITCHES
12406 [default: @code{""}]
12407 creates a variable which can be referred to subsequently by using
12408 the @code{$@{comp_opt@}} notation. This is intended to store the default
12409 switches given to @command{gnatmake} and @command{gcc}.
12411 @item bind_opt=SWITCHES
12412 [default: @code{""}]
12413 creates a variable which can be referred to subsequently by using
12414 the @samp{$@{bind_opt@}} notation. This is intended to store the default
12415 switches given to @command{gnatbind}.
12417 @item link_opt=SWITCHES
12418 [default: @code{""}]
12419 creates a variable which can be referred to subsequently by using
12420 the @samp{$@{link_opt@}} notation. This is intended to store the default
12421 switches given to @command{gnatlink}.
12423 @item main=EXECUTABLE
12424 [default: @code{""}]
12425 specifies the name of the executable for the application. This variable can
12426 be referred to in the following lines by using the @samp{$@{main@}} notation.
12429 @item comp_cmd=COMMAND
12430 [default: @code{"GNAT COMPILE /SEARCH=$@{src_dir@} /DEBUG /TRY_SEMANTICS"}]
12433 @item comp_cmd=COMMAND
12434 [default: @code{"gcc -c -I$@{src_dir@} -g -gnatq"}]
12436 specifies the command used to compile a single file in the application.
12439 @item make_cmd=COMMAND
12440 [default: @code{"GNAT MAKE $@{main@}
12441 /SOURCE_SEARCH=$@{src_dir@} /OBJECT_SEARCH=$@{obj_dir@}
12442 /DEBUG /TRY_SEMANTICS /COMPILER_QUALIFIERS $@{comp_opt@}
12443 /BINDER_QUALIFIERS $@{bind_opt@} /LINKER_QUALIFIERS $@{link_opt@}"}]
12446 @item make_cmd=COMMAND
12447 [default: @code{"gnatmake $@{main@} -aI$@{src_dir@}
12448 -aO$@{obj_dir@} -g -gnatq -cargs $@{comp_opt@}
12449 -bargs $@{bind_opt@} -largs $@{link_opt@}"}]
12451 specifies the command used to recompile the whole application.
12453 @item run_cmd=COMMAND
12454 [default: @code{"$@{main@}"}]
12455 specifies the command used to run the application.
12457 @item debug_cmd=COMMAND
12458 [default: @code{"gdb $@{main@}"}]
12459 specifies the command used to debug the application
12464 @command{gnatxref} and @command{gnatfind} only take into account the
12465 @code{src_dir} and @code{obj_dir} lines, and ignore the others.
12467 @node Regular Expressions in gnatfind and gnatxref
12468 @section Regular Expressions in @code{gnatfind} and @code{gnatxref}
12471 As specified in the section about @command{gnatfind}, the pattern can be a
12472 regular expression. Actually, there are to set of regular expressions
12473 which are recognized by the program:
12476 @item globbing patterns
12477 These are the most usual regular expression. They are the same that you
12478 generally used in a Unix shell command line, or in a DOS session.
12480 Here is a more formal grammar:
12487 term ::= elmt -- matches elmt
12488 term ::= elmt elmt -- concatenation (elmt then elmt)
12489 term ::= * -- any string of 0 or more characters
12490 term ::= ? -- matches any character
12491 term ::= [char @{char@}] -- matches any character listed
12492 term ::= [char - char] -- matches any character in range
12496 @item full regular expression
12497 The second set of regular expressions is much more powerful. This is the
12498 type of regular expressions recognized by utilities such a @file{grep}.
12500 The following is the form of a regular expression, expressed in Ada
12501 reference manual style BNF is as follows
12508 regexp ::= term @{| term@} -- alternation (term or term @dots{})
12510 term ::= item @{item@} -- concatenation (item then item)
12512 item ::= elmt -- match elmt
12513 item ::= elmt * -- zero or more elmt's
12514 item ::= elmt + -- one or more elmt's
12515 item ::= elmt ? -- matches elmt or nothing
12518 elmt ::= nschar -- matches given character
12519 elmt ::= [nschar @{nschar@}] -- matches any character listed
12520 elmt ::= [^^^ nschar @{nschar@}] -- matches any character not listed
12521 elmt ::= [char - char] -- matches chars in given range
12522 elmt ::= \ char -- matches given character
12523 elmt ::= . -- matches any single character
12524 elmt ::= ( regexp ) -- parens used for grouping
12526 char ::= any character, including special characters
12527 nschar ::= any character except ()[].*+?^^^
12531 Following are a few examples:
12535 will match any of the two strings @samp{abcde} and @samp{fghi},
12538 will match any string like @samp{abd}, @samp{abcd}, @samp{abccd},
12539 @samp{abcccd}, and so on,
12542 will match any string which has only lowercase characters in it (and at
12543 least one character.
12548 @node Examples of gnatxref Usage
12549 @section Examples of @code{gnatxref} Usage
12551 @subsection General Usage
12554 For the following examples, we will consider the following units:
12556 @smallexample @c ada
12562 3: procedure Foo (B : in Integer);
12569 1: package body Main is
12570 2: procedure Foo (B : in Integer) is
12581 2: procedure Print (B : Integer);
12590 The first thing to do is to recompile your application (for instance, in
12591 that case just by doing a @samp{gnatmake main}, so that GNAT generates
12592 the cross-referencing information.
12593 You can then issue any of the following commands:
12595 @item gnatxref main.adb
12596 @code{gnatxref} generates cross-reference information for main.adb
12597 and every unit 'with'ed by main.adb.
12599 The output would be:
12607 Decl: main.ads 3:20
12608 Body: main.adb 2:20
12609 Ref: main.adb 4:13 5:13 6:19
12612 Ref: main.adb 6:8 7:8
12622 Decl: main.ads 3:15
12623 Body: main.adb 2:15
12626 Body: main.adb 1:14
12629 Ref: main.adb 6:12 7:12
12633 that is the entity @code{Main} is declared in main.ads, line 2, column 9,
12634 its body is in main.adb, line 1, column 14 and is not referenced any where.
12636 The entity @code{Print} is declared in bar.ads, line 2, column 15 and it
12637 it referenced in main.adb, line 6 column 12 and line 7 column 12.
12639 @item gnatxref package1.adb package2.ads
12640 @code{gnatxref} will generates cross-reference information for
12641 package1.adb, package2.ads and any other package 'with'ed by any
12647 @subsection Using gnatxref with vi
12649 @code{gnatxref} can generate a tags file output, which can be used
12650 directly from @command{vi}. Note that the standard version of @command{vi}
12651 will not work properly with overloaded symbols. Consider using another
12652 free implementation of @command{vi}, such as @command{vim}.
12655 $ gnatxref -v gnatfind.adb > tags
12659 will generate the tags file for @code{gnatfind} itself (if the sources
12660 are in the search path!).
12662 From @command{vi}, you can then use the command @samp{:tag @var{entity}}
12663 (replacing @var{entity} by whatever you are looking for), and vi will
12664 display a new file with the corresponding declaration of entity.
12667 @node Examples of gnatfind Usage
12668 @section Examples of @code{gnatfind} Usage
12672 @item gnatfind ^-f^/FULL_PATHNAME^ xyz:main.adb
12673 Find declarations for all entities xyz referenced at least once in
12674 main.adb. The references are search in every library file in the search
12677 The directories will be printed as well (as the @samp{^-f^/FULL_PATHNAME^}
12680 The output will look like:
12682 ^directory/^[directory]^main.ads:106:14: xyz <= declaration
12683 ^directory/^[directory]^main.adb:24:10: xyz <= body
12684 ^directory/^[directory]^foo.ads:45:23: xyz <= declaration
12688 that is to say, one of the entities xyz found in main.adb is declared at
12689 line 12 of main.ads (and its body is in main.adb), and another one is
12690 declared at line 45 of foo.ads
12692 @item gnatfind ^-fs^/FULL_PATHNAME/SOURCE_LINE^ xyz:main.adb
12693 This is the same command as the previous one, instead @code{gnatfind} will
12694 display the content of the Ada source file lines.
12696 The output will look like:
12699 ^directory/^[directory]^main.ads:106:14: xyz <= declaration
12701 ^directory/^[directory]^main.adb:24:10: xyz <= body
12703 ^directory/^[directory]^foo.ads:45:23: xyz <= declaration
12708 This can make it easier to find exactly the location your are looking
12711 @item gnatfind ^-r^/REFERENCES^ "*x*":main.ads:123 foo.adb
12712 Find references to all entities containing an x that are
12713 referenced on line 123 of main.ads.
12714 The references will be searched only in main.ads and foo.adb.
12716 @item gnatfind main.ads:123
12717 Find declarations and bodies for all entities that are referenced on
12718 line 123 of main.ads.
12720 This is the same as @code{gnatfind "*":main.adb:123}.
12722 @item gnatfind ^mydir/^[mydir]^main.adb:123:45
12723 Find the declaration for the entity referenced at column 45 in
12724 line 123 of file main.adb in directory mydir. Note that it
12725 is usual to omit the identifier name when the column is given,
12726 since the column position identifies a unique reference.
12728 The column has to be the beginning of the identifier, and should not
12729 point to any character in the middle of the identifier.
12733 @c *********************************
12734 @node The GNAT Pretty-Printer gnatpp
12735 @chapter The GNAT Pretty-Printer @command{gnatpp}
12737 @cindex Pretty-Printer
12740 ^The @command{gnatpp} tool^GNAT PRETTY^ is an ASIS-based utility
12741 for source reformatting / pretty-printing.
12742 It takes an Ada source file as input and generates a reformatted
12744 You can specify various style directives via switches; e.g.,
12745 identifier case conventions, rules of indentation, and comment layout.
12747 To produce a reformatted file, @command{gnatpp} generates and uses the ASIS
12748 tree for the input source and thus requires the input to be syntactically and
12749 semantically legal.
12750 If this condition is not met, @command{gnatpp} will terminate with an
12751 error message; no output file will be generated.
12753 If the source files presented to @command{gnatpp} contain
12754 preprocessing directives, then the output file will
12755 correspond to the generated source after all
12756 preprocessing is carried out. There is no way
12757 using @command{gnatpp} to obtain pretty printed files that
12758 include the preprocessing directives.
12760 If the compilation unit
12761 contained in the input source depends semantically upon units located
12762 outside the current directory, you have to provide the source search path
12763 when invoking @command{gnatpp}, if these units are contained in files with
12764 names that do not follow the GNAT file naming rules, you have to provide
12765 the configuration file describing the corresponding naming scheme;
12766 see the description of the @command{gnatpp}
12767 switches below. Another possibility is to use a project file and to
12768 call @command{gnatpp} through the @command{gnat} driver
12770 The @command{gnatpp} command has the form
12773 @c $ gnatpp @ovar{switches} @var{filename}
12774 @c Expanding @ovar macro inline (explanation in macro def comments)
12775 $ gnatpp @r{[}@var{switches}@r{]} @var{filename} @r{[}-cargs @var{gcc_switches}@r{]}
12782 @var{switches} is an optional sequence of switches defining such properties as
12783 the formatting rules, the source search path, and the destination for the
12787 @var{filename} is the name (including the extension) of the source file to
12788 reformat; ``wildcards'' or several file names on the same gnatpp command are
12789 allowed. The file name may contain path information; it does not have to
12790 follow the GNAT file naming rules
12793 @samp{@var{gcc_switches}} is a list of switches for
12794 @command{gcc}. They will be passed on to all compiler invocations made by
12795 @command{gnatelim} to generate the ASIS trees. Here you can provide
12796 @option{^-I^/INCLUDE_DIRS=^} switches to form the source search path,
12797 use the @option{-gnatec} switch to set the configuration file,
12798 use the @option{-gnat05} switch if sources should be compiled in
12803 * Switches for gnatpp::
12804 * Formatting Rules::
12807 @node Switches for gnatpp
12808 @section Switches for @command{gnatpp}
12811 The following subsections describe the various switches accepted by
12812 @command{gnatpp}, organized by category.
12815 You specify a switch by supplying a name and generally also a value.
12816 In many cases the values for a switch with a given name are incompatible with
12818 (for example the switch that controls the casing of a reserved word may have
12819 exactly one value: upper case, lower case, or
12820 mixed case) and thus exactly one such switch can be in effect for an
12821 invocation of @command{gnatpp}.
12822 If more than one is supplied, the last one is used.
12823 However, some values for the same switch are mutually compatible.
12824 You may supply several such switches to @command{gnatpp}, but then
12825 each must be specified in full, with both the name and the value.
12826 Abbreviated forms (the name appearing once, followed by each value) are
12828 For example, to set
12829 the alignment of the assignment delimiter both in declarations and in
12830 assignment statements, you must write @option{-A2A3}
12831 (or @option{-A2 -A3}), but not @option{-A23}.
12835 In many cases the set of options for a given qualifier are incompatible with
12836 each other (for example the qualifier that controls the casing of a reserved
12837 word may have exactly one option, which specifies either upper case, lower
12838 case, or mixed case), and thus exactly one such option can be in effect for
12839 an invocation of @command{gnatpp}.
12840 If more than one is supplied, the last one is used.
12841 However, some qualifiers have options that are mutually compatible,
12842 and then you may then supply several such options when invoking
12846 In most cases, it is obvious whether or not the
12847 ^values for a switch with a given name^options for a given qualifier^
12848 are compatible with each other.
12849 When the semantics might not be evident, the summaries below explicitly
12850 indicate the effect.
12853 * Alignment Control::
12855 * Construct Layout Control::
12856 * General Text Layout Control::
12857 * Other Formatting Options::
12858 * Setting the Source Search Path::
12859 * Output File Control::
12860 * Other gnatpp Switches::
12863 @node Alignment Control
12864 @subsection Alignment Control
12865 @cindex Alignment control in @command{gnatpp}
12868 Programs can be easier to read if certain constructs are vertically aligned.
12869 By default all alignments are set ON.
12870 Through the @option{^-A0^/ALIGN=OFF^} switch you may reset the default to
12871 OFF, and then use one or more of the other
12872 ^@option{-A@var{n}} switches^@option{/ALIGN} options^
12873 to activate alignment for specific constructs.
12876 @cindex @option{^-A@var{n}^/ALIGN^} (@command{gnatpp})
12880 Set all alignments to ON
12883 @item ^-A0^/ALIGN=OFF^
12884 Set all alignments to OFF
12886 @item ^-A1^/ALIGN=COLONS^
12887 Align @code{:} in declarations
12889 @item ^-A2^/ALIGN=DECLARATIONS^
12890 Align @code{:=} in initializations in declarations
12892 @item ^-A3^/ALIGN=STATEMENTS^
12893 Align @code{:=} in assignment statements
12895 @item ^-A4^/ALIGN=ARROWS^
12896 Align @code{=>} in associations
12898 @item ^-A5^/ALIGN=COMPONENT_CLAUSES^
12899 Align @code{at} keywords in the component clauses in record
12900 representation clauses
12904 The @option{^-A^/ALIGN^} switches are mutually compatible; any combination
12907 @node Casing Control
12908 @subsection Casing Control
12909 @cindex Casing control in @command{gnatpp}
12912 @command{gnatpp} allows you to specify the casing for reserved words,
12913 pragma names, attribute designators and identifiers.
12914 For identifiers you may define a
12915 general rule for name casing but also override this rule
12916 via a set of dictionary files.
12918 Three types of casing are supported: lower case, upper case, and mixed case.
12919 Lower and upper case are self-explanatory (but since some letters in
12920 Latin1 and other GNAT-supported character sets
12921 exist only in lower-case form, an upper case conversion will have no
12923 ``Mixed case'' means that the first letter, and also each letter immediately
12924 following an underscore, are converted to their uppercase forms;
12925 all the other letters are converted to their lowercase forms.
12928 @cindex @option{^-a@var{x}^/ATTRIBUTE^} (@command{gnatpp})
12929 @item ^-aL^/ATTRIBUTE_CASING=LOWER_CASE^
12930 Attribute designators are lower case
12932 @item ^-aU^/ATTRIBUTE_CASING=UPPER_CASE^
12933 Attribute designators are upper case
12935 @item ^-aM^/ATTRIBUTE_CASING=MIXED_CASE^
12936 Attribute designators are mixed case (this is the default)
12938 @cindex @option{^-k@var{x}^/KEYWORD_CASING^} (@command{gnatpp})
12939 @item ^-kL^/KEYWORD_CASING=LOWER_CASE^
12940 Keywords (technically, these are known in Ada as @emph{reserved words}) are
12941 lower case (this is the default)
12943 @item ^-kU^/KEYWORD_CASING=UPPER_CASE^
12944 Keywords are upper case
12946 @cindex @option{^-n@var{x}^/NAME_CASING^} (@command{gnatpp})
12947 @item ^-nD^/NAME_CASING=AS_DECLARED^
12948 Name casing for defining occurrences are as they appear in the source file
12949 (this is the default)
12951 @item ^-nU^/NAME_CASING=UPPER_CASE^
12952 Names are in upper case
12954 @item ^-nL^/NAME_CASING=LOWER_CASE^
12955 Names are in lower case
12957 @item ^-nM^/NAME_CASING=MIXED_CASE^
12958 Names are in mixed case
12960 @cindex @option{^-p@var{x}^/PRAGMA_CASING^} (@command{gnatpp})
12961 @item ^-pL^/PRAGMA_CASING=LOWER_CASE^
12962 Pragma names are lower case
12964 @item ^-pU^/PRAGMA_CASING=UPPER_CASE^
12965 Pragma names are upper case
12967 @item ^-pM^/PRAGMA_CASING=MIXED_CASE^
12968 Pragma names are mixed case (this is the default)
12970 @item ^-D@var{file}^/DICTIONARY=@var{file}^
12971 @cindex @option{^-D^/DICTIONARY^} (@command{gnatpp})
12972 Use @var{file} as a @emph{dictionary file} that defines
12973 the casing for a set of specified names,
12974 thereby overriding the effect on these names by
12975 any explicit or implicit
12976 ^-n^/NAME_CASING^ switch.
12977 To supply more than one dictionary file,
12978 use ^several @option{-D} switches^a list of files as options^.
12981 @option{gnatpp} implicitly uses a @emph{default dictionary file}
12982 to define the casing for the Ada predefined names and
12983 the names declared in the GNAT libraries.
12985 @item ^-D-^/SPECIFIC_CASING^
12986 @cindex @option{^-D-^/SPECIFIC_CASING^} (@command{gnatpp})
12987 Do not use the default dictionary file;
12988 instead, use the casing
12989 defined by a @option{^-n^/NAME_CASING^} switch and any explicit
12994 The structure of a dictionary file, and details on the conventions
12995 used in the default dictionary file, are defined in @ref{Name Casing}.
12997 The @option{^-D-^/SPECIFIC_CASING^} and
12998 @option{^-D@var{file}^/DICTIONARY=@var{file}^} switches are mutually
13001 @node Construct Layout Control
13002 @subsection Construct Layout Control
13003 @cindex Layout control in @command{gnatpp}
13006 This group of @command{gnatpp} switches controls the layout of comments and
13007 complex syntactic constructs. See @ref{Formatting Comments} for details
13011 @cindex @option{^-c@var{n}^/COMMENTS_LAYOUT^} (@command{gnatpp})
13012 @item ^-c0^/COMMENTS_LAYOUT=UNTOUCHED^
13013 All the comments remain unchanged
13015 @item ^-c1^/COMMENTS_LAYOUT=DEFAULT^
13016 GNAT-style comment line indentation (this is the default).
13018 @item ^-c2^/COMMENTS_LAYOUT=STANDARD_INDENT^
13019 Reference-manual comment line indentation.
13021 @item ^-c3^/COMMENTS_LAYOUT=GNAT_BEGINNING^
13022 GNAT-style comment beginning
13024 @item ^-c4^/COMMENTS_LAYOUT=REFORMAT^
13025 Reformat comment blocks
13027 @item ^-c5^/COMMENTS_LAYOUT=KEEP_SPECIAL^
13028 Keep unchanged special form comments
13030 Reformat comment blocks
13032 @cindex @option{^-l@var{n}^/CONSTRUCT_LAYOUT^} (@command{gnatpp})
13033 @item ^-l1^/CONSTRUCT_LAYOUT=GNAT^
13034 GNAT-style layout (this is the default)
13036 @item ^-l2^/CONSTRUCT_LAYOUT=COMPACT^
13039 @item ^-l3^/CONSTRUCT_LAYOUT=UNCOMPACT^
13042 @cindex @option{^-N^/NOTABS^} (@command{gnatpp})
13044 All the VT characters are removed from the comment text. All the HT characters
13045 are expanded with the sequences of space characters to get to the next tab
13048 @cindex @option{^--no-separate-is^/NO_SEPARATE_IS^} (@command{gnatpp})
13049 @item ^--no-separate-is^/NO_SEPARATE_IS^
13050 Do not place the keyword @code{is} on a separate line in a subprogram body in
13051 case if the spec occupies more then one line.
13053 @cindex @option{^--separate-label^/SEPARATE_LABEL^} (@command{gnatpp})
13054 @item ^--separate-label^/SEPARATE_LABEL^
13055 Place statement label(s) on a separate line, with the following statement
13058 @cindex @option{^--separate-loop-then^/SEPARATE_LOOP_THEN^} (@command{gnatpp})
13059 @item ^--separate-loop-then^/SEPARATE_LOOP_THEN^
13060 Place the keyword @code{loop} in FOR and WHILE loop statements and the
13061 keyword @code{then} in IF statements on a separate line.
13063 @cindex @option{^--no-separate-loop-then^/NO_SEPARATE_LOOP_THEN^} (@command{gnatpp})
13064 @item ^--no-separate-loop-then^/NO_SEPARATE_LOOP_THEN^
13065 Do not place the keyword @code{loop} in FOR and WHILE loop statements and the
13066 keyword @code{then} in IF statements on a separate line. This option is
13067 incompatible with @option{^--separate-loop-then^/SEPARATE_LOOP_THEN^} option.
13069 @cindex @option{^--use-on-new-line^/USE_ON_NEW_LINE^} (@command{gnatpp})
13070 @item ^--use-on-new-line^/USE_ON_NEW_LINE^
13071 Start each USE clause in a context clause from a separate line.
13073 @cindex @option{^--separate-stmt-name^/STMT_NAME_ON_NEW_LINE^} (@command{gnatpp})
13074 @item ^--separate-stmt-name^/STMT_NAME_ON_NEW_LINE^
13075 Use a separate line for a loop or block statement name, but do not use an extra
13076 indentation level for the statement itself.
13082 The @option{-c1} and @option{-c2} switches are incompatible.
13083 The @option{-c3} and @option{-c4} switches are compatible with each other and
13084 also with @option{-c1} and @option{-c2}. The @option{-c0} switch disables all
13085 the other comment formatting switches.
13087 The @option{-l1}, @option{-l2}, and @option{-l3} switches are incompatible.
13092 For the @option{/COMMENTS_LAYOUT} qualifier:
13095 The @option{DEFAULT} and @option{STANDARD_INDENT} options are incompatible.
13097 The @option{GNAT_BEGINNING} and @option{REFORMAT} options are compatible with
13098 each other and also with @option{DEFAULT} and @option{STANDARD_INDENT}.
13102 The @option{GNAT}, @option{COMPACT}, and @option{UNCOMPACT} options for the
13103 @option{/CONSTRUCT_LAYOUT} qualifier are incompatible.
13106 @node General Text Layout Control
13107 @subsection General Text Layout Control
13110 These switches allow control over line length and indentation.
13113 @item ^-M@var{nnn}^/LINE_LENGTH_MAX=@var{nnn}^
13114 @cindex @option{^-M^/LINE_LENGTH^} (@command{gnatpp})
13115 Maximum line length, @var{nnn} from 32@dots{}256, the default value is 79
13117 @item ^-i@var{nnn}^/INDENTATION_LEVEL=@var{nnn}^
13118 @cindex @option{^-i^/INDENTATION_LEVEL^} (@command{gnatpp})
13119 Indentation level, @var{nnn} from 1@dots{}9, the default value is 3
13121 @item ^-cl@var{nnn}^/CONTINUATION_INDENT=@var{nnn}^
13122 @cindex @option{^-cl^/CONTINUATION_INDENT^} (@command{gnatpp})
13123 Indentation level for continuation lines (relative to the line being
13124 continued), @var{nnn} from 1@dots{}9.
13126 value is one less then the (normal) indentation level, unless the
13127 indentation is set to 1 (in which case the default value for continuation
13128 line indentation is also 1)
13131 @node Other Formatting Options
13132 @subsection Other Formatting Options
13135 These switches control the inclusion of missing end/exit labels, and
13136 the indentation level in @b{case} statements.
13139 @item ^-e^/NO_MISSED_LABELS^
13140 @cindex @option{^-e^/NO_MISSED_LABELS^} (@command{gnatpp})
13141 Do not insert missing end/exit labels. An end label is the name of
13142 a construct that may optionally be repeated at the end of the
13143 construct's declaration;
13144 e.g., the names of packages, subprograms, and tasks.
13145 An exit label is the name of a loop that may appear as target
13146 of an exit statement within the loop.
13147 By default, @command{gnatpp} inserts these end/exit labels when
13148 they are absent from the original source. This option suppresses such
13149 insertion, so that the formatted source reflects the original.
13151 @item ^-ff^/FORM_FEED_AFTER_PRAGMA_PAGE^
13152 @cindex @option{^-ff^/FORM_FEED_AFTER_PRAGMA_PAGE^} (@command{gnatpp})
13153 Insert a Form Feed character after a pragma Page.
13155 @item ^-T@var{nnn}^/MAX_INDENT=@var{nnn}^
13156 @cindex @option{^-T^/MAX_INDENT^} (@command{gnatpp})
13157 Do not use an additional indentation level for @b{case} alternatives
13158 and variants if there are @var{nnn} or more (the default
13160 If @var{nnn} is 0, an additional indentation level is
13161 used for @b{case} alternatives and variants regardless of their number.
13164 @node Setting the Source Search Path
13165 @subsection Setting the Source Search Path
13168 To define the search path for the input source file, @command{gnatpp}
13169 uses the same switches as the GNAT compiler, with the same effects.
13172 @item ^-I^/SEARCH=^@var{dir}
13173 @cindex @option{^-I^/SEARCH^} (@code{gnatpp})
13174 The same as the corresponding gcc switch
13176 @item ^-I-^/NOCURRENT_DIRECTORY^
13177 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@code{gnatpp})
13178 The same as the corresponding gcc switch
13180 @item ^-gnatec^/CONFIGURATION_PRAGMAS_FILE^=@var{path}
13181 @cindex @option{^-gnatec^/CONFIGURATION_PRAGMAS_FILE^} (@code{gnatpp})
13182 The same as the corresponding gcc switch
13184 @item ^--RTS^/RUNTIME_SYSTEM^=@var{path}
13185 @cindex @option{^--RTS^/RUNTIME_SYSTEM^} (@code{gnatpp})
13186 The same as the corresponding gcc switch
13190 @node Output File Control
13191 @subsection Output File Control
13194 By default the output is sent to the file whose name is obtained by appending
13195 the ^@file{.pp}^@file{$PP}^ suffix to the name of the input file
13196 (if the file with this name already exists, it is unconditionally overwritten).
13197 Thus if the input file is @file{^my_ada_proc.adb^MY_ADA_PROC.ADB^} then
13198 @command{gnatpp} will produce @file{^my_ada_proc.adb.pp^MY_ADA_PROC.ADB$PP^}
13200 The output may be redirected by the following switches:
13203 @item ^-pipe^/STANDARD_OUTPUT^
13204 @cindex @option{^-pipe^/STANDARD_OUTPUT^} (@code{gnatpp})
13205 Send the output to @code{Standard_Output}
13207 @item ^-o @var{output_file}^/OUTPUT=@var{output_file}^
13208 @cindex @option{^-o^/OUTPUT^} (@code{gnatpp})
13209 Write the output into @var{output_file}.
13210 If @var{output_file} already exists, @command{gnatpp} terminates without
13211 reading or processing the input file.
13213 @item ^-of ^/FORCED_OUTPUT=^@var{output_file}
13214 @cindex @option{^-of^/FORCED_OUTPUT^} (@code{gnatpp})
13215 Write the output into @var{output_file}, overwriting the existing file
13216 (if one is present).
13218 @item ^-r^/REPLACE^
13219 @cindex @option{^-r^/REPLACE^} (@code{gnatpp})
13220 Replace the input source file with the reformatted output, and copy the
13221 original input source into the file whose name is obtained by appending the
13222 ^@file{.npp}^@file{$NPP}^ suffix to the name of the input file.
13223 If a file with this name already exists, @command{gnatpp} terminates without
13224 reading or processing the input file.
13226 @item ^-rf^/OVERRIDING_REPLACE^
13227 @cindex @option{^-rf^/OVERRIDING_REPLACE^} (@code{gnatpp})
13228 Like @option{^-r^/REPLACE^} except that if the file with the specified name
13229 already exists, it is overwritten.
13231 @item ^-rnb^/REPLACE_NO_BACKUP^
13232 @cindex @option{^-rnb^/REPLACE_NO_BACKUP^} (@code{gnatpp})
13233 Replace the input source file with the reformatted output without
13234 creating any backup copy of the input source.
13236 @item ^--eol=@var{xxx}^/END_OF_LINE=@var{xxx}^
13237 @cindex @option{^--eol^/END_OF_LINE^} (@code{gnatpp})
13238 Specifies the format of the reformatted output file. The @var{xxx}
13239 ^string specified with the switch^option^ may be either
13241 @item ``@option{^dos^DOS^}'' MS DOS style, lines end with CR LF characters
13242 @item ``@option{^crlf^CRLF^}''
13243 the same as @option{^crlf^CRLF^}
13244 @item ``@option{^unix^UNIX^}'' UNIX style, lines end with LF character
13245 @item ``@option{^lf^LF^}''
13246 the same as @option{^unix^UNIX^}
13249 @item ^-W^/RESULT_ENCODING=^@var{e}
13250 @cindex @option{^-W^/RESULT_ENCODING=^} (@command{gnatpp})
13251 Specify the wide character encoding method used to write the code in the
13253 @var{e} is one of the following:
13261 Upper half encoding
13263 @item ^s^SHIFT_JIS^
13273 Brackets encoding (default value)
13279 Options @option{^-pipe^/STANDARD_OUTPUT^},
13280 @option{^-o^/OUTPUT^} and
13281 @option{^-of^/FORCED_OUTPUT^} are allowed only if the call to gnatpp
13282 contains only one file to reformat.
13284 @option{^--eol^/END_OF_LINE^}
13286 @option{^-W^/RESULT_ENCODING^}
13287 cannot be used together
13288 with @option{^-pipe^/STANDARD_OUTPUT^} option.
13290 @node Other gnatpp Switches
13291 @subsection Other @code{gnatpp} Switches
13294 The additional @command{gnatpp} switches are defined in this subsection.
13297 @item ^-files @var{filename}^/FILES=@var{filename}^
13298 @cindex @option{^-files^/FILES^} (@code{gnatpp})
13299 Take the argument source files from the specified file. This file should be an
13300 ordinary text file containing file names separated by spaces or
13301 line breaks. You can use this switch more than once in the same call to
13302 @command{gnatpp}. You also can combine this switch with an explicit list of
13305 @item ^-v^/VERBOSE^
13306 @cindex @option{^-v^/VERBOSE^} (@code{gnatpp})
13308 @command{gnatpp} generates version information and then
13309 a trace of the actions it takes to produce or obtain the ASIS tree.
13311 @item ^-w^/WARNINGS^
13312 @cindex @option{^-w^/WARNINGS^} (@code{gnatpp})
13314 @command{gnatpp} generates a warning whenever it cannot provide
13315 a required layout in the result source.
13318 @node Formatting Rules
13319 @section Formatting Rules
13322 The following subsections show how @command{gnatpp} treats ``white space'',
13323 comments, program layout, and name casing.
13324 They provide the detailed descriptions of the switches shown above.
13327 * White Space and Empty Lines::
13328 * Formatting Comments::
13329 * Construct Layout::
13333 @node White Space and Empty Lines
13334 @subsection White Space and Empty Lines
13337 @command{gnatpp} does not have an option to control space characters.
13338 It will add or remove spaces according to the style illustrated by the
13339 examples in the @cite{Ada Reference Manual}.
13341 The only format effectors
13342 (see @cite{Ada Reference Manual}, paragraph 2.1(13))
13343 that will appear in the output file are platform-specific line breaks,
13344 and also format effectors within (but not at the end of) comments.
13345 In particular, each horizontal tab character that is not inside
13346 a comment will be treated as a space and thus will appear in the
13347 output file as zero or more spaces depending on
13348 the reformatting of the line in which it appears.
13349 The only exception is a Form Feed character, which is inserted after a
13350 pragma @code{Page} when @option{-ff} is set.
13352 The output file will contain no lines with trailing ``white space'' (spaces,
13355 Empty lines in the original source are preserved
13356 only if they separate declarations or statements.
13357 In such contexts, a
13358 sequence of two or more empty lines is replaced by exactly one empty line.
13359 Note that a blank line will be removed if it separates two ``comment blocks''
13360 (a comment block is a sequence of whole-line comments).
13361 In order to preserve a visual separation between comment blocks, use an
13362 ``empty comment'' (a line comprising only hyphens) rather than an empty line.
13363 Likewise, if for some reason you wish to have a sequence of empty lines,
13364 use a sequence of empty comments instead.
13366 @node Formatting Comments
13367 @subsection Formatting Comments
13370 Comments in Ada code are of two kinds:
13373 a @emph{whole-line comment}, which appears by itself (possibly preceded by
13374 ``white space'') on a line
13377 an @emph{end-of-line comment}, which follows some other Ada lexical element
13382 The indentation of a whole-line comment is that of either
13383 the preceding or following line in
13384 the formatted source, depending on switch settings as will be described below.
13386 For an end-of-line comment, @command{gnatpp} leaves the same number of spaces
13387 between the end of the preceding Ada lexical element and the beginning
13388 of the comment as appear in the original source,
13389 unless either the comment has to be split to
13390 satisfy the line length limitation, or else the next line contains a
13391 whole line comment that is considered a continuation of this end-of-line
13392 comment (because it starts at the same position).
13394 cases, the start of the end-of-line comment is moved right to the nearest
13395 multiple of the indentation level.
13396 This may result in a ``line overflow'' (the right-shifted comment extending
13397 beyond the maximum line length), in which case the comment is split as
13400 There is a difference between @option{^-c1^/COMMENTS_LAYOUT=DEFAULT^}
13401 (GNAT-style comment line indentation)
13402 and @option{^-c2^/COMMENTS_LAYOUT=STANDARD_INDENT^}
13403 (reference-manual comment line indentation).
13404 With reference-manual style, a whole-line comment is indented as if it
13405 were a declaration or statement at the same place
13406 (i.e., according to the indentation of the preceding line(s)).
13407 With GNAT style, a whole-line comment that is immediately followed by an
13408 @b{if} or @b{case} statement alternative, a record variant, or the reserved
13409 word @b{begin}, is indented based on the construct that follows it.
13412 @smallexample @c ada
13424 Reference-manual indentation produces:
13426 @smallexample @c ada
13438 while GNAT-style indentation produces:
13440 @smallexample @c ada
13452 The @option{^-c3^/COMMENTS_LAYOUT=GNAT_BEGINNING^} switch
13453 (GNAT style comment beginning) has the following
13458 For each whole-line comment that does not end with two hyphens,
13459 @command{gnatpp} inserts spaces if necessary after the starting two hyphens
13460 to ensure that there are at least two spaces between these hyphens and the
13461 first non-blank character of the comment.
13465 For an end-of-line comment, if in the original source the next line is a
13466 whole-line comment that starts at the same position
13467 as the end-of-line comment,
13468 then the whole-line comment (and all whole-line comments
13469 that follow it and that start at the same position)
13470 will start at this position in the output file.
13473 That is, if in the original source we have:
13475 @smallexample @c ada
13478 A := B + C; -- B must be in the range Low1..High1
13479 -- C must be in the range Low2..High2
13480 --B+C will be in the range Low1+Low2..High1+High2
13486 Then in the formatted source we get
13488 @smallexample @c ada
13491 A := B + C; -- B must be in the range Low1..High1
13492 -- C must be in the range Low2..High2
13493 -- B+C will be in the range Low1+Low2..High1+High2
13499 A comment that exceeds the line length limit will be split.
13501 @option{^-c4^/COMMENTS_LAYOUT=REFORMAT^} (reformat comment blocks) is set and
13502 the line belongs to a reformattable block, splitting the line generates a
13503 @command{gnatpp} warning.
13504 The @option{^-c4^/COMMENTS_LAYOUT=REFORMAT^} switch specifies that whole-line
13505 comments may be reformatted in typical
13506 word processor style (that is, moving words between lines and putting as
13507 many words in a line as possible).
13510 The @option{^-c5^/COMMENTS_LAYOUT=KEEP_SPECIAL^} switch specifies, that comments
13511 that has a special format (that is, a character that is neither a letter nor digit
13512 not white space nor line break immediately following the leading @code{--} of
13513 the comment) should be without any change moved from the argument source
13514 into reformatted source. This switch allows to preserve comments that are used
13515 as a special marks in the code (e.g.@: SPARK annotation).
13517 @node Construct Layout
13518 @subsection Construct Layout
13521 In several cases the suggested layout in the Ada Reference Manual includes
13522 an extra level of indentation that many programmers prefer to avoid. The
13523 affected cases include:
13527 @item Record type declaration (RM 3.8)
13529 @item Record representation clause (RM 13.5.1)
13531 @item Loop statement in case if a loop has a statement identifier (RM 5.6)
13533 @item Block statement in case if a block has a statement identifier (RM 5.6)
13537 In compact mode (when GNAT style layout or compact layout is set),
13538 the pretty printer uses one level of indentation instead
13539 of two. This is achieved in the record definition and record representation
13540 clause cases by putting the @code{record} keyword on the same line as the
13541 start of the declaration or representation clause, and in the block and loop
13542 case by putting the block or loop header on the same line as the statement
13546 The difference between GNAT style @option{^-l1^/CONSTRUCT_LAYOUT=GNAT^}
13547 and compact @option{^-l2^/CONSTRUCT_LAYOUT=COMPACT^}
13548 layout on the one hand, and uncompact layout
13549 @option{^-l3^/CONSTRUCT_LAYOUT=UNCOMPACT^} on the other hand,
13550 can be illustrated by the following examples:
13554 @multitable @columnfractions .5 .5
13555 @item @i{GNAT style, compact layout} @tab @i{Uncompact layout}
13558 @smallexample @c ada
13565 @smallexample @c ada
13574 @smallexample @c ada
13576 a at 0 range 0 .. 31;
13577 b at 4 range 0 .. 31;
13581 @smallexample @c ada
13584 a at 0 range 0 .. 31;
13585 b at 4 range 0 .. 31;
13590 @smallexample @c ada
13598 @smallexample @c ada
13608 @smallexample @c ada
13609 Clear : for J in 1 .. 10 loop
13614 @smallexample @c ada
13616 for J in 1 .. 10 loop
13627 GNAT style, compact layout Uncompact layout
13629 type q is record type q is
13630 a : integer; record
13631 b : integer; a : integer;
13632 end record; b : integer;
13635 for q use record for q use
13636 a at 0 range 0 .. 31; record
13637 b at 4 range 0 .. 31; a at 0 range 0 .. 31;
13638 end record; b at 4 range 0 .. 31;
13641 Block : declare Block :
13642 A : Integer := 3; declare
13643 begin A : Integer := 3;
13645 end Block; Proc (A, A);
13648 Clear : for J in 1 .. 10 loop Clear :
13649 A (J) := 0; for J in 1 .. 10 loop
13650 end loop Clear; A (J) := 0;
13657 A further difference between GNAT style layout and compact layout is that
13658 GNAT style layout inserts empty lines as separation for
13659 compound statements, return statements and bodies.
13661 Note that the layout specified by
13662 @option{^--separate-stmt-name^/STMT_NAME_ON_NEW_LINE^}
13663 for named block and loop statements overrides the layout defined by these
13664 constructs by @option{^-l1^/CONSTRUCT_LAYOUT=GNAT^},
13665 @option{^-l2^/CONSTRUCT_LAYOUT=COMPACT^} or
13666 @option{^-l3^/CONSTRUCT_LAYOUT=UNCOMPACT^} option.
13669 @subsection Name Casing
13672 @command{gnatpp} always converts the usage occurrence of a (simple) name to
13673 the same casing as the corresponding defining identifier.
13675 You control the casing for defining occurrences via the
13676 @option{^-n^/NAME_CASING^} switch.
13678 With @option{-nD} (``as declared'', which is the default),
13681 With @option{/NAME_CASING=AS_DECLARED}, which is the default,
13683 defining occurrences appear exactly as in the source file
13684 where they are declared.
13685 The other ^values for this switch^options for this qualifier^ ---
13686 @option{^-nU^UPPER_CASE^},
13687 @option{^-nL^LOWER_CASE^},
13688 @option{^-nM^MIXED_CASE^} ---
13690 ^upper, lower, or mixed case, respectively^the corresponding casing^.
13691 If @command{gnatpp} changes the casing of a defining
13692 occurrence, it analogously changes the casing of all the
13693 usage occurrences of this name.
13695 If the defining occurrence of a name is not in the source compilation unit
13696 currently being processed by @command{gnatpp}, the casing of each reference to
13697 this name is changed according to the value of the @option{^-n^/NAME_CASING^}
13698 switch (subject to the dictionary file mechanism described below).
13699 Thus @command{gnatpp} acts as though the @option{^-n^/NAME_CASING^} switch
13701 casing for the defining occurrence of the name.
13703 Some names may need to be spelled with casing conventions that are not
13704 covered by the upper-, lower-, and mixed-case transformations.
13705 You can arrange correct casing by placing such names in a
13706 @emph{dictionary file},
13707 and then supplying a @option{^-D^/DICTIONARY^} switch.
13708 The casing of names from dictionary files overrides
13709 any @option{^-n^/NAME_CASING^} switch.
13711 To handle the casing of Ada predefined names and the names from GNAT libraries,
13712 @command{gnatpp} assumes a default dictionary file.
13713 The name of each predefined entity is spelled with the same casing as is used
13714 for the entity in the @cite{Ada Reference Manual}.
13715 The name of each entity in the GNAT libraries is spelled with the same casing
13716 as is used in the declaration of that entity.
13718 The @w{@option{^-D-^/SPECIFIC_CASING^}} switch suppresses the use of the
13719 default dictionary file.
13720 Instead, the casing for predefined and GNAT-defined names will be established
13721 by the @option{^-n^/NAME_CASING^} switch or explicit dictionary files.
13722 For example, by default the names @code{Ada.Text_IO} and @code{GNAT.OS_Lib}
13723 will appear as just shown,
13724 even in the presence of a @option{^-nU^/NAME_CASING=UPPER_CASE^} switch.
13725 To ensure that even such names are rendered in uppercase,
13726 additionally supply the @w{@option{^-D-^/SPECIFIC_CASING^}} switch
13727 (or else, less conveniently, place these names in upper case in a dictionary
13730 A dictionary file is
13731 a plain text file; each line in this file can be either a blank line
13732 (containing only space characters and ASCII.HT characters), an Ada comment
13733 line, or the specification of exactly one @emph{casing schema}.
13735 A casing schema is a string that has the following syntax:
13739 @var{casing_schema} ::= @var{identifier} | *@var{simple_identifier}*
13741 @var{simple_identifier} ::= @var{letter}@{@var{letter_or_digit}@}
13746 (See @cite{Ada Reference Manual}, Section 2.3) for the definition of the
13747 @var{identifier} lexical element and the @var{letter_or_digit} category.)
13749 The casing schema string can be followed by white space and/or an Ada-style
13750 comment; any amount of white space is allowed before the string.
13752 If a dictionary file is passed as
13754 the value of a @option{-D@var{file}} switch
13757 an option to the @option{/DICTIONARY} qualifier
13760 simple name and every identifier, @command{gnatpp} checks if the dictionary
13761 defines the casing for the name or for some of its parts (the term ``subword''
13762 is used below to denote the part of a name which is delimited by ``_'' or by
13763 the beginning or end of the word and which does not contain any ``_'' inside):
13767 if the whole name is in the dictionary, @command{gnatpp} uses for this name
13768 the casing defined by the dictionary; no subwords are checked for this word
13771 for every subword @command{gnatpp} checks if the dictionary contains the
13772 corresponding string of the form @code{*@var{simple_identifier}*},
13773 and if it does, the casing of this @var{simple_identifier} is used
13777 if the whole name does not contain any ``_'' inside, and if for this name
13778 the dictionary contains two entries - one of the form @var{identifier},
13779 and another - of the form *@var{simple_identifier}*, then the first one
13780 is applied to define the casing of this name
13783 if more than one dictionary file is passed as @command{gnatpp} switches, each
13784 dictionary adds new casing exceptions and overrides all the existing casing
13785 exceptions set by the previous dictionaries
13788 when @command{gnatpp} checks if the word or subword is in the dictionary,
13789 this check is not case sensitive
13793 For example, suppose we have the following source to reformat:
13795 @smallexample @c ada
13798 name1 : integer := 1;
13799 name4_name3_name2 : integer := 2;
13800 name2_name3_name4 : Boolean;
13803 name2_name3_name4 := name4_name3_name2 > name1;
13809 And suppose we have two dictionaries:
13826 If @command{gnatpp} is called with the following switches:
13830 @command{gnatpp -nM -D dict1 -D dict2 test.adb}
13833 @command{gnatpp test.adb /NAME_CASING=MIXED_CASE /DICTIONARY=(dict1, dict2)}
13838 then we will get the following name casing in the @command{gnatpp} output:
13840 @smallexample @c ada
13843 NAME1 : Integer := 1;
13844 Name4_NAME3_Name2 : Integer := 2;
13845 Name2_NAME3_Name4 : Boolean;
13848 Name2_NAME3_Name4 := Name4_NAME3_Name2 > NAME1;
13853 @c *********************************
13854 @node The GNAT Metric Tool gnatmetric
13855 @chapter The GNAT Metric Tool @command{gnatmetric}
13857 @cindex Metric tool
13860 ^The @command{gnatmetric} tool^@command{GNAT METRIC}^ is an ASIS-based utility
13861 for computing various program metrics.
13862 It takes an Ada source file as input and generates a file containing the
13863 metrics data as output. Various switches control which
13864 metrics are computed and output.
13866 @command{gnatmetric} generates and uses the ASIS
13867 tree for the input source and thus requires the input to be syntactically and
13868 semantically legal.
13869 If this condition is not met, @command{gnatmetric} will generate
13870 an error message; no metric information for this file will be
13871 computed and reported.
13873 If the compilation unit contained in the input source depends semantically
13874 upon units in files located outside the current directory, you have to provide
13875 the source search path when invoking @command{gnatmetric}.
13876 If it depends semantically upon units that are contained
13877 in files with names that do not follow the GNAT file naming rules, you have to
13878 provide the configuration file describing the corresponding naming scheme (see
13879 the description of the @command{gnatmetric} switches below.)
13880 Alternatively, you may use a project file and invoke @command{gnatmetric}
13881 through the @command{gnat} driver.
13883 The @command{gnatmetric} command has the form
13886 @c $ gnatmetric @ovar{switches} @{@var{filename}@} @r{[}-cargs @var{gcc_switches}@r{]}
13887 @c Expanding @ovar macro inline (explanation in macro def comments)
13888 $ gnatmetric @r{[}@var{switches}@r{]} @{@var{filename}@} @r{[}-cargs @var{gcc_switches}@r{]}
13895 @var{switches} specify the metrics to compute and define the destination for
13899 Each @var{filename} is the name (including the extension) of a source
13900 file to process. ``Wildcards'' are allowed, and
13901 the file name may contain path information.
13902 If no @var{filename} is supplied, then the @var{switches} list must contain
13904 @option{-files} switch (@pxref{Other gnatmetric Switches}).
13905 Including both a @option{-files} switch and one or more
13906 @var{filename} arguments is permitted.
13909 @samp{@var{gcc_switches}} is a list of switches for
13910 @command{gcc}. They will be passed on to all compiler invocations made by
13911 @command{gnatmetric} to generate the ASIS trees. Here you can provide
13912 @option{^-I^/INCLUDE_DIRS=^} switches to form the source search path,
13913 and use the @option{-gnatec} switch to set the configuration file,
13914 use the @option{-gnat05} switch if sources should be compiled in
13919 * Switches for gnatmetric::
13922 @node Switches for gnatmetric
13923 @section Switches for @command{gnatmetric}
13926 The following subsections describe the various switches accepted by
13927 @command{gnatmetric}, organized by category.
13930 * Output Files Control::
13931 * Disable Metrics For Local Units::
13932 * Specifying a set of metrics to compute::
13933 * Other gnatmetric Switches::
13934 * Generate project-wide metrics::
13937 @node Output Files Control
13938 @subsection Output File Control
13939 @cindex Output file control in @command{gnatmetric}
13942 @command{gnatmetric} has two output formats. It can generate a
13943 textual (human-readable) form, and also XML. By default only textual
13944 output is generated.
13946 When generating the output in textual form, @command{gnatmetric} creates
13947 for each Ada source file a corresponding text file
13948 containing the computed metrics, except for the case when the set of metrics
13949 specified by gnatmetric parameters consists only of metrics that are computed
13950 for the whole set of analyzed sources, but not for each Ada source.
13951 By default, this file is placed in the same directory as where the source
13952 file is located, and its name is obtained
13953 by appending the ^@file{.metrix}^@file{$METRIX}^ suffix to the name of the
13956 All the output information generated in XML format is placed in a single
13957 file. By default this file is placed in the current directory and has the
13958 name ^@file{metrix.xml}^@file{METRIX$XML}^.
13960 Some of the computed metrics are summed over the units passed to
13961 @command{gnatmetric}; for example, the total number of lines of code.
13962 By default this information is sent to @file{stdout}, but a file
13963 can be specified with the @option{-og} switch.
13965 The following switches control the @command{gnatmetric} output:
13968 @cindex @option{^-x^/XML^} (@command{gnatmetric})
13970 Generate the XML output
13972 @cindex @option{^-xs^/XSD^} (@command{gnatmetric})
13974 Generate the XML output and the XML schema file that describes the structure
13975 of the XML metric report, this schema is assigned to the XML file. The schema
13976 file has the same name as the XML output file with @file{.xml} suffix replaced
13979 @cindex @option{^-nt^/NO_TEXT^} (@command{gnatmetric})
13980 @item ^-nt^/NO_TEXT^
13981 Do not generate the output in text form (implies @option{^-x^/XML^})
13983 @cindex @option{^-d^/DIRECTORY^} (@command{gnatmetric})
13984 @item ^-d @var{output_dir}^/DIRECTORY=@var{output_dir}^
13985 Put text files with detailed metrics into @var{output_dir}
13987 @cindex @option{^-o^/SUFFIX_DETAILS^} (@command{gnatmetric})
13988 @item ^-o @var{file_suffix}^/SUFFIX_DETAILS=@var{file_suffix}^
13989 Use @var{file_suffix}, instead of ^@file{.metrix}^@file{$METRIX}^
13990 in the name of the output file.
13992 @cindex @option{^-og^/GLOBAL_OUTPUT^} (@command{gnatmetric})
13993 @item ^-og @var{file_name}^/GLOBAL_OUTPUT=@var{file_name}^
13994 Put global metrics into @var{file_name}
13996 @cindex @option{^-ox^/XML_OUTPUT^} (@command{gnatmetric})
13997 @item ^-ox @var{file_name}^/XML_OUTPUT=@var{file_name}^
13998 Put the XML output into @var{file_name} (also implies @option{^-x^/XML^})
14000 @cindex @option{^-sfn^/SHORT_SOURCE_FILE_NAME^} (@command{gnatmetric})
14001 @item ^-sfn^/SHORT_SOURCE_FILE_NAME^
14002 Use ``short'' source file names in the output. (The @command{gnatmetric}
14003 output includes the name(s) of the Ada source file(s) from which the metrics
14004 are computed. By default each name includes the absolute path. The
14005 @option{^-sfn^/SHORT_SOURCE_FILE_NAME^} switch causes @command{gnatmetric}
14006 to exclude all directory information from the file names that are output.)
14010 @node Disable Metrics For Local Units
14011 @subsection Disable Metrics For Local Units
14012 @cindex Disable Metrics For Local Units in @command{gnatmetric}
14015 @command{gnatmetric} relies on the GNAT compilation model @minus{}
14017 unit per one source file. It computes line metrics for the whole source
14018 file, and it also computes syntax
14019 and complexity metrics for the file's outermost unit.
14021 By default, @command{gnatmetric} will also compute all metrics for certain
14022 kinds of locally declared program units:
14026 subprogram (and generic subprogram) bodies;
14029 package (and generic package) specs and bodies;
14032 task object and type specifications and bodies;
14035 protected object and type specifications and bodies.
14039 These kinds of entities will be referred to as
14040 @emph{eligible local program units}, or simply @emph{eligible local units},
14041 @cindex Eligible local unit (for @command{gnatmetric})
14042 in the discussion below.
14044 Note that a subprogram declaration, generic instantiation,
14045 or renaming declaration only receives metrics
14046 computation when it appear as the outermost entity
14049 Suppression of metrics computation for eligible local units can be
14050 obtained via the following switch:
14053 @cindex @option{^-n@var{x}^/SUPPRESS^} (@command{gnatmetric})
14054 @item ^-nolocal^/SUPPRESS=LOCAL_DETAILS^
14055 Do not compute detailed metrics for eligible local program units
14059 @node Specifying a set of metrics to compute
14060 @subsection Specifying a set of metrics to compute
14063 By default all the metrics are computed and reported. The switches
14064 described in this subsection allow you to control, on an individual
14065 basis, whether metrics are computed and
14066 reported. If at least one positive metric
14067 switch is specified (that is, a switch that defines that a given
14068 metric or set of metrics is to be computed), then only
14069 explicitly specified metrics are reported.
14072 * Line Metrics Control::
14073 * Syntax Metrics Control::
14074 * Complexity Metrics Control::
14075 * Object-Oriented Metrics Control::
14078 @node Line Metrics Control
14079 @subsubsection Line Metrics Control
14080 @cindex Line metrics control in @command{gnatmetric}
14083 For any (legal) source file, and for each of its
14084 eligible local program units, @command{gnatmetric} computes the following
14089 the total number of lines;
14092 the total number of code lines (i.e., non-blank lines that are not comments)
14095 the number of comment lines
14098 the number of code lines containing end-of-line comments;
14101 the comment percentage: the ratio between the number of lines that contain
14102 comments and the number of all non-blank lines, expressed as a percentage;
14105 the number of empty lines and lines containing only space characters and/or
14106 format effectors (blank lines)
14109 the average number of code lines in subprogram bodies, task bodies, entry
14110 bodies and statement sequences in package bodies (this metric is only computed
14111 across the whole set of the analyzed units)
14116 @command{gnatmetric} sums the values of the line metrics for all the
14117 files being processed and then generates the cumulative results. The tool
14118 also computes for all the files being processed the average number of code
14121 You can use the following switches to select the specific line metrics
14122 to be computed and reported.
14125 @cindex @option{^--lines@var{x}^/LINE_COUNT_METRICS^} (@command{gnatmetric})
14128 @cindex @option{--no-lines@var{x}}
14131 @item ^--lines-all^/LINE_COUNT_METRICS=ALL^
14132 Report all the line metrics
14134 @item ^--no-lines-all^/LINE_COUNT_METRICS=NONE^
14135 Do not report any of line metrics
14137 @item ^--lines^/LINE_COUNT_METRICS=ALL_LINES^
14138 Report the number of all lines
14140 @item ^--no-lines^/LINE_COUNT_METRICS=NOALL_LINES^
14141 Do not report the number of all lines
14143 @item ^--lines-code^/LINE_COUNT_METRICS=CODE_LINES^
14144 Report the number of code lines
14146 @item ^--no-lines-code^/LINE_COUNT_METRICS=NOCODE_LINES^
14147 Do not report the number of code lines
14149 @item ^--lines-comment^/LINE_COUNT_METRICS=COMMENT_LINES^
14150 Report the number of comment lines
14152 @item ^--no-lines-comment^/LINE_COUNT_METRICS=NOCOMMENT_LINES^
14153 Do not report the number of comment lines
14155 @item ^--lines-eol-comment^/LINE_COUNT_METRICS=CODE_COMMENT_LINES^
14156 Report the number of code lines containing
14157 end-of-line comments
14159 @item ^--no-lines-eol-comment^/LINE_COUNT_METRICS=NOCODE_COMMENT_LINES^
14160 Do not report the number of code lines containing
14161 end-of-line comments
14163 @item ^--lines-ratio^/LINE_COUNT_METRICS=COMMENT_PERCENTAGE^
14164 Report the comment percentage in the program text
14166 @item ^--no-lines-ratio^/LINE_COUNT_METRICS=NOCOMMENT_PERCENTAGE^
14167 Do not report the comment percentage in the program text
14169 @item ^--lines-blank^/LINE_COUNT_METRICS=BLANK_LINES^
14170 Report the number of blank lines
14172 @item ^--no-lines-blank^/LINE_COUNT_METRICS=NOBLANK_LINES^
14173 Do not report the number of blank lines
14175 @item ^--lines-average^/LINE_COUNT_METRICS=AVERAGE_BODY_LINES^
14176 Report the average number of code lines in subprogram bodies, task bodies,
14177 entry bodies and statement sequences in package bodies. The metric is computed
14178 and reported for the whole set of processed Ada sources only.
14180 @item ^--no-lines-average^/LINE_COUNT_METRICS=NOAVERAGE_BODY_LINES^
14181 Do not report the average number of code lines in subprogram bodies,
14182 task bodies, entry bodies and statement sequences in package bodies.
14186 @node Syntax Metrics Control
14187 @subsubsection Syntax Metrics Control
14188 @cindex Syntax metrics control in @command{gnatmetric}
14191 @command{gnatmetric} computes various syntactic metrics for the
14192 outermost unit and for each eligible local unit:
14195 @item LSLOC (``Logical Source Lines Of Code'')
14196 The total number of declarations and the total number of statements
14198 @item Maximal static nesting level of inner program units
14200 @cite{Ada Reference Manual}, 10.1(1), ``A program unit is either a
14201 package, a task unit, a protected unit, a
14202 protected entry, a generic unit, or an explicitly declared subprogram other
14203 than an enumeration literal.''
14205 @item Maximal nesting level of composite syntactic constructs
14206 This corresponds to the notion of the
14207 maximum nesting level in the GNAT built-in style checks
14208 (@pxref{Style Checking})
14212 For the outermost unit in the file, @command{gnatmetric} additionally computes
14213 the following metrics:
14216 @item Public subprograms
14217 This metric is computed for package specs. It is the
14218 number of subprograms and generic subprograms declared in the visible
14219 part (including the visible part of nested packages, protected objects, and
14222 @item All subprograms
14223 This metric is computed for bodies and subunits. The
14224 metric is equal to a total number of subprogram bodies in the compilation
14226 Neither generic instantiations nor renamings-as-a-body nor body stubs
14227 are counted. Any subprogram body is counted, independently of its nesting
14228 level and enclosing constructs. Generic bodies and bodies of protected
14229 subprograms are counted in the same way as ``usual'' subprogram bodies.
14232 This metric is computed for package specs and
14233 generic package declarations. It is the total number of types
14234 that can be referenced from outside this compilation unit, plus the
14235 number of types from all the visible parts of all the visible generic
14236 packages. Generic formal types are not counted. Only types, not subtypes,
14240 Along with the total number of public types, the following
14241 types are counted and reported separately:
14248 Root tagged types (abstract, non-abstract, private, non-private). Type
14249 extensions are @emph{not} counted
14252 Private types (including private extensions)
14263 This metric is computed for any compilation unit. It is equal to the total
14264 number of the declarations of different types given in the compilation unit.
14265 The private and the corresponding full type declaration are counted as one
14266 type declaration. Incomplete type declarations and generic formal types
14268 No distinction is made among different kinds of types (abstract,
14269 private etc.); the total number of types is computed and reported.
14274 By default, all the syntax metrics are computed and reported. You can use the
14275 following switches to select specific syntax metrics.
14279 @cindex @option{^--syntax@var{x}^/SYNTAX_METRICS^} (@command{gnatmetric})
14282 @cindex @option{--no-syntax@var{x}} (@command{gnatmetric})
14285 @item ^--syntax-all^/SYNTAX_METRICS=ALL^
14286 Report all the syntax metrics
14288 @item ^--no-syntax-all^/SYNTAX_METRICS=NONE^
14289 Do not report any of syntax metrics
14291 @item ^--declarations^/SYNTAX_METRICS=DECLARATIONS^
14292 Report the total number of declarations
14294 @item ^--no-declarations^/SYNTAX_METRICS=NODECLARATIONS^
14295 Do not report the total number of declarations
14297 @item ^--statements^/SYNTAX_METRICS=STATEMENTS^
14298 Report the total number of statements
14300 @item ^--no-statements^/SYNTAX_METRICS=NOSTATEMENTS^
14301 Do not report the total number of statements
14303 @item ^--public-subprograms^/SYNTAX_METRICS=PUBLIC_SUBPROGRAMS^
14304 Report the number of public subprograms in a compilation unit
14306 @item ^--no-public-subprograms^/SYNTAX_METRICS=NOPUBLIC_SUBPROGRAMS^
14307 Do not report the number of public subprograms in a compilation unit
14309 @item ^--all-subprograms^/SYNTAX_METRICS=ALL_SUBPROGRAMS^
14310 Report the number of all the subprograms in a compilation unit
14312 @item ^--no-all-subprograms^/SYNTAX_METRICS=NOALL_SUBPROGRAMS^
14313 Do not report the number of all the subprograms in a compilation unit
14315 @item ^--public-types^/SYNTAX_METRICS=PUBLIC_TYPES^
14316 Report the number of public types in a compilation unit
14318 @item ^--no-public-types^/SYNTAX_METRICS=NOPUBLIC_TYPES^
14319 Do not report the number of public types in a compilation unit
14321 @item ^--all-types^/SYNTAX_METRICS=ALL_TYPES^
14322 Report the number of all the types in a compilation unit
14324 @item ^--no-all-types^/SYNTAX_METRICS=NOALL_TYPES^
14325 Do not report the number of all the types in a compilation unit
14327 @item ^--unit-nesting^/SYNTAX_METRICS=UNIT_NESTING^
14328 Report the maximal program unit nesting level
14330 @item ^--no-unit-nesting^/SYNTAX_METRICS=UNIT_NESTING_OFF^
14331 Do not report the maximal program unit nesting level
14333 @item ^--construct-nesting^/SYNTAX_METRICS=CONSTRUCT_NESTING^
14334 Report the maximal construct nesting level
14336 @item ^--no-construct-nesting^/SYNTAX_METRICS=NOCONSTRUCT_NESTING^
14337 Do not report the maximal construct nesting level
14341 @node Complexity Metrics Control
14342 @subsubsection Complexity Metrics Control
14343 @cindex Complexity metrics control in @command{gnatmetric}
14346 For a program unit that is an executable body (a subprogram body (including
14347 generic bodies), task body, entry body or a package body containing
14348 its own statement sequence) @command{gnatmetric} computes the following
14349 complexity metrics:
14353 McCabe cyclomatic complexity;
14356 McCabe essential complexity;
14359 maximal loop nesting level
14364 The McCabe complexity metrics are defined
14365 in @url{http://www.mccabe.com/pdf/nist235r.pdf}
14367 According to McCabe, both control statements and short-circuit control forms
14368 should be taken into account when computing cyclomatic complexity. For each
14369 body, we compute three metric values:
14373 the complexity introduced by control
14374 statements only, without taking into account short-circuit forms,
14377 the complexity introduced by short-circuit control forms only, and
14381 cyclomatic complexity, which is the sum of these two values.
14385 When computing cyclomatic and essential complexity, @command{gnatmetric} skips
14386 the code in the exception handlers and in all the nested program units.
14388 By default, all the complexity metrics are computed and reported.
14389 For more fine-grained control you can use
14390 the following switches:
14393 @cindex @option{^-complexity@var{x}^/COMPLEXITY_METRICS^} (@command{gnatmetric})
14396 @cindex @option{--no-complexity@var{x}}
14399 @item ^--complexity-all^/COMPLEXITY_METRICS=ALL^
14400 Report all the complexity metrics
14402 @item ^--no-complexity-all^/COMPLEXITY_METRICS=NONE^
14403 Do not report any of complexity metrics
14405 @item ^--complexity-cyclomatic^/COMPLEXITY_METRICS=CYCLOMATIC^
14406 Report the McCabe Cyclomatic Complexity
14408 @item ^--no-complexity-cyclomatic^/COMPLEXITY_METRICS=NOCYCLOMATIC^
14409 Do not report the McCabe Cyclomatic Complexity
14411 @item ^--complexity-essential^/COMPLEXITY_METRICS=ESSENTIAL^
14412 Report the Essential Complexity
14414 @item ^--no-complexity-essential^/COMPLEXITY_METRICS=NOESSENTIAL^
14415 Do not report the Essential Complexity
14417 @item ^--loop-nesting^/COMPLEXITY_METRICS=LOOP_NESTING_ON^
14418 Report maximal loop nesting level
14420 @item ^--no-loop-nesting^/COMPLEXITY_METRICS=NOLOOP_NESTING^
14421 Do not report maximal loop nesting level
14423 @item ^--complexity-average^/COMPLEXITY_METRICS=AVERAGE_COMPLEXITY^
14424 Report the average McCabe Cyclomatic Complexity for all the subprogram bodies,
14425 task bodies, entry bodies and statement sequences in package bodies.
14426 The metric is computed and reported for whole set of processed Ada sources
14429 @item ^--no-complexity-average^/COMPLEXITY_METRICS=NOAVERAGE_COMPLEXITY^
14430 Do not report the average McCabe Cyclomatic Complexity for all the subprogram
14431 bodies, task bodies, entry bodies and statement sequences in package bodies
14433 @cindex @option{^-ne^/NO_EXITS_AS_GOTOS^} (@command{gnatmetric})
14434 @item ^-ne^/NO_EXITS_AS_GOTOS^
14435 Do not consider @code{exit} statements as @code{goto}s when
14436 computing Essential Complexity
14438 @item ^--extra-exit-points^/EXTRA_EXIT_POINTS^
14439 Report the extra exit points for subprogram bodies. As an exit point, this
14440 metric counts @code{return} statements and raise statements in case when the
14441 raised exception is not handled in the same body. In case of a function this
14442 metric subtracts 1 from the number of exit points, because a function body
14443 must contain at least one @code{return} statement.
14445 @item ^--no-extra-exit-points^/NOEXTRA_EXIT_POINTS^
14446 Do not report the extra exit points for subprogram bodies
14450 @node Object-Oriented Metrics Control
14451 @subsubsection Object-Oriented Metrics Control
14452 @cindex Object-Oriented metrics control in @command{gnatmetric}
14455 @cindex Coupling metrics (in in @command{gnatmetric})
14456 Coupling metrics are object-oriented metrics that measure the
14457 dependencies between a given class (or a group of classes) and the
14458 ``external world'' (that is, the other classes in the program). In this
14459 subsection the term ``class'' is used in its
14460 traditional object-oriented programming sense
14461 (an instantiable module that contains data and/or method members).
14462 A @emph{category} (of classes)
14463 is a group of closely related classes that are reused and/or
14466 A class @code{K}'s @emph{efferent coupling} is the number of classes
14467 that @code{K} depends upon.
14468 A category's efferent coupling is the number of classes outside the
14469 category that the classes inside the category depend upon.
14471 A class @code{K}'s @emph{afferent coupling} is the number of classes
14472 that depend upon @code{K}.
14473 A category's afferent coupling is the number of classes outside the
14474 category that depend on classes belonging to the category.
14476 Ada's implementation of the object-oriented paradigm does not use the
14477 traditional class notion, so the definition of the coupling
14478 metrics for Ada maps the class and class category notions
14479 onto Ada constructs.
14481 For the coupling metrics, several kinds of modules -- a library package,
14482 a library generic package, and a library generic package instantiation --
14483 that define a tagged type or an interface type are
14484 considered to be a class. A category consists of a library package (or
14485 a library generic package) that defines a tagged or an interface type,
14486 together with all its descendant (generic) packages that define tagged
14487 or interface types. For any package counted as a class,
14488 its body and subunits (if any) are considered
14489 together with its spec when counting the dependencies, and coupling
14490 metrics are reported for spec units only. For dependencies
14491 between classes, the Ada semantic dependencies are considered.
14492 For coupling metrics, only dependencies on units that are considered as
14493 classes, are considered.
14495 When computing coupling metrics, @command{gnatmetric} counts only
14496 dependencies between units that are arguments of the gnatmetric call.
14497 Coupling metrics are program-wide (or project-wide) metrics, so to
14498 get a valid result, you should call @command{gnatmetric} for
14499 the whole set of sources that make up your program. It can be done
14500 by calling @command{gnatmetric} from the GNAT driver with @option{-U}
14501 option (see See @ref{The GNAT Driver and Project Files} for details.
14503 By default, all the coupling metrics are disabled. You can use the following
14504 switches to specify the coupling metrics to be computed and reported:
14509 @cindex @option{--package@var{x}} (@command{gnatmetric})
14510 @cindex @option{--no-package@var{x}} (@command{gnatmetric})
14511 @cindex @option{--category@var{x}} (@command{gnatmetric})
14512 @cindex @option{--no-category@var{x}} (@command{gnatmetric})
14516 @cindex @option{/COUPLING_METRICS} (@command{gnatmetric})
14519 @item ^--coupling-all^/COUPLING_METRICS=ALL^
14520 Report all the coupling metrics
14522 @item ^--no-coupling-all^/COUPLING_METRICS=NONE^
14523 Do not report any of metrics
14525 @item ^--package-efferent-coupling^/COUPLING_METRICS=PACKAGE_EFFERENT^
14526 Report package efferent coupling
14528 @item ^--no-package-efferent-coupling^/COUPLING_METRICS=NOPACKAGE_EFFERENT^
14529 Do not report package efferent coupling
14531 @item ^--package-afferent-coupling^/COUPLING_METRICS=PACKAGE_AFFERENT^
14532 Report package afferent coupling
14534 @item ^--no-package-afferent-coupling^/COUPLING_METRICS=NOPACKAGE_AFFERENT^
14535 Do not report package afferent coupling
14537 @item ^--category-efferent-coupling^/COUPLING_METRICS=CATEGORY_EFFERENT^
14538 Report category efferent coupling
14540 @item ^--no-category-efferent-coupling^/COUPLING_METRICS=NOCATEGORY_EFFERENT^
14541 Do not report category efferent coupling
14543 @item ^--category-afferent-coupling^/COUPLING_METRICS=CATEGORY_AFFERENT^
14544 Report category afferent coupling
14546 @item ^--no-category-afferent-coupling^/COUPLING_METRICS=NOCATEGORY_AFFERENT^
14547 Do not report category afferent coupling
14551 @node Other gnatmetric Switches
14552 @subsection Other @code{gnatmetric} Switches
14555 Additional @command{gnatmetric} switches are as follows:
14558 @item ^-files @var{filename}^/FILES=@var{filename}^
14559 @cindex @option{^-files^/FILES^} (@code{gnatmetric})
14560 Take the argument source files from the specified file. This file should be an
14561 ordinary text file containing file names separated by spaces or
14562 line breaks. You can use this switch more than once in the same call to
14563 @command{gnatmetric}. You also can combine this switch with
14564 an explicit list of files.
14566 @item ^-v^/VERBOSE^
14567 @cindex @option{^-v^/VERBOSE^} (@code{gnatmetric})
14569 @command{gnatmetric} generates version information and then
14570 a trace of sources being processed.
14573 @cindex @option{^-q^/QUIET^} (@code{gnatmetric})
14577 @node Generate project-wide metrics
14578 @subsection Generate project-wide metrics
14580 In order to compute metrics on all units of a given project, you can use
14581 the @command{gnat} driver along with the @option{-P} option:
14587 If the project @code{proj} depends upon other projects, you can compute
14588 the metrics on the project closure using the @option{-U} option:
14590 gnat metric -Pproj -U
14594 Finally, if not all the units are relevant to a particular main
14595 program in the project closure, you can generate metrics for the set
14596 of units needed to create a given main program (unit closure) using
14597 the @option{-U} option followed by the name of the main unit:
14599 gnat metric -Pproj -U main
14603 @c ***********************************
14604 @node File Name Krunching Using gnatkr
14605 @chapter File Name Krunching Using @code{gnatkr}
14609 This chapter discusses the method used by the compiler to shorten
14610 the default file names chosen for Ada units so that they do not
14611 exceed the maximum length permitted. It also describes the
14612 @code{gnatkr} utility that can be used to determine the result of
14613 applying this shortening.
14617 * Krunching Method::
14618 * Examples of gnatkr Usage::
14622 @section About @code{gnatkr}
14625 The default file naming rule in GNAT
14626 is that the file name must be derived from
14627 the unit name. The exact default rule is as follows:
14630 Take the unit name and replace all dots by hyphens.
14632 If such a replacement occurs in the
14633 second character position of a name, and the first character is
14634 ^@samp{a}, @samp{g}, @samp{s}, or @samp{i}, ^@samp{A}, @samp{G}, @samp{S}, or @samp{I},^
14635 then replace the dot by the character
14636 ^@samp{~} (tilde)^@samp{$} (dollar sign)^
14637 instead of a minus.
14639 The reason for this exception is to avoid clashes
14640 with the standard names for children of System, Ada, Interfaces,
14641 and GNAT, which use the prefixes
14642 ^@samp{s-}, @samp{a-}, @samp{i-}, and @samp{g-},^@samp{S-}, @samp{A-}, @samp{I-}, and @samp{G-},^
14645 The @option{^-gnatk^/FILE_NAME_MAX_LENGTH=^@var{nn}}
14646 switch of the compiler activates a ``krunching''
14647 circuit that limits file names to nn characters (where nn is a decimal
14648 integer). For example, using OpenVMS,
14649 where the maximum file name length is
14650 39, the value of nn is usually set to 39, but if you want to generate
14651 a set of files that would be usable if ported to a system with some
14652 different maximum file length, then a different value can be specified.
14653 The default value of 39 for OpenVMS need not be specified.
14655 The @code{gnatkr} utility can be used to determine the krunched name for
14656 a given file, when krunched to a specified maximum length.
14659 @section Using @code{gnatkr}
14662 The @code{gnatkr} command has the form
14666 @c $ gnatkr @var{name} @ovar{length}
14667 @c Expanding @ovar macro inline (explanation in macro def comments)
14668 $ gnatkr @var{name} @r{[}@var{length}@r{]}
14674 $ gnatkr @var{name} /COUNT=nn
14679 @var{name} is the uncrunched file name, derived from the name of the unit
14680 in the standard manner described in the previous section (i.e., in particular
14681 all dots are replaced by hyphens). The file name may or may not have an
14682 extension (defined as a suffix of the form period followed by arbitrary
14683 characters other than period). If an extension is present then it will
14684 be preserved in the output. For example, when krunching @file{hellofile.ads}
14685 to eight characters, the result will be hellofil.ads.
14687 Note: for compatibility with previous versions of @code{gnatkr} dots may
14688 appear in the name instead of hyphens, but the last dot will always be
14689 taken as the start of an extension. So if @code{gnatkr} is given an argument
14690 such as @file{Hello.World.adb} it will be treated exactly as if the first
14691 period had been a hyphen, and for example krunching to eight characters
14692 gives the result @file{hellworl.adb}.
14694 Note that the result is always all lower case (except on OpenVMS where it is
14695 all upper case). Characters of the other case are folded as required.
14697 @var{length} represents the length of the krunched name. The default
14698 when no argument is given is ^8^39^ characters. A length of zero stands for
14699 unlimited, in other words do not chop except for system files where the
14700 implied crunching length is always eight characters.
14703 The output is the krunched name. The output has an extension only if the
14704 original argument was a file name with an extension.
14706 @node Krunching Method
14707 @section Krunching Method
14710 The initial file name is determined by the name of the unit that the file
14711 contains. The name is formed by taking the full expanded name of the
14712 unit and replacing the separating dots with hyphens and
14713 using ^lowercase^uppercase^
14714 for all letters, except that a hyphen in the second character position is
14715 replaced by a ^tilde^dollar sign^ if the first character is
14716 ^@samp{a}, @samp{i}, @samp{g}, or @samp{s}^@samp{A}, @samp{I}, @samp{G}, or @samp{S}^.
14717 The extension is @code{.ads} for a
14718 spec and @code{.adb} for a body.
14719 Krunching does not affect the extension, but the file name is shortened to
14720 the specified length by following these rules:
14724 The name is divided into segments separated by hyphens, tildes or
14725 underscores and all hyphens, tildes, and underscores are
14726 eliminated. If this leaves the name short enough, we are done.
14729 If the name is too long, the longest segment is located (left-most
14730 if there are two of equal length), and shortened by dropping
14731 its last character. This is repeated until the name is short enough.
14733 As an example, consider the krunching of @*@file{our-strings-wide_fixed.adb}
14734 to fit the name into 8 characters as required by some operating systems.
14737 our-strings-wide_fixed 22
14738 our strings wide fixed 19
14739 our string wide fixed 18
14740 our strin wide fixed 17
14741 our stri wide fixed 16
14742 our stri wide fixe 15
14743 our str wide fixe 14
14744 our str wid fixe 13
14750 Final file name: oustwifi.adb
14754 The file names for all predefined units are always krunched to eight
14755 characters. The krunching of these predefined units uses the following
14756 special prefix replacements:
14760 replaced by @file{^a^A^-}
14763 replaced by @file{^g^G^-}
14766 replaced by @file{^i^I^-}
14769 replaced by @file{^s^S^-}
14772 These system files have a hyphen in the second character position. That
14773 is why normal user files replace such a character with a
14774 ^tilde^dollar sign^, to
14775 avoid confusion with system file names.
14777 As an example of this special rule, consider
14778 @*@file{ada-strings-wide_fixed.adb}, which gets krunched as follows:
14781 ada-strings-wide_fixed 22
14782 a- strings wide fixed 18
14783 a- string wide fixed 17
14784 a- strin wide fixed 16
14785 a- stri wide fixed 15
14786 a- stri wide fixe 14
14787 a- str wide fixe 13
14793 Final file name: a-stwifi.adb
14797 Of course no file shortening algorithm can guarantee uniqueness over all
14798 possible unit names, and if file name krunching is used then it is your
14799 responsibility to ensure that no name clashes occur. The utility
14800 program @code{gnatkr} is supplied for conveniently determining the
14801 krunched name of a file.
14803 @node Examples of gnatkr Usage
14804 @section Examples of @code{gnatkr} Usage
14811 $ gnatkr very_long_unit_name.ads --> velounna.ads
14812 $ gnatkr grandparent-parent-child.ads --> grparchi.ads
14813 $ gnatkr Grandparent.Parent.Child.ads --> grparchi.ads
14814 $ gnatkr grandparent-parent-child --> grparchi
14816 $ gnatkr very_long_unit_name.ads/count=6 --> vlunna.ads
14817 $ gnatkr very_long_unit_name.ads/count=0 --> very_long_unit_name.ads
14820 @node Preprocessing Using gnatprep
14821 @chapter Preprocessing Using @code{gnatprep}
14825 This chapter discusses how to use GNAT's @code{gnatprep} utility for simple
14827 Although designed for use with GNAT, @code{gnatprep} does not depend on any
14828 special GNAT features.
14829 For further discussion of conditional compilation in general, see
14830 @ref{Conditional Compilation}.
14833 * Preprocessing Symbols::
14835 * Switches for gnatprep::
14836 * Form of Definitions File::
14837 * Form of Input Text for gnatprep::
14840 @node Preprocessing Symbols
14841 @section Preprocessing Symbols
14844 Preprocessing symbols are defined in definition files and referred to in
14845 sources to be preprocessed. A Preprocessing symbol is an identifier, following
14846 normal Ada (case-insensitive) rules for its syntax, with the restriction that
14847 all characters need to be in the ASCII set (no accented letters).
14849 @node Using gnatprep
14850 @section Using @code{gnatprep}
14853 To call @code{gnatprep} use
14856 @c $ gnatprep @ovar{switches} @var{infile} @var{outfile} @ovar{deffile}
14857 @c Expanding @ovar macro inline (explanation in macro def comments)
14858 $ gnatprep @r{[}@var{switches}@r{]} @var{infile} @var{outfile} @r{[}@var{deffile}@r{]}
14865 is an optional sequence of switches as described in the next section.
14868 is the full name of the input file, which is an Ada source
14869 file containing preprocessor directives.
14872 is the full name of the output file, which is an Ada source
14873 in standard Ada form. When used with GNAT, this file name will
14874 normally have an ads or adb suffix.
14877 is the full name of a text file containing definitions of
14878 preprocessing symbols to be referenced by the preprocessor. This argument is
14879 optional, and can be replaced by the use of the @option{-D} switch.
14883 @node Switches for gnatprep
14884 @section Switches for @code{gnatprep}
14889 @item ^-b^/BLANK_LINES^
14890 @cindex @option{^-b^/BLANK_LINES^} (@command{gnatprep})
14891 Causes both preprocessor lines and the lines deleted by
14892 preprocessing to be replaced by blank lines in the output source file,
14893 preserving line numbers in the output file.
14895 @item ^-c^/COMMENTS^
14896 @cindex @option{^-c^/COMMENTS^} (@command{gnatprep})
14897 Causes both preprocessor lines and the lines deleted
14898 by preprocessing to be retained in the output source as comments marked
14899 with the special string @code{"--! "}. This option will result in line numbers
14900 being preserved in the output file.
14902 @item ^-C^/REPLACE_IN_COMMENTS^
14903 @cindex @option{^-C^/REPLACE_IN_COMMENTS^} (@command{gnatprep})
14904 Causes comments to be scanned. Normally comments are ignored by gnatprep.
14905 If this option is specified, then comments are scanned and any $symbol
14906 substitutions performed as in program text. This is particularly useful
14907 when structured comments are used (e.g., when writing programs in the
14908 SPARK dialect of Ada). Note that this switch is not available when
14909 doing integrated preprocessing (it would be useless in this context
14910 since comments are ignored by the compiler in any case).
14912 @item ^-Dsymbol=value^/ASSOCIATE="symbol=value"^
14913 @cindex @option{^-D^/ASSOCIATE^} (@command{gnatprep})
14914 Defines a new preprocessing symbol, associated with value. If no value is given
14915 on the command line, then symbol is considered to be @code{True}. This switch
14916 can be used in place of a definition file.
14920 @cindex @option{/REMOVE} (@command{gnatprep})
14921 This is the default setting which causes lines deleted by preprocessing
14922 to be entirely removed from the output file.
14925 @item ^-r^/REFERENCE^
14926 @cindex @option{^-r^/REFERENCE^} (@command{gnatprep})
14927 Causes a @code{Source_Reference} pragma to be generated that
14928 references the original input file, so that error messages will use
14929 the file name of this original file. The use of this switch implies
14930 that preprocessor lines are not to be removed from the file, so its
14931 use will force @option{^-b^/BLANK_LINES^} mode if
14932 @option{^-c^/COMMENTS^}
14933 has not been specified explicitly.
14935 Note that if the file to be preprocessed contains multiple units, then
14936 it will be necessary to @code{gnatchop} the output file from
14937 @code{gnatprep}. If a @code{Source_Reference} pragma is present
14938 in the preprocessed file, it will be respected by
14939 @code{gnatchop ^-r^/REFERENCE^}
14940 so that the final chopped files will correctly refer to the original
14941 input source file for @code{gnatprep}.
14943 @item ^-s^/SYMBOLS^
14944 @cindex @option{^-s^/SYMBOLS^} (@command{gnatprep})
14945 Causes a sorted list of symbol names and values to be
14946 listed on the standard output file.
14948 @item ^-u^/UNDEFINED^
14949 @cindex @option{^-u^/UNDEFINED^} (@command{gnatprep})
14950 Causes undefined symbols to be treated as having the value FALSE in the context
14951 of a preprocessor test. In the absence of this option, an undefined symbol in
14952 a @code{#if} or @code{#elsif} test will be treated as an error.
14958 Note: if neither @option{-b} nor @option{-c} is present,
14959 then preprocessor lines and
14960 deleted lines are completely removed from the output, unless -r is
14961 specified, in which case -b is assumed.
14964 @node Form of Definitions File
14965 @section Form of Definitions File
14968 The definitions file contains lines of the form
14975 where symbol is a preprocessing symbol, and value is one of the following:
14979 Empty, corresponding to a null substitution
14981 A string literal using normal Ada syntax
14983 Any sequence of characters from the set
14984 (letters, digits, period, underline).
14988 Comment lines may also appear in the definitions file, starting with
14989 the usual @code{--},
14990 and comments may be added to the definitions lines.
14992 @node Form of Input Text for gnatprep
14993 @section Form of Input Text for @code{gnatprep}
14996 The input text may contain preprocessor conditional inclusion lines,
14997 as well as general symbol substitution sequences.
14999 The preprocessor conditional inclusion commands have the form
15004 #if @i{expression} @r{[}then@r{]}
15006 #elsif @i{expression} @r{[}then@r{]}
15008 #elsif @i{expression} @r{[}then@r{]}
15019 In this example, @i{expression} is defined by the following grammar:
15021 @i{expression} ::= <symbol>
15022 @i{expression} ::= <symbol> = "<value>"
15023 @i{expression} ::= <symbol> = <symbol>
15024 @i{expression} ::= <symbol> 'Defined
15025 @i{expression} ::= not @i{expression}
15026 @i{expression} ::= @i{expression} and @i{expression}
15027 @i{expression} ::= @i{expression} or @i{expression}
15028 @i{expression} ::= @i{expression} and then @i{expression}
15029 @i{expression} ::= @i{expression} or else @i{expression}
15030 @i{expression} ::= ( @i{expression} )
15033 The following restriction exists: it is not allowed to have "and" or "or"
15034 following "not" in the same expression without parentheses. For example, this
15041 This should be one of the following:
15049 For the first test (@i{expression} ::= <symbol>) the symbol must have
15050 either the value true or false, that is to say the right-hand of the
15051 symbol definition must be one of the (case-insensitive) literals
15052 @code{True} or @code{False}. If the value is true, then the
15053 corresponding lines are included, and if the value is false, they are
15056 The test (@i{expression} ::= <symbol> @code{'Defined}) is true only if
15057 the symbol has been defined in the definition file or by a @option{-D}
15058 switch on the command line. Otherwise, the test is false.
15060 The equality tests are case insensitive, as are all the preprocessor lines.
15062 If the symbol referenced is not defined in the symbol definitions file,
15063 then the effect depends on whether or not switch @option{-u}
15064 is specified. If so, then the symbol is treated as if it had the value
15065 false and the test fails. If this switch is not specified, then
15066 it is an error to reference an undefined symbol. It is also an error to
15067 reference a symbol that is defined with a value other than @code{True}
15070 The use of the @code{not} operator inverts the sense of this logical test.
15071 The @code{not} operator cannot be combined with the @code{or} or @code{and}
15072 operators, without parentheses. For example, "if not X or Y then" is not
15073 allowed, but "if (not X) or Y then" and "if not (X or Y) then" are.
15075 The @code{then} keyword is optional as shown
15077 The @code{#} must be the first non-blank character on a line, but
15078 otherwise the format is free form. Spaces or tabs may appear between
15079 the @code{#} and the keyword. The keywords and the symbols are case
15080 insensitive as in normal Ada code. Comments may be used on a
15081 preprocessor line, but other than that, no other tokens may appear on a
15082 preprocessor line. Any number of @code{elsif} clauses can be present,
15083 including none at all. The @code{else} is optional, as in Ada.
15085 The @code{#} marking the start of a preprocessor line must be the first
15086 non-blank character on the line, i.e., it must be preceded only by
15087 spaces or horizontal tabs.
15089 Symbol substitution outside of preprocessor lines is obtained by using
15097 anywhere within a source line, except in a comment or within a
15098 string literal. The identifier
15099 following the @code{$} must match one of the symbols defined in the symbol
15100 definition file, and the result is to substitute the value of the
15101 symbol in place of @code{$symbol} in the output file.
15103 Note that although the substitution of strings within a string literal
15104 is not possible, it is possible to have a symbol whose defined value is
15105 a string literal. So instead of setting XYZ to @code{hello} and writing:
15108 Header : String := "$XYZ";
15112 you should set XYZ to @code{"hello"} and write:
15115 Header : String := $XYZ;
15119 and then the substitution will occur as desired.
15121 @node The GNAT Library Browser gnatls
15122 @chapter The GNAT Library Browser @code{gnatls}
15124 @cindex Library browser
15127 @code{gnatls} is a tool that outputs information about compiled
15128 units. It gives the relationship between objects, unit names and source
15129 files. It can also be used to check the source dependencies of a unit
15130 as well as various characteristics.
15132 Note: to invoke @code{gnatls} with a project file, use the @code{gnat}
15133 driver (see @ref{The GNAT Driver and Project Files}).
15137 * Switches for gnatls::
15138 * Examples of gnatls Usage::
15141 @node Running gnatls
15142 @section Running @code{gnatls}
15145 The @code{gnatls} command has the form
15148 $ gnatls switches @var{object_or_ali_file}
15152 The main argument is the list of object or @file{ali} files
15153 (@pxref{The Ada Library Information Files})
15154 for which information is requested.
15156 In normal mode, without additional option, @code{gnatls} produces a
15157 four-column listing. Each line represents information for a specific
15158 object. The first column gives the full path of the object, the second
15159 column gives the name of the principal unit in this object, the third
15160 column gives the status of the source and the fourth column gives the
15161 full path of the source representing this unit.
15162 Here is a simple example of use:
15166 ^./^[]^demo1.o demo1 DIF demo1.adb
15167 ^./^[]^demo2.o demo2 OK demo2.adb
15168 ^./^[]^hello.o h1 OK hello.adb
15169 ^./^[]^instr-child.o instr.child MOK instr-child.adb
15170 ^./^[]^instr.o instr OK instr.adb
15171 ^./^[]^tef.o tef DIF tef.adb
15172 ^./^[]^text_io_example.o text_io_example OK text_io_example.adb
15173 ^./^[]^tgef.o tgef DIF tgef.adb
15177 The first line can be interpreted as follows: the main unit which is
15179 object file @file{demo1.o} is demo1, whose main source is in
15180 @file{demo1.adb}. Furthermore, the version of the source used for the
15181 compilation of demo1 has been modified (DIF). Each source file has a status
15182 qualifier which can be:
15185 @item OK (unchanged)
15186 The version of the source file used for the compilation of the
15187 specified unit corresponds exactly to the actual source file.
15189 @item MOK (slightly modified)
15190 The version of the source file used for the compilation of the
15191 specified unit differs from the actual source file but not enough to
15192 require recompilation. If you use gnatmake with the qualifier
15193 @option{^-m (minimal recompilation)^/MINIMAL_RECOMPILATION^}, a file marked
15194 MOK will not be recompiled.
15196 @item DIF (modified)
15197 No version of the source found on the path corresponds to the source
15198 used to build this object.
15200 @item ??? (file not found)
15201 No source file was found for this unit.
15203 @item HID (hidden, unchanged version not first on PATH)
15204 The version of the source that corresponds exactly to the source used
15205 for compilation has been found on the path but it is hidden by another
15206 version of the same source that has been modified.
15210 @node Switches for gnatls
15211 @section Switches for @code{gnatls}
15214 @code{gnatls} recognizes the following switches:
15218 @cindex @option{--version} @command{gnatls}
15219 Display Copyright and version, then exit disregarding all other options.
15222 @cindex @option{--help} @command{gnatls}
15223 If @option{--version} was not used, display usage, then exit disregarding
15226 @item ^-a^/ALL_UNITS^
15227 @cindex @option{^-a^/ALL_UNITS^} (@code{gnatls})
15228 Consider all units, including those of the predefined Ada library.
15229 Especially useful with @option{^-d^/DEPENDENCIES^}.
15231 @item ^-d^/DEPENDENCIES^
15232 @cindex @option{^-d^/DEPENDENCIES^} (@code{gnatls})
15233 List sources from which specified units depend on.
15235 @item ^-h^/OUTPUT=OPTIONS^
15236 @cindex @option{^-h^/OUTPUT=OPTIONS^} (@code{gnatls})
15237 Output the list of options.
15239 @item ^-o^/OUTPUT=OBJECTS^
15240 @cindex @option{^-o^/OUTPUT=OBJECTS^} (@code{gnatls})
15241 Only output information about object files.
15243 @item ^-s^/OUTPUT=SOURCES^
15244 @cindex @option{^-s^/OUTPUT=SOURCES^} (@code{gnatls})
15245 Only output information about source files.
15247 @item ^-u^/OUTPUT=UNITS^
15248 @cindex @option{^-u^/OUTPUT=UNITS^} (@code{gnatls})
15249 Only output information about compilation units.
15251 @item ^-files^/FILES^=@var{file}
15252 @cindex @option{^-files^/FILES^} (@code{gnatls})
15253 Take as arguments the files listed in text file @var{file}.
15254 Text file @var{file} may contain empty lines that are ignored.
15255 Each nonempty line should contain the name of an existing file.
15256 Several such switches may be specified simultaneously.
15258 @item ^-aO^/OBJECT_SEARCH=^@var{dir}
15259 @itemx ^-aI^/SOURCE_SEARCH=^@var{dir}
15260 @itemx ^-I^/SEARCH=^@var{dir}
15261 @itemx ^-I-^/NOCURRENT_DIRECTORY^
15263 @cindex @option{^-aO^/OBJECT_SEARCH^} (@code{gnatls})
15264 @cindex @option{^-aI^/SOURCE_SEARCH^} (@code{gnatls})
15265 @cindex @option{^-I^/SEARCH^} (@code{gnatls})
15266 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@code{gnatls})
15267 Source path manipulation. Same meaning as the equivalent @command{gnatmake}
15268 flags (@pxref{Switches for gnatmake}).
15270 @item --RTS=@var{rts-path}
15271 @cindex @option{--RTS} (@code{gnatls})
15272 Specifies the default location of the runtime library. Same meaning as the
15273 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
15275 @item ^-v^/OUTPUT=VERBOSE^
15276 @cindex @option{^-v^/OUTPUT=VERBOSE^} (@code{gnatls})
15277 Verbose mode. Output the complete source, object and project paths. Do not use
15278 the default column layout but instead use long format giving as much as
15279 information possible on each requested units, including special
15280 characteristics such as:
15283 @item Preelaborable
15284 The unit is preelaborable in the Ada sense.
15287 No elaboration code has been produced by the compiler for this unit.
15290 The unit is pure in the Ada sense.
15292 @item Elaborate_Body
15293 The unit contains a pragma Elaborate_Body.
15296 The unit contains a pragma Remote_Types.
15298 @item Shared_Passive
15299 The unit contains a pragma Shared_Passive.
15302 This unit is part of the predefined environment and cannot be modified
15305 @item Remote_Call_Interface
15306 The unit contains a pragma Remote_Call_Interface.
15312 @node Examples of gnatls Usage
15313 @section Example of @code{gnatls} Usage
15317 Example of using the verbose switch. Note how the source and
15318 object paths are affected by the -I switch.
15321 $ gnatls -v -I.. demo1.o
15323 GNATLS 5.03w (20041123-34)
15324 Copyright 1997-2004 Free Software Foundation, Inc.
15326 Source Search Path:
15327 <Current_Directory>
15329 /home/comar/local/adainclude/
15331 Object Search Path:
15332 <Current_Directory>
15334 /home/comar/local/lib/gcc-lib/x86-linux/3.4.3/adalib/
15336 Project Search Path:
15337 <Current_Directory>
15338 /home/comar/local/lib/gnat/
15343 Kind => subprogram body
15344 Flags => No_Elab_Code
15345 Source => demo1.adb modified
15349 The following is an example of use of the dependency list.
15350 Note the use of the -s switch
15351 which gives a straight list of source files. This can be useful for
15352 building specialized scripts.
15355 $ gnatls -d demo2.o
15356 ./demo2.o demo2 OK demo2.adb
15362 $ gnatls -d -s -a demo1.o
15364 /home/comar/local/adainclude/ada.ads
15365 /home/comar/local/adainclude/a-finali.ads
15366 /home/comar/local/adainclude/a-filico.ads
15367 /home/comar/local/adainclude/a-stream.ads
15368 /home/comar/local/adainclude/a-tags.ads
15371 /home/comar/local/adainclude/gnat.ads
15372 /home/comar/local/adainclude/g-io.ads
15374 /home/comar/local/adainclude/system.ads
15375 /home/comar/local/adainclude/s-exctab.ads
15376 /home/comar/local/adainclude/s-finimp.ads
15377 /home/comar/local/adainclude/s-finroo.ads
15378 /home/comar/local/adainclude/s-secsta.ads
15379 /home/comar/local/adainclude/s-stalib.ads
15380 /home/comar/local/adainclude/s-stoele.ads
15381 /home/comar/local/adainclude/s-stratt.ads
15382 /home/comar/local/adainclude/s-tasoli.ads
15383 /home/comar/local/adainclude/s-unstyp.ads
15384 /home/comar/local/adainclude/unchconv.ads
15390 GNAT LIST /DEPENDENCIES /OUTPUT=SOURCES /ALL_UNITS DEMO1.ADB
15392 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]ada.ads
15393 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-finali.ads
15394 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-filico.ads
15395 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-stream.ads
15396 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-tags.ads
15400 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]gnat.ads
15401 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]g-io.ads
15403 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]system.ads
15404 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-exctab.ads
15405 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-finimp.ads
15406 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-finroo.ads
15407 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-secsta.ads
15408 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-stalib.ads
15409 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-stoele.ads
15410 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-stratt.ads
15411 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-tasoli.ads
15412 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-unstyp.ads
15413 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]unchconv.ads
15417 @node Cleaning Up Using gnatclean
15418 @chapter Cleaning Up Using @code{gnatclean}
15420 @cindex Cleaning tool
15423 @code{gnatclean} is a tool that allows the deletion of files produced by the
15424 compiler, binder and linker, including ALI files, object files, tree files,
15425 expanded source files, library files, interface copy source files, binder
15426 generated files and executable files.
15429 * Running gnatclean::
15430 * Switches for gnatclean::
15431 @c * Examples of gnatclean Usage::
15434 @node Running gnatclean
15435 @section Running @code{gnatclean}
15438 The @code{gnatclean} command has the form:
15441 $ gnatclean switches @var{names}
15445 @var{names} is a list of source file names. Suffixes @code{.^ads^ADS^} and
15446 @code{^adb^ADB^} may be omitted. If a project file is specified using switch
15447 @code{^-P^/PROJECT_FILE=^}, then @var{names} may be completely omitted.
15450 In normal mode, @code{gnatclean} delete the files produced by the compiler and,
15451 if switch @code{^-c^/COMPILER_FILES_ONLY^} is not specified, by the binder and
15452 the linker. In informative-only mode, specified by switch
15453 @code{^-n^/NODELETE^}, the list of files that would have been deleted in
15454 normal mode is listed, but no file is actually deleted.
15456 @node Switches for gnatclean
15457 @section Switches for @code{gnatclean}
15460 @code{gnatclean} recognizes the following switches:
15464 @cindex @option{--version} @command{gnatclean}
15465 Display Copyright and version, then exit disregarding all other options.
15468 @cindex @option{--help} @command{gnatclean}
15469 If @option{--version} was not used, display usage, then exit disregarding
15472 @item ^--subdirs^/SUBDIRS^=subdir
15473 Actual object directory of each project file is the subdirectory subdir of the
15474 object directory specified or defauted in the project file.
15476 @item ^--unchecked-shared-lib-imports^/UNCHECKED_SHARED_LIB_IMPORTS^
15477 By default, shared library projects are not allowed to import static library
15478 projects. When this switch is used on the command line, this restriction is
15481 @item ^-c^/COMPILER_FILES_ONLY^
15482 @cindex @option{^-c^/COMPILER_FILES_ONLY^} (@code{gnatclean})
15483 Only attempt to delete the files produced by the compiler, not those produced
15484 by the binder or the linker. The files that are not to be deleted are library
15485 files, interface copy files, binder generated files and executable files.
15487 @item ^-D ^/DIRECTORY_OBJECTS=^@var{dir}
15488 @cindex @option{^-D^/DIRECTORY_OBJECTS^} (@code{gnatclean})
15489 Indicate that ALI and object files should normally be found in directory
15492 @item ^-F^/FULL_PATH_IN_BRIEF_MESSAGES^
15493 @cindex @option{^-F^/FULL_PATH_IN_BRIEF_MESSAGES^} (@code{gnatclean})
15494 When using project files, if some errors or warnings are detected during
15495 parsing and verbose mode is not in effect (no use of switch
15496 ^-v^/VERBOSE^), then error lines start with the full path name of the project
15497 file, rather than its simple file name.
15500 @cindex @option{^-h^/HELP^} (@code{gnatclean})
15501 Output a message explaining the usage of @code{^gnatclean^gnatclean^}.
15503 @item ^-n^/NODELETE^
15504 @cindex @option{^-n^/NODELETE^} (@code{gnatclean})
15505 Informative-only mode. Do not delete any files. Output the list of the files
15506 that would have been deleted if this switch was not specified.
15508 @item ^-P^/PROJECT_FILE=^@var{project}
15509 @cindex @option{^-P^/PROJECT_FILE^} (@code{gnatclean})
15510 Use project file @var{project}. Only one such switch can be used.
15511 When cleaning a project file, the files produced by the compilation of the
15512 immediate sources or inherited sources of the project files are to be
15513 deleted. This is not depending on the presence or not of executable names
15514 on the command line.
15517 @cindex @option{^-q^/QUIET^} (@code{gnatclean})
15518 Quiet output. If there are no errors, do not output anything, except in
15519 verbose mode (switch ^-v^/VERBOSE^) or in informative-only mode
15520 (switch ^-n^/NODELETE^).
15522 @item ^-r^/RECURSIVE^
15523 @cindex @option{^-r^/RECURSIVE^} (@code{gnatclean})
15524 When a project file is specified (using switch ^-P^/PROJECT_FILE=^),
15525 clean all imported and extended project files, recursively. If this switch
15526 is not specified, only the files related to the main project file are to be
15527 deleted. This switch has no effect if no project file is specified.
15529 @item ^-v^/VERBOSE^
15530 @cindex @option{^-v^/VERBOSE^} (@code{gnatclean})
15533 @item ^-vP^/MESSAGES_PROJECT_FILE=^@emph{x}
15534 @cindex @option{^-vP^/MESSAGES_PROJECT_FILE^} (@code{gnatclean})
15535 Indicates the verbosity of the parsing of GNAT project files.
15536 @xref{Switches Related to Project Files}.
15538 @item ^-X^/EXTERNAL_REFERENCE=^@var{name=value}
15539 @cindex @option{^-X^/EXTERNAL_REFERENCE^} (@code{gnatclean})
15540 Indicates that external variable @var{name} has the value @var{value}.
15541 The Project Manager will use this value for occurrences of
15542 @code{external(name)} when parsing the project file.
15543 @xref{Switches Related to Project Files}.
15545 @item ^-aO^/OBJECT_SEARCH=^@var{dir}
15546 @cindex @option{^-aO^/OBJECT_SEARCH^} (@code{gnatclean})
15547 When searching for ALI and object files, look in directory
15550 @item ^-I^/SEARCH=^@var{dir}
15551 @cindex @option{^-I^/SEARCH^} (@code{gnatclean})
15552 Equivalent to @option{^-aO^/OBJECT_SEARCH=^@var{dir}}.
15554 @item ^-I-^/NOCURRENT_DIRECTORY^
15555 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@code{gnatclean})
15556 @cindex Source files, suppressing search
15557 Do not look for ALI or object files in the directory
15558 where @code{gnatclean} was invoked.
15562 @c @node Examples of gnatclean Usage
15563 @c @section Examples of @code{gnatclean} Usage
15566 @node GNAT and Libraries
15567 @chapter GNAT and Libraries
15568 @cindex Library, building, installing, using
15571 This chapter describes how to build and use libraries with GNAT, and also shows
15572 how to recompile the GNAT run-time library. You should be familiar with the
15573 Project Manager facility (@pxref{GNAT Project Manager}) before reading this
15577 * Introduction to Libraries in GNAT::
15578 * General Ada Libraries::
15579 * Stand-alone Ada Libraries::
15580 * Rebuilding the GNAT Run-Time Library::
15583 @node Introduction to Libraries in GNAT
15584 @section Introduction to Libraries in GNAT
15587 A library is, conceptually, a collection of objects which does not have its
15588 own main thread of execution, but rather provides certain services to the
15589 applications that use it. A library can be either statically linked with the
15590 application, in which case its code is directly included in the application,
15591 or, on platforms that support it, be dynamically linked, in which case
15592 its code is shared by all applications making use of this library.
15594 GNAT supports both types of libraries.
15595 In the static case, the compiled code can be provided in different ways. The
15596 simplest approach is to provide directly the set of objects resulting from
15597 compilation of the library source files. Alternatively, you can group the
15598 objects into an archive using whatever commands are provided by the operating
15599 system. For the latter case, the objects are grouped into a shared library.
15601 In the GNAT environment, a library has three types of components:
15607 @xref{The Ada Library Information Files}.
15609 Object files, an archive or a shared library.
15613 A GNAT library may expose all its source files, which is useful for
15614 documentation purposes. Alternatively, it may expose only the units needed by
15615 an external user to make use of the library. That is to say, the specs
15616 reflecting the library services along with all the units needed to compile
15617 those specs, which can include generic bodies or any body implementing an
15618 inlined routine. In the case of @emph{stand-alone libraries} those exposed
15619 units are called @emph{interface units} (@pxref{Stand-alone Ada Libraries}).
15621 All compilation units comprising an application, including those in a library,
15622 need to be elaborated in an order partially defined by Ada's semantics. GNAT
15623 computes the elaboration order from the @file{ALI} files and this is why they
15624 constitute a mandatory part of GNAT libraries.
15625 @emph{Stand-alone libraries} are the exception to this rule because a specific
15626 library elaboration routine is produced independently of the application(s)
15629 @node General Ada Libraries
15630 @section General Ada Libraries
15633 * Building a library::
15634 * Installing a library::
15635 * Using a library::
15638 @node Building a library
15639 @subsection Building a library
15642 The easiest way to build a library is to use the Project Manager,
15643 which supports a special type of project called a @emph{Library Project}
15644 (@pxref{Library Projects}).
15646 A project is considered a library project, when two project-level attributes
15647 are defined in it: @code{Library_Name} and @code{Library_Dir}. In order to
15648 control different aspects of library configuration, additional optional
15649 project-level attributes can be specified:
15652 This attribute controls whether the library is to be static or dynamic
15654 @item Library_Version
15655 This attribute specifies the library version; this value is used
15656 during dynamic linking of shared libraries to determine if the currently
15657 installed versions of the binaries are compatible.
15659 @item Library_Options
15661 These attributes specify additional low-level options to be used during
15662 library generation, and redefine the actual application used to generate
15667 The GNAT Project Manager takes full care of the library maintenance task,
15668 including recompilation of the source files for which objects do not exist
15669 or are not up to date, assembly of the library archive, and installation of
15670 the library (i.e., copying associated source, object and @file{ALI} files
15671 to the specified location).
15673 Here is a simple library project file:
15674 @smallexample @c ada
15676 for Source_Dirs use ("src1", "src2");
15677 for Object_Dir use "obj";
15678 for Library_Name use "mylib";
15679 for Library_Dir use "lib";
15680 for Library_Kind use "dynamic";
15685 and the compilation command to build and install the library:
15687 @smallexample @c ada
15688 $ gnatmake -Pmy_lib
15692 It is not entirely trivial to perform manually all the steps required to
15693 produce a library. We recommend that you use the GNAT Project Manager
15694 for this task. In special cases where this is not desired, the necessary
15695 steps are discussed below.
15697 There are various possibilities for compiling the units that make up the
15698 library: for example with a Makefile (@pxref{Using the GNU make Utility}) or
15699 with a conventional script. For simple libraries, it is also possible to create
15700 a dummy main program which depends upon all the packages that comprise the
15701 interface of the library. This dummy main program can then be given to
15702 @command{gnatmake}, which will ensure that all necessary objects are built.
15704 After this task is accomplished, you should follow the standard procedure
15705 of the underlying operating system to produce the static or shared library.
15707 Here is an example of such a dummy program:
15708 @smallexample @c ada
15710 with My_Lib.Service1;
15711 with My_Lib.Service2;
15712 with My_Lib.Service3;
15713 procedure My_Lib_Dummy is
15721 Here are the generic commands that will build an archive or a shared library.
15724 # compiling the library
15725 $ gnatmake -c my_lib_dummy.adb
15727 # we don't need the dummy object itself
15728 $ rm my_lib_dummy.o my_lib_dummy.ali
15730 # create an archive with the remaining objects
15731 $ ar rc libmy_lib.a *.o
15732 # some systems may require "ranlib" to be run as well
15734 # or create a shared library
15735 $ gcc -shared -o libmy_lib.so *.o
15736 # some systems may require the code to have been compiled with -fPIC
15738 # remove the object files that are now in the library
15741 # Make the ALI files read-only so that gnatmake will not try to
15742 # regenerate the objects that are in the library
15747 Please note that the library must have a name of the form @file{lib@var{xxx}.a}
15748 or @file{lib@var{xxx}.so} (or @file{lib@var{xxx}.dll} on Windows) in order to
15749 be accessed by the directive @option{-l@var{xxx}} at link time.
15751 @node Installing a library
15752 @subsection Installing a library
15753 @cindex @code{ADA_PROJECT_PATH}
15754 @cindex @code{GPR_PROJECT_PATH}
15757 If you use project files, library installation is part of the library build
15758 process (@pxref{Installing a library with project files}).
15760 When project files are not an option, it is also possible, but not recommended,
15761 to install the library so that the sources needed to use the library are on the
15762 Ada source path and the ALI files & libraries be on the Ada Object path (see
15763 @ref{Search Paths and the Run-Time Library (RTL)}. Alternatively, the system
15764 administrator can place general-purpose libraries in the default compiler
15765 paths, by specifying the libraries' location in the configuration files
15766 @file{ada_source_path} and @file{ada_object_path}. These configuration files
15767 must be located in the GNAT installation tree at the same place as the gcc spec
15768 file. The location of the gcc spec file can be determined as follows:
15774 The configuration files mentioned above have a simple format: each line
15775 must contain one unique directory name.
15776 Those names are added to the corresponding path
15777 in their order of appearance in the file. The names can be either absolute
15778 or relative; in the latter case, they are relative to where theses files
15781 The files @file{ada_source_path} and @file{ada_object_path} might not be
15783 GNAT installation, in which case, GNAT will look for its run-time library in
15784 the directories @file{adainclude} (for the sources) and @file{adalib} (for the
15785 objects and @file{ALI} files). When the files exist, the compiler does not
15786 look in @file{adainclude} and @file{adalib}, and thus the
15787 @file{ada_source_path} file
15788 must contain the location for the GNAT run-time sources (which can simply
15789 be @file{adainclude}). In the same way, the @file{ada_object_path} file must
15790 contain the location for the GNAT run-time objects (which can simply
15793 You can also specify a new default path to the run-time library at compilation
15794 time with the switch @option{--RTS=rts-path}. You can thus choose / change
15795 the run-time library you want your program to be compiled with. This switch is
15796 recognized by @command{gcc}, @command{gnatmake}, @command{gnatbind},
15797 @command{gnatls}, @command{gnatfind} and @command{gnatxref}.
15799 It is possible to install a library before or after the standard GNAT
15800 library, by reordering the lines in the configuration files. In general, a
15801 library must be installed before the GNAT library if it redefines
15804 @node Using a library
15805 @subsection Using a library
15807 @noindent Once again, the project facility greatly simplifies the use of
15808 libraries. In this context, using a library is just a matter of adding a
15809 @code{with} clause in the user project. For instance, to make use of the
15810 library @code{My_Lib} shown in examples in earlier sections, you can
15813 @smallexample @c projectfile
15820 Even if you have a third-party, non-Ada library, you can still use GNAT's
15821 Project Manager facility to provide a wrapper for it. For example, the
15822 following project, when @code{with}ed by your main project, will link with the
15823 third-party library @file{liba.a}:
15825 @smallexample @c projectfile
15828 for Externally_Built use "true";
15829 for Source_Files use ();
15830 for Library_Dir use "lib";
15831 for Library_Name use "a";
15832 for Library_Kind use "static";
15836 This is an alternative to the use of @code{pragma Linker_Options}. It is
15837 especially interesting in the context of systems with several interdependent
15838 static libraries where finding a proper linker order is not easy and best be
15839 left to the tools having visibility over project dependence information.
15842 In order to use an Ada library manually, you need to make sure that this
15843 library is on both your source and object path
15844 (see @ref{Search Paths and the Run-Time Library (RTL)}
15845 and @ref{Search Paths for gnatbind}). Furthermore, when the objects are grouped
15846 in an archive or a shared library, you need to specify the desired
15847 library at link time.
15849 For example, you can use the library @file{mylib} installed in
15850 @file{/dir/my_lib_src} and @file{/dir/my_lib_obj} with the following commands:
15853 $ gnatmake -aI/dir/my_lib_src -aO/dir/my_lib_obj my_appl \
15858 This can be expressed more simply:
15863 when the following conditions are met:
15866 @file{/dir/my_lib_src} has been added by the user to the environment
15867 variable @env{ADA_INCLUDE_PATH}, or by the administrator to the file
15868 @file{ada_source_path}
15870 @file{/dir/my_lib_obj} has been added by the user to the environment
15871 variable @env{ADA_OBJECTS_PATH}, or by the administrator to the file
15872 @file{ada_object_path}
15874 a pragma @code{Linker_Options} has been added to one of the sources.
15877 @smallexample @c ada
15878 pragma Linker_Options ("-lmy_lib");
15882 @node Stand-alone Ada Libraries
15883 @section Stand-alone Ada Libraries
15884 @cindex Stand-alone library, building, using
15887 * Introduction to Stand-alone Libraries::
15888 * Building a Stand-alone Library::
15889 * Creating a Stand-alone Library to be used in a non-Ada context::
15890 * Restrictions in Stand-alone Libraries::
15893 @node Introduction to Stand-alone Libraries
15894 @subsection Introduction to Stand-alone Libraries
15897 A Stand-alone Library (abbreviated ``SAL'') is a library that contains the
15899 elaborate the Ada units that are included in the library. In contrast with
15900 an ordinary library, which consists of all sources, objects and @file{ALI}
15902 library, a SAL may specify a restricted subset of compilation units
15903 to serve as a library interface. In this case, the fully
15904 self-sufficient set of files will normally consist of an objects
15905 archive, the sources of interface units' specs, and the @file{ALI}
15906 files of interface units.
15907 If an interface spec contains a generic unit or an inlined subprogram,
15909 source must also be provided; if the units that must be provided in the source
15910 form depend on other units, the source and @file{ALI} files of those must
15913 The main purpose of a SAL is to minimize the recompilation overhead of client
15914 applications when a new version of the library is installed. Specifically,
15915 if the interface sources have not changed, client applications do not need to
15916 be recompiled. If, furthermore, a SAL is provided in the shared form and its
15917 version, controlled by @code{Library_Version} attribute, is not changed,
15918 then the clients do not need to be relinked.
15920 SALs also allow the library providers to minimize the amount of library source
15921 text exposed to the clients. Such ``information hiding'' might be useful or
15922 necessary for various reasons.
15924 Stand-alone libraries are also well suited to be used in an executable whose
15925 main routine is not written in Ada.
15927 @node Building a Stand-alone Library
15928 @subsection Building a Stand-alone Library
15931 GNAT's Project facility provides a simple way of building and installing
15932 stand-alone libraries; see @ref{Stand-alone Library Projects}.
15933 To be a Stand-alone Library Project, in addition to the two attributes
15934 that make a project a Library Project (@code{Library_Name} and
15935 @code{Library_Dir}; see @ref{Library Projects}), the attribute
15936 @code{Library_Interface} must be defined. For example:
15938 @smallexample @c projectfile
15940 for Library_Dir use "lib_dir";
15941 for Library_Name use "dummy";
15942 for Library_Interface use ("int1", "int1.child");
15947 Attribute @code{Library_Interface} has a non-empty string list value,
15948 each string in the list designating a unit contained in an immediate source
15949 of the project file.
15951 When a Stand-alone Library is built, first the binder is invoked to build
15952 a package whose name depends on the library name
15953 (@file{^b~dummy.ads/b^B$DUMMY.ADS/B^} in the example above).
15954 This binder-generated package includes initialization and
15955 finalization procedures whose
15956 names depend on the library name (@code{dummyinit} and @code{dummyfinal}
15958 above). The object corresponding to this package is included in the library.
15960 You must ensure timely (e.g., prior to any use of interfaces in the SAL)
15961 calling of these procedures if a static SAL is built, or if a shared SAL
15963 with the project-level attribute @code{Library_Auto_Init} set to
15966 For a Stand-Alone Library, only the @file{ALI} files of the Interface Units
15967 (those that are listed in attribute @code{Library_Interface}) are copied to
15968 the Library Directory. As a consequence, only the Interface Units may be
15969 imported from Ada units outside of the library. If other units are imported,
15970 the binding phase will fail.
15972 The attribute @code{Library_Src_Dir} may be specified for a
15973 Stand-Alone Library. @code{Library_Src_Dir} is a simple attribute that has a
15974 single string value. Its value must be the path (absolute or relative to the
15975 project directory) of an existing directory. This directory cannot be the
15976 object directory or one of the source directories, but it can be the same as
15977 the library directory. The sources of the Interface
15978 Units of the library that are needed by an Ada client of the library will be
15979 copied to the designated directory, called the Interface Copy directory.
15980 These sources include the specs of the Interface Units, but they may also
15981 include bodies and subunits, when pragmas @code{Inline} or @code{Inline_Always}
15982 are used, or when there is a generic unit in the spec. Before the sources
15983 are copied to the Interface Copy directory, an attempt is made to delete all
15984 files in the Interface Copy directory.
15986 Building stand-alone libraries by hand is somewhat tedious, but for those
15987 occasions when it is necessary here are the steps that you need to perform:
15990 Compile all library sources.
15993 Invoke the binder with the switch @option{-n} (No Ada main program),
15994 with all the @file{ALI} files of the interfaces, and
15995 with the switch @option{-L} to give specific names to the @code{init}
15996 and @code{final} procedures. For example:
15998 gnatbind -n int1.ali int2.ali -Lsal1
16002 Compile the binder generated file:
16008 Link the dynamic library with all the necessary object files,
16009 indicating to the linker the names of the @code{init} (and possibly
16010 @code{final}) procedures for automatic initialization (and finalization).
16011 The built library should be placed in a directory different from
16012 the object directory.
16015 Copy the @code{ALI} files of the interface to the library directory,
16016 add in this copy an indication that it is an interface to a SAL
16017 (i.e., add a word @option{SL} on the line in the @file{ALI} file that starts
16018 with letter ``P'') and make the modified copy of the @file{ALI} file
16023 Using SALs is not different from using other libraries
16024 (see @ref{Using a library}).
16026 @node Creating a Stand-alone Library to be used in a non-Ada context
16027 @subsection Creating a Stand-alone Library to be used in a non-Ada context
16030 It is easy to adapt the SAL build procedure discussed above for use of a SAL in
16033 The only extra step required is to ensure that library interface subprograms
16034 are compatible with the main program, by means of @code{pragma Export}
16035 or @code{pragma Convention}.
16037 Here is an example of simple library interface for use with C main program:
16039 @smallexample @c ada
16040 package My_Package is
16042 procedure Do_Something;
16043 pragma Export (C, Do_Something, "do_something");
16045 procedure Do_Something_Else;
16046 pragma Export (C, Do_Something_Else, "do_something_else");
16052 On the foreign language side, you must provide a ``foreign'' view of the
16053 library interface; remember that it should contain elaboration routines in
16054 addition to interface subprograms.
16056 The example below shows the content of @code{mylib_interface.h} (note
16057 that there is no rule for the naming of this file, any name can be used)
16059 /* the library elaboration procedure */
16060 extern void mylibinit (void);
16062 /* the library finalization procedure */
16063 extern void mylibfinal (void);
16065 /* the interface exported by the library */
16066 extern void do_something (void);
16067 extern void do_something_else (void);
16071 Libraries built as explained above can be used from any program, provided
16072 that the elaboration procedures (named @code{mylibinit} in the previous
16073 example) are called before the library services are used. Any number of
16074 libraries can be used simultaneously, as long as the elaboration
16075 procedure of each library is called.
16077 Below is an example of a C program that uses the @code{mylib} library.
16080 #include "mylib_interface.h"
16085 /* First, elaborate the library before using it */
16088 /* Main program, using the library exported entities */
16090 do_something_else ();
16092 /* Library finalization at the end of the program */
16099 Note that invoking any library finalization procedure generated by
16100 @code{gnatbind} shuts down the Ada run-time environment.
16102 finalization of all Ada libraries must be performed at the end of the program.
16103 No call to these libraries or to the Ada run-time library should be made
16104 after the finalization phase.
16106 @node Restrictions in Stand-alone Libraries
16107 @subsection Restrictions in Stand-alone Libraries
16110 The pragmas listed below should be used with caution inside libraries,
16111 as they can create incompatibilities with other Ada libraries:
16113 @item pragma @code{Locking_Policy}
16114 @item pragma @code{Queuing_Policy}
16115 @item pragma @code{Task_Dispatching_Policy}
16116 @item pragma @code{Unreserve_All_Interrupts}
16120 When using a library that contains such pragmas, the user must make sure
16121 that all libraries use the same pragmas with the same values. Otherwise,
16122 @code{Program_Error} will
16123 be raised during the elaboration of the conflicting
16124 libraries. The usage of these pragmas and its consequences for the user
16125 should therefore be well documented.
16127 Similarly, the traceback in the exception occurrence mechanism should be
16128 enabled or disabled in a consistent manner across all libraries.
16129 Otherwise, Program_Error will be raised during the elaboration of the
16130 conflicting libraries.
16132 If the @code{Version} or @code{Body_Version}
16133 attributes are used inside a library, then you need to
16134 perform a @code{gnatbind} step that specifies all @file{ALI} files in all
16135 libraries, so that version identifiers can be properly computed.
16136 In practice these attributes are rarely used, so this is unlikely
16137 to be a consideration.
16139 @node Rebuilding the GNAT Run-Time Library
16140 @section Rebuilding the GNAT Run-Time Library
16141 @cindex GNAT Run-Time Library, rebuilding
16142 @cindex Building the GNAT Run-Time Library
16143 @cindex Rebuilding the GNAT Run-Time Library
16144 @cindex Run-Time Library, rebuilding
16147 It may be useful to recompile the GNAT library in various contexts, the
16148 most important one being the use of partition-wide configuration pragmas
16149 such as @code{Normalize_Scalars}. A special Makefile called
16150 @code{Makefile.adalib} is provided to that effect and can be found in
16151 the directory containing the GNAT library. The location of this
16152 directory depends on the way the GNAT environment has been installed and can
16153 be determined by means of the command:
16160 The last entry in the object search path usually contains the
16161 gnat library. This Makefile contains its own documentation and in
16162 particular the set of instructions needed to rebuild a new library and
16165 @node Using the GNU make Utility
16166 @chapter Using the GNU @code{make} Utility
16170 This chapter offers some examples of makefiles that solve specific
16171 problems. It does not explain how to write a makefile (@pxref{Top,, GNU
16172 make, make, GNU @code{make}}), nor does it try to replace the
16173 @command{gnatmake} utility (@pxref{The GNAT Make Program gnatmake}).
16175 All the examples in this section are specific to the GNU version of
16176 make. Although @command{make} is a standard utility, and the basic language
16177 is the same, these examples use some advanced features found only in
16181 * Using gnatmake in a Makefile::
16182 * Automatically Creating a List of Directories::
16183 * Generating the Command Line Switches::
16184 * Overcoming Command Line Length Limits::
16187 @node Using gnatmake in a Makefile
16188 @section Using gnatmake in a Makefile
16193 Complex project organizations can be handled in a very powerful way by
16194 using GNU make combined with gnatmake. For instance, here is a Makefile
16195 which allows you to build each subsystem of a big project into a separate
16196 shared library. Such a makefile allows you to significantly reduce the link
16197 time of very big applications while maintaining full coherence at
16198 each step of the build process.
16200 The list of dependencies are handled automatically by
16201 @command{gnatmake}. The Makefile is simply used to call gnatmake in each of
16202 the appropriate directories.
16204 Note that you should also read the example on how to automatically
16205 create the list of directories
16206 (@pxref{Automatically Creating a List of Directories})
16207 which might help you in case your project has a lot of subdirectories.
16212 @font@heightrm=cmr8
16215 ## This Makefile is intended to be used with the following directory
16217 ## - The sources are split into a series of csc (computer software components)
16218 ## Each of these csc is put in its own directory.
16219 ## Their name are referenced by the directory names.
16220 ## They will be compiled into shared library (although this would also work
16221 ## with static libraries
16222 ## - The main program (and possibly other packages that do not belong to any
16223 ## csc is put in the top level directory (where the Makefile is).
16224 ## toplevel_dir __ first_csc (sources) __ lib (will contain the library)
16225 ## \_ second_csc (sources) __ lib (will contain the library)
16227 ## Although this Makefile is build for shared library, it is easy to modify
16228 ## to build partial link objects instead (modify the lines with -shared and
16231 ## With this makefile, you can change any file in the system or add any new
16232 ## file, and everything will be recompiled correctly (only the relevant shared
16233 ## objects will be recompiled, and the main program will be re-linked).
16235 # The list of computer software component for your project. This might be
16236 # generated automatically.
16239 # Name of the main program (no extension)
16242 # If we need to build objects with -fPIC, uncomment the following line
16245 # The following variable should give the directory containing libgnat.so
16246 # You can get this directory through 'gnatls -v'. This is usually the last
16247 # directory in the Object_Path.
16250 # The directories for the libraries
16251 # (This macro expands the list of CSC to the list of shared libraries, you
16252 # could simply use the expanded form:
16253 # LIB_DIR=aa/lib/libaa.so bb/lib/libbb.so cc/lib/libcc.so
16254 LIB_DIR=$@{foreach dir,$@{CSC_LIST@},$@{dir@}/lib/lib$@{dir@}.so@}
16256 $@{MAIN@}: objects $@{LIB_DIR@}
16257 gnatbind $@{MAIN@} $@{CSC_LIST:%=-aO%/lib@} -shared
16258 gnatlink $@{MAIN@} $@{CSC_LIST:%=-l%@}
16261 # recompile the sources
16262 gnatmake -c -i $@{MAIN@}.adb $@{NEED_FPIC@} $@{CSC_LIST:%=-I%@}
16264 # Note: In a future version of GNAT, the following commands will be simplified
16265 # by a new tool, gnatmlib
16267 mkdir -p $@{dir $@@ @}
16268 cd $@{dir $@@ @} && gcc -shared -o $@{notdir $@@ @} ../*.o -L$@{GLIB@} -lgnat
16269 cd $@{dir $@@ @} && cp -f ../*.ali .
16271 # The dependencies for the modules
16272 # Note that we have to force the expansion of *.o, since in some cases
16273 # make won't be able to do it itself.
16274 aa/lib/libaa.so: $@{wildcard aa/*.o@}
16275 bb/lib/libbb.so: $@{wildcard bb/*.o@}
16276 cc/lib/libcc.so: $@{wildcard cc/*.o@}
16278 # Make sure all of the shared libraries are in the path before starting the
16281 LD_LIBRARY_PATH=`pwd`/aa/lib:`pwd`/bb/lib:`pwd`/cc/lib ./$@{MAIN@}
16284 $@{RM@} -rf $@{CSC_LIST:%=%/lib@}
16285 $@{RM@} $@{CSC_LIST:%=%/*.ali@}
16286 $@{RM@} $@{CSC_LIST:%=%/*.o@}
16287 $@{RM@} *.o *.ali $@{MAIN@}
16290 @node Automatically Creating a List of Directories
16291 @section Automatically Creating a List of Directories
16294 In most makefiles, you will have to specify a list of directories, and
16295 store it in a variable. For small projects, it is often easier to
16296 specify each of them by hand, since you then have full control over what
16297 is the proper order for these directories, which ones should be
16300 However, in larger projects, which might involve hundreds of
16301 subdirectories, it might be more convenient to generate this list
16304 The example below presents two methods. The first one, although less
16305 general, gives you more control over the list. It involves wildcard
16306 characters, that are automatically expanded by @command{make}. Its
16307 shortcoming is that you need to explicitly specify some of the
16308 organization of your project, such as for instance the directory tree
16309 depth, whether some directories are found in a separate tree, @enddots{}
16311 The second method is the most general one. It requires an external
16312 program, called @command{find}, which is standard on all Unix systems. All
16313 the directories found under a given root directory will be added to the
16319 @font@heightrm=cmr8
16322 # The examples below are based on the following directory hierarchy:
16323 # All the directories can contain any number of files
16324 # ROOT_DIRECTORY -> a -> aa -> aaa
16327 # -> b -> ba -> baa
16330 # This Makefile creates a variable called DIRS, that can be reused any time
16331 # you need this list (see the other examples in this section)
16333 # The root of your project's directory hierarchy
16337 # First method: specify explicitly the list of directories
16338 # This allows you to specify any subset of all the directories you need.
16341 DIRS := a/aa/ a/ab/ b/ba/
16344 # Second method: use wildcards
16345 # Note that the argument(s) to wildcard below should end with a '/'.
16346 # Since wildcards also return file names, we have to filter them out
16347 # to avoid duplicate directory names.
16348 # We thus use make's @code{dir} and @code{sort} functions.
16349 # It sets DIRs to the following value (note that the directories aaa and baa
16350 # are not given, unless you change the arguments to wildcard).
16351 # DIRS= ./a/a/ ./b/ ./a/aa/ ./a/ab/ ./a/ac/ ./b/ba/ ./b/bb/ ./b/bc/
16354 DIRS := $@{sort $@{dir $@{wildcard $@{ROOT_DIRECTORY@}/*/
16355 $@{ROOT_DIRECTORY@}/*/*/@}@}@}
16358 # Third method: use an external program
16359 # This command is much faster if run on local disks, avoiding NFS slowdowns.
16360 # This is the most complete command: it sets DIRs to the following value:
16361 # DIRS= ./a ./a/aa ./a/aa/aaa ./a/ab ./a/ac ./b ./b/ba ./b/ba/baa ./b/bb ./b/bc
16364 DIRS := $@{shell find $@{ROOT_DIRECTORY@} -type d -print@}
16368 @node Generating the Command Line Switches
16369 @section Generating the Command Line Switches
16372 Once you have created the list of directories as explained in the
16373 previous section (@pxref{Automatically Creating a List of Directories}),
16374 you can easily generate the command line arguments to pass to gnatmake.
16376 For the sake of completeness, this example assumes that the source path
16377 is not the same as the object path, and that you have two separate lists
16381 # see "Automatically creating a list of directories" to create
16386 GNATMAKE_SWITCHES := $@{patsubst %,-aI%,$@{SOURCE_DIRS@}@}
16387 GNATMAKE_SWITCHES += $@{patsubst %,-aO%,$@{OBJECT_DIRS@}@}
16390 gnatmake $@{GNATMAKE_SWITCHES@} main_unit
16393 @node Overcoming Command Line Length Limits
16394 @section Overcoming Command Line Length Limits
16397 One problem that might be encountered on big projects is that many
16398 operating systems limit the length of the command line. It is thus hard to give
16399 gnatmake the list of source and object directories.
16401 This example shows how you can set up environment variables, which will
16402 make @command{gnatmake} behave exactly as if the directories had been
16403 specified on the command line, but have a much higher length limit (or
16404 even none on most systems).
16406 It assumes that you have created a list of directories in your Makefile,
16407 using one of the methods presented in
16408 @ref{Automatically Creating a List of Directories}.
16409 For the sake of completeness, we assume that the object
16410 path (where the ALI files are found) is different from the sources patch.
16412 Note a small trick in the Makefile below: for efficiency reasons, we
16413 create two temporary variables (SOURCE_LIST and OBJECT_LIST), that are
16414 expanded immediately by @code{make}. This way we overcome the standard
16415 make behavior which is to expand the variables only when they are
16418 On Windows, if you are using the standard Windows command shell, you must
16419 replace colons with semicolons in the assignments to these variables.
16424 @font@heightrm=cmr8
16427 # In this example, we create both ADA_INCLUDE_PATH and ADA_OBJECT_PATH.
16428 # This is the same thing as putting the -I arguments on the command line.
16429 # (the equivalent of using -aI on the command line would be to define
16430 # only ADA_INCLUDE_PATH, the equivalent of -aO is ADA_OBJECT_PATH).
16431 # You can of course have different values for these variables.
16433 # Note also that we need to keep the previous values of these variables, since
16434 # they might have been set before running 'make' to specify where the GNAT
16435 # library is installed.
16437 # see "Automatically creating a list of directories" to create these
16443 space:=$@{empty@} $@{empty@}
16444 SOURCE_LIST := $@{subst $@{space@},:,$@{SOURCE_DIRS@}@}
16445 OBJECT_LIST := $@{subst $@{space@},:,$@{OBJECT_DIRS@}@}
16446 ADA_INCLUDE_PATH += $@{SOURCE_LIST@}
16447 ADA_OBJECT_PATH += $@{OBJECT_LIST@}
16448 export ADA_INCLUDE_PATH
16449 export ADA_OBJECT_PATH
16456 @node Memory Management Issues
16457 @chapter Memory Management Issues
16460 This chapter describes some useful memory pools provided in the GNAT library
16461 and in particular the GNAT Debug Pool facility, which can be used to detect
16462 incorrect uses of access values (including ``dangling references'').
16464 It also describes the @command{gnatmem} tool, which can be used to track down
16469 * Some Useful Memory Pools::
16470 * The GNAT Debug Pool Facility::
16472 * The gnatmem Tool::
16476 @node Some Useful Memory Pools
16477 @section Some Useful Memory Pools
16478 @findex Memory Pool
16479 @cindex storage, pool
16482 The @code{System.Pool_Global} package offers the Unbounded_No_Reclaim_Pool
16483 storage pool. Allocations use the standard system call @code{malloc} while
16484 deallocations use the standard system call @code{free}. No reclamation is
16485 performed when the pool goes out of scope. For performance reasons, the
16486 standard default Ada allocators/deallocators do not use any explicit storage
16487 pools but if they did, they could use this storage pool without any change in
16488 behavior. That is why this storage pool is used when the user
16489 manages to make the default implicit allocator explicit as in this example:
16490 @smallexample @c ada
16491 type T1 is access Something;
16492 -- no Storage pool is defined for T2
16493 type T2 is access Something_Else;
16494 for T2'Storage_Pool use T1'Storage_Pool;
16495 -- the above is equivalent to
16496 for T2'Storage_Pool use System.Pool_Global.Global_Pool_Object;
16500 The @code{System.Pool_Local} package offers the Unbounded_Reclaim_Pool storage
16501 pool. The allocation strategy is similar to @code{Pool_Local}'s
16502 except that the all
16503 storage allocated with this pool is reclaimed when the pool object goes out of
16504 scope. This pool provides a explicit mechanism similar to the implicit one
16505 provided by several Ada 83 compilers for allocations performed through a local
16506 access type and whose purpose was to reclaim memory when exiting the
16507 scope of a given local access. As an example, the following program does not
16508 leak memory even though it does not perform explicit deallocation:
16510 @smallexample @c ada
16511 with System.Pool_Local;
16512 procedure Pooloc1 is
16513 procedure Internal is
16514 type A is access Integer;
16515 X : System.Pool_Local.Unbounded_Reclaim_Pool;
16516 for A'Storage_Pool use X;
16519 for I in 1 .. 50 loop
16524 for I in 1 .. 100 loop
16531 The @code{System.Pool_Size} package implements the Stack_Bounded_Pool used when
16532 @code{Storage_Size} is specified for an access type.
16533 The whole storage for the pool is
16534 allocated at once, usually on the stack at the point where the access type is
16535 elaborated. It is automatically reclaimed when exiting the scope where the
16536 access type is defined. This package is not intended to be used directly by the
16537 user and it is implicitly used for each such declaration:
16539 @smallexample @c ada
16540 type T1 is access Something;
16541 for T1'Storage_Size use 10_000;
16544 @node The GNAT Debug Pool Facility
16545 @section The GNAT Debug Pool Facility
16547 @cindex storage, pool, memory corruption
16550 The use of unchecked deallocation and unchecked conversion can easily
16551 lead to incorrect memory references. The problems generated by such
16552 references are usually difficult to tackle because the symptoms can be
16553 very remote from the origin of the problem. In such cases, it is
16554 very helpful to detect the problem as early as possible. This is the
16555 purpose of the Storage Pool provided by @code{GNAT.Debug_Pools}.
16557 In order to use the GNAT specific debugging pool, the user must
16558 associate a debug pool object with each of the access types that may be
16559 related to suspected memory problems. See Ada Reference Manual 13.11.
16560 @smallexample @c ada
16561 type Ptr is access Some_Type;
16562 Pool : GNAT.Debug_Pools.Debug_Pool;
16563 for Ptr'Storage_Pool use Pool;
16567 @code{GNAT.Debug_Pools} is derived from a GNAT-specific kind of
16568 pool: the @code{Checked_Pool}. Such pools, like standard Ada storage pools,
16569 allow the user to redefine allocation and deallocation strategies. They
16570 also provide a checkpoint for each dereference, through the use of
16571 the primitive operation @code{Dereference} which is implicitly called at
16572 each dereference of an access value.
16574 Once an access type has been associated with a debug pool, operations on
16575 values of the type may raise four distinct exceptions,
16576 which correspond to four potential kinds of memory corruption:
16579 @code{GNAT.Debug_Pools.Accessing_Not_Allocated_Storage}
16581 @code{GNAT.Debug_Pools.Accessing_Deallocated_Storage}
16583 @code{GNAT.Debug_Pools.Freeing_Not_Allocated_Storage}
16585 @code{GNAT.Debug_Pools.Freeing_Deallocated_Storage }
16589 For types associated with a Debug_Pool, dynamic allocation is performed using
16590 the standard GNAT allocation routine. References to all allocated chunks of
16591 memory are kept in an internal dictionary. Several deallocation strategies are
16592 provided, whereupon the user can choose to release the memory to the system,
16593 keep it allocated for further invalid access checks, or fill it with an easily
16594 recognizable pattern for debug sessions. The memory pattern is the old IBM
16595 hexadecimal convention: @code{16#DEADBEEF#}.
16597 See the documentation in the file g-debpoo.ads for more information on the
16598 various strategies.
16600 Upon each dereference, a check is made that the access value denotes a
16601 properly allocated memory location. Here is a complete example of use of
16602 @code{Debug_Pools}, that includes typical instances of memory corruption:
16603 @smallexample @c ada
16607 with Gnat.Io; use Gnat.Io;
16608 with Unchecked_Deallocation;
16609 with Unchecked_Conversion;
16610 with GNAT.Debug_Pools;
16611 with System.Storage_Elements;
16612 with Ada.Exceptions; use Ada.Exceptions;
16613 procedure Debug_Pool_Test is
16615 type T is access Integer;
16616 type U is access all T;
16618 P : GNAT.Debug_Pools.Debug_Pool;
16619 for T'Storage_Pool use P;
16621 procedure Free is new Unchecked_Deallocation (Integer, T);
16622 function UC is new Unchecked_Conversion (U, T);
16625 procedure Info is new GNAT.Debug_Pools.Print_Info(Put_Line);
16635 Put_Line (Integer'Image(B.all));
16637 when E : others => Put_Line ("raised: " & Exception_Name (E));
16642 when E : others => Put_Line ("raised: " & Exception_Name (E));
16646 Put_Line (Integer'Image(B.all));
16648 when E : others => Put_Line ("raised: " & Exception_Name (E));
16653 when E : others => Put_Line ("raised: " & Exception_Name (E));
16656 end Debug_Pool_Test;
16660 The debug pool mechanism provides the following precise diagnostics on the
16661 execution of this erroneous program:
16664 Total allocated bytes : 0
16665 Total deallocated bytes : 0
16666 Current Water Mark: 0
16670 Total allocated bytes : 8
16671 Total deallocated bytes : 0
16672 Current Water Mark: 8
16675 raised: GNAT.DEBUG_POOLS.ACCESSING_DEALLOCATED_STORAGE
16676 raised: GNAT.DEBUG_POOLS.FREEING_DEALLOCATED_STORAGE
16677 raised: GNAT.DEBUG_POOLS.ACCESSING_NOT_ALLOCATED_STORAGE
16678 raised: GNAT.DEBUG_POOLS.FREEING_NOT_ALLOCATED_STORAGE
16680 Total allocated bytes : 8
16681 Total deallocated bytes : 4
16682 Current Water Mark: 4
16687 @node The gnatmem Tool
16688 @section The @command{gnatmem} Tool
16692 The @code{gnatmem} utility monitors dynamic allocation and
16693 deallocation activity in a program, and displays information about
16694 incorrect deallocations and possible sources of memory leaks.
16695 It is designed to work in association with a static runtime library
16696 only and in this context provides three types of information:
16699 General information concerning memory management, such as the total
16700 number of allocations and deallocations, the amount of allocated
16701 memory and the high water mark, i.e.@: the largest amount of allocated
16702 memory in the course of program execution.
16705 Backtraces for all incorrect deallocations, that is to say deallocations
16706 which do not correspond to a valid allocation.
16709 Information on each allocation that is potentially the origin of a memory
16714 * Running gnatmem::
16715 * Switches for gnatmem::
16716 * Example of gnatmem Usage::
16719 @node Running gnatmem
16720 @subsection Running @code{gnatmem}
16723 @code{gnatmem} makes use of the output created by the special version of
16724 allocation and deallocation routines that record call information. This
16725 allows to obtain accurate dynamic memory usage history at a minimal cost to
16726 the execution speed. Note however, that @code{gnatmem} is not supported on
16727 all platforms (currently, it is supported on AIX, HP-UX, GNU/Linux,
16728 Solaris and Windows NT/2000/XP (x86).
16731 The @code{gnatmem} command has the form
16734 @c $ gnatmem @ovar{switches} user_program
16735 @c Expanding @ovar macro inline (explanation in macro def comments)
16736 $ gnatmem @r{[}@var{switches}@r{]} @var{user_program}
16740 The program must have been linked with the instrumented version of the
16741 allocation and deallocation routines. This is done by linking with the
16742 @file{libgmem.a} library. For correct symbolic backtrace information,
16743 the user program should be compiled with debugging options
16744 (see @ref{Switches for gcc}). For example to build @file{my_program}:
16747 $ gnatmake -g my_program -largs -lgmem
16751 As library @file{libgmem.a} contains an alternate body for package
16752 @code{System.Memory}, @file{s-memory.adb} should not be compiled and linked
16753 when an executable is linked with library @file{libgmem.a}. It is then not
16754 recommended to use @command{gnatmake} with switch @option{^-a^/ALL_FILES^}.
16757 When @file{my_program} is executed, the file @file{gmem.out} is produced.
16758 This file contains information about all allocations and deallocations
16759 performed by the program. It is produced by the instrumented allocations and
16760 deallocations routines and will be used by @code{gnatmem}.
16762 In order to produce symbolic backtrace information for allocations and
16763 deallocations performed by the GNAT run-time library, you need to use a
16764 version of that library that has been compiled with the @option{-g} switch
16765 (see @ref{Rebuilding the GNAT Run-Time Library}).
16767 Gnatmem must be supplied with the @file{gmem.out} file and the executable to
16768 examine. If the location of @file{gmem.out} file was not explicitly supplied by
16769 @option{-i} switch, gnatmem will assume that this file can be found in the
16770 current directory. For example, after you have executed @file{my_program},
16771 @file{gmem.out} can be analyzed by @code{gnatmem} using the command:
16774 $ gnatmem my_program
16778 This will produce the output with the following format:
16780 *************** debut cc
16782 $ gnatmem my_program
16786 Total number of allocations : 45
16787 Total number of deallocations : 6
16788 Final Water Mark (non freed mem) : 11.29 Kilobytes
16789 High Water Mark : 11.40 Kilobytes
16794 Allocation Root # 2
16795 -------------------
16796 Number of non freed allocations : 11
16797 Final Water Mark (non freed mem) : 1.16 Kilobytes
16798 High Water Mark : 1.27 Kilobytes
16800 my_program.adb:23 my_program.alloc
16806 The first block of output gives general information. In this case, the
16807 Ada construct ``@code{@b{new}}'' was executed 45 times, and only 6 calls to an
16808 Unchecked_Deallocation routine occurred.
16811 Subsequent paragraphs display information on all allocation roots.
16812 An allocation root is a specific point in the execution of the program
16813 that generates some dynamic allocation, such as a ``@code{@b{new}}''
16814 construct. This root is represented by an execution backtrace (or subprogram
16815 call stack). By default the backtrace depth for allocations roots is 1, so
16816 that a root corresponds exactly to a source location. The backtrace can
16817 be made deeper, to make the root more specific.
16819 @node Switches for gnatmem
16820 @subsection Switches for @code{gnatmem}
16823 @code{gnatmem} recognizes the following switches:
16828 @cindex @option{-q} (@code{gnatmem})
16829 Quiet. Gives the minimum output needed to identify the origin of the
16830 memory leaks. Omits statistical information.
16833 @cindex @var{N} (@code{gnatmem})
16834 N is an integer literal (usually between 1 and 10) which controls the
16835 depth of the backtraces defining allocation root. The default value for
16836 N is 1. The deeper the backtrace, the more precise the localization of
16837 the root. Note that the total number of roots can depend on this
16838 parameter. This parameter must be specified @emph{before} the name of the
16839 executable to be analyzed, to avoid ambiguity.
16842 @cindex @option{-b} (@code{gnatmem})
16843 This switch has the same effect as just depth parameter.
16845 @item -i @var{file}
16846 @cindex @option{-i} (@code{gnatmem})
16847 Do the @code{gnatmem} processing starting from @file{file}, rather than
16848 @file{gmem.out} in the current directory.
16851 @cindex @option{-m} (@code{gnatmem})
16852 This switch causes @code{gnatmem} to mask the allocation roots that have less
16853 than n leaks. The default value is 1. Specifying the value of 0 will allow to
16854 examine even the roots that didn't result in leaks.
16857 @cindex @option{-s} (@code{gnatmem})
16858 This switch causes @code{gnatmem} to sort the allocation roots according to the
16859 specified order of sort criteria, each identified by a single letter. The
16860 currently supported criteria are @code{n, h, w} standing respectively for
16861 number of unfreed allocations, high watermark, and final watermark
16862 corresponding to a specific root. The default order is @code{nwh}.
16866 @node Example of gnatmem Usage
16867 @subsection Example of @code{gnatmem} Usage
16870 The following example shows the use of @code{gnatmem}
16871 on a simple memory-leaking program.
16872 Suppose that we have the following Ada program:
16874 @smallexample @c ada
16877 with Unchecked_Deallocation;
16878 procedure Test_Gm is
16880 type T is array (1..1000) of Integer;
16881 type Ptr is access T;
16882 procedure Free is new Unchecked_Deallocation (T, Ptr);
16885 procedure My_Alloc is
16890 procedure My_DeAlloc is
16898 for I in 1 .. 5 loop
16899 for J in I .. 5 loop
16910 The program needs to be compiled with debugging option and linked with
16911 @code{gmem} library:
16914 $ gnatmake -g test_gm -largs -lgmem
16918 Then we execute the program as usual:
16925 Then @code{gnatmem} is invoked simply with
16931 which produces the following output (result may vary on different platforms):
16936 Total number of allocations : 18
16937 Total number of deallocations : 5
16938 Final Water Mark (non freed mem) : 53.00 Kilobytes
16939 High Water Mark : 56.90 Kilobytes
16941 Allocation Root # 1
16942 -------------------
16943 Number of non freed allocations : 11
16944 Final Water Mark (non freed mem) : 42.97 Kilobytes
16945 High Water Mark : 46.88 Kilobytes
16947 test_gm.adb:11 test_gm.my_alloc
16949 Allocation Root # 2
16950 -------------------
16951 Number of non freed allocations : 1
16952 Final Water Mark (non freed mem) : 10.02 Kilobytes
16953 High Water Mark : 10.02 Kilobytes
16955 s-secsta.adb:81 system.secondary_stack.ss_init
16957 Allocation Root # 3
16958 -------------------
16959 Number of non freed allocations : 1
16960 Final Water Mark (non freed mem) : 12 Bytes
16961 High Water Mark : 12 Bytes
16963 s-secsta.adb:181 system.secondary_stack.ss_init
16967 Note that the GNAT run time contains itself a certain number of
16968 allocations that have no corresponding deallocation,
16969 as shown here for root #2 and root
16970 #3. This is a normal behavior when the number of non-freed allocations
16971 is one, it allocates dynamic data structures that the run time needs for
16972 the complete lifetime of the program. Note also that there is only one
16973 allocation root in the user program with a single line back trace:
16974 test_gm.adb:11 test_gm.my_alloc, whereas a careful analysis of the
16975 program shows that 'My_Alloc' is called at 2 different points in the
16976 source (line 21 and line 24). If those two allocation roots need to be
16977 distinguished, the backtrace depth parameter can be used:
16980 $ gnatmem 3 test_gm
16984 which will give the following output:
16989 Total number of allocations : 18
16990 Total number of deallocations : 5
16991 Final Water Mark (non freed mem) : 53.00 Kilobytes
16992 High Water Mark : 56.90 Kilobytes
16994 Allocation Root # 1
16995 -------------------
16996 Number of non freed allocations : 10
16997 Final Water Mark (non freed mem) : 39.06 Kilobytes
16998 High Water Mark : 42.97 Kilobytes
17000 test_gm.adb:11 test_gm.my_alloc
17001 test_gm.adb:24 test_gm
17002 b_test_gm.c:52 main
17004 Allocation Root # 2
17005 -------------------
17006 Number of non freed allocations : 1
17007 Final Water Mark (non freed mem) : 10.02 Kilobytes
17008 High Water Mark : 10.02 Kilobytes
17010 s-secsta.adb:81 system.secondary_stack.ss_init
17011 s-secsta.adb:283 <system__secondary_stack___elabb>
17012 b_test_gm.c:33 adainit
17014 Allocation Root # 3
17015 -------------------
17016 Number of non freed allocations : 1
17017 Final Water Mark (non freed mem) : 3.91 Kilobytes
17018 High Water Mark : 3.91 Kilobytes
17020 test_gm.adb:11 test_gm.my_alloc
17021 test_gm.adb:21 test_gm
17022 b_test_gm.c:52 main
17024 Allocation Root # 4
17025 -------------------
17026 Number of non freed allocations : 1
17027 Final Water Mark (non freed mem) : 12 Bytes
17028 High Water Mark : 12 Bytes
17030 s-secsta.adb:181 system.secondary_stack.ss_init
17031 s-secsta.adb:283 <system__secondary_stack___elabb>
17032 b_test_gm.c:33 adainit
17036 The allocation root #1 of the first example has been split in 2 roots #1
17037 and #3 thanks to the more precise associated backtrace.
17041 @node Stack Related Facilities
17042 @chapter Stack Related Facilities
17045 This chapter describes some useful tools associated with stack
17046 checking and analysis. In
17047 particular, it deals with dynamic and static stack usage measurements.
17050 * Stack Overflow Checking::
17051 * Static Stack Usage Analysis::
17052 * Dynamic Stack Usage Analysis::
17055 @node Stack Overflow Checking
17056 @section Stack Overflow Checking
17057 @cindex Stack Overflow Checking
17058 @cindex -fstack-check
17061 For most operating systems, @command{gcc} does not perform stack overflow
17062 checking by default. This means that if the main environment task or
17063 some other task exceeds the available stack space, then unpredictable
17064 behavior will occur. Most native systems offer some level of protection by
17065 adding a guard page at the end of each task stack. This mechanism is usually
17066 not enough for dealing properly with stack overflow situations because
17067 a large local variable could ``jump'' above the guard page.
17068 Furthermore, when the
17069 guard page is hit, there may not be any space left on the stack for executing
17070 the exception propagation code. Enabling stack checking avoids
17073 To activate stack checking, compile all units with the gcc option
17074 @option{-fstack-check}. For example:
17077 gcc -c -fstack-check package1.adb
17081 Units compiled with this option will generate extra instructions to check
17082 that any use of the stack (for procedure calls or for declaring local
17083 variables in declare blocks) does not exceed the available stack space.
17084 If the space is exceeded, then a @code{Storage_Error} exception is raised.
17086 For declared tasks, the stack size is controlled by the size
17087 given in an applicable @code{Storage_Size} pragma or by the value specified
17088 at bind time with @option{-d} (@pxref{Switches for gnatbind}) or is set to
17089 the default size as defined in the GNAT runtime otherwise.
17091 For the environment task, the stack size depends on
17092 system defaults and is unknown to the compiler. Stack checking
17093 may still work correctly if a fixed
17094 size stack is allocated, but this cannot be guaranteed.
17096 To ensure that a clean exception is signalled for stack
17097 overflow, set the environment variable
17098 @env{GNAT_STACK_LIMIT} to indicate the maximum
17099 stack area that can be used, as in:
17100 @cindex GNAT_STACK_LIMIT
17103 SET GNAT_STACK_LIMIT 1600
17107 The limit is given in kilobytes, so the above declaration would
17108 set the stack limit of the environment task to 1.6 megabytes.
17109 Note that the only purpose of this usage is to limit the amount
17110 of stack used by the environment task. If it is necessary to
17111 increase the amount of stack for the environment task, then this
17112 is an operating systems issue, and must be addressed with the
17113 appropriate operating systems commands.
17116 To have a fixed size stack in the environment task, the stack must be put
17117 in the P0 address space and its size specified. Use these switches to
17121 gnatmake my_progs -largs "-Wl,--opt=STACK=4000,/p0image"
17125 The quotes are required to keep case. The number after @samp{STACK=} is the
17126 size of the environmental task stack in pagelets (512 bytes). In this example
17127 the stack size is about 2 megabytes.
17130 A consequence of the @option{/p0image} qualifier is also to makes RMS buffers
17131 be placed in P0 space. Refer to @cite{HP OpenVMS Linker Utility Manual} for
17132 more details about the @option{/p0image} qualifier and the @option{stack}
17136 @node Static Stack Usage Analysis
17137 @section Static Stack Usage Analysis
17138 @cindex Static Stack Usage Analysis
17139 @cindex -fstack-usage
17142 A unit compiled with @option{-fstack-usage} will generate an extra file
17144 the maximum amount of stack used, on a per-function basis.
17145 The file has the same
17146 basename as the target object file with a @file{.su} extension.
17147 Each line of this file is made up of three fields:
17151 The name of the function.
17155 One or more qualifiers: @code{static}, @code{dynamic}, @code{bounded}.
17158 The second field corresponds to the size of the known part of the function
17161 The qualifier @code{static} means that the function frame size
17163 It usually means that all local variables have a static size.
17164 In this case, the second field is a reliable measure of the function stack
17167 The qualifier @code{dynamic} means that the function frame size is not static.
17168 It happens mainly when some local variables have a dynamic size. When this
17169 qualifier appears alone, the second field is not a reliable measure
17170 of the function stack analysis. When it is qualified with @code{bounded}, it
17171 means that the second field is a reliable maximum of the function stack
17174 @node Dynamic Stack Usage Analysis
17175 @section Dynamic Stack Usage Analysis
17178 It is possible to measure the maximum amount of stack used by a task, by
17179 adding a switch to @command{gnatbind}, as:
17182 $ gnatbind -u0 file
17186 With this option, at each task termination, its stack usage is output on
17188 It is not always convenient to output the stack usage when the program
17189 is still running. Hence, it is possible to delay this output until program
17190 termination. for a given number of tasks specified as the argument of the
17191 @option{-u} option. For instance:
17194 $ gnatbind -u100 file
17198 will buffer the stack usage information of the first 100 tasks to terminate and
17199 output this info at program termination. Results are displayed in four
17203 Index | Task Name | Stack Size | Stack Usage [Value +/- Variation]
17210 is a number associated with each task.
17213 is the name of the task analyzed.
17216 is the maximum size for the stack.
17219 is the measure done by the stack analyzer. In order to prevent overflow, the stack
17220 is not entirely analyzed, and it's not possible to know exactly how
17221 much has actually been used. The report thus contains the theoretical stack usage
17222 (Value) and the possible variation (Variation) around this value.
17227 The environment task stack, e.g., the stack that contains the main unit, is
17228 only processed when the environment variable GNAT_STACK_LIMIT is set.
17231 @c *********************************
17233 @c *********************************
17234 @node Verifying Properties Using gnatcheck
17235 @chapter Verifying Properties Using @command{gnatcheck}
17237 @cindex @command{gnatcheck}
17240 The @command{gnatcheck} tool is an ASIS-based utility that checks properties
17241 of Ada source files according to a given set of semantic rules.
17244 In order to check compliance with a given rule, @command{gnatcheck} has to
17245 semantically analyze the Ada sources.
17246 Therefore, checks can only be performed on
17247 legal Ada units. Moreover, when a unit depends semantically upon units located
17248 outside the current directory, the source search path has to be provided when
17249 calling @command{gnatcheck}, either through a specified project file or
17250 through @command{gnatcheck} switches.
17252 A number of rules are predefined in @command{gnatcheck} and are described
17253 later in this chapter.
17255 For full details, refer to @cite{GNATcheck Reference Manual} document.
17258 @c *********************************
17259 @node Creating Sample Bodies Using gnatstub
17260 @chapter Creating Sample Bodies Using @command{gnatstub}
17264 @command{gnatstub} creates body stubs, that is, empty but compilable bodies
17265 for library unit declarations.
17267 Note: to invoke @code{gnatstub} with a project file, use the @code{gnat}
17268 driver (see @ref{The GNAT Driver and Project Files}).
17270 To create a body stub, @command{gnatstub} has to compile the library
17271 unit declaration. Therefore, bodies can be created only for legal
17272 library units. Moreover, if a library unit depends semantically upon
17273 units located outside the current directory, you have to provide
17274 the source search path when calling @command{gnatstub}, see the description
17275 of @command{gnatstub} switches below.
17277 By default, all the program unit body stubs generated by @code{gnatstub}
17278 raise the predefined @code{Program_Error} exception, which will catch
17279 accidental calls of generated stubs. This behavior can be changed with
17280 option @option{^--no-exception^/NO_EXCEPTION^} (see below).
17283 * Running gnatstub::
17284 * Switches for gnatstub::
17287 @node Running gnatstub
17288 @section Running @command{gnatstub}
17291 @command{gnatstub} has the command-line interface of the form
17294 @c $ gnatstub @ovar{switches} @var{filename} @ovar{directory}
17295 @c Expanding @ovar macro inline (explanation in macro def comments)
17296 $ gnatstub @r{[}@var{switches}@r{]} @var{filename} @r{[}@var{directory}@r{]} @r{[}-cargs @var{gcc_switches}@r{]}
17303 is the name of the source file that contains a library unit declaration
17304 for which a body must be created. The file name may contain the path
17306 The file name does not have to follow the GNAT file name conventions. If the
17308 does not follow GNAT file naming conventions, the name of the body file must
17310 explicitly as the value of the @option{^-o^/BODY=^@var{body-name}} option.
17311 If the file name follows the GNAT file naming
17312 conventions and the name of the body file is not provided,
17315 of the body file from the argument file name by replacing the @file{.ads}
17317 with the @file{.adb} suffix.
17320 indicates the directory in which the body stub is to be placed (the default
17324 @item @samp{@var{gcc_switches}} is a list of switches for
17325 @command{gcc}. They will be passed on to all compiler invocations made by
17326 @command{gnatelim} to generate the ASIS trees. Here you can provide
17327 @option{^-I^/INCLUDE_DIRS=^} switches to form the source search path,
17328 use the @option{-gnatec} switch to set the configuration file,
17329 use the @option{-gnat05} switch if sources should be compiled in
17333 is an optional sequence of switches as described in the next section
17336 @node Switches for gnatstub
17337 @section Switches for @command{gnatstub}
17343 @cindex @option{^-f^/FULL^} (@command{gnatstub})
17344 If the destination directory already contains a file with the name of the
17346 for the argument spec file, replace it with the generated body stub.
17348 @item ^-hs^/HEADER=SPEC^
17349 @cindex @option{^-hs^/HEADER=SPEC^} (@command{gnatstub})
17350 Put the comment header (i.e., all the comments preceding the
17351 compilation unit) from the source of the library unit declaration
17352 into the body stub.
17354 @item ^-hg^/HEADER=GENERAL^
17355 @cindex @option{^-hg^/HEADER=GENERAL^} (@command{gnatstub})
17356 Put a sample comment header into the body stub.
17358 @item ^--header-file=@var{filename}^/FROM_HEADER_FILE=@var{filename}^
17359 @cindex @option{^--header-file^/FROM_HEADER_FILE=^} (@command{gnatstub})
17360 Use the content of the file as the comment header for a generated body stub.
17364 @cindex @option{-IDIR} (@command{gnatstub})
17366 @cindex @option{-I-} (@command{gnatstub})
17369 @item /NOCURRENT_DIRECTORY
17370 @cindex @option{/NOCURRENT_DIRECTORY} (@command{gnatstub})
17372 ^These switches have ^This switch has^ the same meaning as in calls to
17374 ^They define ^It defines ^ the source search path in the call to
17375 @command{gcc} issued
17376 by @command{gnatstub} to compile an argument source file.
17378 @item ^-gnatec^/CONFIGURATION_PRAGMAS_FILE=^@var{PATH}
17379 @cindex @option{^-gnatec^/CONFIGURATION_PRAGMAS_FILE^} (@command{gnatstub})
17380 This switch has the same meaning as in calls to @command{gcc}.
17381 It defines the additional configuration file to be passed to the call to
17382 @command{gcc} issued
17383 by @command{gnatstub} to compile an argument source file.
17385 @item ^-gnatyM^/MAX_LINE_LENGTH=^@var{n}
17386 @cindex @option{^-gnatyM^/MAX_LINE_LENGTH^} (@command{gnatstub})
17387 (@var{n} is a non-negative integer). Set the maximum line length in the
17388 body stub to @var{n}; the default is 79. The maximum value that can be
17389 specified is 32767. Note that in the special case of configuration
17390 pragma files, the maximum is always 32767 regardless of whether or
17391 not this switch appears.
17393 @item ^-gnaty^/STYLE_CHECKS=^@var{n}
17394 @cindex @option{^-gnaty^/STYLE_CHECKS=^} (@command{gnatstub})
17395 (@var{n} is a non-negative integer from 1 to 9). Set the indentation level in
17396 the generated body sample to @var{n}.
17397 The default indentation is 3.
17399 @item ^-gnatyo^/ORDERED_SUBPROGRAMS^
17400 @cindex @option{^-gnato^/ORDERED_SUBPROGRAMS^} (@command{gnatstub})
17401 Order local bodies alphabetically. (By default local bodies are ordered
17402 in the same way as the corresponding local specs in the argument spec file.)
17404 @item ^-i^/INDENTATION=^@var{n}
17405 @cindex @option{^-i^/INDENTATION^} (@command{gnatstub})
17406 Same as @option{^-gnaty^/STYLE_CHECKS=^@var{n}}
17408 @item ^-k^/TREE_FILE=SAVE^
17409 @cindex @option{^-k^/TREE_FILE=SAVE^} (@command{gnatstub})
17410 Do not remove the tree file (i.e., the snapshot of the compiler internal
17411 structures used by @command{gnatstub}) after creating the body stub.
17413 @item ^-l^/LINE_LENGTH=^@var{n}
17414 @cindex @option{^-l^/LINE_LENGTH^} (@command{gnatstub})
17415 Same as @option{^-gnatyM^/MAX_LINE_LENGTH=^@var{n}}
17417 @item ^--no-exception^/NO_EXCEPTION^
17418 @cindex @option{^--no-exception^/NO_EXCEPTION^} (@command{gnatstub})
17419 Avoind raising PROGRAM_ERROR in the generated bodies of program unit stubs.
17420 This is not always possible for function stubs.
17422 @item ^--no-local-header^/NO_LOCAL_HEADER^
17423 @cindex @option{^--no-local-header^/NO_LOCAL_HEADER^} (@command{gnatstub})
17424 Do not place local comment header with unit name before body stub for a
17427 @item ^-o ^/BODY=^@var{body-name}
17428 @cindex @option{^-o^/BODY^} (@command{gnatstub})
17429 Body file name. This should be set if the argument file name does not
17431 the GNAT file naming
17432 conventions. If this switch is omitted the default name for the body will be
17434 from the argument file name according to the GNAT file naming conventions.
17437 @cindex @option{^-q^/QUIET^} (@command{gnatstub})
17438 Quiet mode: do not generate a confirmation when a body is
17439 successfully created, and do not generate a message when a body is not
17443 @item ^-r^/TREE_FILE=REUSE^
17444 @cindex @option{^-r^/TREE_FILE=REUSE^} (@command{gnatstub})
17445 Reuse the tree file (if it exists) instead of creating it. Instead of
17446 creating the tree file for the library unit declaration, @command{gnatstub}
17447 tries to find it in the current directory and use it for creating
17448 a body. If the tree file is not found, no body is created. This option
17449 also implies @option{^-k^/SAVE^}, whether or not
17450 the latter is set explicitly.
17452 @item ^-t^/TREE_FILE=OVERWRITE^
17453 @cindex @option{^-t^/TREE_FILE=OVERWRITE^} (@command{gnatstub})
17454 Overwrite the existing tree file. If the current directory already
17455 contains the file which, according to the GNAT file naming rules should
17456 be considered as a tree file for the argument source file,
17458 will refuse to create the tree file needed to create a sample body
17459 unless this option is set.
17461 @item ^-v^/VERBOSE^
17462 @cindex @option{^-v^/VERBOSE^} (@command{gnatstub})
17463 Verbose mode: generate version information.
17467 @c *********************************
17468 @node Generating Ada Bindings for C and C++ headers
17469 @chapter Generating Ada Bindings for C and C++ headers
17473 GNAT now comes with a binding generator for C and C++ headers which is
17474 intended to do 95% of the tedious work of generating Ada specs from C
17475 or C++ header files.
17477 Note that this capability is not intended to generate 100% correct Ada specs,
17478 and will is some cases require manual adjustments, although it can often
17479 be used out of the box in practice.
17481 Some of the known limitations include:
17484 @item only very simple character constant macros are translated into Ada
17485 constants. Function macros (macros with arguments) are partially translated
17486 as comments, to be completed manually if needed.
17487 @item some extensions (e.g. vector types) are not supported
17488 @item pointers to pointers or complex structures are mapped to System.Address
17489 @item identifiers with identical name (except casing) will generate compilation
17490 errors (e.g. @code{shm_get} vs @code{SHM_GET}).
17493 The code generated is using the Ada 2005 syntax, which makes it
17494 easier to interface with other languages than previous versions of Ada.
17497 * Running the binding generator::
17498 * Generating bindings for C++ headers::
17502 @node Running the binding generator
17503 @section Running the binding generator
17506 The binding generator is part of the @command{gcc} compiler and can be
17507 invoked via the @option{-fdump-ada-spec} switch, which will generate Ada
17508 spec files for the header files specified on the command line, and all
17509 header files needed by these files transitivitely. For example:
17512 $ g++ -c -fdump-ada-spec -C /usr/include/time.h
17513 $ gcc -c -gnat05 *.ads
17516 will generate, under GNU/Linux, the following files: @file{time_h.ads},
17517 @file{bits_time_h.ads}, @file{stddef_h.ads}, @file{bits_types_h.ads} which
17518 correspond to the files @file{/usr/include/time.h},
17519 @file{/usr/include/bits/time.h}, etc@dots{}, and will then compile in Ada 2005
17520 mode these Ada specs.
17522 The @code{-C} switch tells @command{gcc} to extract comments from headers,
17523 and will attempt to generate corresponding Ada comments.
17525 If you want to generate a single Ada file and not the transitive closure, you
17526 can use instead the @option{-fdump-ada-spec-slim} switch.
17528 Note that we recommend when possible to use the @command{g++} driver to
17529 generate bindings, even for most C headers, since this will in general
17530 generate better Ada specs. For generating bindings for C++ headers, it is
17531 mandatory to use the @command{g++} command, or @command{gcc -x c++} which
17532 is equivalent in this case. If @command{g++} cannot work on your C headers
17533 because of incompatibilities between C and C++, then you can fallback to
17534 @command{gcc} instead.
17536 For an example of better bindings generated from the C++ front-end,
17537 the name of the parameters (when available) are actually ignored by the C
17538 front-end. Consider the following C header:
17541 extern void foo (int variable);
17544 with the C front-end, @code{variable} is ignored, and the above is handled as:
17547 extern void foo (int);
17550 generating a generic:
17553 procedure foo (param1 : int);
17556 with the C++ front-end, the name is available, and we generate:
17559 procedure foo (variable : int);
17562 In some cases, the generated bindings will be more complete or more meaningful
17563 when defining some macros, which you can do via the @option{-D} switch. This
17564 is for example the case with @file{Xlib.h} under GNU/Linux:
17567 g++ -c -fdump-ada-spec -DXLIB_ILLEGAL_ACCESS -C /usr/include/X11/Xlib.h
17570 The above will generate more complete bindings than a straight call without
17571 the @option{-DXLIB_ILLEGAL_ACCESS} switch.
17573 In other cases, it is not possible to parse a header file in a stand alone
17574 manner, because other include files need to be included first. In this
17575 case, the solution is to create a small header file including the needed
17576 @code{#include} and possible @code{#define} directives. For example, to
17577 generate Ada bindings for @file{readline/readline.h}, you need to first
17578 include @file{stdio.h}, so you can create a file with the following two
17579 lines in e.g. @file{readline1.h}:
17583 #include <readline/readline.h>
17586 and then generate Ada bindings from this file:
17589 $ g++ -c -fdump-ada-spec readline1.h
17592 @node Generating bindings for C++ headers
17593 @section Generating bindings for C++ headers
17596 Generating bindings for C++ headers is done using the same options, always
17597 with the @command{g++} compiler.
17599 In this mode, C++ classes will be mapped to Ada tagged types, constructors
17600 will be mapped using the @code{CPP_Constructor} pragma, and when possible,
17601 multiple inheritance of abstract classes will be mapped to Ada interfaces
17602 (@xref{Interfacing to C++,,,gnat_rm, GNAT Reference Manual}, for additional
17603 information on interfacing to C++).
17605 For example, given the following C++ header file:
17612 virtual int Number_Of_Teeth () = 0;
17617 virtual void Set_Owner (char* Name) = 0;
17623 virtual void Set_Age (int New_Age);
17626 class Dog : Animal, Carnivore, Domestic @{
17631 virtual int Number_Of_Teeth ();
17632 virtual void Set_Owner (char* Name);
17640 The corresponding Ada code is generated:
17642 @smallexample @c ada
17645 package Class_Carnivore is
17646 type Carnivore is limited interface;
17647 pragma Import (CPP, Carnivore);
17649 function Number_Of_Teeth (this : access Carnivore) return int is abstract;
17651 use Class_Carnivore;
17653 package Class_Domestic is
17654 type Domestic is limited interface;
17655 pragma Import (CPP, Domestic);
17657 procedure Set_Owner
17658 (this : access Domestic;
17659 Name : Interfaces.C.Strings.chars_ptr) is abstract;
17661 use Class_Domestic;
17663 package Class_Animal is
17664 type Animal is tagged limited record
17665 Age_Count : aliased int;
17667 pragma Import (CPP, Animal);
17669 procedure Set_Age (this : access Animal; New_Age : int);
17670 pragma Import (CPP, Set_Age, "_ZN6Animal7Set_AgeEi");
17674 package Class_Dog is
17675 type Dog is new Animal and Carnivore and Domestic with record
17676 Tooth_Count : aliased int;
17677 Owner : Interfaces.C.Strings.chars_ptr;
17679 pragma Import (CPP, Dog);
17681 function Number_Of_Teeth (this : access Dog) return int;
17682 pragma Import (CPP, Number_Of_Teeth, "_ZN3Dog15Number_Of_TeethEv");
17684 procedure Set_Owner
17685 (this : access Dog; Name : Interfaces.C.Strings.chars_ptr);
17686 pragma Import (CPP, Set_Owner, "_ZN3Dog9Set_OwnerEPc");
17688 function New_Dog return Dog;
17689 pragma CPP_Constructor (New_Dog);
17690 pragma Import (CPP, New_Dog, "_ZN3DogC1Ev");
17701 @item -fdump-ada-spec
17702 @cindex @option{-fdump-ada-spec} (@command{gcc})
17703 Generate Ada spec files for the given header files transitively (including
17704 all header files that these headers depend upon).
17706 @item -fdump-ada-spec-slim
17707 @cindex @option{-fdump-ada-spec-slim} (@command{gcc})
17708 Generate Ada spec files for the header files specified on the command line
17712 @cindex @option{-C} (@command{gcc})
17713 Extract comments from headers and generate Ada comments in the Ada spec files.
17716 @node Other Utility Programs
17717 @chapter Other Utility Programs
17720 This chapter discusses some other utility programs available in the Ada
17724 * Using Other Utility Programs with GNAT::
17725 * The External Symbol Naming Scheme of GNAT::
17726 * Converting Ada Files to html with gnathtml::
17727 * Installing gnathtml::
17734 @node Using Other Utility Programs with GNAT
17735 @section Using Other Utility Programs with GNAT
17738 The object files generated by GNAT are in standard system format and in
17739 particular the debugging information uses this format. This means
17740 programs generated by GNAT can be used with existing utilities that
17741 depend on these formats.
17744 In general, any utility program that works with C will also often work with
17745 Ada programs generated by GNAT. This includes software utilities such as
17746 gprof (a profiling program), @code{gdb} (the FSF debugger), and utilities such
17750 @node The External Symbol Naming Scheme of GNAT
17751 @section The External Symbol Naming Scheme of GNAT
17754 In order to interpret the output from GNAT, when using tools that are
17755 originally intended for use with other languages, it is useful to
17756 understand the conventions used to generate link names from the Ada
17759 All link names are in all lowercase letters. With the exception of library
17760 procedure names, the mechanism used is simply to use the full expanded
17761 Ada name with dots replaced by double underscores. For example, suppose
17762 we have the following package spec:
17764 @smallexample @c ada
17775 The variable @code{MN} has a full expanded Ada name of @code{QRS.MN}, so
17776 the corresponding link name is @code{qrs__mn}.
17778 Of course if a @code{pragma Export} is used this may be overridden:
17780 @smallexample @c ada
17785 pragma Export (Var1, C, External_Name => "var1_name");
17787 pragma Export (Var2, C, Link_Name => "var2_link_name");
17794 In this case, the link name for @var{Var1} is whatever link name the
17795 C compiler would assign for the C function @var{var1_name}. This typically
17796 would be either @var{var1_name} or @var{_var1_name}, depending on operating
17797 system conventions, but other possibilities exist. The link name for
17798 @var{Var2} is @var{var2_link_name}, and this is not operating system
17802 One exception occurs for library level procedures. A potential ambiguity
17803 arises between the required name @code{_main} for the C main program,
17804 and the name we would otherwise assign to an Ada library level procedure
17805 called @code{Main} (which might well not be the main program).
17807 To avoid this ambiguity, we attach the prefix @code{_ada_} to such
17808 names. So if we have a library level procedure such as
17810 @smallexample @c ada
17813 procedure Hello (S : String);
17819 the external name of this procedure will be @var{_ada_hello}.
17822 @node Converting Ada Files to html with gnathtml
17823 @section Converting Ada Files to HTML with @code{gnathtml}
17826 This @code{Perl} script allows Ada source files to be browsed using
17827 standard Web browsers. For installation procedure, see the section
17828 @xref{Installing gnathtml}.
17830 Ada reserved keywords are highlighted in a bold font and Ada comments in
17831 a blue font. Unless your program was compiled with the gcc @option{-gnatx}
17832 switch to suppress the generation of cross-referencing information, user
17833 defined variables and types will appear in a different color; you will
17834 be able to click on any identifier and go to its declaration.
17836 The command line is as follow:
17838 @c $ perl gnathtml.pl @ovar{^switches^options^} @var{ada-files}
17839 @c Expanding @ovar macro inline (explanation in macro def comments)
17840 $ perl gnathtml.pl @r{[}@var{^switches^options^}@r{]} @var{ada-files}
17844 You can pass it as many Ada files as you want. @code{gnathtml} will generate
17845 an html file for every ada file, and a global file called @file{index.htm}.
17846 This file is an index of every identifier defined in the files.
17848 The available ^switches^options^ are the following ones:
17852 @cindex @option{-83} (@code{gnathtml})
17853 Only the Ada 83 subset of keywords will be highlighted.
17855 @item -cc @var{color}
17856 @cindex @option{-cc} (@code{gnathtml})
17857 This option allows you to change the color used for comments. The default
17858 value is green. The color argument can be any name accepted by html.
17861 @cindex @option{-d} (@code{gnathtml})
17862 If the Ada files depend on some other files (for instance through
17863 @code{with} clauses, the latter files will also be converted to html.
17864 Only the files in the user project will be converted to html, not the files
17865 in the run-time library itself.
17868 @cindex @option{-D} (@code{gnathtml})
17869 This command is the same as @option{-d} above, but @command{gnathtml} will
17870 also look for files in the run-time library, and generate html files for them.
17872 @item -ext @var{extension}
17873 @cindex @option{-ext} (@code{gnathtml})
17874 This option allows you to change the extension of the generated HTML files.
17875 If you do not specify an extension, it will default to @file{htm}.
17878 @cindex @option{-f} (@code{gnathtml})
17879 By default, gnathtml will generate html links only for global entities
17880 ('with'ed units, global variables and types,@dots{}). If you specify
17881 @option{-f} on the command line, then links will be generated for local
17884 @item -l @var{number}
17885 @cindex @option{-l} (@code{gnathtml})
17886 If this ^switch^option^ is provided and @var{number} is not 0, then
17887 @code{gnathtml} will number the html files every @var{number} line.
17890 @cindex @option{-I} (@code{gnathtml})
17891 Specify a directory to search for library files (@file{.ALI} files) and
17892 source files. You can provide several -I switches on the command line,
17893 and the directories will be parsed in the order of the command line.
17896 @cindex @option{-o} (@code{gnathtml})
17897 Specify the output directory for html files. By default, gnathtml will
17898 saved the generated html files in a subdirectory named @file{html/}.
17900 @item -p @var{file}
17901 @cindex @option{-p} (@code{gnathtml})
17902 If you are using Emacs and the most recent Emacs Ada mode, which provides
17903 a full Integrated Development Environment for compiling, checking,
17904 running and debugging applications, you may use @file{.gpr} files
17905 to give the directories where Emacs can find sources and object files.
17907 Using this ^switch^option^, you can tell gnathtml to use these files.
17908 This allows you to get an html version of your application, even if it
17909 is spread over multiple directories.
17911 @item -sc @var{color}
17912 @cindex @option{-sc} (@code{gnathtml})
17913 This ^switch^option^ allows you to change the color used for symbol
17915 The default value is red. The color argument can be any name accepted by html.
17917 @item -t @var{file}
17918 @cindex @option{-t} (@code{gnathtml})
17919 This ^switch^option^ provides the name of a file. This file contains a list of
17920 file names to be converted, and the effect is exactly as though they had
17921 appeared explicitly on the command line. This
17922 is the recommended way to work around the command line length limit on some
17927 @node Installing gnathtml
17928 @section Installing @code{gnathtml}
17931 @code{Perl} needs to be installed on your machine to run this script.
17932 @code{Perl} is freely available for almost every architecture and
17933 Operating System via the Internet.
17935 On Unix systems, you may want to modify the first line of the script
17936 @code{gnathtml}, to explicitly tell the Operating system where Perl
17937 is. The syntax of this line is:
17939 #!full_path_name_to_perl
17943 Alternatively, you may run the script using the following command line:
17946 @c $ perl gnathtml.pl @ovar{switches} @var{files}
17947 @c Expanding @ovar macro inline (explanation in macro def comments)
17948 $ perl gnathtml.pl @r{[}@var{switches}@r{]} @var{files}
17957 The GNAT distribution provides an Ada 95 template for the HP Language
17958 Sensitive Editor (LSE), a component of DECset. In order to
17959 access it, invoke LSE with the qualifier /ENVIRONMENT=GNU:[LIB]ADA95.ENV.
17966 GNAT supports The HP Performance Coverage Analyzer (PCA), a component
17967 of DECset. To use it proceed as outlined under ``HELP PCA'', except for running
17968 the collection phase with the /DEBUG qualifier.
17971 $ GNAT MAKE /DEBUG <PROGRAM_NAME>
17972 $ DEFINE LIB$DEBUG PCA$COLLECTOR
17973 $ RUN/DEBUG <PROGRAM_NAME>
17979 @c ******************************
17980 @node Code Coverage and Profiling
17981 @chapter Code Coverage and Profiling
17982 @cindex Code Coverage
17986 This chapter describes how to use @code{gcov} - coverage testing tool - and
17987 @code{gprof} - profiler tool - on your Ada programs.
17990 * Code Coverage of Ada Programs using gcov::
17991 * Profiling an Ada Program using gprof::
17994 @node Code Coverage of Ada Programs using gcov
17995 @section Code Coverage of Ada Programs using gcov
17997 @cindex -fprofile-arcs
17998 @cindex -ftest-coverage
18000 @cindex Code Coverage
18003 @code{gcov} is a test coverage program: it analyzes the execution of a given
18004 program on selected tests, to help you determine the portions of the program
18005 that are still untested.
18007 @code{gcov} is part of the GCC suite, and is described in detail in the GCC
18008 User's Guide. You can refer to this documentation for a more complete
18011 This chapter provides a quick startup guide, and
18012 details some Gnat-specific features.
18015 * Quick startup guide::
18019 @node Quick startup guide
18020 @subsection Quick startup guide
18022 In order to perform coverage analysis of a program using @code{gcov}, 3
18027 Code instrumentation during the compilation process
18029 Execution of the instrumented program
18031 Execution of the @code{gcov} tool to generate the result.
18034 The code instrumentation needed by gcov is created at the object level:
18035 The source code is not modified in any way, because the instrumentation code is
18036 inserted by gcc during the compilation process. To compile your code with code
18037 coverage activated, you need to recompile your whole project using the
18039 @code{-fprofile-arcs} and @code{-ftest-coverage}, and link it using
18040 @code{-fprofile-arcs}.
18043 $ gnatmake -P my_project.gpr -f -cargs -fprofile-arcs -ftest-coverage \
18044 -largs -fprofile-arcs
18047 This compilation process will create @file{.gcno} files together with
18048 the usual object files.
18050 Once the program is compiled with coverage instrumentation, you can
18051 run it as many times as needed - on portions of a test suite for
18052 example. The first execution will produce @file{.gcda} files at the
18053 same location as the @file{.gcno} files. The following executions
18054 will update those files, so that a cumulative result of the covered
18055 portions of the program is generated.
18057 Finally, you need to call the @code{gcov} tool. The different options of
18058 @code{gcov} are available in the GCC User's Guide, section 'Invoking gcov'.
18060 This will create annotated source files with a @file{.gcov} extension:
18061 @file{my_main.adb} file will be analysed in @file{my_main.adb.gcov}.
18063 @node Gnat specifics
18064 @subsection Gnat specifics
18066 Because Ada semantics, portions of the source code may be shared among
18067 several object files. This is the case for example when generics are
18068 involved, when inlining is active or when declarations generate initialisation
18069 calls. In order to take
18070 into account this shared code, you need to call @code{gcov} on all
18071 source files of the tested program at once.
18073 The list of source files might exceed the system's maximum command line
18074 length. In order to bypass this limitation, a new mechanism has been
18075 implemented in @code{gcov}: you can now list all your project's files into a
18076 text file, and provide this file to gcov as a parameter, preceded by a @@
18077 (e.g. @samp{gcov @@mysrclist.txt}).
18079 Note that on AIX compiling a static library with @code{-fprofile-arcs} is
18080 not supported as there can be unresolved symbols during the final link.
18082 @node Profiling an Ada Program using gprof
18083 @section Profiling an Ada Program using gprof
18089 This section is not meant to be an exhaustive documentation of @code{gprof}.
18090 Full documentation for it can be found in the GNU Profiler User's Guide
18091 documentation that is part of this GNAT distribution.
18093 Profiling a program helps determine the parts of a program that are executed
18094 most often, and are therefore the most time-consuming.
18096 @code{gprof} is the standard GNU profiling tool; it has been enhanced to
18097 better handle Ada programs and multitasking.
18098 It is currently supported on the following platforms
18103 solaris sparc/sparc64/x86
18109 In order to profile a program using @code{gprof}, 3 steps are needed:
18113 Code instrumentation, requiring a full recompilation of the project with the
18116 Execution of the program under the analysis conditions, i.e. with the desired
18119 Analysis of the results using the @code{gprof} tool.
18123 The following sections detail the different steps, and indicate how
18124 to interpret the results:
18126 * Compilation for profiling::
18127 * Program execution::
18129 * Interpretation of profiling results::
18132 @node Compilation for profiling
18133 @subsection Compilation for profiling
18137 In order to profile a program the first step is to tell the compiler
18138 to generate the necessary profiling information. The compiler switch to be used
18139 is @code{-pg}, which must be added to other compilation switches. This
18140 switch needs to be specified both during compilation and link stages, and can
18141 be specified once when using gnatmake:
18144 gnatmake -f -pg -P my_project
18148 Note that only the objects that were compiled with the @samp{-pg} switch will
18149 be profiled; if you need to profile your whole project, use the @samp{-f}
18150 gnatmake switch to force full recompilation.
18152 @node Program execution
18153 @subsection Program execution
18156 Once the program has been compiled for profiling, you can run it as usual.
18158 The only constraint imposed by profiling is that the program must terminate
18159 normally. An interrupted program (via a Ctrl-C, kill, etc.) will not be
18162 Once the program completes execution, a data file called @file{gmon.out} is
18163 generated in the directory where the program was launched from. If this file
18164 already exists, it will be overwritten.
18166 @node Running gprof
18167 @subsection Running gprof
18170 The @code{gprof} tool is called as follow:
18173 gprof my_prog gmon.out
18184 The complete form of the gprof command line is the following:
18187 gprof [^switches^options^] [executable [data-file]]
18191 @code{gprof} supports numerous ^switch^options^. The order of these
18192 ^switch^options^ does not matter. The full list of options can be found in
18193 the GNU Profiler User's Guide documentation that comes with this documentation.
18195 The following is the subset of those switches that is most relevant:
18199 @item --demangle[=@var{style}]
18200 @itemx --no-demangle
18201 @cindex @option{--demangle} (@code{gprof})
18202 These options control whether symbol names should be demangled when
18203 printing output. The default is to demangle C++ symbols. The
18204 @code{--no-demangle} option may be used to turn off demangling. Different
18205 compilers have different mangling styles. The optional demangling style
18206 argument can be used to choose an appropriate demangling style for your
18207 compiler, in particular Ada symbols generated by GNAT can be demangled using
18208 @code{--demangle=gnat}.
18210 @item -e @var{function_name}
18211 @cindex @option{-e} (@code{gprof})
18212 The @samp{-e @var{function}} option tells @code{gprof} not to print
18213 information about the function @var{function_name} (and its
18214 children@dots{}) in the call graph. The function will still be listed
18215 as a child of any functions that call it, but its index number will be
18216 shown as @samp{[not printed]}. More than one @samp{-e} option may be
18217 given; only one @var{function_name} may be indicated with each @samp{-e}
18220 @item -E @var{function_name}
18221 @cindex @option{-E} (@code{gprof})
18222 The @code{-E @var{function}} option works like the @code{-e} option, but
18223 execution time spent in the function (and children who were not called from
18224 anywhere else), will not be used to compute the percentages-of-time for
18225 the call graph. More than one @samp{-E} option may be given; only one
18226 @var{function_name} may be indicated with each @samp{-E} option.
18228 @item -f @var{function_name}
18229 @cindex @option{-f} (@code{gprof})
18230 The @samp{-f @var{function}} option causes @code{gprof} to limit the
18231 call graph to the function @var{function_name} and its children (and
18232 their children@dots{}). More than one @samp{-f} option may be given;
18233 only one @var{function_name} may be indicated with each @samp{-f}
18236 @item -F @var{function_name}
18237 @cindex @option{-F} (@code{gprof})
18238 The @samp{-F @var{function}} option works like the @code{-f} option, but
18239 only time spent in the function and its children (and their
18240 children@dots{}) will be used to determine total-time and
18241 percentages-of-time for the call graph. More than one @samp{-F} option
18242 may be given; only one @var{function_name} may be indicated with each
18243 @samp{-F} option. The @samp{-F} option overrides the @samp{-E} option.
18247 @node Interpretation of profiling results
18248 @subsection Interpretation of profiling results
18252 The results of the profiling analysis are represented by two arrays: the
18253 'flat profile' and the 'call graph'. Full documentation of those outputs
18254 can be found in the GNU Profiler User's Guide.
18256 The flat profile shows the time spent in each function of the program, and how
18257 many time it has been called. This allows you to locate easily the most
18258 time-consuming functions.
18260 The call graph shows, for each subprogram, the subprograms that call it,
18261 and the subprograms that it calls. It also provides an estimate of the time
18262 spent in each of those callers/called subprograms.
18265 @c ******************************
18266 @node Running and Debugging Ada Programs
18267 @chapter Running and Debugging Ada Programs
18271 This chapter discusses how to debug Ada programs.
18273 It applies to GNAT on the Alpha OpenVMS platform;
18274 for I64 OpenVMS please refer to the @cite{OpenVMS Debugger Manual},
18275 since HP has implemented Ada support in the OpenVMS debugger on I64.
18278 An incorrect Ada program may be handled in three ways by the GNAT compiler:
18282 The illegality may be a violation of the static semantics of Ada. In
18283 that case GNAT diagnoses the constructs in the program that are illegal.
18284 It is then a straightforward matter for the user to modify those parts of
18288 The illegality may be a violation of the dynamic semantics of Ada. In
18289 that case the program compiles and executes, but may generate incorrect
18290 results, or may terminate abnormally with some exception.
18293 When presented with a program that contains convoluted errors, GNAT
18294 itself may terminate abnormally without providing full diagnostics on
18295 the incorrect user program.
18299 * The GNAT Debugger GDB::
18301 * Introduction to GDB Commands::
18302 * Using Ada Expressions::
18303 * Calling User-Defined Subprograms::
18304 * Using the Next Command in a Function::
18307 * Debugging Generic Units::
18308 * Remote Debugging using gdbserver::
18309 * GNAT Abnormal Termination or Failure to Terminate::
18310 * Naming Conventions for GNAT Source Files::
18311 * Getting Internal Debugging Information::
18312 * Stack Traceback::
18318 @node The GNAT Debugger GDB
18319 @section The GNAT Debugger GDB
18322 @code{GDB} is a general purpose, platform-independent debugger that
18323 can be used to debug mixed-language programs compiled with @command{gcc},
18324 and in particular is capable of debugging Ada programs compiled with
18325 GNAT. The latest versions of @code{GDB} are Ada-aware and can handle
18326 complex Ada data structures.
18328 @xref{Top,, Debugging with GDB, gdb, Debugging with GDB},
18330 located in the GNU:[DOCS] directory,
18332 for full details on the usage of @code{GDB}, including a section on
18333 its usage on programs. This manual should be consulted for full
18334 details. The section that follows is a brief introduction to the
18335 philosophy and use of @code{GDB}.
18337 When GNAT programs are compiled, the compiler optionally writes debugging
18338 information into the generated object file, including information on
18339 line numbers, and on declared types and variables. This information is
18340 separate from the generated code. It makes the object files considerably
18341 larger, but it does not add to the size of the actual executable that
18342 will be loaded into memory, and has no impact on run-time performance. The
18343 generation of debug information is triggered by the use of the
18344 ^-g^/DEBUG^ switch in the @command{gcc} or @command{gnatmake} command
18345 used to carry out the compilations. It is important to emphasize that
18346 the use of these options does not change the generated code.
18348 The debugging information is written in standard system formats that
18349 are used by many tools, including debuggers and profilers. The format
18350 of the information is typically designed to describe C types and
18351 semantics, but GNAT implements a translation scheme which allows full
18352 details about Ada types and variables to be encoded into these
18353 standard C formats. Details of this encoding scheme may be found in
18354 the file exp_dbug.ads in the GNAT source distribution. However, the
18355 details of this encoding are, in general, of no interest to a user,
18356 since @code{GDB} automatically performs the necessary decoding.
18358 When a program is bound and linked, the debugging information is
18359 collected from the object files, and stored in the executable image of
18360 the program. Again, this process significantly increases the size of
18361 the generated executable file, but it does not increase the size of
18362 the executable program itself. Furthermore, if this program is run in
18363 the normal manner, it runs exactly as if the debug information were
18364 not present, and takes no more actual memory.
18366 However, if the program is run under control of @code{GDB}, the
18367 debugger is activated. The image of the program is loaded, at which
18368 point it is ready to run. If a run command is given, then the program
18369 will run exactly as it would have if @code{GDB} were not present. This
18370 is a crucial part of the @code{GDB} design philosophy. @code{GDB} is
18371 entirely non-intrusive until a breakpoint is encountered. If no
18372 breakpoint is ever hit, the program will run exactly as it would if no
18373 debugger were present. When a breakpoint is hit, @code{GDB} accesses
18374 the debugging information and can respond to user commands to inspect
18375 variables, and more generally to report on the state of execution.
18379 @section Running GDB
18382 This section describes how to initiate the debugger.
18383 @c The above sentence is really just filler, but it was otherwise
18384 @c clumsy to get the first paragraph nonindented given the conditional
18385 @c nature of the description
18388 The debugger can be launched from a @code{GPS} menu or
18389 directly from the command line. The description below covers the latter use.
18390 All the commands shown can be used in the @code{GPS} debug console window,
18391 but there are usually more GUI-based ways to achieve the same effect.
18394 The command to run @code{GDB} is
18397 $ ^gdb program^GDB PROGRAM^
18401 where @code{^program^PROGRAM^} is the name of the executable file. This
18402 activates the debugger and results in a prompt for debugger commands.
18403 The simplest command is simply @code{run}, which causes the program to run
18404 exactly as if the debugger were not present. The following section
18405 describes some of the additional commands that can be given to @code{GDB}.
18407 @c *******************************
18408 @node Introduction to GDB Commands
18409 @section Introduction to GDB Commands
18412 @code{GDB} contains a large repertoire of commands. @xref{Top,,
18413 Debugging with GDB, gdb, Debugging with GDB},
18415 located in the GNU:[DOCS] directory,
18417 for extensive documentation on the use
18418 of these commands, together with examples of their use. Furthermore,
18419 the command @command{help} invoked from within GDB activates a simple help
18420 facility which summarizes the available commands and their options.
18421 In this section we summarize a few of the most commonly
18422 used commands to give an idea of what @code{GDB} is about. You should create
18423 a simple program with debugging information and experiment with the use of
18424 these @code{GDB} commands on the program as you read through the
18428 @item set args @var{arguments}
18429 The @var{arguments} list above is a list of arguments to be passed to
18430 the program on a subsequent run command, just as though the arguments
18431 had been entered on a normal invocation of the program. The @code{set args}
18432 command is not needed if the program does not require arguments.
18435 The @code{run} command causes execution of the program to start from
18436 the beginning. If the program is already running, that is to say if
18437 you are currently positioned at a breakpoint, then a prompt will ask
18438 for confirmation that you want to abandon the current execution and
18441 @item breakpoint @var{location}
18442 The breakpoint command sets a breakpoint, that is to say a point at which
18443 execution will halt and @code{GDB} will await further
18444 commands. @var{location} is
18445 either a line number within a file, given in the format @code{file:linenumber},
18446 or it is the name of a subprogram. If you request that a breakpoint be set on
18447 a subprogram that is overloaded, a prompt will ask you to specify on which of
18448 those subprograms you want to breakpoint. You can also
18449 specify that all of them should be breakpointed. If the program is run
18450 and execution encounters the breakpoint, then the program
18451 stops and @code{GDB} signals that the breakpoint was encountered by
18452 printing the line of code before which the program is halted.
18454 @item catch exception @var{name}
18455 This command causes the program execution to stop whenever exception
18456 @var{name} is raised. If @var{name} is omitted, then the execution is
18457 suspended when any exception is raised.
18459 @item print @var{expression}
18460 This will print the value of the given expression. Most simple
18461 Ada expression formats are properly handled by @code{GDB}, so the expression
18462 can contain function calls, variables, operators, and attribute references.
18465 Continues execution following a breakpoint, until the next breakpoint or the
18466 termination of the program.
18469 Executes a single line after a breakpoint. If the next statement
18470 is a subprogram call, execution continues into (the first statement of)
18471 the called subprogram.
18474 Executes a single line. If this line is a subprogram call, executes and
18475 returns from the call.
18478 Lists a few lines around the current source location. In practice, it
18479 is usually more convenient to have a separate edit window open with the
18480 relevant source file displayed. Successive applications of this command
18481 print subsequent lines. The command can be given an argument which is a
18482 line number, in which case it displays a few lines around the specified one.
18485 Displays a backtrace of the call chain. This command is typically
18486 used after a breakpoint has occurred, to examine the sequence of calls that
18487 leads to the current breakpoint. The display includes one line for each
18488 activation record (frame) corresponding to an active subprogram.
18491 At a breakpoint, @code{GDB} can display the values of variables local
18492 to the current frame. The command @code{up} can be used to
18493 examine the contents of other active frames, by moving the focus up
18494 the stack, that is to say from callee to caller, one frame at a time.
18497 Moves the focus of @code{GDB} down from the frame currently being
18498 examined to the frame of its callee (the reverse of the previous command),
18500 @item frame @var{n}
18501 Inspect the frame with the given number. The value 0 denotes the frame
18502 of the current breakpoint, that is to say the top of the call stack.
18507 The above list is a very short introduction to the commands that
18508 @code{GDB} provides. Important additional capabilities, including conditional
18509 breakpoints, the ability to execute command sequences on a breakpoint,
18510 the ability to debug at the machine instruction level and many other
18511 features are described in detail in @ref{Top,, Debugging with GDB, gdb,
18512 Debugging with GDB}. Note that most commands can be abbreviated
18513 (for example, c for continue, bt for backtrace).
18515 @node Using Ada Expressions
18516 @section Using Ada Expressions
18517 @cindex Ada expressions
18520 @code{GDB} supports a fairly large subset of Ada expression syntax, with some
18521 extensions. The philosophy behind the design of this subset is
18525 That @code{GDB} should provide basic literals and access to operations for
18526 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
18527 leaving more sophisticated computations to subprograms written into the
18528 program (which therefore may be called from @code{GDB}).
18531 That type safety and strict adherence to Ada language restrictions
18532 are not particularly important to the @code{GDB} user.
18535 That brevity is important to the @code{GDB} user.
18539 Thus, for brevity, the debugger acts as if there were
18540 implicit @code{with} and @code{use} clauses in effect for all user-written
18541 packages, thus making it unnecessary to fully qualify most names with
18542 their packages, regardless of context. Where this causes ambiguity,
18543 @code{GDB} asks the user's intent.
18545 For details on the supported Ada syntax, see @ref{Top,, Debugging with
18546 GDB, gdb, Debugging with GDB}.
18548 @node Calling User-Defined Subprograms
18549 @section Calling User-Defined Subprograms
18552 An important capability of @code{GDB} is the ability to call user-defined
18553 subprograms while debugging. This is achieved simply by entering
18554 a subprogram call statement in the form:
18557 call subprogram-name (parameters)
18561 The keyword @code{call} can be omitted in the normal case where the
18562 @code{subprogram-name} does not coincide with any of the predefined
18563 @code{GDB} commands.
18565 The effect is to invoke the given subprogram, passing it the
18566 list of parameters that is supplied. The parameters can be expressions and
18567 can include variables from the program being debugged. The
18568 subprogram must be defined
18569 at the library level within your program, and @code{GDB} will call the
18570 subprogram within the environment of your program execution (which
18571 means that the subprogram is free to access or even modify variables
18572 within your program).
18574 The most important use of this facility is in allowing the inclusion of
18575 debugging routines that are tailored to particular data structures
18576 in your program. Such debugging routines can be written to provide a suitably
18577 high-level description of an abstract type, rather than a low-level dump
18578 of its physical layout. After all, the standard
18579 @code{GDB print} command only knows the physical layout of your
18580 types, not their abstract meaning. Debugging routines can provide information
18581 at the desired semantic level and are thus enormously useful.
18583 For example, when debugging GNAT itself, it is crucial to have access to
18584 the contents of the tree nodes used to represent the program internally.
18585 But tree nodes are represented simply by an integer value (which in turn
18586 is an index into a table of nodes).
18587 Using the @code{print} command on a tree node would simply print this integer
18588 value, which is not very useful. But the PN routine (defined in file
18589 treepr.adb in the GNAT sources) takes a tree node as input, and displays
18590 a useful high level representation of the tree node, which includes the
18591 syntactic category of the node, its position in the source, the integers
18592 that denote descendant nodes and parent node, as well as varied
18593 semantic information. To study this example in more detail, you might want to
18594 look at the body of the PN procedure in the stated file.
18596 @node Using the Next Command in a Function
18597 @section Using the Next Command in a Function
18600 When you use the @code{next} command in a function, the current source
18601 location will advance to the next statement as usual. A special case
18602 arises in the case of a @code{return} statement.
18604 Part of the code for a return statement is the ``epilog'' of the function.
18605 This is the code that returns to the caller. There is only one copy of
18606 this epilog code, and it is typically associated with the last return
18607 statement in the function if there is more than one return. In some
18608 implementations, this epilog is associated with the first statement
18611 The result is that if you use the @code{next} command from a return
18612 statement that is not the last return statement of the function you
18613 may see a strange apparent jump to the last return statement or to
18614 the start of the function. You should simply ignore this odd jump.
18615 The value returned is always that from the first return statement
18616 that was stepped through.
18618 @node Ada Exceptions
18619 @section Stopping when Ada Exceptions are Raised
18623 You can set catchpoints that stop the program execution when your program
18624 raises selected exceptions.
18627 @item catch exception
18628 Set a catchpoint that stops execution whenever (any task in the) program
18629 raises any exception.
18631 @item catch exception @var{name}
18632 Set a catchpoint that stops execution whenever (any task in the) program
18633 raises the exception @var{name}.
18635 @item catch exception unhandled
18636 Set a catchpoint that stops executino whenever (any task in the) program
18637 raises an exception for which there is no handler.
18639 @item info exceptions
18640 @itemx info exceptions @var{regexp}
18641 The @code{info exceptions} command permits the user to examine all defined
18642 exceptions within Ada programs. With a regular expression, @var{regexp}, as
18643 argument, prints out only those exceptions whose name matches @var{regexp}.
18651 @code{GDB} allows the following task-related commands:
18655 This command shows a list of current Ada tasks, as in the following example:
18662 ID TID P-ID Thread Pri State Name
18663 1 8088000 0 807e000 15 Child Activation Wait main_task
18664 2 80a4000 1 80ae000 15 Accept/Select Wait b
18665 3 809a800 1 80a4800 15 Child Activation Wait a
18666 * 4 80ae800 3 80b8000 15 Running c
18670 In this listing, the asterisk before the first task indicates it to be the
18671 currently running task. The first column lists the task ID that is used
18672 to refer to tasks in the following commands.
18674 @item break @var{linespec} task @var{taskid}
18675 @itemx break @var{linespec} task @var{taskid} if @dots{}
18676 @cindex Breakpoints and tasks
18677 These commands are like the @code{break @dots{} thread @dots{}}.
18678 @var{linespec} specifies source lines.
18680 Use the qualifier @samp{task @var{taskid}} with a breakpoint command
18681 to specify that you only want @code{GDB} to stop the program when a
18682 particular Ada task reaches this breakpoint. @var{taskid} is one of the
18683 numeric task identifiers assigned by @code{GDB}, shown in the first
18684 column of the @samp{info tasks} display.
18686 If you do not specify @samp{task @var{taskid}} when you set a
18687 breakpoint, the breakpoint applies to @emph{all} tasks of your
18690 You can use the @code{task} qualifier on conditional breakpoints as
18691 well; in this case, place @samp{task @var{taskid}} before the
18692 breakpoint condition (before the @code{if}).
18694 @item task @var{taskno}
18695 @cindex Task switching
18697 This command allows to switch to the task referred by @var{taskno}. In
18698 particular, This allows to browse the backtrace of the specified
18699 task. It is advised to switch back to the original task before
18700 continuing execution otherwise the scheduling of the program may be
18705 For more detailed information on the tasking support,
18706 see @ref{Top,, Debugging with GDB, gdb, Debugging with GDB}.
18708 @node Debugging Generic Units
18709 @section Debugging Generic Units
18710 @cindex Debugging Generic Units
18714 GNAT always uses code expansion for generic instantiation. This means that
18715 each time an instantiation occurs, a complete copy of the original code is
18716 made, with appropriate substitutions of formals by actuals.
18718 It is not possible to refer to the original generic entities in
18719 @code{GDB}, but it is always possible to debug a particular instance of
18720 a generic, by using the appropriate expanded names. For example, if we have
18722 @smallexample @c ada
18727 generic package k is
18728 procedure kp (v1 : in out integer);
18732 procedure kp (v1 : in out integer) is
18738 package k1 is new k;
18739 package k2 is new k;
18741 var : integer := 1;
18754 Then to break on a call to procedure kp in the k2 instance, simply
18758 (gdb) break g.k2.kp
18762 When the breakpoint occurs, you can step through the code of the
18763 instance in the normal manner and examine the values of local variables, as for
18766 @node Remote Debugging using gdbserver
18767 @section Remote Debugging using gdbserver
18768 @cindex Remote Debugging using gdbserver
18771 On platforms where gdbserver is supported, it is possible to use this tool
18772 to debug your application remotely. This can be useful in situations
18773 where the program needs to be run on a target host that is different
18774 from the host used for development, particularly when the target has
18775 a limited amount of resources (either CPU and/or memory).
18777 To do so, start your program using gdbserver on the target machine.
18778 gdbserver then automatically suspends the execution of your program
18779 at its entry point, waiting for a debugger to connect to it. The
18780 following commands starts an application and tells gdbserver to
18781 wait for a connection with the debugger on localhost port 4444.
18784 $ gdbserver localhost:4444 program
18785 Process program created; pid = 5685
18786 Listening on port 4444
18789 Once gdbserver has started listening, we can tell the debugger to establish
18790 a connection with this gdbserver, and then start the same debugging session
18791 as if the program was being debugged on the same host, directly under
18792 the control of GDB.
18796 (gdb) target remote targethost:4444
18797 Remote debugging using targethost:4444
18798 0x00007f29936d0af0 in ?? () from /lib64/ld-linux-x86-64.so.
18800 Breakpoint 1 at 0x401f0c: file foo.adb, line 3.
18804 Breakpoint 1, foo () at foo.adb:4
18808 It is also possible to use gdbserver to attach to an already running
18809 program, in which case the execution of that program is simply suspended
18810 until the connection between the debugger and gdbserver is established.
18812 For more information on how to use gdbserver, @ref{Top, Server, Using
18813 the gdbserver Program, gdb, Debugging with GDB}. GNAT Pro provides support
18814 for gdbserver on x86-linux, x86-windows and x86_64-linux.
18816 @node GNAT Abnormal Termination or Failure to Terminate
18817 @section GNAT Abnormal Termination or Failure to Terminate
18818 @cindex GNAT Abnormal Termination or Failure to Terminate
18821 When presented with programs that contain serious errors in syntax
18823 GNAT may on rare occasions experience problems in operation, such
18825 segmentation fault or illegal memory access, raising an internal
18826 exception, terminating abnormally, or failing to terminate at all.
18827 In such cases, you can activate
18828 various features of GNAT that can help you pinpoint the construct in your
18829 program that is the likely source of the problem.
18831 The following strategies are presented in increasing order of
18832 difficulty, corresponding to your experience in using GNAT and your
18833 familiarity with compiler internals.
18837 Run @command{gcc} with the @option{-gnatf}. This first
18838 switch causes all errors on a given line to be reported. In its absence,
18839 only the first error on a line is displayed.
18841 The @option{-gnatdO} switch causes errors to be displayed as soon as they
18842 are encountered, rather than after compilation is terminated. If GNAT
18843 terminates prematurely or goes into an infinite loop, the last error
18844 message displayed may help to pinpoint the culprit.
18847 Run @command{gcc} with the @option{^-v (verbose)^/VERBOSE^} switch. In this
18848 mode, @command{gcc} produces ongoing information about the progress of the
18849 compilation and provides the name of each procedure as code is
18850 generated. This switch allows you to find which Ada procedure was being
18851 compiled when it encountered a code generation problem.
18854 @cindex @option{-gnatdc} switch
18855 Run @command{gcc} with the @option{-gnatdc} switch. This is a GNAT specific
18856 switch that does for the front-end what @option{^-v^VERBOSE^} does
18857 for the back end. The system prints the name of each unit,
18858 either a compilation unit or nested unit, as it is being analyzed.
18860 Finally, you can start
18861 @code{gdb} directly on the @code{gnat1} executable. @code{gnat1} is the
18862 front-end of GNAT, and can be run independently (normally it is just
18863 called from @command{gcc}). You can use @code{gdb} on @code{gnat1} as you
18864 would on a C program (but @pxref{The GNAT Debugger GDB} for caveats). The
18865 @code{where} command is the first line of attack; the variable
18866 @code{lineno} (seen by @code{print lineno}), used by the second phase of
18867 @code{gnat1} and by the @command{gcc} backend, indicates the source line at
18868 which the execution stopped, and @code{input_file name} indicates the name of
18872 @node Naming Conventions for GNAT Source Files
18873 @section Naming Conventions for GNAT Source Files
18876 In order to examine the workings of the GNAT system, the following
18877 brief description of its organization may be helpful:
18881 Files with prefix @file{^sc^SC^} contain the lexical scanner.
18884 All files prefixed with @file{^par^PAR^} are components of the parser. The
18885 numbers correspond to chapters of the Ada Reference Manual. For example,
18886 parsing of select statements can be found in @file{par-ch9.adb}.
18889 All files prefixed with @file{^sem^SEM^} perform semantic analysis. The
18890 numbers correspond to chapters of the Ada standard. For example, all
18891 issues involving context clauses can be found in @file{sem_ch10.adb}. In
18892 addition, some features of the language require sufficient special processing
18893 to justify their own semantic files: sem_aggr for aggregates, sem_disp for
18894 dynamic dispatching, etc.
18897 All files prefixed with @file{^exp^EXP^} perform normalization and
18898 expansion of the intermediate representation (abstract syntax tree, or AST).
18899 these files use the same numbering scheme as the parser and semantics files.
18900 For example, the construction of record initialization procedures is done in
18901 @file{exp_ch3.adb}.
18904 The files prefixed with @file{^bind^BIND^} implement the binder, which
18905 verifies the consistency of the compilation, determines an order of
18906 elaboration, and generates the bind file.
18909 The files @file{atree.ads} and @file{atree.adb} detail the low-level
18910 data structures used by the front-end.
18913 The files @file{sinfo.ads} and @file{sinfo.adb} detail the structure of
18914 the abstract syntax tree as produced by the parser.
18917 The files @file{einfo.ads} and @file{einfo.adb} detail the attributes of
18918 all entities, computed during semantic analysis.
18921 Library management issues are dealt with in files with prefix
18927 Ada files with the prefix @file{^a-^A-^} are children of @code{Ada}, as
18928 defined in Annex A.
18933 Files with prefix @file{^i-^I-^} are children of @code{Interfaces}, as
18934 defined in Annex B.
18938 Files with prefix @file{^s-^S-^} are children of @code{System}. This includes
18939 both language-defined children and GNAT run-time routines.
18943 Files with prefix @file{^g-^G-^} are children of @code{GNAT}. These are useful
18944 general-purpose packages, fully documented in their specs. All
18945 the other @file{.c} files are modifications of common @command{gcc} files.
18948 @node Getting Internal Debugging Information
18949 @section Getting Internal Debugging Information
18952 Most compilers have internal debugging switches and modes. GNAT
18953 does also, except GNAT internal debugging switches and modes are not
18954 secret. A summary and full description of all the compiler and binder
18955 debug flags are in the file @file{debug.adb}. You must obtain the
18956 sources of the compiler to see the full detailed effects of these flags.
18958 The switches that print the source of the program (reconstructed from
18959 the internal tree) are of general interest for user programs, as are the
18961 the full internal tree, and the entity table (the symbol table
18962 information). The reconstructed source provides a readable version of the
18963 program after the front-end has completed analysis and expansion,
18964 and is useful when studying the performance of specific constructs.
18965 For example, constraint checks are indicated, complex aggregates
18966 are replaced with loops and assignments, and tasking primitives
18967 are replaced with run-time calls.
18969 @node Stack Traceback
18970 @section Stack Traceback
18972 @cindex stack traceback
18973 @cindex stack unwinding
18976 Traceback is a mechanism to display the sequence of subprogram calls that
18977 leads to a specified execution point in a program. Often (but not always)
18978 the execution point is an instruction at which an exception has been raised.
18979 This mechanism is also known as @i{stack unwinding} because it obtains
18980 its information by scanning the run-time stack and recovering the activation
18981 records of all active subprograms. Stack unwinding is one of the most
18982 important tools for program debugging.
18984 The first entry stored in traceback corresponds to the deepest calling level,
18985 that is to say the subprogram currently executing the instruction
18986 from which we want to obtain the traceback.
18988 Note that there is no runtime performance penalty when stack traceback
18989 is enabled, and no exception is raised during program execution.
18992 * Non-Symbolic Traceback::
18993 * Symbolic Traceback::
18996 @node Non-Symbolic Traceback
18997 @subsection Non-Symbolic Traceback
18998 @cindex traceback, non-symbolic
19001 Note: this feature is not supported on all platforms. See
19002 @file{GNAT.Traceback spec in g-traceb.ads} for a complete list of supported
19006 * Tracebacks From an Unhandled Exception::
19007 * Tracebacks From Exception Occurrences (non-symbolic)::
19008 * Tracebacks From Anywhere in a Program (non-symbolic)::
19011 @node Tracebacks From an Unhandled Exception
19012 @subsubsection Tracebacks From an Unhandled Exception
19015 A runtime non-symbolic traceback is a list of addresses of call instructions.
19016 To enable this feature you must use the @option{-E}
19017 @code{gnatbind}'s option. With this option a stack traceback is stored as part
19018 of exception information. You can retrieve this information using the
19019 @code{addr2line} tool.
19021 Here is a simple example:
19023 @smallexample @c ada
19029 raise Constraint_Error;
19044 $ gnatmake stb -bargs -E
19047 Execution terminated by unhandled exception
19048 Exception name: CONSTRAINT_ERROR
19050 Call stack traceback locations:
19051 0x401373 0x40138b 0x40139c 0x401335 0x4011c4 0x4011f1 0x77e892a4
19055 As we see the traceback lists a sequence of addresses for the unhandled
19056 exception @code{CONSTRAINT_ERROR} raised in procedure P1. It is easy to
19057 guess that this exception come from procedure P1. To translate these
19058 addresses into the source lines where the calls appear, the
19059 @code{addr2line} tool, described below, is invaluable. The use of this tool
19060 requires the program to be compiled with debug information.
19063 $ gnatmake -g stb -bargs -E
19066 Execution terminated by unhandled exception
19067 Exception name: CONSTRAINT_ERROR
19069 Call stack traceback locations:
19070 0x401373 0x40138b 0x40139c 0x401335 0x4011c4 0x4011f1 0x77e892a4
19072 $ addr2line --exe=stb 0x401373 0x40138b 0x40139c 0x401335 0x4011c4
19073 0x4011f1 0x77e892a4
19075 00401373 at d:/stb/stb.adb:5
19076 0040138B at d:/stb/stb.adb:10
19077 0040139C at d:/stb/stb.adb:14
19078 00401335 at d:/stb/b~stb.adb:104
19079 004011C4 at /build/@dots{}/crt1.c:200
19080 004011F1 at /build/@dots{}/crt1.c:222
19081 77E892A4 in ?? at ??:0
19085 The @code{addr2line} tool has several other useful options:
19089 to get the function name corresponding to any location
19091 @item --demangle=gnat
19092 to use the gnat decoding mode for the function names. Note that
19093 for binutils version 2.9.x the option is simply @option{--demangle}.
19097 $ addr2line --exe=stb --functions --demangle=gnat 0x401373 0x40138b
19098 0x40139c 0x401335 0x4011c4 0x4011f1
19100 00401373 in stb.p1 at d:/stb/stb.adb:5
19101 0040138B in stb.p2 at d:/stb/stb.adb:10
19102 0040139C in stb at d:/stb/stb.adb:14
19103 00401335 in main at d:/stb/b~stb.adb:104
19104 004011C4 in <__mingw_CRTStartup> at /build/@dots{}/crt1.c:200
19105 004011F1 in <mainCRTStartup> at /build/@dots{}/crt1.c:222
19109 From this traceback we can see that the exception was raised in
19110 @file{stb.adb} at line 5, which was reached from a procedure call in
19111 @file{stb.adb} at line 10, and so on. The @file{b~std.adb} is the binder file,
19112 which contains the call to the main program.
19113 @xref{Running gnatbind}. The remaining entries are assorted runtime routines,
19114 and the output will vary from platform to platform.
19116 It is also possible to use @code{GDB} with these traceback addresses to debug
19117 the program. For example, we can break at a given code location, as reported
19118 in the stack traceback:
19124 Furthermore, this feature is not implemented inside Windows DLL. Only
19125 the non-symbolic traceback is reported in this case.
19128 (gdb) break *0x401373
19129 Breakpoint 1 at 0x401373: file stb.adb, line 5.
19133 It is important to note that the stack traceback addresses
19134 do not change when debug information is included. This is particularly useful
19135 because it makes it possible to release software without debug information (to
19136 minimize object size), get a field report that includes a stack traceback
19137 whenever an internal bug occurs, and then be able to retrieve the sequence
19138 of calls with the same program compiled with debug information.
19140 @node Tracebacks From Exception Occurrences (non-symbolic)
19141 @subsubsection Tracebacks From Exception Occurrences
19144 Non-symbolic tracebacks are obtained by using the @option{-E} binder argument.
19145 The stack traceback is attached to the exception information string, and can
19146 be retrieved in an exception handler within the Ada program, by means of the
19147 Ada facilities defined in @code{Ada.Exceptions}. Here is a simple example:
19149 @smallexample @c ada
19151 with Ada.Exceptions;
19156 use Ada.Exceptions;
19164 Text_IO.Put_Line (Exception_Information (E));
19178 This program will output:
19183 Exception name: CONSTRAINT_ERROR
19184 Message: stb.adb:12
19185 Call stack traceback locations:
19186 0x4015e4 0x401633 0x401644 0x401461 0x4011c4 0x4011f1 0x77e892a4
19189 @node Tracebacks From Anywhere in a Program (non-symbolic)
19190 @subsubsection Tracebacks From Anywhere in a Program
19193 It is also possible to retrieve a stack traceback from anywhere in a
19194 program. For this you need to
19195 use the @code{GNAT.Traceback} API. This package includes a procedure called
19196 @code{Call_Chain} that computes a complete stack traceback, as well as useful
19197 display procedures described below. It is not necessary to use the
19198 @option{-E gnatbind} option in this case, because the stack traceback mechanism
19199 is invoked explicitly.
19202 In the following example we compute a traceback at a specific location in
19203 the program, and we display it using @code{GNAT.Debug_Utilities.Image} to
19204 convert addresses to strings:
19206 @smallexample @c ada
19208 with GNAT.Traceback;
19209 with GNAT.Debug_Utilities;
19215 use GNAT.Traceback;
19218 TB : Tracebacks_Array (1 .. 10);
19219 -- We are asking for a maximum of 10 stack frames.
19221 -- Len will receive the actual number of stack frames returned.
19223 Call_Chain (TB, Len);
19225 Text_IO.Put ("In STB.P1 : ");
19227 for K in 1 .. Len loop
19228 Text_IO.Put (Debug_Utilities.Image (TB (K)));
19249 In STB.P1 : 16#0040_F1E4# 16#0040_14F2# 16#0040_170B# 16#0040_171C#
19250 16#0040_1461# 16#0040_11C4# 16#0040_11F1# 16#77E8_92A4#
19254 You can then get further information by invoking the @code{addr2line}
19255 tool as described earlier (note that the hexadecimal addresses
19256 need to be specified in C format, with a leading ``0x'').
19258 @node Symbolic Traceback
19259 @subsection Symbolic Traceback
19260 @cindex traceback, symbolic
19263 A symbolic traceback is a stack traceback in which procedure names are
19264 associated with each code location.
19267 Note that this feature is not supported on all platforms. See
19268 @file{GNAT.Traceback.Symbolic spec in g-trasym.ads} for a complete
19269 list of currently supported platforms.
19272 Note that the symbolic traceback requires that the program be compiled
19273 with debug information. If it is not compiled with debug information
19274 only the non-symbolic information will be valid.
19277 * Tracebacks From Exception Occurrences (symbolic)::
19278 * Tracebacks From Anywhere in a Program (symbolic)::
19281 @node Tracebacks From Exception Occurrences (symbolic)
19282 @subsubsection Tracebacks From Exception Occurrences
19284 @smallexample @c ada
19286 with GNAT.Traceback.Symbolic;
19292 raise Constraint_Error;
19309 Ada.Text_IO.Put_Line (GNAT.Traceback.Symbolic.Symbolic_Traceback (E));
19314 $ gnatmake -g .\stb -bargs -E -largs -lgnat -laddr2line -lintl
19317 0040149F in stb.p1 at stb.adb:8
19318 004014B7 in stb.p2 at stb.adb:13
19319 004014CF in stb.p3 at stb.adb:18
19320 004015DD in ada.stb at stb.adb:22
19321 00401461 in main at b~stb.adb:168
19322 004011C4 in __mingw_CRTStartup at crt1.c:200
19323 004011F1 in mainCRTStartup at crt1.c:222
19324 77E892A4 in ?? at ??:0
19328 In the above example the ``.\'' syntax in the @command{gnatmake} command
19329 is currently required by @command{addr2line} for files that are in
19330 the current working directory.
19331 Moreover, the exact sequence of linker options may vary from platform
19333 The above @option{-largs} section is for Windows platforms. By contrast,
19334 under Unix there is no need for the @option{-largs} section.
19335 Differences across platforms are due to details of linker implementation.
19337 @node Tracebacks From Anywhere in a Program (symbolic)
19338 @subsubsection Tracebacks From Anywhere in a Program
19341 It is possible to get a symbolic stack traceback
19342 from anywhere in a program, just as for non-symbolic tracebacks.
19343 The first step is to obtain a non-symbolic
19344 traceback, and then call @code{Symbolic_Traceback} to compute the symbolic
19345 information. Here is an example:
19347 @smallexample @c ada
19349 with GNAT.Traceback;
19350 with GNAT.Traceback.Symbolic;
19355 use GNAT.Traceback;
19356 use GNAT.Traceback.Symbolic;
19359 TB : Tracebacks_Array (1 .. 10);
19360 -- We are asking for a maximum of 10 stack frames.
19362 -- Len will receive the actual number of stack frames returned.
19364 Call_Chain (TB, Len);
19365 Text_IO.Put_Line (Symbolic_Traceback (TB (1 .. Len)));
19378 @c ******************************
19380 @node Compatibility with HP Ada
19381 @chapter Compatibility with HP Ada
19382 @cindex Compatibility
19387 @cindex Compatibility between GNAT and HP Ada
19388 This chapter compares HP Ada (formerly known as ``DEC Ada'')
19389 for OpenVMS Alpha and GNAT for OpenVMS for Alpha and for I64.
19390 GNAT is highly compatible
19391 with HP Ada, and it should generally be straightforward to port code
19392 from the HP Ada environment to GNAT. However, there are a few language
19393 and implementation differences of which the user must be aware. These
19394 differences are discussed in this chapter. In
19395 addition, the operating environment and command structure for the
19396 compiler are different, and these differences are also discussed.
19398 For further details on these and other compatibility issues,
19399 see Appendix E of the HP publication
19400 @cite{HP Ada, Technical Overview and Comparison on HP Platforms}.
19402 Except where otherwise indicated, the description of GNAT for OpenVMS
19403 applies to both the Alpha and I64 platforms.
19405 For information on porting Ada code from GNAT on Alpha OpenVMS to GNAT on
19406 I64 OpenVMS, see @ref{Transitioning to 64-Bit GNAT for OpenVMS}.
19408 The discussion in this chapter addresses specifically the implementation
19409 of Ada 83 for HP OpenVMS Alpha Systems. In cases where the implementation
19410 of HP Ada differs between OpenVMS Alpha Systems and OpenVMS VAX Systems,
19411 GNAT always follows the Alpha implementation.
19413 For GNAT running on other than VMS systems, all the HP Ada 83 pragmas and
19414 attributes are recognized, although only a subset of them can sensibly
19415 be implemented. The description of pragmas in
19416 @xref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference Manual},
19417 indicates whether or not they are applicable to non-VMS systems.
19420 * Ada Language Compatibility::
19421 * Differences in the Definition of Package System::
19422 * Language-Related Features::
19423 * The Package STANDARD::
19424 * The Package SYSTEM::
19425 * Tasking and Task-Related Features::
19426 * Pragmas and Pragma-Related Features::
19427 * Library of Predefined Units::
19429 * Main Program Definition::
19430 * Implementation-Defined Attributes::
19431 * Compiler and Run-Time Interfacing::
19432 * Program Compilation and Library Management::
19434 * Implementation Limits::
19435 * Tools and Utilities::
19438 @node Ada Language Compatibility
19439 @section Ada Language Compatibility
19442 GNAT handles Ada 95 and Ada 2005 as well as Ada 83, whereas HP Ada is only
19443 for Ada 83. Ada 95 and Ada 2005 are almost completely upwards compatible
19444 with Ada 83, and therefore Ada 83 programs will compile
19445 and run under GNAT with
19446 no changes or only minor changes. The @cite{Annotated Ada Reference Manual}
19447 provides details on specific incompatibilities.
19449 GNAT provides the switch @option{/83} on the @command{GNAT COMPILE} command,
19450 as well as the pragma @code{ADA_83}, to force the compiler to
19451 operate in Ada 83 mode. This mode does not guarantee complete
19452 conformance to Ada 83, but in practice is sufficient to
19453 eliminate most sources of incompatibilities.
19454 In particular, it eliminates the recognition of the
19455 additional Ada 95 and Ada 2005 keywords, so that their use as identifiers
19456 in Ada 83 programs is legal, and handles the cases of packages
19457 with optional bodies, and generics that instantiate unconstrained
19458 types without the use of @code{(<>)}.
19460 @node Differences in the Definition of Package System
19461 @section Differences in the Definition of Package @code{System}
19464 An Ada compiler is allowed to add
19465 implementation-dependent declarations to package @code{System}.
19467 GNAT does not take advantage of this permission, and the version of
19468 @code{System} provided by GNAT exactly matches that defined in the Ada
19471 However, HP Ada adds an extensive set of declarations to package
19473 as fully documented in the HP Ada manuals. To minimize changes required
19474 for programs that make use of these extensions, GNAT provides the pragma
19475 @code{Extend_System} for extending the definition of package System. By using:
19476 @cindex pragma @code{Extend_System}
19477 @cindex @code{Extend_System} pragma
19479 @smallexample @c ada
19482 pragma Extend_System (Aux_DEC);
19488 the set of definitions in @code{System} is extended to include those in
19489 package @code{System.Aux_DEC}.
19490 @cindex @code{System.Aux_DEC} package
19491 @cindex @code{Aux_DEC} package (child of @code{System})
19492 These definitions are incorporated directly into package @code{System},
19493 as though they had been declared there. For a
19494 list of the declarations added, see the spec of this package,
19495 which can be found in the file @file{s-auxdec.ads} in the GNAT library.
19496 @cindex @file{s-auxdec.ads} file
19497 The pragma @code{Extend_System} is a configuration pragma, which means that
19498 it can be placed in the file @file{gnat.adc}, so that it will automatically
19499 apply to all subsequent compilations. See @ref{Configuration Pragmas},
19500 for further details.
19502 An alternative approach that avoids the use of the non-standard
19503 @code{Extend_System} pragma is to add a context clause to the unit that
19504 references these facilities:
19506 @smallexample @c ada
19508 with System.Aux_DEC;
19509 use System.Aux_DEC;
19514 The effect is not quite semantically identical to incorporating
19515 the declarations directly into package @code{System},
19516 but most programs will not notice a difference
19517 unless they use prefix notation (e.g.@: @code{System.Integer_8})
19518 to reference the entities directly in package @code{System}.
19519 For units containing such references,
19520 the prefixes must either be removed, or the pragma @code{Extend_System}
19523 @node Language-Related Features
19524 @section Language-Related Features
19527 The following sections highlight differences in types,
19528 representations of types, operations, alignment, and
19532 * Integer Types and Representations::
19533 * Floating-Point Types and Representations::
19534 * Pragmas Float_Representation and Long_Float::
19535 * Fixed-Point Types and Representations::
19536 * Record and Array Component Alignment::
19537 * Address Clauses::
19538 * Other Representation Clauses::
19541 @node Integer Types and Representations
19542 @subsection Integer Types and Representations
19545 The set of predefined integer types is identical in HP Ada and GNAT.
19546 Furthermore the representation of these integer types is also identical,
19547 including the capability of size clauses forcing biased representation.
19550 HP Ada for OpenVMS Alpha systems has defined the
19551 following additional integer types in package @code{System}:
19568 @code{LARGEST_INTEGER}
19572 In GNAT, the first four of these types may be obtained from the
19573 standard Ada package @code{Interfaces}.
19574 Alternatively, by use of the pragma @code{Extend_System}, identical
19575 declarations can be referenced directly in package @code{System}.
19576 On both GNAT and HP Ada, the maximum integer size is 64 bits.
19578 @node Floating-Point Types and Representations
19579 @subsection Floating-Point Types and Representations
19580 @cindex Floating-Point types
19583 The set of predefined floating-point types is identical in HP Ada and GNAT.
19584 Furthermore the representation of these floating-point
19585 types is also identical. One important difference is that the default
19586 representation for HP Ada is @code{VAX_Float}, but the default representation
19589 Specific types may be declared to be @code{VAX_Float} or IEEE, using the
19590 pragma @code{Float_Representation} as described in the HP Ada
19592 For example, the declarations:
19594 @smallexample @c ada
19596 type F_Float is digits 6;
19597 pragma Float_Representation (VAX_Float, F_Float);
19602 declares a type @code{F_Float} that will be represented in @code{VAX_Float}
19604 This set of declarations actually appears in @code{System.Aux_DEC},
19606 the full set of additional floating-point declarations provided in
19607 the HP Ada version of package @code{System}.
19608 This and similar declarations may be accessed in a user program
19609 by using pragma @code{Extend_System}. The use of this
19610 pragma, and the related pragma @code{Long_Float} is described in further
19611 detail in the following section.
19613 @node Pragmas Float_Representation and Long_Float
19614 @subsection Pragmas @code{Float_Representation} and @code{Long_Float}
19617 HP Ada provides the pragma @code{Float_Representation}, which
19618 acts as a program library switch to allow control over
19619 the internal representation chosen for the predefined
19620 floating-point types declared in the package @code{Standard}.
19621 The format of this pragma is as follows:
19623 @smallexample @c ada
19625 pragma Float_Representation(VAX_Float | IEEE_Float);
19630 This pragma controls the representation of floating-point
19635 @code{VAX_Float} specifies that floating-point
19636 types are represented by default with the VAX system hardware types
19637 @code{F-floating}, @code{D-floating}, @code{G-floating}.
19638 Note that the @code{H-floating}
19639 type was available only on VAX systems, and is not available
19640 in either HP Ada or GNAT.
19643 @code{IEEE_Float} specifies that floating-point
19644 types are represented by default with the IEEE single and
19645 double floating-point types.
19649 GNAT provides an identical implementation of the pragma
19650 @code{Float_Representation}, except that it functions as a
19651 configuration pragma. Note that the
19652 notion of configuration pragma corresponds closely to the
19653 HP Ada notion of a program library switch.
19655 When no pragma is used in GNAT, the default is @code{IEEE_Float},
19657 from HP Ada 83, where the default is @code{VAX_Float}. In addition, the
19658 predefined libraries in GNAT are built using @code{IEEE_Float}, so it is not
19659 advisable to change the format of numbers passed to standard library
19660 routines, and if necessary explicit type conversions may be needed.
19662 The use of @code{IEEE_Float} is recommended in GNAT since it is more
19663 efficient, and (given that it conforms to an international standard)
19664 potentially more portable.
19665 The situation in which @code{VAX_Float} may be useful is in interfacing
19666 to existing code and data that expect the use of @code{VAX_Float}.
19667 In such a situation use the predefined @code{VAX_Float}
19668 types in package @code{System}, as extended by
19669 @code{Extend_System}. For example, use @code{System.F_Float}
19670 to specify the 32-bit @code{F-Float} format.
19673 On OpenVMS systems, HP Ada provides the pragma @code{Long_Float}
19674 to allow control over the internal representation chosen
19675 for the predefined type @code{Long_Float} and for floating-point
19676 type declarations with digits specified in the range 7 .. 15.
19677 The format of this pragma is as follows:
19679 @smallexample @c ada
19681 pragma Long_Float (D_FLOAT | G_FLOAT);
19685 @node Fixed-Point Types and Representations
19686 @subsection Fixed-Point Types and Representations
19689 On HP Ada for OpenVMS Alpha systems, rounding is
19690 away from zero for both positive and negative numbers.
19691 Therefore, @code{+0.5} rounds to @code{1},
19692 and @code{-0.5} rounds to @code{-1}.
19694 On GNAT the results of operations
19695 on fixed-point types are in accordance with the Ada
19696 rules. In particular, results of operations on decimal
19697 fixed-point types are truncated.
19699 @node Record and Array Component Alignment
19700 @subsection Record and Array Component Alignment
19703 On HP Ada for OpenVMS Alpha, all non-composite components
19704 are aligned on natural boundaries. For example, 1-byte
19705 components are aligned on byte boundaries, 2-byte
19706 components on 2-byte boundaries, 4-byte components on 4-byte
19707 byte boundaries, and so on. The OpenVMS Alpha hardware
19708 runs more efficiently with naturally aligned data.
19710 On GNAT, alignment rules are compatible
19711 with HP Ada for OpenVMS Alpha.
19713 @node Address Clauses
19714 @subsection Address Clauses
19717 In HP Ada and GNAT, address clauses are supported for
19718 objects and imported subprograms.
19719 The predefined type @code{System.Address} is a private type
19720 in both compilers on Alpha OpenVMS, with the same representation
19721 (it is simply a machine pointer). Addition, subtraction, and comparison
19722 operations are available in the standard Ada package
19723 @code{System.Storage_Elements}, or in package @code{System}
19724 if it is extended to include @code{System.Aux_DEC} using a
19725 pragma @code{Extend_System} as previously described.
19727 Note that code that @code{with}'s both this extended package @code{System}
19728 and the package @code{System.Storage_Elements} should not @code{use}
19729 both packages, or ambiguities will result. In general it is better
19730 not to mix these two sets of facilities. The Ada package was
19731 designed specifically to provide the kind of features that HP Ada
19732 adds directly to package @code{System}.
19734 The type @code{System.Address} is a 64-bit integer type in GNAT for
19735 I64 OpenVMS. For more information,
19736 see @ref{Transitioning to 64-Bit GNAT for OpenVMS}.
19738 GNAT is compatible with HP Ada in its handling of address
19739 clauses, except for some limitations in
19740 the form of address clauses for composite objects with
19741 initialization. Such address clauses are easily replaced
19742 by the use of an explicitly-defined constant as described
19743 in the Ada Reference Manual (13.1(22)). For example, the sequence
19746 @smallexample @c ada
19748 X, Y : Integer := Init_Func;
19749 Q : String (X .. Y) := "abc";
19751 for Q'Address use Compute_Address;
19756 will be rejected by GNAT, since the address cannot be computed at the time
19757 that @code{Q} is declared. To achieve the intended effect, write instead:
19759 @smallexample @c ada
19762 X, Y : Integer := Init_Func;
19763 Q_Address : constant Address := Compute_Address;
19764 Q : String (X .. Y) := "abc";
19766 for Q'Address use Q_Address;
19772 which will be accepted by GNAT (and other Ada compilers), and is also
19773 compatible with Ada 83. A fuller description of the restrictions
19774 on address specifications is found in @ref{Top, GNAT Reference Manual,
19775 About This Guide, gnat_rm, GNAT Reference Manual}.
19777 @node Other Representation Clauses
19778 @subsection Other Representation Clauses
19781 GNAT implements in a compatible manner all the representation
19782 clauses supported by HP Ada. In addition, GNAT
19783 implements the representation clause forms that were introduced in Ada 95,
19784 including @code{COMPONENT_SIZE} and @code{SIZE} clauses for objects.
19786 @node The Package STANDARD
19787 @section The Package @code{STANDARD}
19790 The package @code{STANDARD}, as implemented by HP Ada, is fully
19791 described in the @cite{Ada Reference Manual} and in the
19792 @cite{HP Ada Language Reference Manual}. As implemented by GNAT, the
19793 package @code{STANDARD} is described in the @cite{Ada Reference Manual}.
19795 In addition, HP Ada supports the Latin-1 character set in
19796 the type @code{CHARACTER}. GNAT supports the Latin-1 character set
19797 in the type @code{CHARACTER} and also Unicode (ISO 10646 BMP) in
19798 the type @code{WIDE_CHARACTER}.
19800 The floating-point types supported by GNAT are those
19801 supported by HP Ada, but the defaults are different, and are controlled by
19802 pragmas. See @ref{Floating-Point Types and Representations}, for details.
19804 @node The Package SYSTEM
19805 @section The Package @code{SYSTEM}
19808 HP Ada provides a specific version of the package
19809 @code{SYSTEM} for each platform on which the language is implemented.
19810 For the complete spec of the package @code{SYSTEM}, see
19811 Appendix F of the @cite{HP Ada Language Reference Manual}.
19813 On HP Ada, the package @code{SYSTEM} includes the following conversion
19816 @item @code{TO_ADDRESS(INTEGER)}
19818 @item @code{TO_ADDRESS(UNSIGNED_LONGWORD)}
19820 @item @code{TO_ADDRESS(}@i{universal_integer}@code{)}
19822 @item @code{TO_INTEGER(ADDRESS)}
19824 @item @code{TO_UNSIGNED_LONGWORD(ADDRESS)}
19826 @item Function @code{IMPORT_VALUE return UNSIGNED_LONGWORD} and the
19827 functions @code{IMPORT_ADDRESS} and @code{IMPORT_LARGEST_VALUE}
19831 By default, GNAT supplies a version of @code{SYSTEM} that matches
19832 the definition given in the @cite{Ada Reference Manual}.
19834 is a subset of the HP system definitions, which is as
19835 close as possible to the original definitions. The only difference
19836 is that the definition of @code{SYSTEM_NAME} is different:
19838 @smallexample @c ada
19840 type Name is (SYSTEM_NAME_GNAT);
19841 System_Name : constant Name := SYSTEM_NAME_GNAT;
19846 Also, GNAT adds the Ada declarations for
19847 @code{BIT_ORDER} and @code{DEFAULT_BIT_ORDER}.
19849 However, the use of the following pragma causes GNAT
19850 to extend the definition of package @code{SYSTEM} so that it
19851 encompasses the full set of HP-specific extensions,
19852 including the functions listed above:
19854 @smallexample @c ada
19856 pragma Extend_System (Aux_DEC);
19861 The pragma @code{Extend_System} is a configuration pragma that
19862 is most conveniently placed in the @file{gnat.adc} file. @xref{Pragma
19863 Extend_System,,, gnat_rm, GNAT Reference Manual}, for further details.
19865 HP Ada does not allow the recompilation of the package
19866 @code{SYSTEM}. Instead HP Ada provides several pragmas
19867 (@code{SYSTEM_NAME}, @code{STORAGE_UNIT}, and @code{MEMORY_SIZE})
19868 to modify values in the package @code{SYSTEM}.
19869 On OpenVMS Alpha systems, the pragma
19870 @code{SYSTEM_NAME} takes the enumeration literal @code{OPENVMS_AXP} as
19871 its single argument.
19873 GNAT does permit the recompilation of package @code{SYSTEM} using
19874 the special switch @option{-gnatg}, and this switch can be used if
19875 it is necessary to modify the definitions in @code{SYSTEM}. GNAT does
19876 not permit the specification of @code{SYSTEM_NAME}, @code{STORAGE_UNIT}
19877 or @code{MEMORY_SIZE} by any other means.
19879 On GNAT systems, the pragma @code{SYSTEM_NAME} takes the
19880 enumeration literal @code{SYSTEM_NAME_GNAT}.
19882 The definitions provided by the use of
19884 @smallexample @c ada
19885 pragma Extend_System (AUX_Dec);
19889 are virtually identical to those provided by the HP Ada 83 package
19890 @code{SYSTEM}. One important difference is that the name of the
19892 function for type @code{UNSIGNED_LONGWORD} is changed to
19893 @code{TO_ADDRESS_LONG}.
19894 @xref{Address Clauses,,, gnat_rm, GNAT Reference Manual}, for a
19895 discussion of why this change was necessary.
19898 The version of @code{TO_ADDRESS} taking a @i{universal_integer} argument
19900 an extension to Ada 83 not strictly compatible with the reference manual.
19901 GNAT, in order to be exactly compatible with the standard,
19902 does not provide this capability. In HP Ada 83, the
19903 point of this definition is to deal with a call like:
19905 @smallexample @c ada
19906 TO_ADDRESS (16#12777#);
19910 Normally, according to Ada 83 semantics, one would expect this to be
19911 ambiguous, since it matches both the @code{INTEGER} and
19912 @code{UNSIGNED_LONGWORD} forms of @code{TO_ADDRESS}.
19913 However, in HP Ada 83, there is no ambiguity, since the
19914 definition using @i{universal_integer} takes precedence.
19916 In GNAT, since the version with @i{universal_integer} cannot be supplied,
19918 not possible to be 100% compatible. Since there are many programs using
19919 numeric constants for the argument to @code{TO_ADDRESS}, the decision in
19921 to change the name of the function in the @code{UNSIGNED_LONGWORD} case,
19922 so the declarations provided in the GNAT version of @code{AUX_Dec} are:
19924 @smallexample @c ada
19925 function To_Address (X : Integer) return Address;
19926 pragma Pure_Function (To_Address);
19928 function To_Address_Long (X : Unsigned_Longword) return Address;
19929 pragma Pure_Function (To_Address_Long);
19933 This means that programs using @code{TO_ADDRESS} for
19934 @code{UNSIGNED_LONGWORD} must change the name to @code{TO_ADDRESS_LONG}.
19936 @node Tasking and Task-Related Features
19937 @section Tasking and Task-Related Features
19940 This section compares the treatment of tasking in GNAT
19941 and in HP Ada for OpenVMS Alpha.
19942 The GNAT description applies to both Alpha and I64 OpenVMS.
19943 For detailed information on tasking in
19944 HP Ada, see the @cite{HP Ada Language Reference Manual} and the
19945 relevant run-time reference manual.
19948 * Implementation of Tasks in HP Ada for OpenVMS Alpha Systems::
19949 * Assigning Task IDs::
19950 * Task IDs and Delays::
19951 * Task-Related Pragmas::
19952 * Scheduling and Task Priority::
19954 * External Interrupts::
19957 @node Implementation of Tasks in HP Ada for OpenVMS Alpha Systems
19958 @subsection Implementation of Tasks in HP Ada for OpenVMS Alpha Systems
19961 On OpenVMS Alpha systems, each Ada task (except a passive
19962 task) is implemented as a single stream of execution
19963 that is created and managed by the kernel. On these
19964 systems, HP Ada tasking support is based on DECthreads,
19965 an implementation of the POSIX standard for threads.
19967 Also, on OpenVMS Alpha systems, HP Ada tasks and foreign
19968 code that calls DECthreads routines can be used together.
19969 The interaction between Ada tasks and DECthreads routines
19970 can have some benefits. For example when on OpenVMS Alpha,
19971 HP Ada can call C code that is already threaded.
19973 GNAT uses the facilities of DECthreads,
19974 and Ada tasks are mapped to threads.
19976 @node Assigning Task IDs
19977 @subsection Assigning Task IDs
19980 The HP Ada Run-Time Library always assigns @code{%TASK 1} to
19981 the environment task that executes the main program. On
19982 OpenVMS Alpha systems, @code{%TASK 0} is often used for tasks
19983 that have been created but are not yet activated.
19985 On OpenVMS Alpha systems, task IDs are assigned at
19986 activation. On GNAT systems, task IDs are also assigned at
19987 task creation but do not have the same form or values as
19988 task ID values in HP Ada. There is no null task, and the
19989 environment task does not have a specific task ID value.
19991 @node Task IDs and Delays
19992 @subsection Task IDs and Delays
19995 On OpenVMS Alpha systems, tasking delays are implemented
19996 using Timer System Services. The Task ID is used for the
19997 identification of the timer request (the @code{REQIDT} parameter).
19998 If Timers are used in the application take care not to use
19999 @code{0} for the identification, because cancelling such a timer
20000 will cancel all timers and may lead to unpredictable results.
20002 @node Task-Related Pragmas
20003 @subsection Task-Related Pragmas
20006 Ada supplies the pragma @code{TASK_STORAGE}, which allows
20007 specification of the size of the guard area for a task
20008 stack. (The guard area forms an area of memory that has no
20009 read or write access and thus helps in the detection of
20010 stack overflow.) On OpenVMS Alpha systems, if the pragma
20011 @code{TASK_STORAGE} specifies a value of zero, a minimal guard
20012 area is created. In the absence of a pragma @code{TASK_STORAGE},
20013 a default guard area is created.
20015 GNAT supplies the following task-related pragmas:
20018 @item @code{TASK_INFO}
20020 This pragma appears within a task definition and
20021 applies to the task in which it appears. The argument
20022 must be of type @code{SYSTEM.TASK_INFO.TASK_INFO_TYPE}.
20024 @item @code{TASK_STORAGE}
20026 GNAT implements pragma @code{TASK_STORAGE} in the same way as HP Ada.
20027 Both HP Ada and GNAT supply the pragmas @code{PASSIVE},
20028 @code{SUPPRESS}, and @code{VOLATILE}.
20030 @node Scheduling and Task Priority
20031 @subsection Scheduling and Task Priority
20034 HP Ada implements the Ada language requirement that
20035 when two tasks are eligible for execution and they have
20036 different priorities, the lower priority task does not
20037 execute while the higher priority task is waiting. The HP
20038 Ada Run-Time Library keeps a task running until either the
20039 task is suspended or a higher priority task becomes ready.
20041 On OpenVMS Alpha systems, the default strategy is round-
20042 robin with preemption. Tasks of equal priority take turns
20043 at the processor. A task is run for a certain period of
20044 time and then placed at the tail of the ready queue for
20045 its priority level.
20047 HP Ada provides the implementation-defined pragma @code{TIME_SLICE},
20048 which can be used to enable or disable round-robin
20049 scheduling of tasks with the same priority.
20050 See the relevant HP Ada run-time reference manual for
20051 information on using the pragmas to control HP Ada task
20054 GNAT follows the scheduling rules of Annex D (Real-Time
20055 Annex) of the @cite{Ada Reference Manual}. In general, this
20056 scheduling strategy is fully compatible with HP Ada
20057 although it provides some additional constraints (as
20058 fully documented in Annex D).
20059 GNAT implements time slicing control in a manner compatible with
20060 HP Ada 83, by means of the pragma @code{Time_Slice}, whose semantics
20061 are identical to the HP Ada 83 pragma of the same name.
20062 Note that it is not possible to mix GNAT tasking and
20063 HP Ada 83 tasking in the same program, since the two run-time
20064 libraries are not compatible.
20066 @node The Task Stack
20067 @subsection The Task Stack
20070 In HP Ada, a task stack is allocated each time a
20071 non-passive task is activated. As soon as the task is
20072 terminated, the storage for the task stack is deallocated.
20073 If you specify a size of zero (bytes) with @code{T'STORAGE_SIZE},
20074 a default stack size is used. Also, regardless of the size
20075 specified, some additional space is allocated for task
20076 management purposes. On OpenVMS Alpha systems, at least
20077 one page is allocated.
20079 GNAT handles task stacks in a similar manner. In accordance with
20080 the Ada rules, it provides the pragma @code{STORAGE_SIZE} as
20081 an alternative method for controlling the task stack size.
20082 The specification of the attribute @code{T'STORAGE_SIZE} is also
20083 supported in a manner compatible with HP Ada.
20085 @node External Interrupts
20086 @subsection External Interrupts
20089 On HP Ada, external interrupts can be associated with task entries.
20090 GNAT is compatible with HP Ada in its handling of external interrupts.
20092 @node Pragmas and Pragma-Related Features
20093 @section Pragmas and Pragma-Related Features
20096 Both HP Ada and GNAT supply all language-defined pragmas
20097 as specified by the Ada 83 standard. GNAT also supplies all
20098 language-defined pragmas introduced by Ada 95 and Ada 2005.
20099 In addition, GNAT implements the implementation-defined pragmas
20103 @item @code{AST_ENTRY}
20105 @item @code{COMMON_OBJECT}
20107 @item @code{COMPONENT_ALIGNMENT}
20109 @item @code{EXPORT_EXCEPTION}
20111 @item @code{EXPORT_FUNCTION}
20113 @item @code{EXPORT_OBJECT}
20115 @item @code{EXPORT_PROCEDURE}
20117 @item @code{EXPORT_VALUED_PROCEDURE}
20119 @item @code{FLOAT_REPRESENTATION}
20123 @item @code{IMPORT_EXCEPTION}
20125 @item @code{IMPORT_FUNCTION}
20127 @item @code{IMPORT_OBJECT}
20129 @item @code{IMPORT_PROCEDURE}
20131 @item @code{IMPORT_VALUED_PROCEDURE}
20133 @item @code{INLINE_GENERIC}
20135 @item @code{INTERFACE_NAME}
20137 @item @code{LONG_FLOAT}
20139 @item @code{MAIN_STORAGE}
20141 @item @code{PASSIVE}
20143 @item @code{PSECT_OBJECT}
20145 @item @code{SHARE_GENERIC}
20147 @item @code{SUPPRESS_ALL}
20149 @item @code{TASK_STORAGE}
20151 @item @code{TIME_SLICE}
20157 These pragmas are all fully implemented, with the exception of @code{TITLE},
20158 @code{PASSIVE}, and @code{SHARE_GENERIC}, which are
20159 recognized, but which have no
20160 effect in GNAT. The effect of @code{PASSIVE} may be obtained by the
20161 use of Ada protected objects. In GNAT, all generics are inlined.
20163 Unlike HP Ada, the GNAT ``@code{EXPORT_}@i{subprogram}'' pragmas require
20164 a separate subprogram specification which must appear before the
20167 GNAT also supplies a number of implementation-defined pragmas including the
20171 @item @code{ABORT_DEFER}
20173 @item @code{ADA_83}
20175 @item @code{ADA_95}
20177 @item @code{ADA_05}
20179 @item @code{Ada_2005}
20181 @item @code{Ada_12}
20183 @item @code{Ada_2012}
20185 @item @code{ANNOTATE}
20187 @item @code{ASSERT}
20189 @item @code{C_PASS_BY_COPY}
20191 @item @code{CPP_CLASS}
20193 @item @code{CPP_CONSTRUCTOR}
20195 @item @code{CPP_DESTRUCTOR}
20199 @item @code{EXTEND_SYSTEM}
20201 @item @code{LINKER_ALIAS}
20203 @item @code{LINKER_SECTION}
20205 @item @code{MACHINE_ATTRIBUTE}
20207 @item @code{NO_RETURN}
20209 @item @code{PURE_FUNCTION}
20211 @item @code{SOURCE_FILE_NAME}
20213 @item @code{SOURCE_REFERENCE}
20215 @item @code{TASK_INFO}
20217 @item @code{UNCHECKED_UNION}
20219 @item @code{UNIMPLEMENTED_UNIT}
20221 @item @code{UNIVERSAL_DATA}
20223 @item @code{UNSUPPRESS}
20225 @item @code{WARNINGS}
20227 @item @code{WEAK_EXTERNAL}
20231 For full details on these and other GNAT implementation-defined pragmas,
20232 see @ref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference
20236 * Restrictions on the Pragma INLINE::
20237 * Restrictions on the Pragma INTERFACE::
20238 * Restrictions on the Pragma SYSTEM_NAME::
20241 @node Restrictions on the Pragma INLINE
20242 @subsection Restrictions on Pragma @code{INLINE}
20245 HP Ada enforces the following restrictions on the pragma @code{INLINE}:
20247 @item Parameters cannot have a task type.
20249 @item Function results cannot be task types, unconstrained
20250 array types, or unconstrained types with discriminants.
20252 @item Bodies cannot declare the following:
20254 @item Subprogram body or stub (imported subprogram is allowed)
20258 @item Generic declarations
20260 @item Instantiations
20264 @item Access types (types derived from access types allowed)
20266 @item Array or record types
20268 @item Dependent tasks
20270 @item Direct recursive calls of subprogram or containing
20271 subprogram, directly or via a renaming
20277 In GNAT, the only restriction on pragma @code{INLINE} is that the
20278 body must occur before the call if both are in the same
20279 unit, and the size must be appropriately small. There are
20280 no other specific restrictions which cause subprograms to
20281 be incapable of being inlined.
20283 @node Restrictions on the Pragma INTERFACE
20284 @subsection Restrictions on Pragma @code{INTERFACE}
20287 The following restrictions on pragma @code{INTERFACE}
20288 are enforced by both HP Ada and GNAT:
20290 @item Languages accepted: Ada, Bliss, C, Fortran, Default.
20291 Default is the default on OpenVMS Alpha systems.
20293 @item Parameter passing: Language specifies default
20294 mechanisms but can be overridden with an @code{EXPORT} pragma.
20297 @item Ada: Use internal Ada rules.
20299 @item Bliss, C: Parameters must be mode @code{in}; cannot be
20300 record or task type. Result cannot be a string, an
20301 array, or a record.
20303 @item Fortran: Parameters cannot have a task type. Result cannot
20304 be a string, an array, or a record.
20309 GNAT is entirely upwards compatible with HP Ada, and in addition allows
20310 record parameters for all languages.
20312 @node Restrictions on the Pragma SYSTEM_NAME
20313 @subsection Restrictions on Pragma @code{SYSTEM_NAME}
20316 For HP Ada for OpenVMS Alpha, the enumeration literal
20317 for the type @code{NAME} is @code{OPENVMS_AXP}.
20318 In GNAT, the enumeration
20319 literal for the type @code{NAME} is @code{SYSTEM_NAME_GNAT}.
20321 @node Library of Predefined Units
20322 @section Library of Predefined Units
20325 A library of predefined units is provided as part of the
20326 HP Ada and GNAT implementations. HP Ada does not provide
20327 the package @code{MACHINE_CODE} but instead recommends importing
20330 The GNAT versions of the HP Ada Run-Time Library (@code{ADA$PREDEFINED:})
20331 units are taken from the OpenVMS Alpha version, not the OpenVMS VAX
20333 The HP Ada Predefined Library units are modified to remove post-Ada 83
20334 incompatibilities and to make them interoperable with GNAT
20335 (@pxref{Changes to DECLIB}, for details).
20336 The units are located in the @file{DECLIB} directory.
20338 The GNAT RTL is contained in
20339 the @file{ADALIB} directory, and
20340 the default search path is set up to find @code{DECLIB} units in preference
20341 to @code{ADALIB} units with the same name (@code{TEXT_IO},
20342 @code{SEQUENTIAL_IO}, and @code{DIRECT_IO}, for example).
20345 * Changes to DECLIB::
20348 @node Changes to DECLIB
20349 @subsection Changes to @code{DECLIB}
20352 The changes made to the HP Ada predefined library for GNAT and post-Ada 83
20353 compatibility are minor and include the following:
20356 @item Adjusting the location of pragmas and record representation
20357 clauses to obey Ada 95 (and thus Ada 2005) rules
20359 @item Adding the proper notation to generic formal parameters
20360 that take unconstrained types in instantiation
20362 @item Adding pragma @code{ELABORATE_BODY} to package specs
20363 that have package bodies not otherwise allowed
20365 @item Replacing occurrences of the identifier ``@code{PROTECTED}'' by
20366 ``@code{PROTECTD}''.
20367 Currently these are found only in the @code{STARLET} package spec.
20369 @item Changing @code{SYSTEM.ADDRESS} to @code{SYSTEM.SHORT_ADDRESS}
20370 where the address size is constrained to 32 bits.
20374 None of the above changes is visible to users.
20380 On OpenVMS Alpha, HP Ada provides the following strongly-typed bindings:
20383 @item Command Language Interpreter (CLI interface)
20385 @item DECtalk Run-Time Library (DTK interface)
20387 @item Librarian utility routines (LBR interface)
20389 @item General Purpose Run-Time Library (LIB interface)
20391 @item Math Run-Time Library (MTH interface)
20393 @item National Character Set Run-Time Library (NCS interface)
20395 @item Compiled Code Support Run-Time Library (OTS interface)
20397 @item Parallel Processing Run-Time Library (PPL interface)
20399 @item Screen Management Run-Time Library (SMG interface)
20401 @item Sort Run-Time Library (SOR interface)
20403 @item String Run-Time Library (STR interface)
20405 @item STARLET System Library
20408 @item X Window System Version 11R4 and 11R5 (X, XLIB interface)
20410 @item X Windows Toolkit (XT interface)
20412 @item X/Motif Version 1.1.3 and 1.2 (XM interface)
20416 GNAT provides implementations of these HP bindings in the @code{DECLIB}
20417 directory, on both the Alpha and I64 OpenVMS platforms.
20419 The X/Motif bindings used to build @code{DECLIB} are whatever versions are
20421 HP Ada @file{ADA$PREDEFINED} directory with extension @file{.ADC}.
20422 A pragma @code{Linker_Options} has been added to packages @code{Xm},
20423 @code{Xt}, and @code{X_Lib}
20424 causing the default X/Motif sharable image libraries to be linked in. This
20425 is done via options files named @file{xm.opt}, @file{xt.opt}, and
20426 @file{x_lib.opt} (also located in the @file{DECLIB} directory).
20428 It may be necessary to edit these options files to update or correct the
20429 library names if, for example, the newer X/Motif bindings from
20430 @file{ADA$EXAMPLES}
20431 had been (previous to installing GNAT) copied and renamed to supersede the
20432 default @file{ADA$PREDEFINED} versions.
20435 * Shared Libraries and Options Files::
20436 * Interfaces to C::
20439 @node Shared Libraries and Options Files
20440 @subsection Shared Libraries and Options Files
20443 When using the HP Ada
20444 predefined X and Motif bindings, the linking with their sharable images is
20445 done automatically by @command{GNAT LINK}.
20446 When using other X and Motif bindings, you need
20447 to add the corresponding sharable images to the command line for
20448 @code{GNAT LINK}. When linking with shared libraries, or with
20449 @file{.OPT} files, you must
20450 also add them to the command line for @command{GNAT LINK}.
20452 A shared library to be used with GNAT is built in the same way as other
20453 libraries under VMS. The VMS Link command can be used in standard fashion.
20455 @node Interfaces to C
20456 @subsection Interfaces to C
20460 provides the following Ada types and operations:
20463 @item C types package (@code{C_TYPES})
20465 @item C strings (@code{C_TYPES.NULL_TERMINATED})
20467 @item Other_types (@code{SHORT_INT})
20471 Interfacing to C with GNAT, you can use the above approach
20472 described for HP Ada or the facilities of Annex B of
20473 the @cite{Ada Reference Manual} (packages @code{INTERFACES.C},
20474 @code{INTERFACES.C.STRINGS} and @code{INTERFACES.C.POINTERS}). For more
20475 information, see @ref{Interfacing to C,,, gnat_rm, GNAT Reference Manual}.
20477 The @option{-gnatF} qualifier forces default and explicit
20478 @code{External_Name} parameters in pragmas @code{Import} and @code{Export}
20479 to be uppercased for compatibility with the default behavior
20480 of HP C. The qualifier has no effect on @code{Link_Name} parameters.
20482 @node Main Program Definition
20483 @section Main Program Definition
20486 The following section discusses differences in the
20487 definition of main programs on HP Ada and GNAT.
20488 On HP Ada, main programs are defined to meet the
20489 following conditions:
20491 @item Procedure with no formal parameters (returns @code{0} upon
20494 @item Procedure with no formal parameters (returns @code{42} when
20495 an unhandled exception is raised)
20497 @item Function with no formal parameters whose returned value
20498 is of a discrete type
20500 @item Procedure with one @code{out} formal of a discrete type for
20501 which a specification of pragma @code{EXPORT_VALUED_PROCEDURE} is given.
20506 When declared with the pragma @code{EXPORT_VALUED_PROCEDURE},
20507 a main function or main procedure returns a discrete
20508 value whose size is less than 64 bits (32 on VAX systems),
20509 the value is zero- or sign-extended as appropriate.
20510 On GNAT, main programs are defined as follows:
20512 @item Must be a non-generic, parameterless subprogram that
20513 is either a procedure or function returning an Ada
20514 @code{STANDARD.INTEGER} (the predefined type)
20516 @item Cannot be a generic subprogram or an instantiation of a
20520 @node Implementation-Defined Attributes
20521 @section Implementation-Defined Attributes
20524 GNAT provides all HP Ada implementation-defined
20527 @node Compiler and Run-Time Interfacing
20528 @section Compiler and Run-Time Interfacing
20531 HP Ada provides the following qualifiers to pass options to the linker
20534 @item @option{/WAIT} and @option{/SUBMIT}
20536 @item @option{/COMMAND}
20538 @item @option{/@r{[}NO@r{]}MAP}
20540 @item @option{/OUTPUT=@var{file-spec}}
20542 @item @option{/@r{[}NO@r{]}DEBUG} and @option{/@r{[}NO@r{]}TRACEBACK}
20546 To pass options to the linker, GNAT provides the following
20550 @item @option{/EXECUTABLE=@var{exec-name}}
20552 @item @option{/VERBOSE}
20554 @item @option{/@r{[}NO@r{]}DEBUG} and @option{/@r{[}NO@r{]}TRACEBACK}
20558 For more information on these switches, see
20559 @ref{Switches for gnatlink}.
20560 In HP Ada, the command-line switch @option{/OPTIMIZE} is available
20561 to control optimization. HP Ada also supplies the
20564 @item @code{OPTIMIZE}
20566 @item @code{INLINE}
20568 @item @code{INLINE_GENERIC}
20570 @item @code{SUPPRESS_ALL}
20572 @item @code{PASSIVE}
20576 In GNAT, optimization is controlled strictly by command
20577 line parameters, as described in the corresponding section of this guide.
20578 The HP pragmas for control of optimization are
20579 recognized but ignored.
20581 Note that in GNAT, the default is optimization off, whereas in HP Ada
20582 the default is that optimization is turned on.
20584 @node Program Compilation and Library Management
20585 @section Program Compilation and Library Management
20588 HP Ada and GNAT provide a comparable set of commands to
20589 build programs. HP Ada also provides a program library,
20590 which is a concept that does not exist on GNAT. Instead,
20591 GNAT provides directories of sources that are compiled as
20594 The following table summarizes
20595 the HP Ada commands and provides
20596 equivalent GNAT commands. In this table, some GNAT
20597 equivalents reflect the fact that GNAT does not use the
20598 concept of a program library. Instead, it uses a model
20599 in which collections of source and object files are used
20600 in a manner consistent with other languages like C and
20601 Fortran. Therefore, standard system file commands are used
20602 to manipulate these elements. Those GNAT commands are marked with
20604 Note that, unlike HP Ada, none of the GNAT commands accepts wild cards.
20607 @multitable @columnfractions .35 .65
20609 @item @emph{HP Ada Command}
20610 @tab @emph{GNAT Equivalent / Description}
20612 @item @command{ADA}
20613 @tab @command{GNAT COMPILE}@*
20614 Invokes the compiler to compile one or more Ada source files.
20616 @item @command{ACS ATTACH}@*
20617 @tab [No equivalent]@*
20618 Switches control of terminal from current process running the program
20621 @item @command{ACS CHECK}
20622 @tab @command{GNAT MAKE /DEPENDENCY_LIST}@*
20623 Forms the execution closure of one
20624 or more compiled units and checks completeness and currency.
20626 @item @command{ACS COMPILE}
20627 @tab @command{GNAT MAKE /ACTIONS=COMPILE}@*
20628 Forms the execution closure of one or
20629 more specified units, checks completeness and currency,
20630 identifies units that have revised source files, compiles same,
20631 and recompiles units that are or will become obsolete.
20632 Also completes incomplete generic instantiations.
20634 @item @command{ACS COPY FOREIGN}
20636 Copies a foreign object file into the program library as a
20639 @item @command{ACS COPY UNIT}
20641 Copies a compiled unit from one program library to another.
20643 @item @command{ACS CREATE LIBRARY}
20644 @tab Create /directory (*)@*
20645 Creates a program library.
20647 @item @command{ACS CREATE SUBLIBRARY}
20648 @tab Create /directory (*)@*
20649 Creates a program sublibrary.
20651 @item @command{ACS DELETE LIBRARY}
20653 Deletes a program library and its contents.
20655 @item @command{ACS DELETE SUBLIBRARY}
20657 Deletes a program sublibrary and its contents.
20659 @item @command{ACS DELETE UNIT}
20660 @tab Delete file (*)@*
20661 On OpenVMS systems, deletes one or more compiled units from
20662 the current program library.
20664 @item @command{ACS DIRECTORY}
20665 @tab Directory (*)@*
20666 On OpenVMS systems, lists units contained in the current
20669 @item @command{ACS ENTER FOREIGN}
20671 Allows the import of a foreign body as an Ada library
20672 spec and enters a reference to a pointer.
20674 @item @command{ACS ENTER UNIT}
20676 Enters a reference (pointer) from the current program library to
20677 a unit compiled into another program library.
20679 @item @command{ACS EXIT}
20680 @tab [No equivalent]@*
20681 Exits from the program library manager.
20683 @item @command{ACS EXPORT}
20685 Creates an object file that contains system-specific object code
20686 for one or more units. With GNAT, object files can simply be copied
20687 into the desired directory.
20689 @item @command{ACS EXTRACT SOURCE}
20691 Allows access to the copied source file for each Ada compilation unit
20693 @item @command{ACS HELP}
20694 @tab @command{HELP GNAT}@*
20695 Provides online help.
20697 @item @command{ACS LINK}
20698 @tab @command{GNAT LINK}@*
20699 Links an object file containing Ada units into an executable file.
20701 @item @command{ACS LOAD}
20703 Loads (partially compiles) Ada units into the program library.
20704 Allows loading a program from a collection of files into a library
20705 without knowing the relationship among units.
20707 @item @command{ACS MERGE}
20709 Merges into the current program library, one or more units from
20710 another library where they were modified.
20712 @item @command{ACS RECOMPILE}
20713 @tab @command{GNAT MAKE /ACTIONS=COMPILE}@*
20714 Recompiles from external or copied source files any obsolete
20715 unit in the closure. Also, completes any incomplete generic
20718 @item @command{ACS REENTER}
20719 @tab @command{GNAT MAKE}@*
20720 Reenters current references to units compiled after last entered
20721 with the @command{ACS ENTER UNIT} command.
20723 @item @command{ACS SET LIBRARY}
20724 @tab Set default (*)@*
20725 Defines a program library to be the compilation context as well
20726 as the target library for compiler output and commands in general.
20728 @item @command{ACS SET PRAGMA}
20729 @tab Edit @file{gnat.adc} (*)@*
20730 Redefines specified values of the library characteristics
20731 @code{LONG_ FLOAT}, @code{MEMORY_SIZE}, @code{SYSTEM_NAME},
20732 and @code{Float_Representation}.
20734 @item @command{ACS SET SOURCE}
20735 @tab Define @code{ADA_INCLUDE_PATH} path (*)@*
20736 Defines the source file search list for the @command{ACS COMPILE} command.
20738 @item @command{ACS SHOW LIBRARY}
20739 @tab Directory (*)@*
20740 Lists information about one or more program libraries.
20742 @item @command{ACS SHOW PROGRAM}
20743 @tab [No equivalent]@*
20744 Lists information about the execution closure of one or
20745 more units in the program library.
20747 @item @command{ACS SHOW SOURCE}
20748 @tab Show logical @code{ADA_INCLUDE_PATH}@*
20749 Shows the source file search used when compiling units.
20751 @item @command{ACS SHOW VERSION}
20752 @tab Compile with @option{VERBOSE} option
20753 Displays the version number of the compiler and program library
20756 @item @command{ACS SPAWN}
20757 @tab [No equivalent]@*
20758 Creates a subprocess of the current process (same as @command{DCL SPAWN}
20761 @item @command{ACS VERIFY}
20762 @tab [No equivalent]@*
20763 Performs a series of consistency checks on a program library to
20764 determine whether the library structure and library files are in
20771 @section Input-Output
20774 On OpenVMS Alpha systems, HP Ada uses OpenVMS Record
20775 Management Services (RMS) to perform operations on
20779 HP Ada and GNAT predefine an identical set of input-
20780 output packages. To make the use of the
20781 generic @code{TEXT_IO} operations more convenient, HP Ada
20782 provides predefined library packages that instantiate the
20783 integer and floating-point operations for the predefined
20784 integer and floating-point types as shown in the following table.
20786 @multitable @columnfractions .45 .55
20787 @item @emph{Package Name} @tab Instantiation
20789 @item @code{INTEGER_TEXT_IO}
20790 @tab @code{INTEGER_IO(INTEGER)}
20792 @item @code{SHORT_INTEGER_TEXT_IO}
20793 @tab @code{INTEGER_IO(SHORT_INTEGER)}
20795 @item @code{SHORT_SHORT_INTEGER_TEXT_IO}
20796 @tab @code{INTEGER_IO(SHORT_SHORT_INTEGER)}
20798 @item @code{FLOAT_TEXT_IO}
20799 @tab @code{FLOAT_IO(FLOAT)}
20801 @item @code{LONG_FLOAT_TEXT_IO}
20802 @tab @code{FLOAT_IO(LONG_FLOAT)}
20806 The HP Ada predefined packages and their operations
20807 are implemented using OpenVMS Alpha files and input-output
20808 facilities. HP Ada supports asynchronous input-output on OpenVMS Alpha.
20809 Familiarity with the following is recommended:
20811 @item RMS file organizations and access methods
20813 @item OpenVMS file specifications and directories
20815 @item OpenVMS File Definition Language (FDL)
20819 GNAT provides I/O facilities that are completely
20820 compatible with HP Ada. The distribution includes the
20821 standard HP Ada versions of all I/O packages, operating
20822 in a manner compatible with HP Ada. In particular, the
20823 following packages are by default the HP Ada (Ada 83)
20824 versions of these packages rather than the renamings
20825 suggested in Annex J of the Ada Reference Manual:
20827 @item @code{TEXT_IO}
20829 @item @code{SEQUENTIAL_IO}
20831 @item @code{DIRECT_IO}
20835 The use of the standard child package syntax (for
20836 example, @code{ADA.TEXT_IO}) retrieves the post-Ada 83 versions of these
20838 GNAT provides HP-compatible predefined instantiations
20839 of the @code{TEXT_IO} packages, and also
20840 provides the standard predefined instantiations required
20841 by the @cite{Ada Reference Manual}.
20843 For further information on how GNAT interfaces to the file
20844 system or how I/O is implemented in programs written in
20845 mixed languages, see @ref{Implementation of the Standard I/O,,,
20846 gnat_rm, GNAT Reference Manual}.
20847 This chapter covers the following:
20849 @item Standard I/O packages
20851 @item @code{FORM} strings
20853 @item @code{ADA.DIRECT_IO}
20855 @item @code{ADA.SEQUENTIAL_IO}
20857 @item @code{ADA.TEXT_IO}
20859 @item Stream pointer positioning
20861 @item Reading and writing non-regular files
20863 @item @code{GET_IMMEDIATE}
20865 @item Treating @code{TEXT_IO} files as streams
20872 @node Implementation Limits
20873 @section Implementation Limits
20876 The following table lists implementation limits for HP Ada
20878 @multitable @columnfractions .60 .20 .20
20880 @item @emph{Compilation Parameter}
20885 @item In a subprogram or entry declaration, maximum number of
20886 formal parameters that are of an unconstrained record type
20891 @item Maximum identifier length (number of characters)
20896 @item Maximum number of characters in a source line
20901 @item Maximum collection size (number of bytes)
20906 @item Maximum number of discriminants for a record type
20911 @item Maximum number of formal parameters in an entry or
20912 subprogram declaration
20917 @item Maximum number of dimensions in an array type
20922 @item Maximum number of library units and subunits in a compilation.
20927 @item Maximum number of library units and subunits in an execution.
20932 @item Maximum number of objects declared with the pragma @code{COMMON_OBJECT}
20933 or @code{PSECT_OBJECT}
20938 @item Maximum number of enumeration literals in an enumeration type
20944 @item Maximum number of lines in a source file
20949 @item Maximum number of bits in any object
20954 @item Maximum size of the static portion of a stack frame (approximate)
20959 @node Tools and Utilities
20960 @section Tools and Utilities
20963 The following table lists some of the OpenVMS development tools
20964 available for HP Ada, and the corresponding tools for
20965 use with @value{EDITION} on Alpha and I64 platforms.
20966 Aside from the debugger, all the OpenVMS tools identified are part
20967 of the DECset package.
20970 @c Specify table in TeX since Texinfo does a poor job
20974 \settabs\+Language-Sensitive Editor\quad
20975 &Product with HP Ada\quad
20978 &\it Product with HP Ada
20979 & \it Product with GNAT Pro\cr
20981 \+Code Management System
20985 \+Language-Sensitive Editor
20987 & emacs or HP LSE (Alpha)\cr
20997 & OpenVMS Debug (I64)\cr
20999 \+Source Code Analyzer /
21016 \+Coverage Analyzer
21020 \+Module Management
21022 & Not applicable\cr
21032 @c This is the Texinfo version of the table. It renders poorly in pdf, hence
21033 @c the TeX version above for the printed version
21035 @c @multitable @columnfractions .3 .4 .4
21036 @multitable {Source Code Analyzer /}{Tool with HP Ada}{Tool with GNAT Pro}
21038 @tab @i{Tool with HP Ada}
21039 @tab @i{Tool with @value{EDITION}}
21040 @item Code Management@*System
21043 @item Language-Sensitive@*Editor
21045 @tab emacs or HP LSE (Alpha)
21054 @tab OpenVMS Debug (I64)
21055 @item Source Code Analyzer /@*Cross Referencer
21059 @tab HP Digital Test@*Manager (DTM)
21061 @item Performance and@*Coverage Analyzer
21064 @item Module Management@*System
21066 @tab Not applicable
21073 @c **************************************
21074 @node Platform-Specific Information for the Run-Time Libraries
21075 @appendix Platform-Specific Information for the Run-Time Libraries
21076 @cindex Tasking and threads libraries
21077 @cindex Threads libraries and tasking
21078 @cindex Run-time libraries (platform-specific information)
21081 The GNAT run-time implementation may vary with respect to both the
21082 underlying threads library and the exception handling scheme.
21083 For threads support, one or more of the following are supplied:
21085 @item @b{native threads library}, a binding to the thread package from
21086 the underlying operating system
21088 @item @b{pthreads library} (Sparc Solaris only), a binding to the Solaris
21089 POSIX thread package
21093 For exception handling, either or both of two models are supplied:
21095 @item @b{Zero-Cost Exceptions} (``ZCX''),@footnote{
21096 Most programs should experience a substantial speed improvement by
21097 being compiled with a ZCX run-time.
21098 This is especially true for
21099 tasking applications or applications with many exception handlers.}
21100 @cindex Zero-Cost Exceptions
21101 @cindex ZCX (Zero-Cost Exceptions)
21102 which uses binder-generated tables that
21103 are interrogated at run time to locate a handler
21105 @item @b{setjmp / longjmp} (``SJLJ''),
21106 @cindex setjmp/longjmp Exception Model
21107 @cindex SJLJ (setjmp/longjmp Exception Model)
21108 which uses dynamically-set data to establish
21109 the set of handlers
21113 This appendix summarizes which combinations of threads and exception support
21114 are supplied on various GNAT platforms.
21115 It then shows how to select a particular library either
21116 permanently or temporarily,
21117 explains the properties of (and tradeoffs among) the various threads
21118 libraries, and provides some additional
21119 information about several specific platforms.
21122 * Summary of Run-Time Configurations::
21123 * Specifying a Run-Time Library::
21124 * Choosing the Scheduling Policy::
21125 * Solaris-Specific Considerations::
21126 * Linux-Specific Considerations::
21127 * AIX-Specific Considerations::
21128 * Irix-Specific Considerations::
21129 * RTX-Specific Considerations::
21130 * HP-UX-Specific Considerations::
21133 @node Summary of Run-Time Configurations
21134 @section Summary of Run-Time Configurations
21136 @multitable @columnfractions .30 .70
21137 @item @b{alpha-openvms}
21138 @item @code{@ @ }@i{rts-native (default)}
21139 @item @code{@ @ @ @ }Tasking @tab native VMS threads
21140 @item @code{@ @ @ @ }Exceptions @tab ZCX
21142 @item @b{alpha-tru64}
21143 @item @code{@ @ }@i{rts-native (default)}
21144 @item @code{@ @ @ @ }Tasking @tab native TRU64 threads
21145 @item @code{@ @ @ @ }Exceptions @tab ZCX
21147 @item @code{@ @ }@i{rts-sjlj}
21148 @item @code{@ @ @ @ }Tasking @tab native TRU64 threads
21149 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21151 @item @b{ia64-hp_linux}
21152 @item @code{@ @ }@i{rts-native (default)}
21153 @item @code{@ @ @ @ }Tasking @tab pthread library
21154 @item @code{@ @ @ @ }Exceptions @tab ZCX
21156 @item @b{ia64-hpux}
21157 @item @code{@ @ }@i{rts-native (default)}
21158 @item @code{@ @ @ @ }Tasking @tab native HP-UX threads
21159 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21161 @item @b{ia64-openvms}
21162 @item @code{@ @ }@i{rts-native (default)}
21163 @item @code{@ @ @ @ }Tasking @tab native VMS threads
21164 @item @code{@ @ @ @ }Exceptions @tab ZCX
21166 @item @b{ia64-sgi_linux}
21167 @item @code{@ @ }@i{rts-native (default)}
21168 @item @code{@ @ @ @ }Tasking @tab pthread library
21169 @item @code{@ @ @ @ }Exceptions @tab ZCX
21171 @item @b{mips-irix}
21172 @item @code{@ @ }@i{rts-native (default)}
21173 @item @code{@ @ @ @ }Tasking @tab native IRIX threads
21174 @item @code{@ @ @ @ }Exceptions @tab ZCX
21177 @item @code{@ @ }@i{rts-native (default)}
21178 @item @code{@ @ @ @ }Tasking @tab native HP-UX threads
21179 @item @code{@ @ @ @ }Exceptions @tab ZCX
21181 @item @code{@ @ }@i{rts-sjlj}
21182 @item @code{@ @ @ @ }Tasking @tab native HP-UX threads
21183 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21186 @item @code{@ @ }@i{rts-native (default)}
21187 @item @code{@ @ @ @ }Tasking @tab native AIX threads
21188 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21190 @item @b{ppc-darwin}
21191 @item @code{@ @ }@i{rts-native (default)}
21192 @item @code{@ @ @ @ }Tasking @tab native MacOS threads
21193 @item @code{@ @ @ @ }Exceptions @tab ZCX
21195 @item @b{sparc-solaris} @tab
21196 @item @code{@ @ }@i{rts-native (default)}
21197 @item @code{@ @ @ @ }Tasking @tab native Solaris threads library
21198 @item @code{@ @ @ @ }Exceptions @tab ZCX
21200 @item @code{@ @ }@i{rts-pthread}
21201 @item @code{@ @ @ @ }Tasking @tab pthread library
21202 @item @code{@ @ @ @ }Exceptions @tab ZCX
21204 @item @code{@ @ }@i{rts-sjlj}
21205 @item @code{@ @ @ @ }Tasking @tab native Solaris threads library
21206 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21208 @item @b{sparc64-solaris} @tab
21209 @item @code{@ @ }@i{rts-native (default)}
21210 @item @code{@ @ @ @ }Tasking @tab native Solaris threads library
21211 @item @code{@ @ @ @ }Exceptions @tab ZCX
21213 @item @b{x86-linux}
21214 @item @code{@ @ }@i{rts-native (default)}
21215 @item @code{@ @ @ @ }Tasking @tab pthread library
21216 @item @code{@ @ @ @ }Exceptions @tab ZCX
21218 @item @code{@ @ }@i{rts-sjlj}
21219 @item @code{@ @ @ @ }Tasking @tab pthread library
21220 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21223 @item @code{@ @ }@i{rts-native (default)}
21224 @item @code{@ @ @ @ }Tasking @tab native LynxOS threads
21225 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21227 @item @b{x86-solaris}
21228 @item @code{@ @ }@i{rts-native (default)}
21229 @item @code{@ @ @ @ }Tasking @tab native Solaris threads
21230 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21232 @item @b{x86-windows}
21233 @item @code{@ @ }@i{rts-native (default)}
21234 @item @code{@ @ @ @ }Tasking @tab native Win32 threads
21235 @item @code{@ @ @ @ }Exceptions @tab ZCX
21237 @item @code{@ @ }@i{rts-sjlj (default)}
21238 @item @code{@ @ @ @ }Tasking @tab native Win32 threads
21239 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21241 @item @b{x86-windows-rtx}
21242 @item @code{@ @ }@i{rts-rtx-rtss (default)}
21243 @item @code{@ @ @ @ }Tasking @tab RTX real-time subsystem RTSS threads (kernel mode)
21244 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21246 @item @code{@ @ }@i{rts-rtx-w32}
21247 @item @code{@ @ @ @ }Tasking @tab RTX Win32 threads (user mode)
21248 @item @code{@ @ @ @ }Exceptions @tab ZCX
21250 @item @b{x86_64-linux}
21251 @item @code{@ @ }@i{rts-native (default)}
21252 @item @code{@ @ @ @ }Tasking @tab pthread library
21253 @item @code{@ @ @ @ }Exceptions @tab ZCX
21255 @item @code{@ @ }@i{rts-sjlj}
21256 @item @code{@ @ @ @ }Tasking @tab pthread library
21257 @item @code{@ @ @ @ }Exceptions @tab SJLJ
21261 @node Specifying a Run-Time Library
21262 @section Specifying a Run-Time Library
21265 The @file{adainclude} subdirectory containing the sources of the GNAT
21266 run-time library, and the @file{adalib} subdirectory containing the
21267 @file{ALI} files and the static and/or shared GNAT library, are located
21268 in the gcc target-dependent area:
21271 target=$prefix/lib/gcc/gcc-@i{dumpmachine}/gcc-@i{dumpversion}/
21275 As indicated above, on some platforms several run-time libraries are supplied.
21276 These libraries are installed in the target dependent area and
21277 contain a complete source and binary subdirectory. The detailed description
21278 below explains the differences between the different libraries in terms of
21279 their thread support.
21281 The default run-time library (when GNAT is installed) is @emph{rts-native}.
21282 This default run time is selected by the means of soft links.
21283 For example on x86-linux:
21289 +--- adainclude----------+
21291 +--- adalib-----------+ |
21293 +--- rts-native | |
21295 | +--- adainclude <---+
21297 | +--- adalib <----+
21308 If the @i{rts-sjlj} library is to be selected on a permanent basis,
21309 these soft links can be modified with the following commands:
21313 $ rm -f adainclude adalib
21314 $ ln -s rts-sjlj/adainclude adainclude
21315 $ ln -s rts-sjlj/adalib adalib
21319 Alternatively, you can specify @file{rts-sjlj/adainclude} in the file
21320 @file{$target/ada_source_path} and @file{rts-sjlj/adalib} in
21321 @file{$target/ada_object_path}.
21323 Selecting another run-time library temporarily can be
21324 achieved by using the @option{--RTS} switch, e.g., @option{--RTS=sjlj}
21325 @cindex @option{--RTS} option
21327 @node Choosing the Scheduling Policy
21328 @section Choosing the Scheduling Policy
21331 When using a POSIX threads implementation, you have a choice of several
21332 scheduling policies: @code{SCHED_FIFO},
21333 @cindex @code{SCHED_FIFO} scheduling policy
21335 @cindex @code{SCHED_RR} scheduling policy
21336 and @code{SCHED_OTHER}.
21337 @cindex @code{SCHED_OTHER} scheduling policy
21338 Typically, the default is @code{SCHED_OTHER}, while using @code{SCHED_FIFO}
21339 or @code{SCHED_RR} requires special (e.g., root) privileges.
21341 By default, GNAT uses the @code{SCHED_OTHER} policy. To specify
21343 @cindex @code{SCHED_FIFO} scheduling policy
21344 you can use one of the following:
21348 @code{pragma Time_Slice (0.0)}
21349 @cindex pragma Time_Slice
21351 the corresponding binder option @option{-T0}
21352 @cindex @option{-T0} option
21354 @code{pragma Task_Dispatching_Policy (FIFO_Within_Priorities)}
21355 @cindex pragma Task_Dispatching_Policy
21359 To specify @code{SCHED_RR},
21360 @cindex @code{SCHED_RR} scheduling policy
21361 you should use @code{pragma Time_Slice} with a
21362 value greater than @code{0.0}, or else use the corresponding @option{-T}
21365 @node Solaris-Specific Considerations
21366 @section Solaris-Specific Considerations
21367 @cindex Solaris Sparc threads libraries
21370 This section addresses some topics related to the various threads libraries
21374 * Solaris Threads Issues::
21377 @node Solaris Threads Issues
21378 @subsection Solaris Threads Issues
21381 GNAT under Solaris/Sparc 32 bits comes with an alternate tasking run-time
21382 library based on POSIX threads --- @emph{rts-pthread}.
21383 @cindex rts-pthread threads library
21384 This run-time library has the advantage of being mostly shared across all
21385 POSIX-compliant thread implementations, and it also provides under
21386 @w{Solaris 8} the @code{PTHREAD_PRIO_INHERIT}
21387 @cindex @code{PTHREAD_PRIO_INHERIT} policy (under rts-pthread)
21388 and @code{PTHREAD_PRIO_PROTECT}
21389 @cindex @code{PTHREAD_PRIO_PROTECT} policy (under rts-pthread)
21390 semantics that can be selected using the predefined pragma
21391 @code{Locking_Policy}
21392 @cindex pragma Locking_Policy (under rts-pthread)
21394 @code{Inheritance_Locking} and @code{Ceiling_Locking} as the policy.
21395 @cindex @code{Inheritance_Locking} (under rts-pthread)
21396 @cindex @code{Ceiling_Locking} (under rts-pthread)
21398 As explained above, the native run-time library is based on the Solaris thread
21399 library (@code{libthread}) and is the default library.
21401 When the Solaris threads library is used (this is the default), programs
21402 compiled with GNAT can automatically take advantage of
21403 and can thus execute on multiple processors.
21404 The user can alternatively specify a processor on which the program should run
21405 to emulate a single-processor system. The multiprocessor / uniprocessor choice
21407 setting the environment variable @env{GNAT_PROCESSOR}
21408 @cindex @env{GNAT_PROCESSOR} environment variable (on Sparc Solaris)
21409 to one of the following:
21413 Use the default configuration (run the program on all
21414 available processors) - this is the same as having @code{GNAT_PROCESSOR}
21418 Let the run-time implementation choose one processor and run the program on
21421 @item 0 .. Last_Proc
21422 Run the program on the specified processor.
21423 @code{Last_Proc} is equal to @code{_SC_NPROCESSORS_CONF - 1}
21424 (where @code{_SC_NPROCESSORS_CONF} is a system variable).
21427 @node Linux-Specific Considerations
21428 @section Linux-Specific Considerations
21429 @cindex Linux threads libraries
21432 On GNU/Linux without NPTL support (usually system with GNU C Library
21433 older than 2.3), the signal model is not POSIX compliant, which means
21434 that to send a signal to the process, you need to send the signal to all
21435 threads, e.g.@: by using @code{killpg()}.
21437 @node AIX-Specific Considerations
21438 @section AIX-Specific Considerations
21439 @cindex AIX resolver library
21442 On AIX, the resolver library initializes some internal structure on
21443 the first call to @code{get*by*} functions, which are used to implement
21444 @code{GNAT.Sockets.Get_Host_By_Name} and
21445 @code{GNAT.Sockets.Get_Host_By_Address}.
21446 If such initialization occurs within an Ada task, and the stack size for
21447 the task is the default size, a stack overflow may occur.
21449 To avoid this overflow, the user should either ensure that the first call
21450 to @code{GNAT.Sockets.Get_Host_By_Name} or
21451 @code{GNAT.Sockets.Get_Host_By_Addrss}
21452 occurs in the environment task, or use @code{pragma Storage_Size} to
21453 specify a sufficiently large size for the stack of the task that contains
21456 @node Irix-Specific Considerations
21457 @section Irix-Specific Considerations
21458 @cindex Irix libraries
21461 The GCC support libraries coming with the Irix compiler have moved to
21462 their canonical place with respect to the general Irix ABI related
21463 conventions. Running applications built with the default shared GNAT
21464 run-time now requires the LD_LIBRARY_PATH environment variable to
21465 include this location. A possible way to achieve this is to issue the
21466 following command line on a bash prompt:
21470 $ LD_LIBRARY_PATH=$LD_LIBRARY_PATH:`dirname \`gcc --print-file-name=libgcc_s.so\``
21474 @node RTX-Specific Considerations
21475 @section RTX-Specific Considerations
21476 @cindex RTX libraries
21479 The Real-time Extension (RTX) to Windows is based on the Windows Win32
21480 API. Applications can be built to work in two different modes:
21484 Windows executables that run in Ring 3 to utilize memory protection
21485 (@emph{rts-rtx-w32}).
21488 Real-time subsystem (RTSS) executables that run in Ring 0, where
21489 performance can be optimized with RTSS applications taking precedent
21490 over all Windows applications (@emph{rts-rtx-rtss}). This mode requires
21491 the Microsoft linker to handle RTSS libraries.
21495 @node HP-UX-Specific Considerations
21496 @section HP-UX-Specific Considerations
21497 @cindex HP-UX Scheduling
21500 On HP-UX, appropriate privileges are required to change the scheduling
21501 parameters of a task. The calling process must have appropriate
21502 privileges or be a member of a group having @code{PRIV_RTSCHED} access to
21503 successfully change the scheduling parameters.
21505 By default, GNAT uses the @code{SCHED_HPUX} policy. To have access to the
21506 priority range 0-31 either the @code{FIFO_Within_Priorities} or the
21507 @code{Round_Robin_Within_Priorities} scheduling policies need to be set.
21509 To specify the @code{FIFO_Within_Priorities} scheduling policy you can use
21510 one of the following:
21514 @code{pragma Time_Slice (0.0)}
21515 @cindex pragma Time_Slice
21517 the corresponding binder option @option{-T0}
21518 @cindex @option{-T0} option
21520 @code{pragma Task_Dispatching_Policy (FIFO_Within_Priorities)}
21521 @cindex pragma Task_Dispatching_Policy
21525 To specify the @code{Round_Robin_Within_Priorities}, scheduling policy
21526 you should use @code{pragma Time_Slice} with a
21527 value greater than @code{0.0}, or use the corresponding @option{-T}
21528 binder option, or set the @code{pragma Task_Dispatching_Policy
21529 (Round_Robin_Within_Priorities)}.
21531 @c *******************************
21532 @node Example of Binder Output File
21533 @appendix Example of Binder Output File
21536 This Appendix displays the source code for @command{gnatbind}'s output
21537 file generated for a simple ``Hello World'' program.
21538 Comments have been added for clarification purposes.
21540 @smallexample @c adanocomment
21544 -- The package is called Ada_Main unless this name is actually used
21545 -- as a unit name in the partition, in which case some other unique
21549 package ada_main is
21551 Elab_Final_Code : Integer;
21552 pragma Import (C, Elab_Final_Code, "__gnat_inside_elab_final_code");
21554 -- The main program saves the parameters (argument count,
21555 -- argument values, environment pointer) in global variables
21556 -- for later access by other units including
21557 -- Ada.Command_Line.
21559 gnat_argc : Integer;
21560 gnat_argv : System.Address;
21561 gnat_envp : System.Address;
21563 -- The actual variables are stored in a library routine. This
21564 -- is useful for some shared library situations, where there
21565 -- are problems if variables are not in the library.
21567 pragma Import (C, gnat_argc);
21568 pragma Import (C, gnat_argv);
21569 pragma Import (C, gnat_envp);
21571 -- The exit status is similarly an external location
21573 gnat_exit_status : Integer;
21574 pragma Import (C, gnat_exit_status);
21576 GNAT_Version : constant String :=
21577 "GNAT Version: 6.0.0w (20061115)";
21578 pragma Export (C, GNAT_Version, "__gnat_version");
21580 -- This is the generated adafinal routine that performs
21581 -- finalization at the end of execution. In the case where
21582 -- Ada is the main program, this main program makes a call
21583 -- to adafinal at program termination.
21585 procedure adafinal;
21586 pragma Export (C, adafinal, "adafinal");
21588 -- This is the generated adainit routine that performs
21589 -- initialization at the start of execution. In the case
21590 -- where Ada is the main program, this main program makes
21591 -- a call to adainit at program startup.
21594 pragma Export (C, adainit, "adainit");
21596 -- This routine is called at the start of execution. It is
21597 -- a dummy routine that is used by the debugger to breakpoint
21598 -- at the start of execution.
21600 procedure Break_Start;
21601 pragma Import (C, Break_Start, "__gnat_break_start");
21603 -- This is the actual generated main program (it would be
21604 -- suppressed if the no main program switch were used). As
21605 -- required by standard system conventions, this program has
21606 -- the external name main.
21610 argv : System.Address;
21611 envp : System.Address)
21613 pragma Export (C, main, "main");
21615 -- The following set of constants give the version
21616 -- identification values for every unit in the bound
21617 -- partition. This identification is computed from all
21618 -- dependent semantic units, and corresponds to the
21619 -- string that would be returned by use of the
21620 -- Body_Version or Version attributes.
21622 type Version_32 is mod 2 ** 32;
21623 u00001 : constant Version_32 := 16#7880BEB3#;
21624 u00002 : constant Version_32 := 16#0D24CBD0#;
21625 u00003 : constant Version_32 := 16#3283DBEB#;
21626 u00004 : constant Version_32 := 16#2359F9ED#;
21627 u00005 : constant Version_32 := 16#664FB847#;
21628 u00006 : constant Version_32 := 16#68E803DF#;
21629 u00007 : constant Version_32 := 16#5572E604#;
21630 u00008 : constant Version_32 := 16#46B173D8#;
21631 u00009 : constant Version_32 := 16#156A40CF#;
21632 u00010 : constant Version_32 := 16#033DABE0#;
21633 u00011 : constant Version_32 := 16#6AB38FEA#;
21634 u00012 : constant Version_32 := 16#22B6217D#;
21635 u00013 : constant Version_32 := 16#68A22947#;
21636 u00014 : constant Version_32 := 16#18CC4A56#;
21637 u00015 : constant Version_32 := 16#08258E1B#;
21638 u00016 : constant Version_32 := 16#367D5222#;
21639 u00017 : constant Version_32 := 16#20C9ECA4#;
21640 u00018 : constant Version_32 := 16#50D32CB6#;
21641 u00019 : constant Version_32 := 16#39A8BB77#;
21642 u00020 : constant Version_32 := 16#5CF8FA2B#;
21643 u00021 : constant Version_32 := 16#2F1EB794#;
21644 u00022 : constant Version_32 := 16#31AB6444#;
21645 u00023 : constant Version_32 := 16#1574B6E9#;
21646 u00024 : constant Version_32 := 16#5109C189#;
21647 u00025 : constant Version_32 := 16#56D770CD#;
21648 u00026 : constant Version_32 := 16#02F9DE3D#;
21649 u00027 : constant Version_32 := 16#08AB6B2C#;
21650 u00028 : constant Version_32 := 16#3FA37670#;
21651 u00029 : constant Version_32 := 16#476457A0#;
21652 u00030 : constant Version_32 := 16#731E1B6E#;
21653 u00031 : constant Version_32 := 16#23C2E789#;
21654 u00032 : constant Version_32 := 16#0F1BD6A1#;
21655 u00033 : constant Version_32 := 16#7C25DE96#;
21656 u00034 : constant Version_32 := 16#39ADFFA2#;
21657 u00035 : constant Version_32 := 16#571DE3E7#;
21658 u00036 : constant Version_32 := 16#5EB646AB#;
21659 u00037 : constant Version_32 := 16#4249379B#;
21660 u00038 : constant Version_32 := 16#0357E00A#;
21661 u00039 : constant Version_32 := 16#3784FB72#;
21662 u00040 : constant Version_32 := 16#2E723019#;
21663 u00041 : constant Version_32 := 16#623358EA#;
21664 u00042 : constant Version_32 := 16#107F9465#;
21665 u00043 : constant Version_32 := 16#6843F68A#;
21666 u00044 : constant Version_32 := 16#63305874#;
21667 u00045 : constant Version_32 := 16#31E56CE1#;
21668 u00046 : constant Version_32 := 16#02917970#;
21669 u00047 : constant Version_32 := 16#6CCBA70E#;
21670 u00048 : constant Version_32 := 16#41CD4204#;
21671 u00049 : constant Version_32 := 16#572E3F58#;
21672 u00050 : constant Version_32 := 16#20729FF5#;
21673 u00051 : constant Version_32 := 16#1D4F93E8#;
21674 u00052 : constant Version_32 := 16#30B2EC3D#;
21675 u00053 : constant Version_32 := 16#34054F96#;
21676 u00054 : constant Version_32 := 16#5A199860#;
21677 u00055 : constant Version_32 := 16#0E7F912B#;
21678 u00056 : constant Version_32 := 16#5760634A#;
21679 u00057 : constant Version_32 := 16#5D851835#;
21681 -- The following Export pragmas export the version numbers
21682 -- with symbolic names ending in B (for body) or S
21683 -- (for spec) so that they can be located in a link. The
21684 -- information provided here is sufficient to track down
21685 -- the exact versions of units used in a given build.
21687 pragma Export (C, u00001, "helloB");
21688 pragma Export (C, u00002, "system__standard_libraryB");
21689 pragma Export (C, u00003, "system__standard_libraryS");
21690 pragma Export (C, u00004, "adaS");
21691 pragma Export (C, u00005, "ada__text_ioB");
21692 pragma Export (C, u00006, "ada__text_ioS");
21693 pragma Export (C, u00007, "ada__exceptionsB");
21694 pragma Export (C, u00008, "ada__exceptionsS");
21695 pragma Export (C, u00009, "gnatS");
21696 pragma Export (C, u00010, "gnat__heap_sort_aB");
21697 pragma Export (C, u00011, "gnat__heap_sort_aS");
21698 pragma Export (C, u00012, "systemS");
21699 pragma Export (C, u00013, "system__exception_tableB");
21700 pragma Export (C, u00014, "system__exception_tableS");
21701 pragma Export (C, u00015, "gnat__htableB");
21702 pragma Export (C, u00016, "gnat__htableS");
21703 pragma Export (C, u00017, "system__exceptionsS");
21704 pragma Export (C, u00018, "system__machine_state_operationsB");
21705 pragma Export (C, u00019, "system__machine_state_operationsS");
21706 pragma Export (C, u00020, "system__machine_codeS");
21707 pragma Export (C, u00021, "system__storage_elementsB");
21708 pragma Export (C, u00022, "system__storage_elementsS");
21709 pragma Export (C, u00023, "system__secondary_stackB");
21710 pragma Export (C, u00024, "system__secondary_stackS");
21711 pragma Export (C, u00025, "system__parametersB");
21712 pragma Export (C, u00026, "system__parametersS");
21713 pragma Export (C, u00027, "system__soft_linksB");
21714 pragma Export (C, u00028, "system__soft_linksS");
21715 pragma Export (C, u00029, "system__stack_checkingB");
21716 pragma Export (C, u00030, "system__stack_checkingS");
21717 pragma Export (C, u00031, "system__tracebackB");
21718 pragma Export (C, u00032, "system__tracebackS");
21719 pragma Export (C, u00033, "ada__streamsS");
21720 pragma Export (C, u00034, "ada__tagsB");
21721 pragma Export (C, u00035, "ada__tagsS");
21722 pragma Export (C, u00036, "system__string_opsB");
21723 pragma Export (C, u00037, "system__string_opsS");
21724 pragma Export (C, u00038, "interfacesS");
21725 pragma Export (C, u00039, "interfaces__c_streamsB");
21726 pragma Export (C, u00040, "interfaces__c_streamsS");
21727 pragma Export (C, u00041, "system__file_ioB");
21728 pragma Export (C, u00042, "system__file_ioS");
21729 pragma Export (C, u00043, "ada__finalizationB");
21730 pragma Export (C, u00044, "ada__finalizationS");
21731 pragma Export (C, u00045, "system__finalization_rootB");
21732 pragma Export (C, u00046, "system__finalization_rootS");
21733 pragma Export (C, u00047, "system__finalization_implementationB");
21734 pragma Export (C, u00048, "system__finalization_implementationS");
21735 pragma Export (C, u00049, "system__string_ops_concat_3B");
21736 pragma Export (C, u00050, "system__string_ops_concat_3S");
21737 pragma Export (C, u00051, "system__stream_attributesB");
21738 pragma Export (C, u00052, "system__stream_attributesS");
21739 pragma Export (C, u00053, "ada__io_exceptionsS");
21740 pragma Export (C, u00054, "system__unsigned_typesS");
21741 pragma Export (C, u00055, "system__file_control_blockS");
21742 pragma Export (C, u00056, "ada__finalization__list_controllerB");
21743 pragma Export (C, u00057, "ada__finalization__list_controllerS");
21745 -- BEGIN ELABORATION ORDER
21748 -- gnat.heap_sort_a (spec)
21749 -- gnat.heap_sort_a (body)
21750 -- gnat.htable (spec)
21751 -- gnat.htable (body)
21752 -- interfaces (spec)
21754 -- system.machine_code (spec)
21755 -- system.parameters (spec)
21756 -- system.parameters (body)
21757 -- interfaces.c_streams (spec)
21758 -- interfaces.c_streams (body)
21759 -- system.standard_library (spec)
21760 -- ada.exceptions (spec)
21761 -- system.exception_table (spec)
21762 -- system.exception_table (body)
21763 -- ada.io_exceptions (spec)
21764 -- system.exceptions (spec)
21765 -- system.storage_elements (spec)
21766 -- system.storage_elements (body)
21767 -- system.machine_state_operations (spec)
21768 -- system.machine_state_operations (body)
21769 -- system.secondary_stack (spec)
21770 -- system.stack_checking (spec)
21771 -- system.soft_links (spec)
21772 -- system.soft_links (body)
21773 -- system.stack_checking (body)
21774 -- system.secondary_stack (body)
21775 -- system.standard_library (body)
21776 -- system.string_ops (spec)
21777 -- system.string_ops (body)
21780 -- ada.streams (spec)
21781 -- system.finalization_root (spec)
21782 -- system.finalization_root (body)
21783 -- system.string_ops_concat_3 (spec)
21784 -- system.string_ops_concat_3 (body)
21785 -- system.traceback (spec)
21786 -- system.traceback (body)
21787 -- ada.exceptions (body)
21788 -- system.unsigned_types (spec)
21789 -- system.stream_attributes (spec)
21790 -- system.stream_attributes (body)
21791 -- system.finalization_implementation (spec)
21792 -- system.finalization_implementation (body)
21793 -- ada.finalization (spec)
21794 -- ada.finalization (body)
21795 -- ada.finalization.list_controller (spec)
21796 -- ada.finalization.list_controller (body)
21797 -- system.file_control_block (spec)
21798 -- system.file_io (spec)
21799 -- system.file_io (body)
21800 -- ada.text_io (spec)
21801 -- ada.text_io (body)
21803 -- END ELABORATION ORDER
21807 -- The following source file name pragmas allow the generated file
21808 -- names to be unique for different main programs. They are needed
21809 -- since the package name will always be Ada_Main.
21811 pragma Source_File_Name (ada_main, Spec_File_Name => "b~hello.ads");
21812 pragma Source_File_Name (ada_main, Body_File_Name => "b~hello.adb");
21814 -- Generated package body for Ada_Main starts here
21816 package body ada_main is
21818 -- The actual finalization is performed by calling the
21819 -- library routine in System.Standard_Library.Adafinal
21821 procedure Do_Finalize;
21822 pragma Import (C, Do_Finalize, "system__standard_library__adafinal");
21829 procedure adainit is
21831 -- These booleans are set to True once the associated unit has
21832 -- been elaborated. It is also used to avoid elaborating the
21833 -- same unit twice.
21836 pragma Import (Ada, E040, "interfaces__c_streams_E");
21839 pragma Import (Ada, E008, "ada__exceptions_E");
21842 pragma Import (Ada, E014, "system__exception_table_E");
21845 pragma Import (Ada, E053, "ada__io_exceptions_E");
21848 pragma Import (Ada, E017, "system__exceptions_E");
21851 pragma Import (Ada, E024, "system__secondary_stack_E");
21854 pragma Import (Ada, E030, "system__stack_checking_E");
21857 pragma Import (Ada, E028, "system__soft_links_E");
21860 pragma Import (Ada, E035, "ada__tags_E");
21863 pragma Import (Ada, E033, "ada__streams_E");
21866 pragma Import (Ada, E046, "system__finalization_root_E");
21869 pragma Import (Ada, E048, "system__finalization_implementation_E");
21872 pragma Import (Ada, E044, "ada__finalization_E");
21875 pragma Import (Ada, E057, "ada__finalization__list_controller_E");
21878 pragma Import (Ada, E055, "system__file_control_block_E");
21881 pragma Import (Ada, E042, "system__file_io_E");
21884 pragma Import (Ada, E006, "ada__text_io_E");
21886 -- Set_Globals is a library routine that stores away the
21887 -- value of the indicated set of global values in global
21888 -- variables within the library.
21890 procedure Set_Globals
21891 (Main_Priority : Integer;
21892 Time_Slice_Value : Integer;
21893 WC_Encoding : Character;
21894 Locking_Policy : Character;
21895 Queuing_Policy : Character;
21896 Task_Dispatching_Policy : Character;
21897 Adafinal : System.Address;
21898 Unreserve_All_Interrupts : Integer;
21899 Exception_Tracebacks : Integer);
21900 @findex __gnat_set_globals
21901 pragma Import (C, Set_Globals, "__gnat_set_globals");
21903 -- SDP_Table_Build is a library routine used to build the
21904 -- exception tables. See unit Ada.Exceptions in files
21905 -- a-except.ads/adb for full details of how zero cost
21906 -- exception handling works. This procedure, the call to
21907 -- it, and the two following tables are all omitted if the
21908 -- build is in longjmp/setjmp exception mode.
21910 @findex SDP_Table_Build
21911 @findex Zero Cost Exceptions
21912 procedure SDP_Table_Build
21913 (SDP_Addresses : System.Address;
21914 SDP_Count : Natural;
21915 Elab_Addresses : System.Address;
21916 Elab_Addr_Count : Natural);
21917 pragma Import (C, SDP_Table_Build, "__gnat_SDP_Table_Build");
21919 -- Table of Unit_Exception_Table addresses. Used for zero
21920 -- cost exception handling to build the top level table.
21922 ST : aliased constant array (1 .. 23) of System.Address := (
21924 Ada.Text_Io'UET_Address,
21925 Ada.Exceptions'UET_Address,
21926 Gnat.Heap_Sort_A'UET_Address,
21927 System.Exception_Table'UET_Address,
21928 System.Machine_State_Operations'UET_Address,
21929 System.Secondary_Stack'UET_Address,
21930 System.Parameters'UET_Address,
21931 System.Soft_Links'UET_Address,
21932 System.Stack_Checking'UET_Address,
21933 System.Traceback'UET_Address,
21934 Ada.Streams'UET_Address,
21935 Ada.Tags'UET_Address,
21936 System.String_Ops'UET_Address,
21937 Interfaces.C_Streams'UET_Address,
21938 System.File_Io'UET_Address,
21939 Ada.Finalization'UET_Address,
21940 System.Finalization_Root'UET_Address,
21941 System.Finalization_Implementation'UET_Address,
21942 System.String_Ops_Concat_3'UET_Address,
21943 System.Stream_Attributes'UET_Address,
21944 System.File_Control_Block'UET_Address,
21945 Ada.Finalization.List_Controller'UET_Address);
21947 -- Table of addresses of elaboration routines. Used for
21948 -- zero cost exception handling to make sure these
21949 -- addresses are included in the top level procedure
21952 EA : aliased constant array (1 .. 23) of System.Address := (
21953 adainit'Code_Address,
21954 Do_Finalize'Code_Address,
21955 Ada.Exceptions'Elab_Spec'Address,
21956 System.Exceptions'Elab_Spec'Address,
21957 Interfaces.C_Streams'Elab_Spec'Address,
21958 System.Exception_Table'Elab_Body'Address,
21959 Ada.Io_Exceptions'Elab_Spec'Address,
21960 System.Stack_Checking'Elab_Spec'Address,
21961 System.Soft_Links'Elab_Body'Address,
21962 System.Secondary_Stack'Elab_Body'Address,
21963 Ada.Tags'Elab_Spec'Address,
21964 Ada.Tags'Elab_Body'Address,
21965 Ada.Streams'Elab_Spec'Address,
21966 System.Finalization_Root'Elab_Spec'Address,
21967 Ada.Exceptions'Elab_Body'Address,
21968 System.Finalization_Implementation'Elab_Spec'Address,
21969 System.Finalization_Implementation'Elab_Body'Address,
21970 Ada.Finalization'Elab_Spec'Address,
21971 Ada.Finalization.List_Controller'Elab_Spec'Address,
21972 System.File_Control_Block'Elab_Spec'Address,
21973 System.File_Io'Elab_Body'Address,
21974 Ada.Text_Io'Elab_Spec'Address,
21975 Ada.Text_Io'Elab_Body'Address);
21977 -- Start of processing for adainit
21981 -- Call SDP_Table_Build to build the top level procedure
21982 -- table for zero cost exception handling (omitted in
21983 -- longjmp/setjmp mode).
21985 SDP_Table_Build (ST'Address, 23, EA'Address, 23);
21987 -- Call Set_Globals to record various information for
21988 -- this partition. The values are derived by the binder
21989 -- from information stored in the ali files by the compiler.
21991 @findex __gnat_set_globals
21993 (Main_Priority => -1,
21994 -- Priority of main program, -1 if no pragma Priority used
21996 Time_Slice_Value => -1,
21997 -- Time slice from Time_Slice pragma, -1 if none used
21999 WC_Encoding => 'b',
22000 -- Wide_Character encoding used, default is brackets
22002 Locking_Policy => ' ',
22003 -- Locking_Policy used, default of space means not
22004 -- specified, otherwise it is the first character of
22005 -- the policy name.
22007 Queuing_Policy => ' ',
22008 -- Queuing_Policy used, default of space means not
22009 -- specified, otherwise it is the first character of
22010 -- the policy name.
22012 Task_Dispatching_Policy => ' ',
22013 -- Task_Dispatching_Policy used, default of space means
22014 -- not specified, otherwise first character of the
22017 Adafinal => System.Null_Address,
22018 -- Address of Adafinal routine, not used anymore
22020 Unreserve_All_Interrupts => 0,
22021 -- Set true if pragma Unreserve_All_Interrupts was used
22023 Exception_Tracebacks => 0);
22024 -- Indicates if exception tracebacks are enabled
22026 Elab_Final_Code := 1;
22028 -- Now we have the elaboration calls for all units in the partition.
22029 -- The Elab_Spec and Elab_Body attributes generate references to the
22030 -- implicit elaboration procedures generated by the compiler for
22031 -- each unit that requires elaboration.
22034 Interfaces.C_Streams'Elab_Spec;
22038 Ada.Exceptions'Elab_Spec;
22041 System.Exception_Table'Elab_Body;
22045 Ada.Io_Exceptions'Elab_Spec;
22049 System.Exceptions'Elab_Spec;
22053 System.Stack_Checking'Elab_Spec;
22056 System.Soft_Links'Elab_Body;
22061 System.Secondary_Stack'Elab_Body;
22065 Ada.Tags'Elab_Spec;
22068 Ada.Tags'Elab_Body;
22072 Ada.Streams'Elab_Spec;
22076 System.Finalization_Root'Elab_Spec;
22080 Ada.Exceptions'Elab_Body;
22084 System.Finalization_Implementation'Elab_Spec;
22087 System.Finalization_Implementation'Elab_Body;
22091 Ada.Finalization'Elab_Spec;
22095 Ada.Finalization.List_Controller'Elab_Spec;
22099 System.File_Control_Block'Elab_Spec;
22103 System.File_Io'Elab_Body;
22107 Ada.Text_Io'Elab_Spec;
22110 Ada.Text_Io'Elab_Body;
22114 Elab_Final_Code := 0;
22122 procedure adafinal is
22131 -- main is actually a function, as in the ANSI C standard,
22132 -- defined to return the exit status. The three parameters
22133 -- are the argument count, argument values and environment
22136 @findex Main Program
22139 argv : System.Address;
22140 envp : System.Address)
22143 -- The initialize routine performs low level system
22144 -- initialization using a standard library routine which
22145 -- sets up signal handling and performs any other
22146 -- required setup. The routine can be found in file
22149 @findex __gnat_initialize
22150 procedure initialize;
22151 pragma Import (C, initialize, "__gnat_initialize");
22153 -- The finalize routine performs low level system
22154 -- finalization using a standard library routine. The
22155 -- routine is found in file a-final.c and in the standard
22156 -- distribution is a dummy routine that does nothing, so
22157 -- really this is a hook for special user finalization.
22159 @findex __gnat_finalize
22160 procedure finalize;
22161 pragma Import (C, finalize, "__gnat_finalize");
22163 -- We get to the main program of the partition by using
22164 -- pragma Import because if we try to with the unit and
22165 -- call it Ada style, then not only do we waste time
22166 -- recompiling it, but also, we don't really know the right
22167 -- switches (e.g.@: identifier character set) to be used
22170 procedure Ada_Main_Program;
22171 pragma Import (Ada, Ada_Main_Program, "_ada_hello");
22173 -- Start of processing for main
22176 -- Save global variables
22182 -- Call low level system initialization
22186 -- Call our generated Ada initialization routine
22190 -- This is the point at which we want the debugger to get
22195 -- Now we call the main program of the partition
22199 -- Perform Ada finalization
22203 -- Perform low level system finalization
22207 -- Return the proper exit status
22208 return (gnat_exit_status);
22211 -- This section is entirely comments, so it has no effect on the
22212 -- compilation of the Ada_Main package. It provides the list of
22213 -- object files and linker options, as well as some standard
22214 -- libraries needed for the link. The gnatlink utility parses
22215 -- this b~hello.adb file to read these comment lines to generate
22216 -- the appropriate command line arguments for the call to the
22217 -- system linker. The BEGIN/END lines are used for sentinels for
22218 -- this parsing operation.
22220 -- The exact file names will of course depend on the environment,
22221 -- host/target and location of files on the host system.
22223 @findex Object file list
22224 -- BEGIN Object file/option list
22227 -- -L/usr/local/gnat/lib/gcc-lib/i686-pc-linux-gnu/2.8.1/adalib/
22228 -- /usr/local/gnat/lib/gcc-lib/i686-pc-linux-gnu/2.8.1/adalib/libgnat.a
22229 -- END Object file/option list
22235 The Ada code in the above example is exactly what is generated by the
22236 binder. We have added comments to more clearly indicate the function
22237 of each part of the generated @code{Ada_Main} package.
22239 The code is standard Ada in all respects, and can be processed by any
22240 tools that handle Ada. In particular, it is possible to use the debugger
22241 in Ada mode to debug the generated @code{Ada_Main} package. For example,
22242 suppose that for reasons that you do not understand, your program is crashing
22243 during elaboration of the body of @code{Ada.Text_IO}. To locate this bug,
22244 you can place a breakpoint on the call:
22246 @smallexample @c ada
22247 Ada.Text_Io'Elab_Body;
22251 and trace the elaboration routine for this package to find out where
22252 the problem might be (more usually of course you would be debugging
22253 elaboration code in your own application).
22255 @node Elaboration Order Handling in GNAT
22256 @appendix Elaboration Order Handling in GNAT
22257 @cindex Order of elaboration
22258 @cindex Elaboration control
22261 * Elaboration Code::
22262 * Checking the Elaboration Order::
22263 * Controlling the Elaboration Order::
22264 * Controlling Elaboration in GNAT - Internal Calls::
22265 * Controlling Elaboration in GNAT - External Calls::
22266 * Default Behavior in GNAT - Ensuring Safety::
22267 * Treatment of Pragma Elaborate::
22268 * Elaboration Issues for Library Tasks::
22269 * Mixing Elaboration Models::
22270 * What to Do If the Default Elaboration Behavior Fails::
22271 * Elaboration for Access-to-Subprogram Values::
22272 * Summary of Procedures for Elaboration Control::
22273 * Other Elaboration Order Considerations::
22277 This chapter describes the handling of elaboration code in Ada and
22278 in GNAT, and discusses how the order of elaboration of program units can
22279 be controlled in GNAT, either automatically or with explicit programming
22282 @node Elaboration Code
22283 @section Elaboration Code
22286 Ada provides rather general mechanisms for executing code at elaboration
22287 time, that is to say before the main program starts executing. Such code arises
22291 @item Initializers for variables.
22292 Variables declared at the library level, in package specs or bodies, can
22293 require initialization that is performed at elaboration time, as in:
22294 @smallexample @c ada
22296 Sqrt_Half : Float := Sqrt (0.5);
22300 @item Package initialization code
22301 Code in a @code{BEGIN-END} section at the outer level of a package body is
22302 executed as part of the package body elaboration code.
22304 @item Library level task allocators
22305 Tasks that are declared using task allocators at the library level
22306 start executing immediately and hence can execute at elaboration time.
22310 Subprogram calls are possible in any of these contexts, which means that
22311 any arbitrary part of the program may be executed as part of the elaboration
22312 code. It is even possible to write a program which does all its work at
22313 elaboration time, with a null main program, although stylistically this
22314 would usually be considered an inappropriate way to structure
22317 An important concern arises in the context of elaboration code:
22318 we have to be sure that it is executed in an appropriate order. What we
22319 have is a series of elaboration code sections, potentially one section
22320 for each unit in the program. It is important that these execute
22321 in the correct order. Correctness here means that, taking the above
22322 example of the declaration of @code{Sqrt_Half},
22323 if some other piece of
22324 elaboration code references @code{Sqrt_Half},
22325 then it must run after the
22326 section of elaboration code that contains the declaration of
22329 There would never be any order of elaboration problem if we made a rule
22330 that whenever you @code{with} a unit, you must elaborate both the spec and body
22331 of that unit before elaborating the unit doing the @code{with}'ing:
22333 @smallexample @c ada
22337 package Unit_2 is @dots{}
22343 would require that both the body and spec of @code{Unit_1} be elaborated
22344 before the spec of @code{Unit_2}. However, a rule like that would be far too
22345 restrictive. In particular, it would make it impossible to have routines
22346 in separate packages that were mutually recursive.
22348 You might think that a clever enough compiler could look at the actual
22349 elaboration code and determine an appropriate correct order of elaboration,
22350 but in the general case, this is not possible. Consider the following
22353 In the body of @code{Unit_1}, we have a procedure @code{Func_1}
22355 the variable @code{Sqrt_1}, which is declared in the elaboration code
22356 of the body of @code{Unit_1}:
22358 @smallexample @c ada
22360 Sqrt_1 : Float := Sqrt (0.1);
22365 The elaboration code of the body of @code{Unit_1} also contains:
22367 @smallexample @c ada
22370 if expression_1 = 1 then
22371 Q := Unit_2.Func_2;
22378 @code{Unit_2} is exactly parallel,
22379 it has a procedure @code{Func_2} that references
22380 the variable @code{Sqrt_2}, which is declared in the elaboration code of
22381 the body @code{Unit_2}:
22383 @smallexample @c ada
22385 Sqrt_2 : Float := Sqrt (0.1);
22390 The elaboration code of the body of @code{Unit_2} also contains:
22392 @smallexample @c ada
22395 if expression_2 = 2 then
22396 Q := Unit_1.Func_1;
22403 Now the question is, which of the following orders of elaboration is
22428 If you carefully analyze the flow here, you will see that you cannot tell
22429 at compile time the answer to this question.
22430 If @code{expression_1} is not equal to 1,
22431 and @code{expression_2} is not equal to 2,
22432 then either order is acceptable, because neither of the function calls is
22433 executed. If both tests evaluate to true, then neither order is acceptable
22434 and in fact there is no correct order.
22436 If one of the two expressions is true, and the other is false, then one
22437 of the above orders is correct, and the other is incorrect. For example,
22438 if @code{expression_1} /= 1 and @code{expression_2} = 2,
22439 then the call to @code{Func_1}
22440 will occur, but not the call to @code{Func_2.}
22441 This means that it is essential
22442 to elaborate the body of @code{Unit_1} before
22443 the body of @code{Unit_2}, so the first
22444 order of elaboration is correct and the second is wrong.
22446 By making @code{expression_1} and @code{expression_2}
22447 depend on input data, or perhaps
22448 the time of day, we can make it impossible for the compiler or binder
22449 to figure out which of these expressions will be true, and hence it
22450 is impossible to guarantee a safe order of elaboration at run time.
22452 @node Checking the Elaboration Order
22453 @section Checking the Elaboration Order
22456 In some languages that involve the same kind of elaboration problems,
22457 e.g.@: Java and C++, the programmer is expected to worry about these
22458 ordering problems himself, and it is common to
22459 write a program in which an incorrect elaboration order gives
22460 surprising results, because it references variables before they
22462 Ada is designed to be a safe language, and a programmer-beware approach is
22463 clearly not sufficient. Consequently, the language provides three lines
22467 @item Standard rules
22468 Some standard rules restrict the possible choice of elaboration
22469 order. In particular, if you @code{with} a unit, then its spec is always
22470 elaborated before the unit doing the @code{with}. Similarly, a parent
22471 spec is always elaborated before the child spec, and finally
22472 a spec is always elaborated before its corresponding body.
22474 @item Dynamic elaboration checks
22475 @cindex Elaboration checks
22476 @cindex Checks, elaboration
22477 Dynamic checks are made at run time, so that if some entity is accessed
22478 before it is elaborated (typically by means of a subprogram call)
22479 then the exception (@code{Program_Error}) is raised.
22481 @item Elaboration control
22482 Facilities are provided for the programmer to specify the desired order
22486 Let's look at these facilities in more detail. First, the rules for
22487 dynamic checking. One possible rule would be simply to say that the
22488 exception is raised if you access a variable which has not yet been
22489 elaborated. The trouble with this approach is that it could require
22490 expensive checks on every variable reference. Instead Ada has two
22491 rules which are a little more restrictive, but easier to check, and
22495 @item Restrictions on calls
22496 A subprogram can only be called at elaboration time if its body
22497 has been elaborated. The rules for elaboration given above guarantee
22498 that the spec of the subprogram has been elaborated before the
22499 call, but not the body. If this rule is violated, then the
22500 exception @code{Program_Error} is raised.
22502 @item Restrictions on instantiations
22503 A generic unit can only be instantiated if the body of the generic
22504 unit has been elaborated. Again, the rules for elaboration given above
22505 guarantee that the spec of the generic unit has been elaborated
22506 before the instantiation, but not the body. If this rule is
22507 violated, then the exception @code{Program_Error} is raised.
22511 The idea is that if the body has been elaborated, then any variables
22512 it references must have been elaborated; by checking for the body being
22513 elaborated we guarantee that none of its references causes any
22514 trouble. As we noted above, this is a little too restrictive, because a
22515 subprogram that has no non-local references in its body may in fact be safe
22516 to call. However, it really would be unsafe to rely on this, because
22517 it would mean that the caller was aware of details of the implementation
22518 in the body. This goes against the basic tenets of Ada.
22520 A plausible implementation can be described as follows.
22521 A Boolean variable is associated with each subprogram
22522 and each generic unit. This variable is initialized to False, and is set to
22523 True at the point body is elaborated. Every call or instantiation checks the
22524 variable, and raises @code{Program_Error} if the variable is False.
22526 Note that one might think that it would be good enough to have one Boolean
22527 variable for each package, but that would not deal with cases of trying
22528 to call a body in the same package as the call
22529 that has not been elaborated yet.
22530 Of course a compiler may be able to do enough analysis to optimize away
22531 some of the Boolean variables as unnecessary, and @code{GNAT} indeed
22532 does such optimizations, but still the easiest conceptual model is to
22533 think of there being one variable per subprogram.
22535 @node Controlling the Elaboration Order
22536 @section Controlling the Elaboration Order
22539 In the previous section we discussed the rules in Ada which ensure
22540 that @code{Program_Error} is raised if an incorrect elaboration order is
22541 chosen. This prevents erroneous executions, but we need mechanisms to
22542 specify a correct execution and avoid the exception altogether.
22543 To achieve this, Ada provides a number of features for controlling
22544 the order of elaboration. We discuss these features in this section.
22546 First, there are several ways of indicating to the compiler that a given
22547 unit has no elaboration problems:
22550 @item packages that do not require a body
22551 A library package that does not require a body does not permit
22552 a body (this rule was introduced in Ada 95).
22553 Thus if we have a such a package, as in:
22555 @smallexample @c ada
22558 package Definitions is
22560 type m is new integer;
22562 type a is array (1 .. 10) of m;
22563 type b is array (1 .. 20) of m;
22571 A package that @code{with}'s @code{Definitions} may safely instantiate
22572 @code{Definitions.Subp} because the compiler can determine that there
22573 definitely is no package body to worry about in this case
22576 @cindex pragma Pure
22578 Places sufficient restrictions on a unit to guarantee that
22579 no call to any subprogram in the unit can result in an
22580 elaboration problem. This means that the compiler does not need
22581 to worry about the point of elaboration of such units, and in
22582 particular, does not need to check any calls to any subprograms
22585 @item pragma Preelaborate
22586 @findex Preelaborate
22587 @cindex pragma Preelaborate
22588 This pragma places slightly less stringent restrictions on a unit than
22590 but these restrictions are still sufficient to ensure that there
22591 are no elaboration problems with any calls to the unit.
22593 @item pragma Elaborate_Body
22594 @findex Elaborate_Body
22595 @cindex pragma Elaborate_Body
22596 This pragma requires that the body of a unit be elaborated immediately
22597 after its spec. Suppose a unit @code{A} has such a pragma,
22598 and unit @code{B} does
22599 a @code{with} of unit @code{A}. Recall that the standard rules require
22600 the spec of unit @code{A}
22601 to be elaborated before the @code{with}'ing unit; given the pragma in
22602 @code{A}, we also know that the body of @code{A}
22603 will be elaborated before @code{B}, so
22604 that calls to @code{A} are safe and do not need a check.
22609 unlike pragma @code{Pure} and pragma @code{Preelaborate},
22611 @code{Elaborate_Body} does not guarantee that the program is
22612 free of elaboration problems, because it may not be possible
22613 to satisfy the requested elaboration order.
22614 Let's go back to the example with @code{Unit_1} and @code{Unit_2}.
22616 marks @code{Unit_1} as @code{Elaborate_Body},
22617 and not @code{Unit_2,} then the order of
22618 elaboration will be:
22630 Now that means that the call to @code{Func_1} in @code{Unit_2}
22631 need not be checked,
22632 it must be safe. But the call to @code{Func_2} in
22633 @code{Unit_1} may still fail if
22634 @code{Expression_1} is equal to 1,
22635 and the programmer must still take
22636 responsibility for this not being the case.
22638 If all units carry a pragma @code{Elaborate_Body}, then all problems are
22639 eliminated, except for calls entirely within a body, which are
22640 in any case fully under programmer control. However, using the pragma
22641 everywhere is not always possible.
22642 In particular, for our @code{Unit_1}/@code{Unit_2} example, if
22643 we marked both of them as having pragma @code{Elaborate_Body}, then
22644 clearly there would be no possible elaboration order.
22646 The above pragmas allow a server to guarantee safe use by clients, and
22647 clearly this is the preferable approach. Consequently a good rule
22648 is to mark units as @code{Pure} or @code{Preelaborate} if possible,
22649 and if this is not possible,
22650 mark them as @code{Elaborate_Body} if possible.
22651 As we have seen, there are situations where neither of these
22652 three pragmas can be used.
22653 So we also provide methods for clients to control the
22654 order of elaboration of the servers on which they depend:
22657 @item pragma Elaborate (unit)
22659 @cindex pragma Elaborate
22660 This pragma is placed in the context clause, after a @code{with} clause,
22661 and it requires that the body of the named unit be elaborated before
22662 the unit in which the pragma occurs. The idea is to use this pragma
22663 if the current unit calls at elaboration time, directly or indirectly,
22664 some subprogram in the named unit.
22666 @item pragma Elaborate_All (unit)
22667 @findex Elaborate_All
22668 @cindex pragma Elaborate_All
22669 This is a stronger version of the Elaborate pragma. Consider the
22673 Unit A @code{with}'s unit B and calls B.Func in elab code
22674 Unit B @code{with}'s unit C, and B.Func calls C.Func
22678 Now if we put a pragma @code{Elaborate (B)}
22679 in unit @code{A}, this ensures that the
22680 body of @code{B} is elaborated before the call, but not the
22681 body of @code{C}, so
22682 the call to @code{C.Func} could still cause @code{Program_Error} to
22685 The effect of a pragma @code{Elaborate_All} is stronger, it requires
22686 not only that the body of the named unit be elaborated before the
22687 unit doing the @code{with}, but also the bodies of all units that the
22688 named unit uses, following @code{with} links transitively. For example,
22689 if we put a pragma @code{Elaborate_All (B)} in unit @code{A},
22691 not only that the body of @code{B} be elaborated before @code{A},
22693 body of @code{C}, because @code{B} @code{with}'s @code{C}.
22697 We are now in a position to give a usage rule in Ada for avoiding
22698 elaboration problems, at least if dynamic dispatching and access to
22699 subprogram values are not used. We will handle these cases separately
22702 The rule is simple. If a unit has elaboration code that can directly or
22703 indirectly make a call to a subprogram in a @code{with}'ed unit, or instantiate
22704 a generic package in a @code{with}'ed unit,
22705 then if the @code{with}'ed unit does not have
22706 pragma @code{Pure} or @code{Preelaborate}, then the client should have
22707 a pragma @code{Elaborate_All}
22708 for the @code{with}'ed unit. By following this rule a client is
22709 assured that calls can be made without risk of an exception.
22711 For generic subprogram instantiations, the rule can be relaxed to
22712 require only a pragma @code{Elaborate} since elaborating the body
22713 of a subprogram cannot cause any transitive elaboration (we are
22714 not calling the subprogram in this case, just elaborating its
22717 If this rule is not followed, then a program may be in one of four
22721 @item No order exists
22722 No order of elaboration exists which follows the rules, taking into
22723 account any @code{Elaborate}, @code{Elaborate_All},
22724 or @code{Elaborate_Body} pragmas. In
22725 this case, an Ada compiler must diagnose the situation at bind
22726 time, and refuse to build an executable program.
22728 @item One or more orders exist, all incorrect
22729 One or more acceptable elaboration orders exist, and all of them
22730 generate an elaboration order problem. In this case, the binder
22731 can build an executable program, but @code{Program_Error} will be raised
22732 when the program is run.
22734 @item Several orders exist, some right, some incorrect
22735 One or more acceptable elaboration orders exists, and some of them
22736 work, and some do not. The programmer has not controlled
22737 the order of elaboration, so the binder may or may not pick one of
22738 the correct orders, and the program may or may not raise an
22739 exception when it is run. This is the worst case, because it means
22740 that the program may fail when moved to another compiler, or even
22741 another version of the same compiler.
22743 @item One or more orders exists, all correct
22744 One ore more acceptable elaboration orders exist, and all of them
22745 work. In this case the program runs successfully. This state of
22746 affairs can be guaranteed by following the rule we gave above, but
22747 may be true even if the rule is not followed.
22751 Note that one additional advantage of following our rules on the use
22752 of @code{Elaborate} and @code{Elaborate_All}
22753 is that the program continues to stay in the ideal (all orders OK) state
22754 even if maintenance
22755 changes some bodies of some units. Conversely, if a program that does
22756 not follow this rule happens to be safe at some point, this state of affairs
22757 may deteriorate silently as a result of maintenance changes.
22759 You may have noticed that the above discussion did not mention
22760 the use of @code{Elaborate_Body}. This was a deliberate omission. If you
22761 @code{with} an @code{Elaborate_Body} unit, it still may be the case that
22762 code in the body makes calls to some other unit, so it is still necessary
22763 to use @code{Elaborate_All} on such units.
22765 @node Controlling Elaboration in GNAT - Internal Calls
22766 @section Controlling Elaboration in GNAT - Internal Calls
22769 In the case of internal calls, i.e., calls within a single package, the
22770 programmer has full control over the order of elaboration, and it is up
22771 to the programmer to elaborate declarations in an appropriate order. For
22774 @smallexample @c ada
22777 function One return Float;
22781 function One return Float is
22790 will obviously raise @code{Program_Error} at run time, because function
22791 One will be called before its body is elaborated. In this case GNAT will
22792 generate a warning that the call will raise @code{Program_Error}:
22798 2. function One return Float;
22800 4. Q : Float := One;
22802 >>> warning: cannot call "One" before body is elaborated
22803 >>> warning: Program_Error will be raised at run time
22806 6. function One return Float is
22819 Note that in this particular case, it is likely that the call is safe, because
22820 the function @code{One} does not access any global variables.
22821 Nevertheless in Ada, we do not want the validity of the check to depend on
22822 the contents of the body (think about the separate compilation case), so this
22823 is still wrong, as we discussed in the previous sections.
22825 The error is easily corrected by rearranging the declarations so that the
22826 body of @code{One} appears before the declaration containing the call
22827 (note that in Ada 95 and Ada 2005,
22828 declarations can appear in any order, so there is no restriction that
22829 would prevent this reordering, and if we write:
22831 @smallexample @c ada
22834 function One return Float;
22836 function One return Float is
22847 then all is well, no warning is generated, and no
22848 @code{Program_Error} exception
22850 Things are more complicated when a chain of subprograms is executed:
22852 @smallexample @c ada
22855 function A return Integer;
22856 function B return Integer;
22857 function C return Integer;
22859 function B return Integer is begin return A; end;
22860 function C return Integer is begin return B; end;
22864 function A return Integer is begin return 1; end;
22870 Now the call to @code{C}
22871 at elaboration time in the declaration of @code{X} is correct, because
22872 the body of @code{C} is already elaborated,
22873 and the call to @code{B} within the body of
22874 @code{C} is correct, but the call
22875 to @code{A} within the body of @code{B} is incorrect, because the body
22876 of @code{A} has not been elaborated, so @code{Program_Error}
22877 will be raised on the call to @code{A}.
22878 In this case GNAT will generate a
22879 warning that @code{Program_Error} may be
22880 raised at the point of the call. Let's look at the warning:
22886 2. function A return Integer;
22887 3. function B return Integer;
22888 4. function C return Integer;
22890 6. function B return Integer is begin return A; end;
22892 >>> warning: call to "A" before body is elaborated may
22893 raise Program_Error
22894 >>> warning: "B" called at line 7
22895 >>> warning: "C" called at line 9
22897 7. function C return Integer is begin return B; end;
22899 9. X : Integer := C;
22901 11. function A return Integer is begin return 1; end;
22911 Note that the message here says ``may raise'', instead of the direct case,
22912 where the message says ``will be raised''. That's because whether
22914 actually called depends in general on run-time flow of control.
22915 For example, if the body of @code{B} said
22917 @smallexample @c ada
22920 function B return Integer is
22922 if some-condition-depending-on-input-data then
22933 then we could not know until run time whether the incorrect call to A would
22934 actually occur, so @code{Program_Error} might
22935 or might not be raised. It is possible for a compiler to
22936 do a better job of analyzing bodies, to
22937 determine whether or not @code{Program_Error}
22938 might be raised, but it certainly
22939 couldn't do a perfect job (that would require solving the halting problem
22940 and is provably impossible), and because this is a warning anyway, it does
22941 not seem worth the effort to do the analysis. Cases in which it
22942 would be relevant are rare.
22944 In practice, warnings of either of the forms given
22945 above will usually correspond to
22946 real errors, and should be examined carefully and eliminated.
22947 In the rare case where a warning is bogus, it can be suppressed by any of
22948 the following methods:
22952 Compile with the @option{-gnatws} switch set
22955 Suppress @code{Elaboration_Check} for the called subprogram
22958 Use pragma @code{Warnings_Off} to turn warnings off for the call
22962 For the internal elaboration check case,
22963 GNAT by default generates the
22964 necessary run-time checks to ensure
22965 that @code{Program_Error} is raised if any
22966 call fails an elaboration check. Of course this can only happen if a
22967 warning has been issued as described above. The use of pragma
22968 @code{Suppress (Elaboration_Check)} may (but is not guaranteed to) suppress
22969 some of these checks, meaning that it may be possible (but is not
22970 guaranteed) for a program to be able to call a subprogram whose body
22971 is not yet elaborated, without raising a @code{Program_Error} exception.
22973 @node Controlling Elaboration in GNAT - External Calls
22974 @section Controlling Elaboration in GNAT - External Calls
22977 The previous section discussed the case in which the execution of a
22978 particular thread of elaboration code occurred entirely within a
22979 single unit. This is the easy case to handle, because a programmer
22980 has direct and total control over the order of elaboration, and
22981 furthermore, checks need only be generated in cases which are rare
22982 and which the compiler can easily detect.
22983 The situation is more complex when separate compilation is taken into account.
22984 Consider the following:
22986 @smallexample @c ada
22990 function Sqrt (Arg : Float) return Float;
22993 package body Math is
22994 function Sqrt (Arg : Float) return Float is
23003 X : Float := Math.Sqrt (0.5);
23016 where @code{Main} is the main program. When this program is executed, the
23017 elaboration code must first be executed, and one of the jobs of the
23018 binder is to determine the order in which the units of a program are
23019 to be elaborated. In this case we have four units: the spec and body
23021 the spec of @code{Stuff} and the body of @code{Main}).
23022 In what order should the four separate sections of elaboration code
23025 There are some restrictions in the order of elaboration that the binder
23026 can choose. In particular, if unit U has a @code{with}
23027 for a package @code{X}, then you
23028 are assured that the spec of @code{X}
23029 is elaborated before U , but you are
23030 not assured that the body of @code{X}
23031 is elaborated before U.
23032 This means that in the above case, the binder is allowed to choose the
23043 but that's not good, because now the call to @code{Math.Sqrt}
23044 that happens during
23045 the elaboration of the @code{Stuff}
23046 spec happens before the body of @code{Math.Sqrt} is
23047 elaborated, and hence causes @code{Program_Error} exception to be raised.
23048 At first glance, one might say that the binder is misbehaving, because
23049 obviously you want to elaborate the body of something you @code{with}
23051 that is not a general rule that can be followed in all cases. Consider
23053 @smallexample @c ada
23056 package X is @dots{}
23058 package Y is @dots{}
23061 package body Y is @dots{}
23064 package body X is @dots{}
23070 This is a common arrangement, and, apart from the order of elaboration
23071 problems that might arise in connection with elaboration code, this works fine.
23072 A rule that says that you must first elaborate the body of anything you
23073 @code{with} cannot work in this case:
23074 the body of @code{X} @code{with}'s @code{Y},
23075 which means you would have to
23076 elaborate the body of @code{Y} first, but that @code{with}'s @code{X},
23078 you have to elaborate the body of @code{X} first, but @dots{} and we have a
23079 loop that cannot be broken.
23081 It is true that the binder can in many cases guess an order of elaboration
23082 that is unlikely to cause a @code{Program_Error}
23083 exception to be raised, and it tries to do so (in the
23084 above example of @code{Math/Stuff/Spec}, the GNAT binder will
23086 elaborate the body of @code{Math} right after its spec, so all will be well).
23088 However, a program that blindly relies on the binder to be helpful can
23089 get into trouble, as we discussed in the previous sections, so
23091 provides a number of facilities for assisting the programmer in
23092 developing programs that are robust with respect to elaboration order.
23094 @node Default Behavior in GNAT - Ensuring Safety
23095 @section Default Behavior in GNAT - Ensuring Safety
23098 The default behavior in GNAT ensures elaboration safety. In its
23099 default mode GNAT implements the
23100 rule we previously described as the right approach. Let's restate it:
23104 @emph{If a unit has elaboration code that can directly or indirectly make a
23105 call to a subprogram in a @code{with}'ed unit, or instantiate a generic
23106 package in a @code{with}'ed unit, then if the @code{with}'ed unit
23107 does not have pragma @code{Pure} or
23108 @code{Preelaborate}, then the client should have an
23109 @code{Elaborate_All} pragma for the @code{with}'ed unit.}
23111 @emph{In the case of instantiating a generic subprogram, it is always
23112 sufficient to have only an @code{Elaborate} pragma for the
23113 @code{with}'ed unit.}
23117 By following this rule a client is assured that calls and instantiations
23118 can be made without risk of an exception.
23120 In this mode GNAT traces all calls that are potentially made from
23121 elaboration code, and puts in any missing implicit @code{Elaborate}
23122 and @code{Elaborate_All} pragmas.
23123 The advantage of this approach is that no elaboration problems
23124 are possible if the binder can find an elaboration order that is
23125 consistent with these implicit @code{Elaborate} and
23126 @code{Elaborate_All} pragmas. The
23127 disadvantage of this approach is that no such order may exist.
23129 If the binder does not generate any diagnostics, then it means that it has
23130 found an elaboration order that is guaranteed to be safe. However, the binder
23131 may still be relying on implicitly generated @code{Elaborate} and
23132 @code{Elaborate_All} pragmas so portability to other compilers than GNAT is not
23135 If it is important to guarantee portability, then the compilations should
23138 (warn on elaboration problems) switch. This will cause warning messages
23139 to be generated indicating the missing @code{Elaborate} and
23140 @code{Elaborate_All} pragmas.
23141 Consider the following source program:
23143 @smallexample @c ada
23148 m : integer := k.r;
23155 where it is clear that there
23156 should be a pragma @code{Elaborate_All}
23157 for unit @code{k}. An implicit pragma will be generated, and it is
23158 likely that the binder will be able to honor it. However, if you want
23159 to port this program to some other Ada compiler than GNAT.
23160 it is safer to include the pragma explicitly in the source. If this
23161 unit is compiled with the
23163 switch, then the compiler outputs a warning:
23170 3. m : integer := k.r;
23172 >>> warning: call to "r" may raise Program_Error
23173 >>> warning: missing pragma Elaborate_All for "k"
23181 and these warnings can be used as a guide for supplying manually
23182 the missing pragmas. It is usually a bad idea to use this warning
23183 option during development. That's because it will warn you when
23184 you need to put in a pragma, but cannot warn you when it is time
23185 to take it out. So the use of pragma @code{Elaborate_All} may lead to
23186 unnecessary dependencies and even false circularities.
23188 This default mode is more restrictive than the Ada Reference
23189 Manual, and it is possible to construct programs which will compile
23190 using the dynamic model described there, but will run into a
23191 circularity using the safer static model we have described.
23193 Of course any Ada compiler must be able to operate in a mode
23194 consistent with the requirements of the Ada Reference Manual,
23195 and in particular must have the capability of implementing the
23196 standard dynamic model of elaboration with run-time checks.
23198 In GNAT, this standard mode can be achieved either by the use of
23199 the @option{-gnatE} switch on the compiler (@command{gcc} or
23200 @command{gnatmake}) command, or by the use of the configuration pragma:
23202 @smallexample @c ada
23203 pragma Elaboration_Checks (DYNAMIC);
23207 Either approach will cause the unit affected to be compiled using the
23208 standard dynamic run-time elaboration checks described in the Ada
23209 Reference Manual. The static model is generally preferable, since it
23210 is clearly safer to rely on compile and link time checks rather than
23211 run-time checks. However, in the case of legacy code, it may be
23212 difficult to meet the requirements of the static model. This
23213 issue is further discussed in
23214 @ref{What to Do If the Default Elaboration Behavior Fails}.
23216 Note that the static model provides a strict subset of the allowed
23217 behavior and programs of the Ada Reference Manual, so if you do
23218 adhere to the static model and no circularities exist,
23219 then you are assured that your program will
23220 work using the dynamic model, providing that you remove any
23221 pragma Elaborate statements from the source.
23223 @node Treatment of Pragma Elaborate
23224 @section Treatment of Pragma Elaborate
23225 @cindex Pragma Elaborate
23228 The use of @code{pragma Elaborate}
23229 should generally be avoided in Ada 95 and Ada 2005 programs,
23230 since there is no guarantee that transitive calls
23231 will be properly handled. Indeed at one point, this pragma was placed
23232 in Annex J (Obsolescent Features), on the grounds that it is never useful.
23234 Now that's a bit restrictive. In practice, the case in which
23235 @code{pragma Elaborate} is useful is when the caller knows that there
23236 are no transitive calls, or that the called unit contains all necessary
23237 transitive @code{pragma Elaborate} statements, and legacy code often
23238 contains such uses.
23240 Strictly speaking the static mode in GNAT should ignore such pragmas,
23241 since there is no assurance at compile time that the necessary safety
23242 conditions are met. In practice, this would cause GNAT to be incompatible
23243 with correctly written Ada 83 code that had all necessary
23244 @code{pragma Elaborate} statements in place. Consequently, we made the
23245 decision that GNAT in its default mode will believe that if it encounters
23246 a @code{pragma Elaborate} then the programmer knows what they are doing,
23247 and it will trust that no elaboration errors can occur.
23249 The result of this decision is two-fold. First to be safe using the
23250 static mode, you should remove all @code{pragma Elaborate} statements.
23251 Second, when fixing circularities in existing code, you can selectively
23252 use @code{pragma Elaborate} statements to convince the static mode of
23253 GNAT that it need not generate an implicit @code{pragma Elaborate_All}
23256 When using the static mode with @option{-gnatwl}, any use of
23257 @code{pragma Elaborate} will generate a warning about possible
23260 @node Elaboration Issues for Library Tasks
23261 @section Elaboration Issues for Library Tasks
23262 @cindex Library tasks, elaboration issues
23263 @cindex Elaboration of library tasks
23266 In this section we examine special elaboration issues that arise for
23267 programs that declare library level tasks.
23269 Generally the model of execution of an Ada program is that all units are
23270 elaborated, and then execution of the program starts. However, the
23271 declaration of library tasks definitely does not fit this model. The
23272 reason for this is that library tasks start as soon as they are declared
23273 (more precisely, as soon as the statement part of the enclosing package
23274 body is reached), that is to say before elaboration
23275 of the program is complete. This means that if such a task calls a
23276 subprogram, or an entry in another task, the callee may or may not be
23277 elaborated yet, and in the standard
23278 Reference Manual model of dynamic elaboration checks, you can even
23279 get timing dependent Program_Error exceptions, since there can be
23280 a race between the elaboration code and the task code.
23282 The static model of elaboration in GNAT seeks to avoid all such
23283 dynamic behavior, by being conservative, and the conservative
23284 approach in this particular case is to assume that all the code
23285 in a task body is potentially executed at elaboration time if
23286 a task is declared at the library level.
23288 This can definitely result in unexpected circularities. Consider
23289 the following example
23291 @smallexample @c ada
23297 type My_Int is new Integer;
23299 function Ident (M : My_Int) return My_Int;
23303 package body Decls is
23304 task body Lib_Task is
23310 function Ident (M : My_Int) return My_Int is
23318 procedure Put_Val (Arg : Decls.My_Int);
23322 package body Utils is
23323 procedure Put_Val (Arg : Decls.My_Int) is
23325 Text_IO.Put_Line (Decls.My_Int'Image (Decls.Ident (Arg)));
23332 Decls.Lib_Task.Start;
23337 If the above example is compiled in the default static elaboration
23338 mode, then a circularity occurs. The circularity comes from the call
23339 @code{Utils.Put_Val} in the task body of @code{Decls.Lib_Task}. Since
23340 this call occurs in elaboration code, we need an implicit pragma
23341 @code{Elaborate_All} for @code{Utils}. This means that not only must
23342 the spec and body of @code{Utils} be elaborated before the body
23343 of @code{Decls}, but also the spec and body of any unit that is
23344 @code{with'ed} by the body of @code{Utils} must also be elaborated before
23345 the body of @code{Decls}. This is the transitive implication of
23346 pragma @code{Elaborate_All} and it makes sense, because in general
23347 the body of @code{Put_Val} might have a call to something in a
23348 @code{with'ed} unit.
23350 In this case, the body of Utils (actually its spec) @code{with's}
23351 @code{Decls}. Unfortunately this means that the body of @code{Decls}
23352 must be elaborated before itself, in case there is a call from the
23353 body of @code{Utils}.
23355 Here is the exact chain of events we are worrying about:
23359 In the body of @code{Decls} a call is made from within the body of a library
23360 task to a subprogram in the package @code{Utils}. Since this call may
23361 occur at elaboration time (given that the task is activated at elaboration
23362 time), we have to assume the worst, i.e., that the
23363 call does happen at elaboration time.
23366 This means that the body and spec of @code{Util} must be elaborated before
23367 the body of @code{Decls} so that this call does not cause an access before
23371 Within the body of @code{Util}, specifically within the body of
23372 @code{Util.Put_Val} there may be calls to any unit @code{with}'ed
23376 One such @code{with}'ed package is package @code{Decls}, so there
23377 might be a call to a subprogram in @code{Decls} in @code{Put_Val}.
23378 In fact there is such a call in this example, but we would have to
23379 assume that there was such a call even if it were not there, since
23380 we are not supposed to write the body of @code{Decls} knowing what
23381 is in the body of @code{Utils}; certainly in the case of the
23382 static elaboration model, the compiler does not know what is in
23383 other bodies and must assume the worst.
23386 This means that the spec and body of @code{Decls} must also be
23387 elaborated before we elaborate the unit containing the call, but
23388 that unit is @code{Decls}! This means that the body of @code{Decls}
23389 must be elaborated before itself, and that's a circularity.
23393 Indeed, if you add an explicit pragma @code{Elaborate_All} for @code{Utils} in
23394 the body of @code{Decls} you will get a true Ada Reference Manual
23395 circularity that makes the program illegal.
23397 In practice, we have found that problems with the static model of
23398 elaboration in existing code often arise from library tasks, so
23399 we must address this particular situation.
23401 Note that if we compile and run the program above, using the dynamic model of
23402 elaboration (that is to say use the @option{-gnatE} switch),
23403 then it compiles, binds,
23404 links, and runs, printing the expected result of 2. Therefore in some sense
23405 the circularity here is only apparent, and we need to capture
23406 the properties of this program that distinguish it from other library-level
23407 tasks that have real elaboration problems.
23409 We have four possible answers to this question:
23414 Use the dynamic model of elaboration.
23416 If we use the @option{-gnatE} switch, then as noted above, the program works.
23417 Why is this? If we examine the task body, it is apparent that the task cannot
23419 @code{accept} statement until after elaboration has been completed, because
23420 the corresponding entry call comes from the main program, not earlier.
23421 This is why the dynamic model works here. But that's really giving
23422 up on a precise analysis, and we prefer to take this approach only if we cannot
23424 problem in any other manner. So let us examine two ways to reorganize
23425 the program to avoid the potential elaboration problem.
23428 Split library tasks into separate packages.
23430 Write separate packages, so that library tasks are isolated from
23431 other declarations as much as possible. Let us look at a variation on
23434 @smallexample @c ada
23442 package body Decls1 is
23443 task body Lib_Task is
23451 type My_Int is new Integer;
23452 function Ident (M : My_Int) return My_Int;
23456 package body Decls2 is
23457 function Ident (M : My_Int) return My_Int is
23465 procedure Put_Val (Arg : Decls2.My_Int);
23469 package body Utils is
23470 procedure Put_Val (Arg : Decls2.My_Int) is
23472 Text_IO.Put_Line (Decls2.My_Int'Image (Decls2.Ident (Arg)));
23479 Decls1.Lib_Task.Start;
23484 All we have done is to split @code{Decls} into two packages, one
23485 containing the library task, and one containing everything else. Now
23486 there is no cycle, and the program compiles, binds, links and executes
23487 using the default static model of elaboration.
23490 Declare separate task types.
23492 A significant part of the problem arises because of the use of the
23493 single task declaration form. This means that the elaboration of
23494 the task type, and the elaboration of the task itself (i.e.@: the
23495 creation of the task) happen at the same time. A good rule
23496 of style in Ada is to always create explicit task types. By
23497 following the additional step of placing task objects in separate
23498 packages from the task type declaration, many elaboration problems
23499 are avoided. Here is another modified example of the example program:
23501 @smallexample @c ada
23503 task type Lib_Task_Type is
23507 type My_Int is new Integer;
23509 function Ident (M : My_Int) return My_Int;
23513 package body Decls is
23514 task body Lib_Task_Type is
23520 function Ident (M : My_Int) return My_Int is
23528 procedure Put_Val (Arg : Decls.My_Int);
23532 package body Utils is
23533 procedure Put_Val (Arg : Decls.My_Int) is
23535 Text_IO.Put_Line (Decls.My_Int'Image (Decls.Ident (Arg)));
23541 Lib_Task : Decls.Lib_Task_Type;
23547 Declst.Lib_Task.Start;
23552 What we have done here is to replace the @code{task} declaration in
23553 package @code{Decls} with a @code{task type} declaration. Then we
23554 introduce a separate package @code{Declst} to contain the actual
23555 task object. This separates the elaboration issues for
23556 the @code{task type}
23557 declaration, which causes no trouble, from the elaboration issues
23558 of the task object, which is also unproblematic, since it is now independent
23559 of the elaboration of @code{Utils}.
23560 This separation of concerns also corresponds to
23561 a generally sound engineering principle of separating declarations
23562 from instances. This version of the program also compiles, binds, links,
23563 and executes, generating the expected output.
23566 Use No_Entry_Calls_In_Elaboration_Code restriction.
23567 @cindex No_Entry_Calls_In_Elaboration_Code
23569 The previous two approaches described how a program can be restructured
23570 to avoid the special problems caused by library task bodies. in practice,
23571 however, such restructuring may be difficult to apply to existing legacy code,
23572 so we must consider solutions that do not require massive rewriting.
23574 Let us consider more carefully why our original sample program works
23575 under the dynamic model of elaboration. The reason is that the code
23576 in the task body blocks immediately on the @code{accept}
23577 statement. Now of course there is nothing to prohibit elaboration
23578 code from making entry calls (for example from another library level task),
23579 so we cannot tell in isolation that
23580 the task will not execute the accept statement during elaboration.
23582 However, in practice it is very unusual to see elaboration code
23583 make any entry calls, and the pattern of tasks starting
23584 at elaboration time and then immediately blocking on @code{accept} or
23585 @code{select} statements is very common. What this means is that
23586 the compiler is being too pessimistic when it analyzes the
23587 whole package body as though it might be executed at elaboration
23590 If we know that the elaboration code contains no entry calls, (a very safe
23591 assumption most of the time, that could almost be made the default
23592 behavior), then we can compile all units of the program under control
23593 of the following configuration pragma:
23596 pragma Restrictions (No_Entry_Calls_In_Elaboration_Code);
23600 This pragma can be placed in the @file{gnat.adc} file in the usual
23601 manner. If we take our original unmodified program and compile it
23602 in the presence of a @file{gnat.adc} containing the above pragma,
23603 then once again, we can compile, bind, link, and execute, obtaining
23604 the expected result. In the presence of this pragma, the compiler does
23605 not trace calls in a task body, that appear after the first @code{accept}
23606 or @code{select} statement, and therefore does not report a potential
23607 circularity in the original program.
23609 The compiler will check to the extent it can that the above
23610 restriction is not violated, but it is not always possible to do a
23611 complete check at compile time, so it is important to use this
23612 pragma only if the stated restriction is in fact met, that is to say
23613 no task receives an entry call before elaboration of all units is completed.
23617 @node Mixing Elaboration Models
23618 @section Mixing Elaboration Models
23620 So far, we have assumed that the entire program is either compiled
23621 using the dynamic model or static model, ensuring consistency. It
23622 is possible to mix the two models, but rules have to be followed
23623 if this mixing is done to ensure that elaboration checks are not
23626 The basic rule is that @emph{a unit compiled with the static model cannot
23627 be @code{with'ed} by a unit compiled with the dynamic model}. The
23628 reason for this is that in the static model, a unit assumes that
23629 its clients guarantee to use (the equivalent of) pragma
23630 @code{Elaborate_All} so that no elaboration checks are required
23631 in inner subprograms, and this assumption is violated if the
23632 client is compiled with dynamic checks.
23634 The precise rule is as follows. A unit that is compiled with dynamic
23635 checks can only @code{with} a unit that meets at least one of the
23636 following criteria:
23641 The @code{with'ed} unit is itself compiled with dynamic elaboration
23642 checks (that is with the @option{-gnatE} switch.
23645 The @code{with'ed} unit is an internal GNAT implementation unit from
23646 the System, Interfaces, Ada, or GNAT hierarchies.
23649 The @code{with'ed} unit has pragma Preelaborate or pragma Pure.
23652 The @code{with'ing} unit (that is the client) has an explicit pragma
23653 @code{Elaborate_All} for the @code{with'ed} unit.
23658 If this rule is violated, that is if a unit with dynamic elaboration
23659 checks @code{with's} a unit that does not meet one of the above four
23660 criteria, then the binder (@code{gnatbind}) will issue a warning
23661 similar to that in the following example:
23664 warning: "x.ads" has dynamic elaboration checks and with's
23665 warning: "y.ads" which has static elaboration checks
23669 These warnings indicate that the rule has been violated, and that as a result
23670 elaboration checks may be missed in the resulting executable file.
23671 This warning may be suppressed using the @option{-ws} binder switch
23672 in the usual manner.
23674 One useful application of this mixing rule is in the case of a subsystem
23675 which does not itself @code{with} units from the remainder of the
23676 application. In this case, the entire subsystem can be compiled with
23677 dynamic checks to resolve a circularity in the subsystem, while
23678 allowing the main application that uses this subsystem to be compiled
23679 using the more reliable default static model.
23681 @node What to Do If the Default Elaboration Behavior Fails
23682 @section What to Do If the Default Elaboration Behavior Fails
23685 If the binder cannot find an acceptable order, it outputs detailed
23686 diagnostics. For example:
23692 error: elaboration circularity detected
23693 info: "proc (body)" must be elaborated before "pack (body)"
23694 info: reason: Elaborate_All probably needed in unit "pack (body)"
23695 info: recompile "pack (body)" with -gnatwl
23696 info: for full details
23697 info: "proc (body)"
23698 info: is needed by its spec:
23699 info: "proc (spec)"
23700 info: which is withed by:
23701 info: "pack (body)"
23702 info: "pack (body)" must be elaborated before "proc (body)"
23703 info: reason: pragma Elaborate in unit "proc (body)"
23709 In this case we have a cycle that the binder cannot break. On the one
23710 hand, there is an explicit pragma Elaborate in @code{proc} for
23711 @code{pack}. This means that the body of @code{pack} must be elaborated
23712 before the body of @code{proc}. On the other hand, there is elaboration
23713 code in @code{pack} that calls a subprogram in @code{proc}. This means
23714 that for maximum safety, there should really be a pragma
23715 Elaborate_All in @code{pack} for @code{proc} which would require that
23716 the body of @code{proc} be elaborated before the body of
23717 @code{pack}. Clearly both requirements cannot be satisfied.
23718 Faced with a circularity of this kind, you have three different options.
23721 @item Fix the program
23722 The most desirable option from the point of view of long-term maintenance
23723 is to rearrange the program so that the elaboration problems are avoided.
23724 One useful technique is to place the elaboration code into separate
23725 child packages. Another is to move some of the initialization code to
23726 explicitly called subprograms, where the program controls the order
23727 of initialization explicitly. Although this is the most desirable option,
23728 it may be impractical and involve too much modification, especially in
23729 the case of complex legacy code.
23731 @item Perform dynamic checks
23732 If the compilations are done using the
23734 (dynamic elaboration check) switch, then GNAT behaves in a quite different
23735 manner. Dynamic checks are generated for all calls that could possibly result
23736 in raising an exception. With this switch, the compiler does not generate
23737 implicit @code{Elaborate} or @code{Elaborate_All} pragmas. The behavior then is
23738 exactly as specified in the @cite{Ada Reference Manual}.
23739 The binder will generate
23740 an executable program that may or may not raise @code{Program_Error}, and then
23741 it is the programmer's job to ensure that it does not raise an exception. Note
23742 that it is important to compile all units with the switch, it cannot be used
23745 @item Suppress checks
23746 The drawback of dynamic checks is that they generate a
23747 significant overhead at run time, both in space and time. If you
23748 are absolutely sure that your program cannot raise any elaboration
23749 exceptions, and you still want to use the dynamic elaboration model,
23750 then you can use the configuration pragma
23751 @code{Suppress (Elaboration_Check)} to suppress all such checks. For
23752 example this pragma could be placed in the @file{gnat.adc} file.
23754 @item Suppress checks selectively
23755 When you know that certain calls or instantiations in elaboration code cannot
23756 possibly lead to an elaboration error, and the binder nevertheless complains
23757 about implicit @code{Elaborate} and @code{Elaborate_All} pragmas that lead to
23758 elaboration circularities, it is possible to remove those warnings locally and
23759 obtain a program that will bind. Clearly this can be unsafe, and it is the
23760 responsibility of the programmer to make sure that the resulting program has no
23761 elaboration anomalies. The pragma @code{Suppress (Elaboration_Check)} can be
23762 used with different granularity to suppress warnings and break elaboration
23767 Place the pragma that names the called subprogram in the declarative part
23768 that contains the call.
23771 Place the pragma in the declarative part, without naming an entity. This
23772 disables warnings on all calls in the corresponding declarative region.
23775 Place the pragma in the package spec that declares the called subprogram,
23776 and name the subprogram. This disables warnings on all elaboration calls to
23780 Place the pragma in the package spec that declares the called subprogram,
23781 without naming any entity. This disables warnings on all elaboration calls to
23782 all subprograms declared in this spec.
23784 @item Use Pragma Elaborate
23785 As previously described in section @xref{Treatment of Pragma Elaborate},
23786 GNAT in static mode assumes that a @code{pragma} Elaborate indicates correctly
23787 that no elaboration checks are required on calls to the designated unit.
23788 There may be cases in which the caller knows that no transitive calls
23789 can occur, so that a @code{pragma Elaborate} will be sufficient in a
23790 case where @code{pragma Elaborate_All} would cause a circularity.
23794 These five cases are listed in order of decreasing safety, and therefore
23795 require increasing programmer care in their application. Consider the
23798 @smallexample @c adanocomment
23800 function F1 return Integer;
23805 function F2 return Integer;
23806 function Pure (x : integer) return integer;
23807 -- pragma Suppress (Elaboration_Check, On => Pure); -- (3)
23808 -- pragma Suppress (Elaboration_Check); -- (4)
23812 package body Pack1 is
23813 function F1 return Integer is
23817 Val : integer := Pack2.Pure (11); -- Elab. call (1)
23820 -- pragma Suppress(Elaboration_Check, Pack2.F2); -- (1)
23821 -- pragma Suppress(Elaboration_Check); -- (2)
23823 X1 := Pack2.F2 + 1; -- Elab. call (2)
23828 package body Pack2 is
23829 function F2 return Integer is
23833 function Pure (x : integer) return integer is
23835 return x ** 3 - 3 * x;
23839 with Pack1, Ada.Text_IO;
23842 Ada.Text_IO.Put_Line(Pack1.X1'Img); -- 101
23845 In the absence of any pragmas, an attempt to bind this program produces
23846 the following diagnostics:
23852 error: elaboration circularity detected
23853 info: "pack1 (body)" must be elaborated before "pack1 (body)"
23854 info: reason: Elaborate_All probably needed in unit "pack1 (body)"
23855 info: recompile "pack1 (body)" with -gnatwl for full details
23856 info: "pack1 (body)"
23857 info: must be elaborated along with its spec:
23858 info: "pack1 (spec)"
23859 info: which is withed by:
23860 info: "pack2 (body)"
23861 info: which must be elaborated along with its spec:
23862 info: "pack2 (spec)"
23863 info: which is withed by:
23864 info: "pack1 (body)"
23867 The sources of the circularity are the two calls to @code{Pack2.Pure} and
23868 @code{Pack2.F2} in the body of @code{Pack1}. We can see that the call to
23869 F2 is safe, even though F2 calls F1, because the call appears after the
23870 elaboration of the body of F1. Therefore the pragma (1) is safe, and will
23871 remove the warning on the call. It is also possible to use pragma (2)
23872 because there are no other potentially unsafe calls in the block.
23875 The call to @code{Pure} is safe because this function does not depend on the
23876 state of @code{Pack2}. Therefore any call to this function is safe, and it
23877 is correct to place pragma (3) in the corresponding package spec.
23880 Finally, we could place pragma (4) in the spec of @code{Pack2} to disable
23881 warnings on all calls to functions declared therein. Note that this is not
23882 necessarily safe, and requires more detailed examination of the subprogram
23883 bodies involved. In particular, a call to @code{F2} requires that @code{F1}
23884 be already elaborated.
23888 It is hard to generalize on which of these four approaches should be
23889 taken. Obviously if it is possible to fix the program so that the default
23890 treatment works, this is preferable, but this may not always be practical.
23891 It is certainly simple enough to use
23893 but the danger in this case is that, even if the GNAT binder
23894 finds a correct elaboration order, it may not always do so,
23895 and certainly a binder from another Ada compiler might not. A
23896 combination of testing and analysis (for which the warnings generated
23899 switch can be useful) must be used to ensure that the program is free
23900 of errors. One switch that is useful in this testing is the
23901 @option{^-p (pessimistic elaboration order)^/PESSIMISTIC_ELABORATION_ORDER^}
23904 Normally the binder tries to find an order that has the best chance
23905 of avoiding elaboration problems. However, if this switch is used, the binder
23906 plays a devil's advocate role, and tries to choose the order that
23907 has the best chance of failing. If your program works even with this
23908 switch, then it has a better chance of being error free, but this is still
23911 For an example of this approach in action, consider the C-tests (executable
23912 tests) from the ACVC suite. If these are compiled and run with the default
23913 treatment, then all but one of them succeed without generating any error
23914 diagnostics from the binder. However, there is one test that fails, and
23915 this is not surprising, because the whole point of this test is to ensure
23916 that the compiler can handle cases where it is impossible to determine
23917 a correct order statically, and it checks that an exception is indeed
23918 raised at run time.
23920 This one test must be compiled and run using the
23922 switch, and then it passes. Alternatively, the entire suite can
23923 be run using this switch. It is never wrong to run with the dynamic
23924 elaboration switch if your code is correct, and we assume that the
23925 C-tests are indeed correct (it is less efficient, but efficiency is
23926 not a factor in running the ACVC tests.)
23928 @node Elaboration for Access-to-Subprogram Values
23929 @section Elaboration for Access-to-Subprogram Values
23930 @cindex Access-to-subprogram
23933 Access-to-subprogram types (introduced in Ada 95) complicate
23934 the handling of elaboration. The trouble is that it becomes
23935 impossible to tell at compile time which procedure
23936 is being called. This means that it is not possible for the binder
23937 to analyze the elaboration requirements in this case.
23939 If at the point at which the access value is created
23940 (i.e., the evaluation of @code{P'Access} for a subprogram @code{P}),
23941 the body of the subprogram is
23942 known to have been elaborated, then the access value is safe, and its use
23943 does not require a check. This may be achieved by appropriate arrangement
23944 of the order of declarations if the subprogram is in the current unit,
23945 or, if the subprogram is in another unit, by using pragma
23946 @code{Pure}, @code{Preelaborate}, or @code{Elaborate_Body}
23947 on the referenced unit.
23949 If the referenced body is not known to have been elaborated at the point
23950 the access value is created, then any use of the access value must do a
23951 dynamic check, and this dynamic check will fail and raise a
23952 @code{Program_Error} exception if the body has not been elaborated yet.
23953 GNAT will generate the necessary checks, and in addition, if the
23955 switch is set, will generate warnings that such checks are required.
23957 The use of dynamic dispatching for tagged types similarly generates
23958 a requirement for dynamic checks, and premature calls to any primitive
23959 operation of a tagged type before the body of the operation has been
23960 elaborated, will result in the raising of @code{Program_Error}.
23962 @node Summary of Procedures for Elaboration Control
23963 @section Summary of Procedures for Elaboration Control
23964 @cindex Elaboration control
23967 First, compile your program with the default options, using none of
23968 the special elaboration control switches. If the binder successfully
23969 binds your program, then you can be confident that, apart from issues
23970 raised by the use of access-to-subprogram types and dynamic dispatching,
23971 the program is free of elaboration errors. If it is important that the
23972 program be portable, then use the
23974 switch to generate warnings about missing @code{Elaborate} or
23975 @code{Elaborate_All} pragmas, and supply the missing pragmas.
23977 If the program fails to bind using the default static elaboration
23978 handling, then you can fix the program to eliminate the binder
23979 message, or recompile the entire program with the
23980 @option{-gnatE} switch to generate dynamic elaboration checks,
23981 and, if you are sure there really are no elaboration problems,
23982 use a global pragma @code{Suppress (Elaboration_Check)}.
23984 @node Other Elaboration Order Considerations
23985 @section Other Elaboration Order Considerations
23987 This section has been entirely concerned with the issue of finding a valid
23988 elaboration order, as defined by the Ada Reference Manual. In a case
23989 where several elaboration orders are valid, the task is to find one
23990 of the possible valid elaboration orders (and the static model in GNAT
23991 will ensure that this is achieved).
23993 The purpose of the elaboration rules in the Ada Reference Manual is to
23994 make sure that no entity is accessed before it has been elaborated. For
23995 a subprogram, this means that the spec and body must have been elaborated
23996 before the subprogram is called. For an object, this means that the object
23997 must have been elaborated before its value is read or written. A violation
23998 of either of these two requirements is an access before elaboration order,
23999 and this section has been all about avoiding such errors.
24001 In the case where more than one order of elaboration is possible, in the
24002 sense that access before elaboration errors are avoided, then any one of
24003 the orders is ``correct'' in the sense that it meets the requirements of
24004 the Ada Reference Manual, and no such error occurs.
24006 However, it may be the case for a given program, that there are
24007 constraints on the order of elaboration that come not from consideration
24008 of avoiding elaboration errors, but rather from extra-lingual logic
24009 requirements. Consider this example:
24011 @smallexample @c ada
24012 with Init_Constants;
24013 package Constants is
24018 package Init_Constants is
24019 procedure P; -- require a body
24020 end Init_Constants;
24023 package body Init_Constants is
24024 procedure P is begin null; end;
24028 end Init_Constants;
24032 Z : Integer := Constants.X + Constants.Y;
24036 with Text_IO; use Text_IO;
24039 Put_Line (Calc.Z'Img);
24044 In this example, there is more than one valid order of elaboration. For
24045 example both the following are correct orders:
24048 Init_Constants spec
24051 Init_Constants body
24056 Init_Constants spec
24057 Init_Constants body
24064 There is no language rule to prefer one or the other, both are correct
24065 from an order of elaboration point of view. But the programmatic effects
24066 of the two orders are very different. In the first, the elaboration routine
24067 of @code{Calc} initializes @code{Z} to zero, and then the main program
24068 runs with this value of zero. But in the second order, the elaboration
24069 routine of @code{Calc} runs after the body of Init_Constants has set
24070 @code{X} and @code{Y} and thus @code{Z} is set to 7 before @code{Main}
24073 One could perhaps by applying pretty clever non-artificial intelligence
24074 to the situation guess that it is more likely that the second order of
24075 elaboration is the one desired, but there is no formal linguistic reason
24076 to prefer one over the other. In fact in this particular case, GNAT will
24077 prefer the second order, because of the rule that bodies are elaborated
24078 as soon as possible, but it's just luck that this is what was wanted
24079 (if indeed the second order was preferred).
24081 If the program cares about the order of elaboration routines in a case like
24082 this, it is important to specify the order required. In this particular
24083 case, that could have been achieved by adding to the spec of Calc:
24085 @smallexample @c ada
24086 pragma Elaborate_All (Constants);
24090 which requires that the body (if any) and spec of @code{Constants},
24091 as well as the body and spec of any unit @code{with}'ed by
24092 @code{Constants} be elaborated before @code{Calc} is elaborated.
24094 Clearly no automatic method can always guess which alternative you require,
24095 and if you are working with legacy code that had constraints of this kind
24096 which were not properly specified by adding @code{Elaborate} or
24097 @code{Elaborate_All} pragmas, then indeed it is possible that two different
24098 compilers can choose different orders.
24100 However, GNAT does attempt to diagnose the common situation where there
24101 are uninitialized variables in the visible part of a package spec, and the
24102 corresponding package body has an elaboration block that directly or
24103 indirectly initialized one or more of these variables. This is the situation
24104 in which a pragma Elaborate_Body is usually desirable, and GNAT will generate
24105 a warning that suggests this addition if it detects this situation.
24107 The @code{gnatbind}
24108 @option{^-p^/PESSIMISTIC_ELABORATION^} switch may be useful in smoking
24109 out problems. This switch causes bodies to be elaborated as late as possible
24110 instead of as early as possible. In the example above, it would have forced
24111 the choice of the first elaboration order. If you get different results
24112 when using this switch, and particularly if one set of results is right,
24113 and one is wrong as far as you are concerned, it shows that you have some
24114 missing @code{Elaborate} pragmas. For the example above, we have the
24118 gnatmake -f -q main
24121 gnatmake -f -q main -bargs -p
24127 It is of course quite unlikely that both these results are correct, so
24128 it is up to you in a case like this to investigate the source of the
24129 difference, by looking at the two elaboration orders that are chosen,
24130 and figuring out which is correct, and then adding the necessary
24131 @code{Elaborate} or @code{Elaborate_All} pragmas to ensure the desired order.
24135 @c *******************************
24136 @node Conditional Compilation
24137 @appendix Conditional Compilation
24138 @c *******************************
24139 @cindex Conditional compilation
24142 It is often necessary to arrange for a single source program
24143 to serve multiple purposes, where it is compiled in different
24144 ways to achieve these different goals. Some examples of the
24145 need for this feature are
24148 @item Adapting a program to a different hardware environment
24149 @item Adapting a program to a different target architecture
24150 @item Turning debugging features on and off
24151 @item Arranging for a program to compile with different compilers
24155 In C, or C++, the typical approach would be to use the preprocessor
24156 that is defined as part of the language. The Ada language does not
24157 contain such a feature. This is not an oversight, but rather a very
24158 deliberate design decision, based on the experience that overuse of
24159 the preprocessing features in C and C++ can result in programs that
24160 are extremely difficult to maintain. For example, if we have ten
24161 switches that can be on or off, this means that there are a thousand
24162 separate programs, any one of which might not even be syntactically
24163 correct, and even if syntactically correct, the resulting program
24164 might not work correctly. Testing all combinations can quickly become
24167 Nevertheless, the need to tailor programs certainly exists, and in
24168 this Appendix we will discuss how this can
24169 be achieved using Ada in general, and GNAT in particular.
24172 * Use of Boolean Constants::
24173 * Debugging - A Special Case::
24174 * Conditionalizing Declarations::
24175 * Use of Alternative Implementations::
24179 @node Use of Boolean Constants
24180 @section Use of Boolean Constants
24183 In the case where the difference is simply which code
24184 sequence is executed, the cleanest solution is to use Boolean
24185 constants to control which code is executed.
24187 @smallexample @c ada
24189 FP_Initialize_Required : constant Boolean := True;
24191 if FP_Initialize_Required then
24198 Not only will the code inside the @code{if} statement not be executed if
24199 the constant Boolean is @code{False}, but it will also be completely
24200 deleted from the program.
24201 However, the code is only deleted after the @code{if} statement
24202 has been checked for syntactic and semantic correctness.
24203 (In contrast, with preprocessors the code is deleted before the
24204 compiler ever gets to see it, so it is not checked until the switch
24206 @cindex Preprocessors (contrasted with conditional compilation)
24208 Typically the Boolean constants will be in a separate package,
24211 @smallexample @c ada
24214 FP_Initialize_Required : constant Boolean := True;
24215 Reset_Available : constant Boolean := False;
24222 The @code{Config} package exists in multiple forms for the various targets,
24223 with an appropriate script selecting the version of @code{Config} needed.
24224 Then any other unit requiring conditional compilation can do a @code{with}
24225 of @code{Config} to make the constants visible.
24228 @node Debugging - A Special Case
24229 @section Debugging - A Special Case
24232 A common use of conditional code is to execute statements (for example
24233 dynamic checks, or output of intermediate results) under control of a
24234 debug switch, so that the debugging behavior can be turned on and off.
24235 This can be done using a Boolean constant to control whether the code
24238 @smallexample @c ada
24241 Put_Line ("got to the first stage!");
24249 @smallexample @c ada
24251 if Debugging and then Temperature > 999.0 then
24252 raise Temperature_Crazy;
24258 Since this is a common case, there are special features to deal with
24259 this in a convenient manner. For the case of tests, Ada 2005 has added
24260 a pragma @code{Assert} that can be used for such tests. This pragma is modeled
24261 @cindex pragma @code{Assert}
24262 on the @code{Assert} pragma that has always been available in GNAT, so this
24263 feature may be used with GNAT even if you are not using Ada 2005 features.
24264 The use of pragma @code{Assert} is described in
24265 @ref{Pragma Assert,,, gnat_rm, GNAT Reference Manual}, but as an
24266 example, the last test could be written:
24268 @smallexample @c ada
24269 pragma Assert (Temperature <= 999.0, "Temperature Crazy");
24275 @smallexample @c ada
24276 pragma Assert (Temperature <= 999.0);
24280 In both cases, if assertions are active and the temperature is excessive,
24281 the exception @code{Assert_Failure} will be raised, with the given string in
24282 the first case or a string indicating the location of the pragma in the second
24283 case used as the exception message.
24285 You can turn assertions on and off by using the @code{Assertion_Policy}
24287 @cindex pragma @code{Assertion_Policy}
24288 This is an Ada 2005 pragma which is implemented in all modes by
24289 GNAT, but only in the latest versions of GNAT which include Ada 2005
24290 capability. Alternatively, you can use the @option{-gnata} switch
24291 @cindex @option{-gnata} switch
24292 to enable assertions from the command line (this is recognized by all versions
24295 For the example above with the @code{Put_Line}, the GNAT-specific pragma
24296 @code{Debug} can be used:
24297 @cindex pragma @code{Debug}
24299 @smallexample @c ada
24300 pragma Debug (Put_Line ("got to the first stage!"));
24304 If debug pragmas are enabled, the argument, which must be of the form of
24305 a procedure call, is executed (in this case, @code{Put_Line} will be called).
24306 Only one call can be present, but of course a special debugging procedure
24307 containing any code you like can be included in the program and then
24308 called in a pragma @code{Debug} argument as needed.
24310 One advantage of pragma @code{Debug} over the @code{if Debugging then}
24311 construct is that pragma @code{Debug} can appear in declarative contexts,
24312 such as at the very beginning of a procedure, before local declarations have
24315 Debug pragmas are enabled using either the @option{-gnata} switch that also
24316 controls assertions, or with a separate Debug_Policy pragma.
24317 @cindex pragma @code{Debug_Policy}
24318 The latter pragma is new in the Ada 2005 versions of GNAT (but it can be used
24319 in Ada 95 and Ada 83 programs as well), and is analogous to
24320 pragma @code{Assertion_Policy} to control assertions.
24322 @code{Assertion_Policy} and @code{Debug_Policy} are configuration pragmas,
24323 and thus they can appear in @file{gnat.adc} if you are not using a
24324 project file, or in the file designated to contain configuration pragmas
24326 They then apply to all subsequent compilations. In practice the use of
24327 the @option{-gnata} switch is often the most convenient method of controlling
24328 the status of these pragmas.
24330 Note that a pragma is not a statement, so in contexts where a statement
24331 sequence is required, you can't just write a pragma on its own. You have
24332 to add a @code{null} statement.
24334 @smallexample @c ada
24337 @dots{} -- some statements
24339 pragma Assert (Num_Cases < 10);
24346 @node Conditionalizing Declarations
24347 @section Conditionalizing Declarations
24350 In some cases, it may be necessary to conditionalize declarations to meet
24351 different requirements. For example we might want a bit string whose length
24352 is set to meet some hardware message requirement.
24354 In some cases, it may be possible to do this using declare blocks controlled
24355 by conditional constants:
24357 @smallexample @c ada
24359 if Small_Machine then
24361 X : Bit_String (1 .. 10);
24367 X : Large_Bit_String (1 .. 1000);
24376 Note that in this approach, both declarations are analyzed by the
24377 compiler so this can only be used where both declarations are legal,
24378 even though one of them will not be used.
24380 Another approach is to define integer constants, e.g.@: @code{Bits_Per_Word},
24381 or Boolean constants, e.g.@: @code{Little_Endian}, and then write declarations
24382 that are parameterized by these constants. For example
24384 @smallexample @c ada
24387 Field1 at 0 range Boolean'Pos (Little_Endian) * 10 .. Bits_Per_Word;
24393 If @code{Bits_Per_Word} is set to 32, this generates either
24395 @smallexample @c ada
24398 Field1 at 0 range 0 .. 32;
24404 for the big endian case, or
24406 @smallexample @c ada
24409 Field1 at 0 range 10 .. 32;
24415 for the little endian case. Since a powerful subset of Ada expression
24416 notation is usable for creating static constants, clever use of this
24417 feature can often solve quite difficult problems in conditionalizing
24418 compilation (note incidentally that in Ada 95, the little endian
24419 constant was introduced as @code{System.Default_Bit_Order}, so you do not
24420 need to define this one yourself).
24423 @node Use of Alternative Implementations
24424 @section Use of Alternative Implementations
24427 In some cases, none of the approaches described above are adequate. This
24428 can occur for example if the set of declarations required is radically
24429 different for two different configurations.
24431 In this situation, the official Ada way of dealing with conditionalizing
24432 such code is to write separate units for the different cases. As long as
24433 this does not result in excessive duplication of code, this can be done
24434 without creating maintenance problems. The approach is to share common
24435 code as far as possible, and then isolate the code and declarations
24436 that are different. Subunits are often a convenient method for breaking
24437 out a piece of a unit that is to be conditionalized, with separate files
24438 for different versions of the subunit for different targets, where the
24439 build script selects the right one to give to the compiler.
24440 @cindex Subunits (and conditional compilation)
24442 As an example, consider a situation where a new feature in Ada 2005
24443 allows something to be done in a really nice way. But your code must be able
24444 to compile with an Ada 95 compiler. Conceptually you want to say:
24446 @smallexample @c ada
24449 @dots{} neat Ada 2005 code
24451 @dots{} not quite as neat Ada 95 code
24457 where @code{Ada_2005} is a Boolean constant.
24459 But this won't work when @code{Ada_2005} is set to @code{False},
24460 since the @code{then} clause will be illegal for an Ada 95 compiler.
24461 (Recall that although such unreachable code would eventually be deleted
24462 by the compiler, it still needs to be legal. If it uses features
24463 introduced in Ada 2005, it will be illegal in Ada 95.)
24465 So instead we write
24467 @smallexample @c ada
24468 procedure Insert is separate;
24472 Then we have two files for the subunit @code{Insert}, with the two sets of
24474 If the package containing this is called @code{File_Queries}, then we might
24478 @item @file{file_queries-insert-2005.adb}
24479 @item @file{file_queries-insert-95.adb}
24483 and the build script renames the appropriate file to
24486 file_queries-insert.adb
24490 and then carries out the compilation.
24492 This can also be done with project files' naming schemes. For example:
24494 @smallexample @c project
24495 For Body ("File_Queries.Insert") use "file_queries-insert-2005.ada";
24499 Note also that with project files it is desirable to use a different extension
24500 than @file{ads} / @file{adb} for alternative versions. Otherwise a naming
24501 conflict may arise through another commonly used feature: to declare as part
24502 of the project a set of directories containing all the sources obeying the
24503 default naming scheme.
24505 The use of alternative units is certainly feasible in all situations,
24506 and for example the Ada part of the GNAT run-time is conditionalized
24507 based on the target architecture using this approach. As a specific example,
24508 consider the implementation of the AST feature in VMS. There is one
24516 which is the same for all architectures, and three bodies:
24520 used for all non-VMS operating systems
24521 @item s-asthan-vms-alpha.adb
24522 used for VMS on the Alpha
24523 @item s-asthan-vms-ia64.adb
24524 used for VMS on the ia64
24528 The dummy version @file{s-asthan.adb} simply raises exceptions noting that
24529 this operating system feature is not available, and the two remaining
24530 versions interface with the corresponding versions of VMS to provide
24531 VMS-compatible AST handling. The GNAT build script knows the architecture
24532 and operating system, and automatically selects the right version,
24533 renaming it if necessary to @file{s-asthan.adb} before the run-time build.
24535 Another style for arranging alternative implementations is through Ada's
24536 access-to-subprogram facility.
24537 In case some functionality is to be conditionally included,
24538 you can declare an access-to-procedure variable @code{Ref} that is initialized
24539 to designate a ``do nothing'' procedure, and then invoke @code{Ref.all}
24541 In some library package, set @code{Ref} to @code{Proc'Access} for some
24542 procedure @code{Proc} that performs the relevant processing.
24543 The initialization only occurs if the library package is included in the
24545 The same idea can also be implemented using tagged types and dispatching
24549 @node Preprocessing
24550 @section Preprocessing
24551 @cindex Preprocessing
24554 Although it is quite possible to conditionalize code without the use of
24555 C-style preprocessing, as described earlier in this section, it is
24556 nevertheless convenient in some cases to use the C approach. Moreover,
24557 older Ada compilers have often provided some preprocessing capability,
24558 so legacy code may depend on this approach, even though it is not
24561 To accommodate such use, GNAT provides a preprocessor (modeled to a large
24562 extent on the various preprocessors that have been used
24563 with legacy code on other compilers, to enable easier transition).
24565 The preprocessor may be used in two separate modes. It can be used quite
24566 separately from the compiler, to generate a separate output source file
24567 that is then fed to the compiler as a separate step. This is the
24568 @code{gnatprep} utility, whose use is fully described in
24569 @ref{Preprocessing Using gnatprep}.
24570 @cindex @code{gnatprep}
24572 The preprocessing language allows such constructs as
24576 #if DEBUG or PRIORITY > 4 then
24577 bunch of declarations
24579 completely different bunch of declarations
24585 The values of the symbols @code{DEBUG} and @code{PRIORITY} can be
24586 defined either on the command line or in a separate file.
24588 The other way of running the preprocessor is even closer to the C style and
24589 often more convenient. In this approach the preprocessing is integrated into
24590 the compilation process. The compiler is fed the preprocessor input which
24591 includes @code{#if} lines etc, and then the compiler carries out the
24592 preprocessing internally and processes the resulting output.
24593 For more details on this approach, see @ref{Integrated Preprocessing}.
24596 @c *******************************
24597 @node Inline Assembler
24598 @appendix Inline Assembler
24599 @c *******************************
24602 If you need to write low-level software that interacts directly
24603 with the hardware, Ada provides two ways to incorporate assembly
24604 language code into your program. First, you can import and invoke
24605 external routines written in assembly language, an Ada feature fully
24606 supported by GNAT@. However, for small sections of code it may be simpler
24607 or more efficient to include assembly language statements directly
24608 in your Ada source program, using the facilities of the implementation-defined
24609 package @code{System.Machine_Code}, which incorporates the gcc
24610 Inline Assembler. The Inline Assembler approach offers a number of advantages,
24611 including the following:
24614 @item No need to use non-Ada tools
24615 @item Consistent interface over different targets
24616 @item Automatic usage of the proper calling conventions
24617 @item Access to Ada constants and variables
24618 @item Definition of intrinsic routines
24619 @item Possibility of inlining a subprogram comprising assembler code
24620 @item Code optimizer can take Inline Assembler code into account
24623 This chapter presents a series of examples to show you how to use
24624 the Inline Assembler. Although it focuses on the Intel x86,
24625 the general approach applies also to other processors.
24626 It is assumed that you are familiar with Ada
24627 and with assembly language programming.
24630 * Basic Assembler Syntax::
24631 * A Simple Example of Inline Assembler::
24632 * Output Variables in Inline Assembler::
24633 * Input Variables in Inline Assembler::
24634 * Inlining Inline Assembler Code::
24635 * Other Asm Functionality::
24638 @c ---------------------------------------------------------------------------
24639 @node Basic Assembler Syntax
24640 @section Basic Assembler Syntax
24643 The assembler used by GNAT and gcc is based not on the Intel assembly
24644 language, but rather on a language that descends from the AT&T Unix
24645 assembler @emph{as} (and which is often referred to as ``AT&T syntax'').
24646 The following table summarizes the main features of @emph{as} syntax
24647 and points out the differences from the Intel conventions.
24648 See the gcc @emph{as} and @emph{gas} (an @emph{as} macro
24649 pre-processor) documentation for further information.
24652 @item Register names
24653 gcc / @emph{as}: Prefix with ``%''; for example @code{%eax}
24655 Intel: No extra punctuation; for example @code{eax}
24657 @item Immediate operand
24658 gcc / @emph{as}: Prefix with ``$''; for example @code{$4}
24660 Intel: No extra punctuation; for example @code{4}
24663 gcc / @emph{as}: Prefix with ``$''; for example @code{$loc}
24665 Intel: No extra punctuation; for example @code{loc}
24667 @item Memory contents
24668 gcc / @emph{as}: No extra punctuation; for example @code{loc}
24670 Intel: Square brackets; for example @code{[loc]}
24672 @item Register contents
24673 gcc / @emph{as}: Parentheses; for example @code{(%eax)}
24675 Intel: Square brackets; for example @code{[eax]}
24677 @item Hexadecimal numbers
24678 gcc / @emph{as}: Leading ``0x'' (C language syntax); for example @code{0xA0}
24680 Intel: Trailing ``h''; for example @code{A0h}
24683 gcc / @emph{as}: Explicit in op code; for example @code{movw} to move
24686 Intel: Implicit, deduced by assembler; for example @code{mov}
24688 @item Instruction repetition
24689 gcc / @emph{as}: Split into two lines; for example
24695 Intel: Keep on one line; for example @code{rep stosl}
24697 @item Order of operands
24698 gcc / @emph{as}: Source first; for example @code{movw $4, %eax}
24700 Intel: Destination first; for example @code{mov eax, 4}
24703 @c ---------------------------------------------------------------------------
24704 @node A Simple Example of Inline Assembler
24705 @section A Simple Example of Inline Assembler
24708 The following example will generate a single assembly language statement,
24709 @code{nop}, which does nothing. Despite its lack of run-time effect,
24710 the example will be useful in illustrating the basics of
24711 the Inline Assembler facility.
24713 @smallexample @c ada
24715 with System.Machine_Code; use System.Machine_Code;
24716 procedure Nothing is
24723 @code{Asm} is a procedure declared in package @code{System.Machine_Code};
24724 here it takes one parameter, a @emph{template string} that must be a static
24725 expression and that will form the generated instruction.
24726 @code{Asm} may be regarded as a compile-time procedure that parses
24727 the template string and additional parameters (none here),
24728 from which it generates a sequence of assembly language instructions.
24730 The examples in this chapter will illustrate several of the forms
24731 for invoking @code{Asm}; a complete specification of the syntax
24732 is found in @ref{Machine Code Insertions,,, gnat_rm, GNAT Reference
24735 Under the standard GNAT conventions, the @code{Nothing} procedure
24736 should be in a file named @file{nothing.adb}.
24737 You can build the executable in the usual way:
24741 However, the interesting aspect of this example is not its run-time behavior
24742 but rather the generated assembly code.
24743 To see this output, invoke the compiler as follows:
24745 gcc -c -S -fomit-frame-pointer -gnatp @file{nothing.adb}
24747 where the options are:
24751 compile only (no bind or link)
24753 generate assembler listing
24754 @item -fomit-frame-pointer
24755 do not set up separate stack frames
24757 do not add runtime checks
24760 This gives a human-readable assembler version of the code. The resulting
24761 file will have the same name as the Ada source file, but with a @code{.s}
24762 extension. In our example, the file @file{nothing.s} has the following
24767 .file "nothing.adb"
24769 ___gnu_compiled_ada:
24772 .globl __ada_nothing
24784 The assembly code you included is clearly indicated by
24785 the compiler, between the @code{#APP} and @code{#NO_APP}
24786 delimiters. The character before the 'APP' and 'NOAPP'
24787 can differ on different targets. For example, GNU/Linux uses '#APP' while
24788 on NT you will see '/APP'.
24790 If you make a mistake in your assembler code (such as using the
24791 wrong size modifier, or using a wrong operand for the instruction) GNAT
24792 will report this error in a temporary file, which will be deleted when
24793 the compilation is finished. Generating an assembler file will help
24794 in such cases, since you can assemble this file separately using the
24795 @emph{as} assembler that comes with gcc.
24797 Assembling the file using the command
24800 as @file{nothing.s}
24803 will give you error messages whose lines correspond to the assembler
24804 input file, so you can easily find and correct any mistakes you made.
24805 If there are no errors, @emph{as} will generate an object file
24806 @file{nothing.out}.
24808 @c ---------------------------------------------------------------------------
24809 @node Output Variables in Inline Assembler
24810 @section Output Variables in Inline Assembler
24813 The examples in this section, showing how to access the processor flags,
24814 illustrate how to specify the destination operands for assembly language
24817 @smallexample @c ada
24819 with Interfaces; use Interfaces;
24820 with Ada.Text_IO; use Ada.Text_IO;
24821 with System.Machine_Code; use System.Machine_Code;
24822 procedure Get_Flags is
24823 Flags : Unsigned_32;
24826 Asm ("pushfl" & LF & HT & -- push flags on stack
24827 "popl %%eax" & LF & HT & -- load eax with flags
24828 "movl %%eax, %0", -- store flags in variable
24829 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
24830 Put_Line ("Flags register:" & Flags'Img);
24835 In order to have a nicely aligned assembly listing, we have separated
24836 multiple assembler statements in the Asm template string with linefeed
24837 (ASCII.LF) and horizontal tab (ASCII.HT) characters.
24838 The resulting section of the assembly output file is:
24845 movl %eax, -40(%ebp)
24850 It would have been legal to write the Asm invocation as:
24853 Asm ("pushfl popl %%eax movl %%eax, %0")
24856 but in the generated assembler file, this would come out as:
24860 pushfl popl %eax movl %eax, -40(%ebp)
24864 which is not so convenient for the human reader.
24866 We use Ada comments
24867 at the end of each line to explain what the assembler instructions
24868 actually do. This is a useful convention.
24870 When writing Inline Assembler instructions, you need to precede each register
24871 and variable name with a percent sign. Since the assembler already requires
24872 a percent sign at the beginning of a register name, you need two consecutive
24873 percent signs for such names in the Asm template string, thus @code{%%eax}.
24874 In the generated assembly code, one of the percent signs will be stripped off.
24876 Names such as @code{%0}, @code{%1}, @code{%2}, etc., denote input or output
24877 variables: operands you later define using @code{Input} or @code{Output}
24878 parameters to @code{Asm}.
24879 An output variable is illustrated in
24880 the third statement in the Asm template string:
24884 The intent is to store the contents of the eax register in a variable that can
24885 be accessed in Ada. Simply writing @code{movl %%eax, Flags} would not
24886 necessarily work, since the compiler might optimize by using a register
24887 to hold Flags, and the expansion of the @code{movl} instruction would not be
24888 aware of this optimization. The solution is not to store the result directly
24889 but rather to advise the compiler to choose the correct operand form;
24890 that is the purpose of the @code{%0} output variable.
24892 Information about the output variable is supplied in the @code{Outputs}
24893 parameter to @code{Asm}:
24895 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
24898 The output is defined by the @code{Asm_Output} attribute of the target type;
24899 the general format is
24901 Type'Asm_Output (constraint_string, variable_name)
24904 The constraint string directs the compiler how
24905 to store/access the associated variable. In the example
24907 Unsigned_32'Asm_Output ("=m", Flags);
24909 the @code{"m"} (memory) constraint tells the compiler that the variable
24910 @code{Flags} should be stored in a memory variable, thus preventing
24911 the optimizer from keeping it in a register. In contrast,
24913 Unsigned_32'Asm_Output ("=r", Flags);
24915 uses the @code{"r"} (register) constraint, telling the compiler to
24916 store the variable in a register.
24918 If the constraint is preceded by the equal character (@strong{=}), it tells
24919 the compiler that the variable will be used to store data into it.
24921 In the @code{Get_Flags} example, we used the @code{"g"} (global) constraint,
24922 allowing the optimizer to choose whatever it deems best.
24924 There are a fairly large number of constraints, but the ones that are
24925 most useful (for the Intel x86 processor) are the following:
24931 global (i.e.@: can be stored anywhere)
24949 use one of eax, ebx, ecx or edx
24951 use one of eax, ebx, ecx, edx, esi or edi
24954 The full set of constraints is described in the gcc and @emph{as}
24955 documentation; note that it is possible to combine certain constraints
24956 in one constraint string.
24958 You specify the association of an output variable with an assembler operand
24959 through the @code{%}@emph{n} notation, where @emph{n} is a non-negative
24961 @smallexample @c ada
24963 Asm ("pushfl" & LF & HT & -- push flags on stack
24964 "popl %%eax" & LF & HT & -- load eax with flags
24965 "movl %%eax, %0", -- store flags in variable
24966 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
24970 @code{%0} will be replaced in the expanded code by the appropriate operand,
24972 the compiler decided for the @code{Flags} variable.
24974 In general, you may have any number of output variables:
24977 Count the operands starting at 0; thus @code{%0}, @code{%1}, etc.
24979 Specify the @code{Outputs} parameter as a parenthesized comma-separated list
24980 of @code{Asm_Output} attributes
24984 @smallexample @c ada
24986 Asm ("movl %%eax, %0" & LF & HT &
24987 "movl %%ebx, %1" & LF & HT &
24989 Outputs => (Unsigned_32'Asm_Output ("=g", Var_A), -- %0 = Var_A
24990 Unsigned_32'Asm_Output ("=g", Var_B), -- %1 = Var_B
24991 Unsigned_32'Asm_Output ("=g", Var_C))); -- %2 = Var_C
24995 where @code{Var_A}, @code{Var_B}, and @code{Var_C} are variables
24996 in the Ada program.
24998 As a variation on the @code{Get_Flags} example, we can use the constraints
24999 string to direct the compiler to store the eax register into the @code{Flags}
25000 variable, instead of including the store instruction explicitly in the
25001 @code{Asm} template string:
25003 @smallexample @c ada
25005 with Interfaces; use Interfaces;
25006 with Ada.Text_IO; use Ada.Text_IO;
25007 with System.Machine_Code; use System.Machine_Code;
25008 procedure Get_Flags_2 is
25009 Flags : Unsigned_32;
25012 Asm ("pushfl" & LF & HT & -- push flags on stack
25013 "popl %%eax", -- save flags in eax
25014 Outputs => Unsigned_32'Asm_Output ("=a", Flags));
25015 Put_Line ("Flags register:" & Flags'Img);
25021 The @code{"a"} constraint tells the compiler that the @code{Flags}
25022 variable will come from the eax register. Here is the resulting code:
25030 movl %eax,-40(%ebp)
25035 The compiler generated the store of eax into Flags after
25036 expanding the assembler code.
25038 Actually, there was no need to pop the flags into the eax register;
25039 more simply, we could just pop the flags directly into the program variable:
25041 @smallexample @c ada
25043 with Interfaces; use Interfaces;
25044 with Ada.Text_IO; use Ada.Text_IO;
25045 with System.Machine_Code; use System.Machine_Code;
25046 procedure Get_Flags_3 is
25047 Flags : Unsigned_32;
25050 Asm ("pushfl" & LF & HT & -- push flags on stack
25051 "pop %0", -- save flags in Flags
25052 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
25053 Put_Line ("Flags register:" & Flags'Img);
25058 @c ---------------------------------------------------------------------------
25059 @node Input Variables in Inline Assembler
25060 @section Input Variables in Inline Assembler
25063 The example in this section illustrates how to specify the source operands
25064 for assembly language statements.
25065 The program simply increments its input value by 1:
25067 @smallexample @c ada
25069 with Interfaces; use Interfaces;
25070 with Ada.Text_IO; use Ada.Text_IO;
25071 with System.Machine_Code; use System.Machine_Code;
25072 procedure Increment is
25074 function Incr (Value : Unsigned_32) return Unsigned_32 is
25075 Result : Unsigned_32;
25078 Inputs => Unsigned_32'Asm_Input ("a", Value),
25079 Outputs => Unsigned_32'Asm_Output ("=a", Result));
25083 Value : Unsigned_32;
25087 Put_Line ("Value before is" & Value'Img);
25088 Value := Incr (Value);
25089 Put_Line ("Value after is" & Value'Img);
25094 The @code{Outputs} parameter to @code{Asm} specifies
25095 that the result will be in the eax register and that it is to be stored
25096 in the @code{Result} variable.
25098 The @code{Inputs} parameter looks much like the @code{Outputs} parameter,
25099 but with an @code{Asm_Input} attribute.
25100 The @code{"="} constraint, indicating an output value, is not present.
25102 You can have multiple input variables, in the same way that you can have more
25103 than one output variable.
25105 The parameter count (%0, %1) etc, now starts at the first input
25106 statement, and continues with the output statements.
25107 When both parameters use the same variable, the
25108 compiler will treat them as the same %n operand, which is the case here.
25110 Just as the @code{Outputs} parameter causes the register to be stored into the
25111 target variable after execution of the assembler statements, so does the
25112 @code{Inputs} parameter cause its variable to be loaded into the register
25113 before execution of the assembler statements.
25115 Thus the effect of the @code{Asm} invocation is:
25117 @item load the 32-bit value of @code{Value} into eax
25118 @item execute the @code{incl %eax} instruction
25119 @item store the contents of eax into the @code{Result} variable
25122 The resulting assembler file (with @option{-O2} optimization) contains:
25125 _increment__incr.1:
25138 @c ---------------------------------------------------------------------------
25139 @node Inlining Inline Assembler Code
25140 @section Inlining Inline Assembler Code
25143 For a short subprogram such as the @code{Incr} function in the previous
25144 section, the overhead of the call and return (creating / deleting the stack
25145 frame) can be significant, compared to the amount of code in the subprogram
25146 body. A solution is to apply Ada's @code{Inline} pragma to the subprogram,
25147 which directs the compiler to expand invocations of the subprogram at the
25148 point(s) of call, instead of setting up a stack frame for out-of-line calls.
25149 Here is the resulting program:
25151 @smallexample @c ada
25153 with Interfaces; use Interfaces;
25154 with Ada.Text_IO; use Ada.Text_IO;
25155 with System.Machine_Code; use System.Machine_Code;
25156 procedure Increment_2 is
25158 function Incr (Value : Unsigned_32) return Unsigned_32 is
25159 Result : Unsigned_32;
25162 Inputs => Unsigned_32'Asm_Input ("a", Value),
25163 Outputs => Unsigned_32'Asm_Output ("=a", Result));
25166 pragma Inline (Increment);
25168 Value : Unsigned_32;
25172 Put_Line ("Value before is" & Value'Img);
25173 Value := Increment (Value);
25174 Put_Line ("Value after is" & Value'Img);
25179 Compile the program with both optimization (@option{-O2}) and inlining
25180 (@option{-gnatn}) enabled.
25182 The @code{Incr} function is still compiled as usual, but at the
25183 point in @code{Increment} where our function used to be called:
25188 call _increment__incr.1
25193 the code for the function body directly appears:
25206 thus saving the overhead of stack frame setup and an out-of-line call.
25208 @c ---------------------------------------------------------------------------
25209 @node Other Asm Functionality
25210 @section Other @code{Asm} Functionality
25213 This section describes two important parameters to the @code{Asm}
25214 procedure: @code{Clobber}, which identifies register usage;
25215 and @code{Volatile}, which inhibits unwanted optimizations.
25218 * The Clobber Parameter::
25219 * The Volatile Parameter::
25222 @c ---------------------------------------------------------------------------
25223 @node The Clobber Parameter
25224 @subsection The @code{Clobber} Parameter
25227 One of the dangers of intermixing assembly language and a compiled language
25228 such as Ada is that the compiler needs to be aware of which registers are
25229 being used by the assembly code. In some cases, such as the earlier examples,
25230 the constraint string is sufficient to indicate register usage (e.g.,
25232 the eax register). But more generally, the compiler needs an explicit
25233 identification of the registers that are used by the Inline Assembly
25236 Using a register that the compiler doesn't know about
25237 could be a side effect of an instruction (like @code{mull}
25238 storing its result in both eax and edx).
25239 It can also arise from explicit register usage in your
25240 assembly code; for example:
25243 Asm ("movl %0, %%ebx" & LF & HT &
25245 Inputs => Unsigned_32'Asm_Input ("g", Var_In),
25246 Outputs => Unsigned_32'Asm_Output ("=g", Var_Out));
25250 where the compiler (since it does not analyze the @code{Asm} template string)
25251 does not know you are using the ebx register.
25253 In such cases you need to supply the @code{Clobber} parameter to @code{Asm},
25254 to identify the registers that will be used by your assembly code:
25258 Asm ("movl %0, %%ebx" & LF & HT &
25260 Inputs => Unsigned_32'Asm_Input ("g", Var_In),
25261 Outputs => Unsigned_32'Asm_Output ("=g", Var_Out),
25266 The Clobber parameter is a static string expression specifying the
25267 register(s) you are using. Note that register names are @emph{not} prefixed
25268 by a percent sign. Also, if more than one register is used then their names
25269 are separated by commas; e.g., @code{"eax, ebx"}
25271 The @code{Clobber} parameter has several additional uses:
25273 @item Use ``register'' name @code{cc} to indicate that flags might have changed
25274 @item Use ``register'' name @code{memory} if you changed a memory location
25277 @c ---------------------------------------------------------------------------
25278 @node The Volatile Parameter
25279 @subsection The @code{Volatile} Parameter
25280 @cindex Volatile parameter
25283 Compiler optimizations in the presence of Inline Assembler may sometimes have
25284 unwanted effects. For example, when an @code{Asm} invocation with an input
25285 variable is inside a loop, the compiler might move the loading of the input
25286 variable outside the loop, regarding it as a one-time initialization.
25288 If this effect is not desired, you can disable such optimizations by setting
25289 the @code{Volatile} parameter to @code{True}; for example:
25291 @smallexample @c ada
25293 Asm ("movl %0, %%ebx" & LF & HT &
25295 Inputs => Unsigned_32'Asm_Input ("g", Var_In),
25296 Outputs => Unsigned_32'Asm_Output ("=g", Var_Out),
25302 By default, @code{Volatile} is set to @code{False} unless there is no
25303 @code{Outputs} parameter.
25305 Although setting @code{Volatile} to @code{True} prevents unwanted
25306 optimizations, it will also disable other optimizations that might be
25307 important for efficiency. In general, you should set @code{Volatile}
25308 to @code{True} only if the compiler's optimizations have created
25310 @c END OF INLINE ASSEMBLER CHAPTER
25311 @c ===============================
25313 @c ***********************************
25314 @c * Compatibility and Porting Guide *
25315 @c ***********************************
25316 @node Compatibility and Porting Guide
25317 @appendix Compatibility and Porting Guide
25320 This chapter describes the compatibility issues that may arise between
25321 GNAT and other Ada compilation systems (including those for Ada 83),
25322 and shows how GNAT can expedite porting
25323 applications developed in other Ada environments.
25326 * Compatibility with Ada 83::
25327 * Compatibility between Ada 95 and Ada 2005::
25328 * Implementation-dependent characteristics::
25329 * Compatibility with Other Ada Systems::
25330 * Representation Clauses::
25332 @c Brief section is only in non-VMS version
25333 @c Full chapter is in VMS version
25334 * Compatibility with HP Ada 83::
25337 * Transitioning to 64-Bit GNAT for OpenVMS::
25341 @node Compatibility with Ada 83
25342 @section Compatibility with Ada 83
25343 @cindex Compatibility (between Ada 83 and Ada 95 / Ada 2005)
25346 Ada 95 and Ada 2005 are highly upwards compatible with Ada 83. In
25347 particular, the design intention was that the difficulties associated
25348 with moving from Ada 83 to Ada 95 or Ada 2005 should be no greater than those
25349 that occur when moving from one Ada 83 system to another.
25351 However, there are a number of points at which there are minor
25352 incompatibilities. The @cite{Ada 95 Annotated Reference Manual} contains
25353 full details of these issues,
25354 and should be consulted for a complete treatment.
25356 following subsections treat the most likely issues to be encountered.
25359 * Legal Ada 83 programs that are illegal in Ada 95::
25360 * More deterministic semantics::
25361 * Changed semantics::
25362 * Other language compatibility issues::
25365 @node Legal Ada 83 programs that are illegal in Ada 95
25366 @subsection Legal Ada 83 programs that are illegal in Ada 95
25368 Some legal Ada 83 programs are illegal (i.e., they will fail to compile) in
25369 Ada 95 and thus also in Ada 2005:
25372 @item Character literals
25373 Some uses of character literals are ambiguous. Since Ada 95 has introduced
25374 @code{Wide_Character} as a new predefined character type, some uses of
25375 character literals that were legal in Ada 83 are illegal in Ada 95.
25377 @smallexample @c ada
25378 for Char in 'A' .. 'Z' loop @dots{} end loop;
25382 The problem is that @code{'A'} and @code{'Z'} could be from either
25383 @code{Character} or @code{Wide_Character}. The simplest correction
25384 is to make the type explicit; e.g.:
25385 @smallexample @c ada
25386 for Char in Character range 'A' .. 'Z' loop @dots{} end loop;
25389 @item New reserved words
25390 The identifiers @code{abstract}, @code{aliased}, @code{protected},
25391 @code{requeue}, @code{tagged}, and @code{until} are reserved in Ada 95.
25392 Existing Ada 83 code using any of these identifiers must be edited to
25393 use some alternative name.
25395 @item Freezing rules
25396 The rules in Ada 95 are slightly different with regard to the point at
25397 which entities are frozen, and representation pragmas and clauses are
25398 not permitted past the freeze point. This shows up most typically in
25399 the form of an error message complaining that a representation item
25400 appears too late, and the appropriate corrective action is to move
25401 the item nearer to the declaration of the entity to which it refers.
25403 A particular case is that representation pragmas
25406 extended HP Ada 83 compatibility pragmas such as @code{Export_Procedure})
25408 cannot be applied to a subprogram body. If necessary, a separate subprogram
25409 declaration must be introduced to which the pragma can be applied.
25411 @item Optional bodies for library packages
25412 In Ada 83, a package that did not require a package body was nevertheless
25413 allowed to have one. This lead to certain surprises in compiling large
25414 systems (situations in which the body could be unexpectedly ignored by the
25415 binder). In Ada 95, if a package does not require a body then it is not
25416 permitted to have a body. To fix this problem, simply remove a redundant
25417 body if it is empty, or, if it is non-empty, introduce a dummy declaration
25418 into the spec that makes the body required. One approach is to add a private
25419 part to the package declaration (if necessary), and define a parameterless
25420 procedure called @code{Requires_Body}, which must then be given a dummy
25421 procedure body in the package body, which then becomes required.
25422 Another approach (assuming that this does not introduce elaboration
25423 circularities) is to add an @code{Elaborate_Body} pragma to the package spec,
25424 since one effect of this pragma is to require the presence of a package body.
25426 @item @code{Numeric_Error} is now the same as @code{Constraint_Error}
25427 In Ada 95, the exception @code{Numeric_Error} is a renaming of
25428 @code{Constraint_Error}.
25429 This means that it is illegal to have separate exception handlers for
25430 the two exceptions. The fix is simply to remove the handler for the
25431 @code{Numeric_Error} case (since even in Ada 83, a compiler was free to raise
25432 @code{Constraint_Error} in place of @code{Numeric_Error} in all cases).
25434 @item Indefinite subtypes in generics
25435 In Ada 83, it was permissible to pass an indefinite type (e.g.@: @code{String})
25436 as the actual for a generic formal private type, but then the instantiation
25437 would be illegal if there were any instances of declarations of variables
25438 of this type in the generic body. In Ada 95, to avoid this clear violation
25439 of the methodological principle known as the ``contract model'',
25440 the generic declaration explicitly indicates whether
25441 or not such instantiations are permitted. If a generic formal parameter
25442 has explicit unknown discriminants, indicated by using @code{(<>)} after the
25443 type name, then it can be instantiated with indefinite types, but no
25444 stand-alone variables can be declared of this type. Any attempt to declare
25445 such a variable will result in an illegality at the time the generic is
25446 declared. If the @code{(<>)} notation is not used, then it is illegal
25447 to instantiate the generic with an indefinite type.
25448 This is the potential incompatibility issue when porting Ada 83 code to Ada 95.
25449 It will show up as a compile time error, and
25450 the fix is usually simply to add the @code{(<>)} to the generic declaration.
25453 @node More deterministic semantics
25454 @subsection More deterministic semantics
25458 Conversions from real types to integer types round away from 0. In Ada 83
25459 the conversion Integer(2.5) could deliver either 2 or 3 as its value. This
25460 implementation freedom was intended to support unbiased rounding in
25461 statistical applications, but in practice it interfered with portability.
25462 In Ada 95 the conversion semantics are unambiguous, and rounding away from 0
25463 is required. Numeric code may be affected by this change in semantics.
25464 Note, though, that this issue is no worse than already existed in Ada 83
25465 when porting code from one vendor to another.
25468 The Real-Time Annex introduces a set of policies that define the behavior of
25469 features that were implementation dependent in Ada 83, such as the order in
25470 which open select branches are executed.
25473 @node Changed semantics
25474 @subsection Changed semantics
25477 The worst kind of incompatibility is one where a program that is legal in
25478 Ada 83 is also legal in Ada 95 but can have an effect in Ada 95 that was not
25479 possible in Ada 83. Fortunately this is extremely rare, but the one
25480 situation that you should be alert to is the change in the predefined type
25481 @code{Character} from 7-bit ASCII to 8-bit Latin-1.
25484 @item Range of type @code{Character}
25485 The range of @code{Standard.Character} is now the full 256 characters
25486 of Latin-1, whereas in most Ada 83 implementations it was restricted
25487 to 128 characters. Although some of the effects of
25488 this change will be manifest in compile-time rejection of legal
25489 Ada 83 programs it is possible for a working Ada 83 program to have
25490 a different effect in Ada 95, one that was not permitted in Ada 83.
25491 As an example, the expression
25492 @code{Character'Pos(Character'Last)} returned @code{127} in Ada 83 and now
25493 delivers @code{255} as its value.
25494 In general, you should look at the logic of any
25495 character-processing Ada 83 program and see whether it needs to be adapted
25496 to work correctly with Latin-1. Note that the predefined Ada 95 API has a
25497 character handling package that may be relevant if code needs to be adapted
25498 to account for the additional Latin-1 elements.
25499 The desirable fix is to
25500 modify the program to accommodate the full character set, but in some cases
25501 it may be convenient to define a subtype or derived type of Character that
25502 covers only the restricted range.
25506 @node Other language compatibility issues
25507 @subsection Other language compatibility issues
25510 @item @option{-gnat83} switch
25511 All implementations of GNAT provide a switch that causes GNAT to operate
25512 in Ada 83 mode. In this mode, some but not all compatibility problems
25513 of the type described above are handled automatically. For example, the
25514 new reserved words introduced in Ada 95 and Ada 2005 are treated simply
25515 as identifiers as in Ada 83.
25517 in practice, it is usually advisable to make the necessary modifications
25518 to the program to remove the need for using this switch.
25519 See @ref{Compiling Different Versions of Ada}.
25521 @item Support for removed Ada 83 pragmas and attributes
25522 A number of pragmas and attributes from Ada 83 were removed from Ada 95,
25523 generally because they were replaced by other mechanisms. Ada 95 and Ada 2005
25524 compilers are allowed, but not required, to implement these missing
25525 elements. In contrast with some other compilers, GNAT implements all
25526 such pragmas and attributes, eliminating this compatibility concern. These
25527 include @code{pragma Interface} and the floating point type attributes
25528 (@code{Emax}, @code{Mantissa}, etc.), among other items.
25532 @node Compatibility between Ada 95 and Ada 2005
25533 @section Compatibility between Ada 95 and Ada 2005
25534 @cindex Compatibility between Ada 95 and Ada 2005
25537 Although Ada 2005 was designed to be upwards compatible with Ada 95, there are
25538 a number of incompatibilities. Several are enumerated below;
25539 for a complete description please see the
25540 Annotated Ada 2005 Reference Manual, or section 9.1.1 in
25541 @cite{Rationale for Ada 2005}.
25544 @item New reserved words.
25545 The words @code{interface}, @code{overriding} and @code{synchronized} are
25546 reserved in Ada 2005.
25547 A pre-Ada 2005 program that uses any of these as an identifier will be
25550 @item New declarations in predefined packages.
25551 A number of packages in the predefined environment contain new declarations:
25552 @code{Ada.Exceptions}, @code{Ada.Real_Time}, @code{Ada.Strings},
25553 @code{Ada.Strings.Fixed}, @code{Ada.Strings.Bounded},
25554 @code{Ada.Strings.Unbounded}, @code{Ada.Strings.Wide_Fixed},
25555 @code{Ada.Strings.Wide_Bounded}, @code{Ada.Strings.Wide_Unbounded},
25556 @code{Ada.Tags}, @code{Ada.Text_IO}, and @code{Interfaces.C}.
25557 If an Ada 95 program does a @code{with} and @code{use} of any of these
25558 packages, the new declarations may cause name clashes.
25560 @item Access parameters.
25561 A nondispatching subprogram with an access parameter cannot be renamed
25562 as a dispatching operation. This was permitted in Ada 95.
25564 @item Access types, discriminants, and constraints.
25565 Rule changes in this area have led to some incompatibilities; for example,
25566 constrained subtypes of some access types are not permitted in Ada 2005.
25568 @item Aggregates for limited types.
25569 The allowance of aggregates for limited types in Ada 2005 raises the
25570 possibility of ambiguities in legal Ada 95 programs, since additional types
25571 now need to be considered in expression resolution.
25573 @item Fixed-point multiplication and division.
25574 Certain expressions involving ``*'' or ``/'' for a fixed-point type, which
25575 were legal in Ada 95 and invoked the predefined versions of these operations,
25577 The ambiguity may be resolved either by applying a type conversion to the
25578 expression, or by explicitly invoking the operation from package
25581 @item Return-by-reference types.
25582 The Ada 95 return-by-reference mechanism has been removed. Instead, the user
25583 can declare a function returning a value from an anonymous access type.
25587 @node Implementation-dependent characteristics
25588 @section Implementation-dependent characteristics
25590 Although the Ada language defines the semantics of each construct as
25591 precisely as practical, in some situations (for example for reasons of
25592 efficiency, or where the effect is heavily dependent on the host or target
25593 platform) the implementation is allowed some freedom. In porting Ada 83
25594 code to GNAT, you need to be aware of whether / how the existing code
25595 exercised such implementation dependencies. Such characteristics fall into
25596 several categories, and GNAT offers specific support in assisting the
25597 transition from certain Ada 83 compilers.
25600 * Implementation-defined pragmas::
25601 * Implementation-defined attributes::
25603 * Elaboration order::
25604 * Target-specific aspects::
25607 @node Implementation-defined pragmas
25608 @subsection Implementation-defined pragmas
25611 Ada compilers are allowed to supplement the language-defined pragmas, and
25612 these are a potential source of non-portability. All GNAT-defined pragmas
25613 are described in @ref{Implementation Defined Pragmas,,, gnat_rm, GNAT
25614 Reference Manual}, and these include several that are specifically
25615 intended to correspond to other vendors' Ada 83 pragmas.
25616 For migrating from VADS, the pragma @code{Use_VADS_Size} may be useful.
25617 For compatibility with HP Ada 83, GNAT supplies the pragmas
25618 @code{Extend_System}, @code{Ident}, @code{Inline_Generic},
25619 @code{Interface_Name}, @code{Passive}, @code{Suppress_All},
25620 and @code{Volatile}.
25621 Other relevant pragmas include @code{External} and @code{Link_With}.
25622 Some vendor-specific
25623 Ada 83 pragmas (@code{Share_Generic}, @code{Subtitle}, and @code{Title}) are
25625 avoiding compiler rejection of units that contain such pragmas; they are not
25626 relevant in a GNAT context and hence are not otherwise implemented.
25628 @node Implementation-defined attributes
25629 @subsection Implementation-defined attributes
25631 Analogous to pragmas, the set of attributes may be extended by an
25632 implementation. All GNAT-defined attributes are described in
25633 @ref{Implementation Defined Attributes,,, gnat_rm, GNAT Reference
25634 Manual}, and these include several that are specifically intended
25635 to correspond to other vendors' Ada 83 attributes. For migrating from VADS,
25636 the attribute @code{VADS_Size} may be useful. For compatibility with HP
25637 Ada 83, GNAT supplies the attributes @code{Bit}, @code{Machine_Size} and
25641 @subsection Libraries
25643 Vendors may supply libraries to supplement the standard Ada API. If Ada 83
25644 code uses vendor-specific libraries then there are several ways to manage
25645 this in Ada 95 or Ada 2005:
25648 If the source code for the libraries (specs and bodies) are
25649 available, then the libraries can be migrated in the same way as the
25652 If the source code for the specs but not the bodies are
25653 available, then you can reimplement the bodies.
25655 Some features introduced by Ada 95 obviate the need for library support. For
25656 example most Ada 83 vendors supplied a package for unsigned integers. The
25657 Ada 95 modular type feature is the preferred way to handle this need, so
25658 instead of migrating or reimplementing the unsigned integer package it may
25659 be preferable to retrofit the application using modular types.
25662 @node Elaboration order
25663 @subsection Elaboration order
25665 The implementation can choose any elaboration order consistent with the unit
25666 dependency relationship. This freedom means that some orders can result in
25667 Program_Error being raised due to an ``Access Before Elaboration'': an attempt
25668 to invoke a subprogram its body has been elaborated, or to instantiate a
25669 generic before the generic body has been elaborated. By default GNAT
25670 attempts to choose a safe order (one that will not encounter access before
25671 elaboration problems) by implicitly inserting @code{Elaborate} or
25672 @code{Elaborate_All} pragmas where
25673 needed. However, this can lead to the creation of elaboration circularities
25674 and a resulting rejection of the program by gnatbind. This issue is
25675 thoroughly described in @ref{Elaboration Order Handling in GNAT}.
25676 In brief, there are several
25677 ways to deal with this situation:
25681 Modify the program to eliminate the circularities, e.g.@: by moving
25682 elaboration-time code into explicitly-invoked procedures
25684 Constrain the elaboration order by including explicit @code{Elaborate_Body} or
25685 @code{Elaborate} pragmas, and then inhibit the generation of implicit
25686 @code{Elaborate_All}
25687 pragmas either globally (as an effect of the @option{-gnatE} switch) or locally
25688 (by selectively suppressing elaboration checks via pragma
25689 @code{Suppress(Elaboration_Check)} when it is safe to do so).
25692 @node Target-specific aspects
25693 @subsection Target-specific aspects
25695 Low-level applications need to deal with machine addresses, data
25696 representations, interfacing with assembler code, and similar issues. If
25697 such an Ada 83 application is being ported to different target hardware (for
25698 example where the byte endianness has changed) then you will need to
25699 carefully examine the program logic; the porting effort will heavily depend
25700 on the robustness of the original design. Moreover, Ada 95 (and thus
25701 Ada 2005) are sometimes
25702 incompatible with typical Ada 83 compiler practices regarding implicit
25703 packing, the meaning of the Size attribute, and the size of access values.
25704 GNAT's approach to these issues is described in @ref{Representation Clauses}.
25706 @node Compatibility with Other Ada Systems
25707 @section Compatibility with Other Ada Systems
25710 If programs avoid the use of implementation dependent and
25711 implementation defined features, as documented in the @cite{Ada
25712 Reference Manual}, there should be a high degree of portability between
25713 GNAT and other Ada systems. The following are specific items which
25714 have proved troublesome in moving Ada 95 programs from GNAT to other Ada 95
25715 compilers, but do not affect porting code to GNAT@.
25716 (As of @value{NOW}, GNAT is the only compiler available for Ada 2005;
25717 the following issues may or may not arise for Ada 2005 programs
25718 when other compilers appear.)
25721 @item Ada 83 Pragmas and Attributes
25722 Ada 95 compilers are allowed, but not required, to implement the missing
25723 Ada 83 pragmas and attributes that are no longer defined in Ada 95.
25724 GNAT implements all such pragmas and attributes, eliminating this as
25725 a compatibility concern, but some other Ada 95 compilers reject these
25726 pragmas and attributes.
25728 @item Specialized Needs Annexes
25729 GNAT implements the full set of special needs annexes. At the
25730 current time, it is the only Ada 95 compiler to do so. This means that
25731 programs making use of these features may not be portable to other Ada
25732 95 compilation systems.
25734 @item Representation Clauses
25735 Some other Ada 95 compilers implement only the minimal set of
25736 representation clauses required by the Ada 95 reference manual. GNAT goes
25737 far beyond this minimal set, as described in the next section.
25740 @node Representation Clauses
25741 @section Representation Clauses
25744 The Ada 83 reference manual was quite vague in describing both the minimal
25745 required implementation of representation clauses, and also their precise
25746 effects. Ada 95 (and thus also Ada 2005) are much more explicit, but the
25747 minimal set of capabilities required is still quite limited.
25749 GNAT implements the full required set of capabilities in
25750 Ada 95 and Ada 2005, but also goes much further, and in particular
25751 an effort has been made to be compatible with existing Ada 83 usage to the
25752 greatest extent possible.
25754 A few cases exist in which Ada 83 compiler behavior is incompatible with
25755 the requirements in Ada 95 (and thus also Ada 2005). These are instances of
25756 intentional or accidental dependence on specific implementation dependent
25757 characteristics of these Ada 83 compilers. The following is a list of
25758 the cases most likely to arise in existing Ada 83 code.
25761 @item Implicit Packing
25762 Some Ada 83 compilers allowed a Size specification to cause implicit
25763 packing of an array or record. This could cause expensive implicit
25764 conversions for change of representation in the presence of derived
25765 types, and the Ada design intends to avoid this possibility.
25766 Subsequent AI's were issued to make it clear that such implicit
25767 change of representation in response to a Size clause is inadvisable,
25768 and this recommendation is represented explicitly in the Ada 95 (and Ada 2005)
25769 Reference Manuals as implementation advice that is followed by GNAT@.
25770 The problem will show up as an error
25771 message rejecting the size clause. The fix is simply to provide
25772 the explicit pragma @code{Pack}, or for more fine tuned control, provide
25773 a Component_Size clause.
25775 @item Meaning of Size Attribute
25776 The Size attribute in Ada 95 (and Ada 2005) for discrete types is defined as
25777 the minimal number of bits required to hold values of the type. For example,
25778 on a 32-bit machine, the size of @code{Natural} will typically be 31 and not
25779 32 (since no sign bit is required). Some Ada 83 compilers gave 31, and
25780 some 32 in this situation. This problem will usually show up as a compile
25781 time error, but not always. It is a good idea to check all uses of the
25782 'Size attribute when porting Ada 83 code. The GNAT specific attribute
25783 Object_Size can provide a useful way of duplicating the behavior of
25784 some Ada 83 compiler systems.
25786 @item Size of Access Types
25787 A common assumption in Ada 83 code is that an access type is in fact a pointer,
25788 and that therefore it will be the same size as a System.Address value. This
25789 assumption is true for GNAT in most cases with one exception. For the case of
25790 a pointer to an unconstrained array type (where the bounds may vary from one
25791 value of the access type to another), the default is to use a ``fat pointer'',
25792 which is represented as two separate pointers, one to the bounds, and one to
25793 the array. This representation has a number of advantages, including improved
25794 efficiency. However, it may cause some difficulties in porting existing Ada 83
25795 code which makes the assumption that, for example, pointers fit in 32 bits on
25796 a machine with 32-bit addressing.
25798 To get around this problem, GNAT also permits the use of ``thin pointers'' for
25799 access types in this case (where the designated type is an unconstrained array
25800 type). These thin pointers are indeed the same size as a System.Address value.
25801 To specify a thin pointer, use a size clause for the type, for example:
25803 @smallexample @c ada
25804 type X is access all String;
25805 for X'Size use Standard'Address_Size;
25809 which will cause the type X to be represented using a single pointer.
25810 When using this representation, the bounds are right behind the array.
25811 This representation is slightly less efficient, and does not allow quite
25812 such flexibility in the use of foreign pointers or in using the
25813 Unrestricted_Access attribute to create pointers to non-aliased objects.
25814 But for any standard portable use of the access type it will work in
25815 a functionally correct manner and allow porting of existing code.
25816 Note that another way of forcing a thin pointer representation
25817 is to use a component size clause for the element size in an array,
25818 or a record representation clause for an access field in a record.
25822 @c This brief section is only in the non-VMS version
25823 @c The complete chapter on HP Ada is in the VMS version
25824 @node Compatibility with HP Ada 83
25825 @section Compatibility with HP Ada 83
25828 The VMS version of GNAT fully implements all the pragmas and attributes
25829 provided by HP Ada 83, as well as providing the standard HP Ada 83
25830 libraries, including Starlet. In addition, data layouts and parameter
25831 passing conventions are highly compatible. This means that porting
25832 existing HP Ada 83 code to GNAT in VMS systems should be easier than
25833 most other porting efforts. The following are some of the most
25834 significant differences between GNAT and HP Ada 83.
25837 @item Default floating-point representation
25838 In GNAT, the default floating-point format is IEEE, whereas in HP Ada 83,
25839 it is VMS format. GNAT does implement the necessary pragmas
25840 (Long_Float, Float_Representation) for changing this default.
25843 The package System in GNAT exactly corresponds to the definition in the
25844 Ada 95 reference manual, which means that it excludes many of the
25845 HP Ada 83 extensions. However, a separate package Aux_DEC is provided
25846 that contains the additional definitions, and a special pragma,
25847 Extend_System allows this package to be treated transparently as an
25848 extension of package System.
25851 The definitions provided by Aux_DEC are exactly compatible with those
25852 in the HP Ada 83 version of System, with one exception.
25853 HP Ada provides the following declarations:
25855 @smallexample @c ada
25856 TO_ADDRESS (INTEGER)
25857 TO_ADDRESS (UNSIGNED_LONGWORD)
25858 TO_ADDRESS (@i{universal_integer})
25862 The version of TO_ADDRESS taking a @i{universal integer} argument is in fact
25863 an extension to Ada 83 not strictly compatible with the reference manual.
25864 In GNAT, we are constrained to be exactly compatible with the standard,
25865 and this means we cannot provide this capability. In HP Ada 83, the
25866 point of this definition is to deal with a call like:
25868 @smallexample @c ada
25869 TO_ADDRESS (16#12777#);
25873 Normally, according to the Ada 83 standard, one would expect this to be
25874 ambiguous, since it matches both the INTEGER and UNSIGNED_LONGWORD forms
25875 of TO_ADDRESS@. However, in HP Ada 83, there is no ambiguity, since the
25876 definition using @i{universal_integer} takes precedence.
25878 In GNAT, since the version with @i{universal_integer} cannot be supplied, it
25879 is not possible to be 100% compatible. Since there are many programs using
25880 numeric constants for the argument to TO_ADDRESS, the decision in GNAT was
25881 to change the name of the function in the UNSIGNED_LONGWORD case, so the
25882 declarations provided in the GNAT version of AUX_Dec are:
25884 @smallexample @c ada
25885 function To_Address (X : Integer) return Address;
25886 pragma Pure_Function (To_Address);
25888 function To_Address_Long (X : Unsigned_Longword)
25890 pragma Pure_Function (To_Address_Long);
25894 This means that programs using TO_ADDRESS for UNSIGNED_LONGWORD must
25895 change the name to TO_ADDRESS_LONG@.
25897 @item Task_Id values
25898 The Task_Id values assigned will be different in the two systems, and GNAT
25899 does not provide a specified value for the Task_Id of the environment task,
25900 which in GNAT is treated like any other declared task.
25904 For full details on these and other less significant compatibility issues,
25905 see appendix E of the HP publication entitled @cite{HP Ada, Technical
25906 Overview and Comparison on HP Platforms}.
25908 For GNAT running on other than VMS systems, all the HP Ada 83 pragmas and
25909 attributes are recognized, although only a subset of them can sensibly
25910 be implemented. The description of pragmas in @ref{Implementation
25911 Defined Pragmas,,, gnat_rm, GNAT Reference Manual}
25912 indicates whether or not they are applicable to non-VMS systems.
25916 @node Transitioning to 64-Bit GNAT for OpenVMS
25917 @section Transitioning to 64-Bit @value{EDITION} for OpenVMS
25920 This section is meant to assist users of pre-2006 @value{EDITION}
25921 for Alpha OpenVMS who are transitioning to 64-bit @value{EDITION},
25922 the version of the GNAT technology supplied in 2006 and later for
25923 OpenVMS on both Alpha and I64.
25926 * Introduction to transitioning::
25927 * Migration of 32 bit code::
25928 * Taking advantage of 64 bit addressing::
25929 * Technical details::
25932 @node Introduction to transitioning
25933 @subsection Introduction
25936 64-bit @value{EDITION} for Open VMS has been designed to meet
25941 Providing a full conforming implementation of Ada 95 and Ada 2005
25944 Allowing maximum backward compatibility, thus easing migration of existing
25948 Supplying a path for exploiting the full 64-bit address range
25952 Ada's strong typing semantics has made it
25953 impractical to have different 32-bit and 64-bit modes. As soon as
25954 one object could possibly be outside the 32-bit address space, this
25955 would make it necessary for the @code{System.Address} type to be 64 bits.
25956 In particular, this would cause inconsistencies if 32-bit code is
25957 called from 64-bit code that raises an exception.
25959 This issue has been resolved by always using 64-bit addressing
25960 at the system level, but allowing for automatic conversions between
25961 32-bit and 64-bit addresses where required. Thus users who
25962 do not currently require 64-bit addressing capabilities, can
25963 recompile their code with only minimal changes (and indeed
25964 if the code is written in portable Ada, with no assumptions about
25965 the size of the @code{Address} type, then no changes at all are necessary).
25967 this approach provides a simple, gradual upgrade path to future
25968 use of larger memories than available for 32-bit systems.
25969 Also, newly written applications or libraries will by default
25970 be fully compatible with future systems exploiting 64-bit
25971 addressing capabilities.
25973 @ref{Migration of 32 bit code}, will focus on porting applications
25974 that do not require more than 2 GB of
25975 addressable memory. This code will be referred to as
25976 @emph{32-bit code}.
25977 For applications intending to exploit the full 64-bit address space,
25978 @ref{Taking advantage of 64 bit addressing},
25979 will consider further changes that may be required.
25980 Such code will be referred to below as @emph{64-bit code}.
25982 @node Migration of 32 bit code
25983 @subsection Migration of 32-bit code
25987 * Access types and 32/64-bit allocation::
25988 * Unchecked conversions::
25989 * Predefined constants::
25990 * Interfacing with C::
25991 * 32/64-bit descriptors::
25992 * Experience with source compatibility::
25995 @node Address types
25996 @subsubsection Address types
25999 To solve the problem of mixing 64-bit and 32-bit addressing,
26000 while maintaining maximum backward compatibility, the following
26001 approach has been taken:
26005 @code{System.Address} always has a size of 64 bits
26006 @cindex @code{System.Address} size
26007 @cindex @code{Address} size
26010 @code{System.Short_Address} is a 32-bit subtype of @code{System.Address}
26011 @cindex @code{System.Short_Address} size
26012 @cindex @code{Short_Address} size
26016 Since @code{System.Short_Address} is a subtype of @code{System.Address},
26017 a @code{Short_Address}
26018 may be used where an @code{Address} is required, and vice versa, without
26019 needing explicit type conversions.
26020 By virtue of the Open VMS parameter passing conventions,
26022 and exported subprograms that have 32-bit address parameters are
26023 compatible with those that have 64-bit address parameters.
26024 (See @ref{Making code 64 bit clean} for details.)
26026 The areas that may need attention are those where record types have
26027 been defined that contain components of the type @code{System.Address}, and
26028 where objects of this type are passed to code expecting a record layout with
26031 Different compilers on different platforms cannot be
26032 expected to represent the same type in the same way,
26033 since alignment constraints
26034 and other system-dependent properties affect the compiler's decision.
26035 For that reason, Ada code
26036 generally uses representation clauses to specify the expected
26037 layout where required.
26039 If such a representation clause uses 32 bits for a component having
26040 the type @code{System.Address}, 64-bit @value{EDITION} for OpenVMS
26041 will detect that error and produce a specific diagnostic message.
26042 The developer should then determine whether the representation
26043 should be 64 bits or not and make either of two changes:
26044 change the size to 64 bits and leave the type as @code{System.Address}, or
26045 leave the size as 32 bits and change the type to @code{System.Short_Address}.
26046 Since @code{Short_Address} is a subtype of @code{Address}, no changes are
26047 required in any code setting or accessing the field; the compiler will
26048 automatically perform any needed conversions between address
26051 @node Access types and 32/64-bit allocation
26052 @subsubsection Access types and 32/64-bit allocation
26053 @cindex 32-bit allocation
26054 @cindex 64-bit allocation
26057 By default, objects designated by access values are always allocated in
26058 the 64-bit address space, and access values themselves are represented
26059 in 64 bits. If these defaults are not appropriate, and 32-bit allocation
26060 is required (for example if the address of an allocated object is assigned
26061 to a @code{Short_Address} variable), then several alternatives are available:
26065 A pool-specific access type (ie, an @w{Ada 83} access type, whose
26066 definition is @code{access T} versus @code{access all T} or
26067 @code{access constant T}), may be declared with a @code{'Size} representation
26068 clause that establishes the size as 32 bits.
26069 In such circumstances allocations for that type will
26070 be from the 32-bit heap. Such a clause is not permitted
26071 for a general access type (declared with @code{access all} or
26072 @code{access constant}) as values of such types must be able to refer
26073 to any object of the designated type, including objects residing outside
26074 the 32-bit address range. Existing @w{Ada 83} code will not contain such
26075 type definitions, however, since general access types were introduced
26079 Switches for @command{GNAT BIND} control whether the internal GNAT
26080 allocation routine @code{__gnat_malloc} uses 64-bit or 32-bit allocations.
26081 @cindex @code{__gnat_malloc}
26082 The switches are respectively @option{-H64} (the default) and
26084 @cindex @option{-H32} (@command{gnatbind})
26085 @cindex @option{-H64} (@command{gnatbind})
26088 The environment variable (logical name) @code{GNAT$NO_MALLOC_64}
26089 @cindex @code{GNAT$NO_MALLOC_64} environment variable
26090 may be used to force @code{__gnat_malloc} to use 32-bit allocation.
26091 If this variable is left
26092 undefined, or defined as @code{"DISABLE"}, @code{"FALSE"}, or @code{"0"},
26093 then the default (64-bit) allocation is used.
26094 If defined as @code{"ENABLE"}, @code{"TRUE"}, or @code{"1"},
26095 then 32-bit allocation is used. The gnatbind qualifiers described above
26096 override this logical name.
26099 A ^gcc switch^gcc switch^ for OpenVMS, @option{-mno-malloc64}, operates
26100 @cindex @option{-mno-malloc64} (^gcc^gcc^)
26101 at a low level to convert explicit calls to @code{malloc} and related
26102 functions from the C run-time library so that they perform allocations
26103 in the 32-bit heap.
26104 Since all internal allocations from GNAT use @code{__gnat_malloc},
26105 this switch is not required unless the program makes explicit calls on
26106 @code{malloc} (or related functions) from interfaced C code.
26110 @node Unchecked conversions
26111 @subsubsection Unchecked conversions
26114 In the case of an @code{Unchecked_Conversion} where the source type is a
26115 64-bit access type or the type @code{System.Address}, and the target
26116 type is a 32-bit type, the compiler will generate a warning.
26117 Even though the generated code will still perform the required
26118 conversions, it is highly recommended in these cases to use
26119 respectively a 32-bit access type or @code{System.Short_Address}
26120 as the source type.
26122 @node Predefined constants
26123 @subsubsection Predefined constants
26126 The following table shows the correspondence between pre-2006 versions of
26127 @value{EDITION} on Alpha OpenVMS (``Old'') and 64-bit @value{EDITION}
26130 @multitable {@code{System.Short_Memory_Size}} {2**32} {2**64}
26131 @item @b{Constant} @tab @b{Old} @tab @b{New}
26132 @item @code{System.Word_Size} @tab 32 @tab 64
26133 @item @code{System.Memory_Size} @tab 2**32 @tab 2**64
26134 @item @code{System.Short_Memory_Size} @tab 2**32 @tab 2**32
26135 @item @code{System.Address_Size} @tab 32 @tab 64
26139 If you need to refer to the specific
26140 memory size of a 32-bit implementation, instead of the
26141 actual memory size, use @code{System.Short_Memory_Size}
26142 rather than @code{System.Memory_Size}.
26143 Similarly, references to @code{System.Address_Size} may need
26144 to be replaced by @code{System.Short_Address'Size}.
26145 The program @command{gnatfind} may be useful for locating
26146 references to the above constants, so that you can verify that they
26149 @node Interfacing with C
26150 @subsubsection Interfacing with C
26153 In order to minimize the impact of the transition to 64-bit addresses on
26154 legacy programs, some fundamental types in the @code{Interfaces.C}
26155 package hierarchy continue to be represented in 32 bits.
26156 These types are: @code{ptrdiff_t}, @code{size_t}, and @code{chars_ptr}.
26157 This eases integration with the default HP C layout choices, for example
26158 as found in the system routines in @code{DECC$SHR.EXE}.
26159 Because of this implementation choice, the type fully compatible with
26160 @code{chars_ptr} is now @code{Short_Address} and not @code{Address}.
26161 Depending on the context the compiler will issue a
26162 warning or an error when type @code{Address} is used, alerting the user to a
26163 potential problem. Otherwise 32-bit programs that use
26164 @code{Interfaces.C} should normally not require code modifications
26166 The other issue arising with C interfacing concerns pragma @code{Convention}.
26167 For VMS 64-bit systems, there is an issue of the appropriate default size
26168 of C convention pointers in the absence of an explicit size clause. The HP
26169 C compiler can choose either 32 or 64 bits depending on compiler options.
26170 GNAT chooses 32-bits rather than 64-bits in the default case where no size
26171 clause is given. This proves a better choice for porting 32-bit legacy
26172 applications. In order to have a 64-bit representation, it is necessary to
26173 specify a size representation clause. For example:
26175 @smallexample @c ada
26176 type int_star is access Interfaces.C.int;
26177 pragma Convention(C, int_star);
26178 for int_star'Size use 64; -- Necessary to get 64 and not 32 bits
26181 @node 32/64-bit descriptors
26182 @subsubsection 32/64-bit descriptors
26185 By default, GNAT uses a 64-bit descriptor mechanism. For an imported
26186 subprogram (i.e., a subprogram identified by pragma @code{Import_Function},
26187 @code{Import_Procedure}, or @code{Import_Valued_Procedure}) that specifies
26188 @code{Short_Descriptor} as its mechanism, a 32-bit descriptor is used.
26189 @cindex @code{Short_Descriptor} mechanism for imported subprograms
26191 If the configuration pragma @code{Short_Descriptors} is supplied, then
26192 all descriptors will be 32 bits.
26193 @cindex pragma @code{Short_Descriptors}
26195 @node Experience with source compatibility
26196 @subsubsection Experience with source compatibility
26199 The Security Server and STARLET on I64 provide an interesting ``test case''
26200 for source compatibility issues, since it is in such system code
26201 where assumptions about @code{Address} size might be expected to occur.
26202 Indeed, there were a small number of occasions in the Security Server
26203 file @file{jibdef.ads}
26204 where a representation clause for a record type specified
26205 32 bits for a component of type @code{Address}.
26206 All of these errors were detected by the compiler.
26207 The repair was obvious and immediate; to simply replace @code{Address} by
26208 @code{Short_Address}.
26210 In the case of STARLET, there were several record types that should
26211 have had representation clauses but did not. In these record types
26212 there was an implicit assumption that an @code{Address} value occupied
26214 These compiled without error, but their usage resulted in run-time error
26215 returns from STARLET system calls.
26216 Future GNAT technology enhancements may include a tool that detects and flags
26217 these sorts of potential source code porting problems.
26219 @c ****************************************
26220 @node Taking advantage of 64 bit addressing
26221 @subsection Taking advantage of 64-bit addressing
26224 * Making code 64 bit clean::
26225 * Allocating memory from the 64 bit storage pool::
26226 * Restrictions on use of 64 bit objects::
26227 * STARLET and other predefined libraries::
26230 @node Making code 64 bit clean
26231 @subsubsection Making code 64-bit clean
26234 In order to prevent problems that may occur when (parts of) a
26235 system start using memory outside the 32-bit address range,
26236 we recommend some additional guidelines:
26240 For imported subprograms that take parameters of the
26241 type @code{System.Address}, ensure that these subprograms can
26242 indeed handle 64-bit addresses. If not, or when in doubt,
26243 change the subprogram declaration to specify
26244 @code{System.Short_Address} instead.
26247 Resolve all warnings related to size mismatches in
26248 unchecked conversions. Failing to do so causes
26249 erroneous execution if the source object is outside
26250 the 32-bit address space.
26253 (optional) Explicitly use the 32-bit storage pool
26254 for access types used in a 32-bit context, or use
26255 generic access types where possible
26256 (@pxref{Restrictions on use of 64 bit objects}).
26260 If these rules are followed, the compiler will automatically insert
26261 any necessary checks to ensure that no addresses or access values
26262 passed to 32-bit code ever refer to objects outside the 32-bit
26264 Any attempt to do this will raise @code{Constraint_Error}.
26266 @node Allocating memory from the 64 bit storage pool
26267 @subsubsection Allocating memory from the 64-bit storage pool
26270 By default, all allocations -- for both pool-specific and general
26271 access types -- use the 64-bit storage pool. To override
26272 this default, for an individual access type or globally, see
26273 @ref{Access types and 32/64-bit allocation}.
26275 @node Restrictions on use of 64 bit objects
26276 @subsubsection Restrictions on use of 64-bit objects
26279 Taking the address of an object allocated from a 64-bit storage pool,
26280 and then passing this address to a subprogram expecting
26281 @code{System.Short_Address},
26282 or assigning it to a variable of type @code{Short_Address}, will cause
26283 @code{Constraint_Error} to be raised. In case the code is not 64-bit clean
26284 (@pxref{Making code 64 bit clean}), or checks are suppressed,
26285 no exception is raised and execution
26286 will become erroneous.
26288 @node STARLET and other predefined libraries
26289 @subsubsection STARLET and other predefined libraries
26292 All code that comes as part of GNAT is 64-bit clean, but the
26293 restrictions given in @ref{Restrictions on use of 64 bit objects},
26294 still apply. Look at the package
26295 specs to see in which contexts objects allocated
26296 in 64-bit address space are acceptable.
26298 @node Technical details
26299 @subsection Technical details
26302 64-bit @value{EDITION} for Open VMS takes advantage of the freedom given in the
26303 Ada standard with respect to the type of @code{System.Address}. Previous
26304 versions of GNAT Pro have defined this type as private and implemented it as a
26307 In order to allow defining @code{System.Short_Address} as a proper subtype,
26308 and to match the implicit sign extension in parameter passing,
26309 in 64-bit @value{EDITION} for Open VMS, @code{System.Address} is defined as a
26310 visible (i.e., non-private) integer type.
26311 Standard operations on the type, such as the binary operators ``+'', ``-'',
26312 etc., that take @code{Address} operands and return an @code{Address} result,
26313 have been hidden by declaring these
26314 @code{abstract}, a feature introduced in Ada 95 that helps avoid the potential
26315 ambiguities that would otherwise result from overloading.
26316 (Note that, although @code{Address} is a visible integer type,
26317 good programming practice dictates against exploiting the type's
26318 integer properties such as literals, since this will compromise
26321 Defining @code{Address} as a visible integer type helps achieve
26322 maximum compatibility for existing Ada code,
26323 without sacrificing the capabilities of the 64-bit architecture.
26326 @c ************************************************
26328 @node Microsoft Windows Topics
26329 @appendix Microsoft Windows Topics
26335 This chapter describes topics that are specific to the Microsoft Windows
26336 platforms (NT, 2000, and XP Professional).
26339 * Using GNAT on Windows::
26340 * Using a network installation of GNAT::
26341 * CONSOLE and WINDOWS subsystems::
26342 * Temporary Files::
26343 * Mixed-Language Programming on Windows::
26344 * Windows Calling Conventions::
26345 * Introduction to Dynamic Link Libraries (DLLs)::
26346 * Using DLLs with GNAT::
26347 * Building DLLs with GNAT Project files::
26348 * Building DLLs with GNAT::
26349 * Building DLLs with gnatdll::
26350 * GNAT and Windows Resources::
26351 * Debugging a DLL::
26352 * Setting Stack Size from gnatlink::
26353 * Setting Heap Size from gnatlink::
26356 @node Using GNAT on Windows
26357 @section Using GNAT on Windows
26360 One of the strengths of the GNAT technology is that its tool set
26361 (@command{gcc}, @command{gnatbind}, @command{gnatlink}, @command{gnatmake}, the
26362 @code{gdb} debugger, etc.) is used in the same way regardless of the
26365 On Windows this tool set is complemented by a number of Microsoft-specific
26366 tools that have been provided to facilitate interoperability with Windows
26367 when this is required. With these tools:
26372 You can build applications using the @code{CONSOLE} or @code{WINDOWS}
26376 You can use any Dynamically Linked Library (DLL) in your Ada code (both
26377 relocatable and non-relocatable DLLs are supported).
26380 You can build Ada DLLs for use in other applications. These applications
26381 can be written in a language other than Ada (e.g., C, C++, etc). Again both
26382 relocatable and non-relocatable Ada DLLs are supported.
26385 You can include Windows resources in your Ada application.
26388 You can use or create COM/DCOM objects.
26392 Immediately below are listed all known general GNAT-for-Windows restrictions.
26393 Other restrictions about specific features like Windows Resources and DLLs
26394 are listed in separate sections below.
26399 It is not possible to use @code{GetLastError} and @code{SetLastError}
26400 when tasking, protected records, or exceptions are used. In these
26401 cases, in order to implement Ada semantics, the GNAT run-time system
26402 calls certain Win32 routines that set the last error variable to 0 upon
26403 success. It should be possible to use @code{GetLastError} and
26404 @code{SetLastError} when tasking, protected record, and exception
26405 features are not used, but it is not guaranteed to work.
26408 It is not possible to link against Microsoft libraries except for
26409 import libraries. Interfacing must be done by the mean of DLLs.
26412 When the compilation environment is located on FAT32 drives, users may
26413 experience recompilations of the source files that have not changed if
26414 Daylight Saving Time (DST) state has changed since the last time files
26415 were compiled. NTFS drives do not have this problem.
26418 No components of the GNAT toolset use any entries in the Windows
26419 registry. The only entries that can be created are file associations and
26420 PATH settings, provided the user has chosen to create them at installation
26421 time, as well as some minimal book-keeping information needed to correctly
26422 uninstall or integrate different GNAT products.
26425 @node Using a network installation of GNAT
26426 @section Using a network installation of GNAT
26429 Make sure the system on which GNAT is installed is accessible from the
26430 current machine, i.e., the install location is shared over the network.
26431 Shared resources are accessed on Windows by means of UNC paths, which
26432 have the format @code{\\server\sharename\path}
26434 In order to use such a network installation, simply add the UNC path of the
26435 @file{bin} directory of your GNAT installation in front of your PATH. For
26436 example, if GNAT is installed in @file{\GNAT} directory of a share location
26437 called @file{c-drive} on a machine @file{LOKI}, the following command will
26440 @code{@ @ @ path \\loki\c-drive\gnat\bin;%path%}
26442 Be aware that every compilation using the network installation results in the
26443 transfer of large amounts of data across the network and will likely cause
26444 serious performance penalty.
26446 @node CONSOLE and WINDOWS subsystems
26447 @section CONSOLE and WINDOWS subsystems
26448 @cindex CONSOLE Subsystem
26449 @cindex WINDOWS Subsystem
26453 There are two main subsystems under Windows. The @code{CONSOLE} subsystem
26454 (which is the default subsystem) will always create a console when
26455 launching the application. This is not something desirable when the
26456 application has a Windows GUI. To get rid of this console the
26457 application must be using the @code{WINDOWS} subsystem. To do so
26458 the @option{-mwindows} linker option must be specified.
26461 $ gnatmake winprog -largs -mwindows
26464 @node Temporary Files
26465 @section Temporary Files
26466 @cindex Temporary files
26469 It is possible to control where temporary files gets created by setting
26470 the @env{TMP} environment variable. The file will be created:
26473 @item Under the directory pointed to by the @env{TMP} environment variable if
26474 this directory exists.
26476 @item Under @file{c:\temp}, if the @env{TMP} environment variable is not
26477 set (or not pointing to a directory) and if this directory exists.
26479 @item Under the current working directory otherwise.
26483 This allows you to determine exactly where the temporary
26484 file will be created. This is particularly useful in networked
26485 environments where you may not have write access to some
26488 @node Mixed-Language Programming on Windows
26489 @section Mixed-Language Programming on Windows
26492 Developing pure Ada applications on Windows is no different than on
26493 other GNAT-supported platforms. However, when developing or porting an
26494 application that contains a mix of Ada and C/C++, the choice of your
26495 Windows C/C++ development environment conditions your overall
26496 interoperability strategy.
26498 If you use @command{gcc} to compile the non-Ada part of your application,
26499 there are no Windows-specific restrictions that affect the overall
26500 interoperability with your Ada code. If you do want to use the
26501 Microsoft tools for your non-Ada code, you have two choices:
26505 Encapsulate your non-Ada code in a DLL to be linked with your Ada
26506 application. In this case, use the Microsoft or whatever environment to
26507 build the DLL and use GNAT to build your executable
26508 (@pxref{Using DLLs with GNAT}).
26511 Or you can encapsulate your Ada code in a DLL to be linked with the
26512 other part of your application. In this case, use GNAT to build the DLL
26513 (@pxref{Building DLLs with GNAT Project files}) and use the Microsoft
26514 or whatever environment to build your executable.
26517 @node Windows Calling Conventions
26518 @section Windows Calling Conventions
26522 This section pertain only to Win32. On Win64 there is a single native
26523 calling convention. All convention specifiers are ignored on this
26527 * C Calling Convention::
26528 * Stdcall Calling Convention::
26529 * Win32 Calling Convention::
26530 * DLL Calling Convention::
26534 When a subprogram @code{F} (caller) calls a subprogram @code{G}
26535 (callee), there are several ways to push @code{G}'s parameters on the
26536 stack and there are several possible scenarios to clean up the stack
26537 upon @code{G}'s return. A calling convention is an agreed upon software
26538 protocol whereby the responsibilities between the caller (@code{F}) and
26539 the callee (@code{G}) are clearly defined. Several calling conventions
26540 are available for Windows:
26544 @code{C} (Microsoft defined)
26547 @code{Stdcall} (Microsoft defined)
26550 @code{Win32} (GNAT specific)
26553 @code{DLL} (GNAT specific)
26556 @node C Calling Convention
26557 @subsection @code{C} Calling Convention
26560 This is the default calling convention used when interfacing to C/C++
26561 routines compiled with either @command{gcc} or Microsoft Visual C++.
26563 In the @code{C} calling convention subprogram parameters are pushed on the
26564 stack by the caller from right to left. The caller itself is in charge of
26565 cleaning up the stack after the call. In addition, the name of a routine
26566 with @code{C} calling convention is mangled by adding a leading underscore.
26568 The name to use on the Ada side when importing (or exporting) a routine
26569 with @code{C} calling convention is the name of the routine. For
26570 instance the C function:
26573 int get_val (long);
26577 should be imported from Ada as follows:
26579 @smallexample @c ada
26581 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
26582 pragma Import (C, Get_Val, External_Name => "get_val");
26587 Note that in this particular case the @code{External_Name} parameter could
26588 have been omitted since, when missing, this parameter is taken to be the
26589 name of the Ada entity in lower case. When the @code{Link_Name} parameter
26590 is missing, as in the above example, this parameter is set to be the
26591 @code{External_Name} with a leading underscore.
26593 When importing a variable defined in C, you should always use the @code{C}
26594 calling convention unless the object containing the variable is part of a
26595 DLL (in which case you should use the @code{Stdcall} calling
26596 convention, @pxref{Stdcall Calling Convention}).
26598 @node Stdcall Calling Convention
26599 @subsection @code{Stdcall} Calling Convention
26602 This convention, which was the calling convention used for Pascal
26603 programs, is used by Microsoft for all the routines in the Win32 API for
26604 efficiency reasons. It must be used to import any routine for which this
26605 convention was specified.
26607 In the @code{Stdcall} calling convention subprogram parameters are pushed
26608 on the stack by the caller from right to left. The callee (and not the
26609 caller) is in charge of cleaning the stack on routine exit. In addition,
26610 the name of a routine with @code{Stdcall} calling convention is mangled by
26611 adding a leading underscore (as for the @code{C} calling convention) and a
26612 trailing @code{@@}@code{@var{nn}}, where @var{nn} is the overall size (in
26613 bytes) of the parameters passed to the routine.
26615 The name to use on the Ada side when importing a C routine with a
26616 @code{Stdcall} calling convention is the name of the C routine. The leading
26617 underscore and trailing @code{@@}@code{@var{nn}} are added automatically by
26618 the compiler. For instance the Win32 function:
26621 @b{APIENTRY} int get_val (long);
26625 should be imported from Ada as follows:
26627 @smallexample @c ada
26629 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
26630 pragma Import (Stdcall, Get_Val);
26631 -- On the x86 a long is 4 bytes, so the Link_Name is "_get_val@@4"
26636 As for the @code{C} calling convention, when the @code{External_Name}
26637 parameter is missing, it is taken to be the name of the Ada entity in lower
26638 case. If instead of writing the above import pragma you write:
26640 @smallexample @c ada
26642 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
26643 pragma Import (Stdcall, Get_Val, External_Name => "retrieve_val");
26648 then the imported routine is @code{_retrieve_val@@4}. However, if instead
26649 of specifying the @code{External_Name} parameter you specify the
26650 @code{Link_Name} as in the following example:
26652 @smallexample @c ada
26654 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
26655 pragma Import (Stdcall, Get_Val, Link_Name => "retrieve_val");
26660 then the imported routine is @code{retrieve_val}, that is, there is no
26661 decoration at all. No leading underscore and no Stdcall suffix
26662 @code{@@}@code{@var{nn}}.
26665 This is especially important as in some special cases a DLL's entry
26666 point name lacks a trailing @code{@@}@code{@var{nn}} while the exported
26667 name generated for a call has it.
26670 It is also possible to import variables defined in a DLL by using an
26671 import pragma for a variable. As an example, if a DLL contains a
26672 variable defined as:
26679 then, to access this variable from Ada you should write:
26681 @smallexample @c ada
26683 My_Var : Interfaces.C.int;
26684 pragma Import (Stdcall, My_Var);
26689 Note that to ease building cross-platform bindings this convention
26690 will be handled as a @code{C} calling convention on non-Windows platforms.
26692 @node Win32 Calling Convention
26693 @subsection @code{Win32} Calling Convention
26696 This convention, which is GNAT-specific is fully equivalent to the
26697 @code{Stdcall} calling convention described above.
26699 @node DLL Calling Convention
26700 @subsection @code{DLL} Calling Convention
26703 This convention, which is GNAT-specific is fully equivalent to the
26704 @code{Stdcall} calling convention described above.
26706 @node Introduction to Dynamic Link Libraries (DLLs)
26707 @section Introduction to Dynamic Link Libraries (DLLs)
26711 A Dynamically Linked Library (DLL) is a library that can be shared by
26712 several applications running under Windows. A DLL can contain any number of
26713 routines and variables.
26715 One advantage of DLLs is that you can change and enhance them without
26716 forcing all the applications that depend on them to be relinked or
26717 recompiled. However, you should be aware than all calls to DLL routines are
26718 slower since, as you will understand below, such calls are indirect.
26720 To illustrate the remainder of this section, suppose that an application
26721 wants to use the services of a DLL @file{API.dll}. To use the services
26722 provided by @file{API.dll} you must statically link against the DLL or
26723 an import library which contains a jump table with an entry for each
26724 routine and variable exported by the DLL. In the Microsoft world this
26725 import library is called @file{API.lib}. When using GNAT this import
26726 library is called either @file{libAPI.dll.a}, @file{libapi.dll.a},
26727 @file{libAPI.a} or @file{libapi.a} (names are case insensitive).
26729 After you have linked your application with the DLL or the import library
26730 and you run your application, here is what happens:
26734 Your application is loaded into memory.
26737 The DLL @file{API.dll} is mapped into the address space of your
26738 application. This means that:
26742 The DLL will use the stack of the calling thread.
26745 The DLL will use the virtual address space of the calling process.
26748 The DLL will allocate memory from the virtual address space of the calling
26752 Handles (pointers) can be safely exchanged between routines in the DLL
26753 routines and routines in the application using the DLL.
26757 The entries in the jump table (from the import library @file{libAPI.dll.a}
26758 or @file{API.lib} or automatically created when linking against a DLL)
26759 which is part of your application are initialized with the addresses
26760 of the routines and variables in @file{API.dll}.
26763 If present in @file{API.dll}, routines @code{DllMain} or
26764 @code{DllMainCRTStartup} are invoked. These routines typically contain
26765 the initialization code needed for the well-being of the routines and
26766 variables exported by the DLL.
26770 There is an additional point which is worth mentioning. In the Windows
26771 world there are two kind of DLLs: relocatable and non-relocatable
26772 DLLs. Non-relocatable DLLs can only be loaded at a very specific address
26773 in the target application address space. If the addresses of two
26774 non-relocatable DLLs overlap and these happen to be used by the same
26775 application, a conflict will occur and the application will run
26776 incorrectly. Hence, when possible, it is always preferable to use and
26777 build relocatable DLLs. Both relocatable and non-relocatable DLLs are
26778 supported by GNAT. Note that the @option{-s} linker option (see GNU Linker
26779 User's Guide) removes the debugging symbols from the DLL but the DLL can
26780 still be relocated.
26782 As a side note, an interesting difference between Microsoft DLLs and
26783 Unix shared libraries, is the fact that on most Unix systems all public
26784 routines are exported by default in a Unix shared library, while under
26785 Windows it is possible (but not required) to list exported routines in
26786 a definition file (@pxref{The Definition File}).
26788 @node Using DLLs with GNAT
26789 @section Using DLLs with GNAT
26792 * Creating an Ada Spec for the DLL Services::
26793 * Creating an Import Library::
26797 To use the services of a DLL, say @file{API.dll}, in your Ada application
26802 The Ada spec for the routines and/or variables you want to access in
26803 @file{API.dll}. If not available this Ada spec must be built from the C/C++
26804 header files provided with the DLL.
26807 The import library (@file{libAPI.dll.a} or @file{API.lib}). As previously
26808 mentioned an import library is a statically linked library containing the
26809 import table which will be filled at load time to point to the actual
26810 @file{API.dll} routines. Sometimes you don't have an import library for the
26811 DLL you want to use. The following sections will explain how to build
26812 one. Note that this is optional.
26815 The actual DLL, @file{API.dll}.
26819 Once you have all the above, to compile an Ada application that uses the
26820 services of @file{API.dll} and whose main subprogram is @code{My_Ada_App},
26821 you simply issue the command
26824 $ gnatmake my_ada_app -largs -lAPI
26828 The argument @option{-largs -lAPI} at the end of the @command{gnatmake} command
26829 tells the GNAT linker to look for an import library. The linker will
26830 look for a library name in this specific order:
26833 @item @file{libAPI.dll.a}
26834 @item @file{API.dll.a}
26835 @item @file{libAPI.a}
26836 @item @file{API.lib}
26837 @item @file{libAPI.dll}
26838 @item @file{API.dll}
26841 The first three are the GNU style import libraries. The third is the
26842 Microsoft style import libraries. The last two are the DLL themself.
26844 Note that if the Ada package spec for @file{API.dll} contains the
26847 @smallexample @c ada
26848 pragma Linker_Options ("-lAPI");
26852 you do not have to add @option{-largs -lAPI} at the end of the
26853 @command{gnatmake} command.
26855 If any one of the items above is missing you will have to create it
26856 yourself. The following sections explain how to do so using as an
26857 example a fictitious DLL called @file{API.dll}.
26859 @node Creating an Ada Spec for the DLL Services
26860 @subsection Creating an Ada Spec for the DLL Services
26863 A DLL typically comes with a C/C++ header file which provides the
26864 definitions of the routines and variables exported by the DLL. The Ada
26865 equivalent of this header file is a package spec that contains definitions
26866 for the imported entities. If the DLL you intend to use does not come with
26867 an Ada spec you have to generate one such spec yourself. For example if
26868 the header file of @file{API.dll} is a file @file{api.h} containing the
26869 following two definitions:
26881 then the equivalent Ada spec could be:
26883 @smallexample @c ada
26886 with Interfaces.C.Strings;
26891 function Get (Str : C.Strings.Chars_Ptr) return C.int;
26894 pragma Import (C, Get);
26895 pragma Import (DLL, Some_Var);
26902 Note that a variable is
26903 @strong{always imported with a DLL convention}. A function
26904 can have @code{C} or @code{Stdcall} convention.
26905 (@pxref{Windows Calling Conventions}).
26907 @node Creating an Import Library
26908 @subsection Creating an Import Library
26909 @cindex Import library
26912 * The Definition File::
26913 * GNAT-Style Import Library::
26914 * Microsoft-Style Import Library::
26918 If a Microsoft-style import library @file{API.lib} or a GNAT-style
26919 import library @file{libAPI.dll.a} or @file{libAPI.a} is available
26920 with @file{API.dll} you can skip this section. You can also skip this
26921 section if @file{API.dll} or @file{libAPI.dll} is built with GNU tools
26922 as in this case it is possible to link directly against the
26923 DLL. Otherwise read on.
26925 @node The Definition File
26926 @subsubsection The Definition File
26927 @cindex Definition file
26931 As previously mentioned, and unlike Unix systems, the list of symbols
26932 that are exported from a DLL must be provided explicitly in Windows.
26933 The main goal of a definition file is precisely that: list the symbols
26934 exported by a DLL. A definition file (usually a file with a @code{.def}
26935 suffix) has the following structure:
26940 @r{[}LIBRARY @var{name}@r{]}
26941 @r{[}DESCRIPTION @var{string}@r{]}
26951 @item LIBRARY @var{name}
26952 This section, which is optional, gives the name of the DLL.
26954 @item DESCRIPTION @var{string}
26955 This section, which is optional, gives a description string that will be
26956 embedded in the import library.
26959 This section gives the list of exported symbols (procedures, functions or
26960 variables). For instance in the case of @file{API.dll} the @code{EXPORTS}
26961 section of @file{API.def} looks like:
26975 Note that you must specify the correct suffix (@code{@@}@code{@var{nn}})
26976 (@pxref{Windows Calling Conventions}) for a Stdcall
26977 calling convention function in the exported symbols list.
26980 There can actually be other sections in a definition file, but these
26981 sections are not relevant to the discussion at hand.
26983 @node GNAT-Style Import Library
26984 @subsubsection GNAT-Style Import Library
26987 To create a static import library from @file{API.dll} with the GNAT tools
26988 you should proceed as follows:
26992 Create the definition file @file{API.def} (@pxref{The Definition File}).
26993 For that use the @code{dll2def} tool as follows:
26996 $ dll2def API.dll > API.def
27000 @code{dll2def} is a very simple tool: it takes as input a DLL and prints
27001 to standard output the list of entry points in the DLL. Note that if
27002 some routines in the DLL have the @code{Stdcall} convention
27003 (@pxref{Windows Calling Conventions}) with stripped @code{@@}@var{nn}
27004 suffix then you'll have to edit @file{api.def} to add it, and specify
27005 @option{-k} to @command{gnatdll} when creating the import library.
27008 Here are some hints to find the right @code{@@}@var{nn} suffix.
27012 If you have the Microsoft import library (.lib), it is possible to get
27013 the right symbols by using Microsoft @code{dumpbin} tool (see the
27014 corresponding Microsoft documentation for further details).
27017 $ dumpbin /exports api.lib
27021 If you have a message about a missing symbol at link time the compiler
27022 tells you what symbol is expected. You just have to go back to the
27023 definition file and add the right suffix.
27027 Build the import library @code{libAPI.dll.a}, using @code{gnatdll}
27028 (@pxref{Using gnatdll}) as follows:
27031 $ gnatdll -e API.def -d API.dll
27035 @code{gnatdll} takes as input a definition file @file{API.def} and the
27036 name of the DLL containing the services listed in the definition file
27037 @file{API.dll}. The name of the static import library generated is
27038 computed from the name of the definition file as follows: if the
27039 definition file name is @var{xyz}@code{.def}, the import library name will
27040 be @code{lib}@var{xyz}@code{.a}. Note that in the previous example option
27041 @option{-e} could have been removed because the name of the definition
27042 file (before the ``@code{.def}'' suffix) is the same as the name of the
27043 DLL (@pxref{Using gnatdll} for more information about @code{gnatdll}).
27046 @node Microsoft-Style Import Library
27047 @subsubsection Microsoft-Style Import Library
27050 With GNAT you can either use a GNAT-style or Microsoft-style import
27051 library. A Microsoft import library is needed only if you plan to make an
27052 Ada DLL available to applications developed with Microsoft
27053 tools (@pxref{Mixed-Language Programming on Windows}).
27055 To create a Microsoft-style import library for @file{API.dll} you
27056 should proceed as follows:
27060 Create the definition file @file{API.def} from the DLL. For this use either
27061 the @code{dll2def} tool as described above or the Microsoft @code{dumpbin}
27062 tool (see the corresponding Microsoft documentation for further details).
27065 Build the actual import library using Microsoft's @code{lib} utility:
27068 $ lib -machine:IX86 -def:API.def -out:API.lib
27072 If you use the above command the definition file @file{API.def} must
27073 contain a line giving the name of the DLL:
27080 See the Microsoft documentation for further details about the usage of
27084 @node Building DLLs with GNAT Project files
27085 @section Building DLLs with GNAT Project files
27086 @cindex DLLs, building
27089 There is nothing specific to Windows in the build process.
27090 @pxref{Library Projects}.
27093 Due to a system limitation, it is not possible under Windows to create threads
27094 when inside the @code{DllMain} routine which is used for auto-initialization
27095 of shared libraries, so it is not possible to have library level tasks in SALs.
27097 @node Building DLLs with GNAT
27098 @section Building DLLs with GNAT
27099 @cindex DLLs, building
27102 This section explain how to build DLLs using the GNAT built-in DLL
27103 support. With the following procedure it is straight forward to build
27104 and use DLLs with GNAT.
27108 @item building object files
27110 The first step is to build all objects files that are to be included
27111 into the DLL. This is done by using the standard @command{gnatmake} tool.
27113 @item building the DLL
27115 To build the DLL you must use @command{gcc}'s @option{-shared} and
27116 @option{-shared-libgcc} options. It is quite simple to use this method:
27119 $ gcc -shared -shared-libgcc -o api.dll obj1.o obj2.o @dots{}
27122 It is important to note that in this case all symbols found in the
27123 object files are automatically exported. It is possible to restrict
27124 the set of symbols to export by passing to @command{gcc} a definition
27125 file, @pxref{The Definition File}. For example:
27128 $ gcc -shared -shared-libgcc -o api.dll api.def obj1.o obj2.o @dots{}
27131 If you use a definition file you must export the elaboration procedures
27132 for every package that required one. Elaboration procedures are named
27133 using the package name followed by "_E".
27135 @item preparing DLL to be used
27137 For the DLL to be used by client programs the bodies must be hidden
27138 from it and the .ali set with read-only attribute. This is very important
27139 otherwise GNAT will recompile all packages and will not actually use
27140 the code in the DLL. For example:
27144 $ copy *.ads *.ali api.dll apilib
27145 $ attrib +R apilib\*.ali
27150 At this point it is possible to use the DLL by directly linking
27151 against it. Note that you must use the GNAT shared runtime when using
27152 GNAT shared libraries. This is achieved by using @option{-shared} binder's
27156 $ gnatmake main -Iapilib -bargs -shared -largs -Lapilib -lAPI
27159 @node Building DLLs with gnatdll
27160 @section Building DLLs with gnatdll
27161 @cindex DLLs, building
27164 * Limitations When Using Ada DLLs from Ada::
27165 * Exporting Ada Entities::
27166 * Ada DLLs and Elaboration::
27167 * Ada DLLs and Finalization::
27168 * Creating a Spec for Ada DLLs::
27169 * Creating the Definition File::
27174 Note that it is preferred to use GNAT Project files
27175 (@pxref{Building DLLs with GNAT Project files}) or the built-in GNAT
27176 DLL support (@pxref{Building DLLs with GNAT}) or to build DLLs.
27178 This section explains how to build DLLs containing Ada code using
27179 @code{gnatdll}. These DLLs will be referred to as Ada DLLs in the
27180 remainder of this section.
27182 The steps required to build an Ada DLL that is to be used by Ada as well as
27183 non-Ada applications are as follows:
27187 You need to mark each Ada @i{entity} exported by the DLL with a @code{C} or
27188 @code{Stdcall} calling convention to avoid any Ada name mangling for the
27189 entities exported by the DLL (@pxref{Exporting Ada Entities}). You can
27190 skip this step if you plan to use the Ada DLL only from Ada applications.
27193 Your Ada code must export an initialization routine which calls the routine
27194 @code{adainit} generated by @command{gnatbind} to perform the elaboration of
27195 the Ada code in the DLL (@pxref{Ada DLLs and Elaboration}). The initialization
27196 routine exported by the Ada DLL must be invoked by the clients of the DLL
27197 to initialize the DLL.
27200 When useful, the DLL should also export a finalization routine which calls
27201 routine @code{adafinal} generated by @command{gnatbind} to perform the
27202 finalization of the Ada code in the DLL (@pxref{Ada DLLs and Finalization}).
27203 The finalization routine exported by the Ada DLL must be invoked by the
27204 clients of the DLL when the DLL services are no further needed.
27207 You must provide a spec for the services exported by the Ada DLL in each
27208 of the programming languages to which you plan to make the DLL available.
27211 You must provide a definition file listing the exported entities
27212 (@pxref{The Definition File}).
27215 Finally you must use @code{gnatdll} to produce the DLL and the import
27216 library (@pxref{Using gnatdll}).
27220 Note that a relocatable DLL stripped using the @code{strip}
27221 binutils tool will not be relocatable anymore. To build a DLL without
27222 debug information pass @code{-largs -s} to @code{gnatdll}. This
27223 restriction does not apply to a DLL built using a Library Project.
27224 @pxref{Library Projects}.
27226 @node Limitations When Using Ada DLLs from Ada
27227 @subsection Limitations When Using Ada DLLs from Ada
27230 When using Ada DLLs from Ada applications there is a limitation users
27231 should be aware of. Because on Windows the GNAT run time is not in a DLL of
27232 its own, each Ada DLL includes a part of the GNAT run time. Specifically,
27233 each Ada DLL includes the services of the GNAT run time that are necessary
27234 to the Ada code inside the DLL. As a result, when an Ada program uses an
27235 Ada DLL there are two independent GNAT run times: one in the Ada DLL and
27236 one in the main program.
27238 It is therefore not possible to exchange GNAT run-time objects between the
27239 Ada DLL and the main Ada program. Example of GNAT run-time objects are file
27240 handles (e.g.@: @code{Text_IO.File_Type}), tasks types, protected objects
27243 It is completely safe to exchange plain elementary, array or record types,
27244 Windows object handles, etc.
27246 @node Exporting Ada Entities
27247 @subsection Exporting Ada Entities
27248 @cindex Export table
27251 Building a DLL is a way to encapsulate a set of services usable from any
27252 application. As a result, the Ada entities exported by a DLL should be
27253 exported with the @code{C} or @code{Stdcall} calling conventions to avoid
27254 any Ada name mangling. As an example here is an Ada package
27255 @code{API}, spec and body, exporting two procedures, a function, and a
27258 @smallexample @c ada
27261 with Interfaces.C; use Interfaces;
27263 Count : C.int := 0;
27264 function Factorial (Val : C.int) return C.int;
27266 procedure Initialize_API;
27267 procedure Finalize_API;
27268 -- Initialization & Finalization routines. More in the next section.
27270 pragma Export (C, Initialize_API);
27271 pragma Export (C, Finalize_API);
27272 pragma Export (C, Count);
27273 pragma Export (C, Factorial);
27279 @smallexample @c ada
27282 package body API is
27283 function Factorial (Val : C.int) return C.int is
27286 Count := Count + 1;
27287 for K in 1 .. Val loop
27293 procedure Initialize_API is
27295 pragma Import (C, Adainit);
27298 end Initialize_API;
27300 procedure Finalize_API is
27301 procedure Adafinal;
27302 pragma Import (C, Adafinal);
27312 If the Ada DLL you are building will only be used by Ada applications
27313 you do not have to export Ada entities with a @code{C} or @code{Stdcall}
27314 convention. As an example, the previous package could be written as
27317 @smallexample @c ada
27321 Count : Integer := 0;
27322 function Factorial (Val : Integer) return Integer;
27324 procedure Initialize_API;
27325 procedure Finalize_API;
27326 -- Initialization and Finalization routines.
27332 @smallexample @c ada
27335 package body API is
27336 function Factorial (Val : Integer) return Integer is
27337 Fact : Integer := 1;
27339 Count := Count + 1;
27340 for K in 1 .. Val loop
27347 -- The remainder of this package body is unchanged.
27354 Note that if you do not export the Ada entities with a @code{C} or
27355 @code{Stdcall} convention you will have to provide the mangled Ada names
27356 in the definition file of the Ada DLL
27357 (@pxref{Creating the Definition File}).
27359 @node Ada DLLs and Elaboration
27360 @subsection Ada DLLs and Elaboration
27361 @cindex DLLs and elaboration
27364 The DLL that you are building contains your Ada code as well as all the
27365 routines in the Ada library that are needed by it. The first thing a
27366 user of your DLL must do is elaborate the Ada code
27367 (@pxref{Elaboration Order Handling in GNAT}).
27369 To achieve this you must export an initialization routine
27370 (@code{Initialize_API} in the previous example), which must be invoked
27371 before using any of the DLL services. This elaboration routine must call
27372 the Ada elaboration routine @code{adainit} generated by the GNAT binder
27373 (@pxref{Binding with Non-Ada Main Programs}). See the body of
27374 @code{Initialize_Api} for an example. Note that the GNAT binder is
27375 automatically invoked during the DLL build process by the @code{gnatdll}
27376 tool (@pxref{Using gnatdll}).
27378 When a DLL is loaded, Windows systematically invokes a routine called
27379 @code{DllMain}. It would therefore be possible to call @code{adainit}
27380 directly from @code{DllMain} without having to provide an explicit
27381 initialization routine. Unfortunately, it is not possible to call
27382 @code{adainit} from the @code{DllMain} if your program has library level
27383 tasks because access to the @code{DllMain} entry point is serialized by
27384 the system (that is, only a single thread can execute ``through'' it at a
27385 time), which means that the GNAT run time will deadlock waiting for the
27386 newly created task to complete its initialization.
27388 @node Ada DLLs and Finalization
27389 @subsection Ada DLLs and Finalization
27390 @cindex DLLs and finalization
27393 When the services of an Ada DLL are no longer needed, the client code should
27394 invoke the DLL finalization routine, if available. The DLL finalization
27395 routine is in charge of releasing all resources acquired by the DLL. In the
27396 case of the Ada code contained in the DLL, this is achieved by calling
27397 routine @code{adafinal} generated by the GNAT binder
27398 (@pxref{Binding with Non-Ada Main Programs}).
27399 See the body of @code{Finalize_Api} for an
27400 example. As already pointed out the GNAT binder is automatically invoked
27401 during the DLL build process by the @code{gnatdll} tool
27402 (@pxref{Using gnatdll}).
27404 @node Creating a Spec for Ada DLLs
27405 @subsection Creating a Spec for Ada DLLs
27408 To use the services exported by the Ada DLL from another programming
27409 language (e.g.@: C), you have to translate the specs of the exported Ada
27410 entities in that language. For instance in the case of @code{API.dll},
27411 the corresponding C header file could look like:
27416 extern int *_imp__count;
27417 #define count (*_imp__count)
27418 int factorial (int);
27424 It is important to understand that when building an Ada DLL to be used by
27425 other Ada applications, you need two different specs for the packages
27426 contained in the DLL: one for building the DLL and the other for using
27427 the DLL. This is because the @code{DLL} calling convention is needed to
27428 use a variable defined in a DLL, but when building the DLL, the variable
27429 must have either the @code{Ada} or @code{C} calling convention. As an
27430 example consider a DLL comprising the following package @code{API}:
27432 @smallexample @c ada
27436 Count : Integer := 0;
27438 -- Remainder of the package omitted.
27445 After producing a DLL containing package @code{API}, the spec that
27446 must be used to import @code{API.Count} from Ada code outside of the
27449 @smallexample @c ada
27454 pragma Import (DLL, Count);
27460 @node Creating the Definition File
27461 @subsection Creating the Definition File
27464 The definition file is the last file needed to build the DLL. It lists
27465 the exported symbols. As an example, the definition file for a DLL
27466 containing only package @code{API} (where all the entities are exported
27467 with a @code{C} calling convention) is:
27482 If the @code{C} calling convention is missing from package @code{API},
27483 then the definition file contains the mangled Ada names of the above
27484 entities, which in this case are:
27493 api__initialize_api
27498 @node Using gnatdll
27499 @subsection Using @code{gnatdll}
27503 * gnatdll Example::
27504 * gnatdll behind the Scenes::
27509 @code{gnatdll} is a tool to automate the DLL build process once all the Ada
27510 and non-Ada sources that make up your DLL have been compiled.
27511 @code{gnatdll} is actually in charge of two distinct tasks: build the
27512 static import library for the DLL and the actual DLL. The form of the
27513 @code{gnatdll} command is
27517 @c $ gnatdll @ovar{switches} @var{list-of-files} @r{[}-largs @var{opts}@r{]}
27518 @c Expanding @ovar macro inline (explanation in macro def comments)
27519 $ gnatdll @r{[}@var{switches}@r{]} @var{list-of-files} @r{[}-largs @var{opts}@r{]}
27524 where @var{list-of-files} is a list of ALI and object files. The object
27525 file list must be the exact list of objects corresponding to the non-Ada
27526 sources whose services are to be included in the DLL. The ALI file list
27527 must be the exact list of ALI files for the corresponding Ada sources
27528 whose services are to be included in the DLL. If @var{list-of-files} is
27529 missing, only the static import library is generated.
27532 You may specify any of the following switches to @code{gnatdll}:
27535 @c @item -a@ovar{address}
27536 @c Expanding @ovar macro inline (explanation in macro def comments)
27537 @item -a@r{[}@var{address}@r{]}
27538 @cindex @option{-a} (@code{gnatdll})
27539 Build a non-relocatable DLL at @var{address}. If @var{address} is not
27540 specified the default address @var{0x11000000} will be used. By default,
27541 when this switch is missing, @code{gnatdll} builds relocatable DLL. We
27542 advise the reader to build relocatable DLL.
27544 @item -b @var{address}
27545 @cindex @option{-b} (@code{gnatdll})
27546 Set the relocatable DLL base address. By default the address is
27549 @item -bargs @var{opts}
27550 @cindex @option{-bargs} (@code{gnatdll})
27551 Binder options. Pass @var{opts} to the binder.
27553 @item -d @var{dllfile}
27554 @cindex @option{-d} (@code{gnatdll})
27555 @var{dllfile} is the name of the DLL. This switch must be present for
27556 @code{gnatdll} to do anything. The name of the generated import library is
27557 obtained algorithmically from @var{dllfile} as shown in the following
27558 example: if @var{dllfile} is @code{xyz.dll}, the import library name is
27559 @code{libxyz.dll.a}. The name of the definition file to use (if not specified
27560 by option @option{-e}) is obtained algorithmically from @var{dllfile}
27561 as shown in the following example:
27562 if @var{dllfile} is @code{xyz.dll}, the definition
27563 file used is @code{xyz.def}.
27565 @item -e @var{deffile}
27566 @cindex @option{-e} (@code{gnatdll})
27567 @var{deffile} is the name of the definition file.
27570 @cindex @option{-g} (@code{gnatdll})
27571 Generate debugging information. This information is stored in the object
27572 file and copied from there to the final DLL file by the linker,
27573 where it can be read by the debugger. You must use the
27574 @option{-g} switch if you plan on using the debugger or the symbolic
27578 @cindex @option{-h} (@code{gnatdll})
27579 Help mode. Displays @code{gnatdll} switch usage information.
27582 @cindex @option{-I} (@code{gnatdll})
27583 Direct @code{gnatdll} to search the @var{dir} directory for source and
27584 object files needed to build the DLL.
27585 (@pxref{Search Paths and the Run-Time Library (RTL)}).
27588 @cindex @option{-k} (@code{gnatdll})
27589 Removes the @code{@@}@var{nn} suffix from the import library's exported
27590 names, but keeps them for the link names. You must specify this
27591 option if you want to use a @code{Stdcall} function in a DLL for which
27592 the @code{@@}@var{nn} suffix has been removed. This is the case for most
27593 of the Windows NT DLL for example. This option has no effect when
27594 @option{-n} option is specified.
27596 @item -l @var{file}
27597 @cindex @option{-l} (@code{gnatdll})
27598 The list of ALI and object files used to build the DLL are listed in
27599 @var{file}, instead of being given in the command line. Each line in
27600 @var{file} contains the name of an ALI or object file.
27603 @cindex @option{-n} (@code{gnatdll})
27604 No Import. Do not create the import library.
27607 @cindex @option{-q} (@code{gnatdll})
27608 Quiet mode. Do not display unnecessary messages.
27611 @cindex @option{-v} (@code{gnatdll})
27612 Verbose mode. Display extra information.
27614 @item -largs @var{opts}
27615 @cindex @option{-largs} (@code{gnatdll})
27616 Linker options. Pass @var{opts} to the linker.
27619 @node gnatdll Example
27620 @subsubsection @code{gnatdll} Example
27623 As an example the command to build a relocatable DLL from @file{api.adb}
27624 once @file{api.adb} has been compiled and @file{api.def} created is
27627 $ gnatdll -d api.dll api.ali
27631 The above command creates two files: @file{libapi.dll.a} (the import
27632 library) and @file{api.dll} (the actual DLL). If you want to create
27633 only the DLL, just type:
27636 $ gnatdll -d api.dll -n api.ali
27640 Alternatively if you want to create just the import library, type:
27643 $ gnatdll -d api.dll
27646 @node gnatdll behind the Scenes
27647 @subsubsection @code{gnatdll} behind the Scenes
27650 This section details the steps involved in creating a DLL. @code{gnatdll}
27651 does these steps for you. Unless you are interested in understanding what
27652 goes on behind the scenes, you should skip this section.
27654 We use the previous example of a DLL containing the Ada package @code{API},
27655 to illustrate the steps necessary to build a DLL. The starting point is a
27656 set of objects that will make up the DLL and the corresponding ALI
27657 files. In the case of this example this means that @file{api.o} and
27658 @file{api.ali} are available. To build a relocatable DLL, @code{gnatdll} does
27663 @code{gnatdll} builds the base file (@file{api.base}). A base file gives
27664 the information necessary to generate relocation information for the
27670 $ gnatlink api -o api.jnk -mdll -Wl,--base-file,api.base
27675 In addition to the base file, the @command{gnatlink} command generates an
27676 output file @file{api.jnk} which can be discarded. The @option{-mdll} switch
27677 asks @command{gnatlink} to generate the routines @code{DllMain} and
27678 @code{DllMainCRTStartup} that are called by the Windows loader when the DLL
27679 is loaded into memory.
27682 @code{gnatdll} uses @code{dlltool} (@pxref{Using dlltool}) to build the
27683 export table (@file{api.exp}). The export table contains the relocation
27684 information in a form which can be used during the final link to ensure
27685 that the Windows loader is able to place the DLL anywhere in memory.
27689 $ dlltool --dllname api.dll --def api.def --base-file api.base \
27690 --output-exp api.exp
27695 @code{gnatdll} builds the base file using the new export table. Note that
27696 @command{gnatbind} must be called once again since the binder generated file
27697 has been deleted during the previous call to @command{gnatlink}.
27702 $ gnatlink api -o api.jnk api.exp -mdll
27703 -Wl,--base-file,api.base
27708 @code{gnatdll} builds the new export table using the new base file and
27709 generates the DLL import library @file{libAPI.dll.a}.
27713 $ dlltool --dllname api.dll --def api.def --base-file api.base \
27714 --output-exp api.exp --output-lib libAPI.a
27719 Finally @code{gnatdll} builds the relocatable DLL using the final export
27725 $ gnatlink api api.exp -o api.dll -mdll
27730 @node Using dlltool
27731 @subsubsection Using @code{dlltool}
27734 @code{dlltool} is the low-level tool used by @code{gnatdll} to build
27735 DLLs and static import libraries. This section summarizes the most
27736 common @code{dlltool} switches. The form of the @code{dlltool} command
27740 @c $ dlltool @ovar{switches}
27741 @c Expanding @ovar macro inline (explanation in macro def comments)
27742 $ dlltool @r{[}@var{switches}@r{]}
27746 @code{dlltool} switches include:
27749 @item --base-file @var{basefile}
27750 @cindex @option{--base-file} (@command{dlltool})
27751 Read the base file @var{basefile} generated by the linker. This switch
27752 is used to create a relocatable DLL.
27754 @item --def @var{deffile}
27755 @cindex @option{--def} (@command{dlltool})
27756 Read the definition file.
27758 @item --dllname @var{name}
27759 @cindex @option{--dllname} (@command{dlltool})
27760 Gives the name of the DLL. This switch is used to embed the name of the
27761 DLL in the static import library generated by @code{dlltool} with switch
27762 @option{--output-lib}.
27765 @cindex @option{-k} (@command{dlltool})
27766 Kill @code{@@}@var{nn} from exported names
27767 (@pxref{Windows Calling Conventions}
27768 for a discussion about @code{Stdcall}-style symbols.
27771 @cindex @option{--help} (@command{dlltool})
27772 Prints the @code{dlltool} switches with a concise description.
27774 @item --output-exp @var{exportfile}
27775 @cindex @option{--output-exp} (@command{dlltool})
27776 Generate an export file @var{exportfile}. The export file contains the
27777 export table (list of symbols in the DLL) and is used to create the DLL.
27779 @item --output-lib @var{libfile}
27780 @cindex @option{--output-lib} (@command{dlltool})
27781 Generate a static import library @var{libfile}.
27784 @cindex @option{-v} (@command{dlltool})
27787 @item --as @var{assembler-name}
27788 @cindex @option{--as} (@command{dlltool})
27789 Use @var{assembler-name} as the assembler. The default is @code{as}.
27792 @node GNAT and Windows Resources
27793 @section GNAT and Windows Resources
27794 @cindex Resources, windows
27797 * Building Resources::
27798 * Compiling Resources::
27799 * Using Resources::
27803 Resources are an easy way to add Windows specific objects to your
27804 application. The objects that can be added as resources include:
27833 This section explains how to build, compile and use resources.
27835 @node Building Resources
27836 @subsection Building Resources
27837 @cindex Resources, building
27840 A resource file is an ASCII file. By convention resource files have an
27841 @file{.rc} extension.
27842 The easiest way to build a resource file is to use Microsoft tools
27843 such as @code{imagedit.exe} to build bitmaps, icons and cursors and
27844 @code{dlgedit.exe} to build dialogs.
27845 It is always possible to build an @file{.rc} file yourself by writing a
27848 It is not our objective to explain how to write a resource file. A
27849 complete description of the resource script language can be found in the
27850 Microsoft documentation.
27852 @node Compiling Resources
27853 @subsection Compiling Resources
27856 @cindex Resources, compiling
27859 This section describes how to build a GNAT-compatible (COFF) object file
27860 containing the resources. This is done using the Resource Compiler
27861 @code{windres} as follows:
27864 $ windres -i myres.rc -o myres.o
27868 By default @code{windres} will run @command{gcc} to preprocess the @file{.rc}
27869 file. You can specify an alternate preprocessor (usually named
27870 @file{cpp.exe}) using the @code{windres} @option{--preprocessor}
27871 parameter. A list of all possible options may be obtained by entering
27872 the command @code{windres} @option{--help}.
27874 It is also possible to use the Microsoft resource compiler @code{rc.exe}
27875 to produce a @file{.res} file (binary resource file). See the
27876 corresponding Microsoft documentation for further details. In this case
27877 you need to use @code{windres} to translate the @file{.res} file to a
27878 GNAT-compatible object file as follows:
27881 $ windres -i myres.res -o myres.o
27884 @node Using Resources
27885 @subsection Using Resources
27886 @cindex Resources, using
27889 To include the resource file in your program just add the
27890 GNAT-compatible object file for the resource(s) to the linker
27891 arguments. With @command{gnatmake} this is done by using the @option{-largs}
27895 $ gnatmake myprog -largs myres.o
27898 @node Debugging a DLL
27899 @section Debugging a DLL
27900 @cindex DLL debugging
27903 * Program and DLL Both Built with GCC/GNAT::
27904 * Program Built with Foreign Tools and DLL Built with GCC/GNAT::
27908 Debugging a DLL is similar to debugging a standard program. But
27909 we have to deal with two different executable parts: the DLL and the
27910 program that uses it. We have the following four possibilities:
27914 The program and the DLL are built with @code{GCC/GNAT}.
27916 The program is built with foreign tools and the DLL is built with
27919 The program is built with @code{GCC/GNAT} and the DLL is built with
27924 In this section we address only cases one and two above.
27925 There is no point in trying to debug
27926 a DLL with @code{GNU/GDB}, if there is no GDB-compatible debugging
27927 information in it. To do so you must use a debugger compatible with the
27928 tools suite used to build the DLL.
27930 @node Program and DLL Both Built with GCC/GNAT
27931 @subsection Program and DLL Both Built with GCC/GNAT
27934 This is the simplest case. Both the DLL and the program have @code{GDB}
27935 compatible debugging information. It is then possible to break anywhere in
27936 the process. Let's suppose here that the main procedure is named
27937 @code{ada_main} and that in the DLL there is an entry point named
27941 The DLL (@pxref{Introduction to Dynamic Link Libraries (DLLs)}) and
27942 program must have been built with the debugging information (see GNAT -g
27943 switch). Here are the step-by-step instructions for debugging it:
27946 @item Launch @code{GDB} on the main program.
27952 @item Start the program and stop at the beginning of the main procedure
27959 This step is required to be able to set a breakpoint inside the DLL. As long
27960 as the program is not run, the DLL is not loaded. This has the
27961 consequence that the DLL debugging information is also not loaded, so it is not
27962 possible to set a breakpoint in the DLL.
27964 @item Set a breakpoint inside the DLL
27967 (gdb) break ada_dll
27974 At this stage a breakpoint is set inside the DLL. From there on
27975 you can use the standard approach to debug the whole program
27976 (@pxref{Running and Debugging Ada Programs}).
27979 @c This used to work, probably because the DLLs were non-relocatable
27980 @c keep this section around until the problem is sorted out.
27982 To break on the @code{DllMain} routine it is not possible to follow
27983 the procedure above. At the time the program stop on @code{ada_main}
27984 the @code{DllMain} routine as already been called. Either you can use
27985 the procedure below @pxref{Debugging the DLL Directly} or this procedure:
27988 @item Launch @code{GDB} on the main program.
27994 @item Load DLL symbols
27997 (gdb) add-sym api.dll
28000 @item Set a breakpoint inside the DLL
28003 (gdb) break ada_dll.adb:45
28006 Note that at this point it is not possible to break using the routine symbol
28007 directly as the program is not yet running. The solution is to break
28008 on the proper line (break in @file{ada_dll.adb} line 45).
28010 @item Start the program
28019 @node Program Built with Foreign Tools and DLL Built with GCC/GNAT
28020 @subsection Program Built with Foreign Tools and DLL Built with GCC/GNAT
28023 * Debugging the DLL Directly::
28024 * Attaching to a Running Process::
28028 In this case things are slightly more complex because it is not possible to
28029 start the main program and then break at the beginning to load the DLL and the
28030 associated DLL debugging information. It is not possible to break at the
28031 beginning of the program because there is no @code{GDB} debugging information,
28032 and therefore there is no direct way of getting initial control. This
28033 section addresses this issue by describing some methods that can be used
28034 to break somewhere in the DLL to debug it.
28037 First suppose that the main procedure is named @code{main} (this is for
28038 example some C code built with Microsoft Visual C) and that there is a
28039 DLL named @code{test.dll} containing an Ada entry point named
28043 The DLL (@pxref{Introduction to Dynamic Link Libraries (DLLs)}) must have
28044 been built with debugging information (see GNAT -g option).
28046 @node Debugging the DLL Directly
28047 @subsubsection Debugging the DLL Directly
28051 Find out the executable starting address
28054 $ objdump --file-header main.exe
28057 The starting address is reported on the last line. For example:
28060 main.exe: file format pei-i386
28061 architecture: i386, flags 0x0000010a:
28062 EXEC_P, HAS_DEBUG, D_PAGED
28063 start address 0x00401010
28067 Launch the debugger on the executable.
28074 Set a breakpoint at the starting address, and launch the program.
28077 $ (gdb) break *0x00401010
28081 The program will stop at the given address.
28084 Set a breakpoint on a DLL subroutine.
28087 (gdb) break ada_dll.adb:45
28090 Or if you want to break using a symbol on the DLL, you need first to
28091 select the Ada language (language used by the DLL).
28094 (gdb) set language ada
28095 (gdb) break ada_dll
28099 Continue the program.
28106 This will run the program until it reaches the breakpoint that has been
28107 set. From that point you can use the standard way to debug a program
28108 as described in (@pxref{Running and Debugging Ada Programs}).
28113 It is also possible to debug the DLL by attaching to a running process.
28115 @node Attaching to a Running Process
28116 @subsubsection Attaching to a Running Process
28117 @cindex DLL debugging, attach to process
28120 With @code{GDB} it is always possible to debug a running process by
28121 attaching to it. It is possible to debug a DLL this way. The limitation
28122 of this approach is that the DLL must run long enough to perform the
28123 attach operation. It may be useful for instance to insert a time wasting
28124 loop in the code of the DLL to meet this criterion.
28128 @item Launch the main program @file{main.exe}.
28134 @item Use the Windows @i{Task Manager} to find the process ID. Let's say
28135 that the process PID for @file{main.exe} is 208.
28143 @item Attach to the running process to be debugged.
28149 @item Load the process debugging information.
28152 (gdb) symbol-file main.exe
28155 @item Break somewhere in the DLL.
28158 (gdb) break ada_dll
28161 @item Continue process execution.
28170 This last step will resume the process execution, and stop at
28171 the breakpoint we have set. From there you can use the standard
28172 approach to debug a program as described in
28173 (@pxref{Running and Debugging Ada Programs}).
28175 @node Setting Stack Size from gnatlink
28176 @section Setting Stack Size from @command{gnatlink}
28179 It is possible to specify the program stack size at link time. On modern
28180 versions of Windows, starting with XP, this is mostly useful to set the size of
28181 the main stack (environment task). The other task stacks are set with pragma
28182 Storage_Size or with the @command{gnatbind -d} command.
28184 Since older versions of Windows (2000, NT4, etc.) do not allow setting the
28185 reserve size of individual tasks, the link-time stack size applies to all
28186 tasks, and pragma Storage_Size has no effect.
28187 In particular, Stack Overflow checks are made against this
28188 link-time specified size.
28190 This setting can be done with
28191 @command{gnatlink} using either:
28195 @item using @option{-Xlinker} linker option
28198 $ gnatlink hello -Xlinker --stack=0x10000,0x1000
28201 This sets the stack reserve size to 0x10000 bytes and the stack commit
28202 size to 0x1000 bytes.
28204 @item using @option{-Wl} linker option
28207 $ gnatlink hello -Wl,--stack=0x1000000
28210 This sets the stack reserve size to 0x1000000 bytes. Note that with
28211 @option{-Wl} option it is not possible to set the stack commit size
28212 because the coma is a separator for this option.
28216 @node Setting Heap Size from gnatlink
28217 @section Setting Heap Size from @command{gnatlink}
28220 Under Windows systems, it is possible to specify the program heap size from
28221 @command{gnatlink} using either:
28225 @item using @option{-Xlinker} linker option
28228 $ gnatlink hello -Xlinker --heap=0x10000,0x1000
28231 This sets the heap reserve size to 0x10000 bytes and the heap commit
28232 size to 0x1000 bytes.
28234 @item using @option{-Wl} linker option
28237 $ gnatlink hello -Wl,--heap=0x1000000
28240 This sets the heap reserve size to 0x1000000 bytes. Note that with
28241 @option{-Wl} option it is not possible to set the heap commit size
28242 because the coma is a separator for this option.
28248 @c **********************************
28249 @c * GNU Free Documentation License *
28250 @c **********************************
28252 @c GNU Free Documentation License
28254 @node Index,,GNU Free Documentation License, Top
28260 @c Put table of contents at end, otherwise it precedes the "title page" in
28261 @c the .txt version
28262 @c Edit the pdf file to move the contents to the beginning, after the title