1 \input texinfo @c -*-texinfo-*-
4 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
6 @c GNAT DOCUMENTATION o
10 @c GNAT is maintained by Ada Core Technologies Inc (http://www.gnat.com). o
12 @c oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo
14 @setfilename gnat_ugn.info
17 Copyright @copyright{} 1995-2005, 2006, 2007, 2008 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 1.2 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
111 @settitle @value{EDITION} User's Guide @value{PLATFORM}
112 @dircategory GNU Ada tools
114 * @value{EDITION} User's Guide (gnat_ugn) @value{PLATFORM}
117 @include gcc-common.texi
119 @setchapternewpage odd
124 @title @value{EDITION} User's Guide
128 @titlefont{@i{@value{PLATFORM}}}
134 @subtitle GNAT, The GNU Ada Compiler
139 @vskip 0pt plus 1filll
146 @node Top, About This Guide, (dir), (dir)
147 @top @value{EDITION} User's Guide
150 @value{EDITION} User's Guide @value{PLATFORM}
153 GNAT, The GNU Ada Compiler@*
154 GCC version @value{version-GCC}@*
161 * Getting Started with GNAT::
162 * The GNAT Compilation Model::
163 * Compiling Using gcc::
164 * Binding Using gnatbind::
165 * Linking Using gnatlink::
166 * The GNAT Make Program gnatmake::
167 * Improving Performance::
168 * Renaming Files Using gnatchop::
169 * Configuration Pragmas::
170 * Handling Arbitrary File Naming Conventions Using gnatname::
171 * GNAT Project Manager::
172 * The Cross-Referencing Tools gnatxref and gnatfind::
173 * The GNAT Pretty-Printer gnatpp::
174 * The GNAT Metric Tool gnatmetric::
175 * File Name Krunching Using gnatkr::
176 * Preprocessing Using gnatprep::
178 * The GNAT Run-Time Library Builder gnatlbr::
180 * The GNAT Library Browser gnatls::
181 * Cleaning Up Using gnatclean::
183 * GNAT and Libraries::
184 * Using the GNU make Utility::
186 * Memory Management Issues::
187 * Stack Related Facilities::
188 * Verifying Properties Using gnatcheck::
189 * Creating Sample Bodies Using gnatstub::
190 * Other Utility Programs::
191 * Running and Debugging Ada Programs::
193 * Code Coverage and Profiling::
196 * Compatibility with HP Ada::
198 * Platform-Specific Information for the Run-Time Libraries::
199 * Example of Binder Output File::
200 * Elaboration Order Handling in GNAT::
201 * Conditional Compilation::
203 * Compatibility and Porting Guide::
205 * Microsoft Windows Topics::
207 * GNU Free Documentation License::
210 --- The Detailed Node Listing ---
214 * What This Guide Contains::
215 * What You Should Know before Reading This Guide::
216 * Related Information::
219 Getting Started with GNAT
222 * Running a Simple Ada Program::
223 * Running a Program with Multiple Units::
224 * Using the gnatmake Utility::
226 * Editing with Emacs::
229 * Introduction to GPS::
232 The GNAT Compilation Model
234 * Source Representation::
235 * Foreign Language Representation::
236 * File Naming Rules::
237 * Using Other File Names::
238 * Alternative File Naming Schemes::
239 * Generating Object Files::
240 * Source Dependencies::
241 * The Ada Library Information Files::
242 * Binding an Ada Program::
243 * Mixed Language Programming::
245 * Building Mixed Ada & C++ Programs::
246 * Comparison between GNAT and C/C++ Compilation Models::
248 * Comparison between GNAT and Conventional Ada Library Models::
250 * Placement of temporary files::
253 Foreign Language Representation
256 * Other 8-Bit Codes::
257 * Wide Character Encodings::
259 Compiling Ada Programs With gcc
261 * Compiling Programs::
263 * Search Paths and the Run-Time Library (RTL)::
264 * Order of Compilation Issues::
269 * Output and Error Message Control::
270 * Warning Message Control::
271 * Debugging and Assertion Control::
272 * Validity Checking::
275 * Using gcc for Syntax Checking::
276 * Using gcc for Semantic Checking::
277 * Compiling Different Versions of Ada::
278 * Character Set Control::
279 * File Naming Control::
280 * Subprogram Inlining Control::
281 * Auxiliary Output Control::
282 * Debugging Control::
283 * Exception Handling Control::
284 * Units to Sources Mapping Files::
285 * Integrated Preprocessing::
290 Binding Ada Programs With gnatbind
293 * Switches for gnatbind::
294 * Command-Line Access::
295 * Search Paths for gnatbind::
296 * Examples of gnatbind Usage::
298 Switches for gnatbind
300 * Consistency-Checking Modes::
301 * Binder Error Message Control::
302 * Elaboration Control::
304 * Binding with Non-Ada Main Programs::
305 * Binding Programs with No Main Subprogram::
307 Linking Using gnatlink
310 * Switches for gnatlink::
312 The GNAT Make Program gnatmake
315 * Switches for gnatmake::
316 * Mode Switches for gnatmake::
317 * Notes on the Command Line::
318 * How gnatmake Works::
319 * Examples of gnatmake Usage::
321 Improving Performance
322 * Performance Considerations::
323 * Text_IO Suggestions::
324 * Reducing Size of Ada Executables with gnatelim::
325 * Reducing Size of Executables with unused subprogram/data elimination::
327 Performance Considerations
328 * Controlling Run-Time Checks::
329 * Use of Restrictions::
330 * Optimization Levels::
331 * Debugging Optimized Code::
332 * Inlining of Subprograms::
333 * Other Optimization Switches::
334 * Optimization and Strict Aliasing::
336 * Coverage Analysis::
339 Reducing Size of Ada Executables with gnatelim
342 * Correcting the List of Eliminate Pragmas::
343 * Making Your Executables Smaller::
344 * Summary of the gnatelim Usage Cycle::
346 Reducing Size of Executables with unused subprogram/data elimination
347 * About unused subprogram/data elimination::
348 * Compilation options::
350 Renaming Files Using gnatchop
352 * Handling Files with Multiple Units::
353 * Operating gnatchop in Compilation Mode::
354 * Command Line for gnatchop::
355 * Switches for gnatchop::
356 * Examples of gnatchop Usage::
358 Configuration Pragmas
360 * Handling of Configuration Pragmas::
361 * The Configuration Pragmas Files::
363 Handling Arbitrary File Naming Conventions Using gnatname
365 * Arbitrary File Naming Conventions::
367 * Switches for gnatname::
368 * Examples of gnatname Usage::
373 * Examples of Project Files::
374 * Project File Syntax::
375 * Objects and Sources in Project Files::
376 * Importing Projects::
377 * Project Extension::
378 * Project Hierarchy Extension::
379 * External References in Project Files::
380 * Packages in Project Files::
381 * Variables from Imported Projects::
384 * Stand-alone Library Projects::
385 * Switches Related to Project Files::
386 * Tools Supporting Project Files::
387 * An Extended Example::
388 * Project File Complete Syntax::
390 The Cross-Referencing Tools gnatxref and gnatfind
392 * gnatxref Switches::
393 * gnatfind Switches::
394 * Project Files for gnatxref and gnatfind::
395 * Regular Expressions in gnatfind and gnatxref::
396 * Examples of gnatxref Usage::
397 * Examples of gnatfind Usage::
399 The GNAT Pretty-Printer gnatpp
401 * Switches for gnatpp::
404 The GNAT Metrics Tool gnatmetric
406 * Switches for gnatmetric::
408 File Name Krunching Using gnatkr
413 * Examples of gnatkr Usage::
415 Preprocessing Using gnatprep
416 * Preprocessing Symbols::
418 * Switches for gnatprep::
419 * Form of Definitions File::
420 * Form of Input Text for gnatprep::
423 The GNAT Run-Time Library Builder gnatlbr
426 * Switches for gnatlbr::
427 * Examples of gnatlbr Usage::
430 The GNAT Library Browser gnatls
433 * Switches for gnatls::
434 * Examples of gnatls Usage::
436 Cleaning Up Using gnatclean
438 * Running gnatclean::
439 * Switches for gnatclean::
440 @c * Examples of gnatclean Usage::
446 * Introduction to Libraries in GNAT::
447 * General Ada Libraries::
448 * Stand-alone Ada Libraries::
449 * Rebuilding the GNAT Run-Time Library::
451 Using the GNU make Utility
453 * Using gnatmake in a Makefile::
454 * Automatically Creating a List of Directories::
455 * Generating the Command Line Switches::
456 * Overcoming Command Line Length Limits::
459 Memory Management Issues
461 * Some Useful Memory Pools::
462 * The GNAT Debug Pool Facility::
467 Stack Related Facilities
469 * Stack Overflow Checking::
470 * Static Stack Usage Analysis::
471 * Dynamic Stack Usage Analysis::
473 Some Useful Memory Pools
475 The GNAT Debug Pool Facility
481 * Switches for gnatmem::
482 * Example of gnatmem Usage::
485 Verifying Properties Using gnatcheck
487 * Format of the Report File::
488 * General gnatcheck Switches::
489 * gnatcheck Rule Options::
490 * Adding the Results of Compiler Checks to gnatcheck Output::
491 * Project-Wide Checks::
494 Sample Bodies Using gnatstub
497 * Switches for gnatstub::
499 Other Utility Programs
501 * Using Other Utility Programs with GNAT::
502 * The External Symbol Naming Scheme of GNAT::
503 * Converting Ada Files to html with gnathtml::
506 Code Coverage and Profiling
508 * Code Coverage of Ada Programs using gcov::
509 * Profiling an Ada Program using gprof::
512 Running and Debugging Ada Programs
514 * The GNAT Debugger GDB::
516 * Introduction to GDB Commands::
517 * Using Ada Expressions::
518 * Calling User-Defined Subprograms::
519 * Using the Next Command in a Function::
522 * Debugging Generic Units::
523 * GNAT Abnormal Termination or Failure to Terminate::
524 * Naming Conventions for GNAT Source Files::
525 * Getting Internal Debugging Information::
533 Compatibility with HP Ada
535 * Ada Language Compatibility::
536 * Differences in the Definition of Package System::
537 * Language-Related Features::
538 * The Package STANDARD::
539 * The Package SYSTEM::
540 * Tasking and Task-Related Features::
541 * Pragmas and Pragma-Related Features::
542 * Library of Predefined Units::
544 * Main Program Definition::
545 * Implementation-Defined Attributes::
546 * Compiler and Run-Time Interfacing::
547 * Program Compilation and Library Management::
549 * Implementation Limits::
550 * Tools and Utilities::
552 Language-Related Features
554 * Integer Types and Representations::
555 * Floating-Point Types and Representations::
556 * Pragmas Float_Representation and Long_Float::
557 * Fixed-Point Types and Representations::
558 * Record and Array Component Alignment::
560 * Other Representation Clauses::
562 Tasking and Task-Related Features
564 * Implementation of Tasks in HP Ada for OpenVMS Alpha Systems::
565 * Assigning Task IDs::
566 * Task IDs and Delays::
567 * Task-Related Pragmas::
568 * Scheduling and Task Priority::
570 * External Interrupts::
572 Pragmas and Pragma-Related Features
574 * Restrictions on the Pragma INLINE::
575 * Restrictions on the Pragma INTERFACE::
576 * Restrictions on the Pragma SYSTEM_NAME::
578 Library of Predefined Units
580 * Changes to DECLIB::
584 * Shared Libraries and Options Files::
588 Platform-Specific Information for the Run-Time Libraries
590 * Summary of Run-Time Configurations::
591 * Specifying a Run-Time Library::
592 * Choosing the Scheduling Policy::
593 * Solaris-Specific Considerations::
594 * Linux-Specific Considerations::
595 * AIX-Specific Considerations::
596 * Irix-Specific Considerations::
598 Example of Binder Output File
600 Elaboration Order Handling in GNAT
603 * Checking the Elaboration Order::
604 * Controlling the Elaboration Order::
605 * Controlling Elaboration in GNAT - Internal Calls::
606 * Controlling Elaboration in GNAT - External Calls::
607 * Default Behavior in GNAT - Ensuring Safety::
608 * Treatment of Pragma Elaborate::
609 * Elaboration Issues for Library Tasks::
610 * Mixing Elaboration Models::
611 * What to Do If the Default Elaboration Behavior Fails::
612 * Elaboration for Access-to-Subprogram Values::
613 * Summary of Procedures for Elaboration Control::
614 * Other Elaboration Order Considerations::
616 Conditional Compilation
617 * Use of Boolean Constants::
618 * Debugging - A Special Case::
619 * Conditionalizing Declarations::
620 * Use of Alternative Implementations::
625 * Basic Assembler Syntax::
626 * A Simple Example of Inline Assembler::
627 * Output Variables in Inline Assembler::
628 * Input Variables in Inline Assembler::
629 * Inlining Inline Assembler Code::
630 * Other Asm Functionality::
632 Compatibility and Porting Guide
634 * Compatibility with Ada 83::
635 * Compatibility between Ada 95 and Ada 2005::
636 * Implementation-dependent characteristics::
638 @c This brief section is only in the non-VMS version
639 @c The complete chapter on HP Ada issues is in the VMS version
640 * Compatibility with HP Ada 83::
642 * Compatibility with Other Ada Systems::
643 * Representation Clauses::
645 * Transitioning to 64-Bit GNAT for OpenVMS::
649 Microsoft Windows Topics
651 * Using GNAT on Windows::
652 * CONSOLE and WINDOWS subsystems::
654 * Mixed-Language Programming on Windows::
655 * Windows Calling Conventions::
656 * Introduction to Dynamic Link Libraries (DLLs)::
657 * Using DLLs with GNAT::
658 * Building DLLs with GNAT::
659 * GNAT and Windows Resources::
661 * Setting Stack Size from gnatlink::
662 * Setting Heap Size from gnatlink::
669 @node About This Guide
670 @unnumbered About This Guide
674 This guide describes the use of @value{EDITION},
675 a compiler and software development toolset for the full Ada
676 programming language, implemented on OpenVMS for HP's Alpha and
677 Integrity server (I64) platforms.
680 This guide describes the use of @value{EDITION},
681 a compiler and software development
682 toolset for the full Ada programming language.
684 It documents the features of the compiler and tools, and explains
685 how to use them to build Ada applications.
687 @value{EDITION} implements Ada 95 and Ada 2005, and it may also be invoked in
688 Ada 83 compatibility mode.
689 By default, @value{EDITION} assumes @value{DEFAULTLANGUAGEVERSION},
690 but you can override with a compiler switch
691 (@pxref{Compiling Different Versions of Ada})
692 to explicitly specify the language version.
693 Throughout this manual, references to ``Ada'' without a year suffix
694 apply to both the Ada 95 and Ada 2005 versions of the language.
698 For ease of exposition, ``@value{EDITION}'' will be referred to simply as
699 ``GNAT'' in the remainder of this document.
706 * What This Guide Contains::
707 * What You Should Know before Reading This Guide::
708 * Related Information::
712 @node What This Guide Contains
713 @unnumberedsec What This Guide Contains
716 This guide contains the following chapters:
720 @ref{Getting Started with GNAT}, describes how to get started compiling
721 and running Ada programs with the GNAT Ada programming environment.
723 @ref{The GNAT Compilation Model}, describes the compilation model used
727 @ref{Compiling Using gcc}, describes how to compile
728 Ada programs with @command{gcc}, the Ada compiler.
731 @ref{Binding Using gnatbind}, describes how to
732 perform binding of Ada programs with @code{gnatbind}, the GNAT binding
736 @ref{Linking Using gnatlink},
737 describes @command{gnatlink}, a
738 program that provides for linking using the GNAT run-time library to
739 construct a program. @command{gnatlink} can also incorporate foreign language
740 object units into the executable.
743 @ref{The GNAT Make Program gnatmake}, describes @command{gnatmake}, a
744 utility that automatically determines the set of sources
745 needed by an Ada compilation unit, and executes the necessary compilations
749 @ref{Improving Performance}, shows various techniques for making your
750 Ada program run faster or take less space.
751 It discusses the effect of the compiler's optimization switch and
752 also describes the @command{gnatelim} tool and unused subprogram/data
756 @ref{Renaming Files Using gnatchop}, describes
757 @code{gnatchop}, a utility that allows you to preprocess a file that
758 contains Ada source code, and split it into one or more new files, one
759 for each compilation unit.
762 @ref{Configuration Pragmas}, describes the configuration pragmas
766 @ref{Handling Arbitrary File Naming Conventions Using gnatname},
767 shows how to override the default GNAT file naming conventions,
768 either for an individual unit or globally.
771 @ref{GNAT Project Manager}, describes how to use project files
772 to organize large projects.
775 @ref{The Cross-Referencing Tools gnatxref and gnatfind}, discusses
776 @code{gnatxref} and @code{gnatfind}, two tools that provide an easy
777 way to navigate through sources.
780 @ref{The GNAT Pretty-Printer gnatpp}, shows how to produce a reformatted
781 version of an Ada source file with control over casing, indentation,
782 comment placement, and other elements of program presentation style.
785 @ref{The GNAT Metric Tool gnatmetric}, shows how to compute various
786 metrics for an Ada source file, such as the number of types and subprograms,
787 and assorted complexity measures.
790 @ref{File Name Krunching Using gnatkr}, describes the @code{gnatkr}
791 file name krunching utility, used to handle shortened
792 file names on operating systems with a limit on the length of names.
795 @ref{Preprocessing Using gnatprep}, describes @code{gnatprep}, a
796 preprocessor utility that allows a single source file to be used to
797 generate multiple or parameterized source files by means of macro
802 @ref{The GNAT Run-Time Library Builder gnatlbr}, describes @command{gnatlbr},
803 a tool for rebuilding the GNAT run time with user-supplied
804 configuration pragmas.
808 @ref{The GNAT Library Browser gnatls}, describes @code{gnatls}, a
809 utility that displays information about compiled units, including dependences
810 on the corresponding sources files, and consistency of compilations.
813 @ref{Cleaning Up Using gnatclean}, describes @code{gnatclean}, a utility
814 to delete files that are produced by the compiler, binder and linker.
818 @ref{GNAT and Libraries}, describes the process of creating and using
819 Libraries with GNAT. It also describes how to recompile the GNAT run-time
823 @ref{Using the GNU make Utility}, describes some techniques for using
824 the GNAT toolset in Makefiles.
828 @ref{Memory Management Issues}, describes some useful predefined storage pools
829 and in particular the GNAT Debug Pool facility, which helps detect incorrect
832 It also describes @command{gnatmem}, a utility that monitors dynamic
833 allocation and deallocation and helps detect ``memory leaks''.
837 @ref{Stack Related Facilities}, describes some useful tools associated with
838 stack checking and analysis.
841 @ref{Verifying Properties Using gnatcheck}, discusses @code{gnatcheck},
842 a utility that checks Ada code against a set of rules.
845 @ref{Creating Sample Bodies Using gnatstub}, discusses @code{gnatstub},
846 a utility that generates empty but compilable bodies for library units.
849 @ref{Other Utility Programs}, discusses several other GNAT utilities,
850 including @code{gnathtml}.
854 @ref{Code Coverage and Profiling}, describes how to perform a structural
855 coverage and profile the execution of Ada programs.
859 @ref{Running and Debugging Ada Programs}, describes how to run and debug
864 @ref{Compatibility with HP Ada}, details the compatibility of GNAT with
865 HP Ada 83 @footnote{``HP Ada'' refers to the legacy product originally
866 developed by Digital Equipment Corporation and currently supported by HP.}
867 for OpenVMS Alpha. This product was formerly known as DEC Ada,
870 historical compatibility reasons, the relevant libraries still use the
875 @ref{Platform-Specific Information for the Run-Time Libraries},
876 describes the various run-time
877 libraries supported by GNAT on various platforms and explains how to
878 choose a particular library.
881 @ref{Example of Binder Output File}, shows the source code for the binder
882 output file for a sample program.
885 @ref{Elaboration Order Handling in GNAT}, describes how GNAT helps
886 you deal with elaboration order issues.
889 @ref{Conditional Compilation}, describes how to model conditional compilation,
890 both with Ada in general and with GNAT facilities in particular.
893 @ref{Inline Assembler}, shows how to use the inline assembly facility
897 @ref{Compatibility and Porting Guide}, contains sections on compatibility
898 of GNAT with other Ada development environments (including Ada 83 systems),
899 to assist in porting code from those environments.
903 @ref{Microsoft Windows Topics}, presents information relevant to the
904 Microsoft Windows platform.
908 @c *************************************************
909 @node What You Should Know before Reading This Guide
910 @c *************************************************
911 @unnumberedsec What You Should Know before Reading This Guide
913 @cindex Ada 95 Language Reference Manual
914 @cindex Ada 2005 Language Reference Manual
916 This guide assumes a basic familiarity with the Ada 95 language, as
917 described in the International Standard ANSI/ISO/IEC-8652:1995, January
919 It does not require knowledge of the new features introduced by Ada 2005,
920 (officially known as ISO/IEC 8652:1995 with Technical Corrigendum 1
922 Both reference manuals are included in the GNAT documentation
925 @node Related Information
926 @unnumberedsec Related Information
929 For further information about related tools, refer to the following
934 @xref{Top, GNAT Reference Manual, About This Guide, gnat_rm, GNAT
935 Reference Manual}, which contains all reference material for the GNAT
936 implementation of Ada.
940 @cite{Using the GNAT Programming Studio}, which describes the GPS
941 Integrated Development Environment.
944 @cite{GNAT Programming Studio Tutorial}, which introduces the
945 main GPS features through examples.
949 @cite{Ada 95 Reference Manual}, which contains reference
950 material for the Ada 95 programming language.
953 @cite{Ada 2005 Reference Manual}, which contains reference
954 material for the Ada 2005 programming language.
957 @xref{Top,, Debugging with GDB, gdb, Debugging with GDB},
959 in the GNU:[DOCS] directory,
961 for all details on the use of the GNU source-level debugger.
964 @xref{Top,, The extensible self-documenting text editor, emacs,
967 located in the GNU:[DOCS] directory if the EMACS kit is installed,
969 for full information on the extensible editor and programming
976 @unnumberedsec Conventions
978 @cindex Typographical conventions
981 Following are examples of the typographical and graphic conventions used
986 @code{Functions}, @command{utility program names}, @code{standard names},
990 @option{Option flags}
993 @file{File names}, @samp{button names}, and @samp{field names}.
996 @code{Variables}, @env{environment variables}, and @var{metasyntactic
1003 @r{[}optional information or parameters@r{]}
1006 Examples are described by text
1008 and then shown this way.
1013 Commands that are entered by the user are preceded in this manual by the
1014 characters @w{``@code{$ }''} (dollar sign followed by space). If your system
1015 uses this sequence as a prompt, then the commands will appear exactly as
1016 you see them in the manual. If your system uses some other prompt, then
1017 the command will appear with the @code{$} replaced by whatever prompt
1018 character you are using.
1021 Full file names are shown with the ``@code{/}'' character
1022 as the directory separator; e.g., @file{parent-dir/subdir/myfile.adb}.
1023 If you are using GNAT on a Windows platform, please note that
1024 the ``@code{\}'' character should be used instead.
1027 @c ****************************
1028 @node Getting Started with GNAT
1029 @chapter Getting Started with GNAT
1032 This chapter describes some simple ways of using GNAT to build
1033 executable Ada programs.
1035 @ref{Running GNAT}, through @ref{Using the gnatmake Utility},
1036 show how to use the command line environment.
1037 @ref{Introduction to GPS}, provides a brief
1038 introduction to the GNAT Programming Studio, a visually-oriented
1039 Integrated Development Environment for GNAT.
1040 GPS offers a graphical ``look and feel'', support for development in
1041 other programming languages, comprehensive browsing features, and
1042 many other capabilities.
1043 For information on GPS please refer to
1044 @cite{Using the GNAT Programming Studio}.
1049 * Running a Simple Ada Program::
1050 * Running a Program with Multiple Units::
1051 * Using the gnatmake Utility::
1053 * Editing with Emacs::
1056 * Introduction to GPS::
1061 @section Running GNAT
1064 Three steps are needed to create an executable file from an Ada source
1069 The source file(s) must be compiled.
1071 The file(s) must be bound using the GNAT binder.
1073 All appropriate object files must be linked to produce an executable.
1077 All three steps are most commonly handled by using the @command{gnatmake}
1078 utility program that, given the name of the main program, automatically
1079 performs the necessary compilation, binding and linking steps.
1081 @node Running a Simple Ada Program
1082 @section Running a Simple Ada Program
1085 Any text editor may be used to prepare an Ada program.
1087 used, the optional Ada mode may be helpful in laying out the program.)
1089 program text is a normal text file. We will assume in our initial
1090 example that you have used your editor to prepare the following
1091 standard format text file:
1093 @smallexample @c ada
1095 with Ada.Text_IO; use Ada.Text_IO;
1098 Put_Line ("Hello WORLD!");
1104 This file should be named @file{hello.adb}.
1105 With the normal default file naming conventions, GNAT requires
1107 contain a single compilation unit whose file name is the
1109 with periods replaced by hyphens; the
1110 extension is @file{ads} for a
1111 spec and @file{adb} for a body.
1112 You can override this default file naming convention by use of the
1113 special pragma @code{Source_File_Name} (@pxref{Using Other File Names}).
1114 Alternatively, if you want to rename your files according to this default
1115 convention, which is probably more convenient if you will be using GNAT
1116 for all your compilations, then the @code{gnatchop} utility
1117 can be used to generate correctly-named source files
1118 (@pxref{Renaming Files Using gnatchop}).
1120 You can compile the program using the following command (@code{$} is used
1121 as the command prompt in the examples in this document):
1128 @command{gcc} is the command used to run the compiler. This compiler is
1129 capable of compiling programs in several languages, including Ada and
1130 C. It assumes that you have given it an Ada program if the file extension is
1131 either @file{.ads} or @file{.adb}, and it will then call
1132 the GNAT compiler to compile the specified file.
1135 The @option{-c} switch is required. It tells @command{gcc} to only do a
1136 compilation. (For C programs, @command{gcc} can also do linking, but this
1137 capability is not used directly for Ada programs, so the @option{-c}
1138 switch must always be present.)
1141 This compile command generates a file
1142 @file{hello.o}, which is the object
1143 file corresponding to your Ada program. It also generates
1144 an ``Ada Library Information'' file @file{hello.ali},
1145 which contains additional information used to check
1146 that an Ada program is consistent.
1147 To build an executable file,
1148 use @code{gnatbind} to bind the program
1149 and @command{gnatlink} to link it. The
1150 argument to both @code{gnatbind} and @command{gnatlink} is the name of the
1151 @file{ALI} file, but the default extension of @file{.ali} can
1152 be omitted. This means that in the most common case, the argument
1153 is simply the name of the main program:
1161 A simpler method of carrying out these steps is to use
1163 a master program that invokes all the required
1164 compilation, binding and linking tools in the correct order. In particular,
1165 @command{gnatmake} automatically recompiles any sources that have been
1166 modified since they were last compiled, or sources that depend
1167 on such modified sources, so that ``version skew'' is avoided.
1168 @cindex Version skew (avoided by @command{gnatmake})
1171 $ gnatmake hello.adb
1175 The result is an executable program called @file{hello}, which can be
1183 assuming that the current directory is on the search path
1184 for executable programs.
1187 and, if all has gone well, you will see
1194 appear in response to this command.
1196 @c ****************************************
1197 @node Running a Program with Multiple Units
1198 @section Running a Program with Multiple Units
1201 Consider a slightly more complicated example that has three files: a
1202 main program, and the spec and body of a package:
1204 @smallexample @c ada
1207 package Greetings is
1212 with Ada.Text_IO; use Ada.Text_IO;
1213 package body Greetings is
1216 Put_Line ("Hello WORLD!");
1219 procedure Goodbye is
1221 Put_Line ("Goodbye WORLD!");
1238 Following the one-unit-per-file rule, place this program in the
1239 following three separate files:
1243 spec of package @code{Greetings}
1246 body of package @code{Greetings}
1249 body of main program
1253 To build an executable version of
1254 this program, we could use four separate steps to compile, bind, and link
1255 the program, as follows:
1259 $ gcc -c greetings.adb
1265 Note that there is no required order of compilation when using GNAT.
1266 In particular it is perfectly fine to compile the main program first.
1267 Also, it is not necessary to compile package specs in the case where
1268 there is an accompanying body; you only need to compile the body. If you want
1269 to submit these files to the compiler for semantic checking and not code
1270 generation, then use the
1271 @option{-gnatc} switch:
1274 $ gcc -c greetings.ads -gnatc
1278 Although the compilation can be done in separate steps as in the
1279 above example, in practice it is almost always more convenient
1280 to use the @command{gnatmake} tool. All you need to know in this case
1281 is the name of the main program's source file. The effect of the above four
1282 commands can be achieved with a single one:
1285 $ gnatmake gmain.adb
1289 In the next section we discuss the advantages of using @command{gnatmake} in
1292 @c *****************************
1293 @node Using the gnatmake Utility
1294 @section Using the @command{gnatmake} Utility
1297 If you work on a program by compiling single components at a time using
1298 @command{gcc}, you typically keep track of the units you modify. In order to
1299 build a consistent system, you compile not only these units, but also any
1300 units that depend on the units you have modified.
1301 For example, in the preceding case,
1302 if you edit @file{gmain.adb}, you only need to recompile that file. But if
1303 you edit @file{greetings.ads}, you must recompile both
1304 @file{greetings.adb} and @file{gmain.adb}, because both files contain
1305 units that depend on @file{greetings.ads}.
1307 @code{gnatbind} will warn you if you forget one of these compilation
1308 steps, so that it is impossible to generate an inconsistent program as a
1309 result of forgetting to do a compilation. Nevertheless it is tedious and
1310 error-prone to keep track of dependencies among units.
1311 One approach to handle the dependency-bookkeeping is to use a
1312 makefile. However, makefiles present maintenance problems of their own:
1313 if the dependencies change as you change the program, you must make
1314 sure that the makefile is kept up-to-date manually, which is also an
1315 error-prone process.
1317 The @command{gnatmake} utility takes care of these details automatically.
1318 Invoke it using either one of the following forms:
1321 $ gnatmake gmain.adb
1322 $ gnatmake ^gmain^GMAIN^
1326 The argument is the name of the file containing the main program;
1327 you may omit the extension. @command{gnatmake}
1328 examines the environment, automatically recompiles any files that need
1329 recompiling, and binds and links the resulting set of object files,
1330 generating the executable file, @file{^gmain^GMAIN.EXE^}.
1331 In a large program, it
1332 can be extremely helpful to use @command{gnatmake}, because working out by hand
1333 what needs to be recompiled can be difficult.
1335 Note that @command{gnatmake}
1336 takes into account all the Ada rules that
1337 establish dependencies among units. These include dependencies that result
1338 from inlining subprogram bodies, and from
1339 generic instantiation. Unlike some other
1340 Ada make tools, @command{gnatmake} does not rely on the dependencies that were
1341 found by the compiler on a previous compilation, which may possibly
1342 be wrong when sources change. @command{gnatmake} determines the exact set of
1343 dependencies from scratch each time it is run.
1346 @node Editing with Emacs
1347 @section Editing with Emacs
1351 Emacs is an extensible self-documenting text editor that is available in a
1352 separate VMSINSTAL kit.
1354 Invoke Emacs by typing @kbd{Emacs} at the command prompt. To get started,
1355 click on the Emacs Help menu and run the Emacs Tutorial.
1356 In a character cell terminal, Emacs help is invoked with @kbd{Ctrl-h} (also
1357 written as @kbd{C-h}), and the tutorial by @kbd{C-h t}.
1359 Documentation on Emacs and other tools is available in Emacs under the
1360 pull-down menu button: @code{Help - Info}. After selecting @code{Info},
1361 use the middle mouse button to select a topic (e.g.@: Emacs).
1363 In a character cell terminal, do @kbd{C-h i} to invoke info, and then @kbd{m}
1364 (stands for menu) followed by the menu item desired, as in @kbd{m Emacs}, to
1365 get to the Emacs manual.
1366 Help on Emacs is also available by typing @kbd{HELP EMACS} at the DCL command
1369 The tutorial is highly recommended in order to learn the intricacies of Emacs,
1370 which is sufficiently extensible to provide for a complete programming
1371 environment and shell for the sophisticated user.
1375 @node Introduction to GPS
1376 @section Introduction to GPS
1377 @cindex GPS (GNAT Programming Studio)
1378 @cindex GNAT Programming Studio (GPS)
1380 Although the command line interface (@command{gnatmake}, etc.) alone
1381 is sufficient, a graphical Interactive Development
1382 Environment can make it easier for you to compose, navigate, and debug
1383 programs. This section describes the main features of GPS
1384 (``GNAT Programming Studio''), the GNAT graphical IDE.
1385 You will see how to use GPS to build and debug an executable, and
1386 you will also learn some of the basics of the GNAT ``project'' facility.
1388 GPS enables you to do much more than is presented here;
1389 e.g., you can produce a call graph, interface to a third-party
1390 Version Control System, and inspect the generated assembly language
1392 Indeed, GPS also supports languages other than Ada.
1393 Such additional information, and an explanation of all of the GPS menu
1394 items. may be found in the on-line help, which includes
1395 a user's guide and a tutorial (these are also accessible from the GNAT
1399 * Building a New Program with GPS::
1400 * Simple Debugging with GPS::
1403 @node Building a New Program with GPS
1404 @subsection Building a New Program with GPS
1406 GPS invokes the GNAT compilation tools using information
1407 contained in a @emph{project} (also known as a @emph{project file}):
1408 a collection of properties such
1409 as source directories, identities of main subprograms, tool switches, etc.,
1410 and their associated values.
1411 See @ref{GNAT Project Manager} for details.
1412 In order to run GPS, you will need to either create a new project
1413 or else open an existing one.
1415 This section will explain how you can use GPS to create a project,
1416 to associate Ada source files with a project, and to build and run
1420 @item @emph{Creating a project}
1422 Invoke GPS, either from the command line or the platform's IDE.
1423 After it starts, GPS will display a ``Welcome'' screen with three
1428 @code{Start with default project in directory}
1431 @code{Create new project with wizard}
1434 @code{Open existing project}
1438 Select @code{Create new project with wizard} and press @code{OK}.
1439 A new window will appear. In the text box labeled with
1440 @code{Enter the name of the project to create}, type @file{sample}
1441 as the project name.
1442 In the next box, browse to choose the directory in which you
1443 would like to create the project file.
1444 After selecting an appropriate directory, press @code{Forward}.
1446 A window will appear with the title
1447 @code{Version Control System Configuration}.
1448 Simply press @code{Forward}.
1450 A window will appear with the title
1451 @code{Please select the source directories for this project}.
1452 The directory that you specified for the project file will be selected
1453 by default as the one to use for sources; simply press @code{Forward}.
1455 A window will appear with the title
1456 @code{Please select the build directory for this project}.
1457 The directory that you specified for the project file will be selected
1458 by default for object files and executables;
1459 simply press @code{Forward}.
1461 A window will appear with the title
1462 @code{Please select the main units for this project}.
1463 You will supply this information later, after creating the source file.
1464 Simply press @code{Forward} for now.
1466 A window will appear with the title
1467 @code{Please select the switches to build the project}.
1468 Press @code{Apply}. This will create a project file named
1469 @file{sample.prj} in the directory that you had specified.
1471 @item @emph{Creating and saving the source file}
1473 After you create the new project, a GPS window will appear, which is
1474 partitioned into two main sections:
1478 A @emph{Workspace area}, initially greyed out, which you will use for
1479 creating and editing source files
1482 Directly below, a @emph{Messages area}, which initially displays a
1483 ``Welcome'' message.
1484 (If the Messages area is not visible, drag its border upward to expand it.)
1488 Select @code{File} on the menu bar, and then the @code{New} command.
1489 The Workspace area will become white, and you can now
1490 enter the source program explicitly.
1491 Type the following text
1493 @smallexample @c ada
1495 with Ada.Text_IO; use Ada.Text_IO;
1498 Put_Line("Hello from GPS!");
1504 Select @code{File}, then @code{Save As}, and enter the source file name
1506 The file will be saved in the same directory you specified as the
1507 location of the default project file.
1509 @item @emph{Updating the project file}
1511 You need to add the new source file to the project.
1513 the @code{Project} menu and then @code{Edit project properties}.
1514 Click the @code{Main files} tab on the left, and then the
1516 Choose @file{hello.adb} from the list, and press @code{Open}.
1517 The project settings window will reflect this action.
1520 @item @emph{Building and running the program}
1522 In the main GPS window, now choose the @code{Build} menu, then @code{Make},
1523 and select @file{hello.adb}.
1524 The Messages window will display the resulting invocations of @command{gcc},
1525 @command{gnatbind}, and @command{gnatlink}
1526 (reflecting the default switch settings from the
1527 project file that you created) and then a ``successful compilation/build''
1530 To run the program, choose the @code{Build} menu, then @code{Run}, and
1531 select @command{hello}.
1532 An @emph{Arguments Selection} window will appear.
1533 There are no command line arguments, so just click @code{OK}.
1535 The Messages window will now display the program's output (the string
1536 @code{Hello from GPS}), and at the bottom of the GPS window a status
1537 update is displayed (@code{Run: hello}).
1538 Close the GPS window (or select @code{File}, then @code{Exit}) to
1539 terminate this GPS session.
1542 @node Simple Debugging with GPS
1543 @subsection Simple Debugging with GPS
1545 This section illustrates basic debugging techniques (setting breakpoints,
1546 examining/modifying variables, single stepping).
1549 @item @emph{Opening a project}
1551 Start GPS and select @code{Open existing project}; browse to
1552 specify the project file @file{sample.prj} that you had created in the
1555 @item @emph{Creating a source file}
1557 Select @code{File}, then @code{New}, and type in the following program:
1559 @smallexample @c ada
1561 with Ada.Text_IO; use Ada.Text_IO;
1562 procedure Example is
1563 Line : String (1..80);
1566 Put_Line("Type a line of text at each prompt; an empty line to exit");
1570 Put_Line (Line (1..N) );
1578 Select @code{File}, then @code{Save as}, and enter the file name
1581 @item @emph{Updating the project file}
1583 Add @code{Example} as a new main unit for the project:
1586 Select @code{Project}, then @code{Edit Project Properties}.
1589 Select the @code{Main files} tab, click @code{Add}, then
1590 select the file @file{example.adb} from the list, and
1592 You will see the file name appear in the list of main units
1598 @item @emph{Building/running the executable}
1600 To build the executable
1601 select @code{Build}, then @code{Make}, and then choose @file{example.adb}.
1603 Run the program to see its effect (in the Messages area).
1604 Each line that you enter is displayed; an empty line will
1605 cause the loop to exit and the program to terminate.
1607 @item @emph{Debugging the program}
1609 Note that the @option{-g} switches to @command{gcc} and @command{gnatlink},
1610 which are required for debugging, are on by default when you create
1612 Thus unless you intentionally remove these settings, you will be able
1613 to debug any program that you develop using GPS.
1616 @item @emph{Initializing}
1618 Select @code{Debug}, then @code{Initialize}, then @file{example}
1620 @item @emph{Setting a breakpoint}
1622 After performing the initialization step, you will observe a small
1623 icon to the right of each line number.
1624 This serves as a toggle for breakpoints; clicking the icon will
1625 set a breakpoint at the corresponding line (the icon will change to
1626 a red circle with an ``x''), and clicking it again
1627 will remove the breakpoint / reset the icon.
1629 For purposes of this example, set a breakpoint at line 10 (the
1630 statement @code{Put_Line@ (Line@ (1..N));}
1632 @item @emph{Starting program execution}
1634 Select @code{Debug}, then @code{Run}. When the
1635 @code{Program Arguments} window appears, click @code{OK}.
1636 A console window will appear; enter some line of text,
1637 e.g.@: @code{abcde}, at the prompt.
1638 The program will pause execution when it gets to the
1639 breakpoint, and the corresponding line is highlighted.
1641 @item @emph{Examining a variable}
1643 Move the mouse over one of the occurrences of the variable @code{N}.
1644 You will see the value (5) displayed, in ``tool tip'' fashion.
1645 Right click on @code{N}, select @code{Debug}, then select @code{Display N}.
1646 You will see information about @code{N} appear in the @code{Debugger Data}
1647 pane, showing the value as 5.
1649 @item @emph{Assigning a new value to a variable}
1651 Right click on the @code{N} in the @code{Debugger Data} pane, and
1652 select @code{Set value of N}.
1653 When the input window appears, enter the value @code{4} and click
1655 This value does not automatically appear in the @code{Debugger Data}
1656 pane; to see it, right click again on the @code{N} in the
1657 @code{Debugger Data} pane and select @code{Update value}.
1658 The new value, 4, will appear in red.
1660 @item @emph{Single stepping}
1662 Select @code{Debug}, then @code{Next}.
1663 This will cause the next statement to be executed, in this case the
1664 call of @code{Put_Line} with the string slice.
1665 Notice in the console window that the displayed string is simply
1666 @code{abcd} and not @code{abcde} which you had entered.
1667 This is because the upper bound of the slice is now 4 rather than 5.
1669 @item @emph{Removing a breakpoint}
1671 Toggle the breakpoint icon at line 10.
1673 @item @emph{Resuming execution from a breakpoint}
1675 Select @code{Debug}, then @code{Continue}.
1676 The program will reach the next iteration of the loop, and
1677 wait for input after displaying the prompt.
1678 This time, just hit the @kbd{Enter} key.
1679 The value of @code{N} will be 0, and the program will terminate.
1680 The console window will disappear.
1685 @node The GNAT Compilation Model
1686 @chapter The GNAT Compilation Model
1687 @cindex GNAT compilation model
1688 @cindex Compilation model
1691 * Source Representation::
1692 * Foreign Language Representation::
1693 * File Naming Rules::
1694 * Using Other File Names::
1695 * Alternative File Naming Schemes::
1696 * Generating Object Files::
1697 * Source Dependencies::
1698 * The Ada Library Information Files::
1699 * Binding an Ada Program::
1700 * Mixed Language Programming::
1702 * Building Mixed Ada & C++ Programs::
1703 * Comparison between GNAT and C/C++ Compilation Models::
1705 * Comparison between GNAT and Conventional Ada Library Models::
1707 * Placement of temporary files::
1712 This chapter describes the compilation model used by GNAT. Although
1713 similar to that used by other languages, such as C and C++, this model
1714 is substantially different from the traditional Ada compilation models,
1715 which are based on a library. The model is initially described without
1716 reference to the library-based model. If you have not previously used an
1717 Ada compiler, you need only read the first part of this chapter. The
1718 last section describes and discusses the differences between the GNAT
1719 model and the traditional Ada compiler models. If you have used other
1720 Ada compilers, this section will help you to understand those
1721 differences, and the advantages of the GNAT model.
1723 @node Source Representation
1724 @section Source Representation
1728 Ada source programs are represented in standard text files, using
1729 Latin-1 coding. Latin-1 is an 8-bit code that includes the familiar
1730 7-bit ASCII set, plus additional characters used for
1731 representing foreign languages (@pxref{Foreign Language Representation}
1732 for support of non-USA character sets). The format effector characters
1733 are represented using their standard ASCII encodings, as follows:
1738 Vertical tab, @code{16#0B#}
1742 Horizontal tab, @code{16#09#}
1746 Carriage return, @code{16#0D#}
1750 Line feed, @code{16#0A#}
1754 Form feed, @code{16#0C#}
1758 Source files are in standard text file format. In addition, GNAT will
1759 recognize a wide variety of stream formats, in which the end of
1760 physical lines is marked by any of the following sequences:
1761 @code{LF}, @code{CR}, @code{CR-LF}, or @code{LF-CR}. This is useful
1762 in accommodating files that are imported from other operating systems.
1764 @cindex End of source file
1765 @cindex Source file, end
1767 The end of a source file is normally represented by the physical end of
1768 file. However, the control character @code{16#1A#} (@code{SUB}) is also
1769 recognized as signalling the end of the source file. Again, this is
1770 provided for compatibility with other operating systems where this
1771 code is used to represent the end of file.
1773 Each file contains a single Ada compilation unit, including any pragmas
1774 associated with the unit. For example, this means you must place a
1775 package declaration (a package @dfn{spec}) and the corresponding body in
1776 separate files. An Ada @dfn{compilation} (which is a sequence of
1777 compilation units) is represented using a sequence of files. Similarly,
1778 you will place each subunit or child unit in a separate file.
1780 @node Foreign Language Representation
1781 @section Foreign Language Representation
1784 GNAT supports the standard character sets defined in Ada as well as
1785 several other non-standard character sets for use in localized versions
1786 of the compiler (@pxref{Character Set Control}).
1789 * Other 8-Bit Codes::
1790 * Wide Character Encodings::
1798 The basic character set is Latin-1. This character set is defined by ISO
1799 standard 8859, part 1. The lower half (character codes @code{16#00#}
1800 @dots{} @code{16#7F#)} is identical to standard ASCII coding, but the upper half
1801 is used to represent additional characters. These include extended letters
1802 used by European languages, such as French accents, the vowels with umlauts
1803 used in German, and the extra letter A-ring used in Swedish.
1805 @findex Ada.Characters.Latin_1
1806 For a complete list of Latin-1 codes and their encodings, see the source
1807 file of library unit @code{Ada.Characters.Latin_1} in file
1808 @file{a-chlat1.ads}.
1809 You may use any of these extended characters freely in character or
1810 string literals. In addition, the extended characters that represent
1811 letters can be used in identifiers.
1813 @node Other 8-Bit Codes
1814 @subsection Other 8-Bit Codes
1817 GNAT also supports several other 8-bit coding schemes:
1820 @item ISO 8859-2 (Latin-2)
1823 Latin-2 letters allowed in identifiers, with uppercase and lowercase
1826 @item ISO 8859-3 (Latin-3)
1829 Latin-3 letters allowed in identifiers, with uppercase and lowercase
1832 @item ISO 8859-4 (Latin-4)
1835 Latin-4 letters allowed in identifiers, with uppercase and lowercase
1838 @item ISO 8859-5 (Cyrillic)
1841 ISO 8859-5 letters (Cyrillic) allowed in identifiers, with uppercase and
1842 lowercase equivalence.
1844 @item ISO 8859-15 (Latin-9)
1847 ISO 8859-15 (Latin-9) letters allowed in identifiers, with uppercase and
1848 lowercase equivalence
1850 @item IBM PC (code page 437)
1851 @cindex code page 437
1852 This code page is the normal default for PCs in the U.S. It corresponds
1853 to the original IBM PC character set. This set has some, but not all, of
1854 the extended Latin-1 letters, but these letters do not have the same
1855 encoding as Latin-1. In this mode, these letters are allowed in
1856 identifiers with uppercase and lowercase equivalence.
1858 @item IBM PC (code page 850)
1859 @cindex code page 850
1860 This code page is a modification of 437 extended to include all the
1861 Latin-1 letters, but still not with the usual Latin-1 encoding. In this
1862 mode, all these letters are allowed in identifiers with uppercase and
1863 lowercase equivalence.
1865 @item Full Upper 8-bit
1866 Any character in the range 80-FF allowed in identifiers, and all are
1867 considered distinct. In other words, there are no uppercase and lowercase
1868 equivalences in this range. This is useful in conjunction with
1869 certain encoding schemes used for some foreign character sets (e.g.,
1870 the typical method of representing Chinese characters on the PC).
1873 No upper-half characters in the range 80-FF are allowed in identifiers.
1874 This gives Ada 83 compatibility for identifier names.
1878 For precise data on the encodings permitted, and the uppercase and lowercase
1879 equivalences that are recognized, see the file @file{csets.adb} in
1880 the GNAT compiler sources. You will need to obtain a full source release
1881 of GNAT to obtain this file.
1883 @node Wide Character Encodings
1884 @subsection Wide Character Encodings
1887 GNAT allows wide character codes to appear in character and string
1888 literals, and also optionally in identifiers, by means of the following
1889 possible encoding schemes:
1894 In this encoding, a wide character is represented by the following five
1902 Where @code{a}, @code{b}, @code{c}, @code{d} are the four hexadecimal
1903 characters (using uppercase letters) of the wide character code. For
1904 example, ESC A345 is used to represent the wide character with code
1906 This scheme is compatible with use of the full Wide_Character set.
1908 @item Upper-Half Coding
1909 @cindex Upper-Half Coding
1910 The wide character with encoding @code{16#abcd#} where the upper bit is on
1911 (in other words, ``a'' is in the range 8-F) is represented as two bytes,
1912 @code{16#ab#} and @code{16#cd#}. The second byte cannot be a format control
1913 character, but is not required to be in the upper half. This method can
1914 be also used for shift-JIS or EUC, where the internal coding matches the
1917 @item Shift JIS Coding
1918 @cindex Shift JIS Coding
1919 A wide character is represented by a two-character sequence,
1921 @code{16#cd#}, with the restrictions described for upper-half encoding as
1922 described above. The internal character code is the corresponding JIS
1923 character according to the standard algorithm for Shift-JIS
1924 conversion. Only characters defined in the JIS code set table can be
1925 used with this encoding method.
1929 A wide character is represented by a two-character sequence
1931 @code{16#cd#}, with both characters being in the upper half. The internal
1932 character code is the corresponding JIS character according to the EUC
1933 encoding algorithm. Only characters defined in the JIS code set table
1934 can be used with this encoding method.
1937 A wide character is represented using
1938 UCS Transformation Format 8 (UTF-8) as defined in Annex R of ISO
1939 10646-1/Am.2. Depending on the character value, the representation
1940 is a one, two, or three byte sequence:
1945 16#0000#-16#007f#: 2#0@var{xxxxxxx}#
1946 16#0080#-16#07ff#: 2#110@var{xxxxx}# 2#10@var{xxxxxx}#
1947 16#0800#-16#ffff#: 2#1110@var{xxxx}# 2#10@var{xxxxxx}# 2#10@var{xxxxxx}#
1952 where the @var{xxx} bits correspond to the left-padded bits of the
1953 16-bit character value. Note that all lower half ASCII characters
1954 are represented as ASCII bytes and all upper half characters and
1955 other wide characters are represented as sequences of upper-half
1956 (The full UTF-8 scheme allows for encoding 31-bit characters as
1957 6-byte sequences, but in this implementation, all UTF-8 sequences
1958 of four or more bytes length will be treated as illegal).
1959 @item Brackets Coding
1960 In this encoding, a wide character is represented by the following eight
1968 Where @code{a}, @code{b}, @code{c}, @code{d} are the four hexadecimal
1969 characters (using uppercase letters) of the wide character code. For
1970 example, [``A345''] is used to represent the wide character with code
1971 @code{16#A345#}. It is also possible (though not required) to use the
1972 Brackets coding for upper half characters. For example, the code
1973 @code{16#A3#} can be represented as @code{[``A3'']}.
1975 This scheme is compatible with use of the full Wide_Character set,
1976 and is also the method used for wide character encoding in the standard
1977 ACVC (Ada Compiler Validation Capability) test suite distributions.
1982 Note: Some of these coding schemes do not permit the full use of the
1983 Ada character set. For example, neither Shift JIS, nor EUC allow the
1984 use of the upper half of the Latin-1 set.
1986 @node File Naming Rules
1987 @section File Naming Rules
1990 The default file name is determined by the name of the unit that the
1991 file contains. The name is formed by taking the full expanded name of
1992 the unit and replacing the separating dots with hyphens and using
1993 ^lowercase^uppercase^ for all letters.
1995 An exception arises if the file name generated by the above rules starts
1996 with one of the characters
1998 @samp{A}, @samp{G}, @samp{I}, or @samp{S},
2001 @samp{a}, @samp{g}, @samp{i}, or @samp{s},
2003 and the second character is a
2004 minus. In this case, the character ^tilde^dollar sign^ is used in place
2005 of the minus. The reason for this special rule is to avoid clashes with
2006 the standard names for child units of the packages System, Ada,
2007 Interfaces, and GNAT, which use the prefixes
2009 @samp{S-}, @samp{A-}, @samp{I-}, and @samp{G-},
2012 @samp{s-}, @samp{a-}, @samp{i-}, and @samp{g-},
2016 The file extension is @file{.ads} for a spec and
2017 @file{.adb} for a body. The following list shows some
2018 examples of these rules.
2025 @item arith_functions.ads
2026 Arith_Functions (package spec)
2027 @item arith_functions.adb
2028 Arith_Functions (package body)
2030 Func.Spec (child package spec)
2032 Func.Spec (child package body)
2034 Sub (subunit of Main)
2035 @item ^a~bad.adb^A$BAD.ADB^
2036 A.Bad (child package body)
2040 Following these rules can result in excessively long
2041 file names if corresponding
2042 unit names are long (for example, if child units or subunits are
2043 heavily nested). An option is available to shorten such long file names
2044 (called file name ``krunching''). This may be particularly useful when
2045 programs being developed with GNAT are to be used on operating systems
2046 with limited file name lengths. @xref{Using gnatkr}.
2048 Of course, no file shortening algorithm can guarantee uniqueness over
2049 all possible unit names; if file name krunching is used, it is your
2050 responsibility to ensure no name clashes occur. Alternatively you
2051 can specify the exact file names that you want used, as described
2052 in the next section. Finally, if your Ada programs are migrating from a
2053 compiler with a different naming convention, you can use the gnatchop
2054 utility to produce source files that follow the GNAT naming conventions.
2055 (For details @pxref{Renaming Files Using gnatchop}.)
2057 Note: in the case of @code{Windows NT/XP} or @code{OpenVMS} operating
2058 systems, case is not significant. So for example on @code{Windows XP}
2059 if the canonical name is @code{main-sub.adb}, you can use the file name
2060 @code{Main-Sub.adb} instead. However, case is significant for other
2061 operating systems, so for example, if you want to use other than
2062 canonically cased file names on a Unix system, you need to follow
2063 the procedures described in the next section.
2065 @node Using Other File Names
2066 @section Using Other File Names
2070 In the previous section, we have described the default rules used by
2071 GNAT to determine the file name in which a given unit resides. It is
2072 often convenient to follow these default rules, and if you follow them,
2073 the compiler knows without being explicitly told where to find all
2076 However, in some cases, particularly when a program is imported from
2077 another Ada compiler environment, it may be more convenient for the
2078 programmer to specify which file names contain which units. GNAT allows
2079 arbitrary file names to be used by means of the Source_File_Name pragma.
2080 The form of this pragma is as shown in the following examples:
2081 @cindex Source_File_Name pragma
2083 @smallexample @c ada
2085 pragma Source_File_Name (My_Utilities.Stacks,
2086 Spec_File_Name => "myutilst_a.ada");
2087 pragma Source_File_name (My_Utilities.Stacks,
2088 Body_File_Name => "myutilst.ada");
2093 As shown in this example, the first argument for the pragma is the unit
2094 name (in this example a child unit). The second argument has the form
2095 of a named association. The identifier
2096 indicates whether the file name is for a spec or a body;
2097 the file name itself is given by a string literal.
2099 The source file name pragma is a configuration pragma, which means that
2100 normally it will be placed in the @file{gnat.adc}
2101 file used to hold configuration
2102 pragmas that apply to a complete compilation environment.
2103 For more details on how the @file{gnat.adc} file is created and used
2104 see @ref{Handling of Configuration Pragmas}.
2105 @cindex @file{gnat.adc}
2108 GNAT allows completely arbitrary file names to be specified using the
2109 source file name pragma. However, if the file name specified has an
2110 extension other than @file{.ads} or @file{.adb} it is necessary to use
2111 a special syntax when compiling the file. The name in this case must be
2112 preceded by the special sequence @option{-x} followed by a space and the name
2113 of the language, here @code{ada}, as in:
2116 $ gcc -c -x ada peculiar_file_name.sim
2121 @command{gnatmake} handles non-standard file names in the usual manner (the
2122 non-standard file name for the main program is simply used as the
2123 argument to gnatmake). Note that if the extension is also non-standard,
2124 then it must be included in the @command{gnatmake} command, it may not
2127 @node Alternative File Naming Schemes
2128 @section Alternative File Naming Schemes
2129 @cindex File naming schemes, alternative
2132 In the previous section, we described the use of the @code{Source_File_Name}
2133 pragma to allow arbitrary names to be assigned to individual source files.
2134 However, this approach requires one pragma for each file, and especially in
2135 large systems can result in very long @file{gnat.adc} files, and also create
2136 a maintenance problem.
2138 GNAT also provides a facility for specifying systematic file naming schemes
2139 other than the standard default naming scheme previously described. An
2140 alternative scheme for naming is specified by the use of
2141 @code{Source_File_Name} pragmas having the following format:
2142 @cindex Source_File_Name pragma
2144 @smallexample @c ada
2145 pragma Source_File_Name (
2146 Spec_File_Name => FILE_NAME_PATTERN
2147 @r{[},Casing => CASING_SPEC@r{]}
2148 @r{[},Dot_Replacement => STRING_LITERAL@r{]});
2150 pragma Source_File_Name (
2151 Body_File_Name => FILE_NAME_PATTERN
2152 @r{[},Casing => CASING_SPEC@r{]}
2153 @r{[},Dot_Replacement => STRING_LITERAL@r{]});
2155 pragma Source_File_Name (
2156 Subunit_File_Name => FILE_NAME_PATTERN
2157 @r{[},Casing => CASING_SPEC@r{]}
2158 @r{[},Dot_Replacement => STRING_LITERAL@r{]});
2160 FILE_NAME_PATTERN ::= STRING_LITERAL
2161 CASING_SPEC ::= Lowercase | Uppercase | Mixedcase
2165 The @code{FILE_NAME_PATTERN} string shows how the file name is constructed.
2166 It contains a single asterisk character, and the unit name is substituted
2167 systematically for this asterisk. The optional parameter
2168 @code{Casing} indicates
2169 whether the unit name is to be all upper-case letters, all lower-case letters,
2170 or mixed-case. If no
2171 @code{Casing} parameter is used, then the default is all
2172 ^lower-case^upper-case^.
2174 The optional @code{Dot_Replacement} string is used to replace any periods
2175 that occur in subunit or child unit names. If no @code{Dot_Replacement}
2176 argument is used then separating dots appear unchanged in the resulting
2178 Although the above syntax indicates that the
2179 @code{Casing} argument must appear
2180 before the @code{Dot_Replacement} argument, but it
2181 is also permissible to write these arguments in the opposite order.
2183 As indicated, it is possible to specify different naming schemes for
2184 bodies, specs, and subunits. Quite often the rule for subunits is the
2185 same as the rule for bodies, in which case, there is no need to give
2186 a separate @code{Subunit_File_Name} rule, and in this case the
2187 @code{Body_File_name} rule is used for subunits as well.
2189 The separate rule for subunits can also be used to implement the rather
2190 unusual case of a compilation environment (e.g.@: a single directory) which
2191 contains a subunit and a child unit with the same unit name. Although
2192 both units cannot appear in the same partition, the Ada Reference Manual
2193 allows (but does not require) the possibility of the two units coexisting
2194 in the same environment.
2196 The file name translation works in the following steps:
2201 If there is a specific @code{Source_File_Name} pragma for the given unit,
2202 then this is always used, and any general pattern rules are ignored.
2205 If there is a pattern type @code{Source_File_Name} pragma that applies to
2206 the unit, then the resulting file name will be used if the file exists. If
2207 more than one pattern matches, the latest one will be tried first, and the
2208 first attempt resulting in a reference to a file that exists will be used.
2211 If no pattern type @code{Source_File_Name} pragma that applies to the unit
2212 for which the corresponding file exists, then the standard GNAT default
2213 naming rules are used.
2218 As an example of the use of this mechanism, consider a commonly used scheme
2219 in which file names are all lower case, with separating periods copied
2220 unchanged to the resulting file name, and specs end with @file{.1.ada}, and
2221 bodies end with @file{.2.ada}. GNAT will follow this scheme if the following
2224 @smallexample @c ada
2225 pragma Source_File_Name
2226 (Spec_File_Name => "*.1.ada");
2227 pragma Source_File_Name
2228 (Body_File_Name => "*.2.ada");
2232 The default GNAT scheme is actually implemented by providing the following
2233 default pragmas internally:
2235 @smallexample @c ada
2236 pragma Source_File_Name
2237 (Spec_File_Name => "*.ads", Dot_Replacement => "-");
2238 pragma Source_File_Name
2239 (Body_File_Name => "*.adb", Dot_Replacement => "-");
2243 Our final example implements a scheme typically used with one of the
2244 Ada 83 compilers, where the separator character for subunits was ``__''
2245 (two underscores), specs were identified by adding @file{_.ADA}, bodies
2246 by adding @file{.ADA}, and subunits by
2247 adding @file{.SEP}. All file names were
2248 upper case. Child units were not present of course since this was an
2249 Ada 83 compiler, but it seems reasonable to extend this scheme to use
2250 the same double underscore separator for child units.
2252 @smallexample @c ada
2253 pragma Source_File_Name
2254 (Spec_File_Name => "*_.ADA",
2255 Dot_Replacement => "__",
2256 Casing = Uppercase);
2257 pragma Source_File_Name
2258 (Body_File_Name => "*.ADA",
2259 Dot_Replacement => "__",
2260 Casing = Uppercase);
2261 pragma Source_File_Name
2262 (Subunit_File_Name => "*.SEP",
2263 Dot_Replacement => "__",
2264 Casing = Uppercase);
2267 @node Generating Object Files
2268 @section Generating Object Files
2271 An Ada program consists of a set of source files, and the first step in
2272 compiling the program is to generate the corresponding object files.
2273 These are generated by compiling a subset of these source files.
2274 The files you need to compile are the following:
2278 If a package spec has no body, compile the package spec to produce the
2279 object file for the package.
2282 If a package has both a spec and a body, compile the body to produce the
2283 object file for the package. The source file for the package spec need
2284 not be compiled in this case because there is only one object file, which
2285 contains the code for both the spec and body of the package.
2288 For a subprogram, compile the subprogram body to produce the object file
2289 for the subprogram. The spec, if one is present, is as usual in a
2290 separate file, and need not be compiled.
2294 In the case of subunits, only compile the parent unit. A single object
2295 file is generated for the entire subunit tree, which includes all the
2299 Compile child units independently of their parent units
2300 (though, of course, the spec of all the ancestor unit must be present in order
2301 to compile a child unit).
2305 Compile generic units in the same manner as any other units. The object
2306 files in this case are small dummy files that contain at most the
2307 flag used for elaboration checking. This is because GNAT always handles generic
2308 instantiation by means of macro expansion. However, it is still necessary to
2309 compile generic units, for dependency checking and elaboration purposes.
2313 The preceding rules describe the set of files that must be compiled to
2314 generate the object files for a program. Each object file has the same
2315 name as the corresponding source file, except that the extension is
2318 You may wish to compile other files for the purpose of checking their
2319 syntactic and semantic correctness. For example, in the case where a
2320 package has a separate spec and body, you would not normally compile the
2321 spec. However, it is convenient in practice to compile the spec to make
2322 sure it is error-free before compiling clients of this spec, because such
2323 compilations will fail if there is an error in the spec.
2325 GNAT provides an option for compiling such files purely for the
2326 purposes of checking correctness; such compilations are not required as
2327 part of the process of building a program. To compile a file in this
2328 checking mode, use the @option{-gnatc} switch.
2330 @node Source Dependencies
2331 @section Source Dependencies
2334 A given object file clearly depends on the source file which is compiled
2335 to produce it. Here we are using @dfn{depends} in the sense of a typical
2336 @code{make} utility; in other words, an object file depends on a source
2337 file if changes to the source file require the object file to be
2339 In addition to this basic dependency, a given object may depend on
2340 additional source files as follows:
2344 If a file being compiled @code{with}'s a unit @var{X}, the object file
2345 depends on the file containing the spec of unit @var{X}. This includes
2346 files that are @code{with}'ed implicitly either because they are parents
2347 of @code{with}'ed child units or they are run-time units required by the
2348 language constructs used in a particular unit.
2351 If a file being compiled instantiates a library level generic unit, the
2352 object file depends on both the spec and body files for this generic
2356 If a file being compiled instantiates a generic unit defined within a
2357 package, the object file depends on the body file for the package as
2358 well as the spec file.
2362 @cindex @option{-gnatn} switch
2363 If a file being compiled contains a call to a subprogram for which
2364 pragma @code{Inline} applies and inlining is activated with the
2365 @option{-gnatn} switch, the object file depends on the file containing the
2366 body of this subprogram as well as on the file containing the spec. Note
2367 that for inlining to actually occur as a result of the use of this switch,
2368 it is necessary to compile in optimizing mode.
2370 @cindex @option{-gnatN} switch
2371 The use of @option{-gnatN} activates inlining optimization
2372 that is performed by the front end of the compiler. This inlining does
2373 not require that the code generation be optimized. Like @option{-gnatn},
2374 the use of this switch generates additional dependencies.
2376 @option{-gnatN} automatically implies @option{-gnatn} so it is not necessary
2377 to specify both options.
2379 When using a gcc-based back end (in practice this means using any version
2380 of GNAT other than the JGNAT, .NET or GNAAMP versions), then the use of
2381 @option{-gnatN} is deprecated, and the use of @option{-gnatn} is preferred.
2382 Historically front end inlining was more extensive than the gcc back end
2383 inlining, but that is no longer the case.
2386 If an object file @file{O} depends on the proper body of a subunit through
2387 inlining or instantiation, it depends on the parent unit of the subunit.
2388 This means that any modification of the parent unit or one of its subunits
2389 affects the compilation of @file{O}.
2392 The object file for a parent unit depends on all its subunit body files.
2395 The previous two rules meant that for purposes of computing dependencies and
2396 recompilation, a body and all its subunits are treated as an indivisible whole.
2399 These rules are applied transitively: if unit @code{A} @code{with}'s
2400 unit @code{B}, whose elaboration calls an inlined procedure in package
2401 @code{C}, the object file for unit @code{A} will depend on the body of
2402 @code{C}, in file @file{c.adb}.
2404 The set of dependent files described by these rules includes all the
2405 files on which the unit is semantically dependent, as dictated by the
2406 Ada language standard. However, it is a superset of what the
2407 standard describes, because it includes generic, inline, and subunit
2410 An object file must be recreated by recompiling the corresponding source
2411 file if any of the source files on which it depends are modified. For
2412 example, if the @code{make} utility is used to control compilation,
2413 the rule for an Ada object file must mention all the source files on
2414 which the object file depends, according to the above definition.
2415 The determination of the necessary
2416 recompilations is done automatically when one uses @command{gnatmake}.
2419 @node The Ada Library Information Files
2420 @section The Ada Library Information Files
2421 @cindex Ada Library Information files
2422 @cindex @file{ALI} files
2425 Each compilation actually generates two output files. The first of these
2426 is the normal object file that has a @file{.o} extension. The second is a
2427 text file containing full dependency information. It has the same
2428 name as the source file, but an @file{.ali} extension.
2429 This file is known as the Ada Library Information (@file{ALI}) file.
2430 The following information is contained in the @file{ALI} file.
2434 Version information (indicates which version of GNAT was used to compile
2435 the unit(s) in question)
2438 Main program information (including priority and time slice settings,
2439 as well as the wide character encoding used during compilation).
2442 List of arguments used in the @command{gcc} command for the compilation
2445 Attributes of the unit, including configuration pragmas used, an indication
2446 of whether the compilation was successful, exception model used etc.
2449 A list of relevant restrictions applying to the unit (used for consistency)
2453 Categorization information (e.g.@: use of pragma @code{Pure}).
2456 Information on all @code{with}'ed units, including presence of
2457 @code{Elaborate} or @code{Elaborate_All} pragmas.
2460 Information from any @code{Linker_Options} pragmas used in the unit
2463 Information on the use of @code{Body_Version} or @code{Version}
2464 attributes in the unit.
2467 Dependency information. This is a list of files, together with
2468 time stamp and checksum information. These are files on which
2469 the unit depends in the sense that recompilation is required
2470 if any of these units are modified.
2473 Cross-reference data. Contains information on all entities referenced
2474 in the unit. Used by tools like @code{gnatxref} and @code{gnatfind} to
2475 provide cross-reference information.
2480 For a full detailed description of the format of the @file{ALI} file,
2481 see the source of the body of unit @code{Lib.Writ}, contained in file
2482 @file{lib-writ.adb} in the GNAT compiler sources.
2484 @node Binding an Ada Program
2485 @section Binding an Ada Program
2488 When using languages such as C and C++, once the source files have been
2489 compiled the only remaining step in building an executable program
2490 is linking the object modules together. This means that it is possible to
2491 link an inconsistent version of a program, in which two units have
2492 included different versions of the same header.
2494 The rules of Ada do not permit such an inconsistent program to be built.
2495 For example, if two clients have different versions of the same package,
2496 it is illegal to build a program containing these two clients.
2497 These rules are enforced by the GNAT binder, which also determines an
2498 elaboration order consistent with the Ada rules.
2500 The GNAT binder is run after all the object files for a program have
2501 been created. It is given the name of the main program unit, and from
2502 this it determines the set of units required by the program, by reading the
2503 corresponding ALI files. It generates error messages if the program is
2504 inconsistent or if no valid order of elaboration exists.
2506 If no errors are detected, the binder produces a main program, in Ada by
2507 default, that contains calls to the elaboration procedures of those
2508 compilation unit that require them, followed by
2509 a call to the main program. This Ada program is compiled to generate the
2510 object file for the main program. The name of
2511 the Ada file is @file{b~@var{xxx}.adb} (with the corresponding spec
2512 @file{b~@var{xxx}.ads}) where @var{xxx} is the name of the
2515 Finally, the linker is used to build the resulting executable program,
2516 using the object from the main program from the bind step as well as the
2517 object files for the Ada units of the program.
2519 @node Mixed Language Programming
2520 @section Mixed Language Programming
2521 @cindex Mixed Language Programming
2524 This section describes how to develop a mixed-language program,
2525 specifically one that comprises units in both Ada and C.
2528 * Interfacing to C::
2529 * Calling Conventions::
2532 @node Interfacing to C
2533 @subsection Interfacing to C
2535 Interfacing Ada with a foreign language such as C involves using
2536 compiler directives to import and/or export entity definitions in each
2537 language---using @code{extern} statements in C, for instance, and the
2538 @code{Import}, @code{Export}, and @code{Convention} pragmas in Ada.
2539 A full treatment of these topics is provided in Appendix B, section 1
2540 of the Ada Reference Manual.
2542 There are two ways to build a program using GNAT that contains some Ada
2543 sources and some foreign language sources, depending on whether or not
2544 the main subprogram is written in Ada. Here is a source example with
2545 the main subprogram in Ada:
2551 void print_num (int num)
2553 printf ("num is %d.\n", num);
2559 /* num_from_Ada is declared in my_main.adb */
2560 extern int num_from_Ada;
2564 return num_from_Ada;
2568 @smallexample @c ada
2570 procedure My_Main is
2572 -- Declare then export an Integer entity called num_from_Ada
2573 My_Num : Integer := 10;
2574 pragma Export (C, My_Num, "num_from_Ada");
2576 -- Declare an Ada function spec for Get_Num, then use
2577 -- C function get_num for the implementation.
2578 function Get_Num return Integer;
2579 pragma Import (C, Get_Num, "get_num");
2581 -- Declare an Ada procedure spec for Print_Num, then use
2582 -- C function print_num for the implementation.
2583 procedure Print_Num (Num : Integer);
2584 pragma Import (C, Print_Num, "print_num");
2587 Print_Num (Get_Num);
2593 To build this example, first compile the foreign language files to
2594 generate object files:
2596 ^gcc -c file1.c^gcc -c FILE1.C^
2597 ^gcc -c file2.c^gcc -c FILE2.C^
2601 Then, compile the Ada units to produce a set of object files and ALI
2604 gnatmake ^-c^/ACTIONS=COMPILE^ my_main.adb
2608 Run the Ada binder on the Ada main program:
2610 gnatbind my_main.ali
2614 Link the Ada main program, the Ada objects and the other language
2617 gnatlink my_main.ali file1.o file2.o
2621 The last three steps can be grouped in a single command:
2623 gnatmake my_main.adb -largs file1.o file2.o
2626 @cindex Binder output file
2628 If the main program is in a language other than Ada, then you may have
2629 more than one entry point into the Ada subsystem. You must use a special
2630 binder option to generate callable routines that initialize and
2631 finalize the Ada units (@pxref{Binding with Non-Ada Main Programs}).
2632 Calls to the initialization and finalization routines must be inserted
2633 in the main program, or some other appropriate point in the code. The
2634 call to initialize the Ada units must occur before the first Ada
2635 subprogram is called, and the call to finalize the Ada units must occur
2636 after the last Ada subprogram returns. The binder will place the
2637 initialization and finalization subprograms into the
2638 @file{b~@var{xxx}.adb} file where they can be accessed by your C
2639 sources. To illustrate, we have the following example:
2643 extern void adainit (void);
2644 extern void adafinal (void);
2645 extern int add (int, int);
2646 extern int sub (int, int);
2648 int main (int argc, char *argv[])
2654 /* Should print "21 + 7 = 28" */
2655 printf ("%d + %d = %d\n", a, b, add (a, b));
2656 /* Should print "21 - 7 = 14" */
2657 printf ("%d - %d = %d\n", a, b, sub (a, b));
2663 @smallexample @c ada
2666 function Add (A, B : Integer) return Integer;
2667 pragma Export (C, Add, "add");
2671 package body Unit1 is
2672 function Add (A, B : Integer) return Integer is
2680 function Sub (A, B : Integer) return Integer;
2681 pragma Export (C, Sub, "sub");
2685 package body Unit2 is
2686 function Sub (A, B : Integer) return Integer is
2695 The build procedure for this application is similar to the last
2696 example's. First, compile the foreign language files to generate object
2699 ^gcc -c main.c^gcc -c main.c^
2703 Next, compile the Ada units to produce a set of object files and ALI
2706 gnatmake ^-c^/ACTIONS=COMPILE^ unit1.adb
2707 gnatmake ^-c^/ACTIONS=COMPILE^ unit2.adb
2711 Run the Ada binder on every generated ALI file. Make sure to use the
2712 @option{-n} option to specify a foreign main program:
2714 gnatbind ^-n^/NOMAIN^ unit1.ali unit2.ali
2718 Link the Ada main program, the Ada objects and the foreign language
2719 objects. You need only list the last ALI file here:
2721 gnatlink unit2.ali main.o -o exec_file
2724 This procedure yields a binary executable called @file{exec_file}.
2728 Depending on the circumstances (for example when your non-Ada main object
2729 does not provide symbol @code{main}), you may also need to instruct the
2730 GNAT linker not to include the standard startup objects by passing the
2731 @option{^-nostartfiles^/NOSTART_FILES^} switch to @command{gnatlink}.
2733 @node Calling Conventions
2734 @subsection Calling Conventions
2735 @cindex Foreign Languages
2736 @cindex Calling Conventions
2737 GNAT follows standard calling sequence conventions and will thus interface
2738 to any other language that also follows these conventions. The following
2739 Convention identifiers are recognized by GNAT:
2742 @cindex Interfacing to Ada
2743 @cindex Other Ada compilers
2744 @cindex Convention Ada
2746 This indicates that the standard Ada calling sequence will be
2747 used and all Ada data items may be passed without any limitations in the
2748 case where GNAT is used to generate both the caller and callee. It is also
2749 possible to mix GNAT generated code and code generated by another Ada
2750 compiler. In this case, the data types should be restricted to simple
2751 cases, including primitive types. Whether complex data types can be passed
2752 depends on the situation. Probably it is safe to pass simple arrays, such
2753 as arrays of integers or floats. Records may or may not work, depending
2754 on whether both compilers lay them out identically. Complex structures
2755 involving variant records, access parameters, tasks, or protected types,
2756 are unlikely to be able to be passed.
2758 Note that in the case of GNAT running
2759 on a platform that supports HP Ada 83, a higher degree of compatibility
2760 can be guaranteed, and in particular records are layed out in an identical
2761 manner in the two compilers. Note also that if output from two different
2762 compilers is mixed, the program is responsible for dealing with elaboration
2763 issues. Probably the safest approach is to write the main program in the
2764 version of Ada other than GNAT, so that it takes care of its own elaboration
2765 requirements, and then call the GNAT-generated adainit procedure to ensure
2766 elaboration of the GNAT components. Consult the documentation of the other
2767 Ada compiler for further details on elaboration.
2769 However, it is not possible to mix the tasking run time of GNAT and
2770 HP Ada 83, All the tasking operations must either be entirely within
2771 GNAT compiled sections of the program, or entirely within HP Ada 83
2772 compiled sections of the program.
2774 @cindex Interfacing to Assembly
2775 @cindex Convention Assembler
2777 Specifies assembler as the convention. In practice this has the
2778 same effect as convention Ada (but is not equivalent in the sense of being
2779 considered the same convention).
2781 @cindex Convention Asm
2784 Equivalent to Assembler.
2786 @cindex Interfacing to COBOL
2787 @cindex Convention COBOL
2790 Data will be passed according to the conventions described
2791 in section B.4 of the Ada Reference Manual.
2794 @cindex Interfacing to C
2795 @cindex Convention C
2797 Data will be passed according to the conventions described
2798 in section B.3 of the Ada Reference Manual.
2800 A note on interfacing to a C ``varargs'' function:
2801 @findex C varargs function
2802 @cindex Interfacing to C varargs function
2803 @cindex varargs function interfaces
2807 In C, @code{varargs} allows a function to take a variable number of
2808 arguments. There is no direct equivalent in this to Ada. One
2809 approach that can be used is to create a C wrapper for each
2810 different profile and then interface to this C wrapper. For
2811 example, to print an @code{int} value using @code{printf},
2812 create a C function @code{printfi} that takes two arguments, a
2813 pointer to a string and an int, and calls @code{printf}.
2814 Then in the Ada program, use pragma @code{Import} to
2815 interface to @code{printfi}.
2818 It may work on some platforms to directly interface to
2819 a @code{varargs} function by providing a specific Ada profile
2820 for a particular call. However, this does not work on
2821 all platforms, since there is no guarantee that the
2822 calling sequence for a two argument normal C function
2823 is the same as for calling a @code{varargs} C function with
2824 the same two arguments.
2827 @cindex Convention Default
2832 @cindex Convention External
2839 @cindex Interfacing to C++
2840 @cindex Convention C++
2841 @item C_Plus_Plus (or CPP)
2842 This stands for C++. For most purposes this is identical to C.
2843 See the separate description of the specialized GNAT pragmas relating to
2844 C++ interfacing for further details.
2848 @cindex Interfacing to Fortran
2849 @cindex Convention Fortran
2851 Data will be passed according to the conventions described
2852 in section B.5 of the Ada Reference Manual.
2855 This applies to an intrinsic operation, as defined in the Ada
2856 Reference Manual. If a pragma Import (Intrinsic) applies to a subprogram,
2857 this means that the body of the subprogram is provided by the compiler itself,
2858 usually by means of an efficient code sequence, and that the user does not
2859 supply an explicit body for it. In an application program, the pragma may
2860 be applied to the following sets of names:
2864 Rotate_Left, Rotate_Right, Shift_Left, Shift_Right,
2865 Shift_Right_Arithmetic. The corresponding subprogram declaration must have
2866 two formal parameters. The
2867 first one must be a signed integer type or a modular type with a binary
2868 modulus, and the second parameter must be of type Natural.
2869 The return type must be the same as the type of the first argument. The size
2870 of this type can only be 8, 16, 32, or 64.
2873 Binary arithmetic operators: ``+'', ``-'', ``*'', ``/''
2874 The corresponding operator declaration must have parameters and result type
2875 that have the same root numeric type (for example, all three are long_float
2876 types). This simplifies the definition of operations that use type checking
2877 to perform dimensional checks:
2879 @smallexample @c ada
2880 type Distance is new Long_Float;
2881 type Time is new Long_Float;
2882 type Velocity is new Long_Float;
2883 function "/" (D : Distance; T : Time)
2885 pragma Import (Intrinsic, "/");
2889 This common idiom is often programmed with a generic definition and an
2890 explicit body. The pragma makes it simpler to introduce such declarations.
2891 It incurs no overhead in compilation time or code size, because it is
2892 implemented as a single machine instruction.
2895 General subprogram entities, to bind an Ada subprogram declaration to
2896 a compiler builtin by name with back-ends where such interfaces are
2897 available. A typical example is the set of ``__builtin'' functions
2898 exposed by the GCC back-end, as in the following example:
2900 @smallexample @c ada
2901 function builtin_sqrt (F : Float) return Float;
2902 pragma Import (Intrinsic, builtin_sqrt, "__builtin_sqrtf");
2905 Most of the GCC builtins are accessible this way, and as for other
2906 import conventions (e.g. C), it is the user's responsibility to ensure
2907 that the Ada subprogram profile matches the underlying builtin
2915 @cindex Convention Stdcall
2917 This is relevant only to Windows XP/2000/NT implementations of GNAT,
2918 and specifies that the @code{Stdcall} calling sequence will be used,
2919 as defined by the NT API. Nevertheless, to ease building
2920 cross-platform bindings this convention will be handled as a @code{C} calling
2921 convention on non-Windows platforms.
2924 @cindex Convention DLL
2926 This is equivalent to @code{Stdcall}.
2929 @cindex Convention Win32
2931 This is equivalent to @code{Stdcall}.
2935 @cindex Convention Stubbed
2937 This is a special convention that indicates that the compiler
2938 should provide a stub body that raises @code{Program_Error}.
2942 GNAT additionally provides a useful pragma @code{Convention_Identifier}
2943 that can be used to parametrize conventions and allow additional synonyms
2944 to be specified. For example if you have legacy code in which the convention
2945 identifier Fortran77 was used for Fortran, you can use the configuration
2948 @smallexample @c ada
2949 pragma Convention_Identifier (Fortran77, Fortran);
2953 And from now on the identifier Fortran77 may be used as a convention
2954 identifier (for example in an @code{Import} pragma) with the same
2958 @node Building Mixed Ada & C++ Programs
2959 @section Building Mixed Ada and C++ Programs
2962 A programmer inexperienced with mixed-language development may find that
2963 building an application containing both Ada and C++ code can be a
2964 challenge. This section gives a few
2965 hints that should make this task easier. The first section addresses
2966 the differences between interfacing with C and interfacing with C++.
2968 looks into the delicate problem of linking the complete application from
2969 its Ada and C++ parts. The last section gives some hints on how the GNAT
2970 run-time library can be adapted in order to allow inter-language dispatching
2971 with a new C++ compiler.
2974 * Interfacing to C++::
2975 * Linking a Mixed C++ & Ada Program::
2976 * A Simple Example::
2977 * Interfacing with C++ at the Class Level::
2980 @node Interfacing to C++
2981 @subsection Interfacing to C++
2984 GNAT supports interfacing with the G++ compiler (or any C++ compiler
2985 generating code that is compatible with the G++ Application Binary
2986 Interface ---see http://www.codesourcery.com/archives/cxx-abi).
2989 Interfacing can be done at 3 levels: simple data, subprograms, and
2990 classes. In the first two cases, GNAT offers a specific @code{Convention
2991 C_Plus_Plus} (or @code{CPP}) that behaves exactly like @code{Convention C}.
2992 Usually, C++ mangles the names of subprograms, and currently, GNAT does
2993 not provide any help to solve the demangling problem. This problem can be
2994 addressed in two ways:
2997 by modifying the C++ code in order to force a C convention using
2998 the @code{extern "C"} syntax.
3001 by figuring out the mangled name and use it as the Link_Name argument of
3006 Interfacing at the class level can be achieved by using the GNAT specific
3007 pragmas such as @code{CPP_Constructor}. @xref{Interfacing to C++,,,
3008 gnat_rm, GNAT Reference Manual}, for additional information.
3010 @node Linking a Mixed C++ & Ada Program
3011 @subsection Linking a Mixed C++ & Ada Program
3014 Usually the linker of the C++ development system must be used to link
3015 mixed applications because most C++ systems will resolve elaboration
3016 issues (such as calling constructors on global class instances)
3017 transparently during the link phase. GNAT has been adapted to ease the
3018 use of a foreign linker for the last phase. Three cases can be
3023 Using GNAT and G++ (GNU C++ compiler) from the same GCC installation:
3024 The C++ linker can simply be called by using the C++ specific driver
3025 called @code{c++}. Note that this setup is not very common because it
3026 may involve recompiling the whole GCC tree from sources, which makes it
3027 harder to upgrade the compilation system for one language without
3028 destabilizing the other.
3033 $ gnatmake ada_unit -largs file1.o file2.o --LINK=c++
3037 Using GNAT and G++ from two different GCC installations: If both
3038 compilers are on the @env{PATH}, the previous method may be used. It is
3039 important to note that environment variables such as
3040 @env{C_INCLUDE_PATH}, @env{GCC_EXEC_PREFIX}, @env{BINUTILS_ROOT}, and
3041 @env{GCC_ROOT} will affect both compilers
3042 at the same time and may make one of the two compilers operate
3043 improperly if set during invocation of the wrong compiler. It is also
3044 very important that the linker uses the proper @file{libgcc.a} GCC
3045 library -- that is, the one from the C++ compiler installation. The
3046 implicit link command as suggested in the @command{gnatmake} command
3047 from the former example can be replaced by an explicit link command with
3048 the full-verbosity option in order to verify which library is used:
3051 $ gnatlink -v -v ada_unit file1.o file2.o --LINK=c++
3053 If there is a problem due to interfering environment variables, it can
3054 be worked around by using an intermediate script. The following example
3055 shows the proper script to use when GNAT has not been installed at its
3056 default location and g++ has been installed at its default location:
3064 $ gnatlink -v -v ada_unit file1.o file2.o --LINK=./my_script
3068 Using a non-GNU C++ compiler: The commands previously described can be
3069 used to insure that the C++ linker is used. Nonetheless, you need to add
3070 a few more parameters to the link command line, depending on the exception
3073 If the @code{setjmp/longjmp} exception mechanism is used, only the paths
3074 to the libgcc libraries are required:
3079 CC $* `gcc -print-file-name=libgcc.a` `gcc -print-file-name=libgcc_eh.a`
3080 $ gnatlink ada_unit file1.o file2.o --LINK=./my_script
3083 Where CC is the name of the non-GNU C++ compiler.
3085 If the @code{zero cost} exception mechanism is used, and the platform
3086 supports automatic registration of exception tables (e.g.@: Solaris or IRIX),
3087 paths to more objects are required:
3092 CC `gcc -print-file-name=crtbegin.o` $* \
3093 `gcc -print-file-name=libgcc.a` `gcc -print-file-name=libgcc_eh.a` \
3094 `gcc -print-file-name=crtend.o`
3095 $ gnatlink ada_unit file1.o file2.o --LINK=./my_script
3098 If the @code{zero cost} exception mechanism is used, and the platform
3099 doesn't support automatic registration of exception tables (e.g.@: HP-UX,
3100 Tru64 or AIX), the simple approach described above will not work and
3101 a pre-linking phase using GNAT will be necessary.
3105 @node A Simple Example
3106 @subsection A Simple Example
3108 The following example, provided as part of the GNAT examples, shows how
3109 to achieve procedural interfacing between Ada and C++ in both
3110 directions. The C++ class A has two methods. The first method is exported
3111 to Ada by the means of an extern C wrapper function. The second method
3112 calls an Ada subprogram. On the Ada side, The C++ calls are modelled by
3113 a limited record with a layout comparable to the C++ class. The Ada
3114 subprogram, in turn, calls the C++ method. So, starting from the C++
3115 main program, the process passes back and forth between the two
3119 Here are the compilation commands:
3121 $ gnatmake -c simple_cpp_interface
3124 $ gnatbind -n simple_cpp_interface
3125 $ gnatlink simple_cpp_interface -o cpp_main --LINK=$(CPLUSPLUS)
3126 -lstdc++ ex7.o cpp_main.o
3130 Here are the corresponding sources:
3138 void adainit (void);
3139 void adafinal (void);
3140 void method1 (A *t);
3162 class A : public Origin @{
3164 void method1 (void);
3165 void method2 (int v);
3175 extern "C" @{ void ada_method2 (A *t, int v);@}
3177 void A::method1 (void)
3180 printf ("in A::method1, a_value = %d \n",a_value);
3184 void A::method2 (int v)
3186 ada_method2 (this, v);
3187 printf ("in A::method2, a_value = %d \n",a_value);
3194 printf ("in A::A, a_value = %d \n",a_value);
3198 @smallexample @c ada
3200 package body Simple_Cpp_Interface is
3202 procedure Ada_Method2 (This : in out A; V : Integer) is
3208 end Simple_Cpp_Interface;
3211 package Simple_Cpp_Interface is
3214 Vptr : System.Address;
3218 pragma Convention (C, A);
3220 procedure Method1 (This : in out A);
3221 pragma Import (C, Method1);
3223 procedure Ada_Method2 (This : in out A; V : Integer);
3224 pragma Export (C, Ada_Method2);
3226 end Simple_Cpp_Interface;
3229 @node Interfacing with C++ at the Class Level
3230 @subsection Interfacing with C++ at the Class Level
3232 In this section we demonstrate the GNAT features for interfacing with
3233 C++ by means of an example making use of Ada 2005 abstract interface
3234 types. This example consists of a classification of animals; classes
3235 have been used to model our main classification of animals, and
3236 interfaces provide support for the management of secondary
3237 classifications. We first demonstrate a case in which the types and
3238 constructors are defined on the C++ side and imported from the Ada
3239 side, and latter the reverse case.
3241 The root of our derivation will be the @code{Animal} class, with a
3242 single private attribute (the @code{Age} of the animal) and two public
3243 primitives to set and get the value of this attribute.
3248 @b{virtual} void Set_Age (int New_Age);
3249 @b{virtual} int Age ();
3255 Abstract interface types are defined in C++ by means of classes with pure
3256 virtual functions and no data members. In our example we will use two
3257 interfaces that provide support for the common management of @code{Carnivore}
3258 and @code{Domestic} animals:
3261 @b{class} Carnivore @{
3263 @b{virtual} int Number_Of_Teeth () = 0;
3266 @b{class} Domestic @{
3268 @b{virtual void} Set_Owner (char* Name) = 0;
3272 Using these declarations, we can now say that a @code{Dog} is an animal that is
3273 both Carnivore and Domestic, that is:
3276 @b{class} Dog : Animal, Carnivore, Domestic @{
3278 @b{virtual} int Number_Of_Teeth ();
3279 @b{virtual} void Set_Owner (char* Name);
3281 Dog(); // Constructor
3288 In the following examples we will assume that the previous declarations are
3289 located in a file named @code{animals.h}. The following package demonstrates
3290 how to import these C++ declarations from the Ada side:
3292 @smallexample @c ada
3293 with Interfaces.C.Strings; use Interfaces.C.Strings;
3295 type Carnivore is interface;
3296 pragma Convention (C_Plus_Plus, Carnivore);
3297 function Number_Of_Teeth (X : Carnivore)
3298 return Natural is abstract;
3300 type Domestic is interface;
3301 pragma Convention (C_Plus_Plus, Set_Owner);
3303 (X : in out Domestic;
3304 Name : Chars_Ptr) is abstract;
3306 type Animal is tagged record
3309 pragma Import (C_Plus_Plus, Animal);
3311 procedure Set_Age (X : in out Animal; Age : Integer);
3312 pragma Import (C_Plus_Plus, Set_Age);
3314 function Age (X : Animal) return Integer;
3315 pragma Import (C_Plus_Plus, Age);
3317 type Dog is new Animal and Carnivore and Domestic with record
3318 Tooth_Count : Natural;
3319 Owner : String (1 .. 30);
3321 pragma Import (C_Plus_Plus, Dog);
3323 function Number_Of_Teeth (A : Dog) return Integer;
3324 pragma Import (C_Plus_Plus, Number_Of_Teeth);
3326 procedure Set_Owner (A : in out Dog; Name : Chars_Ptr);
3327 pragma Import (C_Plus_Plus, Set_Owner);
3329 function New_Dog return Dog'Class;
3330 pragma CPP_Constructor (New_Dog);
3331 pragma Import (CPP, New_Dog, "_ZN3DogC2Ev");
3335 Thanks to the compatibility between GNAT run-time structures and the C++ ABI,
3336 interfacing with these C++ classes is easy. The only requirement is that all
3337 the primitives and components must be declared exactly in the same order in
3340 Regarding the abstract interfaces, we must indicate to the GNAT compiler by
3341 means of a @code{pragma Convention (C_Plus_Plus)}, the convention used to pass
3342 the arguments to the called primitives will be the same as for C++. For the
3343 imported classes we use @code{pragma Import} with convention @code{C_Plus_Plus}
3344 to indicate that they have been defined on the C++ side; this is required
3345 because the dispatch table associated with these tagged types will be built
3346 in the C++ side and therefore will not contain the predefined Ada primitives
3347 which Ada would otherwise expect.
3349 As the reader can see there is no need to indicate the C++ mangled names
3350 associated with each subprogram because it is assumed that all the calls to
3351 these primitives will be dispatching calls. The only exception is the
3352 constructor, which must be registered with the compiler by means of
3353 @code{pragma CPP_Constructor} and needs to provide its associated C++
3354 mangled name because the Ada compiler generates direct calls to it.
3356 With the above packages we can now declare objects of type Dog on the Ada side
3357 and dispatch calls to the corresponding subprograms on the C++ side. We can
3358 also extend the tagged type Dog with further fields and primitives, and
3359 override some of its C++ primitives on the Ada side. For example, here we have
3360 a type derivation defined on the Ada side that inherits all the dispatching
3361 primitives of the ancestor from the C++ side.
3364 @b{with} Animals; @b{use} Animals;
3365 @b{package} Vaccinated_Animals @b{is}
3366 @b{type} Vaccinated_Dog @b{is new} Dog @b{with null record};
3367 @b{function} Vaccination_Expired (A : Vaccinated_Dog) @b{return} Boolean;
3368 @b{end} Vaccinated_Animals;
3371 It is important to note that, because of the ABI compatibility, the programmer
3372 does not need to add any further information to indicate either the object
3373 layout or the dispatch table entry associated with each dispatching operation.
3375 Now let us define all the types and constructors on the Ada side and export
3376 them to C++, using the same hierarchy of our previous example:
3378 @smallexample @c ada
3379 with Interfaces.C.Strings;
3380 use Interfaces.C.Strings;
3382 type Carnivore is interface;
3383 pragma Convention (C_Plus_Plus, Carnivore);
3384 function Number_Of_Teeth (X : Carnivore)
3385 return Natural is abstract;
3387 type Domestic is interface;
3388 pragma Convention (C_Plus_Plus, Set_Owner);
3390 (X : in out Domestic;
3391 Name : Chars_Ptr) is abstract;
3393 type Animal is tagged record
3396 pragma Convention (C_Plus_Plus, Animal);
3398 procedure Set_Age (X : in out Animal; Age : Integer);
3399 pragma Export (C_Plus_Plus, Set_Age);
3401 function Age (X : Animal) return Integer;
3402 pragma Export (C_Plus_Plus, Age);
3404 type Dog is new Animal and Carnivore and Domestic with record
3405 Tooth_Count : Natural;
3406 Owner : String (1 .. 30);
3408 pragma Convention (C_Plus_Plus, Dog);
3410 function Number_Of_Teeth (A : Dog) return Integer;
3411 pragma Export (C_Plus_Plus, Number_Of_Teeth);
3413 procedure Set_Owner (A : in out Dog; Name : Chars_Ptr);
3414 pragma Export (C_Plus_Plus, Set_Owner);
3416 function New_Dog return Dog'Class;
3417 pragma Export (C_Plus_Plus, New_Dog);
3421 Compared with our previous example the only difference is the use of
3422 @code{pragma Export} to indicate to the GNAT compiler that the primitives will
3423 be available to C++. Thanks to the ABI compatibility, on the C++ side there is
3424 nothing else to be done; as explained above, the only requirement is that all
3425 the primitives and components are declared in exactly the same order.
3427 For completeness, let us see a brief C++ main program that uses the
3428 declarations available in @code{animals.h} (presented in our first example) to
3429 import and use the declarations from the Ada side, properly initializing and
3430 finalizing the Ada run-time system along the way:
3433 @b{#include} "animals.h"
3434 @b{#include} <iostream>
3435 @b{using namespace} std;
3437 void Check_Carnivore (Carnivore *obj) @{@dots{}@}
3438 void Check_Domestic (Domestic *obj) @{@dots{}@}
3439 void Check_Animal (Animal *obj) @{@dots{}@}
3440 void Check_Dog (Dog *obj) @{@dots{}@}
3443 void adainit (void);
3444 void adafinal (void);
3450 Dog *obj = new_dog(); // Ada constructor
3451 Check_Carnivore (obj); // Check secondary DT
3452 Check_Domestic (obj); // Check secondary DT
3453 Check_Animal (obj); // Check primary DT
3454 Check_Dog (obj); // Check primary DT
3459 adainit (); test(); adafinal ();
3464 @node Comparison between GNAT and C/C++ Compilation Models
3465 @section Comparison between GNAT and C/C++ Compilation Models
3468 The GNAT model of compilation is close to the C and C++ models. You can
3469 think of Ada specs as corresponding to header files in C. As in C, you
3470 don't need to compile specs; they are compiled when they are used. The
3471 Ada @code{with} is similar in effect to the @code{#include} of a C
3474 One notable difference is that, in Ada, you may compile specs separately
3475 to check them for semantic and syntactic accuracy. This is not always
3476 possible with C headers because they are fragments of programs that have
3477 less specific syntactic or semantic rules.
3479 The other major difference is the requirement for running the binder,
3480 which performs two important functions. First, it checks for
3481 consistency. In C or C++, the only defense against assembling
3482 inconsistent programs lies outside the compiler, in a makefile, for
3483 example. The binder satisfies the Ada requirement that it be impossible
3484 to construct an inconsistent program when the compiler is used in normal
3487 @cindex Elaboration order control
3488 The other important function of the binder is to deal with elaboration
3489 issues. There are also elaboration issues in C++ that are handled
3490 automatically. This automatic handling has the advantage of being
3491 simpler to use, but the C++ programmer has no control over elaboration.
3492 Where @code{gnatbind} might complain there was no valid order of
3493 elaboration, a C++ compiler would simply construct a program that
3494 malfunctioned at run time.
3497 @node Comparison between GNAT and Conventional Ada Library Models
3498 @section Comparison between GNAT and Conventional Ada Library Models
3501 This section is intended for Ada programmers who have
3502 used an Ada compiler implementing the traditional Ada library
3503 model, as described in the Ada Reference Manual.
3505 @cindex GNAT library
3506 In GNAT, there is no ``library'' in the normal sense. Instead, the set of
3507 source files themselves acts as the library. Compiling Ada programs does
3508 not generate any centralized information, but rather an object file and
3509 a ALI file, which are of interest only to the binder and linker.
3510 In a traditional system, the compiler reads information not only from
3511 the source file being compiled, but also from the centralized library.
3512 This means that the effect of a compilation depends on what has been
3513 previously compiled. In particular:
3517 When a unit is @code{with}'ed, the unit seen by the compiler corresponds
3518 to the version of the unit most recently compiled into the library.
3521 Inlining is effective only if the necessary body has already been
3522 compiled into the library.
3525 Compiling a unit may obsolete other units in the library.
3529 In GNAT, compiling one unit never affects the compilation of any other
3530 units because the compiler reads only source files. Only changes to source
3531 files can affect the results of a compilation. In particular:
3535 When a unit is @code{with}'ed, the unit seen by the compiler corresponds
3536 to the source version of the unit that is currently accessible to the
3541 Inlining requires the appropriate source files for the package or
3542 subprogram bodies to be available to the compiler. Inlining is always
3543 effective, independent of the order in which units are complied.
3546 Compiling a unit never affects any other compilations. The editing of
3547 sources may cause previous compilations to be out of date if they
3548 depended on the source file being modified.
3552 The most important result of these differences is that order of compilation
3553 is never significant in GNAT. There is no situation in which one is
3554 required to do one compilation before another. What shows up as order of
3555 compilation requirements in the traditional Ada library becomes, in
3556 GNAT, simple source dependencies; in other words, there is only a set
3557 of rules saying what source files must be present when a file is
3561 @node Placement of temporary files
3562 @section Placement of temporary files
3563 @cindex Temporary files (user control over placement)
3566 GNAT creates temporary files in the directory designated by the environment
3567 variable @env{TMPDIR}.
3568 (See the HP @emph{C RTL Reference Manual} on the function @code{getenv()}
3569 for detailed information on how environment variables are resolved.
3570 For most users the easiest way to make use of this feature is to simply
3571 define @env{TMPDIR} as a job level logical name).
3572 For example, if you wish to use a Ramdisk (assuming DECRAM is installed)
3573 for compiler temporary files, then you can include something like the
3574 following command in your @file{LOGIN.COM} file:
3577 $ define/job TMPDIR "/disk$scratchram/000000/temp/"
3581 If @env{TMPDIR} is not defined, then GNAT uses the directory designated by
3582 @env{TMP}; if @env{TMP} is not defined, then GNAT uses the directory
3583 designated by @env{TEMP}.
3584 If none of these environment variables are defined then GNAT uses the
3585 directory designated by the logical name @code{SYS$SCRATCH:}
3586 (by default the user's home directory). If all else fails
3587 GNAT uses the current directory for temporary files.
3590 @c *************************
3591 @node Compiling Using gcc
3592 @chapter Compiling Using @command{gcc}
3595 This chapter discusses how to compile Ada programs using the @command{gcc}
3596 command. It also describes the set of switches
3597 that can be used to control the behavior of the compiler.
3599 * Compiling Programs::
3600 * Switches for gcc::
3601 * Search Paths and the Run-Time Library (RTL)::
3602 * Order of Compilation Issues::
3606 @node Compiling Programs
3607 @section Compiling Programs
3610 The first step in creating an executable program is to compile the units
3611 of the program using the @command{gcc} command. You must compile the
3616 the body file (@file{.adb}) for a library level subprogram or generic
3620 the spec file (@file{.ads}) for a library level package or generic
3621 package that has no body
3624 the body file (@file{.adb}) for a library level package
3625 or generic package that has a body
3630 You need @emph{not} compile the following files
3635 the spec of a library unit which has a body
3642 because they are compiled as part of compiling related units. GNAT
3644 when the corresponding body is compiled, and subunits when the parent is
3647 @cindex cannot generate code
3648 If you attempt to compile any of these files, you will get one of the
3649 following error messages (where @var{fff} is the name of the file you compiled):
3652 cannot generate code for file @var{fff} (package spec)
3653 to check package spec, use -gnatc
3655 cannot generate code for file @var{fff} (missing subunits)
3656 to check parent unit, use -gnatc
3658 cannot generate code for file @var{fff} (subprogram spec)
3659 to check subprogram spec, use -gnatc
3661 cannot generate code for file @var{fff} (subunit)
3662 to check subunit, use -gnatc
3666 As indicated by the above error messages, if you want to submit
3667 one of these files to the compiler to check for correct semantics
3668 without generating code, then use the @option{-gnatc} switch.
3670 The basic command for compiling a file containing an Ada unit is
3673 $ gcc -c @ovar{switches} @file{file name}
3677 where @var{file name} is the name of the Ada file (usually
3679 @file{.ads} for a spec or @file{.adb} for a body).
3682 @option{-c} switch to tell @command{gcc} to compile, but not link, the file.
3684 The result of a successful compilation is an object file, which has the
3685 same name as the source file but an extension of @file{.o} and an Ada
3686 Library Information (ALI) file, which also has the same name as the
3687 source file, but with @file{.ali} as the extension. GNAT creates these
3688 two output files in the current directory, but you may specify a source
3689 file in any directory using an absolute or relative path specification
3690 containing the directory information.
3693 @command{gcc} is actually a driver program that looks at the extensions of
3694 the file arguments and loads the appropriate compiler. For example, the
3695 GNU C compiler is @file{cc1}, and the Ada compiler is @file{gnat1}.
3696 These programs are in directories known to the driver program (in some
3697 configurations via environment variables you set), but need not be in
3698 your path. The @command{gcc} driver also calls the assembler and any other
3699 utilities needed to complete the generation of the required object
3702 It is possible to supply several file names on the same @command{gcc}
3703 command. This causes @command{gcc} to call the appropriate compiler for
3704 each file. For example, the following command lists three separate
3705 files to be compiled:
3708 $ gcc -c x.adb y.adb z.c
3712 calls @code{gnat1} (the Ada compiler) twice to compile @file{x.adb} and
3713 @file{y.adb}, and @code{cc1} (the C compiler) once to compile @file{z.c}.
3714 The compiler generates three object files @file{x.o}, @file{y.o} and
3715 @file{z.o} and the two ALI files @file{x.ali} and @file{y.ali} from the
3716 Ada compilations. Any switches apply to all the files ^listed,^listed.^
3719 @option{-gnat@var{x}} switches, which apply only to Ada compilations.
3722 @node Switches for gcc
3723 @section Switches for @command{gcc}
3726 The @command{gcc} command accepts switches that control the
3727 compilation process. These switches are fully described in this section.
3728 First we briefly list all the switches, in alphabetical order, then we
3729 describe the switches in more detail in functionally grouped sections.
3731 More switches exist for GCC than those documented here, especially
3732 for specific targets. However, their use is not recommended as
3733 they may change code generation in ways that are incompatible with
3734 the Ada run-time library, or can cause inconsistencies between
3738 * Output and Error Message Control::
3739 * Warning Message Control::
3740 * Debugging and Assertion Control::
3741 * Validity Checking::
3744 * Using gcc for Syntax Checking::
3745 * Using gcc for Semantic Checking::
3746 * Compiling Different Versions of Ada::
3747 * Character Set Control::
3748 * File Naming Control::
3749 * Subprogram Inlining Control::
3750 * Auxiliary Output Control::
3751 * Debugging Control::
3752 * Exception Handling Control::
3753 * Units to Sources Mapping Files::
3754 * Integrated Preprocessing::
3755 * Code Generation Control::
3764 @cindex @option{-b} (@command{gcc})
3765 @item -b @var{target}
3766 Compile your program to run on @var{target}, which is the name of a
3767 system configuration. You must have a GNAT cross-compiler built if
3768 @var{target} is not the same as your host system.
3771 @cindex @option{-B} (@command{gcc})
3772 Load compiler executables (for example, @code{gnat1}, the Ada compiler)
3773 from @var{dir} instead of the default location. Only use this switch
3774 when multiple versions of the GNAT compiler are available.
3775 @xref{Directory Options,, Options for Directory Search, gcc, Using the
3776 GNU Compiler Collection (GCC)}, for further details. You would normally
3777 use the @option{-b} or @option{-V} switch instead.
3780 @cindex @option{-c} (@command{gcc})
3781 Compile. Always use this switch when compiling Ada programs.
3783 Note: for some other languages when using @command{gcc}, notably in
3784 the case of C and C++, it is possible to use
3785 use @command{gcc} without a @option{-c} switch to
3786 compile and link in one step. In the case of GNAT, you
3787 cannot use this approach, because the binder must be run
3788 and @command{gcc} cannot be used to run the GNAT binder.
3792 @cindex @option{-fno-inline} (@command{gcc})
3793 Suppresses all back-end inlining, even if other optimization or inlining
3795 This includes suppression of inlining that results
3796 from the use of the pragma @code{Inline_Always}.
3797 Any occurrences of pragma @code{Inline} or @code{Inline_Always}
3798 are ignored, and @option{-gnatn} and @option{-gnatN} have no
3799 effect if this switch is present.
3801 @item -fno-inline-functions
3802 @cindex @option{-fno-inline-functions} (@command{gcc})
3803 Suppresses automatic inlining of small subprograms, which is enabled
3804 if @option{-O3} is used.
3806 @item -fno-inline-functions-called-once
3807 @cindex @option{-fno-inline-functions-called-once} (@command{gcc})
3808 Suppresses inlining of subprograms local to the unit and called once
3809 from within it, which is enabled if @option{-O1} is used.
3811 @item -fno-strict-aliasing
3812 @cindex @option{-fno-strict-aliasing} (@command{gcc})
3813 Causes the compiler to avoid assumptions regarding non-aliasing
3814 of objects of different types. See
3815 @ref{Optimization and Strict Aliasing} for details.
3818 @cindex @option{-fstack-check} (@command{gcc})
3819 Activates stack checking.
3820 See @ref{Stack Overflow Checking} for details.
3823 @cindex @option{-fstack-usage} (@command{gcc})
3824 Makes the compiler output stack usage information for the program, on a
3825 per-function basis. See @ref{Static Stack Usage Analysis} for details.
3827 @item -fcallgraph-info@r{[}=su@r{]}
3828 @cindex @option{-fcallgraph-info} (@command{gcc})
3829 Makes the compiler output callgraph information for the program, on a
3830 per-file basis. The information is generated in the VCG format. It can
3831 be decorated with stack-usage per-node information.
3834 @cindex @option{^-g^/DEBUG^} (@command{gcc})
3835 Generate debugging information. This information is stored in the object
3836 file and copied from there to the final executable file by the linker,
3837 where it can be read by the debugger. You must use the
3838 @option{^-g^/DEBUG^} switch if you plan on using the debugger.
3841 @cindex @option{-gnat83} (@command{gcc})
3842 Enforce Ada 83 restrictions.
3845 @cindex @option{-gnat95} (@command{gcc})
3846 Enforce Ada 95 restrictions.
3849 @cindex @option{-gnat05} (@command{gcc})
3850 Allow full Ada 2005 features.
3853 @cindex @option{-gnata} (@command{gcc})
3854 Assertions enabled. @code{Pragma Assert} and @code{pragma Debug} to be
3855 activated. Note that these pragmas can also be controlled using the
3856 configuration pragmas @code{Assertion_Policy} and @code{Debug_Policy}.
3857 It also activates pragmas @code{Check}, @code{Precondition}, and
3858 @code{Postcondition}. Note that these pragmas can also be controlled
3859 using the configuration pragma @code{Check_Policy}.
3862 @cindex @option{-gnatA} (@command{gcc})
3863 Avoid processing @file{gnat.adc}. If a @file{gnat.adc} file is present,
3867 @cindex @option{-gnatb} (@command{gcc})
3868 Generate brief messages to @file{stderr} even if verbose mode set.
3871 @cindex @option{-gnatc} (@command{gcc})
3872 Check syntax and semantics only (no code generation attempted).
3875 @cindex @option{-gnatd} (@command{gcc})
3876 Specify debug options for the compiler. The string of characters after
3877 the @option{-gnatd} specify the specific debug options. The possible
3878 characters are 0-9, a-z, A-Z, optionally preceded by a dot. See
3879 compiler source file @file{debug.adb} for details of the implemented
3880 debug options. Certain debug options are relevant to applications
3881 programmers, and these are documented at appropriate points in this
3885 @cindex @option{-gnatD} (@command{gcc})
3886 Create expanded source files for source level debugging. This switch
3887 also suppress generation of cross-reference information
3888 (see @option{-gnatx}).
3890 @item -gnatec=@var{path}
3891 @cindex @option{-gnatec} (@command{gcc})
3892 Specify a configuration pragma file
3894 (the equal sign is optional)
3896 (@pxref{The Configuration Pragmas Files}).
3898 @item ^-gnateD^/DATA_PREPROCESSING=^symbol@r{[}=@var{value}@r{]}
3899 @cindex @option{-gnateD} (@command{gcc})
3900 Defines a symbol, associated with @var{value}, for preprocessing.
3901 (@pxref{Integrated Preprocessing}).
3904 @cindex @option{-gnatef} (@command{gcc})
3905 Display full source path name in brief error messages.
3907 @item -gnatem=@var{path}
3908 @cindex @option{-gnatem} (@command{gcc})
3909 Specify a mapping file
3911 (the equal sign is optional)
3913 (@pxref{Units to Sources Mapping Files}).
3915 @item -gnatep=@var{file}
3916 @cindex @option{-gnatep} (@command{gcc})
3917 Specify a preprocessing data file
3919 (the equal sign is optional)
3921 (@pxref{Integrated Preprocessing}).
3924 @cindex @option{-gnatE} (@command{gcc})
3925 Full dynamic elaboration checks.
3928 @cindex @option{-gnatf} (@command{gcc})
3929 Full errors. Multiple errors per line, all undefined references, do not
3930 attempt to suppress cascaded errors.
3933 @cindex @option{-gnatF} (@command{gcc})
3934 Externals names are folded to all uppercase.
3936 @item ^-gnatg^/GNAT_INTERNAL^
3937 @cindex @option{^-gnatg^/GNAT_INTERNAL^} (@command{gcc})
3938 Internal GNAT implementation mode. This should not be used for
3939 applications programs, it is intended only for use by the compiler
3940 and its run-time library. For documentation, see the GNAT sources.
3941 Note that @option{^-gnatg^/GNAT_INTERNAL^} implies
3942 @option{^-gnatwae^/WARNINGS=ALL,ERRORS^} and
3943 @option{^-gnatyg^/STYLE_CHECKS=GNAT^}
3944 so that all standard warnings and all standard style options are turned on.
3945 All warnings and style error messages are treated as errors.
3948 @cindex @option{-gnatG} (@command{gcc})
3949 List generated expanded code in source form.
3951 @item ^-gnath^/HELP^
3952 @cindex @option{^-gnath^/HELP^} (@command{gcc})
3953 Output usage information. The output is written to @file{stdout}.
3955 @item ^-gnati^/IDENTIFIER_CHARACTER_SET=^@var{c}
3956 @cindex @option{^-gnati^/IDENTIFIER_CHARACTER_SET^} (@command{gcc})
3957 Identifier character set
3959 (@var{c}=1/2/3/4/8/9/p/f/n/w).
3961 For details of the possible selections for @var{c},
3962 see @ref{Character Set Control}.
3964 @item ^-gnatI^/IGNORE_REP_CLAUSES^
3965 @cindex @option{^-gnatI^IGNORE_REP_CLAUSES^} (@command{gcc})
3966 Ignore representation clauses. When this switch is used, all
3967 representation clauses are treated as comments. This is useful
3968 when initially porting code where you want to ignore rep clause
3969 problems, and also for compiling foreign code (particularly
3973 @cindex @option{-gnatjnn} (@command{gcc})
3974 Reformat error messages to fit on nn character lines
3976 @item -gnatk=@var{n}
3977 @cindex @option{-gnatk} (@command{gcc})
3978 Limit file names to @var{n} (1-999) characters ^(@code{k} = krunch)^^.
3981 @cindex @option{-gnatl} (@command{gcc})
3982 Output full source listing with embedded error messages.
3985 @cindex @option{-gnatL} (@command{gcc})
3986 Used in conjunction with -gnatG or -gnatD to intersperse original
3987 source lines (as comment lines with line numbers) in the expanded
3990 @item -gnatm=@var{n}
3991 @cindex @option{-gnatm} (@command{gcc})
3992 Limit number of detected error or warning messages to @var{n}
3993 where @var{n} is in the range 1..999_999. The default setting if
3994 no switch is given is 9999. Compilation is terminated if this
3995 limit is exceeded. The equal sign here is optional.
3998 @cindex @option{-gnatn} (@command{gcc})
3999 Activate inlining for subprograms for which
4000 pragma @code{inline} is specified. This inlining is performed
4001 by the GCC back-end.
4004 @cindex @option{-gnatN} (@command{gcc})
4005 Activate front end inlining for subprograms for which
4006 pragma @code{Inline} is specified. This inlining is performed
4007 by the front end and will be visible in the
4008 @option{-gnatG} output.
4009 In some cases, this has proved more effective than the back end
4010 inlining resulting from the use of
4013 @option{-gnatN} automatically implies
4014 @option{-gnatn} so it is not necessary
4015 to specify both options. There are a few cases that the back-end inlining
4016 catches that cannot be dealt with in the front-end.
4019 @cindex @option{-gnato} (@command{gcc})
4020 Enable numeric overflow checking (which is not normally enabled by
4021 default). Not that division by zero is a separate check that is not
4022 controlled by this switch (division by zero checking is on by default).
4025 @cindex @option{-gnatp} (@command{gcc})
4026 Suppress all checks.
4029 @cindex @option{-gnatP} (@command{gcc})
4030 Enable polling. This is required on some systems (notably Windows NT) to
4031 obtain asynchronous abort and asynchronous transfer of control capability.
4032 @xref{Pragma Polling,,, gnat_rm, GNAT Reference Manual}, for full
4036 @cindex @option{-gnatq} (@command{gcc})
4037 Don't quit; try semantics, even if parse errors.
4040 @cindex @option{-gnatQ} (@command{gcc})
4041 Don't quit; generate @file{ALI} and tree files even if illegalities.
4044 @cindex @option{-gnatr} (@command{gcc})
4045 Treat pragma Restrictions as Restriction_Warnings.
4047 @item ^-gnatR@r{[}0@r{/}1@r{/}2@r{/}3@r{[}s@r{]]}^/REPRESENTATION_INFO^
4048 @cindex @option{-gnatR} (@command{gcc})
4049 Output representation information for declared types and objects.
4052 @cindex @option{-gnats} (@command{gcc})
4056 @cindex @option{-gnatS} (@command{gcc})
4057 Print package Standard.
4060 @cindex @option{-gnatt} (@command{gcc})
4061 Generate tree output file.
4063 @item ^-gnatT^/TABLE_MULTIPLIER=^@var{nnn}
4064 @cindex @option{^-gnatT^/TABLE_MULTIPLIER^} (@command{gcc})
4065 All compiler tables start at @var{nnn} times usual starting size.
4068 @cindex @option{-gnatu} (@command{gcc})
4069 List units for this compilation.
4072 @cindex @option{-gnatU} (@command{gcc})
4073 Tag all error messages with the unique string ``error:''
4076 @cindex @option{-gnatv} (@command{gcc})
4077 Verbose mode. Full error output with source lines to @file{stdout}.
4080 @cindex @option{-gnatV} (@command{gcc})
4081 Control level of validity checking. See separate section describing
4084 @item ^-gnatw@var{xxx}^/WARNINGS=(@var{option}@r{[},@dots{}@r{]})^
4085 @cindex @option{^-gnatw^/WARNINGS^} (@command{gcc})
4087 ^@var{xxx} is a string of option letters that^the list of options^ denotes
4088 the exact warnings that
4089 are enabled or disabled (@pxref{Warning Message Control}).
4091 @item ^-gnatW^/WIDE_CHARACTER_ENCODING=^@var{e}
4092 @cindex @option{^-gnatW^/WIDE_CHARACTER_ENCODING^} (@command{gcc})
4093 Wide character encoding method
4095 (@var{e}=n/h/u/s/e/8).
4098 (@var{e}=@code{BRACKETS, NONE, HEX, UPPER, SHIFT_JIS, EUC, UTF8})
4102 @cindex @option{-gnatx} (@command{gcc})
4103 Suppress generation of cross-reference information.
4105 @item ^-gnaty^/STYLE_CHECKS=(option,option@dots{})^
4106 @cindex @option{^-gnaty^/STYLE_CHECKS^} (@command{gcc})
4107 Enable built-in style checks (@pxref{Style Checking}).
4109 @item ^-gnatz^/DISTRIBUTION_STUBS=^@var{m}
4110 @cindex @option{^-gnatz^/DISTRIBUTION_STUBS^} (@command{gcc})
4111 Distribution stub generation and compilation
4113 (@var{m}=r/c for receiver/caller stubs).
4116 (@var{m}=@code{RECEIVER} or @code{CALLER} to specify the type of stubs
4117 to be generated and compiled).
4120 @item ^-I^/SEARCH=^@var{dir}
4121 @cindex @option{^-I^/SEARCH^} (@command{gcc})
4123 Direct GNAT to search the @var{dir} directory for source files needed by
4124 the current compilation
4125 (@pxref{Search Paths and the Run-Time Library (RTL)}).
4127 @item ^-I-^/NOCURRENT_DIRECTORY^
4128 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@command{gcc})
4130 Except for the source file named in the command line, do not look for source
4131 files in the directory containing the source file named in the command line
4132 (@pxref{Search Paths and the Run-Time Library (RTL)}).
4136 @cindex @option{-mbig-switch} (@command{gcc})
4137 @cindex @code{case} statement (effect of @option{-mbig-switch} option)
4138 This standard gcc switch causes the compiler to use larger offsets in its
4139 jump table representation for @code{case} statements.
4140 This may result in less efficient code, but is sometimes necessary
4141 (for example on HP-UX targets)
4142 @cindex HP-UX and @option{-mbig-switch} option
4143 in order to compile large and/or nested @code{case} statements.
4146 @cindex @option{-o} (@command{gcc})
4147 This switch is used in @command{gcc} to redirect the generated object file
4148 and its associated ALI file. Beware of this switch with GNAT, because it may
4149 cause the object file and ALI file to have different names which in turn
4150 may confuse the binder and the linker.
4154 @cindex @option{-nostdinc} (@command{gcc})
4155 Inhibit the search of the default location for the GNAT Run Time
4156 Library (RTL) source files.
4159 @cindex @option{-nostdlib} (@command{gcc})
4160 Inhibit the search of the default location for the GNAT Run Time
4161 Library (RTL) ALI files.
4165 @cindex @option{-O} (@command{gcc})
4166 @var{n} controls the optimization level.
4170 No optimization, the default setting if no @option{-O} appears
4173 Normal optimization, the default if you specify @option{-O} without
4174 an operand. A good compromise between code quality and compilation
4178 Extensive optimization, may improve execution time, possibly at the cost of
4179 substantially increased compilation time.
4182 Same as @option{-O2}, and also includes inline expansion for small subprograms
4186 Optimize space usage
4190 See also @ref{Optimization Levels}.
4195 @cindex @option{/NOOPTIMIZE} (@code{GNAT COMPILE})
4196 Equivalent to @option{/OPTIMIZE=NONE}.
4197 This is the default behavior in the absence of an @option{/OPTIMIZE}
4200 @item /OPTIMIZE@r{[}=(keyword@r{[},@dots{}@r{]})@r{]}
4201 @cindex @option{/OPTIMIZE} (@code{GNAT COMPILE})
4202 Selects the level of optimization for your program. The supported
4203 keywords are as follows:
4206 Perform most optimizations, including those that
4208 This is the default if the @option{/OPTIMIZE} qualifier is supplied
4209 without keyword options.
4212 Do not do any optimizations. Same as @code{/NOOPTIMIZE}.
4215 Perform some optimizations, but omit ones that are costly.
4218 Same as @code{SOME}.
4221 Full optimization as in @option{/OPTIMIZE=ALL}, and also attempts
4222 automatic inlining of small subprograms within a unit
4225 Try to unroll loops. This keyword may be specified together with
4226 any keyword above other than @code{NONE}. Loop unrolling
4227 usually, but not always, improves the performance of programs.
4230 Optimize space usage
4234 See also @ref{Optimization Levels}.
4238 @item -pass-exit-codes
4239 @cindex @option{-pass-exit-codes} (@command{gcc})
4240 Catch exit codes from the compiler and use the most meaningful as
4244 @item --RTS=@var{rts-path}
4245 @cindex @option{--RTS} (@command{gcc})
4246 Specifies the default location of the runtime library. Same meaning as the
4247 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
4250 @cindex @option{^-S^/ASM^} (@command{gcc})
4251 ^Used in place of @option{-c} to^Used to^
4252 cause the assembler source file to be
4253 generated, using @file{^.s^.S^} as the extension,
4254 instead of the object file.
4255 This may be useful if you need to examine the generated assembly code.
4257 @item ^-fverbose-asm^/VERBOSE_ASM^
4258 @cindex @option{^-fverbose-asm^/VERBOSE_ASM^} (@command{gcc})
4259 ^Used in conjunction with @option{-S}^Used in place of @option{/ASM}^
4260 to cause the generated assembly code file to be annotated with variable
4261 names, making it significantly easier to follow.
4264 @cindex @option{^-v^/VERBOSE^} (@command{gcc})
4265 Show commands generated by the @command{gcc} driver. Normally used only for
4266 debugging purposes or if you need to be sure what version of the
4267 compiler you are executing.
4271 @cindex @option{-V} (@command{gcc})
4272 Execute @var{ver} version of the compiler. This is the @command{gcc}
4273 version, not the GNAT version.
4276 @item ^-w^/NO_BACK_END_WARNINGS^
4277 @cindex @option{-w} (@command{gcc})
4278 Turn off warnings generated by the back end of the compiler. Use of
4279 this switch also causes the default for front end warnings to be set
4280 to suppress (as though @option{-gnatws} had appeared at the start of
4286 @c Combining qualifiers does not work on VMS
4287 You may combine a sequence of GNAT switches into a single switch. For
4288 example, the combined switch
4290 @cindex Combining GNAT switches
4296 is equivalent to specifying the following sequence of switches:
4299 -gnato -gnatf -gnati3
4304 The following restrictions apply to the combination of switches
4309 The switch @option{-gnatc} if combined with other switches must come
4310 first in the string.
4313 The switch @option{-gnats} if combined with other switches must come
4314 first in the string.
4318 @option{^-gnatz^/DISTRIBUTION_STUBS^}, @option{-gnatzc}, and @option{-gnatzr}
4319 may not be combined with any other switches.
4323 Once a ``y'' appears in the string (that is a use of the @option{-gnaty}
4324 switch), then all further characters in the switch are interpreted
4325 as style modifiers (see description of @option{-gnaty}).
4328 Once a ``d'' appears in the string (that is a use of the @option{-gnatd}
4329 switch), then all further characters in the switch are interpreted
4330 as debug flags (see description of @option{-gnatd}).
4333 Once a ``w'' appears in the string (that is a use of the @option{-gnatw}
4334 switch), then all further characters in the switch are interpreted
4335 as warning mode modifiers (see description of @option{-gnatw}).
4338 Once a ``V'' appears in the string (that is a use of the @option{-gnatV}
4339 switch), then all further characters in the switch are interpreted
4340 as validity checking options (see description of @option{-gnatV}).
4344 @node Output and Error Message Control
4345 @subsection Output and Error Message Control
4349 The standard default format for error messages is called ``brief format''.
4350 Brief format messages are written to @file{stderr} (the standard error
4351 file) and have the following form:
4354 e.adb:3:04: Incorrect spelling of keyword "function"
4355 e.adb:4:20: ";" should be "is"
4359 The first integer after the file name is the line number in the file,
4360 and the second integer is the column number within the line.
4362 @code{GPS} can parse the error messages
4363 and point to the referenced character.
4365 The following switches provide control over the error message
4371 @cindex @option{-gnatv} (@command{gcc})
4374 The v stands for verbose.
4376 The effect of this setting is to write long-format error
4377 messages to @file{stdout} (the standard output file.
4378 The same program compiled with the
4379 @option{-gnatv} switch would generate:
4383 3. funcion X (Q : Integer)
4385 >>> Incorrect spelling of keyword "function"
4388 >>> ";" should be "is"
4393 The vertical bar indicates the location of the error, and the @samp{>>>}
4394 prefix can be used to search for error messages. When this switch is
4395 used the only source lines output are those with errors.
4398 @cindex @option{-gnatl} (@command{gcc})
4400 The @code{l} stands for list.
4402 This switch causes a full listing of
4403 the file to be generated. In the case where a body is
4404 compiled, the corresponding spec is also listed, along
4405 with any subunits. Typical output from compiling a package
4406 body @file{p.adb} might look like:
4408 @smallexample @c ada
4412 1. package body p is
4414 3. procedure a is separate;
4425 2. pragma Elaborate_Body
4449 When you specify the @option{-gnatv} or @option{-gnatl} switches and
4450 standard output is redirected, a brief summary is written to
4451 @file{stderr} (standard error) giving the number of error messages and
4452 warning messages generated.
4454 @item -^gnatl^OUTPUT_FILE^=file
4455 @cindex @option{^-gnatl^OUTPUT_FILE^=fname} (@command{gcc})
4456 This has the same effect as @option{-gnatl} except that the output is
4457 written to a file instead of to standard output. If the given name
4458 @file{fname} does not start with a period, then it is the full name
4459 of the file to be written. If @file{fname} is an extension, it is
4460 appended to the name of the file being compiled. For example, if
4461 file @file{xyz.adb} is compiled with @option{^-gnatl^OUTPUT_FILE^=.lst},
4462 then the output is written to file ^xyz.adb.lst^xyz.adb_lst^.
4465 @cindex @option{-gnatU} (@command{gcc})
4466 This switch forces all error messages to be preceded by the unique
4467 string ``error:''. This means that error messages take a few more
4468 characters in space, but allows easy searching for and identification
4472 @cindex @option{-gnatb} (@command{gcc})
4474 The @code{b} stands for brief.
4476 This switch causes GNAT to generate the
4477 brief format error messages to @file{stderr} (the standard error
4478 file) as well as the verbose
4479 format message or full listing (which as usual is written to
4480 @file{stdout} (the standard output file).
4482 @item -gnatm=@var{n}
4483 @cindex @option{-gnatm} (@command{gcc})
4485 The @code{m} stands for maximum.
4487 @var{n} is a decimal integer in the
4488 range of 1 to 999 and limits the number of error messages to be
4489 generated. For example, using @option{-gnatm2} might yield
4492 e.adb:3:04: Incorrect spelling of keyword "function"
4493 e.adb:5:35: missing ".."
4494 fatal error: maximum errors reached
4495 compilation abandoned
4499 Note that the equal sign is optional, so the switches
4500 @option{-gnatm2} and @option{-gnatm=2} are equivalent.
4503 @cindex @option{-gnatf} (@command{gcc})
4504 @cindex Error messages, suppressing
4506 The @code{f} stands for full.
4508 Normally, the compiler suppresses error messages that are likely to be
4509 redundant. This switch causes all error
4510 messages to be generated. In particular, in the case of
4511 references to undefined variables. If a given variable is referenced
4512 several times, the normal format of messages is
4514 e.adb:7:07: "V" is undefined (more references follow)
4518 where the parenthetical comment warns that there are additional
4519 references to the variable @code{V}. Compiling the same program with the
4520 @option{-gnatf} switch yields
4523 e.adb:7:07: "V" is undefined
4524 e.adb:8:07: "V" is undefined
4525 e.adb:8:12: "V" is undefined
4526 e.adb:8:16: "V" is undefined
4527 e.adb:9:07: "V" is undefined
4528 e.adb:9:12: "V" is undefined
4532 The @option{-gnatf} switch also generates additional information for
4533 some error messages. Some examples are:
4537 Full details on entities not available in high integrity mode
4539 Details on possibly non-portable unchecked conversion
4541 List possible interpretations for ambiguous calls
4543 Additional details on incorrect parameters
4547 @cindex @option{-gnatjnn} (@command{gcc})
4548 In normal operation mode (or if @option{-gnatj0} is used, then error messages
4549 with continuation lines are treated as though the continuation lines were
4550 separate messages (and so a warning with two continuation lines counts as
4551 three warnings, and is listed as three separate messages).
4553 If the @option{-gnatjnn} switch is used with a positive value for nn, then
4554 messages are output in a different manner. A message and all its continuation
4555 lines are treated as a unit, and count as only one warning or message in the
4556 statistics totals. Furthermore, the message is reformatted so that no line
4557 is longer than nn characters.
4560 @cindex @option{-gnatq} (@command{gcc})
4562 The @code{q} stands for quit (really ``don't quit'').
4564 In normal operation mode, the compiler first parses the program and
4565 determines if there are any syntax errors. If there are, appropriate
4566 error messages are generated and compilation is immediately terminated.
4568 GNAT to continue with semantic analysis even if syntax errors have been
4569 found. This may enable the detection of more errors in a single run. On
4570 the other hand, the semantic analyzer is more likely to encounter some
4571 internal fatal error when given a syntactically invalid tree.
4574 @cindex @option{-gnatQ} (@command{gcc})
4575 In normal operation mode, the @file{ALI} file is not generated if any
4576 illegalities are detected in the program. The use of @option{-gnatQ} forces
4577 generation of the @file{ALI} file. This file is marked as being in
4578 error, so it cannot be used for binding purposes, but it does contain
4579 reasonably complete cross-reference information, and thus may be useful
4580 for use by tools (e.g., semantic browsing tools or integrated development
4581 environments) that are driven from the @file{ALI} file. This switch
4582 implies @option{-gnatq}, since the semantic phase must be run to get a
4583 meaningful ALI file.
4585 In addition, if @option{-gnatt} is also specified, then the tree file is
4586 generated even if there are illegalities. It may be useful in this case
4587 to also specify @option{-gnatq} to ensure that full semantic processing
4588 occurs. The resulting tree file can be processed by ASIS, for the purpose
4589 of providing partial information about illegal units, but if the error
4590 causes the tree to be badly malformed, then ASIS may crash during the
4593 When @option{-gnatQ} is used and the generated @file{ALI} file is marked as
4594 being in error, @command{gnatmake} will attempt to recompile the source when it
4595 finds such an @file{ALI} file, including with switch @option{-gnatc}.
4597 Note that @option{-gnatQ} has no effect if @option{-gnats} is specified,
4598 since ALI files are never generated if @option{-gnats} is set.
4602 @node Warning Message Control
4603 @subsection Warning Message Control
4604 @cindex Warning messages
4606 In addition to error messages, which correspond to illegalities as defined
4607 in the Ada Reference Manual, the compiler detects two kinds of warning
4610 First, the compiler considers some constructs suspicious and generates a
4611 warning message to alert you to a possible error. Second, if the
4612 compiler detects a situation that is sure to raise an exception at
4613 run time, it generates a warning message. The following shows an example
4614 of warning messages:
4616 e.adb:4:24: warning: creation of object may raise Storage_Error
4617 e.adb:10:17: warning: static value out of range
4618 e.adb:10:17: warning: "Constraint_Error" will be raised at run time
4622 GNAT considers a large number of situations as appropriate
4623 for the generation of warning messages. As always, warnings are not
4624 definite indications of errors. For example, if you do an out-of-range
4625 assignment with the deliberate intention of raising a
4626 @code{Constraint_Error} exception, then the warning that may be
4627 issued does not indicate an error. Some of the situations for which GNAT
4628 issues warnings (at least some of the time) are given in the following
4629 list. This list is not complete, and new warnings are often added to
4630 subsequent versions of GNAT. The list is intended to give a general idea
4631 of the kinds of warnings that are generated.
4635 Possible infinitely recursive calls
4638 Out-of-range values being assigned
4641 Possible order of elaboration problems
4644 Assertions (pragma Assert) that are sure to fail
4650 Address clauses with possibly unaligned values, or where an attempt is
4651 made to overlay a smaller variable with a larger one.
4654 Fixed-point type declarations with a null range
4657 Direct_IO or Sequential_IO instantiated with a type that has access values
4660 Variables that are never assigned a value
4663 Variables that are referenced before being initialized
4666 Task entries with no corresponding @code{accept} statement
4669 Duplicate accepts for the same task entry in a @code{select}
4672 Objects that take too much storage
4675 Unchecked conversion between types of differing sizes
4678 Missing @code{return} statement along some execution path in a function
4681 Incorrect (unrecognized) pragmas
4684 Incorrect external names
4687 Allocation from empty storage pool
4690 Potentially blocking operation in protected type
4693 Suspicious parenthesization of expressions
4696 Mismatching bounds in an aggregate
4699 Attempt to return local value by reference
4702 Premature instantiation of a generic body
4705 Attempt to pack aliased components
4708 Out of bounds array subscripts
4711 Wrong length on string assignment
4714 Violations of style rules if style checking is enabled
4717 Unused @code{with} clauses
4720 @code{Bit_Order} usage that does not have any effect
4723 @code{Standard.Duration} used to resolve universal fixed expression
4726 Dereference of possibly null value
4729 Declaration that is likely to cause storage error
4732 Internal GNAT unit @code{with}'ed by application unit
4735 Values known to be out of range at compile time
4738 Unreferenced labels and variables
4741 Address overlays that could clobber memory
4744 Unexpected initialization when address clause present
4747 Bad alignment for address clause
4750 Useless type conversions
4753 Redundant assignment statements and other redundant constructs
4756 Useless exception handlers
4759 Accidental hiding of name by child unit
4762 Access before elaboration detected at compile time
4765 A range in a @code{for} loop that is known to be null or might be null
4770 The following section lists compiler switches that are available
4771 to control the handling of warning messages. It is also possible
4772 to exercise much finer control over what warnings are issued and
4773 suppressed using the GNAT pragma Warnings, @xref{Pragma Warnings,,,
4774 gnat_rm, GNAT Reference manual}.
4779 @emph{Activate all optional errors.}
4780 @cindex @option{-gnatwa} (@command{gcc})
4781 This switch activates most optional warning messages, see remaining list
4782 in this section for details on optional warning messages that can be
4783 individually controlled. The warnings that are not turned on by this
4785 @option{-gnatwd} (implicit dereferencing),
4786 @option{-gnatwh} (hiding),
4787 @option{-gnatwl} (elaboration warnings),
4788 @option{-gnatw.o} (warn on values set by out parameters ignored)
4789 and @option{-gnatwt} (tracking of deleted conditional code).
4790 All other optional warnings are turned on.
4793 @emph{Suppress all optional errors.}
4794 @cindex @option{-gnatwA} (@command{gcc})
4795 This switch suppresses all optional warning messages, see remaining list
4796 in this section for details on optional warning messages that can be
4797 individually controlled.
4800 @emph{Activate warnings on failing assertions.}
4801 @cindex @option{-gnatw.a} (@command{gcc})
4802 @cindex Assert failures
4803 This switch activates warnings for assertions where the compiler can tell at
4804 compile time that the assertion will fail. Note that this warning is given
4805 even if assertions are disabled. The default is that such warnings are
4809 @emph{Suppress warnings on failing assertions.}
4810 @cindex @option{-gnatw.A} (@command{gcc})
4811 @cindex Assert failures
4812 This switch suppresses warnings for assertions where the compiler can tell at
4813 compile time that the assertion will fail.
4816 @emph{Activate warnings on bad fixed values.}
4817 @cindex @option{-gnatwb} (@command{gcc})
4818 @cindex Bad fixed values
4819 @cindex Fixed-point Small value
4821 This switch activates warnings for static fixed-point expressions whose
4822 value is not an exact multiple of Small. Such values are implementation
4823 dependent, since an implementation is free to choose either of the multiples
4824 that surround the value. GNAT always chooses the closer one, but this is not
4825 required behavior, and it is better to specify a value that is an exact
4826 multiple, ensuring predictable execution. The default is that such warnings
4830 @emph{Suppress warnings on bad fixed values.}
4831 @cindex @option{-gnatwB} (@command{gcc})
4832 This switch suppresses warnings for static fixed-point expressions whose
4833 value is not an exact multiple of Small.
4836 @emph{Activate warnings on conditionals.}
4837 @cindex @option{-gnatwc} (@command{gcc})
4838 @cindex Conditionals, constant
4839 This switch activates warnings for conditional expressions used in
4840 tests that are known to be True or False at compile time. The default
4841 is that such warnings are not generated.
4842 Note that this warning does
4843 not get issued for the use of boolean variables or constants whose
4844 values are known at compile time, since this is a standard technique
4845 for conditional compilation in Ada, and this would generate too many
4846 false positive warnings.
4848 This warning option also activates a special test for comparisons using
4849 the operators ``>='' and`` <=''.
4850 If the compiler can tell that only the equality condition is possible,
4851 then it will warn that the ``>'' or ``<'' part of the test
4852 is useless and that the operator could be replaced by ``=''.
4853 An example would be comparing a @code{Natural} variable <= 0.
4855 This warning option also generates warnings if
4856 one or both tests is optimized away in a membership test for integer
4857 values if the result can be determined at compile time. Range tests on
4858 enumeration types are not included, since it is common for such tests
4859 to include an end point.
4861 This warning can also be turned on using @option{-gnatwa}.
4864 @emph{Suppress warnings on conditionals.}
4865 @cindex @option{-gnatwC} (@command{gcc})
4866 This switch suppresses warnings for conditional expressions used in
4867 tests that are known to be True or False at compile time.
4870 @emph{Activate warnings on missing component clauses.}
4871 @cindex @option{-gnatw.c} (@command{gcc})
4872 @cindex Component clause, missing
4873 This switch activates warnings for record components where a record
4874 representation clause is present and has component clauses for the
4875 majority, but not all, of the components. A warning is given for each
4876 component for which no component clause is present.
4878 This warning can also be turned on using @option{-gnatwa}.
4881 @emph{Suppress warnings on missing component clauses.}
4882 @cindex @option{-gnatwC} (@command{gcc})
4883 This switch suppresses warnings for record components that are
4884 missing a component clause in the situation described above.
4887 @emph{Activate warnings on implicit dereferencing.}
4888 @cindex @option{-gnatwd} (@command{gcc})
4889 If this switch is set, then the use of a prefix of an access type
4890 in an indexed component, slice, or selected component without an
4891 explicit @code{.all} will generate a warning. With this warning
4892 enabled, access checks occur only at points where an explicit
4893 @code{.all} appears in the source code (assuming no warnings are
4894 generated as a result of this switch). The default is that such
4895 warnings are not generated.
4896 Note that @option{-gnatwa} does not affect the setting of
4897 this warning option.
4900 @emph{Suppress warnings on implicit dereferencing.}
4901 @cindex @option{-gnatwD} (@command{gcc})
4902 @cindex Implicit dereferencing
4903 @cindex Dereferencing, implicit
4904 This switch suppresses warnings for implicit dereferences in
4905 indexed components, slices, and selected components.
4908 @emph{Treat warnings as errors.}
4909 @cindex @option{-gnatwe} (@command{gcc})
4910 @cindex Warnings, treat as error
4911 This switch causes warning messages to be treated as errors.
4912 The warning string still appears, but the warning messages are counted
4913 as errors, and prevent the generation of an object file.
4916 @emph{Activate every optional warning}
4917 @cindex @option{-gnatw.e} (@command{gcc})
4918 @cindex Warnings, activate every optional warning
4919 This switch activates all optional warnings, including those which
4920 are not activated by @code{-gnatwa}.
4923 @emph{Activate warnings on unreferenced formals.}
4924 @cindex @option{-gnatwf} (@command{gcc})
4925 @cindex Formals, unreferenced
4926 This switch causes a warning to be generated if a formal parameter
4927 is not referenced in the body of the subprogram. This warning can
4928 also be turned on using @option{-gnatwa} or @option{-gnatwu}. The
4929 default is that these warnings are not generated.
4932 @emph{Suppress warnings on unreferenced formals.}
4933 @cindex @option{-gnatwF} (@command{gcc})
4934 This switch suppresses warnings for unreferenced formal
4935 parameters. Note that the
4936 combination @option{-gnatwu} followed by @option{-gnatwF} has the
4937 effect of warning on unreferenced entities other than subprogram
4941 @emph{Activate warnings on unrecognized pragmas.}
4942 @cindex @option{-gnatwg} (@command{gcc})
4943 @cindex Pragmas, unrecognized
4944 This switch causes a warning to be generated if an unrecognized
4945 pragma is encountered. Apart from issuing this warning, the
4946 pragma is ignored and has no effect. This warning can
4947 also be turned on using @option{-gnatwa}. The default
4948 is that such warnings are issued (satisfying the Ada Reference
4949 Manual requirement that such warnings appear).
4952 @emph{Suppress warnings on unrecognized pragmas.}
4953 @cindex @option{-gnatwG} (@command{gcc})
4954 This switch suppresses warnings for unrecognized pragmas.
4957 @emph{Activate warnings on hiding.}
4958 @cindex @option{-gnatwh} (@command{gcc})
4959 @cindex Hiding of Declarations
4960 This switch activates warnings on hiding declarations.
4961 A declaration is considered hiding
4962 if it is for a non-overloadable entity, and it declares an entity with the
4963 same name as some other entity that is directly or use-visible. The default
4964 is that such warnings are not generated.
4965 Note that @option{-gnatwa} does not affect the setting of this warning option.
4968 @emph{Suppress warnings on hiding.}
4969 @cindex @option{-gnatwH} (@command{gcc})
4970 This switch suppresses warnings on hiding declarations.
4973 @emph{Activate warnings on implementation units.}
4974 @cindex @option{-gnatwi} (@command{gcc})
4975 This switch activates warnings for a @code{with} of an internal GNAT
4976 implementation unit, defined as any unit from the @code{Ada},
4977 @code{Interfaces}, @code{GNAT},
4978 ^^@code{DEC},^ or @code{System}
4979 hierarchies that is not
4980 documented in either the Ada Reference Manual or the GNAT
4981 Programmer's Reference Manual. Such units are intended only
4982 for internal implementation purposes and should not be @code{with}'ed
4983 by user programs. The default is that such warnings are generated
4984 This warning can also be turned on using @option{-gnatwa}.
4987 @emph{Disable warnings on implementation units.}
4988 @cindex @option{-gnatwI} (@command{gcc})
4989 This switch disables warnings for a @code{with} of an internal GNAT
4990 implementation unit.
4993 @emph{Activate warnings on obsolescent features (Annex J).}
4994 @cindex @option{-gnatwj} (@command{gcc})
4995 @cindex Features, obsolescent
4996 @cindex Obsolescent features
4997 If this warning option is activated, then warnings are generated for
4998 calls to subprograms marked with @code{pragma Obsolescent} and
4999 for use of features in Annex J of the Ada Reference Manual. In the
5000 case of Annex J, not all features are flagged. In particular use
5001 of the renamed packages (like @code{Text_IO}) and use of package
5002 @code{ASCII} are not flagged, since these are very common and
5003 would generate many annoying positive warnings. The default is that
5004 such warnings are not generated. This warning is also turned on by
5005 the use of @option{-gnatwa}.
5007 In addition to the above cases, warnings are also generated for
5008 GNAT features that have been provided in past versions but which
5009 have been superseded (typically by features in the new Ada standard).
5010 For example, @code{pragma Ravenscar} will be flagged since its
5011 function is replaced by @code{pragma Profile(Ravenscar)}.
5013 Note that this warning option functions differently from the
5014 restriction @code{No_Obsolescent_Features} in two respects.
5015 First, the restriction applies only to annex J features.
5016 Second, the restriction does flag uses of package @code{ASCII}.
5019 @emph{Suppress warnings on obsolescent features (Annex J).}
5020 @cindex @option{-gnatwJ} (@command{gcc})
5021 This switch disables warnings on use of obsolescent features.
5024 @emph{Activate warnings on variables that could be constants.}
5025 @cindex @option{-gnatwk} (@command{gcc})
5026 This switch activates warnings for variables that are initialized but
5027 never modified, and then could be declared constants. The default is that
5028 such warnings are not given.
5029 This warning can also be turned on using @option{-gnatwa}.
5032 @emph{Suppress warnings on variables that could be constants.}
5033 @cindex @option{-gnatwK} (@command{gcc})
5034 This switch disables warnings on variables that could be declared constants.
5037 @emph{Activate warnings for elaboration pragmas.}
5038 @cindex @option{-gnatwl} (@command{gcc})
5039 @cindex Elaboration, warnings
5040 This switch activates warnings on missing
5041 @code{Elaborate_All} and @code{Elaborate} pragmas.
5042 See the section in this guide on elaboration checking for details on
5043 when such pragmas should be used. In dynamic elaboration mode, this switch
5044 generations warnings about the need to add elaboration pragmas. Note however,
5045 that if you blindly follow these warnings, and add @code{Elaborate_All}
5046 warnings wherever they are recommended, you basically end up with the
5047 equivalent of the static elaboration model, which may not be what you want for
5048 legacy code for which the static model does not work.
5050 For the static model, the messages generated are labeled "info:" (for
5051 information messages). They are not warnings to add elaboration pragmas,
5052 merely informational messages showing what implicit elaboration pragmas
5053 have been added, for use in analyzing elaboration circularity problems.
5055 Warnings are also generated if you
5056 are using the static mode of elaboration, and a @code{pragma Elaborate}
5057 is encountered. The default is that such warnings
5059 This warning is not automatically turned on by the use of @option{-gnatwa}.
5062 @emph{Suppress warnings for elaboration pragmas.}
5063 @cindex @option{-gnatwL} (@command{gcc})
5064 This switch suppresses warnings on missing Elaborate and Elaborate_All pragmas.
5065 See the section in this guide on elaboration checking for details on
5066 when such pragmas should be used.
5069 @emph{Activate warnings on modified but unreferenced variables.}
5070 @cindex @option{-gnatwm} (@command{gcc})
5071 This switch activates warnings for variables that are assigned (using
5072 an initialization value or with one or more assignment statements) but
5073 whose value is never read. The warning is suppressed for volatile
5074 variables and also for variables that are renamings of other variables
5075 or for which an address clause is given.
5076 This warning can also be turned on using @option{-gnatwa}.
5077 The default is that these warnings are not given.
5080 @emph{Disable warnings on modified but unreferenced variables.}
5081 @cindex @option{-gnatwM} (@command{gcc})
5082 This switch disables warnings for variables that are assigned or
5083 initialized, but never read.
5086 @emph{Set normal warnings mode.}
5087 @cindex @option{-gnatwn} (@command{gcc})
5088 This switch sets normal warning mode, in which enabled warnings are
5089 issued and treated as warnings rather than errors. This is the default
5090 mode. the switch @option{-gnatwn} can be used to cancel the effect of
5091 an explicit @option{-gnatws} or
5092 @option{-gnatwe}. It also cancels the effect of the
5093 implicit @option{-gnatwe} that is activated by the
5094 use of @option{-gnatg}.
5097 @emph{Activate warnings on address clause overlays.}
5098 @cindex @option{-gnatwo} (@command{gcc})
5099 @cindex Address Clauses, warnings
5100 This switch activates warnings for possibly unintended initialization
5101 effects of defining address clauses that cause one variable to overlap
5102 another. The default is that such warnings are generated.
5103 This warning can also be turned on using @option{-gnatwa}.
5106 @emph{Suppress warnings on address clause overlays.}
5107 @cindex @option{-gnatwO} (@command{gcc})
5108 This switch suppresses warnings on possibly unintended initialization
5109 effects of defining address clauses that cause one variable to overlap
5113 @emph{Activate warnings on modified but unreferenced out parameters.}
5114 @cindex @option{-gnatw.o} (@command{gcc})
5115 This switch activates warnings for variables that are modified by using
5116 them as actuals for a call to a procedure with an out mode formal, where
5117 the resulting assigned value is never read. It is applicable in the case
5118 where there is more than one out mode formal. If there is only one out
5119 mode formal, the warning is issued by default (controlled by -gnatwu).
5120 The warning is suppressed for volatile
5121 variables and also for variables that are renamings of other variables
5122 or for which an address clause is given.
5123 The default is that these warnings are not given. Note that this warning
5124 is not included in -gnatwa, it must be activated explicitly.
5127 @emph{Disable warnings on modified but unreferenced out parameters.}
5128 @cindex @option{-gnatw.O} (@command{gcc})
5129 This switch suppresses warnings for variables that are modified by using
5130 them as actuals for a call to a procedure with an out mode formal, where
5131 the resulting assigned value is never read.
5134 @emph{Activate warnings on ineffective pragma Inlines.}
5135 @cindex @option{-gnatwp} (@command{gcc})
5136 @cindex Inlining, warnings
5137 This switch activates warnings for failure of front end inlining
5138 (activated by @option{-gnatN}) to inline a particular call. There are
5139 many reasons for not being able to inline a call, including most
5140 commonly that the call is too complex to inline. The default is
5141 that such warnings are not given.
5142 This warning can also be turned on using @option{-gnatwa}.
5143 Warnings on ineffective inlining by the gcc back-end can be activated
5144 separately, using the gcc switch -Winline.
5147 @emph{Suppress warnings on ineffective pragma Inlines.}
5148 @cindex @option{-gnatwP} (@command{gcc})
5149 This switch suppresses warnings on ineffective pragma Inlines. If the
5150 inlining mechanism cannot inline a call, it will simply ignore the
5154 @emph{Activate warnings on parameter ordering.}
5155 @cindex @option{-gnatw.p} (@command{gcc})
5156 @cindex Parameter order, warnings
5157 This switch activates warnings for cases of suspicious parameter
5158 ordering when the list of arguments are all simple identifiers that
5159 match the names of the formals, but are in a different order. The
5160 warning is suppressed if any use of named parameter notation is used,
5161 so this is the appropriate way to suppress a false positive (and
5162 serves to emphasize that the "misordering" is deliberate). The
5164 that such warnings are not given.
5165 This warning can also be turned on using @option{-gnatwa}.
5168 @emph{Suppress warnings on parameter ordering.}
5169 @cindex @option{-gnatw.P} (@command{gcc})
5170 This switch suppresses warnings on cases of suspicious parameter
5174 @emph{Activate warnings on questionable missing parentheses.}
5175 @cindex @option{-gnatwq} (@command{gcc})
5176 @cindex Parentheses, warnings
5177 This switch activates warnings for cases where parentheses are not used and
5178 the result is potential ambiguity from a readers point of view. For example
5179 (not a > b) when a and b are modular means ((not a) > b) and very likely the
5180 programmer intended (not (a > b)). Similarly (-x mod 5) means (-(x mod 5)) and
5181 quite likely ((-x) mod 5) was intended. In such situations it seems best to
5182 follow the rule of always parenthesizing to make the association clear, and
5183 this warning switch warns if such parentheses are not present. The default
5184 is that these warnings are given.
5185 This warning can also be turned on using @option{-gnatwa}.
5188 @emph{Suppress warnings on questionable missing parentheses.}
5189 @cindex @option{-gnatwQ} (@command{gcc})
5190 This switch suppresses warnings for cases where the association is not
5191 clear and the use of parentheses is preferred.
5194 @emph{Activate warnings on redundant constructs.}
5195 @cindex @option{-gnatwr} (@command{gcc})
5196 This switch activates warnings for redundant constructs. The following
5197 is the current list of constructs regarded as redundant:
5201 Assignment of an item to itself.
5203 Type conversion that converts an expression to its own type.
5205 Use of the attribute @code{Base} where @code{typ'Base} is the same
5208 Use of pragma @code{Pack} when all components are placed by a record
5209 representation clause.
5211 Exception handler containing only a reraise statement (raise with no
5212 operand) which has no effect.
5214 Use of the operator abs on an operand that is known at compile time
5217 Comparison of boolean expressions to an explicit True value.
5220 This warning can also be turned on using @option{-gnatwa}.
5221 The default is that warnings for redundant constructs are not given.
5224 @emph{Suppress warnings on redundant constructs.}
5225 @cindex @option{-gnatwR} (@command{gcc})
5226 This switch suppresses warnings for redundant constructs.
5229 @emph{Suppress all warnings.}
5230 @cindex @option{-gnatws} (@command{gcc})
5231 This switch completely suppresses the
5232 output of all warning messages from the GNAT front end.
5233 Note that it does not suppress warnings from the @command{gcc} back end.
5234 To suppress these back end warnings as well, use the switch @option{-w}
5235 in addition to @option{-gnatws}.
5238 @emph{Activate warnings for tracking of deleted conditional code.}
5239 @cindex @option{-gnatwt} (@command{gcc})
5240 @cindex Deactivated code, warnings
5241 @cindex Deleted code, warnings
5242 This switch activates warnings for tracking of code in conditionals (IF and
5243 CASE statements) that is detected to be dead code which cannot be executed, and
5244 which is removed by the front end. This warning is off by default, and is not
5245 turned on by @option{-gnatwa}, it has to be turned on explicitly. This may be
5246 useful for detecting deactivated code in certified applications.
5249 @emph{Suppress warnings for tracking of deleted conditional code.}
5250 @cindex @option{-gnatwT} (@command{gcc})
5251 This switch suppresses warnings for tracking of deleted conditional code.
5254 @emph{Activate warnings on unused entities.}
5255 @cindex @option{-gnatwu} (@command{gcc})
5256 This switch activates warnings to be generated for entities that
5257 are declared but not referenced, and for units that are @code{with}'ed
5259 referenced. In the case of packages, a warning is also generated if
5260 no entities in the package are referenced. This means that if the package
5261 is referenced but the only references are in @code{use}
5262 clauses or @code{renames}
5263 declarations, a warning is still generated. A warning is also generated
5264 for a generic package that is @code{with}'ed but never instantiated.
5265 In the case where a package or subprogram body is compiled, and there
5266 is a @code{with} on the corresponding spec
5267 that is only referenced in the body,
5268 a warning is also generated, noting that the
5269 @code{with} can be moved to the body. The default is that
5270 such warnings are not generated.
5271 This switch also activates warnings on unreferenced formals
5272 (it includes the effect of @option{-gnatwf}).
5273 This warning can also be turned on using @option{-gnatwa}.
5276 @emph{Suppress warnings on unused entities.}
5277 @cindex @option{-gnatwU} (@command{gcc})
5278 This switch suppresses warnings for unused entities and packages.
5279 It also turns off warnings on unreferenced formals (and thus includes
5280 the effect of @option{-gnatwF}).
5283 @emph{Activate warnings on unassigned variables.}
5284 @cindex @option{-gnatwv} (@command{gcc})
5285 @cindex Unassigned variable warnings
5286 This switch activates warnings for access to variables which
5287 may not be properly initialized. The default is that
5288 such warnings are generated.
5289 This warning can also be turned on using @option{-gnatwa}.
5292 @emph{Suppress warnings on unassigned variables.}
5293 @cindex @option{-gnatwV} (@command{gcc})
5294 This switch suppresses warnings for access to variables which
5295 may not be properly initialized.
5296 For variables of a composite type, the warning can also be suppressed in
5297 Ada 2005 by using a default initialization with a box. For example, if
5298 Table is an array of records whose components are only partially uninitialized,
5299 then the following code:
5301 @smallexample @c ada
5302 Tab : Table := (others => <>);
5305 will suppress warnings on subsequent statements that access components
5309 @emph{Activate warnings on wrong low bound assumption.}
5310 @cindex @option{-gnatww} (@command{gcc})
5311 @cindex String indexing warnings
5312 This switch activates warnings for indexing an unconstrained string parameter
5313 with a literal or S'Length. This is a case where the code is assuming that the
5314 low bound is one, which is in general not true (for example when a slice is
5315 passed). The default is that such warnings are generated.
5316 This warning can also be turned on using @option{-gnatwa}.
5319 @emph{Suppress warnings on wrong low bound assumption.}
5320 @cindex @option{-gnatwW} (@command{gcc})
5321 This switch suppresses warnings for indexing an unconstrained string parameter
5322 with a literal or S'Length. Note that this warning can also be suppressed
5323 in a particular case by adding an
5324 assertion that the lower bound is 1,
5325 as shown in the following example.
5327 @smallexample @c ada
5328 procedure K (S : String) is
5329 pragma Assert (S'First = 1);
5334 @emph{Activate warnings on unnecessary Warnings Off pragmas}
5335 @cindex @option{-gnatw.w} (@command{gcc})
5336 @cindex Warnings Off control
5337 This switch activates warnings for use of @code{pragma Warnings (Off, entity}
5338 where either the pragma is entirely useless (because it suppresses no
5339 warnings), or it could be replaced by @code{pragma Unreferenced} or
5340 @code{pragma Unmodified}.The default is that these warnings are not given.
5341 Note that this warning is not included in -gnatwa, it must be
5342 activated explicitly.
5345 @emph{Suppress warnings on unnecessary Warnings Off pragmas}
5346 @cindex @option{-gnatw.W} (@command{gcc})
5347 This switch suppresses warnings for use of @code{pragma Warnings (Off, entity}.
5350 @emph{Activate warnings on Export/Import pragmas.}
5351 @cindex @option{-gnatwx} (@command{gcc})
5352 @cindex Export/Import pragma warnings
5353 This switch activates warnings on Export/Import pragmas when
5354 the compiler detects a possible conflict between the Ada and
5355 foreign language calling sequences. For example, the use of
5356 default parameters in a convention C procedure is dubious
5357 because the C compiler cannot supply the proper default, so
5358 a warning is issued. The default is that such warnings are
5360 This warning can also be turned on using @option{-gnatwa}.
5363 @emph{Suppress warnings on Export/Import pragmas.}
5364 @cindex @option{-gnatwX} (@command{gcc})
5365 This switch suppresses warnings on Export/Import pragmas.
5366 The sense of this is that you are telling the compiler that
5367 you know what you are doing in writing the pragma, and it
5368 should not complain at you.
5371 @emph{Activate warnings for No_Exception_Propagation mode.}
5372 @cindex @option{-gnatwm} (@command{gcc})
5373 This switch activates warnings for exception usage when pragma Restrictions
5374 (No_Exception_Propagation) is in effect. Warnings are given for implicit or
5375 explicit exception raises which are not covered by a local handler, and for
5376 exception handlers which do not cover a local raise. The default is that these
5377 warnings are not given.
5380 @emph{Disable warnings for No_Exception_Propagation mode.}
5381 This switch disables warnings for exception usage when pragma Restrictions
5382 (No_Exception_Propagation) is in effect.
5385 @emph{Activate warnings for Ada 2005 compatibility issues.}
5386 @cindex @option{-gnatwy} (@command{gcc})
5387 @cindex Ada 2005 compatibility issues warnings
5388 For the most part Ada 2005 is upwards compatible with Ada 95,
5389 but there are some exceptions (for example the fact that
5390 @code{interface} is now a reserved word in Ada 2005). This
5391 switch activates several warnings to help in identifying
5392 and correcting such incompatibilities. The default is that
5393 these warnings are generated. Note that at one point Ada 2005
5394 was called Ada 0Y, hence the choice of character.
5395 This warning can also be turned on using @option{-gnatwa}.
5398 @emph{Disable warnings for Ada 2005 compatibility issues.}
5399 @cindex @option{-gnatwY} (@command{gcc})
5400 @cindex Ada 2005 compatibility issues warnings
5401 This switch suppresses several warnings intended to help in identifying
5402 incompatibilities between Ada 95 and Ada 2005.
5405 @emph{Activate warnings on unchecked conversions.}
5406 @cindex @option{-gnatwz} (@command{gcc})
5407 @cindex Unchecked_Conversion warnings
5408 This switch activates warnings for unchecked conversions
5409 where the types are known at compile time to have different
5411 is that such warnings are generated. Warnings are also
5412 generated for subprogram pointers with different conventions,
5413 and, on VMS only, for data pointers with different conventions.
5414 This warning can also be turned on using @option{-gnatwa}.
5417 @emph{Suppress warnings on unchecked conversions.}
5418 @cindex @option{-gnatwZ} (@command{gcc})
5419 This switch suppresses warnings for unchecked conversions
5420 where the types are known at compile time to have different
5421 sizes or conventions.
5423 @item ^-Wunused^WARNINGS=UNUSED^
5424 @cindex @option{-Wunused}
5425 The warnings controlled by the @option{-gnatw} switch are generated by
5426 the front end of the compiler. The @option{GCC} back end can provide
5427 additional warnings and they are controlled by the @option{-W} switch.
5428 For example, @option{^-Wunused^WARNINGS=UNUSED^} activates back end
5429 warnings for entities that are declared but not referenced.
5431 @item ^-Wuninitialized^WARNINGS=UNINITIALIZED^
5432 @cindex @option{-Wuninitialized}
5433 Similarly, @option{^-Wuninitialized^WARNINGS=UNINITIALIZED^} activates
5434 the back end warning for uninitialized variables. This switch must be
5435 used in conjunction with an optimization level greater than zero.
5437 @item ^-Wall^/ALL_BACK_END_WARNINGS^
5438 @cindex @option{-Wall}
5439 This switch enables all the above warnings from the @option{GCC} back end.
5440 The code generator detects a number of warning situations that are missed
5441 by the @option{GNAT} front end, and this switch can be used to activate them.
5442 The use of this switch also sets the default front end warning mode to
5443 @option{-gnatwa}, that is, most front end warnings activated as well.
5445 @item ^-w^/NO_BACK_END_WARNINGS^
5447 Conversely, this switch suppresses warnings from the @option{GCC} back end.
5448 The use of this switch also sets the default front end warning mode to
5449 @option{-gnatws}, that is, front end warnings suppressed as well.
5455 A string of warning parameters can be used in the same parameter. For example:
5462 will turn on all optional warnings except for elaboration pragma warnings,
5463 and also specify that warnings should be treated as errors.
5465 When no switch @option{^-gnatw^/WARNINGS^} is used, this is equivalent to:
5490 @node Debugging and Assertion Control
5491 @subsection Debugging and Assertion Control
5495 @cindex @option{-gnata} (@command{gcc})
5501 The pragmas @code{Assert} and @code{Debug} normally have no effect and
5502 are ignored. This switch, where @samp{a} stands for assert, causes
5503 @code{Assert} and @code{Debug} pragmas to be activated.
5505 The pragmas have the form:
5509 @b{pragma} Assert (@var{Boolean-expression} @r{[},
5510 @var{static-string-expression}@r{]})
5511 @b{pragma} Debug (@var{procedure call})
5516 The @code{Assert} pragma causes @var{Boolean-expression} to be tested.
5517 If the result is @code{True}, the pragma has no effect (other than
5518 possible side effects from evaluating the expression). If the result is
5519 @code{False}, the exception @code{Assert_Failure} declared in the package
5520 @code{System.Assertions} is
5521 raised (passing @var{static-string-expression}, if present, as the
5522 message associated with the exception). If no string expression is
5523 given the default is a string giving the file name and line number
5526 The @code{Debug} pragma causes @var{procedure} to be called. Note that
5527 @code{pragma Debug} may appear within a declaration sequence, allowing
5528 debugging procedures to be called between declarations.
5531 @item /DEBUG@r{[}=debug-level@r{]}
5533 Specifies how much debugging information is to be included in
5534 the resulting object file where 'debug-level' is one of the following:
5537 Include both debugger symbol records and traceback
5539 This is the default setting.
5541 Include both debugger symbol records and traceback in
5544 Excludes both debugger symbol records and traceback
5545 the object file. Same as /NODEBUG.
5547 Includes only debugger symbol records in the object
5548 file. Note that this doesn't include traceback information.
5553 @node Validity Checking
5554 @subsection Validity Checking
5555 @findex Validity Checking
5558 The Ada Reference Manual has specific requirements for checking
5559 for invalid values. In particular, RM 13.9.1 requires that the
5560 evaluation of invalid values (for example from unchecked conversions),
5561 not result in erroneous execution. In GNAT, the result of such an
5562 evaluation in normal default mode is to either use the value
5563 unmodified, or to raise Constraint_Error in those cases where use
5564 of the unmodified value would cause erroneous execution. The cases
5565 where unmodified values might lead to erroneous execution are case
5566 statements (where a wild jump might result from an invalid value),
5567 and subscripts on the left hand side (where memory corruption could
5568 occur as a result of an invalid value).
5570 The @option{-gnatV^@var{x}^^} switch allows more control over the validity
5573 The @code{x} argument is a string of letters that
5574 indicate validity checks that are performed or not performed in addition
5575 to the default checks described above.
5578 The options allowed for this qualifier
5579 indicate validity checks that are performed or not performed in addition
5580 to the default checks described above.
5586 @emph{All validity checks.}
5587 @cindex @option{-gnatVa} (@command{gcc})
5588 All validity checks are turned on.
5590 That is, @option{-gnatVa} is
5591 equivalent to @option{gnatVcdfimorst}.
5595 @emph{Validity checks for copies.}
5596 @cindex @option{-gnatVc} (@command{gcc})
5597 The right hand side of assignments, and the initializing values of
5598 object declarations are validity checked.
5601 @emph{Default (RM) validity checks.}
5602 @cindex @option{-gnatVd} (@command{gcc})
5603 Some validity checks are done by default following normal Ada semantics
5605 A check is done in case statements that the expression is within the range
5606 of the subtype. If it is not, Constraint_Error is raised.
5607 For assignments to array components, a check is done that the expression used
5608 as index is within the range. If it is not, Constraint_Error is raised.
5609 Both these validity checks may be turned off using switch @option{-gnatVD}.
5610 They are turned on by default. If @option{-gnatVD} is specified, a subsequent
5611 switch @option{-gnatVd} will leave the checks turned on.
5612 Switch @option{-gnatVD} should be used only if you are sure that all such
5613 expressions have valid values. If you use this switch and invalid values
5614 are present, then the program is erroneous, and wild jumps or memory
5615 overwriting may occur.
5618 @emph{Validity checks for elementary components.}
5619 @cindex @option{-gnatVe} (@command{gcc})
5620 In the absence of this switch, assignments to record or array components are
5621 not validity checked, even if validity checks for assignments generally
5622 (@option{-gnatVc}) are turned on. In Ada, assignment of composite values do not
5623 require valid data, but assignment of individual components does. So for
5624 example, there is a difference between copying the elements of an array with a
5625 slice assignment, compared to assigning element by element in a loop. This
5626 switch allows you to turn off validity checking for components, even when they
5627 are assigned component by component.
5630 @emph{Validity checks for floating-point values.}
5631 @cindex @option{-gnatVf} (@command{gcc})
5632 In the absence of this switch, validity checking occurs only for discrete
5633 values. If @option{-gnatVf} is specified, then validity checking also applies
5634 for floating-point values, and NaNs and infinities are considered invalid,
5635 as well as out of range values for constrained types. Note that this means
5636 that standard IEEE infinity mode is not allowed. The exact contexts
5637 in which floating-point values are checked depends on the setting of other
5638 options. For example,
5639 @option{^-gnatVif^VALIDITY_CHECKING=(IN_PARAMS,FLOATS)^} or
5640 @option{^-gnatVfi^VALIDITY_CHECKING=(FLOATS,IN_PARAMS)^}
5641 (the order does not matter) specifies that floating-point parameters of mode
5642 @code{in} should be validity checked.
5645 @emph{Validity checks for @code{in} mode parameters}
5646 @cindex @option{-gnatVi} (@command{gcc})
5647 Arguments for parameters of mode @code{in} are validity checked in function
5648 and procedure calls at the point of call.
5651 @emph{Validity checks for @code{in out} mode parameters.}
5652 @cindex @option{-gnatVm} (@command{gcc})
5653 Arguments for parameters of mode @code{in out} are validity checked in
5654 procedure calls at the point of call. The @code{'m'} here stands for
5655 modify, since this concerns parameters that can be modified by the call.
5656 Note that there is no specific option to test @code{out} parameters,
5657 but any reference within the subprogram will be tested in the usual
5658 manner, and if an invalid value is copied back, any reference to it
5659 will be subject to validity checking.
5662 @emph{No validity checks.}
5663 @cindex @option{-gnatVn} (@command{gcc})
5664 This switch turns off all validity checking, including the default checking
5665 for case statements and left hand side subscripts. Note that the use of
5666 the switch @option{-gnatp} suppresses all run-time checks, including
5667 validity checks, and thus implies @option{-gnatVn}. When this switch
5668 is used, it cancels any other @option{-gnatV} previously issued.
5671 @emph{Validity checks for operator and attribute operands.}
5672 @cindex @option{-gnatVo} (@command{gcc})
5673 Arguments for predefined operators and attributes are validity checked.
5674 This includes all operators in package @code{Standard},
5675 the shift operators defined as intrinsic in package @code{Interfaces}
5676 and operands for attributes such as @code{Pos}. Checks are also made
5677 on individual component values for composite comparisons, and on the
5678 expressions in type conversions and qualified expressions. Checks are
5679 also made on explicit ranges using @samp{..} (e.g.@: slices, loops etc).
5682 @emph{Validity checks for parameters.}
5683 @cindex @option{-gnatVp} (@command{gcc})
5684 This controls the treatment of parameters within a subprogram (as opposed
5685 to @option{-gnatVi} and @option{-gnatVm} which control validity testing
5686 of parameters on a call. If either of these call options is used, then
5687 normally an assumption is made within a subprogram that the input arguments
5688 have been validity checking at the point of call, and do not need checking
5689 again within a subprogram). If @option{-gnatVp} is set, then this assumption
5690 is not made, and parameters are not assumed to be valid, so their validity
5691 will be checked (or rechecked) within the subprogram.
5694 @emph{Validity checks for function returns.}
5695 @cindex @option{-gnatVr} (@command{gcc})
5696 The expression in @code{return} statements in functions is validity
5700 @emph{Validity checks for subscripts.}
5701 @cindex @option{-gnatVs} (@command{gcc})
5702 All subscripts expressions are checked for validity, whether they appear
5703 on the right side or left side (in default mode only left side subscripts
5704 are validity checked).
5707 @emph{Validity checks for tests.}
5708 @cindex @option{-gnatVt} (@command{gcc})
5709 Expressions used as conditions in @code{if}, @code{while} or @code{exit}
5710 statements are checked, as well as guard expressions in entry calls.
5715 The @option{-gnatV} switch may be followed by
5716 ^a string of letters^a list of options^
5717 to turn on a series of validity checking options.
5719 @option{^-gnatVcr^/VALIDITY_CHECKING=(COPIES, RETURNS)^}
5720 specifies that in addition to the default validity checking, copies and
5721 function return expressions are to be validity checked.
5722 In order to make it easier
5723 to specify the desired combination of effects,
5725 the upper case letters @code{CDFIMORST} may
5726 be used to turn off the corresponding lower case option.
5729 the prefix @code{NO} on an option turns off the corresponding validity
5732 @item @code{NOCOPIES}
5733 @item @code{NODEFAULT}
5734 @item @code{NOFLOATS}
5735 @item @code{NOIN_PARAMS}
5736 @item @code{NOMOD_PARAMS}
5737 @item @code{NOOPERANDS}
5738 @item @code{NORETURNS}
5739 @item @code{NOSUBSCRIPTS}
5740 @item @code{NOTESTS}
5744 @option{^-gnatVaM^/VALIDITY_CHECKING=(ALL, NOMOD_PARAMS)^}
5745 turns on all validity checking options except for
5746 checking of @code{@b{in out}} procedure arguments.
5748 The specification of additional validity checking generates extra code (and
5749 in the case of @option{-gnatVa} the code expansion can be substantial).
5750 However, these additional checks can be very useful in detecting
5751 uninitialized variables, incorrect use of unchecked conversion, and other
5752 errors leading to invalid values. The use of pragma @code{Initialize_Scalars}
5753 is useful in conjunction with the extra validity checking, since this
5754 ensures that wherever possible uninitialized variables have invalid values.
5756 See also the pragma @code{Validity_Checks} which allows modification of
5757 the validity checking mode at the program source level, and also allows for
5758 temporary disabling of validity checks.
5760 @node Style Checking
5761 @subsection Style Checking
5762 @findex Style checking
5765 The @option{-gnaty^x^(option,option,@dots{})^} switch
5766 @cindex @option{-gnaty} (@command{gcc})
5767 causes the compiler to
5768 enforce specified style rules. A limited set of style rules has been used
5769 in writing the GNAT sources themselves. This switch allows user programs
5770 to activate all or some of these checks. If the source program fails a
5771 specified style check, an appropriate warning message is given, preceded by
5772 the character sequence ``(style)''.
5774 @code{(option,option,@dots{})} is a sequence of keywords
5777 The string @var{x} is a sequence of letters or digits
5779 indicating the particular style
5780 checks to be performed. The following checks are defined:
5785 @emph{Specify indentation level.}
5786 If a digit from 1-9 appears
5787 ^in the string after @option{-gnaty}^as an option for /STYLE_CHECKS^
5788 then proper indentation is checked, with the digit indicating the
5789 indentation level required. A value of zero turns off this style check.
5790 The general style of required indentation is as specified by
5791 the examples in the Ada Reference Manual. Full line comments must be
5792 aligned with the @code{--} starting on a column that is a multiple of
5793 the alignment level, or they may be aligned the same way as the following
5794 non-blank line (this is useful when full line comments appear in the middle
5798 @emph{Check attribute casing.}
5799 Attribute names, including the case of keywords such as @code{digits}
5800 used as attributes names, must be written in mixed case, that is, the
5801 initial letter and any letter following an underscore must be uppercase.
5802 All other letters must be lowercase.
5804 @item ^A^ARRAY_INDEXES^
5805 @emph{Use of array index numbers in array attributes.}
5806 When using the array attributes First, Last, Range,
5807 or Length, the index number must be omitted for one-dimensional arrays
5808 and is required for multi-dimensional arrays.
5811 @emph{Blanks not allowed at statement end.}
5812 Trailing blanks are not allowed at the end of statements. The purpose of this
5813 rule, together with h (no horizontal tabs), is to enforce a canonical format
5814 for the use of blanks to separate source tokens.
5817 @emph{Check comments.}
5818 Comments must meet the following set of rules:
5823 The ``@code{--}'' that starts the column must either start in column one,
5824 or else at least one blank must precede this sequence.
5827 Comments that follow other tokens on a line must have at least one blank
5828 following the ``@code{--}'' at the start of the comment.
5831 Full line comments must have two blanks following the ``@code{--}'' that
5832 starts the comment, with the following exceptions.
5835 A line consisting only of the ``@code{--}'' characters, possibly preceded
5836 by blanks is permitted.
5839 A comment starting with ``@code{--x}'' where @code{x} is a special character
5841 This allows proper processing of the output generated by specialized tools
5842 including @command{gnatprep} (where ``@code{--!}'' is used) and the SPARK
5844 language (where ``@code{--#}'' is used). For the purposes of this rule, a
5845 special character is defined as being in one of the ASCII ranges
5846 @code{16#21#@dots{}16#2F#} or @code{16#3A#@dots{}16#3F#}.
5847 Note that this usage is not permitted
5848 in GNAT implementation units (i.e., when @option{-gnatg} is used).
5851 A line consisting entirely of minus signs, possibly preceded by blanks, is
5852 permitted. This allows the construction of box comments where lines of minus
5853 signs are used to form the top and bottom of the box.
5856 A comment that starts and ends with ``@code{--}'' is permitted as long as at
5857 least one blank follows the initial ``@code{--}''. Together with the preceding
5858 rule, this allows the construction of box comments, as shown in the following
5861 ---------------------------
5862 -- This is a box comment --
5863 -- with two text lines. --
5864 ---------------------------
5868 @item ^d^DOS_LINE_ENDINGS^
5869 @emph{Check no DOS line terminators present.}
5870 All lines must be terminated by a single ASCII.LF
5871 character (in particular the DOS line terminator sequence CR/LF is not
5875 @emph{Check end/exit labels.}
5876 Optional labels on @code{end} statements ending subprograms and on
5877 @code{exit} statements exiting named loops, are required to be present.
5880 @emph{No form feeds or vertical tabs.}
5881 Neither form feeds nor vertical tab characters are permitted
5885 @emph{GNAT style mode}
5886 The set of style check switches is set to match that used by the GNAT sources.
5887 This may be useful when developing code that is eventually intended to be
5888 incorporated into GNAT. For further details, see GNAT sources.
5891 @emph{No horizontal tabs.}
5892 Horizontal tab characters are not permitted in the source text.
5893 Together with the b (no blanks at end of line) check, this
5894 enforces a canonical form for the use of blanks to separate
5898 @emph{Check if-then layout.}
5899 The keyword @code{then} must appear either on the same
5900 line as corresponding @code{if}, or on a line on its own, lined
5901 up under the @code{if} with at least one non-blank line in between
5902 containing all or part of the condition to be tested.
5905 @emph{check mode IN keywords}
5906 Mode @code{in} (the default mode) is not
5907 allowed to be given explicitly. @code{in out} is fine,
5908 but not @code{in} on its own.
5911 @emph{Check keyword casing.}
5912 All keywords must be in lower case (with the exception of keywords
5913 such as @code{digits} used as attribute names to which this check
5917 @emph{Check layout.}
5918 Layout of statement and declaration constructs must follow the
5919 recommendations in the Ada Reference Manual, as indicated by the
5920 form of the syntax rules. For example an @code{else} keyword must
5921 be lined up with the corresponding @code{if} keyword.
5923 There are two respects in which the style rule enforced by this check
5924 option are more liberal than those in the Ada Reference Manual. First
5925 in the case of record declarations, it is permissible to put the
5926 @code{record} keyword on the same line as the @code{type} keyword, and
5927 then the @code{end} in @code{end record} must line up under @code{type}.
5928 This is also permitted when the type declaration is split on two lines.
5929 For example, any of the following three layouts is acceptable:
5931 @smallexample @c ada
5954 Second, in the case of a block statement, a permitted alternative
5955 is to put the block label on the same line as the @code{declare} or
5956 @code{begin} keyword, and then line the @code{end} keyword up under
5957 the block label. For example both the following are permitted:
5959 @smallexample @c ada
5977 The same alternative format is allowed for loops. For example, both of
5978 the following are permitted:
5980 @smallexample @c ada
5982 Clear : while J < 10 loop
5993 @item ^Lnnn^MAX_NESTING=nnn^
5994 @emph{Set maximum nesting level}
5995 The maximum level of nesting of constructs (including subprograms, loops,
5996 blocks, packages, and conditionals) may not exceed the given value
5997 @option{nnn}. A value of zero disconnects this style check.
5999 @item ^m^LINE_LENGTH^
6000 @emph{Check maximum line length.}
6001 The length of source lines must not exceed 79 characters, including
6002 any trailing blanks. The value of 79 allows convenient display on an
6003 80 character wide device or window, allowing for possible special
6004 treatment of 80 character lines. Note that this count is of
6005 characters in the source text. This means that a tab character counts
6006 as one character in this count but a wide character sequence counts as
6007 a single character (however many bytes are needed in the encoding).
6009 @item ^Mnnn^MAX_LENGTH=nnn^
6010 @emph{Set maximum line length.}
6011 The length of lines must not exceed the
6012 given value @option{nnn}. The maximum value that can be specified is 32767.
6014 @item ^n^STANDARD_CASING^
6015 @emph{Check casing of entities in Standard.}
6016 Any identifier from Standard must be cased
6017 to match the presentation in the Ada Reference Manual (for example,
6018 @code{Integer} and @code{ASCII.NUL}).
6021 @emph{Turn off all style checks}
6022 All style check options are turned off.
6024 @item ^o^ORDERED_SUBPROGRAMS^
6025 @emph{Check order of subprogram bodies.}
6026 All subprogram bodies in a given scope
6027 (e.g.@: a package body) must be in alphabetical order. The ordering
6028 rule uses normal Ada rules for comparing strings, ignoring casing
6029 of letters, except that if there is a trailing numeric suffix, then
6030 the value of this suffix is used in the ordering (e.g.@: Junk2 comes
6034 @emph{Check pragma casing.}
6035 Pragma names must be written in mixed case, that is, the
6036 initial letter and any letter following an underscore must be uppercase.
6037 All other letters must be lowercase.
6039 @item ^r^REFERENCES^
6040 @emph{Check references.}
6041 All identifier references must be cased in the same way as the
6042 corresponding declaration. No specific casing style is imposed on
6043 identifiers. The only requirement is for consistency of references
6046 @item ^S^STATEMENTS_AFTER_THEN_ELSE^
6047 @emph{Check no statements after THEN/ELSE.}
6048 No statements are allowed
6049 on the same line as a THEN or ELSE keyword following the
6050 keyword in an IF statement. OR ELSE and AND THEN are not affected,
6051 and a special exception allows a pragma to appear after ELSE.
6054 @emph{Check separate specs.}
6055 Separate declarations (``specs'') are required for subprograms (a
6056 body is not allowed to serve as its own declaration). The only
6057 exception is that parameterless library level procedures are
6058 not required to have a separate declaration. This exception covers
6059 the most frequent form of main program procedures.
6062 @emph{Check token spacing.}
6063 The following token spacing rules are enforced:
6068 The keywords @code{@b{abs}} and @code{@b{not}} must be followed by a space.
6071 The token @code{=>} must be surrounded by spaces.
6074 The token @code{<>} must be preceded by a space or a left parenthesis.
6077 Binary operators other than @code{**} must be surrounded by spaces.
6078 There is no restriction on the layout of the @code{**} binary operator.
6081 Colon must be surrounded by spaces.
6084 Colon-equal (assignment, initialization) must be surrounded by spaces.
6087 Comma must be the first non-blank character on the line, or be
6088 immediately preceded by a non-blank character, and must be followed
6092 If the token preceding a left parenthesis ends with a letter or digit, then
6093 a space must separate the two tokens.
6096 A right parenthesis must either be the first non-blank character on
6097 a line, or it must be preceded by a non-blank character.
6100 A semicolon must not be preceded by a space, and must not be followed by
6101 a non-blank character.
6104 A unary plus or minus may not be followed by a space.
6107 A vertical bar must be surrounded by spaces.
6110 @item ^u^UNNECESSARY_BLANK_LINES^
6111 @emph{Check unnecessary blank lines.}
6112 Unnecessary blank lines are not allowed. A blank line is considered
6113 unnecessary if it appears at the end of the file, or if more than
6114 one blank line occurs in sequence.
6116 @item ^x^XTRA_PARENS^
6117 @emph{Check extra parentheses.}
6118 Unnecessary extra level of parentheses (C-style) are not allowed
6119 around conditions in @code{if} statements, @code{while} statements and
6120 @code{exit} statements.
6122 @item ^y^ALL_BUILTIN^
6123 @emph{Set all standard style check options}
6124 This is equivalent to @code{gnaty3aAbcefhiklmnprst}, that is all checking
6125 options enabled with the exception of @option{-gnatyo}, @option{-gnatyI},
6126 @option{-gnatyS}, @option{-gnatyLnnn},
6127 @option{-gnatyd}, @option{-gnatyu}, and @option{-gnatyx}.
6131 @emph{Remove style check options}
6132 This causes any subsequent options in the string to act as canceling the
6133 corresponding style check option. To cancel maximum nesting level control,
6134 use @option{L} parameter witout any integer value after that, because any
6135 digit following @option{-} in the parameter string of the @option{-gnaty}
6136 option will be threated as canceling indentation check. The same is true
6137 for @option{M} parameter. @option{y} and @option{N} parameters are not
6138 allowed after @option{-}.
6141 This causes any subsequent options in the string to enable the corresponding
6142 style check option. That is, it cancels the effect of a previous ^-^REMOVE^,
6148 @emph{Removing style check options}
6149 If the name of a style check is preceded by @option{NO} then the corresponding
6150 style check is turned off. For example @option{NOCOMMENTS} turns off style
6151 checking for comments.
6156 In the above rules, appearing in column one is always permitted, that is,
6157 counts as meeting either a requirement for a required preceding space,
6158 or as meeting a requirement for no preceding space.
6160 Appearing at the end of a line is also always permitted, that is, counts
6161 as meeting either a requirement for a following space, or as meeting
6162 a requirement for no following space.
6165 If any of these style rules is violated, a message is generated giving
6166 details on the violation. The initial characters of such messages are
6167 always ``@code{(style)}''. Note that these messages are treated as warning
6168 messages, so they normally do not prevent the generation of an object
6169 file. The @option{-gnatwe} switch can be used to treat warning messages,
6170 including style messages, as fatal errors.
6174 @option{-gnaty} on its own (that is not
6175 followed by any letters or digits), then the effect is equivalent
6176 to the use of @option{-gnatyy}, as described above, that is all
6177 built-in standard style check options are enabled.
6181 /STYLE_CHECKS=ALL_BUILTIN enables all checking options with
6182 the exception of ORDERED_SUBPROGRAMS, UNNECESSARY_BLANK_LINES,
6183 XTRA_PARENS, and DOS_LINE_ENDINGS. In addition
6195 clears any previously set style checks.
6197 @node Run-Time Checks
6198 @subsection Run-Time Checks
6199 @cindex Division by zero
6200 @cindex Access before elaboration
6201 @cindex Checks, division by zero
6202 @cindex Checks, access before elaboration
6203 @cindex Checks, stack overflow checking
6206 If you compile with the default options, GNAT will insert many run-time
6207 checks into the compiled code, including code that performs range
6208 checking against constraints, but not arithmetic overflow checking for
6209 integer operations (including division by zero), checks for access
6210 before elaboration on subprogram calls, or stack overflow checking. All
6211 other run-time checks, as required by the Ada Reference Manual, are
6212 generated by default. The following @command{gcc} switches refine this
6218 @cindex @option{-gnatp} (@command{gcc})
6219 @cindex Suppressing checks
6220 @cindex Checks, suppressing
6222 Suppress all run-time checks as though @code{pragma Suppress (all_checks})
6223 had been present in the source. Validity checks are also suppressed (in
6224 other words @option{-gnatp} also implies @option{-gnatVn}.
6225 Use this switch to improve the performance
6226 of the code at the expense of safety in the presence of invalid data or
6230 @cindex @option{-gnato} (@command{gcc})
6231 @cindex Overflow checks
6232 @cindex Check, overflow
6233 Enables overflow checking for integer operations.
6234 This causes GNAT to generate slower and larger executable
6235 programs by adding code to check for overflow (resulting in raising
6236 @code{Constraint_Error} as required by standard Ada
6237 semantics). These overflow checks correspond to situations in which
6238 the true value of the result of an operation may be outside the base
6239 range of the result type. The following example shows the distinction:
6241 @smallexample @c ada
6242 X1 : Integer := Integer'Last;
6243 X2 : Integer range 1 .. 5 := 5;
6244 X3 : Integer := Integer'Last;
6245 X4 : Integer range 1 .. 5 := 5;
6246 F : Float := 2.0E+20;
6255 Here the first addition results in a value that is outside the base range
6256 of Integer, and hence requires an overflow check for detection of the
6257 constraint error. Thus the first assignment to @code{X1} raises a
6258 @code{Constraint_Error} exception only if @option{-gnato} is set.
6260 The second increment operation results in a violation
6261 of the explicit range constraint, and such range checks are always
6262 performed (unless specifically suppressed with a pragma @code{suppress}
6263 or the use of @option{-gnatp}).
6265 The two conversions of @code{F} both result in values that are outside
6266 the base range of type @code{Integer} and thus will raise
6267 @code{Constraint_Error} exceptions only if @option{-gnato} is used.
6268 The fact that the result of the second conversion is assigned to
6269 variable @code{X4} with a restricted range is irrelevant, since the problem
6270 is in the conversion, not the assignment.
6272 Basically the rule is that in the default mode (@option{-gnato} not
6273 used), the generated code assures that all integer variables stay
6274 within their declared ranges, or within the base range if there is
6275 no declared range. This prevents any serious problems like indexes
6276 out of range for array operations.
6278 What is not checked in default mode is an overflow that results in
6279 an in-range, but incorrect value. In the above example, the assignments
6280 to @code{X1}, @code{X2}, @code{X3} all give results that are within the
6281 range of the target variable, but the result is wrong in the sense that
6282 it is too large to be represented correctly. Typically the assignment
6283 to @code{X1} will result in wrap around to the largest negative number.
6284 The conversions of @code{F} will result in some @code{Integer} value
6285 and if that integer value is out of the @code{X4} range then the
6286 subsequent assignment would generate an exception.
6288 @findex Machine_Overflows
6289 Note that the @option{-gnato} switch does not affect the code generated
6290 for any floating-point operations; it applies only to integer
6292 For floating-point, GNAT has the @code{Machine_Overflows}
6293 attribute set to @code{False} and the normal mode of operation is to
6294 generate IEEE NaN and infinite values on overflow or invalid operations
6295 (such as dividing 0.0 by 0.0).
6297 The reason that we distinguish overflow checking from other kinds of
6298 range constraint checking is that a failure of an overflow check, unlike
6299 for example the failure of a range check, can result in an incorrect
6300 value, but cannot cause random memory destruction (like an out of range
6301 subscript), or a wild jump (from an out of range case value). Overflow
6302 checking is also quite expensive in time and space, since in general it
6303 requires the use of double length arithmetic.
6305 Note again that @option{-gnato} is off by default, so overflow checking is
6306 not performed in default mode. This means that out of the box, with the
6307 default settings, GNAT does not do all the checks expected from the
6308 language description in the Ada Reference Manual. If you want all constraint
6309 checks to be performed, as described in this Manual, then you must
6310 explicitly use the -gnato switch either on the @command{gnatmake} or
6311 @command{gcc} command.
6314 @cindex @option{-gnatE} (@command{gcc})
6315 @cindex Elaboration checks
6316 @cindex Check, elaboration
6317 Enables dynamic checks for access-before-elaboration
6318 on subprogram calls and generic instantiations.
6319 For full details of the effect and use of this switch,
6320 @xref{Compiling Using gcc}.
6323 @cindex @option{-fstack-check} (@command{gcc})
6324 @cindex Stack Overflow Checking
6325 @cindex Checks, stack overflow checking
6326 Activates stack overflow checking. For full details of the effect and use of
6327 this switch see @ref{Stack Overflow Checking}.
6332 The setting of these switches only controls the default setting of the
6333 checks. You may modify them using either @code{Suppress} (to remove
6334 checks) or @code{Unsuppress} (to add back suppressed checks) pragmas in
6337 @node Using gcc for Syntax Checking
6338 @subsection Using @command{gcc} for Syntax Checking
6341 @cindex @option{-gnats} (@command{gcc})
6345 The @code{s} stands for ``syntax''.
6348 Run GNAT in syntax checking only mode. For
6349 example, the command
6352 $ gcc -c -gnats x.adb
6356 compiles file @file{x.adb} in syntax-check-only mode. You can check a
6357 series of files in a single command
6359 , and can use wild cards to specify such a group of files.
6360 Note that you must specify the @option{-c} (compile
6361 only) flag in addition to the @option{-gnats} flag.
6364 You may use other switches in conjunction with @option{-gnats}. In
6365 particular, @option{-gnatl} and @option{-gnatv} are useful to control the
6366 format of any generated error messages.
6368 When the source file is empty or contains only empty lines and/or comments,
6369 the output is a warning:
6372 $ gcc -c -gnats -x ada toto.txt
6373 toto.txt:1:01: warning: empty file, contains no compilation units
6377 Otherwise, the output is simply the error messages, if any. No object file or
6378 ALI file is generated by a syntax-only compilation. Also, no units other
6379 than the one specified are accessed. For example, if a unit @code{X}
6380 @code{with}'s a unit @code{Y}, compiling unit @code{X} in syntax
6381 check only mode does not access the source file containing unit
6384 @cindex Multiple units, syntax checking
6385 Normally, GNAT allows only a single unit in a source file. However, this
6386 restriction does not apply in syntax-check-only mode, and it is possible
6387 to check a file containing multiple compilation units concatenated
6388 together. This is primarily used by the @code{gnatchop} utility
6389 (@pxref{Renaming Files Using gnatchop}).
6392 @node Using gcc for Semantic Checking
6393 @subsection Using @command{gcc} for Semantic Checking
6396 @cindex @option{-gnatc} (@command{gcc})
6400 The @code{c} stands for ``check''.
6402 Causes the compiler to operate in semantic check mode,
6403 with full checking for all illegalities specified in the
6404 Ada Reference Manual, but without generation of any object code
6405 (no object file is generated).
6407 Because dependent files must be accessed, you must follow the GNAT
6408 semantic restrictions on file structuring to operate in this mode:
6412 The needed source files must be accessible
6413 (@pxref{Search Paths and the Run-Time Library (RTL)}).
6416 Each file must contain only one compilation unit.
6419 The file name and unit name must match (@pxref{File Naming Rules}).
6422 The output consists of error messages as appropriate. No object file is
6423 generated. An @file{ALI} file is generated for use in the context of
6424 cross-reference tools, but this file is marked as not being suitable
6425 for binding (since no object file is generated).
6426 The checking corresponds exactly to the notion of
6427 legality in the Ada Reference Manual.
6429 Any unit can be compiled in semantics-checking-only mode, including
6430 units that would not normally be compiled (subunits,
6431 and specifications where a separate body is present).
6434 @node Compiling Different Versions of Ada
6435 @subsection Compiling Different Versions of Ada
6438 The switches described in this section allow you to explicitly specify
6439 the version of the Ada language that your programs are written in.
6440 By default @value{EDITION} assumes @value{DEFAULTLANGUAGEVERSION},
6441 but you can also specify @value{NONDEFAULTLANGUAGEVERSION} or
6442 indicate Ada 83 compatibility mode.
6445 @cindex Compatibility with Ada 83
6447 @item -gnat83 (Ada 83 Compatibility Mode)
6448 @cindex @option{-gnat83} (@command{gcc})
6449 @cindex ACVC, Ada 83 tests
6453 Although GNAT is primarily an Ada 95 / Ada 2005 compiler, this switch
6454 specifies that the program is to be compiled in Ada 83 mode. With
6455 @option{-gnat83}, GNAT rejects most post-Ada 83 extensions and applies Ada 83
6456 semantics where this can be done easily.
6457 It is not possible to guarantee this switch does a perfect
6458 job; some subtle tests, such as are
6459 found in earlier ACVC tests (and that have been removed from the ACATS suite
6460 for Ada 95), might not compile correctly.
6461 Nevertheless, this switch may be useful in some circumstances, for example
6462 where, due to contractual reasons, existing code needs to be maintained
6463 using only Ada 83 features.
6465 With few exceptions (most notably the need to use @code{<>} on
6466 @cindex Generic formal parameters
6467 unconstrained generic formal parameters, the use of the new Ada 95 / Ada 2005
6468 reserved words, and the use of packages
6469 with optional bodies), it is not necessary to specify the
6470 @option{-gnat83} switch when compiling Ada 83 programs, because, with rare
6471 exceptions, Ada 95 and Ada 2005 are upwardly compatible with Ada 83. Thus
6472 a correct Ada 83 program is usually also a correct program
6473 in these later versions of the language standard.
6474 For further information, please refer to @ref{Compatibility and Porting Guide}.
6476 @item -gnat95 (Ada 95 mode)
6477 @cindex @option{-gnat95} (@command{gcc})
6481 This switch directs the compiler to implement the Ada 95 version of the
6483 Since Ada 95 is almost completely upwards
6484 compatible with Ada 83, Ada 83 programs may generally be compiled using
6485 this switch (see the description of the @option{-gnat83} switch for further
6486 information about Ada 83 mode).
6487 If an Ada 2005 program is compiled in Ada 95 mode,
6488 uses of the new Ada 2005 features will cause error
6489 messages or warnings.
6491 This switch also can be used to cancel the effect of a previous
6492 @option{-gnat83} or @option{-gnat05} switch earlier in the command line.
6494 @item -gnat05 (Ada 2005 mode)
6495 @cindex @option{-gnat05} (@command{gcc})
6496 @cindex Ada 2005 mode
6499 This switch directs the compiler to implement the Ada 2005 version of the
6501 Since Ada 2005 is almost completely upwards
6502 compatible with Ada 95 (and thus also with Ada 83), Ada 83 and Ada 95 programs
6503 may generally be compiled using this switch (see the description of the
6504 @option{-gnat83} and @option{-gnat95} switches for further
6507 For information about the approved ``Ada Issues'' that have been incorporated
6508 into Ada 2005, see @url{http://www.ada-auth.org/cgi-bin/cvsweb.cgi/AIs}.
6509 Included with GNAT releases is a file @file{features-ada0y} that describes
6510 the set of implemented Ada 2005 features.
6514 @node Character Set Control
6515 @subsection Character Set Control
6517 @item ^-gnati^/IDENTIFIER_CHARACTER_SET=^@var{c}
6518 @cindex @option{^-gnati^/IDENTIFIER_CHARACTER_SET^} (@command{gcc})
6521 Normally GNAT recognizes the Latin-1 character set in source program
6522 identifiers, as described in the Ada Reference Manual.
6524 GNAT to recognize alternate character sets in identifiers. @var{c} is a
6525 single character ^^or word^ indicating the character set, as follows:
6529 ISO 8859-1 (Latin-1) identifiers
6532 ISO 8859-2 (Latin-2) letters allowed in identifiers
6535 ISO 8859-3 (Latin-3) letters allowed in identifiers
6538 ISO 8859-4 (Latin-4) letters allowed in identifiers
6541 ISO 8859-5 (Cyrillic) letters allowed in identifiers
6544 ISO 8859-15 (Latin-9) letters allowed in identifiers
6547 IBM PC letters (code page 437) allowed in identifiers
6550 IBM PC letters (code page 850) allowed in identifiers
6552 @item ^f^FULL_UPPER^
6553 Full upper-half codes allowed in identifiers
6556 No upper-half codes allowed in identifiers
6559 Wide-character codes (that is, codes greater than 255)
6560 allowed in identifiers
6563 @xref{Foreign Language Representation}, for full details on the
6564 implementation of these character sets.
6566 @item ^-gnatW^/WIDE_CHARACTER_ENCODING=^@var{e}
6567 @cindex @option{^-gnatW^/WIDE_CHARACTER_ENCODING^} (@command{gcc})
6568 Specify the method of encoding for wide characters.
6569 @var{e} is one of the following:
6574 Hex encoding (brackets coding also recognized)
6577 Upper half encoding (brackets encoding also recognized)
6580 Shift/JIS encoding (brackets encoding also recognized)
6583 EUC encoding (brackets encoding also recognized)
6586 UTF-8 encoding (brackets encoding also recognized)
6589 Brackets encoding only (default value)
6591 For full details on these encoding
6592 methods see @ref{Wide Character Encodings}.
6593 Note that brackets coding is always accepted, even if one of the other
6594 options is specified, so for example @option{-gnatW8} specifies that both
6595 brackets and UTF-8 encodings will be recognized. The units that are
6596 with'ed directly or indirectly will be scanned using the specified
6597 representation scheme, and so if one of the non-brackets scheme is
6598 used, it must be used consistently throughout the program. However,
6599 since brackets encoding is always recognized, it may be conveniently
6600 used in standard libraries, allowing these libraries to be used with
6601 any of the available coding schemes.
6604 If no @option{-gnatW?} parameter is present, then the default
6605 representation is normally Brackets encoding only. However, if the
6606 first three characters of the file are 16#EF# 16#BB# 16#BF# (the standard
6607 byte order mark or BOM for UTF-8), then these three characters are
6608 skipped and the default representation for the file is set to UTF-8.
6610 Note that the wide character representation that is specified (explicitly
6611 or by default) for the main program also acts as the default encoding used
6612 for Wide_Text_IO files if not specifically overridden by a WCEM form
6616 @node File Naming Control
6617 @subsection File Naming Control
6620 @item ^-gnatk^/FILE_NAME_MAX_LENGTH=^@var{n}
6621 @cindex @option{-gnatk} (@command{gcc})
6622 Activates file name ``krunching''. @var{n}, a decimal integer in the range
6623 1-999, indicates the maximum allowable length of a file name (not
6624 including the @file{.ads} or @file{.adb} extension). The default is not
6625 to enable file name krunching.
6627 For the source file naming rules, @xref{File Naming Rules}.
6630 @node Subprogram Inlining Control
6631 @subsection Subprogram Inlining Control
6636 @cindex @option{-gnatn} (@command{gcc})
6638 The @code{n} here is intended to suggest the first syllable of the
6641 GNAT recognizes and processes @code{Inline} pragmas. However, for the
6642 inlining to actually occur, optimization must be enabled. To enable
6643 inlining of subprograms specified by pragma @code{Inline},
6644 you must also specify this switch.
6645 In the absence of this switch, GNAT does not attempt
6646 inlining and does not need to access the bodies of
6647 subprograms for which @code{pragma Inline} is specified if they are not
6648 in the current unit.
6650 If you specify this switch the compiler will access these bodies,
6651 creating an extra source dependency for the resulting object file, and
6652 where possible, the call will be inlined.
6653 For further details on when inlining is possible
6654 see @ref{Inlining of Subprograms}.
6657 @cindex @option{-gnatN} (@command{gcc})
6658 The front end inlining activated by this switch is generally more extensive,
6659 and quite often more effective than the standard @option{-gnatn} inlining mode.
6660 It will also generate additional dependencies.
6662 @option{-gnatN} automatically implies @option{-gnatn} so it is not necessary
6663 to specify both options.
6666 @node Auxiliary Output Control
6667 @subsection Auxiliary Output Control
6671 @cindex @option{-gnatt} (@command{gcc})
6672 @cindex Writing internal trees
6673 @cindex Internal trees, writing to file
6674 Causes GNAT to write the internal tree for a unit to a file (with the
6675 extension @file{.adt}.
6676 This not normally required, but is used by separate analysis tools.
6678 these tools do the necessary compilations automatically, so you should
6679 not have to specify this switch in normal operation.
6682 @cindex @option{-gnatu} (@command{gcc})
6683 Print a list of units required by this compilation on @file{stdout}.
6684 The listing includes all units on which the unit being compiled depends
6685 either directly or indirectly.
6688 @item -pass-exit-codes
6689 @cindex @option{-pass-exit-codes} (@command{gcc})
6690 If this switch is not used, the exit code returned by @command{gcc} when
6691 compiling multiple files indicates whether all source files have
6692 been successfully used to generate object files or not.
6694 When @option{-pass-exit-codes} is used, @command{gcc} exits with an extended
6695 exit status and allows an integrated development environment to better
6696 react to a compilation failure. Those exit status are:
6700 There was an error in at least one source file.
6702 At least one source file did not generate an object file.
6704 The compiler died unexpectedly (internal error for example).
6706 An object file has been generated for every source file.
6711 @node Debugging Control
6712 @subsection Debugging Control
6716 @cindex Debugging options
6719 @cindex @option{-gnatd} (@command{gcc})
6720 Activate internal debugging switches. @var{x} is a letter or digit, or
6721 string of letters or digits, which specifies the type of debugging
6722 outputs desired. Normally these are used only for internal development
6723 or system debugging purposes. You can find full documentation for these
6724 switches in the body of the @code{Debug} unit in the compiler source
6725 file @file{debug.adb}.
6729 @cindex @option{-gnatG} (@command{gcc})
6730 This switch causes the compiler to generate auxiliary output containing
6731 a pseudo-source listing of the generated expanded code. Like most Ada
6732 compilers, GNAT works by first transforming the high level Ada code into
6733 lower level constructs. For example, tasking operations are transformed
6734 into calls to the tasking run-time routines. A unique capability of GNAT
6735 is to list this expanded code in a form very close to normal Ada source.
6736 This is very useful in understanding the implications of various Ada
6737 usage on the efficiency of the generated code. There are many cases in
6738 Ada (e.g.@: the use of controlled types), where simple Ada statements can
6739 generate a lot of run-time code. By using @option{-gnatG} you can identify
6740 these cases, and consider whether it may be desirable to modify the coding
6741 approach to improve efficiency.
6743 The format of the output is very similar to standard Ada source, and is
6744 easily understood by an Ada programmer. The following special syntactic
6745 additions correspond to low level features used in the generated code that
6746 do not have any exact analogies in pure Ada source form. The following
6747 is a partial list of these special constructions. See the spec
6748 of package @code{Sprint} in file @file{sprint.ads} for a full list.
6750 If the switch @option{-gnatL} is used in conjunction with
6751 @cindex @option{-gnatL} (@command{gcc})
6752 @option{-gnatG}, then the original source lines are interspersed
6753 in the expanded source (as comment lines with the original line number).
6756 @item new @var{xxx} @r{[}storage_pool = @var{yyy}@r{]}
6757 Shows the storage pool being used for an allocator.
6759 @item at end @var{procedure-name};
6760 Shows the finalization (cleanup) procedure for a scope.
6762 @item (if @var{expr} then @var{expr} else @var{expr})
6763 Conditional expression equivalent to the @code{x?y:z} construction in C.
6765 @item @var{target}^^^(@var{source})
6766 A conversion with floating-point truncation instead of rounding.
6768 @item @var{target}?(@var{source})
6769 A conversion that bypasses normal Ada semantic checking. In particular
6770 enumeration types and fixed-point types are treated simply as integers.
6772 @item @var{target}?^^^(@var{source})
6773 Combines the above two cases.
6775 @item @var{x} #/ @var{y}
6776 @itemx @var{x} #mod @var{y}
6777 @itemx @var{x} #* @var{y}
6778 @itemx @var{x} #rem @var{y}
6779 A division or multiplication of fixed-point values which are treated as
6780 integers without any kind of scaling.
6782 @item free @var{expr} @r{[}storage_pool = @var{xxx}@r{]}
6783 Shows the storage pool associated with a @code{free} statement.
6785 @item [subtype or type declaration]
6786 Used to list an equivalent declaration for an internally generated
6787 type that is referenced elsewhere in the listing.
6789 @item freeze @var{type-name} @ovar{actions}
6790 Shows the point at which @var{type-name} is frozen, with possible
6791 associated actions to be performed at the freeze point.
6793 @item reference @var{itype}
6794 Reference (and hence definition) to internal type @var{itype}.
6796 @item @var{function-name}! (@var{arg}, @var{arg}, @var{arg})
6797 Intrinsic function call.
6799 @item @var{label-name} : label
6800 Declaration of label @var{labelname}.
6802 @item #$ @var{subprogram-name}
6803 An implicit call to a run-time support routine
6804 (to meet the requirement of H.3.1(9) in a
6807 @item @var{expr} && @var{expr} && @var{expr} @dots{} && @var{expr}
6808 A multiple concatenation (same effect as @var{expr} & @var{expr} &
6809 @var{expr}, but handled more efficiently).
6811 @item [constraint_error]
6812 Raise the @code{Constraint_Error} exception.
6814 @item @var{expression}'reference
6815 A pointer to the result of evaluating @var{expression}.
6817 @item @var{target-type}!(@var{source-expression})
6818 An unchecked conversion of @var{source-expression} to @var{target-type}.
6820 @item [@var{numerator}/@var{denominator}]
6821 Used to represent internal real literals (that) have no exact
6822 representation in base 2-16 (for example, the result of compile time
6823 evaluation of the expression 1.0/27.0).
6827 @cindex @option{-gnatD} (@command{gcc})
6828 When used in conjunction with @option{-gnatG}, this switch causes
6829 the expanded source, as described above for
6830 @option{-gnatG} to be written to files with names
6831 @file{^xxx.dg^XXX_DG^}, where @file{xxx} is the normal file name,
6832 instead of to the standard output file. For
6833 example, if the source file name is @file{hello.adb}, then a file
6834 @file{^hello.adb.dg^HELLO.ADB_DG^} will be written. The debugging
6835 information generated by the @command{gcc} @option{^-g^/DEBUG^} switch
6836 will refer to the generated @file{^xxx.dg^XXX_DG^} file. This allows
6837 you to do source level debugging using the generated code which is
6838 sometimes useful for complex code, for example to find out exactly
6839 which part of a complex construction raised an exception. This switch
6840 also suppress generation of cross-reference information (see
6841 @option{-gnatx}) since otherwise the cross-reference information
6842 would refer to the @file{^.dg^.DG^} file, which would cause
6843 confusion since this is not the original source file.
6845 Note that @option{-gnatD} actually implies @option{-gnatG}
6846 automatically, so it is not necessary to give both options.
6847 In other words @option{-gnatD} is equivalent to @option{-gnatDG}).
6849 If the switch @option{-gnatL} is used in conjunction with
6850 @cindex @option{-gnatL} (@command{gcc})
6851 @option{-gnatDG}, then the original source lines are interspersed
6852 in the expanded source (as comment lines with the original line number).
6855 @cindex @option{-gnatr} (@command{gcc})
6856 @cindex pragma Restrictions
6857 This switch causes pragma Restrictions to be treated as Restriction_Warnings
6858 so that violation of restrictions causes warnings rather than illegalities.
6859 This is useful during the development process when new restrictions are added
6860 or investigated. The switch also causes pragma Profile to be treated as
6861 Profile_Warnings, and pragma Restricted_Run_Time and pragma Ravenscar set
6862 restriction warnings rather than restrictions.
6865 @item -gnatR@r{[}0@r{|}1@r{|}2@r{|}3@r{[}s@r{]]}
6866 @cindex @option{-gnatR} (@command{gcc})
6867 This switch controls output from the compiler of a listing showing
6868 representation information for declared types and objects. For
6869 @option{-gnatR0}, no information is output (equivalent to omitting
6870 the @option{-gnatR} switch). For @option{-gnatR1} (which is the default,
6871 so @option{-gnatR} with no parameter has the same effect), size and alignment
6872 information is listed for declared array and record types. For
6873 @option{-gnatR2}, size and alignment information is listed for all
6874 declared types and objects. Finally @option{-gnatR3} includes symbolic
6875 expressions for values that are computed at run time for
6876 variant records. These symbolic expressions have a mostly obvious
6877 format with #n being used to represent the value of the n'th
6878 discriminant. See source files @file{repinfo.ads/adb} in the
6879 @code{GNAT} sources for full details on the format of @option{-gnatR3}
6880 output. If the switch is followed by an s (e.g.@: @option{-gnatR2s}), then
6881 the output is to a file with the name @file{^file.rep^file_REP^} where
6882 file is the name of the corresponding source file.
6885 @item /REPRESENTATION_INFO
6886 @cindex @option{/REPRESENTATION_INFO} (@command{gcc})
6887 This qualifier controls output from the compiler of a listing showing
6888 representation information for declared types and objects. For
6889 @option{/REPRESENTATION_INFO=NONE}, no information is output
6890 (equivalent to omitting the @option{/REPRESENTATION_INFO} qualifier).
6891 @option{/REPRESENTATION_INFO} without option is equivalent to
6892 @option{/REPRESENTATION_INFO=ARRAYS}.
6893 For @option{/REPRESENTATION_INFO=ARRAYS}, size and alignment
6894 information is listed for declared array and record types. For
6895 @option{/REPRESENTATION_INFO=OBJECTS}, size and alignment information
6896 is listed for all expression information for values that are computed
6897 at run time for variant records. These symbolic expressions have a mostly
6898 obvious format with #n being used to represent the value of the n'th
6899 discriminant. See source files @file{REPINFO.ADS/ADB} in the
6900 @code{GNAT} sources for full details on the format of
6901 @option{/REPRESENTATION_INFO=SYMBOLIC} output.
6902 If _FILE is added at the end of an option
6903 (e.g.@: @option{/REPRESENTATION_INFO=ARRAYS_FILE}),
6904 then the output is to a file with the name @file{file_REP} where
6905 file is the name of the corresponding source file.
6907 Note that it is possible for record components to have zero size. In
6908 this case, the component clause uses an obvious extension of permitted
6909 Ada syntax, for example @code{at 0 range 0 .. -1}.
6911 Representation information requires that code be generated (since it is the
6912 code generator that lays out complex data structures). If an attempt is made
6913 to output representation information when no code is generated, for example
6914 when a subunit is compiled on its own, then no information can be generated
6915 and the compiler outputs a message to this effect.
6918 @cindex @option{-gnatS} (@command{gcc})
6919 The use of the switch @option{-gnatS} for an
6920 Ada compilation will cause the compiler to output a
6921 representation of package Standard in a form very
6922 close to standard Ada. It is not quite possible to
6923 do this entirely in standard Ada (since new
6924 numeric base types cannot be created in standard
6925 Ada), but the output is easily
6926 readable to any Ada programmer, and is useful to
6927 determine the characteristics of target dependent
6928 types in package Standard.
6931 @cindex @option{-gnatx} (@command{gcc})
6932 Normally the compiler generates full cross-referencing information in
6933 the @file{ALI} file. This information is used by a number of tools,
6934 including @code{gnatfind} and @code{gnatxref}. The @option{-gnatx} switch
6935 suppresses this information. This saves some space and may slightly
6936 speed up compilation, but means that these tools cannot be used.
6939 @node Exception Handling Control
6940 @subsection Exception Handling Control
6943 GNAT uses two methods for handling exceptions at run-time. The
6944 @code{setjmp/longjmp} method saves the context when entering
6945 a frame with an exception handler. Then when an exception is
6946 raised, the context can be restored immediately, without the
6947 need for tracing stack frames. This method provides very fast
6948 exception propagation, but introduces significant overhead for
6949 the use of exception handlers, even if no exception is raised.
6951 The other approach is called ``zero cost'' exception handling.
6952 With this method, the compiler builds static tables to describe
6953 the exception ranges. No dynamic code is required when entering
6954 a frame containing an exception handler. When an exception is
6955 raised, the tables are used to control a back trace of the
6956 subprogram invocation stack to locate the required exception
6957 handler. This method has considerably poorer performance for
6958 the propagation of exceptions, but there is no overhead for
6959 exception handlers if no exception is raised. Note that in this
6960 mode and in the context of mixed Ada and C/C++ programming,
6961 to propagate an exception through a C/C++ code, the C/C++ code
6962 must be compiled with the @option{-funwind-tables} GCC's
6965 The following switches may be used to control which of the
6966 two exception handling methods is used.
6972 @cindex @option{--RTS=sjlj} (@command{gnatmake})
6973 This switch causes the setjmp/longjmp run-time (when available) to be used
6974 for exception handling. If the default
6975 mechanism for the target is zero cost exceptions, then
6976 this switch can be used to modify this default, and must be
6977 used for all units in the partition.
6978 This option is rarely used. One case in which it may be
6979 advantageous is if you have an application where exception
6980 raising is common and the overall performance of the
6981 application is improved by favoring exception propagation.
6984 @cindex @option{--RTS=zcx} (@command{gnatmake})
6985 @cindex Zero Cost Exceptions
6986 This switch causes the zero cost approach to be used
6987 for exception handling. If this is the default mechanism for the
6988 target (see below), then this switch is unneeded. If the default
6989 mechanism for the target is setjmp/longjmp exceptions, then
6990 this switch can be used to modify this default, and must be
6991 used for all units in the partition.
6992 This option can only be used if the zero cost approach
6993 is available for the target in use, otherwise it will generate an error.
6997 The same option @option{--RTS} must be used both for @command{gcc}
6998 and @command{gnatbind}. Passing this option to @command{gnatmake}
6999 (@pxref{Switches for gnatmake}) will ensure the required consistency
7000 through the compilation and binding steps.
7002 @node Units to Sources Mapping Files
7003 @subsection Units to Sources Mapping Files
7007 @item -gnatem^^=^@var{path}
7008 @cindex @option{-gnatem} (@command{gcc})
7009 A mapping file is a way to communicate to the compiler two mappings:
7010 from unit names to file names (without any directory information) and from
7011 file names to path names (with full directory information). These mappings
7012 are used by the compiler to short-circuit the path search.
7014 The use of mapping files is not required for correct operation of the
7015 compiler, but mapping files can improve efficiency, particularly when
7016 sources are read over a slow network connection. In normal operation,
7017 you need not be concerned with the format or use of mapping files,
7018 and the @option{-gnatem} switch is not a switch that you would use
7019 explicitly. it is intended only for use by automatic tools such as
7020 @command{gnatmake} running under the project file facility. The
7021 description here of the format of mapping files is provided
7022 for completeness and for possible use by other tools.
7024 A mapping file is a sequence of sets of three lines. In each set,
7025 the first line is the unit name, in lower case, with ``@code{%s}''
7027 specs and ``@code{%b}'' appended for bodies; the second line is the
7028 file name; and the third line is the path name.
7034 /gnat/project1/sources/main.2.ada
7037 When the switch @option{-gnatem} is specified, the compiler will create
7038 in memory the two mappings from the specified file. If there is any problem
7039 (nonexistent file, truncated file or duplicate entries), no mapping will
7042 Several @option{-gnatem} switches may be specified; however, only the last
7043 one on the command line will be taken into account.
7045 When using a project file, @command{gnatmake} create a temporary mapping file
7046 and communicates it to the compiler using this switch.
7050 @node Integrated Preprocessing
7051 @subsection Integrated Preprocessing
7054 GNAT sources may be preprocessed immediately before compilation.
7055 In this case, the actual
7056 text of the source is not the text of the source file, but is derived from it
7057 through a process called preprocessing. Integrated preprocessing is specified
7058 through switches @option{-gnatep} and/or @option{-gnateD}. @option{-gnatep}
7059 indicates, through a text file, the preprocessing data to be used.
7060 @option{-gnateD} specifies or modifies the values of preprocessing symbol.
7063 Note that when integrated preprocessing is used, the output from the
7064 preprocessor is not written to any external file. Instead it is passed
7065 internally to the compiler. If you need to preserve the result of
7066 preprocessing in a file, then you should use @command{gnatprep}
7067 to perform the desired preprocessing in stand-alone mode.
7070 It is recommended that @command{gnatmake} switch ^-s^/SWITCH_CHECK^ should be
7071 used when Integrated Preprocessing is used. The reason is that preprocessing
7072 with another Preprocessing Data file without changing the sources will
7073 not trigger recompilation without this switch.
7076 Note that @command{gnatmake} switch ^-m^/MINIMAL_RECOMPILATION^ will almost
7077 always trigger recompilation for sources that are preprocessed,
7078 because @command{gnatmake} cannot compute the checksum of the source after
7082 The actual preprocessing function is described in details in section
7083 @ref{Preprocessing Using gnatprep}. This section only describes how integrated
7084 preprocessing is triggered and parameterized.
7088 @item -gnatep=@var{file}
7089 @cindex @option{-gnatep} (@command{gcc})
7090 This switch indicates to the compiler the file name (without directory
7091 information) of the preprocessor data file to use. The preprocessor data file
7092 should be found in the source directories.
7095 A preprocessing data file is a text file with significant lines indicating
7096 how should be preprocessed either a specific source or all sources not
7097 mentioned in other lines. A significant line is a nonempty, non-comment line.
7098 Comments are similar to Ada comments.
7101 Each significant line starts with either a literal string or the character '*'.
7102 A literal string is the file name (without directory information) of the source
7103 to preprocess. A character '*' indicates the preprocessing for all the sources
7104 that are not specified explicitly on other lines (order of the lines is not
7105 significant). It is an error to have two lines with the same file name or two
7106 lines starting with the character '*'.
7109 After the file name or the character '*', another optional literal string
7110 indicating the file name of the definition file to be used for preprocessing
7111 (@pxref{Form of Definitions File}). The definition files are found by the
7112 compiler in one of the source directories. In some cases, when compiling
7113 a source in a directory other than the current directory, if the definition
7114 file is in the current directory, it may be necessary to add the current
7115 directory as a source directory through switch ^-I.^/SEARCH=[]^, otherwise
7116 the compiler would not find the definition file.
7119 Then, optionally, ^switches^switches^ similar to those of @code{gnatprep} may
7120 be found. Those ^switches^switches^ are:
7125 Causes both preprocessor lines and the lines deleted by
7126 preprocessing to be replaced by blank lines, preserving the line number.
7127 This ^switch^switch^ is always implied; however, if specified after @option{-c}
7128 it cancels the effect of @option{-c}.
7131 Causes both preprocessor lines and the lines deleted
7132 by preprocessing to be retained as comments marked
7133 with the special string ``@code{--! }''.
7135 @item -Dsymbol=value
7136 Define or redefine a symbol, associated with value. A symbol is an Ada
7137 identifier, or an Ada reserved word, with the exception of @code{if},
7138 @code{else}, @code{elsif}, @code{end}, @code{and}, @code{or} and @code{then}.
7139 @code{value} is either a literal string, an Ada identifier or any Ada reserved
7140 word. A symbol declared with this ^switch^switch^ replaces a symbol with the
7141 same name defined in a definition file.
7144 Causes a sorted list of symbol names and values to be
7145 listed on the standard output file.
7148 Causes undefined symbols to be treated as having the value @code{FALSE}
7150 of a preprocessor test. In the absence of this option, an undefined symbol in
7151 a @code{#if} or @code{#elsif} test will be treated as an error.
7156 Examples of valid lines in a preprocessor data file:
7159 "toto.adb" "prep.def" -u
7160 -- preprocess "toto.adb", using definition file "prep.def",
7161 -- undefined symbol are False.
7164 -- preprocess all other sources without a definition file;
7165 -- suppressed lined are commented; symbol VERSION has the value V101.
7167 "titi.adb" "prep2.def" -s
7168 -- preprocess "titi.adb", using definition file "prep2.def";
7169 -- list all symbols with their values.
7172 @item ^-gnateD^/DATA_PREPROCESSING=^symbol@r{[}=value@r{]}
7173 @cindex @option{-gnateD} (@command{gcc})
7174 Define or redefine a preprocessing symbol, associated with value. If no value
7175 is given on the command line, then the value of the symbol is @code{True}.
7176 A symbol is an identifier, following normal Ada (case-insensitive)
7177 rules for its syntax, and value is any sequence (including an empty sequence)
7178 of characters from the set (letters, digits, period, underline).
7179 Ada reserved words may be used as symbols, with the exceptions of @code{if},
7180 @code{else}, @code{elsif}, @code{end}, @code{and}, @code{or} and @code{then}.
7183 A symbol declared with this ^switch^switch^ on the command line replaces a
7184 symbol with the same name either in a definition file or specified with a
7185 ^switch^switch^ -D in the preprocessor data file.
7188 This switch is similar to switch @option{^-D^/ASSOCIATE^} of @code{gnatprep}.
7192 @node Code Generation Control
7193 @subsection Code Generation Control
7197 The GCC technology provides a wide range of target dependent
7198 @option{-m} switches for controlling
7199 details of code generation with respect to different versions of
7200 architectures. This includes variations in instruction sets (e.g.@:
7201 different members of the power pc family), and different requirements
7202 for optimal arrangement of instructions (e.g.@: different members of
7203 the x86 family). The list of available @option{-m} switches may be
7204 found in the GCC documentation.
7206 Use of these @option{-m} switches may in some cases result in improved
7209 The GNAT Pro technology is tested and qualified without any
7210 @option{-m} switches,
7211 so generally the most reliable approach is to avoid the use of these
7212 switches. However, we generally expect most of these switches to work
7213 successfully with GNAT Pro, and many customers have reported successful
7214 use of these options.
7216 Our general advice is to avoid the use of @option{-m} switches unless
7217 special needs lead to requirements in this area. In particular,
7218 there is no point in using @option{-m} switches to improve performance
7219 unless you actually see a performance improvement.
7223 @subsection Return Codes
7224 @cindex Return Codes
7225 @cindex @option{/RETURN_CODES=VMS}
7228 On VMS, GNAT compiled programs return POSIX-style codes by default,
7229 e.g.@: @option{/RETURN_CODES=POSIX}.
7231 To enable VMS style return codes, use GNAT BIND and LINK with the option
7232 @option{/RETURN_CODES=VMS}. For example:
7235 GNAT BIND MYMAIN.ALI /RETURN_CODES=VMS
7236 GNAT LINK MYMAIN.ALI /RETURN_CODES=VMS
7240 Programs built with /RETURN_CODES=VMS are suitable to be called in
7241 VMS DCL scripts. Programs compiled with the default /RETURN_CODES=POSIX
7242 are suitable for spawning with appropriate GNAT RTL routines.
7246 @node Search Paths and the Run-Time Library (RTL)
7247 @section Search Paths and the Run-Time Library (RTL)
7250 With the GNAT source-based library system, the compiler must be able to
7251 find source files for units that are needed by the unit being compiled.
7252 Search paths are used to guide this process.
7254 The compiler compiles one source file whose name must be given
7255 explicitly on the command line. In other words, no searching is done
7256 for this file. To find all other source files that are needed (the most
7257 common being the specs of units), the compiler examines the following
7258 directories, in the following order:
7262 The directory containing the source file of the main unit being compiled
7263 (the file name on the command line).
7266 Each directory named by an @option{^-I^/SOURCE_SEARCH^} switch given on the
7267 @command{gcc} command line, in the order given.
7270 @findex ADA_PRJ_INCLUDE_FILE
7271 Each of the directories listed in the text file whose name is given
7272 by the @env{ADA_PRJ_INCLUDE_FILE} ^environment variable^logical name^.
7275 @env{ADA_PRJ_INCLUDE_FILE} is normally set by gnatmake or by the ^gnat^GNAT^
7276 driver when project files are used. It should not normally be set
7280 @findex ADA_INCLUDE_PATH
7281 Each of the directories listed in the value of the
7282 @env{ADA_INCLUDE_PATH} ^environment variable^logical name^.
7284 Construct this value
7285 exactly as the @env{PATH} environment variable: a list of directory
7286 names separated by colons (semicolons when working with the NT version).
7289 Normally, define this value as a logical name containing a comma separated
7290 list of directory names.
7292 This variable can also be defined by means of an environment string
7293 (an argument to the HP C exec* set of functions).
7297 DEFINE ANOTHER_PATH FOO:[BAG]
7298 DEFINE ADA_INCLUDE_PATH ANOTHER_PATH,FOO:[BAM],FOO:[BAR]
7301 By default, the path includes GNU:[LIB.OPENVMS7_x.2_8_x.DECLIB]
7302 first, followed by the standard Ada
7303 libraries in GNU:[LIB.OPENVMS7_x.2_8_x.ADAINCLUDE].
7304 If this is not redefined, the user will obtain the HP Ada 83 IO packages
7305 (Text_IO, Sequential_IO, etc)
7306 instead of the standard Ada packages. Thus, in order to get the standard Ada
7307 packages by default, ADA_INCLUDE_PATH must be redefined.
7311 The content of the @file{ada_source_path} file which is part of the GNAT
7312 installation tree and is used to store standard libraries such as the
7313 GNAT Run Time Library (RTL) source files.
7315 @ref{Installing a library}
7320 Specifying the switch @option{^-I-^/NOCURRENT_DIRECTORY^}
7321 inhibits the use of the directory
7322 containing the source file named in the command line. You can still
7323 have this directory on your search path, but in this case it must be
7324 explicitly requested with a @option{^-I^/SOURCE_SEARCH^} switch.
7326 Specifying the switch @option{-nostdinc}
7327 inhibits the search of the default location for the GNAT Run Time
7328 Library (RTL) source files.
7330 The compiler outputs its object files and ALI files in the current
7333 Caution: The object file can be redirected with the @option{-o} switch;
7334 however, @command{gcc} and @code{gnat1} have not been coordinated on this
7335 so the @file{ALI} file will not go to the right place. Therefore, you should
7336 avoid using the @option{-o} switch.
7340 The packages @code{Ada}, @code{System}, and @code{Interfaces} and their
7341 children make up the GNAT RTL, together with the simple @code{System.IO}
7342 package used in the @code{"Hello World"} example. The sources for these units
7343 are needed by the compiler and are kept together in one directory. Not
7344 all of the bodies are needed, but all of the sources are kept together
7345 anyway. In a normal installation, you need not specify these directory
7346 names when compiling or binding. Either the environment variables or
7347 the built-in defaults cause these files to be found.
7349 In addition to the language-defined hierarchies (@code{System}, @code{Ada} and
7350 @code{Interfaces}), the GNAT distribution provides a fourth hierarchy,
7351 consisting of child units of @code{GNAT}. This is a collection of generally
7352 useful types, subprograms, etc. @xref{Top, GNAT Reference Manual, About
7353 This Guid, gnat_rm, GNAT Reference Manual}, for further details.
7355 Besides simplifying access to the RTL, a major use of search paths is
7356 in compiling sources from multiple directories. This can make
7357 development environments much more flexible.
7359 @node Order of Compilation Issues
7360 @section Order of Compilation Issues
7363 If, in our earlier example, there was a spec for the @code{hello}
7364 procedure, it would be contained in the file @file{hello.ads}; yet this
7365 file would not have to be explicitly compiled. This is the result of the
7366 model we chose to implement library management. Some of the consequences
7367 of this model are as follows:
7371 There is no point in compiling specs (except for package
7372 specs with no bodies) because these are compiled as needed by clients. If
7373 you attempt a useless compilation, you will receive an error message.
7374 It is also useless to compile subunits because they are compiled as needed
7378 There are no order of compilation requirements: performing a
7379 compilation never obsoletes anything. The only way you can obsolete
7380 something and require recompilations is to modify one of the
7381 source files on which it depends.
7384 There is no library as such, apart from the ALI files
7385 (@pxref{The Ada Library Information Files}, for information on the format
7386 of these files). For now we find it convenient to create separate ALI files,
7387 but eventually the information therein may be incorporated into the object
7391 When you compile a unit, the source files for the specs of all units
7392 that it @code{with}'s, all its subunits, and the bodies of any generics it
7393 instantiates must be available (reachable by the search-paths mechanism
7394 described above), or you will receive a fatal error message.
7401 The following are some typical Ada compilation command line examples:
7404 @item $ gcc -c xyz.adb
7405 Compile body in file @file{xyz.adb} with all default options.
7408 @item $ gcc -c -O2 -gnata xyz-def.adb
7411 @item $ GNAT COMPILE /OPTIMIZE=ALL -gnata xyz-def.adb
7414 Compile the child unit package in file @file{xyz-def.adb} with extensive
7415 optimizations, and pragma @code{Assert}/@code{Debug} statements
7418 @item $ gcc -c -gnatc abc-def.adb
7419 Compile the subunit in file @file{abc-def.adb} in semantic-checking-only
7423 @node Binding Using gnatbind
7424 @chapter Binding Using @code{gnatbind}
7428 * Running gnatbind::
7429 * Switches for gnatbind::
7430 * Command-Line Access::
7431 * Search Paths for gnatbind::
7432 * Examples of gnatbind Usage::
7436 This chapter describes the GNAT binder, @code{gnatbind}, which is used
7437 to bind compiled GNAT objects.
7439 Note: to invoke @code{gnatbind} with a project file, use the @code{gnat}
7440 driver (see @ref{The GNAT Driver and Project Files}).
7442 The @code{gnatbind} program performs four separate functions:
7446 Checks that a program is consistent, in accordance with the rules in
7447 Chapter 10 of the Ada Reference Manual. In particular, error
7448 messages are generated if a program uses inconsistent versions of a
7452 Checks that an acceptable order of elaboration exists for the program
7453 and issues an error message if it cannot find an order of elaboration
7454 that satisfies the rules in Chapter 10 of the Ada Language Manual.
7457 Generates a main program incorporating the given elaboration order.
7458 This program is a small Ada package (body and spec) that
7459 must be subsequently compiled
7460 using the GNAT compiler. The necessary compilation step is usually
7461 performed automatically by @command{gnatlink}. The two most important
7462 functions of this program
7463 are to call the elaboration routines of units in an appropriate order
7464 and to call the main program.
7467 Determines the set of object files required by the given main program.
7468 This information is output in the forms of comments in the generated program,
7469 to be read by the @command{gnatlink} utility used to link the Ada application.
7472 @node Running gnatbind
7473 @section Running @code{gnatbind}
7476 The form of the @code{gnatbind} command is
7479 $ gnatbind @ovar{switches} @var{mainprog}@r{[}.ali@r{]} @ovar{switches}
7483 where @file{@var{mainprog}.adb} is the Ada file containing the main program
7484 unit body. If no switches are specified, @code{gnatbind} constructs an Ada
7485 package in two files whose names are
7486 @file{b~@var{mainprog}.ads}, and @file{b~@var{mainprog}.adb}.
7487 For example, if given the
7488 parameter @file{hello.ali}, for a main program contained in file
7489 @file{hello.adb}, the binder output files would be @file{b~hello.ads}
7490 and @file{b~hello.adb}.
7492 When doing consistency checking, the binder takes into consideration
7493 any source files it can locate. For example, if the binder determines
7494 that the given main program requires the package @code{Pack}, whose
7496 file is @file{pack.ali} and whose corresponding source spec file is
7497 @file{pack.ads}, it attempts to locate the source file @file{pack.ads}
7498 (using the same search path conventions as previously described for the
7499 @command{gcc} command). If it can locate this source file, it checks that
7501 or source checksums of the source and its references to in @file{ALI} files
7502 match. In other words, any @file{ALI} files that mentions this spec must have
7503 resulted from compiling this version of the source file (or in the case
7504 where the source checksums match, a version close enough that the
7505 difference does not matter).
7507 @cindex Source files, use by binder
7508 The effect of this consistency checking, which includes source files, is
7509 that the binder ensures that the program is consistent with the latest
7510 version of the source files that can be located at bind time. Editing a
7511 source file without compiling files that depend on the source file cause
7512 error messages to be generated by the binder.
7514 For example, suppose you have a main program @file{hello.adb} and a
7515 package @code{P}, from file @file{p.ads} and you perform the following
7520 Enter @code{gcc -c hello.adb} to compile the main program.
7523 Enter @code{gcc -c p.ads} to compile package @code{P}.
7526 Edit file @file{p.ads}.
7529 Enter @code{gnatbind hello}.
7533 At this point, the file @file{p.ali} contains an out-of-date time stamp
7534 because the file @file{p.ads} has been edited. The attempt at binding
7535 fails, and the binder generates the following error messages:
7538 error: "hello.adb" must be recompiled ("p.ads" has been modified)
7539 error: "p.ads" has been modified and must be recompiled
7543 Now both files must be recompiled as indicated, and then the bind can
7544 succeed, generating a main program. You need not normally be concerned
7545 with the contents of this file, but for reference purposes a sample
7546 binder output file is given in @ref{Example of Binder Output File}.
7548 In most normal usage, the default mode of @command{gnatbind} which is to
7549 generate the main package in Ada, as described in the previous section.
7550 In particular, this means that any Ada programmer can read and understand
7551 the generated main program. It can also be debugged just like any other
7552 Ada code provided the @option{^-g^/DEBUG^} switch is used for
7553 @command{gnatbind} and @command{gnatlink}.
7555 However for some purposes it may be convenient to generate the main
7556 program in C rather than Ada. This may for example be helpful when you
7557 are generating a mixed language program with the main program in C. The
7558 GNAT compiler itself is an example.
7559 The use of the @option{^-C^/BIND_FILE=C^} switch
7560 for both @code{gnatbind} and @command{gnatlink} will cause the program to
7561 be generated in C (and compiled using the gnu C compiler).
7563 @node Switches for gnatbind
7564 @section Switches for @command{gnatbind}
7567 The following switches are available with @code{gnatbind}; details will
7568 be presented in subsequent sections.
7571 * Consistency-Checking Modes::
7572 * Binder Error Message Control::
7573 * Elaboration Control::
7575 * Binding with Non-Ada Main Programs::
7576 * Binding Programs with No Main Subprogram::
7583 @cindex @option{--version} @command{gnatbind}
7584 Display Copyright and version, then exit disregarding all other options.
7587 @cindex @option{--help} @command{gnatbind}
7588 If @option{--version} was not used, display usage, then exit disregarding
7592 @cindex @option{-a} @command{gnatbind}
7593 Indicates that, if supported by the platform, the adainit procedure should
7594 be treated as an initialisation routine by the linker (a constructor). This
7595 is intended to be used by the Project Manager to automatically initialize
7596 shared Stand-Alone Libraries.
7598 @item ^-aO^/OBJECT_SEARCH^
7599 @cindex @option{^-aO^/OBJECT_SEARCH^} (@command{gnatbind})
7600 Specify directory to be searched for ALI files.
7602 @item ^-aI^/SOURCE_SEARCH^
7603 @cindex @option{^-aI^/SOURCE_SEARCH^} (@command{gnatbind})
7604 Specify directory to be searched for source file.
7606 @item ^-A^/BIND_FILE=ADA^
7607 @cindex @option{^-A^/BIND_FILE=ADA^} (@command{gnatbind})
7608 Generate binder program in Ada (default)
7610 @item ^-b^/REPORT_ERRORS=BRIEF^
7611 @cindex @option{^-b^/REPORT_ERRORS=BRIEF^} (@command{gnatbind})
7612 Generate brief messages to @file{stderr} even if verbose mode set.
7614 @item ^-c^/NOOUTPUT^
7615 @cindex @option{^-c^/NOOUTPUT^} (@command{gnatbind})
7616 Check only, no generation of binder output file.
7618 @item ^-C^/BIND_FILE=C^
7619 @cindex @option{^-C^/BIND_FILE=C^} (@command{gnatbind})
7620 Generate binder program in C
7622 @item ^-d^/DEFAULT_STACK_SIZE=^@var{nn}@r{[}k@r{|}m@r{]}
7623 @cindex @option{^-d^/DEFAULT_STACK_SIZE=^@var{nn}@r{[}k@r{|}m@r{]}} (@command{gnatbind})
7624 This switch can be used to change the default task stack size value
7625 to a specified size @var{nn}, which is expressed in bytes by default, or
7626 in kilobytes when suffixed with @var{k} or in megabytes when suffixed
7628 In the absence of a @samp{@r{[}k@r{|}m@r{]}} suffix, this switch is equivalent,
7629 in effect, to completing all task specs with
7630 @smallexample @c ada
7631 pragma Storage_Size (nn);
7633 When they do not already have such a pragma.
7635 @item ^-D^/DEFAULT_SECONDARY_STACK_SIZE=^@var{nn}@r{[}k@r{|}m@r{]}
7636 @cindex @option{^-D^/DEFAULT_SECONDARY_STACK_SIZE=nnnnn^} (@command{gnatbind})
7637 This switch can be used to change the default secondary stack size value
7638 to a specified size @var{nn}, which is expressed in bytes by default, or
7639 in kilobytes when suffixed with @var{k} or in megabytes when suffixed
7642 The secondary stack is used to deal with functions that return a variable
7643 sized result, for example a function returning an unconstrained
7644 String. There are two ways in which this secondary stack is allocated.
7646 For most targets, the secondary stack is growing on demand and is allocated
7647 as a chain of blocks in the heap. The -D option is not very
7648 relevant. It only give some control over the size of the allocated
7649 blocks (whose size is the minimum of the default secondary stack size value,
7650 and the actual size needed for the current allocation request).
7652 For certain targets, notably VxWorks 653,
7653 the secondary stack is allocated by carving off a fixed ratio chunk of the
7654 primary task stack. The -D option is used to define the
7655 size of the environment task's secondary stack.
7657 @item ^-e^/ELABORATION_DEPENDENCIES^
7658 @cindex @option{^-e^/ELABORATION_DEPENDENCIES^} (@command{gnatbind})
7659 Output complete list of elaboration-order dependencies.
7661 @item ^-E^/STORE_TRACEBACKS^
7662 @cindex @option{^-E^/STORE_TRACEBACKS^} (@command{gnatbind})
7663 Store tracebacks in exception occurrences when the target supports it.
7664 This is the default with the zero cost exception mechanism.
7666 @c The following may get moved to an appendix
7667 This option is currently supported on the following targets:
7668 all x86 ports, Solaris, Windows, HP-UX, AIX, PowerPC VxWorks and Alpha VxWorks.
7670 See also the packages @code{GNAT.Traceback} and
7671 @code{GNAT.Traceback.Symbolic} for more information.
7673 Note that on x86 ports, you must not use @option{-fomit-frame-pointer}
7674 @command{gcc} option.
7677 @item ^-F^/FORCE_ELABS_FLAGS^
7678 @cindex @option{^-F^/FORCE_ELABS_FLAGS^} (@command{gnatbind})
7679 Force the checks of elaboration flags. @command{gnatbind} does not normally
7680 generate checks of elaboration flags for the main executable, except when
7681 a Stand-Alone Library is used. However, there are cases when this cannot be
7682 detected by gnatbind. An example is importing an interface of a Stand-Alone
7683 Library through a pragma Import and only specifying through a linker switch
7684 this Stand-Alone Library. This switch is used to guarantee that elaboration
7685 flag checks are generated.
7688 @cindex @option{^-h^/HELP^} (@command{gnatbind})
7689 Output usage (help) information
7692 @cindex @option{^-I^/SEARCH^} (@command{gnatbind})
7693 Specify directory to be searched for source and ALI files.
7695 @item ^-I-^/NOCURRENT_DIRECTORY^
7696 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@command{gnatbind})
7697 Do not look for sources in the current directory where @code{gnatbind} was
7698 invoked, and do not look for ALI files in the directory containing the
7699 ALI file named in the @code{gnatbind} command line.
7701 @item ^-l^/ORDER_OF_ELABORATION^
7702 @cindex @option{^-l^/ORDER_OF_ELABORATION^} (@command{gnatbind})
7703 Output chosen elaboration order.
7705 @item ^-L@var{xxx}^/BUILD_LIBRARY=@var{xxx}^
7706 @cindex @option{^-L^/BUILD_LIBRARY^} (@command{gnatbind})
7707 Bind the units for library building. In this case the adainit and
7708 adafinal procedures (@pxref{Binding with Non-Ada Main Programs})
7709 are renamed to ^@var{xxx}init^@var{XXX}INIT^ and
7710 ^@var{xxx}final^@var{XXX}FINAL^.
7711 Implies ^-n^/NOCOMPILE^.
7713 (@xref{GNAT and Libraries}, for more details.)
7716 On OpenVMS, these init and final procedures are exported in uppercase
7717 letters. For example if /BUILD_LIBRARY=toto is used, the exported name of
7718 the init procedure will be "TOTOINIT" and the exported name of the final
7719 procedure will be "TOTOFINAL".
7722 @item ^-Mxyz^/RENAME_MAIN=xyz^
7723 @cindex @option{^-M^/RENAME_MAIN^} (@command{gnatbind})
7724 Rename generated main program from main to xyz. This option is
7725 supported on cross environments only.
7727 @item ^-m^/ERROR_LIMIT=^@var{n}
7728 @cindex @option{^-m^/ERROR_LIMIT^} (@command{gnatbind})
7729 Limit number of detected errors to @var{n}, where @var{n} is
7730 in the range 1..999_999. The default value if no switch is
7731 given is 9999. Binding is terminated if the limit is exceeded.
7733 Furthermore, under Windows, the sources pointed to by the libraries path
7734 set in the registry are not searched for.
7738 @cindex @option{^-n^/NOMAIN^} (@command{gnatbind})
7742 @cindex @option{-nostdinc} (@command{gnatbind})
7743 Do not look for sources in the system default directory.
7746 @cindex @option{-nostdlib} (@command{gnatbind})
7747 Do not look for library files in the system default directory.
7749 @item --RTS=@var{rts-path}
7750 @cindex @option{--RTS} (@code{gnatbind})
7751 Specifies the default location of the runtime library. Same meaning as the
7752 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
7754 @item ^-o ^/OUTPUT=^@var{file}
7755 @cindex @option{^-o ^/OUTPUT^} (@command{gnatbind})
7756 Name the output file @var{file} (default is @file{b~@var{xxx}.adb}).
7757 Note that if this option is used, then linking must be done manually,
7758 gnatlink cannot be used.
7760 @item ^-O^/OBJECT_LIST^
7761 @cindex @option{^-O^/OBJECT_LIST^} (@command{gnatbind})
7764 @item ^-p^/PESSIMISTIC_ELABORATION^
7765 @cindex @option{^-p^/PESSIMISTIC_ELABORATION^} (@command{gnatbind})
7766 Pessimistic (worst-case) elaboration order
7769 @cindex @option{^-R^-R^} (@command{gnatbind})
7770 Output closure source list.
7772 @item ^-s^/READ_SOURCES=ALL^
7773 @cindex @option{^-s^/READ_SOURCES=ALL^} (@command{gnatbind})
7774 Require all source files to be present.
7776 @item ^-S@var{xxx}^/INITIALIZE_SCALARS=@var{xxx}^
7777 @cindex @option{^-S^/INITIALIZE_SCALARS^} (@command{gnatbind})
7778 Specifies the value to be used when detecting uninitialized scalar
7779 objects with pragma Initialize_Scalars.
7780 The @var{xxx} ^string specified with the switch^option^ may be either
7782 @item ``@option{^in^INVALID^}'' requesting an invalid value where possible
7783 @item ``@option{^lo^LOW^}'' for the lowest possible value
7784 @item ``@option{^hi^HIGH^}'' for the highest possible value
7785 @item ``@option{@var{xx}}'' for a value consisting of repeated bytes with the
7786 value @code{16#@var{xx}#} (i.e., @var{xx} is a string of two hexadecimal digits).
7789 In addition, you can specify @option{-Sev} to indicate that the value is
7790 to be set at run time. In this case, the program will look for an environment
7791 @cindex GNAT_INIT_SCALARS
7792 variable of the form @env{GNAT_INIT_SCALARS=@var{xx}}, where @var{xx} is one
7793 of @option{in/lo/hi/@var{xx}} with the same meanings as above.
7794 If no environment variable is found, or if it does not have a valid value,
7795 then the default is @option{in} (invalid values).
7799 @cindex @option{-static} (@code{gnatbind})
7800 Link against a static GNAT run time.
7803 @cindex @option{-shared} (@code{gnatbind})
7804 Link against a shared GNAT run time when available.
7807 @item ^-t^/NOTIME_STAMP_CHECK^
7808 @cindex @option{^-t^/NOTIME_STAMP_CHECK^} (@code{gnatbind})
7809 Tolerate time stamp and other consistency errors
7811 @item ^-T@var{n}^/TIME_SLICE=@var{n}^
7812 @cindex @option{^-T^/TIME_SLICE^} (@code{gnatbind})
7813 Set the time slice value to @var{n} milliseconds. If the system supports
7814 the specification of a specific time slice value, then the indicated value
7815 is used. If the system does not support specific time slice values, but
7816 does support some general notion of round-robin scheduling, then any
7817 nonzero value will activate round-robin scheduling.
7819 A value of zero is treated specially. It turns off time
7820 slicing, and in addition, indicates to the tasking run time that the
7821 semantics should match as closely as possible the Annex D
7822 requirements of the Ada RM, and in particular sets the default
7823 scheduling policy to @code{FIFO_Within_Priorities}.
7825 @item ^-u@var{n}^/DYNAMIC_STACK_USAGE=@var{n}^
7826 @cindex @option{^-u^/DYNAMIC_STACK_USAGE^} (@code{gnatbind})
7827 Enable dynamic stack usage, with @var{n} results stored and displayed
7828 at program termination. A result is generated when a task
7829 terminates. Results that can't be stored are displayed on the fly, at
7830 task termination. This option is currently not supported on Itanium
7831 platforms. (See @ref{Dynamic Stack Usage Analysis} for details.)
7833 @item ^-v^/REPORT_ERRORS=VERBOSE^
7834 @cindex @option{^-v^/REPORT_ERRORS=VERBOSE^} (@code{gnatbind})
7835 Verbose mode. Write error messages, header, summary output to
7840 @cindex @option{-w} (@code{gnatbind})
7841 Warning mode (@var{x}=s/e for suppress/treat as error)
7845 @item /WARNINGS=NORMAL
7846 @cindex @option{/WARNINGS} (@code{gnatbind})
7847 Normal warnings mode. Warnings are issued but ignored
7849 @item /WARNINGS=SUPPRESS
7850 @cindex @option{/WARNINGS} (@code{gnatbind})
7851 All warning messages are suppressed
7853 @item /WARNINGS=ERROR
7854 @cindex @option{/WARNINGS} (@code{gnatbind})
7855 Warning messages are treated as fatal errors
7858 @item ^-Wx^/WIDE_CHARACTER_ENCODING=^@var{e}
7859 @cindex @option{^-Wx^/WIDE_CHARACTER_ENCODING^} (@code{gnatbind})
7860 Override default wide character encoding for standard Text_IO files.
7862 @item ^-x^/READ_SOURCES=NONE^
7863 @cindex @option{^-x^/READ_SOURCES^} (@code{gnatbind})
7864 Exclude source files (check object consistency only).
7867 @item /READ_SOURCES=AVAILABLE
7868 @cindex @option{/READ_SOURCES} (@code{gnatbind})
7869 Default mode, in which sources are checked for consistency only if
7873 @item ^-y^/ENABLE_LEAP_SECONDS^
7874 @cindex @option{^-y^/ENABLE_LEAP_SECONDS^} (@code{gnatbind})
7875 Enable leap seconds support in @code{Ada.Calendar} and its children.
7877 @item ^-z^/ZERO_MAIN^
7878 @cindex @option{^-z^/ZERO_MAIN^} (@code{gnatbind})
7884 You may obtain this listing of switches by running @code{gnatbind} with
7888 @node Consistency-Checking Modes
7889 @subsection Consistency-Checking Modes
7892 As described earlier, by default @code{gnatbind} checks
7893 that object files are consistent with one another and are consistent
7894 with any source files it can locate. The following switches control binder
7899 @item ^-s^/READ_SOURCES=ALL^
7900 @cindex @option{^-s^/READ_SOURCES=ALL^} (@code{gnatbind})
7901 Require source files to be present. In this mode, the binder must be
7902 able to locate all source files that are referenced, in order to check
7903 their consistency. In normal mode, if a source file cannot be located it
7904 is simply ignored. If you specify this switch, a missing source
7907 @item ^-Wx^/WIDE_CHARACTER_ENCODING=^@var{e}
7908 @cindex @option{^-Wx^/WIDE_CHARACTER_ENCODING^} (@code{gnatbind})
7909 Override default wide character encoding for standard Text_IO files.
7910 Normally the default wide character encoding method used for standard
7911 [Wide_[Wide_]]Text_IO files is taken from the encoding specified for
7912 the main source input (see description of switch
7913 @option{^-gnatWx^/WIDE_CHARACTER_ENCODING^} for the compiler). The
7914 use of this switch for the binder (which has the same set of
7915 possible arguments) overrides this default as specified.
7917 @item ^-x^/READ_SOURCES=NONE^
7918 @cindex @option{^-x^/READ_SOURCES=NONE^} (@code{gnatbind})
7919 Exclude source files. In this mode, the binder only checks that ALI
7920 files are consistent with one another. Source files are not accessed.
7921 The binder runs faster in this mode, and there is still a guarantee that
7922 the resulting program is self-consistent.
7923 If a source file has been edited since it was last compiled, and you
7924 specify this switch, the binder will not detect that the object
7925 file is out of date with respect to the source file. Note that this is the
7926 mode that is automatically used by @command{gnatmake} because in this
7927 case the checking against sources has already been performed by
7928 @command{gnatmake} in the course of compilation (i.e.@: before binding).
7931 @item /READ_SOURCES=AVAILABLE
7932 @cindex @code{/READ_SOURCES=AVAILABLE} (@code{gnatbind})
7933 This is the default mode in which source files are checked if they are
7934 available, and ignored if they are not available.
7938 @node Binder Error Message Control
7939 @subsection Binder Error Message Control
7942 The following switches provide control over the generation of error
7943 messages from the binder:
7947 @item ^-v^/REPORT_ERRORS=VERBOSE^
7948 @cindex @option{^-v^/REPORT_ERRORS=VERBOSE^} (@code{gnatbind})
7949 Verbose mode. In the normal mode, brief error messages are generated to
7950 @file{stderr}. If this switch is present, a header is written
7951 to @file{stdout} and any error messages are directed to @file{stdout}.
7952 All that is written to @file{stderr} is a brief summary message.
7954 @item ^-b^/REPORT_ERRORS=BRIEF^
7955 @cindex @option{^-b^/REPORT_ERRORS=BRIEF^} (@code{gnatbind})
7956 Generate brief error messages to @file{stderr} even if verbose mode is
7957 specified. This is relevant only when used with the
7958 @option{^-v^/REPORT_ERRORS=VERBOSE^} switch.
7962 @cindex @option{-m} (@code{gnatbind})
7963 Limits the number of error messages to @var{n}, a decimal integer in the
7964 range 1-999. The binder terminates immediately if this limit is reached.
7967 @cindex @option{-M} (@code{gnatbind})
7968 Renames the generated main program from @code{main} to @code{xxx}.
7969 This is useful in the case of some cross-building environments, where
7970 the actual main program is separate from the one generated
7974 @item ^-ws^/WARNINGS=SUPPRESS^
7975 @cindex @option{^-ws^/WARNINGS=SUPPRESS^} (@code{gnatbind})
7977 Suppress all warning messages.
7979 @item ^-we^/WARNINGS=ERROR^
7980 @cindex @option{^-we^/WARNINGS=ERROR^} (@code{gnatbind})
7981 Treat any warning messages as fatal errors.
7984 @item /WARNINGS=NORMAL
7985 Standard mode with warnings generated, but warnings do not get treated
7989 @item ^-t^/NOTIME_STAMP_CHECK^
7990 @cindex @option{^-t^/NOTIME_STAMP_CHECK^} (@code{gnatbind})
7991 @cindex Time stamp checks, in binder
7992 @cindex Binder consistency checks
7993 @cindex Consistency checks, in binder
7994 The binder performs a number of consistency checks including:
7998 Check that time stamps of a given source unit are consistent
8000 Check that checksums of a given source unit are consistent
8002 Check that consistent versions of @code{GNAT} were used for compilation
8004 Check consistency of configuration pragmas as required
8008 Normally failure of such checks, in accordance with the consistency
8009 requirements of the Ada Reference Manual, causes error messages to be
8010 generated which abort the binder and prevent the output of a binder
8011 file and subsequent link to obtain an executable.
8013 The @option{^-t^/NOTIME_STAMP_CHECK^} switch converts these error messages
8014 into warnings, so that
8015 binding and linking can continue to completion even in the presence of such
8016 errors. The result may be a failed link (due to missing symbols), or a
8017 non-functional executable which has undefined semantics.
8018 @emph{This means that
8019 @option{^-t^/NOTIME_STAMP_CHECK^} should be used only in unusual situations,
8023 @node Elaboration Control
8024 @subsection Elaboration Control
8027 The following switches provide additional control over the elaboration
8028 order. For full details see @ref{Elaboration Order Handling in GNAT}.
8031 @item ^-p^/PESSIMISTIC_ELABORATION^
8032 @cindex @option{^-p^/PESSIMISTIC_ELABORATION^} (@code{gnatbind})
8033 Normally the binder attempts to choose an elaboration order that is
8034 likely to minimize the likelihood of an elaboration order error resulting
8035 in raising a @code{Program_Error} exception. This switch reverses the
8036 action of the binder, and requests that it deliberately choose an order
8037 that is likely to maximize the likelihood of an elaboration error.
8038 This is useful in ensuring portability and avoiding dependence on
8039 accidental fortuitous elaboration ordering.
8041 Normally it only makes sense to use the @option{^-p^/PESSIMISTIC_ELABORATION^}
8043 elaboration checking is used (@option{-gnatE} switch used for compilation).
8044 This is because in the default static elaboration mode, all necessary
8045 @code{Elaborate} and @code{Elaborate_All} pragmas are implicitly inserted.
8046 These implicit pragmas are still respected by the binder in
8047 @option{^-p^/PESSIMISTIC_ELABORATION^} mode, so a
8048 safe elaboration order is assured.
8051 @node Output Control
8052 @subsection Output Control
8055 The following switches allow additional control over the output
8056 generated by the binder.
8061 @item ^-A^/BIND_FILE=ADA^
8062 @cindex @option{^-A^/BIND_FILE=ADA^} (@code{gnatbind})
8063 Generate binder program in Ada (default). The binder program is named
8064 @file{b~@var{mainprog}.adb} by default. This can be changed with
8065 @option{^-o^/OUTPUT^} @code{gnatbind} option.
8067 @item ^-c^/NOOUTPUT^
8068 @cindex @option{^-c^/NOOUTPUT^} (@code{gnatbind})
8069 Check only. Do not generate the binder output file. In this mode the
8070 binder performs all error checks but does not generate an output file.
8072 @item ^-C^/BIND_FILE=C^
8073 @cindex @option{^-C^/BIND_FILE=C^} (@code{gnatbind})
8074 Generate binder program in C. The binder program is named
8075 @file{b_@var{mainprog}.c}.
8076 This can be changed with @option{^-o^/OUTPUT^} @code{gnatbind}
8079 @item ^-e^/ELABORATION_DEPENDENCIES^
8080 @cindex @option{^-e^/ELABORATION_DEPENDENCIES^} (@code{gnatbind})
8081 Output complete list of elaboration-order dependencies, showing the
8082 reason for each dependency. This output can be rather extensive but may
8083 be useful in diagnosing problems with elaboration order. The output is
8084 written to @file{stdout}.
8087 @cindex @option{^-h^/HELP^} (@code{gnatbind})
8088 Output usage information. The output is written to @file{stdout}.
8090 @item ^-K^/LINKER_OPTION_LIST^
8091 @cindex @option{^-K^/LINKER_OPTION_LIST^} (@code{gnatbind})
8092 Output linker options to @file{stdout}. Includes library search paths,
8093 contents of pragmas Ident and Linker_Options, and libraries added
8096 @item ^-l^/ORDER_OF_ELABORATION^
8097 @cindex @option{^-l^/ORDER_OF_ELABORATION^} (@code{gnatbind})
8098 Output chosen elaboration order. The output is written to @file{stdout}.
8100 @item ^-O^/OBJECT_LIST^
8101 @cindex @option{^-O^/OBJECT_LIST^} (@code{gnatbind})
8102 Output full names of all the object files that must be linked to provide
8103 the Ada component of the program. The output is written to @file{stdout}.
8104 This list includes the files explicitly supplied and referenced by the user
8105 as well as implicitly referenced run-time unit files. The latter are
8106 omitted if the corresponding units reside in shared libraries. The
8107 directory names for the run-time units depend on the system configuration.
8109 @item ^-o ^/OUTPUT=^@var{file}
8110 @cindex @option{^-o^/OUTPUT^} (@code{gnatbind})
8111 Set name of output file to @var{file} instead of the normal
8112 @file{b~@var{mainprog}.adb} default. Note that @var{file} denote the Ada
8113 binder generated body filename. In C mode you would normally give
8114 @var{file} an extension of @file{.c} because it will be a C source program.
8115 Note that if this option is used, then linking must be done manually.
8116 It is not possible to use gnatlink in this case, since it cannot locate
8119 @item ^-r^/RESTRICTION_LIST^
8120 @cindex @option{^-r^/RESTRICTION_LIST^} (@code{gnatbind})
8121 Generate list of @code{pragma Restrictions} that could be applied to
8122 the current unit. This is useful for code audit purposes, and also may
8123 be used to improve code generation in some cases.
8127 @node Binding with Non-Ada Main Programs
8128 @subsection Binding with Non-Ada Main Programs
8131 In our description so far we have assumed that the main
8132 program is in Ada, and that the task of the binder is to generate a
8133 corresponding function @code{main} that invokes this Ada main
8134 program. GNAT also supports the building of executable programs where
8135 the main program is not in Ada, but some of the called routines are
8136 written in Ada and compiled using GNAT (@pxref{Mixed Language Programming}).
8137 The following switch is used in this situation:
8141 @cindex @option{^-n^/NOMAIN^} (@code{gnatbind})
8142 No main program. The main program is not in Ada.
8146 In this case, most of the functions of the binder are still required,
8147 but instead of generating a main program, the binder generates a file
8148 containing the following callable routines:
8153 You must call this routine to initialize the Ada part of the program by
8154 calling the necessary elaboration routines. A call to @code{adainit} is
8155 required before the first call to an Ada subprogram.
8157 Note that it is assumed that the basic execution environment must be setup
8158 to be appropriate for Ada execution at the point where the first Ada
8159 subprogram is called. In particular, if the Ada code will do any
8160 floating-point operations, then the FPU must be setup in an appropriate
8161 manner. For the case of the x86, for example, full precision mode is
8162 required. The procedure GNAT.Float_Control.Reset may be used to ensure
8163 that the FPU is in the right state.
8167 You must call this routine to perform any library-level finalization
8168 required by the Ada subprograms. A call to @code{adafinal} is required
8169 after the last call to an Ada subprogram, and before the program
8174 If the @option{^-n^/NOMAIN^} switch
8175 @cindex @option{^-n^/NOMAIN^} (@command{gnatbind})
8176 @cindex Binder, multiple input files
8177 is given, more than one ALI file may appear on
8178 the command line for @code{gnatbind}. The normal @dfn{closure}
8179 calculation is performed for each of the specified units. Calculating
8180 the closure means finding out the set of units involved by tracing
8181 @code{with} references. The reason it is necessary to be able to
8182 specify more than one ALI file is that a given program may invoke two or
8183 more quite separate groups of Ada units.
8185 The binder takes the name of its output file from the last specified ALI
8186 file, unless overridden by the use of the @option{^-o file^/OUTPUT=file^}.
8187 @cindex @option{^-o^/OUTPUT^} (@command{gnatbind})
8188 The output is an Ada unit in source form that can
8189 be compiled with GNAT unless the -C switch is used in which case the
8190 output is a C source file, which must be compiled using the C compiler.
8191 This compilation occurs automatically as part of the @command{gnatlink}
8194 Currently the GNAT run time requires a FPU using 80 bits mode
8195 precision. Under targets where this is not the default it is required to
8196 call GNAT.Float_Control.Reset before using floating point numbers (this
8197 include float computation, float input and output) in the Ada code. A
8198 side effect is that this could be the wrong mode for the foreign code
8199 where floating point computation could be broken after this call.
8201 @node Binding Programs with No Main Subprogram
8202 @subsection Binding Programs with No Main Subprogram
8205 It is possible to have an Ada program which does not have a main
8206 subprogram. This program will call the elaboration routines of all the
8207 packages, then the finalization routines.
8209 The following switch is used to bind programs organized in this manner:
8212 @item ^-z^/ZERO_MAIN^
8213 @cindex @option{^-z^/ZERO_MAIN^} (@code{gnatbind})
8214 Normally the binder checks that the unit name given on the command line
8215 corresponds to a suitable main subprogram. When this switch is used,
8216 a list of ALI files can be given, and the execution of the program
8217 consists of elaboration of these units in an appropriate order. Note
8218 that the default wide character encoding method for standard Text_IO
8219 files is always set to Brackets if this switch is set (you can use
8221 @option{^-Wx^WIDE_CHARACTER_ENCODING^} to override this default).
8224 @node Command-Line Access
8225 @section Command-Line Access
8228 The package @code{Ada.Command_Line} provides access to the command-line
8229 arguments and program name. In order for this interface to operate
8230 correctly, the two variables
8242 are declared in one of the GNAT library routines. These variables must
8243 be set from the actual @code{argc} and @code{argv} values passed to the
8244 main program. With no @option{^n^/NOMAIN^} present, @code{gnatbind}
8245 generates the C main program to automatically set these variables.
8246 If the @option{^n^/NOMAIN^} switch is used, there is no automatic way to
8247 set these variables. If they are not set, the procedures in
8248 @code{Ada.Command_Line} will not be available, and any attempt to use
8249 them will raise @code{Constraint_Error}. If command line access is
8250 required, your main program must set @code{gnat_argc} and
8251 @code{gnat_argv} from the @code{argc} and @code{argv} values passed to
8254 @node Search Paths for gnatbind
8255 @section Search Paths for @code{gnatbind}
8258 The binder takes the name of an ALI file as its argument and needs to
8259 locate source files as well as other ALI files to verify object consistency.
8261 For source files, it follows exactly the same search rules as @command{gcc}
8262 (@pxref{Search Paths and the Run-Time Library (RTL)}). For ALI files the
8263 directories searched are:
8267 The directory containing the ALI file named in the command line, unless
8268 the switch @option{^-I-^/NOCURRENT_DIRECTORY^} is specified.
8271 All directories specified by @option{^-I^/SEARCH^}
8272 switches on the @code{gnatbind}
8273 command line, in the order given.
8276 @findex ADA_PRJ_OBJECTS_FILE
8277 Each of the directories listed in the text file whose name is given
8278 by the @env{ADA_PRJ_OBJECTS_FILE} ^environment variable^logical name^.
8281 @env{ADA_PRJ_OBJECTS_FILE} is normally set by gnatmake or by the ^gnat^GNAT^
8282 driver when project files are used. It should not normally be set
8286 @findex ADA_OBJECTS_PATH
8287 Each of the directories listed in the value of the
8288 @env{ADA_OBJECTS_PATH} ^environment variable^logical name^.
8290 Construct this value
8291 exactly as the @env{PATH} environment variable: a list of directory
8292 names separated by colons (semicolons when working with the NT version
8296 Normally, define this value as a logical name containing a comma separated
8297 list of directory names.
8299 This variable can also be defined by means of an environment string
8300 (an argument to the HP C exec* set of functions).
8304 DEFINE ANOTHER_PATH FOO:[BAG]
8305 DEFINE ADA_OBJECTS_PATH ANOTHER_PATH,FOO:[BAM],FOO:[BAR]
8308 By default, the path includes GNU:[LIB.OPENVMS7_x.2_8_x.DECLIB]
8309 first, followed by the standard Ada
8310 libraries in GNU:[LIB.OPENVMS7_x.2_8_x.ADALIB].
8311 If this is not redefined, the user will obtain the HP Ada 83 IO packages
8312 (Text_IO, Sequential_IO, etc)
8313 instead of the standard Ada packages. Thus, in order to get the standard Ada
8314 packages by default, ADA_OBJECTS_PATH must be redefined.
8318 The content of the @file{ada_object_path} file which is part of the GNAT
8319 installation tree and is used to store standard libraries such as the
8320 GNAT Run Time Library (RTL) unless the switch @option{-nostdlib} is
8323 @ref{Installing a library}
8328 In the binder the switch @option{^-I^/SEARCH^}
8329 @cindex @option{^-I^/SEARCH^} (@command{gnatbind})
8330 is used to specify both source and
8331 library file paths. Use @option{^-aI^/SOURCE_SEARCH^}
8332 @cindex @option{^-aI^/SOURCE_SEARCH^} (@command{gnatbind})
8333 instead if you want to specify
8334 source paths only, and @option{^-aO^/LIBRARY_SEARCH^}
8335 @cindex @option{^-aO^/LIBRARY_SEARCH^} (@command{gnatbind})
8336 if you want to specify library paths
8337 only. This means that for the binder
8338 @option{^-I^/SEARCH=^}@var{dir} is equivalent to
8339 @option{^-aI^/SOURCE_SEARCH=^}@var{dir}
8340 @option{^-aO^/OBJECT_SEARCH=^}@var{dir}.
8341 The binder generates the bind file (a C language source file) in the
8342 current working directory.
8348 The packages @code{Ada}, @code{System}, and @code{Interfaces} and their
8349 children make up the GNAT Run-Time Library, together with the package
8350 GNAT and its children, which contain a set of useful additional
8351 library functions provided by GNAT. The sources for these units are
8352 needed by the compiler and are kept together in one directory. The ALI
8353 files and object files generated by compiling the RTL are needed by the
8354 binder and the linker and are kept together in one directory, typically
8355 different from the directory containing the sources. In a normal
8356 installation, you need not specify these directory names when compiling
8357 or binding. Either the environment variables or the built-in defaults
8358 cause these files to be found.
8360 Besides simplifying access to the RTL, a major use of search paths is
8361 in compiling sources from multiple directories. This can make
8362 development environments much more flexible.
8364 @node Examples of gnatbind Usage
8365 @section Examples of @code{gnatbind} Usage
8368 This section contains a number of examples of using the GNAT binding
8369 utility @code{gnatbind}.
8372 @item gnatbind hello
8373 The main program @code{Hello} (source program in @file{hello.adb}) is
8374 bound using the standard switch settings. The generated main program is
8375 @file{b~hello.adb}. This is the normal, default use of the binder.
8378 @item gnatbind hello -o mainprog.adb
8381 @item gnatbind HELLO.ALI /OUTPUT=Mainprog.ADB
8383 The main program @code{Hello} (source program in @file{hello.adb}) is
8384 bound using the standard switch settings. The generated main program is
8385 @file{mainprog.adb} with the associated spec in
8386 @file{mainprog.ads}. Note that you must specify the body here not the
8387 spec, in the case where the output is in Ada. Note that if this option
8388 is used, then linking must be done manually, since gnatlink will not
8389 be able to find the generated file.
8392 @item gnatbind main -C -o mainprog.c -x
8395 @item gnatbind MAIN.ALI /BIND_FILE=C /OUTPUT=Mainprog.C /READ_SOURCES=NONE
8397 The main program @code{Main} (source program in
8398 @file{main.adb}) is bound, excluding source files from the
8399 consistency checking, generating
8400 the file @file{mainprog.c}.
8403 @item gnatbind -x main_program -C -o mainprog.c
8404 This command is exactly the same as the previous example. Switches may
8405 appear anywhere in the command line, and single letter switches may be
8406 combined into a single switch.
8410 @item gnatbind -n math dbase -C -o ada-control.c
8413 @item gnatbind /NOMAIN math dbase /BIND_FILE=C /OUTPUT=ada-control.c
8415 The main program is in a language other than Ada, but calls to
8416 subprograms in packages @code{Math} and @code{Dbase} appear. This call
8417 to @code{gnatbind} generates the file @file{ada-control.c} containing
8418 the @code{adainit} and @code{adafinal} routines to be called before and
8419 after accessing the Ada units.
8422 @c ------------------------------------
8423 @node Linking Using gnatlink
8424 @chapter Linking Using @command{gnatlink}
8425 @c ------------------------------------
8429 This chapter discusses @command{gnatlink}, a tool that links
8430 an Ada program and builds an executable file. This utility
8431 invokes the system linker ^(via the @command{gcc} command)^^
8432 with a correct list of object files and library references.
8433 @command{gnatlink} automatically determines the list of files and
8434 references for the Ada part of a program. It uses the binder file
8435 generated by the @command{gnatbind} to determine this list.
8437 Note: to invoke @code{gnatlink} with a project file, use the @code{gnat}
8438 driver (see @ref{The GNAT Driver and Project Files}).
8441 * Running gnatlink::
8442 * Switches for gnatlink::
8445 @node Running gnatlink
8446 @section Running @command{gnatlink}
8449 The form of the @command{gnatlink} command is
8452 $ gnatlink @ovar{switches} @var{mainprog}@r{[}.ali@r{]}
8453 @ovar{non-Ada objects} @ovar{linker options}
8457 The arguments of @command{gnatlink} (switches, main @file{ALI} file,
8459 or linker options) may be in any order, provided that no non-Ada object may
8460 be mistaken for a main @file{ALI} file.
8461 Any file name @file{F} without the @file{.ali}
8462 extension will be taken as the main @file{ALI} file if a file exists
8463 whose name is the concatenation of @file{F} and @file{.ali}.
8466 @file{@var{mainprog}.ali} references the ALI file of the main program.
8467 The @file{.ali} extension of this file can be omitted. From this
8468 reference, @command{gnatlink} locates the corresponding binder file
8469 @file{b~@var{mainprog}.adb} and, using the information in this file along
8470 with the list of non-Ada objects and linker options, constructs a
8471 linker command file to create the executable.
8473 The arguments other than the @command{gnatlink} switches and the main
8474 @file{ALI} file are passed to the linker uninterpreted.
8475 They typically include the names of
8476 object files for units written in other languages than Ada and any library
8477 references required to resolve references in any of these foreign language
8478 units, or in @code{Import} pragmas in any Ada units.
8480 @var{linker options} is an optional list of linker specific
8482 The default linker called by gnatlink is @command{gcc} which in
8483 turn calls the appropriate system linker.
8484 Standard options for the linker such as @option{-lmy_lib} or
8485 @option{-Ldir} can be added as is.
8486 For options that are not recognized by
8487 @command{gcc} as linker options, use the @command{gcc} switches
8488 @option{-Xlinker} or @option{-Wl,}.
8489 Refer to the GCC documentation for
8490 details. Here is an example showing how to generate a linker map:
8493 $ ^gnatlink my_prog -Wl,-Map,MAPFILE^GNAT LINK my_prog.ali /MAP^
8496 Using @var{linker options} it is possible to set the program stack and
8499 See @ref{Setting Stack Size from gnatlink} and
8500 @ref{Setting Heap Size from gnatlink}.
8503 @command{gnatlink} determines the list of objects required by the Ada
8504 program and prepends them to the list of objects passed to the linker.
8505 @command{gnatlink} also gathers any arguments set by the use of
8506 @code{pragma Linker_Options} and adds them to the list of arguments
8507 presented to the linker.
8510 @command{gnatlink} accepts the following types of extra files on the command
8511 line: objects (@file{.OBJ}), libraries (@file{.OLB}), sharable images
8512 (@file{.EXE}), and options files (@file{.OPT}). These are recognized and
8513 handled according to their extension.
8516 @node Switches for gnatlink
8517 @section Switches for @command{gnatlink}
8520 The following switches are available with the @command{gnatlink} utility:
8526 @cindex @option{--version} @command{gnatlink}
8527 Display Copyright and version, then exit disregarding all other options.
8530 @cindex @option{--help} @command{gnatlink}
8531 If @option{--version} was not used, display usage, then exit disregarding
8534 @item ^-A^/BIND_FILE=ADA^
8535 @cindex @option{^-A^/BIND_FILE=ADA^} (@command{gnatlink})
8536 The binder has generated code in Ada. This is the default.
8538 @item ^-C^/BIND_FILE=C^
8539 @cindex @option{^-C^/BIND_FILE=C^} (@command{gnatlink})
8540 If instead of generating a file in Ada, the binder has generated one in
8541 C, then the linker needs to know about it. Use this switch to signal
8542 to @command{gnatlink} that the binder has generated C code rather than
8545 @item ^-f^/FORCE_OBJECT_FILE_LIST^
8546 @cindex Command line length
8547 @cindex @option{^-f^/FORCE_OBJECT_FILE_LIST^} (@command{gnatlink})
8548 On some targets, the command line length is limited, and @command{gnatlink}
8549 will generate a separate file for the linker if the list of object files
8551 The @option{^-f^/FORCE_OBJECT_FILE_LIST^} switch forces this file
8552 to be generated even if
8553 the limit is not exceeded. This is useful in some cases to deal with
8554 special situations where the command line length is exceeded.
8557 @cindex Debugging information, including
8558 @cindex @option{^-g^/DEBUG^} (@command{gnatlink})
8559 The option to include debugging information causes the Ada bind file (in
8560 other words, @file{b~@var{mainprog}.adb}) to be compiled with
8561 @option{^-g^/DEBUG^}.
8562 In addition, the binder does not delete the @file{b~@var{mainprog}.adb},
8563 @file{b~@var{mainprog}.o} and @file{b~@var{mainprog}.ali} files.
8564 Without @option{^-g^/DEBUG^}, the binder removes these files by
8565 default. The same procedure apply if a C bind file was generated using
8566 @option{^-C^/BIND_FILE=C^} @code{gnatbind} option, in this case the filenames
8567 are @file{b_@var{mainprog}.c} and @file{b_@var{mainprog}.o}.
8569 @item ^-n^/NOCOMPILE^
8570 @cindex @option{^-n^/NOCOMPILE^} (@command{gnatlink})
8571 Do not compile the file generated by the binder. This may be used when
8572 a link is rerun with different options, but there is no need to recompile
8576 @cindex @option{^-v^/VERBOSE^} (@command{gnatlink})
8577 Causes additional information to be output, including a full list of the
8578 included object files. This switch option is most useful when you want
8579 to see what set of object files are being used in the link step.
8581 @item ^-v -v^/VERBOSE/VERBOSE^
8582 @cindex @option{^-v -v^/VERBOSE/VERBOSE^} (@command{gnatlink})
8583 Very verbose mode. Requests that the compiler operate in verbose mode when
8584 it compiles the binder file, and that the system linker run in verbose mode.
8586 @item ^-o ^/EXECUTABLE=^@var{exec-name}
8587 @cindex @option{^-o^/EXECUTABLE^} (@command{gnatlink})
8588 @var{exec-name} specifies an alternate name for the generated
8589 executable program. If this switch is omitted, the executable has the same
8590 name as the main unit. For example, @code{gnatlink try.ali} creates
8591 an executable called @file{^try^TRY.EXE^}.
8594 @item -b @var{target}
8595 @cindex @option{-b} (@command{gnatlink})
8596 Compile your program to run on @var{target}, which is the name of a
8597 system configuration. You must have a GNAT cross-compiler built if
8598 @var{target} is not the same as your host system.
8601 @cindex @option{-B} (@command{gnatlink})
8602 Load compiler executables (for example, @code{gnat1}, the Ada compiler)
8603 from @var{dir} instead of the default location. Only use this switch
8604 when multiple versions of the GNAT compiler are available.
8605 @xref{Directory Options,,, gcc, The GNU Compiler Collection},
8606 for further details. You would normally use the @option{-b} or
8607 @option{-V} switch instead.
8609 @item --GCC=@var{compiler_name}
8610 @cindex @option{--GCC=compiler_name} (@command{gnatlink})
8611 Program used for compiling the binder file. The default is
8612 @command{gcc}. You need to use quotes around @var{compiler_name} if
8613 @code{compiler_name} contains spaces or other separator characters.
8614 As an example @option{--GCC="foo -x -y"} will instruct @command{gnatlink} to
8615 use @code{foo -x -y} as your compiler. Note that switch @option{-c} is always
8616 inserted after your command name. Thus in the above example the compiler
8617 command that will be used by @command{gnatlink} will be @code{foo -c -x -y}.
8618 A limitation of this syntax is that the name and path name of the executable
8619 itself must not include any embedded spaces. If the compiler executable is
8620 different from the default one (gcc or <prefix>-gcc), then the back-end
8621 switches in the ALI file are not used to compile the binder generated source.
8622 For example, this is the case with @option{--GCC="foo -x -y"}. But the back end
8623 switches will be used for @option{--GCC="gcc -gnatv"}. If several
8624 @option{--GCC=compiler_name} are used, only the last @var{compiler_name}
8625 is taken into account. However, all the additional switches are also taken
8627 @option{--GCC="foo -x -y" --GCC="bar -z -t"} is equivalent to
8628 @option{--GCC="bar -x -y -z -t"}.
8630 @item --LINK=@var{name}
8631 @cindex @option{--LINK=} (@command{gnatlink})
8632 @var{name} is the name of the linker to be invoked. This is especially
8633 useful in mixed language programs since languages such as C++ require
8634 their own linker to be used. When this switch is omitted, the default
8635 name for the linker is @command{gcc}. When this switch is used, the
8636 specified linker is called instead of @command{gcc} with exactly the same
8637 parameters that would have been passed to @command{gcc} so if the desired
8638 linker requires different parameters it is necessary to use a wrapper
8639 script that massages the parameters before invoking the real linker. It
8640 may be useful to control the exact invocation by using the verbose
8646 @item /DEBUG=TRACEBACK
8647 @cindex @code{/DEBUG=TRACEBACK} (@command{gnatlink})
8648 This qualifier causes sufficient information to be included in the
8649 executable file to allow a traceback, but does not include the full
8650 symbol information needed by the debugger.
8652 @item /IDENTIFICATION="<string>"
8653 @code{"<string>"} specifies the string to be stored in the image file
8654 identification field in the image header.
8655 It overrides any pragma @code{Ident} specified string.
8657 @item /NOINHIBIT-EXEC
8658 Generate the executable file even if there are linker warnings.
8660 @item /NOSTART_FILES
8661 Don't link in the object file containing the ``main'' transfer address.
8662 Used when linking with a foreign language main program compiled with an
8666 Prefer linking with object libraries over sharable images, even without
8672 @node The GNAT Make Program gnatmake
8673 @chapter The GNAT Make Program @command{gnatmake}
8677 * Running gnatmake::
8678 * Switches for gnatmake::
8679 * Mode Switches for gnatmake::
8680 * Notes on the Command Line::
8681 * How gnatmake Works::
8682 * Examples of gnatmake Usage::
8685 A typical development cycle when working on an Ada program consists of
8686 the following steps:
8690 Edit some sources to fix bugs.
8696 Compile all sources affected.
8706 The third step can be tricky, because not only do the modified files
8707 @cindex Dependency rules
8708 have to be compiled, but any files depending on these files must also be
8709 recompiled. The dependency rules in Ada can be quite complex, especially
8710 in the presence of overloading, @code{use} clauses, generics and inlined
8713 @command{gnatmake} automatically takes care of the third and fourth steps
8714 of this process. It determines which sources need to be compiled,
8715 compiles them, and binds and links the resulting object files.
8717 Unlike some other Ada make programs, the dependencies are always
8718 accurately recomputed from the new sources. The source based approach of
8719 the GNAT compilation model makes this possible. This means that if
8720 changes to the source program cause corresponding changes in
8721 dependencies, they will always be tracked exactly correctly by
8724 @node Running gnatmake
8725 @section Running @command{gnatmake}
8728 The usual form of the @command{gnatmake} command is
8731 $ gnatmake @ovar{switches} @var{file_name}
8732 @ovar{file_names} @ovar{mode_switches}
8736 The only required argument is one @var{file_name}, which specifies
8737 a compilation unit that is a main program. Several @var{file_names} can be
8738 specified: this will result in several executables being built.
8739 If @code{switches} are present, they can be placed before the first
8740 @var{file_name}, between @var{file_names} or after the last @var{file_name}.
8741 If @var{mode_switches} are present, they must always be placed after
8742 the last @var{file_name} and all @code{switches}.
8744 If you are using standard file extensions (@file{.adb} and @file{.ads}), then the
8745 extension may be omitted from the @var{file_name} arguments. However, if
8746 you are using non-standard extensions, then it is required that the
8747 extension be given. A relative or absolute directory path can be
8748 specified in a @var{file_name}, in which case, the input source file will
8749 be searched for in the specified directory only. Otherwise, the input
8750 source file will first be searched in the directory where
8751 @command{gnatmake} was invoked and if it is not found, it will be search on
8752 the source path of the compiler as described in
8753 @ref{Search Paths and the Run-Time Library (RTL)}.
8755 All @command{gnatmake} output (except when you specify
8756 @option{^-M^/DEPENDENCIES_LIST^}) is to
8757 @file{stderr}. The output produced by the
8758 @option{^-M^/DEPENDENCIES_LIST^} switch is send to
8761 @node Switches for gnatmake
8762 @section Switches for @command{gnatmake}
8765 You may specify any of the following switches to @command{gnatmake}:
8771 @cindex @option{--version} @command{gnatmake}
8772 Display Copyright and version, then exit disregarding all other options.
8775 @cindex @option{--help} @command{gnatmake}
8776 If @option{--version} was not used, display usage, then exit disregarding
8780 @item --GCC=@var{compiler_name}
8781 @cindex @option{--GCC=compiler_name} (@command{gnatmake})
8782 Program used for compiling. The default is `@command{gcc}'. You need to use
8783 quotes around @var{compiler_name} if @code{compiler_name} contains
8784 spaces or other separator characters. As an example @option{--GCC="foo -x
8785 -y"} will instruct @command{gnatmake} to use @code{foo -x -y} as your
8786 compiler. A limitation of this syntax is that the name and path name of
8787 the executable itself must not include any embedded spaces. Note that
8788 switch @option{-c} is always inserted after your command name. Thus in the
8789 above example the compiler command that will be used by @command{gnatmake}
8790 will be @code{foo -c -x -y}. If several @option{--GCC=compiler_name} are
8791 used, only the last @var{compiler_name} is taken into account. However,
8792 all the additional switches are also taken into account. Thus,
8793 @option{--GCC="foo -x -y" --GCC="bar -z -t"} is equivalent to
8794 @option{--GCC="bar -x -y -z -t"}.
8796 @item --GNATBIND=@var{binder_name}
8797 @cindex @option{--GNATBIND=binder_name} (@command{gnatmake})
8798 Program used for binding. The default is `@code{gnatbind}'. You need to
8799 use quotes around @var{binder_name} if @var{binder_name} contains spaces
8800 or other separator characters. As an example @option{--GNATBIND="bar -x
8801 -y"} will instruct @command{gnatmake} to use @code{bar -x -y} as your
8802 binder. Binder switches that are normally appended by @command{gnatmake}
8803 to `@code{gnatbind}' are now appended to the end of @code{bar -x -y}.
8804 A limitation of this syntax is that the name and path name of the executable
8805 itself must not include any embedded spaces.
8807 @item --GNATLINK=@var{linker_name}
8808 @cindex @option{--GNATLINK=linker_name} (@command{gnatmake})
8809 Program used for linking. The default is `@command{gnatlink}'. You need to
8810 use quotes around @var{linker_name} if @var{linker_name} contains spaces
8811 or other separator characters. As an example @option{--GNATLINK="lan -x
8812 -y"} will instruct @command{gnatmake} to use @code{lan -x -y} as your
8813 linker. Linker switches that are normally appended by @command{gnatmake} to
8814 `@command{gnatlink}' are now appended to the end of @code{lan -x -y}.
8815 A limitation of this syntax is that the name and path name of the executable
8816 itself must not include any embedded spaces.
8820 @item ^-a^/ALL_FILES^
8821 @cindex @option{^-a^/ALL_FILES^} (@command{gnatmake})
8822 Consider all files in the make process, even the GNAT internal system
8823 files (for example, the predefined Ada library files), as well as any
8824 locked files. Locked files are files whose ALI file is write-protected.
8826 @command{gnatmake} does not check these files,
8827 because the assumption is that the GNAT internal files are properly up
8828 to date, and also that any write protected ALI files have been properly
8829 installed. Note that if there is an installation problem, such that one
8830 of these files is not up to date, it will be properly caught by the
8832 You may have to specify this switch if you are working on GNAT
8833 itself. The switch @option{^-a^/ALL_FILES^} is also useful
8834 in conjunction with @option{^-f^/FORCE_COMPILE^}
8835 if you need to recompile an entire application,
8836 including run-time files, using special configuration pragmas,
8837 such as a @code{Normalize_Scalars} pragma.
8840 @code{gnatmake ^-a^/ALL_FILES^} compiles all GNAT
8843 @code{gcc -c -gnatpg} rather than @code{gcc -c}.
8846 the @code{/CHECKS=SUPPRESS_ALL /STYLE_CHECKS=GNAT} switch.
8849 @item ^-b^/ACTIONS=BIND^
8850 @cindex @option{^-b^/ACTIONS=BIND^} (@command{gnatmake})
8851 Bind only. Can be combined with @option{^-c^/ACTIONS=COMPILE^} to do
8852 compilation and binding, but no link.
8853 Can be combined with @option{^-l^/ACTIONS=LINK^}
8854 to do binding and linking. When not combined with
8855 @option{^-c^/ACTIONS=COMPILE^}
8856 all the units in the closure of the main program must have been previously
8857 compiled and must be up to date. The root unit specified by @var{file_name}
8858 may be given without extension, with the source extension or, if no GNAT
8859 Project File is specified, with the ALI file extension.
8861 @item ^-c^/ACTIONS=COMPILE^
8862 @cindex @option{^-c^/ACTIONS=COMPILE^} (@command{gnatmake})
8863 Compile only. Do not perform binding, except when @option{^-b^/ACTIONS=BIND^}
8864 is also specified. Do not perform linking, except if both
8865 @option{^-b^/ACTIONS=BIND^} and
8866 @option{^-l^/ACTIONS=LINK^} are also specified.
8867 If the root unit specified by @var{file_name} is not a main unit, this is the
8868 default. Otherwise @command{gnatmake} will attempt binding and linking
8869 unless all objects are up to date and the executable is more recent than
8873 @cindex @option{^-C^/MAPPING^} (@command{gnatmake})
8874 Use a temporary mapping file. A mapping file is a way to communicate to the
8875 compiler two mappings: from unit names to file names (without any directory
8876 information) and from file names to path names (with full directory
8877 information). These mappings are used by the compiler to short-circuit the path
8878 search. When @command{gnatmake} is invoked with this switch, it will create
8879 a temporary mapping file, initially populated by the project manager,
8880 if @option{^-P^/PROJECT_FILE^} is used, otherwise initially empty.
8881 Each invocation of the compiler will add the newly accessed sources to the
8882 mapping file. This will improve the source search during the next invocation
8885 @item ^-C=^/USE_MAPPING_FILE=^@var{file}
8886 @cindex @option{^-C=^/USE_MAPPING^} (@command{gnatmake})
8887 Use a specific mapping file. The file, specified as a path name (absolute or
8888 relative) by this switch, should already exist, otherwise the switch is
8889 ineffective. The specified mapping file will be communicated to the compiler.
8890 This switch is not compatible with a project file
8891 (^-P^/PROJECT_FILE=^@var{file}) or with multiple compiling processes
8892 (^-j^/PROCESSES=^nnn, when nnn is greater than 1).
8894 @item ^-d^/DISPLAY_PROGRESS^
8895 @cindex @option{^-d^/DISPLAY_PROGRESS^} (@command{gnatmake})
8896 Display progress for each source, up to date or not, as a single line
8899 completed x out of y (zz%)
8902 If the file needs to be compiled this is displayed after the invocation of
8903 the compiler. These lines are displayed even in quiet output mode.
8905 @item ^-D ^/DIRECTORY_OBJECTS=^@var{dir}
8906 @cindex @option{^-D^/DIRECTORY_OBJECTS^} (@command{gnatmake})
8907 Put all object files and ALI file in directory @var{dir}.
8908 If the @option{^-D^/DIRECTORY_OBJECTS^} switch is not used, all object files
8909 and ALI files go in the current working directory.
8911 This switch cannot be used when using a project file.
8915 @cindex @option{-eL} (@command{gnatmake})
8916 Follow all symbolic links when processing project files.
8919 @item ^-eS^/STANDARD_OUTPUT_FOR_COMMANDS^
8920 @cindex @option{^-eS^/STANDARD_OUTPUT_FOR_COMMANDS^} (@command{gnatmake})
8921 Output the commands for the compiler, the binder and the linker
8922 on ^standard output^SYS$OUTPUT^,
8923 instead of ^standard error^SYS$ERROR^.
8925 @item ^-f^/FORCE_COMPILE^
8926 @cindex @option{^-f^/FORCE_COMPILE^} (@command{gnatmake})
8927 Force recompilations. Recompile all sources, even though some object
8928 files may be up to date, but don't recompile predefined or GNAT internal
8929 files or locked files (files with a write-protected ALI file),
8930 unless the @option{^-a^/ALL_FILES^} switch is also specified.
8932 @item ^-F^/FULL_PATH_IN_BRIEF_MESSAGES^
8933 @cindex @option{^-F^/FULL_PATH_IN_BRIEF_MESSAGES^} (@command{gnatmake})
8934 When using project files, if some errors or warnings are detected during
8935 parsing and verbose mode is not in effect (no use of switch
8936 ^-v^/VERBOSE^), then error lines start with the full path name of the project
8937 file, rather than its simple file name.
8940 @cindex @option{^-g^/DEBUG^} (@command{gnatmake})
8941 Enable debugging. This switch is simply passed to the compiler and to the
8944 @item ^-i^/IN_PLACE^
8945 @cindex @option{^-i^/IN_PLACE^} (@command{gnatmake})
8946 In normal mode, @command{gnatmake} compiles all object files and ALI files
8947 into the current directory. If the @option{^-i^/IN_PLACE^} switch is used,
8948 then instead object files and ALI files that already exist are overwritten
8949 in place. This means that once a large project is organized into separate
8950 directories in the desired manner, then @command{gnatmake} will automatically
8951 maintain and update this organization. If no ALI files are found on the
8952 Ada object path (@ref{Search Paths and the Run-Time Library (RTL)}),
8953 the new object and ALI files are created in the
8954 directory containing the source being compiled. If another organization
8955 is desired, where objects and sources are kept in different directories,
8956 a useful technique is to create dummy ALI files in the desired directories.
8957 When detecting such a dummy file, @command{gnatmake} will be forced to
8958 recompile the corresponding source file, and it will be put the resulting
8959 object and ALI files in the directory where it found the dummy file.
8961 @item ^-j^/PROCESSES=^@var{n}
8962 @cindex @option{^-j^/PROCESSES^} (@command{gnatmake})
8963 @cindex Parallel make
8964 Use @var{n} processes to carry out the (re)compilations. On a
8965 multiprocessor machine compilations will occur in parallel. In the
8966 event of compilation errors, messages from various compilations might
8967 get interspersed (but @command{gnatmake} will give you the full ordered
8968 list of failing compiles at the end). If this is problematic, rerun
8969 the make process with n set to 1 to get a clean list of messages.
8971 @item ^-k^/CONTINUE_ON_ERROR^
8972 @cindex @option{^-k^/CONTINUE_ON_ERROR^} (@command{gnatmake})
8973 Keep going. Continue as much as possible after a compilation error. To
8974 ease the programmer's task in case of compilation errors, the list of
8975 sources for which the compile fails is given when @command{gnatmake}
8978 If @command{gnatmake} is invoked with several @file{file_names} and with this
8979 switch, if there are compilation errors when building an executable,
8980 @command{gnatmake} will not attempt to build the following executables.
8982 @item ^-l^/ACTIONS=LINK^
8983 @cindex @option{^-l^/ACTIONS=LINK^} (@command{gnatmake})
8984 Link only. Can be combined with @option{^-b^/ACTIONS=BIND^} to binding
8985 and linking. Linking will not be performed if combined with
8986 @option{^-c^/ACTIONS=COMPILE^}
8987 but not with @option{^-b^/ACTIONS=BIND^}.
8988 When not combined with @option{^-b^/ACTIONS=BIND^}
8989 all the units in the closure of the main program must have been previously
8990 compiled and must be up to date, and the main program needs to have been bound.
8991 The root unit specified by @var{file_name}
8992 may be given without extension, with the source extension or, if no GNAT
8993 Project File is specified, with the ALI file extension.
8995 @item ^-m^/MINIMAL_RECOMPILATION^
8996 @cindex @option{^-m^/MINIMAL_RECOMPILATION^} (@command{gnatmake})
8997 Specify that the minimum necessary amount of recompilations
8998 be performed. In this mode @command{gnatmake} ignores time
8999 stamp differences when the only
9000 modifications to a source file consist in adding/removing comments,
9001 empty lines, spaces or tabs. This means that if you have changed the
9002 comments in a source file or have simply reformatted it, using this
9003 switch will tell @command{gnatmake} not to recompile files that depend on it
9004 (provided other sources on which these files depend have undergone no
9005 semantic modifications). Note that the debugging information may be
9006 out of date with respect to the sources if the @option{-m} switch causes
9007 a compilation to be switched, so the use of this switch represents a
9008 trade-off between compilation time and accurate debugging information.
9010 @item ^-M^/DEPENDENCIES_LIST^
9011 @cindex Dependencies, producing list
9012 @cindex @option{^-M^/DEPENDENCIES_LIST^} (@command{gnatmake})
9013 Check if all objects are up to date. If they are, output the object
9014 dependences to @file{stdout} in a form that can be directly exploited in
9015 a @file{Makefile}. By default, each source file is prefixed with its
9016 (relative or absolute) directory name. This name is whatever you
9017 specified in the various @option{^-aI^/SOURCE_SEARCH^}
9018 and @option{^-I^/SEARCH^} switches. If you use
9019 @code{gnatmake ^-M^/DEPENDENCIES_LIST^}
9020 @option{^-q^/QUIET^}
9021 (see below), only the source file names,
9022 without relative paths, are output. If you just specify the
9023 @option{^-M^/DEPENDENCIES_LIST^}
9024 switch, dependencies of the GNAT internal system files are omitted. This
9025 is typically what you want. If you also specify
9026 the @option{^-a^/ALL_FILES^} switch,
9027 dependencies of the GNAT internal files are also listed. Note that
9028 dependencies of the objects in external Ada libraries (see switch
9029 @option{^-aL^/SKIP_MISSING=^}@var{dir} in the following list)
9032 @item ^-n^/DO_OBJECT_CHECK^
9033 @cindex @option{^-n^/DO_OBJECT_CHECK^} (@command{gnatmake})
9034 Don't compile, bind, or link. Checks if all objects are up to date.
9035 If they are not, the full name of the first file that needs to be
9036 recompiled is printed.
9037 Repeated use of this option, followed by compiling the indicated source
9038 file, will eventually result in recompiling all required units.
9040 @item ^-o ^/EXECUTABLE=^@var{exec_name}
9041 @cindex @option{^-o^/EXECUTABLE^} (@command{gnatmake})
9042 Output executable name. The name of the final executable program will be
9043 @var{exec_name}. If the @option{^-o^/EXECUTABLE^} switch is omitted the default
9044 name for the executable will be the name of the input file in appropriate form
9045 for an executable file on the host system.
9047 This switch cannot be used when invoking @command{gnatmake} with several
9050 @item ^-p or --create-missing-dirs^/CREATE_MISSING_DIRS^
9051 @cindex @option{^-p^/CREATE_MISSING_DIRS^} (@command{gnatmake})
9052 When using project files (^-P^/PROJECT_FILE=^@var{project}), create
9053 automatically missing object directories, library directories and exec
9056 @item ^-P^/PROJECT_FILE=^@var{project}
9057 @cindex @option{^-P^/PROJECT_FILE^} (@command{gnatmake})
9058 Use project file @var{project}. Only one such switch can be used.
9059 @xref{gnatmake and Project Files}.
9062 @cindex @option{^-q^/QUIET^} (@command{gnatmake})
9063 Quiet. When this flag is not set, the commands carried out by
9064 @command{gnatmake} are displayed.
9066 @item ^-s^/SWITCH_CHECK/^
9067 @cindex @option{^-s^/SWITCH_CHECK^} (@command{gnatmake})
9068 Recompile if compiler switches have changed since last compilation.
9069 All compiler switches but -I and -o are taken into account in the
9071 orders between different ``first letter'' switches are ignored, but
9072 orders between same switches are taken into account. For example,
9073 @option{-O -O2} is different than @option{-O2 -O}, but @option{-g -O}
9074 is equivalent to @option{-O -g}.
9076 This switch is recommended when Integrated Preprocessing is used.
9079 @cindex @option{^-u^/UNIQUE^} (@command{gnatmake})
9080 Unique. Recompile at most the main files. It implies -c. Combined with
9081 -f, it is equivalent to calling the compiler directly. Note that using
9082 ^-u^/UNIQUE^ with a project file and no main has a special meaning
9083 (@pxref{Project Files and Main Subprograms}).
9085 @item ^-U^/ALL_PROJECTS^
9086 @cindex @option{^-U^/ALL_PROJECTS^} (@command{gnatmake})
9087 When used without a project file or with one or several mains on the command
9088 line, is equivalent to ^-u^/UNIQUE^. When used with a project file and no main
9089 on the command line, all sources of all project files are checked and compiled
9090 if not up to date, and libraries are rebuilt, if necessary.
9093 @cindex @option{^-v^/REASONS^} (@command{gnatmake})
9094 Verbose. Display the reason for all recompilations @command{gnatmake}
9095 decides are necessary, with the highest verbosity level.
9097 @item ^-vl^/LOW_VERBOSITY^
9098 @cindex @option{^-vl^/LOW_VERBOSITY^} (@command{gnatmake})
9099 Verbosity level Low. Display fewer lines than in verbosity Medium.
9101 @item ^-vm^/MEDIUM_VERBOSITY^
9102 @cindex @option{^-vm^/MEDIUM_VERBOSITY^} (@command{gnatmake})
9103 Verbosity level Medium. Potentially display fewer lines than in verbosity High.
9105 @item ^-vh^/HIGH_VERBOSITY^
9106 @cindex @option{^-vm^/HIGH_VERBOSITY^} (@command{gnatmake})
9107 Verbosity level High. Equivalent to ^-v^/REASONS^.
9109 @item ^-vP^/MESSAGES_PROJECT_FILE=^@emph{x}
9110 Indicate the verbosity of the parsing of GNAT project files.
9111 @xref{Switches Related to Project Files}.
9113 @item ^-x^/NON_PROJECT_UNIT_COMPILATION^
9114 @cindex @option{^-x^/NON_PROJECT_UNIT_COMPILATION^} (@command{gnatmake})
9115 Indicate that sources that are not part of any Project File may be compiled.
9116 Normally, when using Project Files, only sources that are part of a Project
9117 File may be compile. When this switch is used, a source outside of all Project
9118 Files may be compiled. The ALI file and the object file will be put in the
9119 object directory of the main Project. The compilation switches used will only
9120 be those specified on the command line. Even when
9121 @option{^-x^/NON_PROJECT_UNIT_COMPILATION^} is used, mains specified on the
9122 command line need to be sources of a project file.
9124 @item ^-X^/EXTERNAL_REFERENCE=^@var{name=value}
9125 Indicate that external variable @var{name} has the value @var{value}.
9126 The Project Manager will use this value for occurrences of
9127 @code{external(name)} when parsing the project file.
9128 @xref{Switches Related to Project Files}.
9131 @cindex @option{^-z^/NOMAIN^} (@command{gnatmake})
9132 No main subprogram. Bind and link the program even if the unit name
9133 given on the command line is a package name. The resulting executable
9134 will execute the elaboration routines of the package and its closure,
9135 then the finalization routines.
9140 @item @command{gcc} @asis{switches}
9142 Any uppercase or multi-character switch that is not a @command{gnatmake} switch
9143 is passed to @command{gcc} (e.g.@: @option{-O}, @option{-gnato,} etc.)
9146 Any qualifier that cannot be recognized as a qualifier for @code{GNAT MAKE}
9147 but is recognizable as a valid qualifier for @code{GNAT COMPILE} is
9148 automatically treated as a compiler switch, and passed on to all
9149 compilations that are carried out.
9154 Source and library search path switches:
9158 @item ^-aI^/SOURCE_SEARCH=^@var{dir}
9159 @cindex @option{^-aI^/SOURCE_SEARCH^} (@command{gnatmake})
9160 When looking for source files also look in directory @var{dir}.
9161 The order in which source files search is undertaken is
9162 described in @ref{Search Paths and the Run-Time Library (RTL)}.
9164 @item ^-aL^/SKIP_MISSING=^@var{dir}
9165 @cindex @option{^-aL^/SKIP_MISSING^} (@command{gnatmake})
9166 Consider @var{dir} as being an externally provided Ada library.
9167 Instructs @command{gnatmake} to skip compilation units whose @file{.ALI}
9168 files have been located in directory @var{dir}. This allows you to have
9169 missing bodies for the units in @var{dir} and to ignore out of date bodies
9170 for the same units. You still need to specify
9171 the location of the specs for these units by using the switches
9172 @option{^-aI^/SOURCE_SEARCH=^@var{dir}}
9173 or @option{^-I^/SEARCH=^@var{dir}}.
9174 Note: this switch is provided for compatibility with previous versions
9175 of @command{gnatmake}. The easier method of causing standard libraries
9176 to be excluded from consideration is to write-protect the corresponding
9179 @item ^-aO^/OBJECT_SEARCH=^@var{dir}
9180 @cindex @option{^-aO^/OBJECT_SEARCH^} (@command{gnatmake})
9181 When searching for library and object files, look in directory
9182 @var{dir}. The order in which library files are searched is described in
9183 @ref{Search Paths for gnatbind}.
9185 @item ^-A^/CONDITIONAL_SOURCE_SEARCH=^@var{dir}
9186 @cindex Search paths, for @command{gnatmake}
9187 @cindex @option{^-A^/CONDITIONAL_SOURCE_SEARCH^} (@command{gnatmake})
9188 Equivalent to @option{^-aL^/SKIP_MISSING=^@var{dir}
9189 ^-aI^/SOURCE_SEARCH=^@var{dir}}.
9191 @item ^-I^/SEARCH=^@var{dir}
9192 @cindex @option{^-I^/SEARCH^} (@command{gnatmake})
9193 Equivalent to @option{^-aO^/OBJECT_SEARCH=^@var{dir}
9194 ^-aI^/SOURCE_SEARCH=^@var{dir}}.
9196 @item ^-I-^/NOCURRENT_DIRECTORY^
9197 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@command{gnatmake})
9198 @cindex Source files, suppressing search
9199 Do not look for source files in the directory containing the source
9200 file named in the command line.
9201 Do not look for ALI or object files in the directory
9202 where @command{gnatmake} was invoked.
9204 @item ^-L^/LIBRARY_SEARCH=^@var{dir}
9205 @cindex @option{^-L^/LIBRARY_SEARCH^} (@command{gnatmake})
9206 @cindex Linker libraries
9207 Add directory @var{dir} to the list of directories in which the linker
9208 will search for libraries. This is equivalent to
9209 @option{-largs ^-L^/LIBRARY_SEARCH=^}@var{dir}.
9211 Furthermore, under Windows, the sources pointed to by the libraries path
9212 set in the registry are not searched for.
9216 @cindex @option{-nostdinc} (@command{gnatmake})
9217 Do not look for source files in the system default directory.
9220 @cindex @option{-nostdlib} (@command{gnatmake})
9221 Do not look for library files in the system default directory.
9223 @item --RTS=@var{rts-path}
9224 @cindex @option{--RTS} (@command{gnatmake})
9225 Specifies the default location of the runtime library. GNAT looks for the
9227 in the following directories, and stops as soon as a valid runtime is found
9228 (@file{adainclude} or @file{ada_source_path}, and @file{adalib} or
9229 @file{ada_object_path} present):
9232 @item <current directory>/$rts_path
9234 @item <default-search-dir>/$rts_path
9236 @item <default-search-dir>/rts-$rts_path
9240 The selected path is handled like a normal RTS path.
9244 @node Mode Switches for gnatmake
9245 @section Mode Switches for @command{gnatmake}
9248 The mode switches (referred to as @code{mode_switches}) allow the
9249 inclusion of switches that are to be passed to the compiler itself, the
9250 binder or the linker. The effect of a mode switch is to cause all
9251 subsequent switches up to the end of the switch list, or up to the next
9252 mode switch, to be interpreted as switches to be passed on to the
9253 designated component of GNAT.
9257 @item -cargs @var{switches}
9258 @cindex @option{-cargs} (@command{gnatmake})
9259 Compiler switches. Here @var{switches} is a list of switches
9260 that are valid switches for @command{gcc}. They will be passed on to
9261 all compile steps performed by @command{gnatmake}.
9263 @item -bargs @var{switches}
9264 @cindex @option{-bargs} (@command{gnatmake})
9265 Binder switches. Here @var{switches} is a list of switches
9266 that are valid switches for @code{gnatbind}. They will be passed on to
9267 all bind steps performed by @command{gnatmake}.
9269 @item -largs @var{switches}
9270 @cindex @option{-largs} (@command{gnatmake})
9271 Linker switches. Here @var{switches} is a list of switches
9272 that are valid switches for @command{gnatlink}. They will be passed on to
9273 all link steps performed by @command{gnatmake}.
9275 @item -margs @var{switches}
9276 @cindex @option{-margs} (@command{gnatmake})
9277 Make switches. The switches are directly interpreted by @command{gnatmake},
9278 regardless of any previous occurrence of @option{-cargs}, @option{-bargs}
9282 @node Notes on the Command Line
9283 @section Notes on the Command Line
9286 This section contains some additional useful notes on the operation
9287 of the @command{gnatmake} command.
9291 @cindex Recompilation, by @command{gnatmake}
9292 If @command{gnatmake} finds no ALI files, it recompiles the main program
9293 and all other units required by the main program.
9294 This means that @command{gnatmake}
9295 can be used for the initial compile, as well as during subsequent steps of
9296 the development cycle.
9299 If you enter @code{gnatmake @var{file}.adb}, where @file{@var{file}.adb}
9300 is a subunit or body of a generic unit, @command{gnatmake} recompiles
9301 @file{@var{file}.adb} (because it finds no ALI) and stops, issuing a
9305 In @command{gnatmake} the switch @option{^-I^/SEARCH^}
9306 is used to specify both source and
9307 library file paths. Use @option{^-aI^/SOURCE_SEARCH^}
9308 instead if you just want to specify
9309 source paths only and @option{^-aO^/OBJECT_SEARCH^}
9310 if you want to specify library paths
9314 @command{gnatmake} will ignore any files whose ALI file is write-protected.
9315 This may conveniently be used to exclude standard libraries from
9316 consideration and in particular it means that the use of the
9317 @option{^-f^/FORCE_COMPILE^} switch will not recompile these files
9318 unless @option{^-a^/ALL_FILES^} is also specified.
9321 @command{gnatmake} has been designed to make the use of Ada libraries
9322 particularly convenient. Assume you have an Ada library organized
9323 as follows: @i{^obj-dir^[OBJ_DIR]^} contains the objects and ALI files for
9324 of your Ada compilation units,
9325 whereas @i{^include-dir^[INCLUDE_DIR]^} contains the
9326 specs of these units, but no bodies. Then to compile a unit
9327 stored in @code{main.adb}, which uses this Ada library you would just type
9331 $ gnatmake -aI@var{include-dir} -aL@var{obj-dir} main
9334 $ gnatmake /SOURCE_SEARCH=@i{[INCLUDE_DIR]}
9335 /SKIP_MISSING=@i{[OBJ_DIR]} main
9340 Using @command{gnatmake} along with the
9341 @option{^-m (minimal recompilation)^/MINIMAL_RECOMPILATION^}
9342 switch provides a mechanism for avoiding unnecessary recompilations. Using
9344 you can update the comments/format of your
9345 source files without having to recompile everything. Note, however, that
9346 adding or deleting lines in a source files may render its debugging
9347 info obsolete. If the file in question is a spec, the impact is rather
9348 limited, as that debugging info will only be useful during the
9349 elaboration phase of your program. For bodies the impact can be more
9350 significant. In all events, your debugger will warn you if a source file
9351 is more recent than the corresponding object, and alert you to the fact
9352 that the debugging information may be out of date.
9355 @node How gnatmake Works
9356 @section How @command{gnatmake} Works
9359 Generally @command{gnatmake} automatically performs all necessary
9360 recompilations and you don't need to worry about how it works. However,
9361 it may be useful to have some basic understanding of the @command{gnatmake}
9362 approach and in particular to understand how it uses the results of
9363 previous compilations without incorrectly depending on them.
9365 First a definition: an object file is considered @dfn{up to date} if the
9366 corresponding ALI file exists and if all the source files listed in the
9367 dependency section of this ALI file have time stamps matching those in
9368 the ALI file. This means that neither the source file itself nor any
9369 files that it depends on have been modified, and hence there is no need
9370 to recompile this file.
9372 @command{gnatmake} works by first checking if the specified main unit is up
9373 to date. If so, no compilations are required for the main unit. If not,
9374 @command{gnatmake} compiles the main program to build a new ALI file that
9375 reflects the latest sources. Then the ALI file of the main unit is
9376 examined to find all the source files on which the main program depends,
9377 and @command{gnatmake} recursively applies the above procedure on all these
9380 This process ensures that @command{gnatmake} only trusts the dependencies
9381 in an existing ALI file if they are known to be correct. Otherwise it
9382 always recompiles to determine a new, guaranteed accurate set of
9383 dependencies. As a result the program is compiled ``upside down'' from what may
9384 be more familiar as the required order of compilation in some other Ada
9385 systems. In particular, clients are compiled before the units on which
9386 they depend. The ability of GNAT to compile in any order is critical in
9387 allowing an order of compilation to be chosen that guarantees that
9388 @command{gnatmake} will recompute a correct set of new dependencies if
9391 When invoking @command{gnatmake} with several @var{file_names}, if a unit is
9392 imported by several of the executables, it will be recompiled at most once.
9394 Note: when using non-standard naming conventions
9395 (@pxref{Using Other File Names}), changing through a configuration pragmas
9396 file the version of a source and invoking @command{gnatmake} to recompile may
9397 have no effect, if the previous version of the source is still accessible
9398 by @command{gnatmake}. It may be necessary to use the switch
9399 ^-f^/FORCE_COMPILE^.
9401 @node Examples of gnatmake Usage
9402 @section Examples of @command{gnatmake} Usage
9405 @item gnatmake hello.adb
9406 Compile all files necessary to bind and link the main program
9407 @file{hello.adb} (containing unit @code{Hello}) and bind and link the
9408 resulting object files to generate an executable file @file{^hello^HELLO.EXE^}.
9410 @item gnatmake main1 main2 main3
9411 Compile all files necessary to bind and link the main programs
9412 @file{main1.adb} (containing unit @code{Main1}), @file{main2.adb}
9413 (containing unit @code{Main2}) and @file{main3.adb}
9414 (containing unit @code{Main3}) and bind and link the resulting object files
9415 to generate three executable files @file{^main1^MAIN1.EXE^},
9416 @file{^main2^MAIN2.EXE^}
9417 and @file{^main3^MAIN3.EXE^}.
9420 @item gnatmake -q Main_Unit -cargs -O2 -bargs -l
9424 @item gnatmake Main_Unit /QUIET
9425 /COMPILER_QUALIFIERS /OPTIMIZE=ALL
9426 /BINDER_QUALIFIERS /ORDER_OF_ELABORATION
9428 Compile all files necessary to bind and link the main program unit
9429 @code{Main_Unit} (from file @file{main_unit.adb}). All compilations will
9430 be done with optimization level 2 and the order of elaboration will be
9431 listed by the binder. @command{gnatmake} will operate in quiet mode, not
9432 displaying commands it is executing.
9435 @c *************************
9436 @node Improving Performance
9437 @chapter Improving Performance
9438 @cindex Improving performance
9441 This chapter presents several topics related to program performance.
9442 It first describes some of the tradeoffs that need to be considered
9443 and some of the techniques for making your program run faster.
9444 It then documents the @command{gnatelim} tool and unused subprogram/data
9445 elimination feature, which can reduce the size of program executables.
9447 Note: to invoke @command{gnatelim} with a project file, use the @code{gnat}
9448 driver (see @ref{The GNAT Driver and Project Files}).
9452 * Performance Considerations::
9453 * Text_IO Suggestions::
9454 * Reducing Size of Ada Executables with gnatelim::
9455 * Reducing Size of Executables with unused subprogram/data elimination::
9459 @c *****************************
9460 @node Performance Considerations
9461 @section Performance Considerations
9464 The GNAT system provides a number of options that allow a trade-off
9469 performance of the generated code
9472 speed of compilation
9475 minimization of dependences and recompilation
9478 the degree of run-time checking.
9482 The defaults (if no options are selected) aim at improving the speed
9483 of compilation and minimizing dependences, at the expense of performance
9484 of the generated code:
9491 no inlining of subprogram calls
9494 all run-time checks enabled except overflow and elaboration checks
9498 These options are suitable for most program development purposes. This
9499 chapter describes how you can modify these choices, and also provides
9500 some guidelines on debugging optimized code.
9503 * Controlling Run-Time Checks::
9504 * Use of Restrictions::
9505 * Optimization Levels::
9506 * Debugging Optimized Code::
9507 * Inlining of Subprograms::
9508 * Other Optimization Switches::
9509 * Optimization and Strict Aliasing::
9512 * Coverage Analysis::
9516 @node Controlling Run-Time Checks
9517 @subsection Controlling Run-Time Checks
9520 By default, GNAT generates all run-time checks, except arithmetic overflow
9521 checking for integer operations and checks for access before elaboration on
9522 subprogram calls. The latter are not required in default mode, because all
9523 necessary checking is done at compile time.
9524 @cindex @option{-gnatp} (@command{gcc})
9525 @cindex @option{-gnato} (@command{gcc})
9526 Two gnat switches, @option{-gnatp} and @option{-gnato} allow this default to
9527 be modified. @xref{Run-Time Checks}.
9529 Our experience is that the default is suitable for most development
9532 We treat integer overflow specially because these
9533 are quite expensive and in our experience are not as important as other
9534 run-time checks in the development process. Note that division by zero
9535 is not considered an overflow check, and divide by zero checks are
9536 generated where required by default.
9538 Elaboration checks are off by default, and also not needed by default, since
9539 GNAT uses a static elaboration analysis approach that avoids the need for
9540 run-time checking. This manual contains a full chapter discussing the issue
9541 of elaboration checks, and if the default is not satisfactory for your use,
9542 you should read this chapter.
9544 For validity checks, the minimal checks required by the Ada Reference
9545 Manual (for case statements and assignments to array elements) are on
9546 by default. These can be suppressed by use of the @option{-gnatVn} switch.
9547 Note that in Ada 83, there were no validity checks, so if the Ada 83 mode
9548 is acceptable (or when comparing GNAT performance with an Ada 83 compiler),
9549 it may be reasonable to routinely use @option{-gnatVn}. Validity checks
9550 are also suppressed entirely if @option{-gnatp} is used.
9552 @cindex Overflow checks
9553 @cindex Checks, overflow
9556 @cindex pragma Suppress
9557 @cindex pragma Unsuppress
9558 Note that the setting of the switches controls the default setting of
9559 the checks. They may be modified using either @code{pragma Suppress} (to
9560 remove checks) or @code{pragma Unsuppress} (to add back suppressed
9561 checks) in the program source.
9563 @node Use of Restrictions
9564 @subsection Use of Restrictions
9567 The use of pragma Restrictions allows you to control which features are
9568 permitted in your program. Apart from the obvious point that if you avoid
9569 relatively expensive features like finalization (enforceable by the use
9570 of pragma Restrictions (No_Finalization), the use of this pragma does not
9571 affect the generated code in most cases.
9573 One notable exception to this rule is that the possibility of task abort
9574 results in some distributed overhead, particularly if finalization or
9575 exception handlers are used. The reason is that certain sections of code
9576 have to be marked as non-abortable.
9578 If you use neither the @code{abort} statement, nor asynchronous transfer
9579 of control (@code{select @dots{} then abort}), then this distributed overhead
9580 is removed, which may have a general positive effect in improving
9581 overall performance. Especially code involving frequent use of tasking
9582 constructs and controlled types will show much improved performance.
9583 The relevant restrictions pragmas are
9585 @smallexample @c ada
9586 pragma Restrictions (No_Abort_Statements);
9587 pragma Restrictions (Max_Asynchronous_Select_Nesting => 0);
9591 It is recommended that these restriction pragmas be used if possible. Note
9592 that this also means that you can write code without worrying about the
9593 possibility of an immediate abort at any point.
9595 @node Optimization Levels
9596 @subsection Optimization Levels
9597 @cindex @option{^-O^/OPTIMIZE^} (@command{gcc})
9600 Without any optimization ^option,^qualifier,^
9601 the compiler's goal is to reduce the cost of
9602 compilation and to make debugging produce the expected results.
9603 Statements are independent: if you stop the program with a breakpoint between
9604 statements, you can then assign a new value to any variable or change
9605 the program counter to any other statement in the subprogram and get exactly
9606 the results you would expect from the source code.
9608 Turning on optimization makes the compiler attempt to improve the
9609 performance and/or code size at the expense of compilation time and
9610 possibly the ability to debug the program.
9613 ^-O options, with or without level numbers,^/OPTIMIZE qualifiers,^
9614 the last such option is the one that is effective.
9617 The default is optimization off. This results in the fastest compile
9618 times, but GNAT makes absolutely no attempt to optimize, and the
9619 generated programs are considerably larger and slower than when
9620 optimization is enabled. You can use the
9622 @option{-O} switch (the permitted forms are @option{-O0}, @option{-O1}
9623 @option{-O2}, @option{-O3}, and @option{-Os})
9626 @code{OPTIMIZE} qualifier
9628 to @command{gcc} to control the optimization level:
9631 @item ^-O0^/OPTIMIZE=NONE^
9632 No optimization (the default);
9633 generates unoptimized code but has
9634 the fastest compilation time.
9636 Note that many other compilers do fairly extensive optimization
9637 even if ``no optimization'' is specified. With gcc, it is
9638 very unusual to use ^-O0^/OPTIMIZE=NONE^ for production if
9639 execution time is of any concern, since ^-O0^/OPTIMIZE=NONE^
9640 really does mean no optimization at all. This difference between
9641 gcc and other compilers should be kept in mind when doing
9642 performance comparisons.
9644 @item ^-O1^/OPTIMIZE=SOME^
9645 Moderate optimization;
9646 optimizes reasonably well but does not
9647 degrade compilation time significantly.
9649 @item ^-O2^/OPTIMIZE=ALL^
9651 @itemx /OPTIMIZE=DEVELOPMENT
9654 generates highly optimized code and has
9655 the slowest compilation time.
9657 @item ^-O3^/OPTIMIZE=INLINING^
9658 Full optimization as in @option{-O2},
9659 and also attempts automatic inlining of small
9660 subprograms within a unit (@pxref{Inlining of Subprograms}).
9662 @item ^-Os^/OPTIMIZE=SPACE^
9663 Optimize space usage of resulting program.
9667 Higher optimization levels perform more global transformations on the
9668 program and apply more expensive analysis algorithms in order to generate
9669 faster and more compact code. The price in compilation time, and the
9670 resulting improvement in execution time,
9671 both depend on the particular application and the hardware environment.
9672 You should experiment to find the best level for your application.
9674 Since the precise set of optimizations done at each level will vary from
9675 release to release (and sometime from target to target), it is best to think
9676 of the optimization settings in general terms.
9677 @xref{Optimize Options,, Options That Control Optimization, gcc, Using
9678 the GNU Compiler Collection (GCC)}, for details about
9679 ^the @option{-O} settings and a number of @option{-f} options that^how to^
9680 individually enable or disable specific optimizations.
9682 Unlike some other compilation systems, ^@command{gcc}^GNAT^ has
9683 been tested extensively at all optimization levels. There are some bugs
9684 which appear only with optimization turned on, but there have also been
9685 bugs which show up only in @emph{unoptimized} code. Selecting a lower
9686 level of optimization does not improve the reliability of the code
9687 generator, which in practice is highly reliable at all optimization
9690 Note regarding the use of @option{-O3}: The use of this optimization level
9691 is generally discouraged with GNAT, since it often results in larger
9692 executables which run more slowly. See further discussion of this point
9693 in @ref{Inlining of Subprograms}.
9695 @node Debugging Optimized Code
9696 @subsection Debugging Optimized Code
9697 @cindex Debugging optimized code
9698 @cindex Optimization and debugging
9701 Although it is possible to do a reasonable amount of debugging at
9703 nonzero optimization levels,
9704 the higher the level the more likely that
9707 @option{/OPTIMIZE} settings other than @code{NONE},
9708 such settings will make it more likely that
9710 source-level constructs will have been eliminated by optimization.
9711 For example, if a loop is strength-reduced, the loop
9712 control variable may be completely eliminated and thus cannot be
9713 displayed in the debugger.
9714 This can only happen at @option{-O2} or @option{-O3}.
9715 Explicit temporary variables that you code might be eliminated at
9716 ^level^setting^ @option{-O1} or higher.
9718 The use of the @option{^-g^/DEBUG^} switch,
9719 @cindex @option{^-g^/DEBUG^} (@command{gcc})
9720 which is needed for source-level debugging,
9721 affects the size of the program executable on disk,
9722 and indeed the debugging information can be quite large.
9723 However, it has no effect on the generated code (and thus does not
9724 degrade performance)
9726 Since the compiler generates debugging tables for a compilation unit before
9727 it performs optimizations, the optimizing transformations may invalidate some
9728 of the debugging data. You therefore need to anticipate certain
9729 anomalous situations that may arise while debugging optimized code.
9730 These are the most common cases:
9734 @i{The ``hopping Program Counter'':} Repeated @code{step} or @code{next}
9736 the PC bouncing back and forth in the code. This may result from any of
9737 the following optimizations:
9741 @i{Common subexpression elimination:} using a single instance of code for a
9742 quantity that the source computes several times. As a result you
9743 may not be able to stop on what looks like a statement.
9746 @i{Invariant code motion:} moving an expression that does not change within a
9747 loop, to the beginning of the loop.
9750 @i{Instruction scheduling:} moving instructions so as to
9751 overlap loads and stores (typically) with other code, or in
9752 general to move computations of values closer to their uses. Often
9753 this causes you to pass an assignment statement without the assignment
9754 happening and then later bounce back to the statement when the
9755 value is actually needed. Placing a breakpoint on a line of code
9756 and then stepping over it may, therefore, not always cause all the
9757 expected side-effects.
9761 @i{The ``big leap'':} More commonly known as @emph{cross-jumping}, in which
9762 two identical pieces of code are merged and the program counter suddenly
9763 jumps to a statement that is not supposed to be executed, simply because
9764 it (and the code following) translates to the same thing as the code
9765 that @emph{was} supposed to be executed. This effect is typically seen in
9766 sequences that end in a jump, such as a @code{goto}, a @code{return}, or
9767 a @code{break} in a C @code{^switch^switch^} statement.
9770 @i{The ``roving variable'':} The symptom is an unexpected value in a variable.
9771 There are various reasons for this effect:
9775 In a subprogram prologue, a parameter may not yet have been moved to its
9779 A variable may be dead, and its register re-used. This is
9780 probably the most common cause.
9783 As mentioned above, the assignment of a value to a variable may
9787 A variable may be eliminated entirely by value propagation or
9788 other means. In this case, GCC may incorrectly generate debugging
9789 information for the variable
9793 In general, when an unexpected value appears for a local variable or parameter
9794 you should first ascertain if that value was actually computed by
9795 your program, as opposed to being incorrectly reported by the debugger.
9797 array elements in an object designated by an access value
9798 are generally less of a problem, once you have ascertained that the access
9800 Typically, this means checking variables in the preceding code and in the
9801 calling subprogram to verify that the value observed is explainable from other
9802 values (one must apply the procedure recursively to those
9803 other values); or re-running the code and stopping a little earlier
9804 (perhaps before the call) and stepping to better see how the variable obtained
9805 the value in question; or continuing to step @emph{from} the point of the
9806 strange value to see if code motion had simply moved the variable's
9811 In light of such anomalies, a recommended technique is to use @option{-O0}
9812 early in the software development cycle, when extensive debugging capabilities
9813 are most needed, and then move to @option{-O1} and later @option{-O2} as
9814 the debugger becomes less critical.
9815 Whether to use the @option{^-g^/DEBUG^} switch in the release version is
9816 a release management issue.
9818 Note that if you use @option{-g} you can then use the @command{strip} program
9819 on the resulting executable,
9820 which removes both debugging information and global symbols.
9823 @node Inlining of Subprograms
9824 @subsection Inlining of Subprograms
9827 A call to a subprogram in the current unit is inlined if all the
9828 following conditions are met:
9832 The optimization level is at least @option{-O1}.
9835 The called subprogram is suitable for inlining: It must be small enough
9836 and not contain something that @command{gcc} cannot support in inlined
9840 @cindex pragma Inline
9842 Either @code{pragma Inline} applies to the subprogram, or it is local
9843 to the unit and called once from within it, or it is small and automatic
9844 inlining (optimization level @option{-O3}) is specified.
9848 Calls to subprograms in @code{with}'ed units are normally not inlined.
9849 To achieve actual inlining (that is, replacement of the call by the code
9850 in the body of the subprogram), the following conditions must all be true.
9854 The optimization level is at least @option{-O1}.
9857 The called subprogram is suitable for inlining: It must be small enough
9858 and not contain something that @command{gcc} cannot support in inlined
9862 The call appears in a body (not in a package spec).
9865 There is a @code{pragma Inline} for the subprogram.
9868 @cindex @option{-gnatn} (@command{gcc})
9869 The @option{^-gnatn^/INLINE^} switch
9870 is used in the @command{gcc} command line
9873 Even if all these conditions are met, it may not be possible for
9874 the compiler to inline the call, due to the length of the body,
9875 or features in the body that make it impossible for the compiler
9878 Note that specifying the @option{-gnatn} switch causes additional
9879 compilation dependencies. Consider the following:
9881 @smallexample @c ada
9901 With the default behavior (no @option{-gnatn} switch specified), the
9902 compilation of the @code{Main} procedure depends only on its own source,
9903 @file{main.adb}, and the spec of the package in file @file{r.ads}. This
9904 means that editing the body of @code{R} does not require recompiling
9907 On the other hand, the call @code{R.Q} is not inlined under these
9908 circumstances. If the @option{-gnatn} switch is present when @code{Main}
9909 is compiled, the call will be inlined if the body of @code{Q} is small
9910 enough, but now @code{Main} depends on the body of @code{R} in
9911 @file{r.adb} as well as on the spec. This means that if this body is edited,
9912 the main program must be recompiled. Note that this extra dependency
9913 occurs whether or not the call is in fact inlined by @command{gcc}.
9915 The use of front end inlining with @option{-gnatN} generates similar
9916 additional dependencies.
9918 @cindex @option{^-fno-inline^/INLINE=SUPPRESS^} (@command{gcc})
9919 Note: The @option{^-fno-inline^/INLINE=SUPPRESS^} switch
9920 can be used to prevent
9921 all inlining. This switch overrides all other conditions and ensures
9922 that no inlining occurs. The extra dependences resulting from
9923 @option{-gnatn} will still be active, even if
9924 this switch is used to suppress the resulting inlining actions.
9926 @cindex @option{-fno-inline-functions} (@command{gcc})
9927 Note: The @option{-fno-inline-functions} switch can be used to prevent
9928 automatic inlining of small subprograms if @option{-O3} is used.
9930 @cindex @option{-fno-inline-functions-called-once} (@command{gcc})
9931 Note: The @option{-fno-inline-functions-called-once} switch
9932 can be used to prevent inlining of subprograms local to the unit
9933 and called once from within it if @option{-O1} is used.
9935 Note regarding the use of @option{-O3}: There is no difference in inlining
9936 behavior between @option{-O2} and @option{-O3} for subprograms with an explicit
9937 pragma @code{Inline} assuming the use of @option{-gnatn}
9938 or @option{-gnatN} (the switches that activate inlining). If you have used
9939 pragma @code{Inline} in appropriate cases, then it is usually much better
9940 to use @option{-O2} and @option{-gnatn} and avoid the use of @option{-O3} which
9941 in this case only has the effect of inlining subprograms you did not
9942 think should be inlined. We often find that the use of @option{-O3} slows
9943 down code by performing excessive inlining, leading to increased instruction
9944 cache pressure from the increased code size. So the bottom line here is
9945 that you should not automatically assume that @option{-O3} is better than
9946 @option{-O2}, and indeed you should use @option{-O3} only if tests show that
9947 it actually improves performance.
9949 @node Other Optimization Switches
9950 @subsection Other Optimization Switches
9951 @cindex Optimization Switches
9953 Since @code{GNAT} uses the @command{gcc} back end, all the specialized
9954 @command{gcc} optimization switches are potentially usable. These switches
9955 have not been extensively tested with GNAT but can generally be expected
9956 to work. Examples of switches in this category are
9957 @option{-funroll-loops} and
9958 the various target-specific @option{-m} options (in particular, it has been
9959 observed that @option{-march=pentium4} can significantly improve performance
9960 on appropriate machines). For full details of these switches, see
9961 @ref{Submodel Options,, Hardware Models and Configurations, gcc, Using
9962 the GNU Compiler Collection (GCC)}.
9964 @node Optimization and Strict Aliasing
9965 @subsection Optimization and Strict Aliasing
9967 @cindex Strict Aliasing
9968 @cindex No_Strict_Aliasing
9971 The strong typing capabilities of Ada allow an optimizer to generate
9972 efficient code in situations where other languages would be forced to
9973 make worst case assumptions preventing such optimizations. Consider
9974 the following example:
9976 @smallexample @c ada
9979 type Int1 is new Integer;
9980 type Int2 is new Integer;
9981 type Int1A is access Int1;
9982 type Int2A is access Int2;
9989 for J in Data'Range loop
9990 if Data (J) = Int1V.all then
9991 Int2V.all := Int2V.all + 1;
10000 In this example, since the variable @code{Int1V} can only access objects
10001 of type @code{Int1}, and @code{Int2V} can only access objects of type
10002 @code{Int2}, there is no possibility that the assignment to
10003 @code{Int2V.all} affects the value of @code{Int1V.all}. This means that
10004 the compiler optimizer can "know" that the value @code{Int1V.all} is constant
10005 for all iterations of the loop and avoid the extra memory reference
10006 required to dereference it each time through the loop.
10008 This kind of optimization, called strict aliasing analysis, is
10009 triggered by specifying an optimization level of @option{-O2} or
10010 higher and allows @code{GNAT} to generate more efficient code
10011 when access values are involved.
10013 However, although this optimization is always correct in terms of
10014 the formal semantics of the Ada Reference Manual, difficulties can
10015 arise if features like @code{Unchecked_Conversion} are used to break
10016 the typing system. Consider the following complete program example:
10018 @smallexample @c ada
10021 type int1 is new integer;
10022 type int2 is new integer;
10023 type a1 is access int1;
10024 type a2 is access int2;
10029 function to_a2 (Input : a1) return a2;
10032 with Unchecked_Conversion;
10034 function to_a2 (Input : a1) return a2 is
10036 new Unchecked_Conversion (a1, a2);
10038 return to_a2u (Input);
10044 with Text_IO; use Text_IO;
10046 v1 : a1 := new int1;
10047 v2 : a2 := to_a2 (v1);
10051 put_line (int1'image (v1.all));
10057 This program prints out 0 in @option{-O0} or @option{-O1}
10058 mode, but it prints out 1 in @option{-O2} mode. That's
10059 because in strict aliasing mode, the compiler can and
10060 does assume that the assignment to @code{v2.all} could not
10061 affect the value of @code{v1.all}, since different types
10064 This behavior is not a case of non-conformance with the standard, since
10065 the Ada RM specifies that an unchecked conversion where the resulting
10066 bit pattern is not a correct value of the target type can result in an
10067 abnormal value and attempting to reference an abnormal value makes the
10068 execution of a program erroneous. That's the case here since the result
10069 does not point to an object of type @code{int2}. This means that the
10070 effect is entirely unpredictable.
10072 However, although that explanation may satisfy a language
10073 lawyer, in practice an applications programmer expects an
10074 unchecked conversion involving pointers to create true
10075 aliases and the behavior of printing 1 seems plain wrong.
10076 In this case, the strict aliasing optimization is unwelcome.
10078 Indeed the compiler recognizes this possibility, and the
10079 unchecked conversion generates a warning:
10082 p2.adb:5:07: warning: possible aliasing problem with type "a2"
10083 p2.adb:5:07: warning: use -fno-strict-aliasing switch for references
10084 p2.adb:5:07: warning: or use "pragma No_Strict_Aliasing (a2);"
10088 Unfortunately the problem is recognized when compiling the body of
10089 package @code{p2}, but the actual "bad" code is generated while
10090 compiling the body of @code{m} and this latter compilation does not see
10091 the suspicious @code{Unchecked_Conversion}.
10093 As implied by the warning message, there are approaches you can use to
10094 avoid the unwanted strict aliasing optimization in a case like this.
10096 One possibility is to simply avoid the use of @option{-O2}, but
10097 that is a bit drastic, since it throws away a number of useful
10098 optimizations that do not involve strict aliasing assumptions.
10100 A less drastic approach is to compile the program using the
10101 option @option{-fno-strict-aliasing}. Actually it is only the
10102 unit containing the dereferencing of the suspicious pointer
10103 that needs to be compiled. So in this case, if we compile
10104 unit @code{m} with this switch, then we get the expected
10105 value of zero printed. Analyzing which units might need
10106 the switch can be painful, so a more reasonable approach
10107 is to compile the entire program with options @option{-O2}
10108 and @option{-fno-strict-aliasing}. If the performance is
10109 satisfactory with this combination of options, then the
10110 advantage is that the entire issue of possible "wrong"
10111 optimization due to strict aliasing is avoided.
10113 To avoid the use of compiler switches, the configuration
10114 pragma @code{No_Strict_Aliasing} with no parameters may be
10115 used to specify that for all access types, the strict
10116 aliasing optimization should be suppressed.
10118 However, these approaches are still overkill, in that they causes
10119 all manipulations of all access values to be deoptimized. A more
10120 refined approach is to concentrate attention on the specific
10121 access type identified as problematic.
10123 First, if a careful analysis of uses of the pointer shows
10124 that there are no possible problematic references, then
10125 the warning can be suppressed by bracketing the
10126 instantiation of @code{Unchecked_Conversion} to turn
10129 @smallexample @c ada
10130 pragma Warnings (Off);
10132 new Unchecked_Conversion (a1, a2);
10133 pragma Warnings (On);
10137 Of course that approach is not appropriate for this particular
10138 example, since indeed there is a problematic reference. In this
10139 case we can take one of two other approaches.
10141 The first possibility is to move the instantiation of unchecked
10142 conversion to the unit in which the type is declared. In
10143 this example, we would move the instantiation of
10144 @code{Unchecked_Conversion} from the body of package
10145 @code{p2} to the spec of package @code{p1}. Now the
10146 warning disappears. That's because any use of the
10147 access type knows there is a suspicious unchecked
10148 conversion, and the strict aliasing optimization
10149 is automatically suppressed for the type.
10151 If it is not practical to move the unchecked conversion to the same unit
10152 in which the destination access type is declared (perhaps because the
10153 source type is not visible in that unit), you may use pragma
10154 @code{No_Strict_Aliasing} for the type. This pragma must occur in the
10155 same declarative sequence as the declaration of the access type:
10157 @smallexample @c ada
10158 type a2 is access int2;
10159 pragma No_Strict_Aliasing (a2);
10163 Here again, the compiler now knows that the strict aliasing optimization
10164 should be suppressed for any reference to type @code{a2} and the
10165 expected behavior is obtained.
10167 Finally, note that although the compiler can generate warnings for
10168 simple cases of unchecked conversions, there are tricker and more
10169 indirect ways of creating type incorrect aliases which the compiler
10170 cannot detect. Examples are the use of address overlays and unchecked
10171 conversions involving composite types containing access types as
10172 components. In such cases, no warnings are generated, but there can
10173 still be aliasing problems. One safe coding practice is to forbid the
10174 use of address clauses for type overlaying, and to allow unchecked
10175 conversion only for primitive types. This is not really a significant
10176 restriction since any possible desired effect can be achieved by
10177 unchecked conversion of access values.
10180 @node Coverage Analysis
10181 @subsection Coverage Analysis
10184 GNAT supports the HP Performance Coverage Analyzer (PCA), which allows
10185 the user to determine the distribution of execution time across a program,
10186 @pxref{Profiling} for details of usage.
10190 @node Text_IO Suggestions
10191 @section @code{Text_IO} Suggestions
10192 @cindex @code{Text_IO} and performance
10195 The @code{Ada.Text_IO} package has fairly high overheads due in part to
10196 the requirement of maintaining page and line counts. If performance
10197 is critical, a recommendation is to use @code{Stream_IO} instead of
10198 @code{Text_IO} for volume output, since this package has less overhead.
10200 If @code{Text_IO} must be used, note that by default output to the standard
10201 output and standard error files is unbuffered (this provides better
10202 behavior when output statements are used for debugging, or if the
10203 progress of a program is observed by tracking the output, e.g. by
10204 using the Unix @command{tail -f} command to watch redirected output.
10206 If you are generating large volumes of output with @code{Text_IO} and
10207 performance is an important factor, use a designated file instead
10208 of the standard output file, or change the standard output file to
10209 be buffered using @code{Interfaces.C_Streams.setvbuf}.
10213 @node Reducing Size of Ada Executables with gnatelim
10214 @section Reducing Size of Ada Executables with @code{gnatelim}
10218 This section describes @command{gnatelim}, a tool which detects unused
10219 subprograms and helps the compiler to create a smaller executable for your
10224 * Running gnatelim::
10225 * Correcting the List of Eliminate Pragmas::
10226 * Making Your Executables Smaller::
10227 * Summary of the gnatelim Usage Cycle::
10230 @node About gnatelim
10231 @subsection About @code{gnatelim}
10234 When a program shares a set of Ada
10235 packages with other programs, it may happen that this program uses
10236 only a fraction of the subprograms defined in these packages. The code
10237 created for these unused subprograms increases the size of the executable.
10239 @code{gnatelim} tracks unused subprograms in an Ada program and
10240 outputs a list of GNAT-specific pragmas @code{Eliminate} marking all the
10241 subprograms that are declared but never called. By placing the list of
10242 @code{Eliminate} pragmas in the GNAT configuration file @file{gnat.adc} and
10243 recompiling your program, you may decrease the size of its executable,
10244 because the compiler will not generate the code for 'eliminated' subprograms.
10245 @xref{Pragma Eliminate,,, gnat_rm, GNAT Reference Manual}, for more
10246 information about this pragma.
10248 @code{gnatelim} needs as its input data the name of the main subprogram
10249 and a bind file for a main subprogram.
10251 To create a bind file for @code{gnatelim}, run @code{gnatbind} for
10252 the main subprogram. @code{gnatelim} can work with both Ada and C
10253 bind files; when both are present, it uses the Ada bind file.
10254 The following commands will build the program and create the bind file:
10257 $ gnatmake ^-c Main_Prog^/ACTIONS=COMPILE MAIN_PROG^
10258 $ gnatbind main_prog
10261 Note that @code{gnatelim} needs neither object nor ALI files.
10263 @node Running gnatelim
10264 @subsection Running @code{gnatelim}
10267 @code{gnatelim} has the following command-line interface:
10270 $ gnatelim @ovar{options} name
10274 @code{name} should be a name of a source file that contains the main subprogram
10275 of a program (partition).
10277 @code{gnatelim} has the following switches:
10282 @cindex @option{^-q^/QUIET^} (@command{gnatelim})
10283 Quiet mode: by default @code{gnatelim} outputs to the standard error
10284 stream the number of program units left to be processed. This option turns
10287 @item ^-v^/VERBOSE^
10288 @cindex @option{^-v^/VERBOSE^} (@command{gnatelim})
10289 Verbose mode: @code{gnatelim} version information is printed as Ada
10290 comments to the standard output stream. Also, in addition to the number of
10291 program units left @code{gnatelim} will output the name of the current unit
10295 @cindex @option{^-a^/ALL^} (@command{gnatelim})
10296 Also look for subprograms from the GNAT run time that can be eliminated. Note
10297 that when @file{gnat.adc} is produced using this switch, the entire program
10298 must be recompiled with switch @option{^-a^/ALL_FILES^} to @command{gnatmake}.
10300 @item ^-I^/INCLUDE_DIRS=^@var{dir}
10301 @cindex @option{^-I^/INCLUDE_DIRS^} (@command{gnatelim})
10302 When looking for source files also look in directory @var{dir}. Specifying
10303 @option{^-I-^/INCLUDE_DIRS=-^} instructs @code{gnatelim} not to look for
10304 sources in the current directory.
10306 @item ^-b^/BIND_FILE=^@var{bind_file}
10307 @cindex @option{^-b^/BIND_FILE^} (@command{gnatelim})
10308 Specifies @var{bind_file} as the bind file to process. If not set, the name
10309 of the bind file is computed from the full expanded Ada name
10310 of a main subprogram.
10312 @item ^-C^/CONFIG_FILE=^@var{config_file}
10313 @cindex @option{^-C^/CONFIG_FILE^} (@command{gnatelim})
10314 Specifies a file @var{config_file} that contains configuration pragmas. The
10315 file must be specified with full path.
10317 @item ^--GCC^/COMPILER^=@var{compiler_name}
10318 @cindex @option{^-GCC^/COMPILER^} (@command{gnatelim})
10319 Instructs @code{gnatelim} to use specific @command{gcc} compiler instead of one
10320 available on the path.
10322 @item ^--GNATMAKE^/GNATMAKE^=@var{gnatmake_name}
10323 @cindex @option{^--GNATMAKE^/GNATMAKE^} (@command{gnatelim})
10324 Instructs @code{gnatelim} to use specific @command{gnatmake} instead of one
10325 available on the path.
10329 @code{gnatelim} sends its output to the standard output stream, and all the
10330 tracing and debug information is sent to the standard error stream.
10331 In order to produce a proper GNAT configuration file
10332 @file{gnat.adc}, redirection must be used:
10336 $ PIPE GNAT ELIM MAIN_PROG.ADB > GNAT.ADC
10339 $ gnatelim main_prog.adb > gnat.adc
10348 $ gnatelim main_prog.adb >> gnat.adc
10352 in order to append the @code{gnatelim} output to the existing contents of
10356 @node Correcting the List of Eliminate Pragmas
10357 @subsection Correcting the List of Eliminate Pragmas
10360 In some rare cases @code{gnatelim} may try to eliminate
10361 subprograms that are actually called in the program. In this case, the
10362 compiler will generate an error message of the form:
10365 file.adb:106:07: cannot call eliminated subprogram "My_Prog"
10369 You will need to manually remove the wrong @code{Eliminate} pragmas from
10370 the @file{gnat.adc} file. You should recompile your program
10371 from scratch after that, because you need a consistent @file{gnat.adc} file
10372 during the entire compilation.
10374 @node Making Your Executables Smaller
10375 @subsection Making Your Executables Smaller
10378 In order to get a smaller executable for your program you now have to
10379 recompile the program completely with the new @file{gnat.adc} file
10380 created by @code{gnatelim} in your current directory:
10383 $ gnatmake ^-f main_prog^/FORCE_COMPILE MAIN_PROG^
10387 (Use the @option{^-f^/FORCE_COMPILE^} option for @command{gnatmake} to
10388 recompile everything
10389 with the set of pragmas @code{Eliminate} that you have obtained with
10390 @command{gnatelim}).
10392 Be aware that the set of @code{Eliminate} pragmas is specific to each
10393 program. It is not recommended to merge sets of @code{Eliminate}
10394 pragmas created for different programs in one @file{gnat.adc} file.
10396 @node Summary of the gnatelim Usage Cycle
10397 @subsection Summary of the gnatelim Usage Cycle
10400 Here is a quick summary of the steps to be taken in order to reduce
10401 the size of your executables with @code{gnatelim}. You may use
10402 other GNAT options to control the optimization level,
10403 to produce the debugging information, to set search path, etc.
10407 Produce a bind file
10410 $ gnatmake ^-c main_prog^/ACTIONS=COMPILE MAIN_PROG^
10411 $ gnatbind main_prog
10415 Generate a list of @code{Eliminate} pragmas
10418 $ PIPE GNAT ELIM MAIN_PROG > GNAT.ADC
10421 $ gnatelim main_prog >@r{[}>@r{]} gnat.adc
10426 Recompile the application
10429 $ gnatmake ^-f main_prog^/FORCE_COMPILE MAIN_PROG^
10434 @node Reducing Size of Executables with unused subprogram/data elimination
10435 @section Reducing Size of Executables with Unused Subprogram/Data Elimination
10436 @findex unused subprogram/data elimination
10439 This section describes how you can eliminate unused subprograms and data from
10440 your executable just by setting options at compilation time.
10443 * About unused subprogram/data elimination::
10444 * Compilation options::
10445 * Example of unused subprogram/data elimination::
10448 @node About unused subprogram/data elimination
10449 @subsection About unused subprogram/data elimination
10452 By default, an executable contains all code and data of its composing objects
10453 (directly linked or coming from statically linked libraries), even data or code
10454 never used by this executable.
10456 This feature will allow you to eliminate such unused code from your
10457 executable, making it smaller (in disk and in memory).
10459 This functionality is available on all Linux platforms except for the IA-64
10460 architecture and on all cross platforms using the ELF binary file format.
10461 In both cases GNU binutils version 2.16 or later are required to enable it.
10463 @node Compilation options
10464 @subsection Compilation options
10467 The operation of eliminating the unused code and data from the final executable
10468 is directly performed by the linker.
10470 In order to do this, it has to work with objects compiled with the
10472 @option{-ffunction-sections} @option{-fdata-sections}.
10473 @cindex @option{-ffunction-sections} (@command{gcc})
10474 @cindex @option{-fdata-sections} (@command{gcc})
10475 These options are usable with C and Ada files.
10476 They will place respectively each
10477 function or data in a separate section in the resulting object file.
10479 Once the objects and static libraries are created with these options, the
10480 linker can perform the dead code elimination. You can do this by setting
10481 the @option{-Wl,--gc-sections} option to gcc command or in the
10482 @option{-largs} section of @command{gnatmake}. This will perform a
10483 garbage collection of code and data never referenced.
10485 If the linker performs a partial link (@option{-r} ld linker option), then you
10486 will need to provide one or several entry point using the
10487 @option{-e} / @option{--entry} ld option.
10489 Note that objects compiled without the @option{-ffunction-sections} and
10490 @option{-fdata-sections} options can still be linked with the executable.
10491 However, no dead code elimination will be performed on those objects (they will
10494 The GNAT static library is now compiled with -ffunction-sections and
10495 -fdata-sections on some platforms. This allows you to eliminate the unused code
10496 and data of the GNAT library from your executable.
10498 @node Example of unused subprogram/data elimination
10499 @subsection Example of unused subprogram/data elimination
10502 Here is a simple example:
10504 @smallexample @c ada
10513 Used_Data : Integer;
10514 Unused_Data : Integer;
10516 procedure Used (Data : Integer);
10517 procedure Unused (Data : Integer);
10520 package body Aux is
10521 procedure Used (Data : Integer) is
10526 procedure Unused (Data : Integer) is
10528 Unused_Data := Data;
10534 @code{Unused} and @code{Unused_Data} are never referenced in this code
10535 excerpt, and hence they may be safely removed from the final executable.
10540 $ nm test | grep used
10541 020015f0 T aux__unused
10542 02005d88 B aux__unused_data
10543 020015cc T aux__used
10544 02005d84 B aux__used_data
10546 $ gnatmake test -cargs -fdata-sections -ffunction-sections \
10547 -largs -Wl,--gc-sections
10549 $ nm test | grep used
10550 02005350 T aux__used
10551 0201ffe0 B aux__used_data
10555 It can be observed that the procedure @code{Unused} and the object
10556 @code{Unused_Data} are removed by the linker when using the
10557 appropriate options.
10559 @c ********************************
10560 @node Renaming Files Using gnatchop
10561 @chapter Renaming Files Using @code{gnatchop}
10565 This chapter discusses how to handle files with multiple units by using
10566 the @code{gnatchop} utility. This utility is also useful in renaming
10567 files to meet the standard GNAT default file naming conventions.
10570 * Handling Files with Multiple Units::
10571 * Operating gnatchop in Compilation Mode::
10572 * Command Line for gnatchop::
10573 * Switches for gnatchop::
10574 * Examples of gnatchop Usage::
10577 @node Handling Files with Multiple Units
10578 @section Handling Files with Multiple Units
10581 The basic compilation model of GNAT requires that a file submitted to the
10582 compiler have only one unit and there be a strict correspondence
10583 between the file name and the unit name.
10585 The @code{gnatchop} utility allows both of these rules to be relaxed,
10586 allowing GNAT to process files which contain multiple compilation units
10587 and files with arbitrary file names. @code{gnatchop}
10588 reads the specified file and generates one or more output files,
10589 containing one unit per file. The unit and the file name correspond,
10590 as required by GNAT.
10592 If you want to permanently restructure a set of ``foreign'' files so that
10593 they match the GNAT rules, and do the remaining development using the
10594 GNAT structure, you can simply use @command{gnatchop} once, generate the
10595 new set of files and work with them from that point on.
10597 Alternatively, if you want to keep your files in the ``foreign'' format,
10598 perhaps to maintain compatibility with some other Ada compilation
10599 system, you can set up a procedure where you use @command{gnatchop} each
10600 time you compile, regarding the source files that it writes as temporary
10601 files that you throw away.
10603 @node Operating gnatchop in Compilation Mode
10604 @section Operating gnatchop in Compilation Mode
10607 The basic function of @code{gnatchop} is to take a file with multiple units
10608 and split it into separate files. The boundary between files is reasonably
10609 clear, except for the issue of comments and pragmas. In default mode, the
10610 rule is that any pragmas between units belong to the previous unit, except
10611 that configuration pragmas always belong to the following unit. Any comments
10612 belong to the following unit. These rules
10613 almost always result in the right choice of
10614 the split point without needing to mark it explicitly and most users will
10615 find this default to be what they want. In this default mode it is incorrect to
10616 submit a file containing only configuration pragmas, or one that ends in
10617 configuration pragmas, to @code{gnatchop}.
10619 However, using a special option to activate ``compilation mode'',
10621 can perform another function, which is to provide exactly the semantics
10622 required by the RM for handling of configuration pragmas in a compilation.
10623 In the absence of configuration pragmas (at the main file level), this
10624 option has no effect, but it causes such configuration pragmas to be handled
10625 in a quite different manner.
10627 First, in compilation mode, if @code{gnatchop} is given a file that consists of
10628 only configuration pragmas, then this file is appended to the
10629 @file{gnat.adc} file in the current directory. This behavior provides
10630 the required behavior described in the RM for the actions to be taken
10631 on submitting such a file to the compiler, namely that these pragmas
10632 should apply to all subsequent compilations in the same compilation
10633 environment. Using GNAT, the current directory, possibly containing a
10634 @file{gnat.adc} file is the representation
10635 of a compilation environment. For more information on the
10636 @file{gnat.adc} file, see @ref{Handling of Configuration Pragmas}.
10638 Second, in compilation mode, if @code{gnatchop}
10639 is given a file that starts with
10640 configuration pragmas, and contains one or more units, then these
10641 configuration pragmas are prepended to each of the chopped files. This
10642 behavior provides the required behavior described in the RM for the
10643 actions to be taken on compiling such a file, namely that the pragmas
10644 apply to all units in the compilation, but not to subsequently compiled
10647 Finally, if configuration pragmas appear between units, they are appended
10648 to the previous unit. This results in the previous unit being illegal,
10649 since the compiler does not accept configuration pragmas that follow
10650 a unit. This provides the required RM behavior that forbids configuration
10651 pragmas other than those preceding the first compilation unit of a
10654 For most purposes, @code{gnatchop} will be used in default mode. The
10655 compilation mode described above is used only if you need exactly
10656 accurate behavior with respect to compilations, and you have files
10657 that contain multiple units and configuration pragmas. In this
10658 circumstance the use of @code{gnatchop} with the compilation mode
10659 switch provides the required behavior, and is for example the mode
10660 in which GNAT processes the ACVC tests.
10662 @node Command Line for gnatchop
10663 @section Command Line for @code{gnatchop}
10666 The @code{gnatchop} command has the form:
10669 $ gnatchop switches @var{file name} @r{[}@var{file name} @dots{}@r{]}
10674 The only required argument is the file name of the file to be chopped.
10675 There are no restrictions on the form of this file name. The file itself
10676 contains one or more Ada units, in normal GNAT format, concatenated
10677 together. As shown, more than one file may be presented to be chopped.
10679 When run in default mode, @code{gnatchop} generates one output file in
10680 the current directory for each unit in each of the files.
10682 @var{directory}, if specified, gives the name of the directory to which
10683 the output files will be written. If it is not specified, all files are
10684 written to the current directory.
10686 For example, given a
10687 file called @file{hellofiles} containing
10689 @smallexample @c ada
10694 with Text_IO; use Text_IO;
10697 Put_Line ("Hello");
10707 $ gnatchop ^hellofiles^HELLOFILES.^
10711 generates two files in the current directory, one called
10712 @file{hello.ads} containing the single line that is the procedure spec,
10713 and the other called @file{hello.adb} containing the remaining text. The
10714 original file is not affected. The generated files can be compiled in
10718 When gnatchop is invoked on a file that is empty or that contains only empty
10719 lines and/or comments, gnatchop will not fail, but will not produce any
10722 For example, given a
10723 file called @file{toto.txt} containing
10725 @smallexample @c ada
10737 $ gnatchop ^toto.txt^TOT.TXT^
10741 will not produce any new file and will result in the following warnings:
10744 toto.txt:1:01: warning: empty file, contains no compilation units
10745 no compilation units found
10746 no source files written
10749 @node Switches for gnatchop
10750 @section Switches for @code{gnatchop}
10753 @command{gnatchop} recognizes the following switches:
10759 @cindex @option{--version} @command{gnatchop}
10760 Display Copyright and version, then exit disregarding all other options.
10763 @cindex @option{--help} @command{gnatchop}
10764 If @option{--version} was not used, display usage, then exit disregarding
10767 @item ^-c^/COMPILATION^
10768 @cindex @option{^-c^/COMPILATION^} (@code{gnatchop})
10769 Causes @code{gnatchop} to operate in compilation mode, in which
10770 configuration pragmas are handled according to strict RM rules. See
10771 previous section for a full description of this mode.
10774 @item -gnat@var{xxx}
10775 This passes the given @option{-gnat@var{xxx}} switch to @code{gnat} which is
10776 used to parse the given file. Not all @var{xxx} options make sense,
10777 but for example, the use of @option{-gnati2} allows @code{gnatchop} to
10778 process a source file that uses Latin-2 coding for identifiers.
10782 Causes @code{gnatchop} to generate a brief help summary to the standard
10783 output file showing usage information.
10785 @item ^-k@var{mm}^/FILE_NAME_MAX_LENGTH=@var{mm}^
10786 @cindex @option{^-k^/FILE_NAME_MAX_LENGTH^} (@code{gnatchop})
10787 Limit generated file names to the specified number @code{mm}
10789 This is useful if the
10790 resulting set of files is required to be interoperable with systems
10791 which limit the length of file names.
10793 If no value is given, or
10794 if no @code{/FILE_NAME_MAX_LENGTH} qualifier is given,
10795 a default of 39, suitable for OpenVMS Alpha
10796 Systems, is assumed
10799 No space is allowed between the @option{-k} and the numeric value. The numeric
10800 value may be omitted in which case a default of @option{-k8},
10802 with DOS-like file systems, is used. If no @option{-k} switch
10804 there is no limit on the length of file names.
10807 @item ^-p^/PRESERVE^
10808 @cindex @option{^-p^/PRESERVE^} (@code{gnatchop})
10809 Causes the file ^modification^creation^ time stamp of the input file to be
10810 preserved and used for the time stamp of the output file(s). This may be
10811 useful for preserving coherency of time stamps in an environment where
10812 @code{gnatchop} is used as part of a standard build process.
10815 @cindex @option{^-q^/QUIET^} (@code{gnatchop})
10816 Causes output of informational messages indicating the set of generated
10817 files to be suppressed. Warnings and error messages are unaffected.
10819 @item ^-r^/REFERENCE^
10820 @cindex @option{^-r^/REFERENCE^} (@code{gnatchop})
10821 @findex Source_Reference
10822 Generate @code{Source_Reference} pragmas. Use this switch if the output
10823 files are regarded as temporary and development is to be done in terms
10824 of the original unchopped file. This switch causes
10825 @code{Source_Reference} pragmas to be inserted into each of the
10826 generated files to refers back to the original file name and line number.
10827 The result is that all error messages refer back to the original
10829 In addition, the debugging information placed into the object file (when
10830 the @option{^-g^/DEBUG^} switch of @command{gcc} or @command{gnatmake} is
10832 also refers back to this original file so that tools like profilers and
10833 debuggers will give information in terms of the original unchopped file.
10835 If the original file to be chopped itself contains
10836 a @code{Source_Reference}
10837 pragma referencing a third file, then gnatchop respects
10838 this pragma, and the generated @code{Source_Reference} pragmas
10839 in the chopped file refer to the original file, with appropriate
10840 line numbers. This is particularly useful when @code{gnatchop}
10841 is used in conjunction with @code{gnatprep} to compile files that
10842 contain preprocessing statements and multiple units.
10844 @item ^-v^/VERBOSE^
10845 @cindex @option{^-v^/VERBOSE^} (@code{gnatchop})
10846 Causes @code{gnatchop} to operate in verbose mode. The version
10847 number and copyright notice are output, as well as exact copies of
10848 the gnat1 commands spawned to obtain the chop control information.
10850 @item ^-w^/OVERWRITE^
10851 @cindex @option{^-w^/OVERWRITE^} (@code{gnatchop})
10852 Overwrite existing file names. Normally @code{gnatchop} regards it as a
10853 fatal error if there is already a file with the same name as a
10854 file it would otherwise output, in other words if the files to be
10855 chopped contain duplicated units. This switch bypasses this
10856 check, and causes all but the last instance of such duplicated
10857 units to be skipped.
10860 @item --GCC=@var{xxxx}
10861 @cindex @option{--GCC=} (@code{gnatchop})
10862 Specify the path of the GNAT parser to be used. When this switch is used,
10863 no attempt is made to add the prefix to the GNAT parser executable.
10867 @node Examples of gnatchop Usage
10868 @section Examples of @code{gnatchop} Usage
10872 @item gnatchop /OVERWRITE HELLO_S.ADA [PRERELEASE.FILES]
10875 @item gnatchop -w hello_s.ada prerelease/files
10878 Chops the source file @file{hello_s.ada}. The output files will be
10879 placed in the directory @file{^prerelease/files^[PRERELEASE.FILES]^},
10881 files with matching names in that directory (no files in the current
10882 directory are modified).
10884 @item gnatchop ^archive^ARCHIVE.^
10885 Chops the source file @file{^archive^ARCHIVE.^}
10886 into the current directory. One
10887 useful application of @code{gnatchop} is in sending sets of sources
10888 around, for example in email messages. The required sources are simply
10889 concatenated (for example, using a ^Unix @code{cat}^VMS @code{APPEND/NEW}^
10891 @command{gnatchop} is used at the other end to reconstitute the original
10894 @item gnatchop file1 file2 file3 direc
10895 Chops all units in files @file{file1}, @file{file2}, @file{file3}, placing
10896 the resulting files in the directory @file{direc}. Note that if any units
10897 occur more than once anywhere within this set of files, an error message
10898 is generated, and no files are written. To override this check, use the
10899 @option{^-w^/OVERWRITE^} switch,
10900 in which case the last occurrence in the last file will
10901 be the one that is output, and earlier duplicate occurrences for a given
10902 unit will be skipped.
10905 @node Configuration Pragmas
10906 @chapter Configuration Pragmas
10907 @cindex Configuration pragmas
10908 @cindex Pragmas, configuration
10911 Configuration pragmas include those pragmas described as
10912 such in the Ada Reference Manual, as well as
10913 implementation-dependent pragmas that are configuration pragmas.
10914 @xref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference Manual},
10915 for details on these additional GNAT-specific configuration pragmas.
10916 Most notably, the pragma @code{Source_File_Name}, which allows
10917 specifying non-default names for source files, is a configuration
10918 pragma. The following is a complete list of configuration pragmas
10919 recognized by GNAT:
10931 Compile_Time_Warning
10933 Component_Alignment
10940 External_Name_Casing
10943 Float_Representation
10956 Priority_Specific_Dispatching
10959 Propagate_Exceptions
10962 Restricted_Run_Time
10964 Restrictions_Warnings
10967 Source_File_Name_Project
10970 Suppress_Exception_Locations
10971 Task_Dispatching_Policy
10977 Wide_Character_Encoding
10982 * Handling of Configuration Pragmas::
10983 * The Configuration Pragmas Files::
10986 @node Handling of Configuration Pragmas
10987 @section Handling of Configuration Pragmas
10989 Configuration pragmas may either appear at the start of a compilation
10990 unit, in which case they apply only to that unit, or they may apply to
10991 all compilations performed in a given compilation environment.
10993 GNAT also provides the @code{gnatchop} utility to provide an automatic
10994 way to handle configuration pragmas following the semantics for
10995 compilations (that is, files with multiple units), described in the RM.
10996 See @ref{Operating gnatchop in Compilation Mode} for details.
10997 However, for most purposes, it will be more convenient to edit the
10998 @file{gnat.adc} file that contains configuration pragmas directly,
10999 as described in the following section.
11001 @node The Configuration Pragmas Files
11002 @section The Configuration Pragmas Files
11003 @cindex @file{gnat.adc}
11006 In GNAT a compilation environment is defined by the current
11007 directory at the time that a compile command is given. This current
11008 directory is searched for a file whose name is @file{gnat.adc}. If
11009 this file is present, it is expected to contain one or more
11010 configuration pragmas that will be applied to the current compilation.
11011 However, if the switch @option{-gnatA} is used, @file{gnat.adc} is not
11014 Configuration pragmas may be entered into the @file{gnat.adc} file
11015 either by running @code{gnatchop} on a source file that consists only of
11016 configuration pragmas, or more conveniently by
11017 direct editing of the @file{gnat.adc} file, which is a standard format
11020 In addition to @file{gnat.adc}, additional files containing configuration
11021 pragmas may be applied to the current compilation using the switch
11022 @option{-gnatec}@var{path}. @var{path} must designate an existing file that
11023 contains only configuration pragmas. These configuration pragmas are
11024 in addition to those found in @file{gnat.adc} (provided @file{gnat.adc}
11025 is present and switch @option{-gnatA} is not used).
11027 It is allowed to specify several switches @option{-gnatec}, all of which
11028 will be taken into account.
11030 If you are using project file, a separate mechanism is provided using
11031 project attributes, see @ref{Specifying Configuration Pragmas} for more
11035 Of special interest to GNAT OpenVMS Alpha is the following
11036 configuration pragma:
11038 @smallexample @c ada
11040 pragma Extend_System (Aux_DEC);
11045 In the presence of this pragma, GNAT adds to the definition of the
11046 predefined package SYSTEM all the additional types and subprograms that are
11047 defined in HP Ada. See @ref{Compatibility with HP Ada} for details.
11050 @node Handling Arbitrary File Naming Conventions Using gnatname
11051 @chapter Handling Arbitrary File Naming Conventions Using @code{gnatname}
11052 @cindex Arbitrary File Naming Conventions
11055 * Arbitrary File Naming Conventions::
11056 * Running gnatname::
11057 * Switches for gnatname::
11058 * Examples of gnatname Usage::
11061 @node Arbitrary File Naming Conventions
11062 @section Arbitrary File Naming Conventions
11065 The GNAT compiler must be able to know the source file name of a compilation
11066 unit. When using the standard GNAT default file naming conventions
11067 (@code{.ads} for specs, @code{.adb} for bodies), the GNAT compiler
11068 does not need additional information.
11071 When the source file names do not follow the standard GNAT default file naming
11072 conventions, the GNAT compiler must be given additional information through
11073 a configuration pragmas file (@pxref{Configuration Pragmas})
11075 When the non-standard file naming conventions are well-defined,
11076 a small number of pragmas @code{Source_File_Name} specifying a naming pattern
11077 (@pxref{Alternative File Naming Schemes}) may be sufficient. However,
11078 if the file naming conventions are irregular or arbitrary, a number
11079 of pragma @code{Source_File_Name} for individual compilation units
11081 To help maintain the correspondence between compilation unit names and
11082 source file names within the compiler,
11083 GNAT provides a tool @code{gnatname} to generate the required pragmas for a
11086 @node Running gnatname
11087 @section Running @code{gnatname}
11090 The usual form of the @code{gnatname} command is
11093 $ gnatname @ovar{switches} @var{naming_pattern} @ovar{naming_patterns}
11094 @r{[}--and @ovar{switches} @var{naming_pattern} @ovar{naming_patterns}@r{]}
11098 All of the arguments are optional. If invoked without any argument,
11099 @code{gnatname} will display its usage.
11102 When used with at least one naming pattern, @code{gnatname} will attempt to
11103 find all the compilation units in files that follow at least one of the
11104 naming patterns. To find these compilation units,
11105 @code{gnatname} will use the GNAT compiler in syntax-check-only mode on all
11109 One or several Naming Patterns may be given as arguments to @code{gnatname}.
11110 Each Naming Pattern is enclosed between double quotes.
11111 A Naming Pattern is a regular expression similar to the wildcard patterns
11112 used in file names by the Unix shells or the DOS prompt.
11115 @code{gnatname} may be called with several sections of directories/patterns.
11116 Sections are separated by switch @code{--and}. In each section, there must be
11117 at least one pattern. If no directory is specified in a section, the current
11118 directory (or the project directory is @code{-P} is used) is implied.
11119 The options other that the directory switches and the patterns apply globally
11120 even if they are in different sections.
11123 Examples of Naming Patterns are
11132 For a more complete description of the syntax of Naming Patterns,
11133 see the second kind of regular expressions described in @file{g-regexp.ads}
11134 (the ``Glob'' regular expressions).
11137 When invoked with no switch @code{-P}, @code{gnatname} will create a
11138 configuration pragmas file @file{gnat.adc} in the current working directory,
11139 with pragmas @code{Source_File_Name} for each file that contains a valid Ada
11142 @node Switches for gnatname
11143 @section Switches for @code{gnatname}
11146 Switches for @code{gnatname} must precede any specified Naming Pattern.
11149 You may specify any of the following switches to @code{gnatname}:
11155 @cindex @option{--version} @command{gnatname}
11156 Display Copyright and version, then exit disregarding all other options.
11159 @cindex @option{--help} @command{gnatname}
11160 If @option{--version} was not used, display usage, then exit disregarding
11164 Start another section of directories/patterns.
11166 @item ^-c^/CONFIG_FILE=^@file{file}
11167 @cindex @option{^-c^/CONFIG_FILE^} (@code{gnatname})
11168 Create a configuration pragmas file @file{file} (instead of the default
11171 There may be zero, one or more space between @option{-c} and
11174 @file{file} may include directory information. @file{file} must be
11175 writable. There may be only one switch @option{^-c^/CONFIG_FILE^}.
11176 When a switch @option{^-c^/CONFIG_FILE^} is
11177 specified, no switch @option{^-P^/PROJECT_FILE^} may be specified (see below).
11179 @item ^-d^/SOURCE_DIRS=^@file{dir}
11180 @cindex @option{^-d^/SOURCE_DIRS^} (@code{gnatname})
11181 Look for source files in directory @file{dir}. There may be zero, one or more
11182 spaces between @option{^-d^/SOURCE_DIRS=^} and @file{dir}.
11183 When a switch @option{^-d^/SOURCE_DIRS^}
11184 is specified, the current working directory will not be searched for source
11185 files, unless it is explicitly specified with a @option{^-d^/SOURCE_DIRS^}
11186 or @option{^-D^/DIR_FILES^} switch.
11187 Several switches @option{^-d^/SOURCE_DIRS^} may be specified.
11188 If @file{dir} is a relative path, it is relative to the directory of
11189 the configuration pragmas file specified with switch
11190 @option{^-c^/CONFIG_FILE^},
11191 or to the directory of the project file specified with switch
11192 @option{^-P^/PROJECT_FILE^} or,
11193 if neither switch @option{^-c^/CONFIG_FILE^}
11194 nor switch @option{^-P^/PROJECT_FILE^} are specified, it is relative to the
11195 current working directory. The directory
11196 specified with switch @option{^-d^/SOURCE_DIRS^} must exist and be readable.
11198 @item ^-D^/DIRS_FILE=^@file{file}
11199 @cindex @option{^-D^/DIRS_FILE^} (@code{gnatname})
11200 Look for source files in all directories listed in text file @file{file}.
11201 There may be zero, one or more spaces between @option{^-D^/DIRS_FILE=^}
11203 @file{file} must be an existing, readable text file.
11204 Each nonempty line in @file{file} must be a directory.
11205 Specifying switch @option{^-D^/DIRS_FILE^} is equivalent to specifying as many
11206 switches @option{^-d^/SOURCE_DIRS^} as there are nonempty lines in
11209 @item ^-f^/FOREIGN_PATTERN=^@file{pattern}
11210 @cindex @option{^-f^/FOREIGN_PATTERN^} (@code{gnatname})
11211 Foreign patterns. Using this switch, it is possible to add sources of languages
11212 other than Ada to the list of sources of a project file.
11213 It is only useful if a ^-P^/PROJECT_FILE^ switch is used.
11216 gnatname ^-Pprj -f"*.c"^/PROJECT_FILE=PRJ /FOREIGN_PATTERN=*.C^ "*.ada"
11219 will look for Ada units in all files with the @file{.ada} extension,
11220 and will add to the list of file for project @file{prj.gpr} the C files
11221 with extension @file{.^c^C^}.
11224 @cindex @option{^-h^/HELP^} (@code{gnatname})
11225 Output usage (help) information. The output is written to @file{stdout}.
11227 @item ^-P^/PROJECT_FILE=^@file{proj}
11228 @cindex @option{^-P^/PROJECT_FILE^} (@code{gnatname})
11229 Create or update project file @file{proj}. There may be zero, one or more space
11230 between @option{-P} and @file{proj}. @file{proj} may include directory
11231 information. @file{proj} must be writable.
11232 There may be only one switch @option{^-P^/PROJECT_FILE^}.
11233 When a switch @option{^-P^/PROJECT_FILE^} is specified,
11234 no switch @option{^-c^/CONFIG_FILE^} may be specified.
11236 @item ^-v^/VERBOSE^
11237 @cindex @option{^-v^/VERBOSE^} (@code{gnatname})
11238 Verbose mode. Output detailed explanation of behavior to @file{stdout}.
11239 This includes name of the file written, the name of the directories to search
11240 and, for each file in those directories whose name matches at least one of
11241 the Naming Patterns, an indication of whether the file contains a unit,
11242 and if so the name of the unit.
11244 @item ^-v -v^/VERBOSE /VERBOSE^
11245 @cindex @option{^-v -v^/VERBOSE /VERBOSE^} (@code{gnatname})
11246 Very Verbose mode. In addition to the output produced in verbose mode,
11247 for each file in the searched directories whose name matches none of
11248 the Naming Patterns, an indication is given that there is no match.
11250 @item ^-x^/EXCLUDED_PATTERN=^@file{pattern}
11251 @cindex @option{^-x^/EXCLUDED_PATTERN^} (@code{gnatname})
11252 Excluded patterns. Using this switch, it is possible to exclude some files
11253 that would match the name patterns. For example,
11255 gnatname ^-x "*_nt.ada"^/EXCLUDED_PATTERN=*_nt.ada^ "*.ada"
11258 will look for Ada units in all files with the @file{.ada} extension,
11259 except those whose names end with @file{_nt.ada}.
11263 @node Examples of gnatname Usage
11264 @section Examples of @code{gnatname} Usage
11268 $ gnatname /CONFIG_FILE=[HOME.ME]NAMES.ADC /SOURCE_DIRS=SOURCES "[a-z]*.ada*"
11274 $ gnatname -c /home/me/names.adc -d sources "[a-z]*.ada*"
11279 In this example, the directory @file{^/home/me^[HOME.ME]^} must already exist
11280 and be writable. In addition, the directory
11281 @file{^/home/me/sources^[HOME.ME.SOURCES]^} (specified by
11282 @option{^-d sources^/SOURCE_DIRS=SOURCES^}) must exist and be readable.
11285 Note the optional spaces after @option{-c} and @option{-d}.
11290 $ gnatname -P/home/me/proj -x "*_nt_body.ada"
11291 -dsources -dsources/plus -Dcommon_dirs.txt "body_*" "spec_*"
11294 $ gnatname /PROJECT_FILE=[HOME.ME]PROJ
11295 /EXCLUDED_PATTERN=*_nt_body.ada
11296 /SOURCE_DIRS=(SOURCES,[SOURCES.PLUS])
11297 /DIRS_FILE=COMMON_DIRS.TXT "body_*" "spec_*"
11301 Note that several switches @option{^-d^/SOURCE_DIRS^} may be used,
11302 even in conjunction with one or several switches
11303 @option{^-D^/DIRS_FILE^}. Several Naming Patterns and one excluded pattern
11304 are used in this example.
11306 @c *****************************************
11307 @c * G N A T P r o j e c t M a n a g e r *
11308 @c *****************************************
11309 @node GNAT Project Manager
11310 @chapter GNAT Project Manager
11314 * Examples of Project Files::
11315 * Project File Syntax::
11316 * Objects and Sources in Project Files::
11317 * Importing Projects::
11318 * Project Extension::
11319 * Project Hierarchy Extension::
11320 * External References in Project Files::
11321 * Packages in Project Files::
11322 * Variables from Imported Projects::
11324 * Library Projects::
11325 * Stand-alone Library Projects::
11326 * Switches Related to Project Files::
11327 * Tools Supporting Project Files::
11328 * An Extended Example::
11329 * Project File Complete Syntax::
11332 @c ****************
11333 @c * Introduction *
11334 @c ****************
11337 @section Introduction
11340 This chapter describes GNAT's @emph{Project Manager}, a facility that allows
11341 you to manage complex builds involving a number of source files, directories,
11342 and compilation options for different system configurations. In particular,
11343 project files allow you to specify:
11346 The directory or set of directories containing the source files, and/or the
11347 names of the specific source files themselves
11349 The directory in which the compiler's output
11350 (@file{ALI} files, object files, tree files) is to be placed
11352 The directory in which the executable programs is to be placed
11354 ^Switch^Switch^ settings for any of the project-enabled tools
11355 (@command{gnatmake}, compiler, binder, linker, @code{gnatls}, @code{gnatxref},
11356 @code{gnatfind}); you can apply these settings either globally or to individual
11359 The source files containing the main subprogram(s) to be built
11361 The source programming language(s) (currently Ada and/or C)
11363 Source file naming conventions; you can specify these either globally or for
11364 individual compilation units
11371 @node Project Files
11372 @subsection Project Files
11375 Project files are written in a syntax close to that of Ada, using familiar
11376 notions such as packages, context clauses, declarations, default values,
11377 assignments, and inheritance. Finally, project files can be built
11378 hierarchically from other project files, simplifying complex system
11379 integration and project reuse.
11381 A @dfn{project} is a specific set of values for various compilation properties.
11382 The settings for a given project are described by means of
11383 a @dfn{project file}, which is a text file written in an Ada-like syntax.
11384 Property values in project files are either strings or lists of strings.
11385 Properties that are not explicitly set receive default values. A project
11386 file may interrogate the values of @dfn{external variables} (user-defined
11387 command-line switches or environment variables), and it may specify property
11388 settings conditionally, based on the value of such variables.
11390 In simple cases, a project's source files depend only on other source files
11391 in the same project, or on the predefined libraries. (@emph{Dependence} is
11393 the Ada technical sense; as in one Ada unit @code{with}ing another.) However,
11394 the Project Manager also allows more sophisticated arrangements,
11395 where the source files in one project depend on source files in other
11399 One project can @emph{import} other projects containing needed source files.
11401 You can organize GNAT projects in a hierarchy: a @emph{child} project
11402 can extend a @emph{parent} project, inheriting the parent's source files and
11403 optionally overriding any of them with alternative versions
11407 More generally, the Project Manager lets you structure large development
11408 efforts into hierarchical subsystems, where build decisions are delegated
11409 to the subsystem level, and thus different compilation environments
11410 (^switch^switch^ settings) used for different subsystems.
11412 The Project Manager is invoked through the
11413 @option{^-P^/PROJECT_FILE=^@emph{projectfile}}
11414 switch to @command{gnatmake} or to the @command{^gnat^GNAT^} front driver.
11416 There may be zero, one or more spaces between @option{-P} and
11417 @option{@emph{projectfile}}.
11419 If you want to define (on the command line) an external variable that is
11420 queried by the project file, you must use the
11421 @option{^-X^/EXTERNAL_REFERENCE=^@emph{vbl}=@emph{value}} switch.
11422 The Project Manager parses and interprets the project file, and drives the
11423 invoked tool based on the project settings.
11425 The Project Manager supports a wide range of development strategies,
11426 for systems of all sizes. Here are some typical practices that are
11430 Using a common set of source files, but generating object files in different
11431 directories via different ^switch^switch^ settings
11433 Using a mostly-shared set of source files, but with different versions of
11438 The destination of an executable can be controlled inside a project file
11439 using the @option{^-o^-o^}
11441 In the absence of such a ^switch^switch^ either inside
11442 the project file or on the command line, any executable files generated by
11443 @command{gnatmake} are placed in the directory @code{Exec_Dir} specified
11444 in the project file. If no @code{Exec_Dir} is specified, they will be placed
11445 in the object directory of the project.
11447 You can use project files to achieve some of the effects of a source
11448 versioning system (for example, defining separate projects for
11449 the different sets of sources that comprise different releases) but the
11450 Project Manager is independent of any source configuration management tools
11451 that might be used by the developers.
11453 The next section introduces the main features of GNAT's project facility
11454 through a sequence of examples; subsequent sections will present the syntax
11455 and semantics in more detail. A more formal description of the project
11456 facility appears in @ref{Project File Reference,,, gnat_rm, GNAT
11459 @c *****************************
11460 @c * Examples of Project Files *
11461 @c *****************************
11463 @node Examples of Project Files
11464 @section Examples of Project Files
11466 This section illustrates some of the typical uses of project files and
11467 explains their basic structure and behavior.
11470 * Common Sources with Different ^Switches^Switches^ and Directories::
11471 * Using External Variables::
11472 * Importing Other Projects::
11473 * Extending a Project::
11476 @node Common Sources with Different ^Switches^Switches^ and Directories
11477 @subsection Common Sources with Different ^Switches^Switches^ and Directories
11481 * Specifying the Object Directory::
11482 * Specifying the Exec Directory::
11483 * Project File Packages::
11484 * Specifying ^Switch^Switch^ Settings::
11485 * Main Subprograms::
11486 * Executable File Names::
11487 * Source File Naming Conventions::
11488 * Source Language(s)::
11492 Suppose that the Ada source files @file{pack.ads}, @file{pack.adb}, and
11493 @file{proc.adb} are in the @file{/common} directory. The file
11494 @file{proc.adb} contains an Ada main subprogram @code{Proc} that @code{with}s
11495 package @code{Pack}. We want to compile these source files under two sets
11496 of ^switches^switches^:
11499 When debugging, we want to pass the @option{-g} switch to @command{gnatmake},
11500 and the @option{^-gnata^-gnata^},
11501 @option{^-gnato^-gnato^},
11502 and @option{^-gnatE^-gnatE^} switches to the
11503 compiler; the compiler's output is to appear in @file{/common/debug}
11505 When preparing a release version, we want to pass the @option{^-O2^O2^} switch
11506 to the compiler; the compiler's output is to appear in @file{/common/release}
11510 The GNAT project files shown below, respectively @file{debug.gpr} and
11511 @file{release.gpr} in the @file{/common} directory, achieve these effects.
11524 ^/common/debug^[COMMON.DEBUG]^
11529 ^/common/release^[COMMON.RELEASE]^
11534 Here are the corresponding project files:
11536 @smallexample @c projectfile
11539 for Object_Dir use "debug";
11540 for Main use ("proc");
11543 for ^Default_Switches^Default_Switches^ ("Ada")
11545 for Executable ("proc.adb") use "proc1";
11550 package Compiler is
11551 for ^Default_Switches^Default_Switches^ ("Ada")
11552 use ("-fstack-check",
11555 "^-gnatE^-gnatE^");
11561 @smallexample @c projectfile
11564 for Object_Dir use "release";
11565 for Exec_Dir use ".";
11566 for Main use ("proc");
11568 package Compiler is
11569 for ^Default_Switches^Default_Switches^ ("Ada")
11577 The name of the project defined by @file{debug.gpr} is @code{"Debug"} (case
11578 insensitive), and analogously the project defined by @file{release.gpr} is
11579 @code{"Release"}. For consistency the file should have the same name as the
11580 project, and the project file's extension should be @code{"gpr"}. These
11581 conventions are not required, but a warning is issued if they are not followed.
11583 If the current directory is @file{^/temp^[TEMP]^}, then the command
11585 gnatmake ^-P/common/debug.gpr^/PROJECT_FILE=[COMMON]DEBUG^
11589 generates object and ALI files in @file{^/common/debug^[COMMON.DEBUG]^},
11590 as well as the @code{^proc1^PROC1.EXE^} executable,
11591 using the ^switch^switch^ settings defined in the project file.
11593 Likewise, the command
11595 gnatmake ^-P/common/release.gpr^/PROJECT_FILE=[COMMON]RELEASE^
11599 generates object and ALI files in @file{^/common/release^[COMMON.RELEASE]^},
11600 and the @code{^proc^PROC.EXE^}
11601 executable in @file{^/common^[COMMON]^},
11602 using the ^switch^switch^ settings from the project file.
11605 @unnumberedsubsubsec Source Files
11608 If a project file does not explicitly specify a set of source directories or
11609 a set of source files, then by default the project's source files are the
11610 Ada source files in the project file directory. Thus @file{pack.ads},
11611 @file{pack.adb}, and @file{proc.adb} are the source files for both projects.
11613 @node Specifying the Object Directory
11614 @unnumberedsubsubsec Specifying the Object Directory
11617 Several project properties are modeled by Ada-style @emph{attributes};
11618 a property is defined by supplying the equivalent of an Ada attribute
11619 definition clause in the project file.
11620 A project's object directory is another such a property; the corresponding
11621 attribute is @code{Object_Dir}, and its value is also a string expression,
11622 specified either as absolute or relative. In the later case,
11623 it is relative to the project file directory. Thus the compiler's
11624 output is directed to @file{^/common/debug^[COMMON.DEBUG]^}
11625 (for the @code{Debug} project)
11626 and to @file{^/common/release^[COMMON.RELEASE]^}
11627 (for the @code{Release} project).
11628 If @code{Object_Dir} is not specified, then the default is the project file
11631 @node Specifying the Exec Directory
11632 @unnumberedsubsubsec Specifying the Exec Directory
11635 A project's exec directory is another property; the corresponding
11636 attribute is @code{Exec_Dir}, and its value is also a string expression,
11637 either specified as relative or absolute. If @code{Exec_Dir} is not specified,
11638 then the default is the object directory (which may also be the project file
11639 directory if attribute @code{Object_Dir} is not specified). Thus the executable
11640 is placed in @file{^/common/debug^[COMMON.DEBUG]^}
11641 for the @code{Debug} project (attribute @code{Exec_Dir} not specified)
11642 and in @file{^/common^[COMMON]^} for the @code{Release} project.
11644 @node Project File Packages
11645 @unnumberedsubsubsec Project File Packages
11648 A GNAT tool that is integrated with the Project Manager is modeled by a
11649 corresponding package in the project file. In the example above,
11650 The @code{Debug} project defines the packages @code{Builder}
11651 (for @command{gnatmake}) and @code{Compiler};
11652 the @code{Release} project defines only the @code{Compiler} package.
11654 The Ada-like package syntax is not to be taken literally. Although packages in
11655 project files bear a surface resemblance to packages in Ada source code, the
11656 notation is simply a way to convey a grouping of properties for a named
11657 entity. Indeed, the package names permitted in project files are restricted
11658 to a predefined set, corresponding to the project-aware tools, and the contents
11659 of packages are limited to a small set of constructs.
11660 The packages in the example above contain attribute definitions.
11662 @node Specifying ^Switch^Switch^ Settings
11663 @unnumberedsubsubsec Specifying ^Switch^Switch^ Settings
11666 ^Switch^Switch^ settings for a project-aware tool can be specified through
11667 attributes in the package that corresponds to the tool.
11668 The example above illustrates one of the relevant attributes,
11669 @code{^Default_Switches^Default_Switches^}, which is defined in packages
11670 in both project files.
11671 Unlike simple attributes like @code{Source_Dirs},
11672 @code{^Default_Switches^Default_Switches^} is
11673 known as an @emph{associative array}. When you define this attribute, you must
11674 supply an ``index'' (a literal string), and the effect of the attribute
11675 definition is to set the value of the array at the specified index.
11676 For the @code{^Default_Switches^Default_Switches^} attribute,
11677 the index is a programming language (in our case, Ada),
11678 and the value specified (after @code{use}) must be a list
11679 of string expressions.
11681 The attributes permitted in project files are restricted to a predefined set.
11682 Some may appear at project level, others in packages.
11683 For any attribute that is an associative array, the index must always be a
11684 literal string, but the restrictions on this string (e.g., a file name or a
11685 language name) depend on the individual attribute.
11686 Also depending on the attribute, its specified value will need to be either a
11687 string or a string list.
11689 In the @code{Debug} project, we set the switches for two tools,
11690 @command{gnatmake} and the compiler, and thus we include the two corresponding
11691 packages; each package defines the @code{^Default_Switches^Default_Switches^}
11692 attribute with index @code{"Ada"}.
11693 Note that the package corresponding to
11694 @command{gnatmake} is named @code{Builder}. The @code{Release} project is
11695 similar, but only includes the @code{Compiler} package.
11697 In project @code{Debug} above, the ^switches^switches^ starting with
11698 @option{-gnat} that are specified in package @code{Compiler}
11699 could have been placed in package @code{Builder}, since @command{gnatmake}
11700 transmits all such ^switches^switches^ to the compiler.
11702 @node Main Subprograms
11703 @unnumberedsubsubsec Main Subprograms
11706 One of the specifiable properties of a project is a list of files that contain
11707 main subprograms. This property is captured in the @code{Main} attribute,
11708 whose value is a list of strings. If a project defines the @code{Main}
11709 attribute, it is not necessary to identify the main subprogram(s) when
11710 invoking @command{gnatmake} (@pxref{gnatmake and Project Files}).
11712 @node Executable File Names
11713 @unnumberedsubsubsec Executable File Names
11716 By default, the executable file name corresponding to a main source is
11717 deduced from the main source file name. Through the attributes
11718 @code{Executable} and @code{Executable_Suffix} of package @code{Builder},
11719 it is possible to change this default.
11720 In project @code{Debug} above, the executable file name
11721 for main source @file{^proc.adb^PROC.ADB^} is
11722 @file{^proc1^PROC1.EXE^}.
11723 Attribute @code{Executable_Suffix}, when specified, may change the suffix
11724 of the executable files, when no attribute @code{Executable} applies:
11725 its value replace the platform-specific executable suffix.
11726 Attributes @code{Executable} and @code{Executable_Suffix} are the only ways to
11727 specify a non-default executable file name when several mains are built at once
11728 in a single @command{gnatmake} command.
11730 @node Source File Naming Conventions
11731 @unnumberedsubsubsec Source File Naming Conventions
11734 Since the project files above do not specify any source file naming
11735 conventions, the GNAT defaults are used. The mechanism for defining source
11736 file naming conventions -- a package named @code{Naming} --
11737 is described below (@pxref{Naming Schemes}).
11739 @node Source Language(s)
11740 @unnumberedsubsubsec Source Language(s)
11743 Since the project files do not specify a @code{Languages} attribute, by
11744 default the GNAT tools assume that the language of the project file is Ada.
11745 More generally, a project can comprise source files
11746 in Ada, C, and/or other languages.
11748 @node Using External Variables
11749 @subsection Using External Variables
11752 Instead of supplying different project files for debug and release, we can
11753 define a single project file that queries an external variable (set either
11754 on the command line or via an ^environment variable^logical name^) in order to
11755 conditionally define the appropriate settings. Again, assume that the
11756 source files @file{pack.ads}, @file{pack.adb}, and @file{proc.adb} are
11757 located in directory @file{^/common^[COMMON]^}. The following project file,
11758 @file{build.gpr}, queries the external variable named @code{STYLE} and
11759 defines an object directory and ^switch^switch^ settings based on whether
11760 the value is @code{"deb"} (debug) or @code{"rel"} (release), and where
11761 the default is @code{"deb"}.
11763 @smallexample @c projectfile
11766 for Main use ("proc");
11768 type Style_Type is ("deb", "rel");
11769 Style : Style_Type := external ("STYLE", "deb");
11773 for Object_Dir use "debug";
11776 for Object_Dir use "release";
11777 for Exec_Dir use ".";
11786 for ^Default_Switches^Default_Switches^ ("Ada")
11788 for Executable ("proc") use "proc1";
11797 package Compiler is
11801 for ^Default_Switches^Default_Switches^ ("Ada")
11802 use ("^-gnata^-gnata^",
11804 "^-gnatE^-gnatE^");
11807 for ^Default_Switches^Default_Switches^ ("Ada")
11818 @code{Style_Type} is an example of a @emph{string type}, which is the project
11819 file analog of an Ada enumeration type but whose components are string literals
11820 rather than identifiers. @code{Style} is declared as a variable of this type.
11822 The form @code{external("STYLE", "deb")} is known as an
11823 @emph{external reference}; its first argument is the name of an
11824 @emph{external variable}, and the second argument is a default value to be
11825 used if the external variable doesn't exist. You can define an external
11826 variable on the command line via the @option{^-X^/EXTERNAL_REFERENCE^} switch,
11827 or you can use ^an environment variable^a logical name^
11828 as an external variable.
11830 Each @code{case} construct is expanded by the Project Manager based on the
11831 value of @code{Style}. Thus the command
11834 gnatmake -P/common/build.gpr -XSTYLE=deb
11840 gnatmake /PROJECT_FILE=[COMMON]BUILD.GPR /EXTERNAL_REFERENCE=STYLE=deb
11845 is equivalent to the @command{gnatmake} invocation using the project file
11846 @file{debug.gpr} in the earlier example. So is the command
11848 gnatmake ^-P/common/build.gpr^/PROJECT_FILE=[COMMON]BUILD.GPR^
11852 since @code{"deb"} is the default for @code{STYLE}.
11858 gnatmake -P/common/build.gpr -XSTYLE=rel
11864 GNAT MAKE /PROJECT_FILE=[COMMON]BUILD.GPR /EXTERNAL_REFERENCE=STYLE=rel
11869 is equivalent to the @command{gnatmake} invocation using the project file
11870 @file{release.gpr} in the earlier example.
11872 @node Importing Other Projects
11873 @subsection Importing Other Projects
11874 @cindex @code{ADA_PROJECT_PATH}
11877 A compilation unit in a source file in one project may depend on compilation
11878 units in source files in other projects. To compile this unit under
11879 control of a project file, the
11880 dependent project must @emph{import} the projects containing the needed source
11882 This effect is obtained using syntax similar to an Ada @code{with} clause,
11883 but where @code{with}ed entities are strings that denote project files.
11885 As an example, suppose that the two projects @code{GUI_Proj} and
11886 @code{Comm_Proj} are defined in the project files @file{gui_proj.gpr} and
11887 @file{comm_proj.gpr} in directories @file{^/gui^[GUI]^}
11888 and @file{^/comm^[COMM]^}, respectively.
11889 Suppose that the source files for @code{GUI_Proj} are
11890 @file{gui.ads} and @file{gui.adb}, and that the source files for
11891 @code{Comm_Proj} are @file{comm.ads} and @file{comm.adb}, where each set of
11892 files is located in its respective project file directory. Schematically:
11911 We want to develop an application in directory @file{^/app^[APP]^} that
11912 @code{with} the packages @code{GUI} and @code{Comm}, using the properties of
11913 the corresponding project files (e.g.@: the ^switch^switch^ settings
11914 and object directory).
11915 Skeletal code for a main procedure might be something like the following:
11917 @smallexample @c ada
11920 procedure App_Main is
11929 Here is a project file, @file{app_proj.gpr}, that achieves the desired
11932 @smallexample @c projectfile
11934 with "/gui/gui_proj", "/comm/comm_proj";
11935 project App_Proj is
11936 for Main use ("app_main");
11942 Building an executable is achieved through the command:
11944 gnatmake ^-P/app/app_proj^/PROJECT_FILE=[APP]APP_PROJ^
11947 which will generate the @code{^app_main^APP_MAIN.EXE^} executable
11948 in the directory where @file{app_proj.gpr} resides.
11950 If an imported project file uses the standard extension (@code{^gpr^GPR^}) then
11951 (as illustrated above) the @code{with} clause can omit the extension.
11953 Our example specified an absolute path for each imported project file.
11954 Alternatively, the directory name of an imported object can be omitted
11958 The imported project file is in the same directory as the importing project
11961 You have defined ^an environment variable^a logical name^
11962 that includes the directory containing
11963 the needed project file. The syntax of @code{ADA_PROJECT_PATH} is the same as
11964 the syntax of @code{ADA_INCLUDE_PATH} and @code{ADA_OBJECTS_PATH}: a list of
11965 directory names separated by colons (semicolons on Windows).
11969 Thus, if we define @code{ADA_PROJECT_PATH} to include @file{^/gui^[GUI]^} and
11970 @file{^/comm^[COMM]^}, then our project file @file{app_proj.gpr} can be written
11973 @smallexample @c projectfile
11975 with "gui_proj", "comm_proj";
11976 project App_Proj is
11977 for Main use ("app_main");
11983 Importing other projects can create ambiguities.
11984 For example, the same unit might be present in different imported projects, or
11985 it might be present in both the importing project and in an imported project.
11986 Both of these conditions are errors. Note that in the current version of
11987 the Project Manager, it is illegal to have an ambiguous unit even if the
11988 unit is never referenced by the importing project. This restriction may be
11989 relaxed in a future release.
11991 @node Extending a Project
11992 @subsection Extending a Project
11995 In large software systems it is common to have multiple
11996 implementations of a common interface; in Ada terms, multiple versions of a
11997 package body for the same spec. For example, one implementation
11998 might be safe for use in tasking programs, while another might only be used
11999 in sequential applications. This can be modeled in GNAT using the concept
12000 of @emph{project extension}. If one project (the ``child'') @emph{extends}
12001 another project (the ``parent'') then by default all source files of the
12002 parent project are inherited by the child, but the child project can
12003 override any of the parent's source files with new versions, and can also
12004 add new files. This facility is the project analog of a type extension in
12005 Object-Oriented Programming. Project hierarchies are permitted (a child
12006 project may be the parent of yet another project), and a project that
12007 inherits one project can also import other projects.
12009 As an example, suppose that directory @file{^/seq^[SEQ]^} contains the project
12010 file @file{seq_proj.gpr} as well as the source files @file{pack.ads},
12011 @file{pack.adb}, and @file{proc.adb}:
12024 Note that the project file can simply be empty (that is, no attribute or
12025 package is defined):
12027 @smallexample @c projectfile
12029 project Seq_Proj is
12035 implying that its source files are all the Ada source files in the project
12038 Suppose we want to supply an alternate version of @file{pack.adb}, in
12039 directory @file{^/tasking^[TASKING]^}, but use the existing versions of
12040 @file{pack.ads} and @file{proc.adb}. We can define a project
12041 @code{Tasking_Proj} that inherits @code{Seq_Proj}:
12045 ^/tasking^[TASKING]^
12051 project Tasking_Proj extends "/seq/seq_proj" is
12057 The version of @file{pack.adb} used in a build depends on which project file
12060 Note that we could have obtained the desired behavior using project import
12061 rather than project inheritance; a @code{base} project would contain the
12062 sources for @file{pack.ads} and @file{proc.adb}, a sequential project would
12063 import @code{base} and add @file{pack.adb}, and likewise a tasking project
12064 would import @code{base} and add a different version of @file{pack.adb}. The
12065 choice depends on whether other sources in the original project need to be
12066 overridden. If they do, then project extension is necessary, otherwise,
12067 importing is sufficient.
12070 In a project file that extends another project file, it is possible to
12071 indicate that an inherited source is not part of the sources of the extending
12072 project. This is necessary sometimes when a package spec has been overloaded
12073 and no longer requires a body: in this case, it is necessary to indicate that
12074 the inherited body is not part of the sources of the project, otherwise there
12075 will be a compilation error when compiling the spec.
12077 For that purpose, the attribute @code{Excluded_Source_Files} is used.
12078 Its value is a string list: a list of file names. It is also possible to use
12079 attribute @code{Excluded_Source_List_File}. Its value is a single string:
12080 the file name of a text file containing a list of file names, one per line.
12082 @smallexample @c @projectfile
12083 project B extends "a" is
12084 for Source_Files use ("pkg.ads");
12085 -- New spec of Pkg does not need a completion
12086 for Excluded_Source_Files use ("pkg.adb");
12090 Attribute @code{Excluded_Source_Files} may also be used to check if a source
12091 is still needed: if it is possible to build using @command{gnatmake} when such
12092 a source is put in attribute @code{Excluded_Source_Files} of a project P, then
12093 it is possible to remove the source completely from a system that includes
12096 @c ***********************
12097 @c * Project File Syntax *
12098 @c ***********************
12100 @node Project File Syntax
12101 @section Project File Syntax
12105 * Qualified Projects::
12111 * Associative Array Attributes::
12112 * case Constructions::
12116 This section describes the structure of project files.
12118 A project may be an @emph{independent project}, entirely defined by a single
12119 project file. Any Ada source file in an independent project depends only
12120 on the predefined library and other Ada source files in the same project.
12123 A project may also @dfn{depend on} other projects, in either or both of
12124 the following ways:
12126 @item It may import any number of projects
12127 @item It may extend at most one other project
12131 The dependence relation is a directed acyclic graph (the subgraph reflecting
12132 the ``extends'' relation is a tree).
12134 A project's @dfn{immediate sources} are the source files directly defined by
12135 that project, either implicitly by residing in the project file's directory,
12136 or explicitly through any of the source-related attributes described below.
12137 More generally, a project @var{proj}'s @dfn{sources} are the immediate sources
12138 of @var{proj} together with the immediate sources (unless overridden) of any
12139 project on which @var{proj} depends (either directly or indirectly).
12142 @subsection Basic Syntax
12145 As seen in the earlier examples, project files have an Ada-like syntax.
12146 The minimal project file is:
12147 @smallexample @c projectfile
12156 The identifier @code{Empty} is the name of the project.
12157 This project name must be present after the reserved
12158 word @code{end} at the end of the project file, followed by a semi-colon.
12160 Any name in a project file, such as the project name or a variable name,
12161 has the same syntax as an Ada identifier.
12163 The reserved words of project files are the Ada 95 reserved words plus
12164 @code{extends}, @code{external}, and @code{project}. Note that the only Ada
12165 reserved words currently used in project file syntax are:
12201 Comments in project files have the same syntax as in Ada, two consecutive
12202 hyphens through the end of the line.
12204 @node Qualified Projects
12205 @subsection Qualified Projects
12208 Before the reserved @code{project}, there may be one or two "qualifiers", that
12209 is identifiers or other reserved words, to qualify the project.
12211 The current list of qualifiers is:
12215 @code{abstract}: qualify a project with no sources. An abstract project must
12216 have a declaration specifying that there are no sources in the project, and,
12217 if it extends another project, the project it extends must also be a qualified
12221 @code{standard}: a standard project is a non library project with sources.
12224 @code{aggregate}: for future extension
12227 @code{aggregate library}: for future extension
12230 @code{library}: a library project must declare both attributes
12231 @code{Library_Name} and @code{Library_Dir}.
12234 @code{configuration}: a configuration project cannot be in a project tree.
12238 @subsection Packages
12241 A project file may contain @emph{packages}. The name of a package must be one
12242 of the identifiers from the following list. A package
12243 with a given name may only appear once in a project file. Package names are
12244 case insensitive. The following package names are legal:
12260 @code{Cross_Reference}
12264 @code{Pretty_Printer}
12274 @code{Language_Processing}
12278 In its simplest form, a package may be empty:
12280 @smallexample @c projectfile
12290 A package may contain @emph{attribute declarations},
12291 @emph{variable declarations} and @emph{case constructions}, as will be
12294 When there is ambiguity between a project name and a package name,
12295 the name always designates the project. To avoid possible confusion, it is
12296 always a good idea to avoid naming a project with one of the
12297 names allowed for packages or any name that starts with @code{gnat}.
12300 @subsection Expressions
12303 An @emph{expression} is either a @emph{string expression} or a
12304 @emph{string list expression}.
12306 A @emph{string expression} is either a @emph{simple string expression} or a
12307 @emph{compound string expression}.
12309 A @emph{simple string expression} is one of the following:
12311 @item A literal string; e.g.@: @code{"comm/my_proj.gpr"}
12312 @item A string-valued variable reference (@pxref{Variables})
12313 @item A string-valued attribute reference (@pxref{Attributes})
12314 @item An external reference (@pxref{External References in Project Files})
12318 A @emph{compound string expression} is a concatenation of string expressions,
12319 using the operator @code{"&"}
12321 Path & "/" & File_Name & ".ads"
12325 A @emph{string list expression} is either a
12326 @emph{simple string list expression} or a
12327 @emph{compound string list expression}.
12329 A @emph{simple string list expression} is one of the following:
12331 @item A parenthesized list of zero or more string expressions,
12332 separated by commas
12334 File_Names := (File_Name, "gnat.adc", File_Name & ".orig");
12337 @item A string list-valued variable reference
12338 @item A string list-valued attribute reference
12342 A @emph{compound string list expression} is the concatenation (using
12343 @code{"&"}) of a simple string list expression and an expression. Note that
12344 each term in a compound string list expression, except the first, may be
12345 either a string expression or a string list expression.
12347 @smallexample @c projectfile
12349 File_Name_List := () & File_Name; -- One string in this list
12350 Extended_File_Name_List := File_Name_List & (File_Name & ".orig");
12352 Big_List := File_Name_List & Extended_File_Name_List;
12353 -- Concatenation of two string lists: three strings
12354 Illegal_List := "gnat.adc" & Extended_File_Name_List;
12355 -- Illegal: must start with a string list
12360 @subsection String Types
12363 A @emph{string type declaration} introduces a discrete set of string literals.
12364 If a string variable is declared to have this type, its value
12365 is restricted to the given set of literals.
12367 Here is an example of a string type declaration:
12369 @smallexample @c projectfile
12370 type OS is ("NT", "nt", "Unix", "GNU/Linux", "other OS");
12374 Variables of a string type are called @emph{typed variables}; all other
12375 variables are called @emph{untyped variables}. Typed variables are
12376 particularly useful in @code{case} constructions, to support conditional
12377 attribute declarations.
12378 (@pxref{case Constructions}).
12380 The string literals in the list are case sensitive and must all be different.
12381 They may include any graphic characters allowed in Ada, including spaces.
12383 A string type may only be declared at the project level, not inside a package.
12385 A string type may be referenced by its name if it has been declared in the same
12386 project file, or by an expanded name whose prefix is the name of the project
12387 in which it is declared.
12390 @subsection Variables
12393 A variable may be declared at the project file level, or within a package.
12394 Here are some examples of variable declarations:
12396 @smallexample @c projectfile
12398 This_OS : OS := external ("OS"); -- a typed variable declaration
12399 That_OS := "GNU/Linux"; -- an untyped variable declaration
12404 The syntax of a @emph{typed variable declaration} is identical to the Ada
12405 syntax for an object declaration. By contrast, the syntax of an untyped
12406 variable declaration is identical to an Ada assignment statement. In fact,
12407 variable declarations in project files have some of the characteristics of
12408 an assignment, in that successive declarations for the same variable are
12409 allowed. Untyped variable declarations do establish the expected kind of the
12410 variable (string or string list), and successive declarations for it must
12411 respect the initial kind.
12414 A string variable declaration (typed or untyped) declares a variable
12415 whose value is a string. This variable may be used as a string expression.
12416 @smallexample @c projectfile
12417 File_Name := "readme.txt";
12418 Saved_File_Name := File_Name & ".saved";
12422 A string list variable declaration declares a variable whose value is a list
12423 of strings. The list may contain any number (zero or more) of strings.
12425 @smallexample @c projectfile
12427 List_With_One_Element := ("^-gnaty^-gnaty^");
12428 List_With_Two_Elements := List_With_One_Element & "^-gnatg^-gnatg^";
12429 Long_List := ("main.ada", "pack1_.ada", "pack1.ada", "pack2_.ada"
12430 "pack2.ada", "util_.ada", "util.ada");
12434 The same typed variable may not be declared more than once at project level,
12435 and it may not be declared more than once in any package; it is in effect
12438 The same untyped variable may be declared several times. Declarations are
12439 elaborated in the order in which they appear, so the new value replaces
12440 the old one, and any subsequent reference to the variable uses the new value.
12441 However, as noted above, if a variable has been declared as a string, all
12443 declarations must give it a string value. Similarly, if a variable has
12444 been declared as a string list, all subsequent declarations
12445 must give it a string list value.
12447 A @emph{variable reference} may take several forms:
12450 @item The simple variable name, for a variable in the current package (if any)
12451 or in the current project
12452 @item An expanded name, whose prefix is a context name.
12456 A @emph{context} may be one of the following:
12459 @item The name of an existing package in the current project
12460 @item The name of an imported project of the current project
12461 @item The name of an ancestor project (i.e., a project extended by the current
12462 project, either directly or indirectly)
12463 @item An expanded name whose prefix is an imported/parent project name, and
12464 whose selector is a package name in that project.
12468 A variable reference may be used in an expression.
12471 @subsection Attributes
12474 A project (and its packages) may have @emph{attributes} that define
12475 the project's properties. Some attributes have values that are strings;
12476 others have values that are string lists.
12478 There are two categories of attributes: @emph{simple attributes}
12479 and @emph{associative arrays} (@pxref{Associative Array Attributes}).
12481 Legal project attribute names, and attribute names for each legal package are
12482 listed below. Attributes names are case-insensitive.
12484 The following attributes are defined on projects (all are simple attributes):
12486 @multitable @columnfractions .4 .3
12487 @item @emph{Attribute Name}
12489 @item @code{Source_Files}
12491 @item @code{Source_Dirs}
12493 @item @code{Source_List_File}
12495 @item @code{Object_Dir}
12497 @item @code{Exec_Dir}
12499 @item @code{Excluded_Source_Dirs}
12501 @item @code{Excluded_Source_Files}
12503 @item @code{Excluded_Source_List_File}
12505 @item @code{Languages}
12509 @item @code{Library_Dir}
12511 @item @code{Library_Name}
12513 @item @code{Library_Kind}
12515 @item @code{Library_Version}
12517 @item @code{Library_Interface}
12519 @item @code{Library_Auto_Init}
12521 @item @code{Library_Options}
12523 @item @code{Library_Src_Dir}
12525 @item @code{Library_ALI_Dir}
12527 @item @code{Library_GCC}
12529 @item @code{Library_Symbol_File}
12531 @item @code{Library_Symbol_Policy}
12533 @item @code{Library_Reference_Symbol_File}
12535 @item @code{Externally_Built}
12540 The following attributes are defined for package @code{Naming}
12541 (@pxref{Naming Schemes}):
12543 @multitable @columnfractions .4 .2 .2 .2
12544 @item Attribute Name @tab Category @tab Index @tab Value
12545 @item @code{Spec_Suffix}
12546 @tab associative array
12549 @item @code{Body_Suffix}
12550 @tab associative array
12553 @item @code{Separate_Suffix}
12554 @tab simple attribute
12557 @item @code{Casing}
12558 @tab simple attribute
12561 @item @code{Dot_Replacement}
12562 @tab simple attribute
12566 @tab associative array
12570 @tab associative array
12573 @item @code{Specification_Exceptions}
12574 @tab associative array
12577 @item @code{Implementation_Exceptions}
12578 @tab associative array
12584 The following attributes are defined for packages @code{Builder},
12585 @code{Compiler}, @code{Binder},
12586 @code{Linker}, @code{Cross_Reference}, and @code{Finder}
12587 (@pxref{^Switches^Switches^ and Project Files}).
12589 @multitable @columnfractions .4 .2 .2 .2
12590 @item Attribute Name @tab Category @tab Index @tab Value
12591 @item @code{^Default_Switches^Default_Switches^}
12592 @tab associative array
12595 @item @code{^Switches^Switches^}
12596 @tab associative array
12602 In addition, package @code{Compiler} has a single string attribute
12603 @code{Local_Configuration_Pragmas} and package @code{Builder} has a single
12604 string attribute @code{Global_Configuration_Pragmas}.
12607 Each simple attribute has a default value: the empty string (for string-valued
12608 attributes) and the empty list (for string list-valued attributes).
12610 An attribute declaration defines a new value for an attribute.
12612 Examples of simple attribute declarations:
12614 @smallexample @c projectfile
12615 for Object_Dir use "objects";
12616 for Source_Dirs use ("units", "test/drivers");
12620 The syntax of a @dfn{simple attribute declaration} is similar to that of an
12621 attribute definition clause in Ada.
12623 Attributes references may be appear in expressions.
12624 The general form for such a reference is @code{<entity>'<attribute>}:
12625 Associative array attributes are functions. Associative
12626 array attribute references must have an argument that is a string literal.
12630 @smallexample @c projectfile
12632 Naming'Dot_Replacement
12633 Imported_Project'Source_Dirs
12634 Imported_Project.Naming'Casing
12635 Builder'^Default_Switches^Default_Switches^("Ada")
12639 The prefix of an attribute may be:
12641 @item @code{project} for an attribute of the current project
12642 @item The name of an existing package of the current project
12643 @item The name of an imported project
12644 @item The name of a parent project that is extended by the current project
12645 @item An expanded name whose prefix is imported/parent project name,
12646 and whose selector is a package name
12651 @smallexample @c projectfile
12654 for Source_Dirs use project'Source_Dirs & "units";
12655 for Source_Dirs use project'Source_Dirs & "test/drivers"
12661 In the first attribute declaration, initially the attribute @code{Source_Dirs}
12662 has the default value: an empty string list. After this declaration,
12663 @code{Source_Dirs} is a string list of one element: @code{"units"}.
12664 After the second attribute declaration @code{Source_Dirs} is a string list of
12665 two elements: @code{"units"} and @code{"test/drivers"}.
12667 Note: this example is for illustration only. In practice,
12668 the project file would contain only one attribute declaration:
12670 @smallexample @c projectfile
12671 for Source_Dirs use ("units", "test/drivers");
12674 @node Associative Array Attributes
12675 @subsection Associative Array Attributes
12678 Some attributes are defined as @emph{associative arrays}. An associative
12679 array may be regarded as a function that takes a string as a parameter
12680 and delivers a string or string list value as its result.
12682 Here are some examples of single associative array attribute associations:
12684 @smallexample @c projectfile
12685 for Body ("main") use "Main.ada";
12686 for ^Switches^Switches^ ("main.ada")
12688 "^-gnatv^-gnatv^");
12689 for ^Switches^Switches^ ("main.ada")
12690 use Builder'^Switches^Switches^ ("main.ada")
12695 Like untyped variables and simple attributes, associative array attributes
12696 may be declared several times. Each declaration supplies a new value for the
12697 attribute, and replaces the previous setting.
12700 An associative array attribute may be declared as a full associative array
12701 declaration, with the value of the same attribute in an imported or extended
12704 @smallexample @c projectfile
12706 for Default_Switches use Default.Builder'Default_Switches;
12711 In this example, @code{Default} must be either a project imported by the
12712 current project, or the project that the current project extends. If the
12713 attribute is in a package (in this case, in package @code{Builder}), the same
12714 package needs to be specified.
12717 A full associative array declaration replaces any other declaration for the
12718 attribute, including other full associative array declaration. Single
12719 associative array associations may be declare after a full associative
12720 declaration, modifying the value for a single association of the attribute.
12722 @node case Constructions
12723 @subsection @code{case} Constructions
12726 A @code{case} construction is used in a project file to effect conditional
12728 Here is a typical example:
12730 @smallexample @c projectfile
12733 type OS_Type is ("GNU/Linux", "Unix", "NT", "VMS");
12735 OS : OS_Type := external ("OS", "GNU/Linux");
12739 package Compiler is
12741 when "GNU/Linux" | "Unix" =>
12742 for ^Default_Switches^Default_Switches^ ("Ada")
12743 use ("^-gnath^-gnath^");
12745 for ^Default_Switches^Default_Switches^ ("Ada")
12746 use ("^-gnatP^-gnatP^");
12755 The syntax of a @code{case} construction is based on the Ada case statement
12756 (although there is no @code{null} construction for empty alternatives).
12758 The case expression must be a typed string variable.
12759 Each alternative comprises the reserved word @code{when}, either a list of
12760 literal strings separated by the @code{"|"} character or the reserved word
12761 @code{others}, and the @code{"=>"} token.
12762 Each literal string must belong to the string type that is the type of the
12764 An @code{others} alternative, if present, must occur last.
12766 After each @code{=>}, there are zero or more constructions. The only
12767 constructions allowed in a case construction are other case constructions,
12768 attribute declarations and variable declarations. String type declarations and
12769 package declarations are not allowed. Variable declarations are restricted to
12770 variables that have already been declared before the case construction.
12772 The value of the case variable is often given by an external reference
12773 (@pxref{External References in Project Files}).
12775 @c ****************************************
12776 @c * Objects and Sources in Project Files *
12777 @c ****************************************
12779 @node Objects and Sources in Project Files
12780 @section Objects and Sources in Project Files
12783 * Object Directory::
12785 * Source Directories::
12786 * Source File Names::
12790 Each project has exactly one object directory and one or more source
12791 directories. The source directories must contain at least one source file,
12792 unless the project file explicitly specifies that no source files are present
12793 (@pxref{Source File Names}).
12795 @node Object Directory
12796 @subsection Object Directory
12799 The object directory for a project is the directory containing the compiler's
12800 output (such as @file{ALI} files and object files) for the project's immediate
12803 The object directory is given by the value of the attribute @code{Object_Dir}
12804 in the project file.
12806 @smallexample @c projectfile
12807 for Object_Dir use "objects";
12811 The attribute @code{Object_Dir} has a string value, the path name of the object
12812 directory. The path name may be absolute or relative to the directory of the
12813 project file. This directory must already exist, and be readable and writable.
12815 By default, when the attribute @code{Object_Dir} is not given an explicit value
12816 or when its value is the empty string, the object directory is the same as the
12817 directory containing the project file.
12819 @node Exec Directory
12820 @subsection Exec Directory
12823 The exec directory for a project is the directory containing the executables
12824 for the project's main subprograms.
12826 The exec directory is given by the value of the attribute @code{Exec_Dir}
12827 in the project file.
12829 @smallexample @c projectfile
12830 for Exec_Dir use "executables";
12834 The attribute @code{Exec_Dir} has a string value, the path name of the exec
12835 directory. The path name may be absolute or relative to the directory of the
12836 project file. This directory must already exist, and be writable.
12838 By default, when the attribute @code{Exec_Dir} is not given an explicit value
12839 or when its value is the empty string, the exec directory is the same as the
12840 object directory of the project file.
12842 @node Source Directories
12843 @subsection Source Directories
12846 The source directories of a project are specified by the project file
12847 attribute @code{Source_Dirs}.
12849 This attribute's value is a string list. If the attribute is not given an
12850 explicit value, then there is only one source directory, the one where the
12851 project file resides.
12853 A @code{Source_Dirs} attribute that is explicitly defined to be the empty list,
12856 @smallexample @c projectfile
12857 for Source_Dirs use ();
12861 indicates that the project contains no source files.
12863 Otherwise, each string in the string list designates one or more
12864 source directories.
12866 @smallexample @c projectfile
12867 for Source_Dirs use ("sources", "test/drivers");
12871 If a string in the list ends with @code{"/**"}, then the directory whose path
12872 name precedes the two asterisks, as well as all its subdirectories
12873 (recursively), are source directories.
12875 @smallexample @c projectfile
12876 for Source_Dirs use ("/system/sources/**");
12880 Here the directory @code{/system/sources} and all of its subdirectories
12881 (recursively) are source directories.
12883 To specify that the source directories are the directory of the project file
12884 and all of its subdirectories, you can declare @code{Source_Dirs} as follows:
12885 @smallexample @c projectfile
12886 for Source_Dirs use ("./**");
12890 Each of the source directories must exist and be readable.
12892 @node Source File Names
12893 @subsection Source File Names
12896 In a project that contains source files, their names may be specified by the
12897 attributes @code{Source_Files} (a string list) or @code{Source_List_File}
12898 (a string). Source file names never include any directory information.
12900 If the attribute @code{Source_Files} is given an explicit value, then each
12901 element of the list is a source file name.
12903 @smallexample @c projectfile
12904 for Source_Files use ("main.adb");
12905 for Source_Files use ("main.adb", "pack1.ads", "pack2.adb");
12909 If the attribute @code{Source_Files} is not given an explicit value,
12910 but the attribute @code{Source_List_File} is given a string value,
12911 then the source file names are contained in the text file whose path name
12912 (absolute or relative to the directory of the project file) is the
12913 value of the attribute @code{Source_List_File}.
12915 Each line in the file that is not empty or is not a comment
12916 contains a source file name.
12918 @smallexample @c projectfile
12919 for Source_List_File use "source_list.txt";
12923 By default, if neither the attribute @code{Source_Files} nor the attribute
12924 @code{Source_List_File} is given an explicit value, then each file in the
12925 source directories that conforms to the project's naming scheme
12926 (@pxref{Naming Schemes}) is an immediate source of the project.
12928 A warning is issued if both attributes @code{Source_Files} and
12929 @code{Source_List_File} are given explicit values. In this case, the attribute
12930 @code{Source_Files} prevails.
12932 Each source file name must be the name of one existing source file
12933 in one of the source directories.
12935 A @code{Source_Files} attribute whose value is an empty list
12936 indicates that there are no source files in the project.
12938 If the order of the source directories is known statically, that is if
12939 @code{"/**"} is not used in the string list @code{Source_Dirs}, then there may
12940 be several files with the same source file name. In this case, only the file
12941 in the first directory is considered as an immediate source of the project
12942 file. If the order of the source directories is not known statically, it is
12943 an error to have several files with the same source file name.
12945 Projects can be specified to have no Ada source
12946 files: the value of (@code{Source_Dirs} or @code{Source_Files} may be an empty
12947 list, or the @code{"Ada"} may be absent from @code{Languages}:
12949 @smallexample @c projectfile
12950 for Source_Dirs use ();
12951 for Source_Files use ();
12952 for Languages use ("C", "C++");
12956 Otherwise, a project must contain at least one immediate source.
12958 Projects with no source files are useful as template packages
12959 (@pxref{Packages in Project Files}) for other projects; in particular to
12960 define a package @code{Naming} (@pxref{Naming Schemes}).
12962 @c ****************************
12963 @c * Importing Projects *
12964 @c ****************************
12966 @node Importing Projects
12967 @section Importing Projects
12968 @cindex @code{ADA_PROJECT_PATH}
12971 An immediate source of a project P may depend on source files that
12972 are neither immediate sources of P nor in the predefined library.
12973 To get this effect, P must @emph{import} the projects that contain the needed
12976 @smallexample @c projectfile
12978 with "project1", "utilities.gpr";
12979 with "/namings/apex.gpr";
12986 As can be seen in this example, the syntax for importing projects is similar
12987 to the syntax for importing compilation units in Ada. However, project files
12988 use literal strings instead of names, and the @code{with} clause identifies
12989 project files rather than packages.
12991 Each literal string is the file name or path name (absolute or relative) of a
12992 project file. If a string corresponds to a file name, with no path or a
12993 relative path, then its location is determined by the @emph{project path}. The
12994 latter can be queried using @code{gnatls -v}. It contains:
12998 In first position, the directory containing the current project file.
13000 In last position, the default project directory. This default project directory
13001 is part of the GNAT installation and is the standard place to install project
13002 files giving access to standard support libraries.
13004 @ref{Installing a library}
13008 In between, all the directories referenced in the
13009 ^environment variable^logical name^ @env{ADA_PROJECT_PATH} if it exists.
13013 If a relative pathname is used, as in
13015 @smallexample @c projectfile
13020 then the full path for the project is constructed by concatenating this
13021 relative path to those in the project path, in order, until a matching file is
13022 found. Any symbolic link will be fully resolved in the directory of the
13023 importing project file before the imported project file is examined.
13025 If the @code{with}'ed project file name does not have an extension,
13026 the default is @file{^.gpr^.GPR^}. If a file with this extension is not found,
13027 then the file name as specified in the @code{with} clause (no extension) will
13028 be used. In the above example, if a file @code{project1.gpr} is found, then it
13029 will be used; otherwise, if a file @code{^project1^PROJECT1^} exists
13030 then it will be used; if neither file exists, this is an error.
13032 A warning is issued if the name of the project file does not match the
13033 name of the project; this check is case insensitive.
13035 Any source file that is an immediate source of the imported project can be
13036 used by the immediate sources of the importing project, transitively. Thus
13037 if @code{A} imports @code{B}, and @code{B} imports @code{C}, the immediate
13038 sources of @code{A} may depend on the immediate sources of @code{C}, even if
13039 @code{A} does not import @code{C} explicitly. However, this is not recommended,
13040 because if and when @code{B} ceases to import @code{C}, some sources in
13041 @code{A} will no longer compile.
13043 A side effect of this capability is that normally cyclic dependencies are not
13044 permitted: if @code{A} imports @code{B} (directly or indirectly) then @code{B}
13045 is not allowed to import @code{A}. However, there are cases when cyclic
13046 dependencies would be beneficial. For these cases, another form of import
13047 between projects exists, the @code{limited with}: a project @code{A} that
13048 imports a project @code{B} with a straight @code{with} may also be imported,
13049 directly or indirectly, by @code{B} on the condition that imports from @code{B}
13050 to @code{A} include at least one @code{limited with}.
13052 @smallexample @c 0projectfile
13058 limited with "../a/a.gpr";
13066 limited with "../a/a.gpr";
13072 In the above legal example, there are two project cycles:
13075 @item A -> C -> D -> A
13079 In each of these cycle there is one @code{limited with}: import of @code{A}
13080 from @code{B} and import of @code{A} from @code{D}.
13082 The difference between straight @code{with} and @code{limited with} is that
13083 the name of a project imported with a @code{limited with} cannot be used in the
13084 project that imports it. In particular, its packages cannot be renamed and
13085 its variables cannot be referred to.
13087 An exception to the above rules for @code{limited with} is that for the main
13088 project specified to @command{gnatmake} or to the @command{GNAT} driver a
13089 @code{limited with} is equivalent to a straight @code{with}. For example,
13090 in the example above, projects @code{B} and @code{D} could not be main
13091 projects for @command{gnatmake} or to the @command{GNAT} driver, because they
13092 each have a @code{limited with} that is the only one in a cycle of importing
13095 @c *********************
13096 @c * Project Extension *
13097 @c *********************
13099 @node Project Extension
13100 @section Project Extension
13103 During development of a large system, it is sometimes necessary to use
13104 modified versions of some of the source files, without changing the original
13105 sources. This can be achieved through the @emph{project extension} facility.
13107 @smallexample @c projectfile
13108 project Modified_Utilities extends "/baseline/utilities.gpr" is @dots{}
13112 A project extension declaration introduces an extending project
13113 (the @emph{child}) and a project being extended (the @emph{parent}).
13115 By default, a child project inherits all the sources of its parent.
13116 However, inherited sources can be overridden: a unit in a parent is hidden
13117 by a unit of the same name in the child.
13119 Inherited sources are considered to be sources (but not immediate sources)
13120 of the child project; see @ref{Project File Syntax}.
13122 An inherited source file retains any switches specified in the parent project.
13124 For example if the project @code{Utilities} contains the spec and the
13125 body of an Ada package @code{Util_IO}, then the project
13126 @code{Modified_Utilities} can contain a new body for package @code{Util_IO}.
13127 The original body of @code{Util_IO} will not be considered in program builds.
13128 However, the package spec will still be found in the project
13131 A child project can have only one parent, except when it is qualified as
13132 abstract. But it may import any number of other projects.
13134 A project is not allowed to import directly or indirectly at the same time a
13135 child project and any of its ancestors.
13137 @c *******************************
13138 @c * Project Hierarchy Extension *
13139 @c *******************************
13141 @node Project Hierarchy Extension
13142 @section Project Hierarchy Extension
13145 When extending a large system spanning multiple projects, it is often
13146 inconvenient to extend every project in the hierarchy that is impacted by a
13147 small change introduced. In such cases, it is possible to create a virtual
13148 extension of entire hierarchy using @code{extends all} relationship.
13150 When the project is extended using @code{extends all} inheritance, all projects
13151 that are imported by it, both directly and indirectly, are considered virtually
13152 extended. That is, the Project Manager creates "virtual projects"
13153 that extend every project in the hierarchy; all these virtual projects have
13154 no sources of their own and have as object directory the object directory of
13155 the root of "extending all" project.
13157 It is possible to explicitly extend one or more projects in the hierarchy
13158 in order to modify the sources. These extending projects must be imported by
13159 the "extending all" project, which will replace the corresponding virtual
13160 projects with the explicit ones.
13162 When building such a project hierarchy extension, the Project Manager will
13163 ensure that both modified sources and sources in virtual extending projects
13164 that depend on them, are recompiled.
13166 By means of example, consider the following hierarchy of projects.
13170 project A, containing package P1
13172 project B importing A and containing package P2 which depends on P1
13174 project C importing B and containing package P3 which depends on P2
13178 We want to modify packages P1 and P3.
13180 This project hierarchy will need to be extended as follows:
13184 Create project A1 that extends A, placing modified P1 there:
13186 @smallexample @c 0projectfile
13187 project A1 extends "(@dots{})/A" is
13192 Create project C1 that "extends all" C and imports A1, placing modified
13195 @smallexample @c 0projectfile
13196 with "(@dots{})/A1";
13197 project C1 extends all "(@dots{})/C" is
13202 When you build project C1, your entire modified project space will be
13203 recompiled, including the virtual project B1 that has been impacted by the
13204 "extending all" inheritance of project C.
13206 Note that if a Library Project in the hierarchy is virtually extended,
13207 the virtual project that extends the Library Project is not a Library Project.
13209 @c ****************************************
13210 @c * External References in Project Files *
13211 @c ****************************************
13213 @node External References in Project Files
13214 @section External References in Project Files
13217 A project file may contain references to external variables; such references
13218 are called @emph{external references}.
13220 An external variable is either defined as part of the environment (an
13221 environment variable in Unix, for example) or else specified on the command
13222 line via the @option{^-X^/EXTERNAL_REFERENCE=^@emph{vbl}=@emph{value}} switch.
13223 If both, then the command line value is used.
13225 The value of an external reference is obtained by means of the built-in
13226 function @code{external}, which returns a string value.
13227 This function has two forms:
13229 @item @code{external (external_variable_name)}
13230 @item @code{external (external_variable_name, default_value)}
13234 Each parameter must be a string literal. For example:
13236 @smallexample @c projectfile
13238 external ("OS", "GNU/Linux")
13242 In the form with one parameter, the function returns the value of
13243 the external variable given as parameter. If this name is not present in the
13244 environment, the function returns an empty string.
13246 In the form with two string parameters, the second argument is
13247 the value returned when the variable given as the first argument is not
13248 present in the environment. In the example above, if @code{"OS"} is not
13249 the name of ^an environment variable^a logical name^ and is not passed on
13250 the command line, then the returned value is @code{"GNU/Linux"}.
13252 An external reference may be part of a string expression or of a string
13253 list expression, and can therefore appear in a variable declaration or
13254 an attribute declaration.
13256 @smallexample @c projectfile
13258 type Mode_Type is ("Debug", "Release");
13259 Mode : Mode_Type := external ("MODE");
13266 @c *****************************
13267 @c * Packages in Project Files *
13268 @c *****************************
13270 @node Packages in Project Files
13271 @section Packages in Project Files
13274 A @emph{package} defines the settings for project-aware tools within a
13276 For each such tool one can declare a package; the names for these
13277 packages are preset (@pxref{Packages}).
13278 A package may contain variable declarations, attribute declarations, and case
13281 @smallexample @c projectfile
13284 package Builder is -- used by gnatmake
13285 for ^Default_Switches^Default_Switches^ ("Ada")
13294 The syntax of package declarations mimics that of package in Ada.
13296 Most of the packages have an attribute
13297 @code{^Default_Switches^Default_Switches^}.
13298 This attribute is an associative array, and its value is a string list.
13299 The index of the associative array is the name of a programming language (case
13300 insensitive). This attribute indicates the ^switch^switch^
13301 or ^switches^switches^ to be used
13302 with the corresponding tool.
13304 Some packages also have another attribute, @code{^Switches^Switches^},
13305 an associative array whose value is a string list.
13306 The index is the name of a source file.
13307 This attribute indicates the ^switch^switch^
13308 or ^switches^switches^ to be used by the corresponding
13309 tool when dealing with this specific file.
13311 Further information on these ^switch^switch^-related attributes is found in
13312 @ref{^Switches^Switches^ and Project Files}.
13314 A package may be declared as a @emph{renaming} of another package; e.g., from
13315 the project file for an imported project.
13317 @smallexample @c projectfile
13319 with "/global/apex.gpr";
13321 package Naming renames Apex.Naming;
13328 Packages that are renamed in other project files often come from project files
13329 that have no sources: they are just used as templates. Any modification in the
13330 template will be reflected automatically in all the project files that rename
13331 a package from the template.
13333 In addition to the tool-oriented packages, you can also declare a package
13334 named @code{Naming} to establish specialized source file naming conventions
13335 (@pxref{Naming Schemes}).
13337 @c ************************************
13338 @c * Variables from Imported Projects *
13339 @c ************************************
13341 @node Variables from Imported Projects
13342 @section Variables from Imported Projects
13345 An attribute or variable defined in an imported or parent project can
13346 be used in expressions in the importing / extending project.
13347 Such an attribute or variable is denoted by an expanded name whose prefix
13348 is either the name of the project or the expanded name of a package within
13351 @smallexample @c projectfile
13354 project Main extends "base" is
13355 Var1 := Imported.Var;
13356 Var2 := Base.Var & ".new";
13361 for ^Default_Switches^Default_Switches^ ("Ada")
13362 use Imported.Builder'Ada_^Switches^Switches^ &
13363 "^-gnatg^-gnatg^" &
13369 package Compiler is
13370 for ^Default_Switches^Default_Switches^ ("Ada")
13371 use Base.Compiler'Ada_^Switches^Switches^;
13382 The value of @code{Var1} is a copy of the variable @code{Var} defined
13383 in the project file @file{"imported.gpr"}
13385 the value of @code{Var2} is a copy of the value of variable @code{Var}
13386 defined in the project file @file{base.gpr}, concatenated with @code{".new"}
13388 attribute @code{^Default_Switches^Default_Switches^ ("Ada")} in package
13389 @code{Builder} is a string list that includes in its value a copy of the value
13390 of @code{Ada_^Switches^Switches^} defined in the @code{Builder} package
13391 in project file @file{imported.gpr} plus two new elements:
13392 @option{"^-gnatg^-gnatg^"}
13393 and @option{"^-v^-v^"};
13395 attribute @code{^Default_Switches^Default_Switches^ ("Ada")} in package
13396 @code{Compiler} is a copy of the variable @code{Ada_^Switches^Switches^}
13397 defined in the @code{Compiler} package in project file @file{base.gpr},
13398 the project being extended.
13401 @c ******************
13402 @c * Naming Schemes *
13403 @c ******************
13405 @node Naming Schemes
13406 @section Naming Schemes
13409 Sometimes an Ada software system is ported from a foreign compilation
13410 environment to GNAT, and the file names do not use the default GNAT
13411 conventions. Instead of changing all the file names (which for a variety
13412 of reasons might not be possible), you can define the relevant file
13413 naming scheme in the @code{Naming} package in your project file.
13416 Note that the use of pragmas described in
13417 @ref{Alternative File Naming Schemes} by mean of a configuration
13418 pragmas file is not supported when using project files. You must use
13419 the features described in this paragraph. You can however use specify
13420 other configuration pragmas (@pxref{Specifying Configuration Pragmas}).
13423 For example, the following
13424 package models the Apex file naming rules:
13426 @smallexample @c projectfile
13429 for Casing use "lowercase";
13430 for Dot_Replacement use ".";
13431 for Spec_Suffix ("Ada") use ".1.ada";
13432 for Body_Suffix ("Ada") use ".2.ada";
13439 For example, the following package models the HP Ada file naming rules:
13441 @smallexample @c projectfile
13444 for Casing use "lowercase";
13445 for Dot_Replacement use "__";
13446 for Spec_Suffix ("Ada") use "_.^ada^ada^";
13447 for Body_Suffix ("Ada") use ".^ada^ada^";
13453 (Note that @code{Casing} is @code{"lowercase"} because GNAT gets the file
13454 names in lower case)
13458 You can define the following attributes in package @code{Naming}:
13462 @item @code{Casing}
13463 This must be a string with one of the three values @code{"lowercase"},
13464 @code{"uppercase"} or @code{"mixedcase"}; these strings are case insensitive.
13467 If @code{Casing} is not specified, then the default is @code{"lowercase"}.
13469 @item @code{Dot_Replacement}
13470 This must be a string whose value satisfies the following conditions:
13473 @item It must not be empty
13474 @item It cannot start or end with an alphanumeric character
13475 @item It cannot be a single underscore
13476 @item It cannot start with an underscore followed by an alphanumeric
13477 @item It cannot contain a dot @code{'.'} except if the entire string
13482 If @code{Dot_Replacement} is not specified, then the default is @code{"-"}.
13484 @item @code{Spec_Suffix}
13485 This is an associative array (indexed by the programming language name, case
13486 insensitive) whose value is a string that must satisfy the following
13490 @item It must not be empty
13491 @item It must include at least one dot
13494 If @code{Spec_Suffix ("Ada")} is not specified, then the default is
13495 @code{"^.ads^.ADS^"}.
13497 @item @code{Body_Suffix}
13498 This is an associative array (indexed by the programming language name, case
13499 insensitive) whose value is a string that must satisfy the following
13503 @item It must not be empty
13504 @item It must include at least one dot
13505 @item It cannot be the same as @code{Spec_Suffix ("Ada")}
13508 If @code{Body_Suffix ("Ada")} and @code{Spec_Suffix ("Ada")} end with the
13509 same string, then a file name that ends with the longest of these two suffixes
13510 will be a body if the longest suffix is @code{Body_Suffix ("Ada")} or a spec
13511 if the longest suffix is @code{Spec_Suffix ("Ada")}.
13513 If @code{Body_Suffix ("Ada")} is not specified, then the default is
13514 @code{"^.adb^.ADB^"}.
13516 @item @code{Separate_Suffix}
13517 This must be a string whose value satisfies the same conditions as
13518 @code{Body_Suffix}. The same "longest suffix" rules apply.
13521 If @code{Separate_Suffix ("Ada")} is not specified, then it defaults to same
13522 value as @code{Body_Suffix ("Ada")}.
13526 You can use the associative array attribute @code{Spec} to define
13527 the source file name for an individual Ada compilation unit's spec. The array
13528 index must be a string literal that identifies the Ada unit (case insensitive).
13529 The value of this attribute must be a string that identifies the file that
13530 contains this unit's spec (case sensitive or insensitive depending on the
13533 @smallexample @c projectfile
13534 for Spec ("MyPack.MyChild") use "mypack.mychild.spec";
13539 You can use the associative array attribute @code{Body} to
13540 define the source file name for an individual Ada compilation unit's body
13541 (possibly a subunit). The array index must be a string literal that identifies
13542 the Ada unit (case insensitive). The value of this attribute must be a string
13543 that identifies the file that contains this unit's body or subunit (case
13544 sensitive or insensitive depending on the operating system).
13546 @smallexample @c projectfile
13547 for Body ("MyPack.MyChild") use "mypack.mychild.body";
13551 @c ********************
13552 @c * Library Projects *
13553 @c ********************
13555 @node Library Projects
13556 @section Library Projects
13559 @emph{Library projects} are projects whose object code is placed in a library.
13560 (Note that this facility is not yet supported on all platforms)
13562 To create a library project, you need to define in its project file
13563 two project-level attributes: @code{Library_Name} and @code{Library_Dir}.
13564 Additionally, you may define other library-related attributes such as
13565 @code{Library_Kind}, @code{Library_Version}, @code{Library_Interface},
13566 @code{Library_Auto_Init}, @code{Library_Options} and @code{Library_GCC}.
13568 The @code{Library_Name} attribute has a string value. There is no restriction
13569 on the name of a library. It is the responsibility of the developer to
13570 choose a name that will be accepted by the platform. It is recommended to
13571 choose names that could be Ada identifiers; such names are almost guaranteed
13572 to be acceptable on all platforms.
13574 The @code{Library_Dir} attribute has a string value that designates the path
13575 (absolute or relative) of the directory where the library will reside.
13576 It must designate an existing directory, and this directory must be writable,
13577 different from the project's object directory and from any source directory
13578 in the project tree.
13580 If both @code{Library_Name} and @code{Library_Dir} are specified and
13581 are legal, then the project file defines a library project. The optional
13582 library-related attributes are checked only for such project files.
13584 The @code{Library_Kind} attribute has a string value that must be one of the
13585 following (case insensitive): @code{"static"}, @code{"dynamic"} or
13586 @code{"relocatable"} (which is a synonym for @code{"dynamic"}). If this
13587 attribute is not specified, the library is a static library, that is
13588 an archive of object files that can be potentially linked into a
13589 static executable. Otherwise, the library may be dynamic or
13590 relocatable, that is a library that is loaded only at the start of execution.
13592 If you need to build both a static and a dynamic library, you should use two
13593 different object directories, since in some cases some extra code needs to
13594 be generated for the latter. For such cases, it is recommended to either use
13595 two different project files, or a single one which uses external variables
13596 to indicate what kind of library should be build.
13598 The @code{Library_ALI_Dir} attribute may be specified to indicate the
13599 directory where the ALI files of the library will be copied. When it is
13600 not specified, the ALI files are copied to the directory specified in
13601 attribute @code{Library_Dir}. The directory specified by @code{Library_ALI_Dir}
13602 must be writable and different from the project's object directory and from
13603 any source directory in the project tree.
13605 The @code{Library_Version} attribute has a string value whose interpretation
13606 is platform dependent. It has no effect on VMS and Windows. On Unix, it is
13607 used only for dynamic/relocatable libraries as the internal name of the
13608 library (the @code{"soname"}). If the library file name (built from the
13609 @code{Library_Name}) is different from the @code{Library_Version}, then the
13610 library file will be a symbolic link to the actual file whose name will be
13611 @code{Library_Version}.
13615 @smallexample @c projectfile
13621 for Library_Dir use "lib_dir";
13622 for Library_Name use "dummy";
13623 for Library_Kind use "relocatable";
13624 for Library_Version use "libdummy.so." & Version;
13631 Directory @file{lib_dir} will contain the internal library file whose name
13632 will be @file{libdummy.so.1}, and @file{libdummy.so} will be a symbolic link to
13633 @file{libdummy.so.1}.
13635 When @command{gnatmake} detects that a project file
13636 is a library project file, it will check all immediate sources of the project
13637 and rebuild the library if any of the sources have been recompiled.
13639 Standard project files can import library project files. In such cases,
13640 the libraries will only be rebuilt if some of its sources are recompiled
13641 because they are in the closure of some other source in an importing project.
13642 Sources of the library project files that are not in such a closure will
13643 not be checked, unless the full library is checked, because one of its sources
13644 needs to be recompiled.
13646 For instance, assume the project file @code{A} imports the library project file
13647 @code{L}. The immediate sources of A are @file{a1.adb}, @file{a2.ads} and
13648 @file{a2.adb}. The immediate sources of L are @file{l1.ads}, @file{l1.adb},
13649 @file{l2.ads}, @file{l2.adb}.
13651 If @file{l1.adb} has been modified, then the library associated with @code{L}
13652 will be rebuilt when compiling all the immediate sources of @code{A} only
13653 if @file{a1.ads}, @file{a2.ads} or @file{a2.adb} includes a statement
13656 To be sure that all the sources in the library associated with @code{L} are
13657 up to date, and that all the sources of project @code{A} are also up to date,
13658 the following two commands needs to be used:
13665 When a library is built or rebuilt, an attempt is made first to delete all
13666 files in the library directory.
13667 All @file{ALI} files will also be copied from the object directory to the
13668 library directory. To build executables, @command{gnatmake} will use the
13669 library rather than the individual object files.
13672 It is also possible to create library project files for third-party libraries
13673 that are precompiled and cannot be compiled locally thanks to the
13674 @code{externally_built} attribute. (See @ref{Installing a library}).
13677 @c *******************************
13678 @c * Stand-alone Library Projects *
13679 @c *******************************
13681 @node Stand-alone Library Projects
13682 @section Stand-alone Library Projects
13685 A Stand-alone Library is a library that contains the necessary code to
13686 elaborate the Ada units that are included in the library. A Stand-alone
13687 Library is suitable to be used in an executable when the main is not
13688 in Ada. However, Stand-alone Libraries may also be used with an Ada main
13691 A Stand-alone Library Project is a Library Project where the library is
13692 a Stand-alone Library.
13694 To be a Stand-alone Library Project, in addition to the two attributes
13695 that make a project a Library Project (@code{Library_Name} and
13696 @code{Library_Dir}, see @ref{Library Projects}), the attribute
13697 @code{Library_Interface} must be defined.
13699 @smallexample @c projectfile
13701 for Library_Dir use "lib_dir";
13702 for Library_Name use "dummy";
13703 for Library_Interface use ("int1", "int1.child");
13707 Attribute @code{Library_Interface} has a nonempty string list value,
13708 each string in the list designating a unit contained in an immediate source
13709 of the project file.
13711 When a Stand-alone Library is built, first the binder is invoked to build
13712 a package whose name depends on the library name
13713 (^b~dummy.ads/b^B$DUMMY.ADS/B^ in the example above).
13714 This binder-generated package includes initialization and
13715 finalization procedures whose
13716 names depend on the library name (dummyinit and dummyfinal in the example
13717 above). The object corresponding to this package is included in the library.
13719 A dynamic or relocatable Stand-alone Library is automatically initialized
13720 if automatic initialization of Stand-alone Libraries is supported on the
13721 platform and if attribute @code{Library_Auto_Init} is not specified or
13722 is specified with the value "true". A static Stand-alone Library is never
13723 automatically initialized.
13725 Single string attribute @code{Library_Auto_Init} may be specified with only
13726 two possible values: "false" or "true" (case-insensitive). Specifying
13727 "false" for attribute @code{Library_Auto_Init} will prevent automatic
13728 initialization of dynamic or relocatable libraries.
13730 When a non-automatically initialized Stand-alone Library is used
13731 in an executable, its initialization procedure must be called before
13732 any service of the library is used.
13733 When the main subprogram is in Ada, it may mean that the initialization
13734 procedure has to be called during elaboration of another package.
13736 For a Stand-Alone Library, only the @file{ALI} files of the Interface Units
13737 (those that are listed in attribute @code{Library_Interface}) are copied to
13738 the Library Directory. As a consequence, only the Interface Units may be
13739 imported from Ada units outside of the library. If other units are imported,
13740 the binding phase will fail.
13742 When a Stand-Alone Library is bound, the switches that are specified in
13743 the attribute @code{Default_Switches ("Ada")} in package @code{Binder} are
13744 used in the call to @command{gnatbind}.
13746 The string list attribute @code{Library_Options} may be used to specified
13747 additional switches to the call to @command{gcc} to link the library.
13749 The attribute @code{Library_Src_Dir}, may be specified for a
13750 Stand-Alone Library. @code{Library_Src_Dir} is a simple attribute that has a
13751 single string value. Its value must be the path (absolute or relative to the
13752 project directory) of an existing directory. This directory cannot be the
13753 object directory or one of the source directories, but it can be the same as
13754 the library directory. The sources of the Interface
13755 Units of the library, necessary to an Ada client of the library, will be
13756 copied to the designated directory, called Interface Copy directory.
13757 These sources includes the specs of the Interface Units, but they may also
13758 include bodies and subunits, when pragmas @code{Inline} or @code{Inline_Always}
13759 are used, or when there is a generic units in the spec. Before the sources
13760 are copied to the Interface Copy directory, an attempt is made to delete all
13761 files in the Interface Copy directory.
13763 @c *************************************
13764 @c * Switches Related to Project Files *
13765 @c *************************************
13766 @node Switches Related to Project Files
13767 @section Switches Related to Project Files
13770 The following switches are used by GNAT tools that support project files:
13774 @item ^-P^/PROJECT_FILE=^@var{project}
13775 @cindex @option{^-P^/PROJECT_FILE^} (any project-aware tool)
13776 Indicates the name of a project file. This project file will be parsed with
13777 the verbosity indicated by @option{^-vP^MESSAGE_PROJECT_FILES=^@emph{x}},
13778 if any, and using the external references indicated
13779 by @option{^-X^/EXTERNAL_REFERENCE^} switches, if any.
13781 There may zero, one or more spaces between @option{-P} and @var{project}.
13785 There must be only one @option{^-P^/PROJECT_FILE^} switch on the command line.
13788 Since the Project Manager parses the project file only after all the switches
13789 on the command line are checked, the order of the switches
13790 @option{^-P^/PROJECT_FILE^},
13791 @option{^-vP^/MESSAGES_PROJECT_FILE=^@emph{x}}
13792 or @option{^-X^/EXTERNAL_REFERENCE^} is not significant.
13794 @item ^-X^/EXTERNAL_REFERENCE=^@var{name=value}
13795 @cindex @option{^-X^/EXTERNAL_REFERENCE^} (any project-aware tool)
13796 Indicates that external variable @var{name} has the value @var{value}.
13797 The Project Manager will use this value for occurrences of
13798 @code{external(name)} when parsing the project file.
13802 If @var{name} or @var{value} includes a space, then @var{name=value} should be
13803 put between quotes.
13811 Several @option{^-X^/EXTERNAL_REFERENCE^} switches can be used simultaneously.
13812 If several @option{^-X^/EXTERNAL_REFERENCE^} switches specify the same
13813 @var{name}, only the last one is used.
13816 An external variable specified with a @option{^-X^/EXTERNAL_REFERENCE^} switch
13817 takes precedence over the value of the same name in the environment.
13819 @item ^-vP^/MESSAGES_PROJECT_FILE=^@emph{x}
13820 @cindex @option{^-vP^/MESSAGES_PROJECT_FILE^} (any project-aware tool)
13821 Indicates the verbosity of the parsing of GNAT project files.
13824 @option{-vP0} means Default;
13825 @option{-vP1} means Medium;
13826 @option{-vP2} means High.
13830 There are three possible options for this qualifier: DEFAULT, MEDIUM and
13835 The default is ^Default^DEFAULT^: no output for syntactically correct
13838 If several @option{^-vP^/MESSAGES_PROJECT_FILE=^@emph{x}} switches are present,
13839 only the last one is used.
13841 @item ^-aP^/ADD_PROJECT_SEARCH_DIR=^<dir>
13842 @cindex @option{^-aP^/ADD_PROJECT_SEARCH_DIR=^} (any project-aware tool)
13843 Add directory <dir> at the beginning of the project search path, in order,
13844 after the current working directory.
13848 @cindex @option{-eL} (any project-aware tool)
13849 Follow all symbolic links when processing project files.
13852 @item ^--subdirs^/SUBDIRS^=<subdir>
13853 @cindex @option{^--subdirs^/SUBDIRS^=} (gnatmake and gnatclean)
13854 This switch is recognized by gnatmake and gnatclean. It indicate that the real
13855 directories (except the source directories) are the subdirectories <subdir>
13856 of the directories specified in the project files. This applies in particular
13857 to object directories, library directories and exec directories. If the
13858 subdirectories do not exist, they are created automatically.
13862 @c **********************************
13863 @c * Tools Supporting Project Files *
13864 @c **********************************
13866 @node Tools Supporting Project Files
13867 @section Tools Supporting Project Files
13870 * gnatmake and Project Files::
13871 * The GNAT Driver and Project Files::
13874 @node gnatmake and Project Files
13875 @subsection gnatmake and Project Files
13878 This section covers several topics related to @command{gnatmake} and
13879 project files: defining ^switches^switches^ for @command{gnatmake}
13880 and for the tools that it invokes; specifying configuration pragmas;
13881 the use of the @code{Main} attribute; building and rebuilding library project
13885 * ^Switches^Switches^ and Project Files::
13886 * Specifying Configuration Pragmas::
13887 * Project Files and Main Subprograms::
13888 * Library Project Files::
13891 @node ^Switches^Switches^ and Project Files
13892 @subsubsection ^Switches^Switches^ and Project Files
13895 It is not currently possible to specify VMS style qualifiers in the project
13896 files; only Unix style ^switches^switches^ may be specified.
13900 For each of the packages @code{Builder}, @code{Compiler}, @code{Binder}, and
13901 @code{Linker}, you can specify a @code{^Default_Switches^Default_Switches^}
13902 attribute, a @code{^Switches^Switches^} attribute, or both;
13903 as their names imply, these ^switch^switch^-related
13904 attributes affect the ^switches^switches^ that are used for each of these GNAT
13906 @command{gnatmake} is invoked. As will be explained below, these
13907 component-specific ^switches^switches^ precede
13908 the ^switches^switches^ provided on the @command{gnatmake} command line.
13910 The @code{^Default_Switches^Default_Switches^} attribute is an associative
13911 array indexed by language name (case insensitive) whose value is a string list.
13914 @smallexample @c projectfile
13916 package Compiler is
13917 for ^Default_Switches^Default_Switches^ ("Ada")
13918 use ("^-gnaty^-gnaty^",
13925 The @code{^Switches^Switches^} attribute is also an associative array,
13926 indexed by a file name (which may or may not be case sensitive, depending
13927 on the operating system) whose value is a string list. For example:
13929 @smallexample @c projectfile
13932 for ^Switches^Switches^ ("main1.adb")
13934 for ^Switches^Switches^ ("main2.adb")
13941 For the @code{Builder} package, the file names must designate source files
13942 for main subprograms. For the @code{Binder} and @code{Linker} packages, the
13943 file names must designate @file{ALI} or source files for main subprograms.
13944 In each case just the file name without an explicit extension is acceptable.
13946 For each tool used in a program build (@command{gnatmake}, the compiler, the
13947 binder, and the linker), the corresponding package @dfn{contributes} a set of
13948 ^switches^switches^ for each file on which the tool is invoked, based on the
13949 ^switch^switch^-related attributes defined in the package.
13950 In particular, the ^switches^switches^
13951 that each of these packages contributes for a given file @var{f} comprise:
13955 the value of attribute @code{^Switches^Switches^ (@var{f})},
13956 if it is specified in the package for the given file,
13958 otherwise, the value of @code{^Default_Switches^Default_Switches^ ("Ada")},
13959 if it is specified in the package.
13963 If neither of these attributes is defined in the package, then the package does
13964 not contribute any ^switches^switches^ for the given file.
13966 When @command{gnatmake} is invoked on a file, the ^switches^switches^ comprise
13967 two sets, in the following order: those contributed for the file
13968 by the @code{Builder} package;
13969 and the switches passed on the command line.
13971 When @command{gnatmake} invokes a tool (compiler, binder, linker) on a file,
13972 the ^switches^switches^ passed to the tool comprise three sets,
13973 in the following order:
13977 the applicable ^switches^switches^ contributed for the file
13978 by the @code{Builder} package in the project file supplied on the command line;
13981 those contributed for the file by the package (in the relevant project file --
13982 see below) corresponding to the tool; and
13985 the applicable switches passed on the command line.
13989 The term @emph{applicable ^switches^switches^} reflects the fact that
13990 @command{gnatmake} ^switches^switches^ may or may not be passed to individual
13991 tools, depending on the individual ^switch^switch^.
13993 @command{gnatmake} may invoke the compiler on source files from different
13994 projects. The Project Manager will use the appropriate project file to
13995 determine the @code{Compiler} package for each source file being compiled.
13996 Likewise for the @code{Binder} and @code{Linker} packages.
13998 As an example, consider the following package in a project file:
14000 @smallexample @c projectfile
14003 package Compiler is
14004 for ^Default_Switches^Default_Switches^ ("Ada")
14006 for ^Switches^Switches^ ("a.adb")
14008 for ^Switches^Switches^ ("b.adb")
14010 "^-gnaty^-gnaty^");
14017 If @command{gnatmake} is invoked with this project file, and it needs to
14018 compile, say, the files @file{a.adb}, @file{b.adb}, and @file{c.adb}, then
14019 @file{a.adb} will be compiled with the ^switch^switch^
14020 @option{^-O1^-O1^},
14021 @file{b.adb} with ^switches^switches^
14023 and @option{^-gnaty^-gnaty^},
14024 and @file{c.adb} with @option{^-g^-g^}.
14026 The following example illustrates the ordering of the ^switches^switches^
14027 contributed by different packages:
14029 @smallexample @c projectfile
14033 for ^Switches^Switches^ ("main.adb")
14041 package Compiler is
14042 for ^Switches^Switches^ ("main.adb")
14050 If you issue the command:
14053 gnatmake ^-Pproj2^/PROJECT_FILE=PROJ2^ -O0 main
14057 then the compiler will be invoked on @file{main.adb} with the following
14058 sequence of ^switches^switches^
14061 ^-g -O1 -O2 -O0^-g -O1 -O2 -O0^
14064 with the last @option{^-O^-O^}
14065 ^switch^switch^ having precedence over the earlier ones;
14066 several other ^switches^switches^
14067 (such as @option{^-c^-c^}) are added implicitly.
14069 The ^switches^switches^
14071 and @option{^-O1^-O1^} are contributed by package
14072 @code{Builder}, @option{^-O2^-O2^} is contributed
14073 by the package @code{Compiler}
14074 and @option{^-O0^-O0^} comes from the command line.
14076 The @option{^-g^-g^}
14077 ^switch^switch^ will also be passed in the invocation of
14078 @command{Gnatlink.}
14080 A final example illustrates switch contributions from packages in different
14083 @smallexample @c projectfile
14086 for Source_Files use ("pack.ads", "pack.adb");
14087 package Compiler is
14088 for ^Default_Switches^Default_Switches^ ("Ada")
14089 use ("^-gnata^-gnata^");
14097 for Source_Files use ("foo_main.adb", "bar_main.adb");
14099 for ^Switches^Switches^ ("foo_main.adb")
14107 -- Ada source file:
14109 procedure Foo_Main is
14117 gnatmake ^-PProj4^/PROJECT_FILE=PROJ4^ foo_main.adb -cargs -gnato
14121 then the ^switches^switches^ passed to the compiler for @file{foo_main.adb} are
14122 @option{^-g^-g^} (contributed by the package @code{Proj4.Builder}) and
14123 @option{^-gnato^-gnato^} (passed on the command line).
14124 When the imported package @code{Pack} is compiled, the ^switches^switches^ used
14125 are @option{^-g^-g^} from @code{Proj4.Builder},
14126 @option{^-gnata^-gnata^} (contributed from package @code{Proj3.Compiler},
14127 and @option{^-gnato^-gnato^} from the command line.
14130 When using @command{gnatmake} with project files, some ^switches^switches^ or
14131 arguments may be expressed as relative paths. As the working directory where
14132 compilation occurs may change, these relative paths are converted to absolute
14133 paths. For the ^switches^switches^ found in a project file, the relative paths
14134 are relative to the project file directory, for the switches on the command
14135 line, they are relative to the directory where @command{gnatmake} is invoked.
14136 The ^switches^switches^ for which this occurs are:
14142 ^-aI^-aI^, as well as all arguments that are not switches (arguments to
14144 ^-o^-o^, object files specified in package @code{Linker} or after
14145 -largs on the command line). The exception to this rule is the ^switch^switch^
14146 ^--RTS=^--RTS=^ for which a relative path argument is never converted.
14148 @node Specifying Configuration Pragmas
14149 @subsubsection Specifying Configuration Pragmas
14151 When using @command{gnatmake} with project files, if there exists a file
14152 @file{gnat.adc} that contains configuration pragmas, this file will be
14155 Configuration pragmas can be defined by means of the following attributes in
14156 project files: @code{Global_Configuration_Pragmas} in package @code{Builder}
14157 and @code{Local_Configuration_Pragmas} in package @code{Compiler}.
14159 Both these attributes are single string attributes. Their values is the path
14160 name of a file containing configuration pragmas. If a path name is relative,
14161 then it is relative to the project directory of the project file where the
14162 attribute is defined.
14164 When compiling a source, the configuration pragmas used are, in order,
14165 those listed in the file designated by attribute
14166 @code{Global_Configuration_Pragmas} in package @code{Builder} of the main
14167 project file, if it is specified, and those listed in the file designated by
14168 attribute @code{Local_Configuration_Pragmas} in package @code{Compiler} of
14169 the project file of the source, if it exists.
14171 @node Project Files and Main Subprograms
14172 @subsubsection Project Files and Main Subprograms
14175 When using a project file, you can invoke @command{gnatmake}
14176 with one or several main subprograms, by specifying their source files on the
14180 gnatmake ^-P^/PROJECT_FILE=^prj main1 main2 main3
14184 Each of these needs to be a source file of the same project, except
14185 when the switch ^-u^/UNIQUE^ is used.
14188 When ^-u^/UNIQUE^ is not used, all the mains need to be sources of the
14189 same project, one of the project in the tree rooted at the project specified
14190 on the command line. The package @code{Builder} of this common project, the
14191 "main project" is the one that is considered by @command{gnatmake}.
14194 When ^-u^/UNIQUE^ is used, the specified source files may be in projects
14195 imported directly or indirectly by the project specified on the command line.
14196 Note that if such a source file is not part of the project specified on the
14197 command line, the ^switches^switches^ found in package @code{Builder} of the
14198 project specified on the command line, if any, that are transmitted
14199 to the compiler will still be used, not those found in the project file of
14203 When using a project file, you can also invoke @command{gnatmake} without
14204 explicitly specifying any main, and the effect depends on whether you have
14205 defined the @code{Main} attribute. This attribute has a string list value,
14206 where each element in the list is the name of a source file (the file
14207 extension is optional) that contains a unit that can be a main subprogram.
14209 If the @code{Main} attribute is defined in a project file as a non-empty
14210 string list and the switch @option{^-u^/UNIQUE^} is not used on the command
14211 line, then invoking @command{gnatmake} with this project file but without any
14212 main on the command line is equivalent to invoking @command{gnatmake} with all
14213 the file names in the @code{Main} attribute on the command line.
14216 @smallexample @c projectfile
14219 for Main use ("main1", "main2", "main3");
14225 With this project file, @code{"gnatmake ^-Pprj^/PROJECT_FILE=PRJ^"}
14227 @code{"gnatmake ^-Pprj^/PROJECT_FILE=PRJ^ main1 main2 main3"}.
14229 When the project attribute @code{Main} is not specified, or is specified
14230 as an empty string list, or when the switch @option{-u} is used on the command
14231 line, then invoking @command{gnatmake} with no main on the command line will
14232 result in all immediate sources of the project file being checked, and
14233 potentially recompiled. Depending on the presence of the switch @option{-u},
14234 sources from other project files on which the immediate sources of the main
14235 project file depend are also checked and potentially recompiled. In other
14236 words, the @option{-u} switch is applied to all of the immediate sources of the
14239 When no main is specified on the command line and attribute @code{Main} exists
14240 and includes several mains, or when several mains are specified on the
14241 command line, the default ^switches^switches^ in package @code{Builder} will
14242 be used for all mains, even if there are specific ^switches^switches^
14243 specified for one or several mains.
14245 But the ^switches^switches^ from package @code{Binder} or @code{Linker} will be
14246 the specific ^switches^switches^ for each main, if they are specified.
14248 @node Library Project Files
14249 @subsubsection Library Project Files
14252 When @command{gnatmake} is invoked with a main project file that is a library
14253 project file, it is not allowed to specify one or more mains on the command
14257 When a library project file is specified, switches ^-b^/ACTION=BIND^ and
14258 ^-l^/ACTION=LINK^ have special meanings.
14261 @item ^-b^/ACTION=BIND^ is only allowed for stand-alone libraries. It indicates
14262 to @command{gnatmake} that @command{gnatbind} should be invoked for the
14265 @item ^-l^/ACTION=LINK^ may be used for all library projects. It indicates
14266 to @command{gnatmake} that the binder generated file should be compiled
14267 (in the case of a stand-alone library) and that the library should be built.
14271 @node The GNAT Driver and Project Files
14272 @subsection The GNAT Driver and Project Files
14275 A number of GNAT tools, other than @command{^gnatmake^gnatmake^}
14276 can benefit from project files:
14277 @command{^gnatbind^gnatbind^},
14278 @command{^gnatcheck^gnatcheck^}),
14279 @command{^gnatclean^gnatclean^}),
14280 @command{^gnatelim^gnatelim^},
14281 @command{^gnatfind^gnatfind^},
14282 @command{^gnatlink^gnatlink^},
14283 @command{^gnatls^gnatls^},
14284 @command{^gnatmetric^gnatmetric^},
14285 @command{^gnatpp^gnatpp^},
14286 @command{^gnatstub^gnatstub^},
14287 and @command{^gnatxref^gnatxref^}. However, none of these tools can be invoked
14288 directly with a project file switch (@option{^-P^/PROJECT_FILE=^}).
14289 They must be invoked through the @command{gnat} driver.
14291 The @command{gnat} driver is a wrapper that accepts a number of commands and
14292 calls the corresponding tool. It was designed initially for VMS platforms (to
14293 convert VMS qualifiers to Unix-style switches), but it is now available on all
14296 On non-VMS platforms, the @command{gnat} driver accepts the following commands
14297 (case insensitive):
14301 BIND to invoke @command{^gnatbind^gnatbind^}
14303 CHOP to invoke @command{^gnatchop^gnatchop^}
14305 CLEAN to invoke @command{^gnatclean^gnatclean^}
14307 COMP or COMPILE to invoke the compiler
14309 ELIM to invoke @command{^gnatelim^gnatelim^}
14311 FIND to invoke @command{^gnatfind^gnatfind^}
14313 KR or KRUNCH to invoke @command{^gnatkr^gnatkr^}
14315 LINK to invoke @command{^gnatlink^gnatlink^}
14317 LS or LIST to invoke @command{^gnatls^gnatls^}
14319 MAKE to invoke @command{^gnatmake^gnatmake^}
14321 NAME to invoke @command{^gnatname^gnatname^}
14323 PREP or PREPROCESS to invoke @command{^gnatprep^gnatprep^}
14325 PP or PRETTY to invoke @command{^gnatpp^gnatpp^}
14327 METRIC to invoke @command{^gnatmetric^gnatmetric^}
14329 STUB to invoke @command{^gnatstub^gnatstub^}
14331 XREF to invoke @command{^gnatxref^gnatxref^}
14335 (note that the compiler is invoked using the command
14336 @command{^gnatmake -f -u -c^gnatmake -f -u -c^}).
14339 On non-VMS platforms, between @command{gnat} and the command, two
14340 special switches may be used:
14344 @command{-v} to display the invocation of the tool.
14346 @command{-dn} to prevent the @command{gnat} driver from removing
14347 the temporary files it has created. These temporary files are
14348 configuration files and temporary file list files.
14352 The command may be followed by switches and arguments for the invoked
14356 gnat bind -C main.ali
14362 Switches may also be put in text files, one switch per line, and the text
14363 files may be specified with their path name preceded by '@@'.
14366 gnat bind @@args.txt main.ali
14370 In addition, for commands BIND, COMP or COMPILE, FIND, ELIM, LS or LIST, LINK,
14371 METRIC, PP or PRETTY, STUB and XREF, the project file related switches
14372 (@option{^-P^/PROJECT_FILE^},
14373 @option{^-X^/EXTERNAL_REFERENCE^} and
14374 @option{^-vP^/MESSAGES_PROJECT_FILE=^x}) may be used in addition to
14375 the switches of the invoking tool.
14378 When GNAT PP or GNAT PRETTY is used with a project file, but with no source
14379 specified on the command line, it invokes @command{^gnatpp^gnatpp^} with all
14380 the immediate sources of the specified project file.
14383 When GNAT METRIC is used with a project file, but with no source
14384 specified on the command line, it invokes @command{^gnatmetric^gnatmetric^}
14385 with all the immediate sources of the specified project file and with
14386 @option{^-d^/DIRECTORY^} with the parameter pointing to the object directory
14390 In addition, when GNAT PP, GNAT PRETTY or GNAT METRIC is used with
14391 a project file, no source is specified on the command line and
14392 switch ^-U^/ALL_PROJECTS^ is specified on the command line, then
14393 the underlying tool (^gnatpp^gnatpp^ or
14394 ^gnatmetric^gnatmetric^) is invoked for all sources of all projects,
14395 not only for the immediate sources of the main project.
14397 (-U stands for Universal or Union of the project files of the project tree)
14401 For each of the following commands, there is optionally a corresponding
14402 package in the main project.
14406 package @code{Binder} for command BIND (invoking @code{^gnatbind^gnatbind^})
14409 package @code{Check} for command CHECK (invoking
14410 @code{^gnatcheck^gnatcheck^})
14413 package @code{Compiler} for command COMP or COMPILE (invoking the compiler)
14416 package @code{Cross_Reference} for command XREF (invoking
14417 @code{^gnatxref^gnatxref^})
14420 package @code{Eliminate} for command ELIM (invoking
14421 @code{^gnatelim^gnatelim^})
14424 package @code{Finder} for command FIND (invoking @code{^gnatfind^gnatfind^})
14427 package @code{Gnatls} for command LS or LIST (invoking @code{^gnatls^gnatls^})
14430 package @code{Gnatstub} for command STUB
14431 (invoking @code{^gnatstub^gnatstub^})
14434 package @code{Linker} for command LINK (invoking @code{^gnatlink^gnatlink^})
14437 package @code{Metrics} for command METRIC
14438 (invoking @code{^gnatmetric^gnatmetric^})
14441 package @code{Pretty_Printer} for command PP or PRETTY
14442 (invoking @code{^gnatpp^gnatpp^})
14447 Package @code{Gnatls} has a unique attribute @code{^Switches^Switches^},
14448 a simple variable with a string list value. It contains ^switches^switches^
14449 for the invocation of @code{^gnatls^gnatls^}.
14451 @smallexample @c projectfile
14455 for ^Switches^Switches^
14464 All other packages have two attribute @code{^Switches^Switches^} and
14465 @code{^Default_Switches^Default_Switches^}.
14468 @code{^Switches^Switches^} is an associative array attribute, indexed by the
14469 source file name, that has a string list value: the ^switches^switches^ to be
14470 used when the tool corresponding to the package is invoked for the specific
14474 @code{^Default_Switches^Default_Switches^} is an associative array attribute,
14475 indexed by the programming language that has a string list value.
14476 @code{^Default_Switches^Default_Switches^ ("Ada")} contains the
14477 ^switches^switches^ for the invocation of the tool corresponding
14478 to the package, except if a specific @code{^Switches^Switches^} attribute
14479 is specified for the source file.
14481 @smallexample @c projectfile
14485 for Source_Dirs use ("./**");
14488 for ^Switches^Switches^ use
14495 package Compiler is
14496 for ^Default_Switches^Default_Switches^ ("Ada")
14497 use ("^-gnatv^-gnatv^",
14498 "^-gnatwa^-gnatwa^");
14504 for ^Default_Switches^Default_Switches^ ("Ada")
14512 for ^Default_Switches^Default_Switches^ ("Ada")
14514 for ^Switches^Switches^ ("main.adb")
14523 for ^Default_Switches^Default_Switches^ ("Ada")
14530 package Cross_Reference is
14531 for ^Default_Switches^Default_Switches^ ("Ada")
14536 end Cross_Reference;
14542 With the above project file, commands such as
14545 ^gnat comp -Pproj main^GNAT COMP /PROJECT_FILE=PROJ MAIN^
14546 ^gnat ls -Pproj main^GNAT LIST /PROJECT_FILE=PROJ MAIN^
14547 ^gnat xref -Pproj main^GNAT XREF /PROJECT_FILE=PROJ MAIN^
14548 ^gnat bind -Pproj main.ali^GNAT BIND /PROJECT_FILE=PROJ MAIN.ALI^
14549 ^gnat link -Pproj main.ali^GNAT LINK /PROJECT_FILE=PROJ MAIN.ALI^
14553 will set up the environment properly and invoke the tool with the switches
14554 found in the package corresponding to the tool:
14555 @code{^Default_Switches^Default_Switches^ ("Ada")} for all tools,
14556 except @code{^Switches^Switches^ ("main.adb")}
14557 for @code{^gnatlink^gnatlink^}.
14558 It is also possible to invoke some of the tools,
14559 @code{^gnatcheck^gnatcheck^}),
14560 @code{^gnatmetric^gnatmetric^}),
14561 and @code{^gnatpp^gnatpp^})
14562 on a set of project units thanks to the combination of the switches
14563 @option{-P}, @option{-U} and possibly the main unit when one is interested
14564 in its closure. For instance,
14568 will compute the metrics for all the immediate units of project
14571 gnat metric -Pproj -U
14573 will compute the metrics for all the units of the closure of projects
14574 rooted at @code{proj}.
14576 gnat metric -Pproj -U main_unit
14578 will compute the metrics for the closure of units rooted at
14579 @code{main_unit}. This last possibility relies implicitly
14580 on @command{gnatbind}'s option @option{-R}.
14582 @c **********************
14583 @node An Extended Example
14584 @section An Extended Example
14587 Suppose that we have two programs, @var{prog1} and @var{prog2},
14588 whose sources are in corresponding directories. We would like
14589 to build them with a single @command{gnatmake} command, and we want to place
14590 their object files into @file{build} subdirectories of the source directories.
14591 Furthermore, we want to have to have two separate subdirectories
14592 in @file{build} -- @file{release} and @file{debug} -- which will contain
14593 the object files compiled with different set of compilation flags.
14595 In other words, we have the following structure:
14612 Here are the project files that we must place in a directory @file{main}
14613 to maintain this structure:
14617 @item We create a @code{Common} project with a package @code{Compiler} that
14618 specifies the compilation ^switches^switches^:
14623 @b{project} Common @b{is}
14625 @b{for} Source_Dirs @b{use} (); -- No source files
14629 @b{type} Build_Type @b{is} ("release", "debug");
14630 Build : Build_Type := External ("BUILD", "debug");
14633 @b{package} Compiler @b{is}
14634 @b{case} Build @b{is}
14635 @b{when} "release" =>
14636 @b{for} ^Default_Switches^Default_Switches^ ("Ada")
14637 @b{use} ("^-O2^-O2^");
14638 @b{when} "debug" =>
14639 @b{for} ^Default_Switches^Default_Switches^ ("Ada")
14640 @b{use} ("^-g^-g^");
14648 @item We create separate projects for the two programs:
14655 @b{project} Prog1 @b{is}
14657 @b{for} Source_Dirs @b{use} ("prog1");
14658 @b{for} Object_Dir @b{use} "prog1/build/" & Common.Build;
14660 @b{package} Compiler @b{renames} Common.Compiler;
14671 @b{project} Prog2 @b{is}
14673 @b{for} Source_Dirs @b{use} ("prog2");
14674 @b{for} Object_Dir @b{use} "prog2/build/" & Common.Build;
14676 @b{package} Compiler @b{renames} Common.Compiler;
14682 @item We create a wrapping project @code{Main}:
14691 @b{project} Main @b{is}
14693 @b{package} Compiler @b{renames} Common.Compiler;
14699 @item Finally we need to create a dummy procedure that @code{with}s (either
14700 explicitly or implicitly) all the sources of our two programs.
14705 Now we can build the programs using the command
14708 gnatmake ^-P^/PROJECT_FILE=^main dummy
14712 for the Debug mode, or
14716 gnatmake -Pmain -XBUILD=release
14722 GNAT MAKE /PROJECT_FILE=main /EXTERNAL_REFERENCE=BUILD=release
14727 for the Release mode.
14729 @c ********************************
14730 @c * Project File Complete Syntax *
14731 @c ********************************
14733 @node Project File Complete Syntax
14734 @section Project File Complete Syntax
14738 context_clause project_declaration
14744 @b{with} path_name @{ , path_name @} ;
14749 project_declaration ::=
14750 simple_project_declaration | project_extension
14752 simple_project_declaration ::=
14753 @b{project} <project_>simple_name @b{is}
14754 @{declarative_item@}
14755 @b{end} <project_>simple_name;
14757 project_extension ::=
14758 @b{project} <project_>simple_name @b{extends} path_name @b{is}
14759 @{declarative_item@}
14760 @b{end} <project_>simple_name;
14762 declarative_item ::=
14763 package_declaration |
14764 typed_string_declaration |
14765 other_declarative_item
14767 package_declaration ::=
14768 package_spec | package_renaming
14771 @b{package} package_identifier @b{is}
14772 @{simple_declarative_item@}
14773 @b{end} package_identifier ;
14775 package_identifier ::=
14776 @code{Naming} | @code{Builder} | @code{Compiler} | @code{Binder} |
14777 @code{Linker} | @code{Finder} | @code{Cross_Reference} |
14778 @code{^gnatls^gnatls^} | @code{IDE} | @code{Pretty_Printer}
14780 package_renaming ::==
14781 @b{package} package_identifier @b{renames}
14782 <project_>simple_name.package_identifier ;
14784 typed_string_declaration ::=
14785 @b{type} <typed_string_>_simple_name @b{is}
14786 ( string_literal @{, string_literal@} );
14788 other_declarative_item ::=
14789 attribute_declaration |
14790 typed_variable_declaration |
14791 variable_declaration |
14794 attribute_declaration ::=
14795 full_associative_array_declaration |
14796 @b{for} attribute_designator @b{use} expression ;
14798 full_associative_array_declaration ::=
14799 @b{for} <associative_array_attribute_>simple_name @b{use}
14800 <project_>simple_name [ . <package_>simple_Name ] ' <attribute_>simple_name ;
14802 attribute_designator ::=
14803 <simple_attribute_>simple_name |
14804 <associative_array_attribute_>simple_name ( string_literal )
14806 typed_variable_declaration ::=
14807 <typed_variable_>simple_name : <typed_string_>name := string_expression ;
14809 variable_declaration ::=
14810 <variable_>simple_name := expression;
14820 attribute_reference
14826 ( <string_>expression @{ , <string_>expression @} )
14829 @b{external} ( string_literal [, string_literal] )
14831 attribute_reference ::=
14832 attribute_prefix ' <simple_attribute_>simple_name [ ( literal_string ) ]
14834 attribute_prefix ::=
14836 <project_>simple_name | package_identifier |
14837 <project_>simple_name . package_identifier
14839 case_construction ::=
14840 @b{case} <typed_variable_>name @b{is}
14845 @b{when} discrete_choice_list =>
14846 @{case_construction | attribute_declaration@}
14848 discrete_choice_list ::=
14849 string_literal @{| string_literal@} |
14853 simple_name @{. simple_name@}
14856 identifier (same as Ada)
14860 @node The Cross-Referencing Tools gnatxref and gnatfind
14861 @chapter The Cross-Referencing Tools @code{gnatxref} and @code{gnatfind}
14866 The compiler generates cross-referencing information (unless
14867 you set the @samp{-gnatx} switch), which are saved in the @file{.ali} files.
14868 This information indicates where in the source each entity is declared and
14869 referenced. Note that entities in package Standard are not included, but
14870 entities in all other predefined units are included in the output.
14872 Before using any of these two tools, you need to compile successfully your
14873 application, so that GNAT gets a chance to generate the cross-referencing
14876 The two tools @code{gnatxref} and @code{gnatfind} take advantage of this
14877 information to provide the user with the capability to easily locate the
14878 declaration and references to an entity. These tools are quite similar,
14879 the difference being that @code{gnatfind} is intended for locating
14880 definitions and/or references to a specified entity or entities, whereas
14881 @code{gnatxref} is oriented to generating a full report of all
14884 To use these tools, you must not compile your application using the
14885 @option{-gnatx} switch on the @command{gnatmake} command line
14886 (@pxref{The GNAT Make Program gnatmake}). Otherwise, cross-referencing
14887 information will not be generated.
14889 Note: to invoke @code{gnatxref} or @code{gnatfind} with a project file,
14890 use the @code{gnat} driver (see @ref{The GNAT Driver and Project Files}).
14893 * gnatxref Switches::
14894 * gnatfind Switches::
14895 * Project Files for gnatxref and gnatfind::
14896 * Regular Expressions in gnatfind and gnatxref::
14897 * Examples of gnatxref Usage::
14898 * Examples of gnatfind Usage::
14901 @node gnatxref Switches
14902 @section @code{gnatxref} Switches
14905 The command invocation for @code{gnatxref} is:
14907 $ gnatxref @ovar{switches} @var{sourcefile1} @r{[}@var{sourcefile2} @dots{}@r{]}
14916 identifies the source files for which a report is to be generated. The
14917 ``with''ed units will be processed too. You must provide at least one file.
14919 These file names are considered to be regular expressions, so for instance
14920 specifying @file{source*.adb} is the same as giving every file in the current
14921 directory whose name starts with @file{source} and whose extension is
14924 You shouldn't specify any directory name, just base names. @command{gnatxref}
14925 and @command{gnatfind} will be able to locate these files by themselves using
14926 the source path. If you specify directories, no result is produced.
14931 The switches can be:
14935 @cindex @option{--version} @command{gnatxref}
14936 Display Copyright and version, then exit disregarding all other options.
14939 @cindex @option{--help} @command{gnatxref}
14940 If @option{--version} was not used, display usage, then exit disregarding
14943 @item ^-a^/ALL_FILES^
14944 @cindex @option{^-a^/ALL_FILES^} (@command{gnatxref})
14945 If this switch is present, @code{gnatfind} and @code{gnatxref} will parse
14946 the read-only files found in the library search path. Otherwise, these files
14947 will be ignored. This option can be used to protect Gnat sources or your own
14948 libraries from being parsed, thus making @code{gnatfind} and @code{gnatxref}
14949 much faster, and their output much smaller. Read-only here refers to access
14950 or permissions status in the file system for the current user.
14953 @cindex @option{-aIDIR} (@command{gnatxref})
14954 When looking for source files also look in directory DIR. The order in which
14955 source file search is undertaken is the same as for @command{gnatmake}.
14958 @cindex @option{-aODIR} (@command{gnatxref})
14959 When searching for library and object files, look in directory
14960 DIR. The order in which library files are searched is the same as for
14961 @command{gnatmake}.
14964 @cindex @option{-nostdinc} (@command{gnatxref})
14965 Do not look for sources in the system default directory.
14968 @cindex @option{-nostdlib} (@command{gnatxref})
14969 Do not look for library files in the system default directory.
14971 @item --RTS=@var{rts-path}
14972 @cindex @option{--RTS} (@command{gnatxref})
14973 Specifies the default location of the runtime library. Same meaning as the
14974 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
14976 @item ^-d^/DERIVED_TYPES^
14977 @cindex @option{^-d^/DERIVED_TYPES^} (@command{gnatxref})
14978 If this switch is set @code{gnatxref} will output the parent type
14979 reference for each matching derived types.
14981 @item ^-f^/FULL_PATHNAME^
14982 @cindex @option{^-f^/FULL_PATHNAME^} (@command{gnatxref})
14983 If this switch is set, the output file names will be preceded by their
14984 directory (if the file was found in the search path). If this switch is
14985 not set, the directory will not be printed.
14987 @item ^-g^/IGNORE_LOCALS^
14988 @cindex @option{^-g^/IGNORE_LOCALS^} (@command{gnatxref})
14989 If this switch is set, information is output only for library-level
14990 entities, ignoring local entities. The use of this switch may accelerate
14991 @code{gnatfind} and @code{gnatxref}.
14994 @cindex @option{-IDIR} (@command{gnatxref})
14995 Equivalent to @samp{-aODIR -aIDIR}.
14998 @cindex @option{-pFILE} (@command{gnatxref})
14999 Specify a project file to use @xref{Project Files}.
15000 If you need to use the @file{.gpr}
15001 project files, you should use gnatxref through the GNAT driver
15002 (@command{gnat xref -Pproject}).
15004 By default, @code{gnatxref} and @code{gnatfind} will try to locate a
15005 project file in the current directory.
15007 If a project file is either specified or found by the tools, then the content
15008 of the source directory and object directory lines are added as if they
15009 had been specified respectively by @samp{^-aI^/SOURCE_SEARCH^}
15010 and @samp{^-aO^OBJECT_SEARCH^}.
15012 Output only unused symbols. This may be really useful if you give your
15013 main compilation unit on the command line, as @code{gnatxref} will then
15014 display every unused entity and 'with'ed package.
15018 Instead of producing the default output, @code{gnatxref} will generate a
15019 @file{tags} file that can be used by vi. For examples how to use this
15020 feature, see @ref{Examples of gnatxref Usage}. The tags file is output
15021 to the standard output, thus you will have to redirect it to a file.
15027 All these switches may be in any order on the command line, and may even
15028 appear after the file names. They need not be separated by spaces, thus
15029 you can say @samp{gnatxref ^-ag^/ALL_FILES/IGNORE_LOCALS^} instead of
15030 @samp{gnatxref ^-a -g^/ALL_FILES /IGNORE_LOCALS^}.
15032 @node gnatfind Switches
15033 @section @code{gnatfind} Switches
15036 The command line for @code{gnatfind} is:
15039 $ gnatfind @ovar{switches} @var{pattern}@r{[}:@var{sourcefile}@r{[}:@var{line}@r{[}:@var{column}@r{]]]}
15040 @r{[}@var{file1} @var{file2} @dots{}]
15048 An entity will be output only if it matches the regular expression found
15049 in @var{pattern}, see @ref{Regular Expressions in gnatfind and gnatxref}.
15051 Omitting the pattern is equivalent to specifying @samp{*}, which
15052 will match any entity. Note that if you do not provide a pattern, you
15053 have to provide both a sourcefile and a line.
15055 Entity names are given in Latin-1, with uppercase/lowercase equivalence
15056 for matching purposes. At the current time there is no support for
15057 8-bit codes other than Latin-1, or for wide characters in identifiers.
15060 @code{gnatfind} will look for references, bodies or declarations
15061 of symbols referenced in @file{@var{sourcefile}}, at line @var{line}
15062 and column @var{column}. See @ref{Examples of gnatfind Usage}
15063 for syntax examples.
15066 is a decimal integer identifying the line number containing
15067 the reference to the entity (or entities) to be located.
15070 is a decimal integer identifying the exact location on the
15071 line of the first character of the identifier for the
15072 entity reference. Columns are numbered from 1.
15074 @item file1 file2 @dots{}
15075 The search will be restricted to these source files. If none are given, then
15076 the search will be done for every library file in the search path.
15077 These file must appear only after the pattern or sourcefile.
15079 These file names are considered to be regular expressions, so for instance
15080 specifying @file{source*.adb} is the same as giving every file in the current
15081 directory whose name starts with @file{source} and whose extension is
15084 The location of the spec of the entity will always be displayed, even if it
15085 isn't in one of @file{@var{file1}}, @file{@var{file2}},@enddots{} The
15086 occurrences of the entity in the separate units of the ones given on the
15087 command line will also be displayed.
15089 Note that if you specify at least one file in this part, @code{gnatfind} may
15090 sometimes not be able to find the body of the subprograms.
15095 At least one of 'sourcefile' or 'pattern' has to be present on
15098 The following switches are available:
15102 @cindex @option{--version} @command{gnatfind}
15103 Display Copyright and version, then exit disregarding all other options.
15106 @cindex @option{--help} @command{gnatfind}
15107 If @option{--version} was not used, display usage, then exit disregarding
15110 @item ^-a^/ALL_FILES^
15111 @cindex @option{^-a^/ALL_FILES^} (@command{gnatfind})
15112 If this switch is present, @code{gnatfind} and @code{gnatxref} will parse
15113 the read-only files found in the library search path. Otherwise, these files
15114 will be ignored. This option can be used to protect Gnat sources or your own
15115 libraries from being parsed, thus making @code{gnatfind} and @code{gnatxref}
15116 much faster, and their output much smaller. Read-only here refers to access
15117 or permission status in the file system for the current user.
15120 @cindex @option{-aIDIR} (@command{gnatfind})
15121 When looking for source files also look in directory DIR. The order in which
15122 source file search is undertaken is the same as for @command{gnatmake}.
15125 @cindex @option{-aODIR} (@command{gnatfind})
15126 When searching for library and object files, look in directory
15127 DIR. The order in which library files are searched is the same as for
15128 @command{gnatmake}.
15131 @cindex @option{-nostdinc} (@command{gnatfind})
15132 Do not look for sources in the system default directory.
15135 @cindex @option{-nostdlib} (@command{gnatfind})
15136 Do not look for library files in the system default directory.
15138 @item --RTS=@var{rts-path}
15139 @cindex @option{--RTS} (@command{gnatfind})
15140 Specifies the default location of the runtime library. Same meaning as the
15141 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
15143 @item ^-d^/DERIVED_TYPE_INFORMATION^
15144 @cindex @option{^-d^/DERIVED_TYPE_INFORMATION^} (@code{gnatfind})
15145 If this switch is set, then @code{gnatfind} will output the parent type
15146 reference for each matching derived types.
15148 @item ^-e^/EXPRESSIONS^
15149 @cindex @option{^-e^/EXPRESSIONS^} (@command{gnatfind})
15150 By default, @code{gnatfind} accept the simple regular expression set for
15151 @samp{pattern}. If this switch is set, then the pattern will be
15152 considered as full Unix-style regular expression.
15154 @item ^-f^/FULL_PATHNAME^
15155 @cindex @option{^-f^/FULL_PATHNAME^} (@command{gnatfind})
15156 If this switch is set, the output file names will be preceded by their
15157 directory (if the file was found in the search path). If this switch is
15158 not set, the directory will not be printed.
15160 @item ^-g^/IGNORE_LOCALS^
15161 @cindex @option{^-g^/IGNORE_LOCALS^} (@command{gnatfind})
15162 If this switch is set, information is output only for library-level
15163 entities, ignoring local entities. The use of this switch may accelerate
15164 @code{gnatfind} and @code{gnatxref}.
15167 @cindex @option{-IDIR} (@command{gnatfind})
15168 Equivalent to @samp{-aODIR -aIDIR}.
15171 @cindex @option{-pFILE} (@command{gnatfind})
15172 Specify a project file (@pxref{Project Files}) to use.
15173 By default, @code{gnatxref} and @code{gnatfind} will try to locate a
15174 project file in the current directory.
15176 If a project file is either specified or found by the tools, then the content
15177 of the source directory and object directory lines are added as if they
15178 had been specified respectively by @samp{^-aI^/SOURCE_SEARCH^} and
15179 @samp{^-aO^/OBJECT_SEARCH^}.
15181 @item ^-r^/REFERENCES^
15182 @cindex @option{^-r^/REFERENCES^} (@command{gnatfind})
15183 By default, @code{gnatfind} will output only the information about the
15184 declaration, body or type completion of the entities. If this switch is
15185 set, the @code{gnatfind} will locate every reference to the entities in
15186 the files specified on the command line (or in every file in the search
15187 path if no file is given on the command line).
15189 @item ^-s^/PRINT_LINES^
15190 @cindex @option{^-s^/PRINT_LINES^} (@command{gnatfind})
15191 If this switch is set, then @code{gnatfind} will output the content
15192 of the Ada source file lines were the entity was found.
15194 @item ^-t^/TYPE_HIERARCHY^
15195 @cindex @option{^-t^/TYPE_HIERARCHY^} (@command{gnatfind})
15196 If this switch is set, then @code{gnatfind} will output the type hierarchy for
15197 the specified type. It act like -d option but recursively from parent
15198 type to parent type. When this switch is set it is not possible to
15199 specify more than one file.
15204 All these switches may be in any order on the command line, and may even
15205 appear after the file names. They need not be separated by spaces, thus
15206 you can say @samp{gnatxref ^-ag^/ALL_FILES/IGNORE_LOCALS^} instead of
15207 @samp{gnatxref ^-a -g^/ALL_FILES /IGNORE_LOCALS^}.
15209 As stated previously, gnatfind will search in every directory in the
15210 search path. You can force it to look only in the current directory if
15211 you specify @code{*} at the end of the command line.
15213 @node Project Files for gnatxref and gnatfind
15214 @section Project Files for @command{gnatxref} and @command{gnatfind}
15217 Project files allow a programmer to specify how to compile its
15218 application, where to find sources, etc. These files are used
15220 primarily by GPS, but they can also be used
15223 @code{gnatxref} and @code{gnatfind}.
15225 A project file name must end with @file{.gpr}. If a single one is
15226 present in the current directory, then @code{gnatxref} and @code{gnatfind} will
15227 extract the information from it. If multiple project files are found, none of
15228 them is read, and you have to use the @samp{-p} switch to specify the one
15231 The following lines can be included, even though most of them have default
15232 values which can be used in most cases.
15233 The lines can be entered in any order in the file.
15234 Except for @file{src_dir} and @file{obj_dir}, you can only have one instance of
15235 each line. If you have multiple instances, only the last one is taken into
15240 [default: @code{"^./^[]^"}]
15241 specifies a directory where to look for source files. Multiple @code{src_dir}
15242 lines can be specified and they will be searched in the order they
15246 [default: @code{"^./^[]^"}]
15247 specifies a directory where to look for object and library files. Multiple
15248 @code{obj_dir} lines can be specified, and they will be searched in the order
15251 @item comp_opt=SWITCHES
15252 [default: @code{""}]
15253 creates a variable which can be referred to subsequently by using
15254 the @code{$@{comp_opt@}} notation. This is intended to store the default
15255 switches given to @command{gnatmake} and @command{gcc}.
15257 @item bind_opt=SWITCHES
15258 [default: @code{""}]
15259 creates a variable which can be referred to subsequently by using
15260 the @samp{$@{bind_opt@}} notation. This is intended to store the default
15261 switches given to @command{gnatbind}.
15263 @item link_opt=SWITCHES
15264 [default: @code{""}]
15265 creates a variable which can be referred to subsequently by using
15266 the @samp{$@{link_opt@}} notation. This is intended to store the default
15267 switches given to @command{gnatlink}.
15269 @item main=EXECUTABLE
15270 [default: @code{""}]
15271 specifies the name of the executable for the application. This variable can
15272 be referred to in the following lines by using the @samp{$@{main@}} notation.
15275 @item comp_cmd=COMMAND
15276 [default: @code{"GNAT COMPILE /SEARCH=$@{src_dir@} /DEBUG /TRY_SEMANTICS"}]
15279 @item comp_cmd=COMMAND
15280 [default: @code{"gcc -c -I$@{src_dir@} -g -gnatq"}]
15282 specifies the command used to compile a single file in the application.
15285 @item make_cmd=COMMAND
15286 [default: @code{"GNAT MAKE $@{main@}
15287 /SOURCE_SEARCH=$@{src_dir@} /OBJECT_SEARCH=$@{obj_dir@}
15288 /DEBUG /TRY_SEMANTICS /COMPILER_QUALIFIERS $@{comp_opt@}
15289 /BINDER_QUALIFIERS $@{bind_opt@} /LINKER_QUALIFIERS $@{link_opt@}"}]
15292 @item make_cmd=COMMAND
15293 [default: @code{"gnatmake $@{main@} -aI$@{src_dir@}
15294 -aO$@{obj_dir@} -g -gnatq -cargs $@{comp_opt@}
15295 -bargs $@{bind_opt@} -largs $@{link_opt@}"}]
15297 specifies the command used to recompile the whole application.
15299 @item run_cmd=COMMAND
15300 [default: @code{"$@{main@}"}]
15301 specifies the command used to run the application.
15303 @item debug_cmd=COMMAND
15304 [default: @code{"gdb $@{main@}"}]
15305 specifies the command used to debug the application
15310 @command{gnatxref} and @command{gnatfind} only take into account the
15311 @code{src_dir} and @code{obj_dir} lines, and ignore the others.
15313 @node Regular Expressions in gnatfind and gnatxref
15314 @section Regular Expressions in @code{gnatfind} and @code{gnatxref}
15317 As specified in the section about @command{gnatfind}, the pattern can be a
15318 regular expression. Actually, there are to set of regular expressions
15319 which are recognized by the program:
15322 @item globbing patterns
15323 These are the most usual regular expression. They are the same that you
15324 generally used in a Unix shell command line, or in a DOS session.
15326 Here is a more formal grammar:
15333 term ::= elmt -- matches elmt
15334 term ::= elmt elmt -- concatenation (elmt then elmt)
15335 term ::= * -- any string of 0 or more characters
15336 term ::= ? -- matches any character
15337 term ::= [char @{char@}] -- matches any character listed
15338 term ::= [char - char] -- matches any character in range
15342 @item full regular expression
15343 The second set of regular expressions is much more powerful. This is the
15344 type of regular expressions recognized by utilities such a @file{grep}.
15346 The following is the form of a regular expression, expressed in Ada
15347 reference manual style BNF is as follows
15354 regexp ::= term @{| term@} -- alternation (term or term @dots{})
15356 term ::= item @{item@} -- concatenation (item then item)
15358 item ::= elmt -- match elmt
15359 item ::= elmt * -- zero or more elmt's
15360 item ::= elmt + -- one or more elmt's
15361 item ::= elmt ? -- matches elmt or nothing
15364 elmt ::= nschar -- matches given character
15365 elmt ::= [nschar @{nschar@}] -- matches any character listed
15366 elmt ::= [^^^ nschar @{nschar@}] -- matches any character not listed
15367 elmt ::= [char - char] -- matches chars in given range
15368 elmt ::= \ char -- matches given character
15369 elmt ::= . -- matches any single character
15370 elmt ::= ( regexp ) -- parens used for grouping
15372 char ::= any character, including special characters
15373 nschar ::= any character except ()[].*+?^^^
15377 Following are a few examples:
15381 will match any of the two strings @samp{abcde} and @samp{fghi},
15384 will match any string like @samp{abd}, @samp{abcd}, @samp{abccd},
15385 @samp{abcccd}, and so on,
15388 will match any string which has only lowercase characters in it (and at
15389 least one character.
15394 @node Examples of gnatxref Usage
15395 @section Examples of @code{gnatxref} Usage
15397 @subsection General Usage
15400 For the following examples, we will consider the following units:
15402 @smallexample @c ada
15408 3: procedure Foo (B : in Integer);
15415 1: package body Main is
15416 2: procedure Foo (B : in Integer) is
15427 2: procedure Print (B : Integer);
15436 The first thing to do is to recompile your application (for instance, in
15437 that case just by doing a @samp{gnatmake main}, so that GNAT generates
15438 the cross-referencing information.
15439 You can then issue any of the following commands:
15441 @item gnatxref main.adb
15442 @code{gnatxref} generates cross-reference information for main.adb
15443 and every unit 'with'ed by main.adb.
15445 The output would be:
15453 Decl: main.ads 3:20
15454 Body: main.adb 2:20
15455 Ref: main.adb 4:13 5:13 6:19
15458 Ref: main.adb 6:8 7:8
15468 Decl: main.ads 3:15
15469 Body: main.adb 2:15
15472 Body: main.adb 1:14
15475 Ref: main.adb 6:12 7:12
15479 that is the entity @code{Main} is declared in main.ads, line 2, column 9,
15480 its body is in main.adb, line 1, column 14 and is not referenced any where.
15482 The entity @code{Print} is declared in bar.ads, line 2, column 15 and it
15483 it referenced in main.adb, line 6 column 12 and line 7 column 12.
15485 @item gnatxref package1.adb package2.ads
15486 @code{gnatxref} will generates cross-reference information for
15487 package1.adb, package2.ads and any other package 'with'ed by any
15493 @subsection Using gnatxref with vi
15495 @code{gnatxref} can generate a tags file output, which can be used
15496 directly from @command{vi}. Note that the standard version of @command{vi}
15497 will not work properly with overloaded symbols. Consider using another
15498 free implementation of @command{vi}, such as @command{vim}.
15501 $ gnatxref -v gnatfind.adb > tags
15505 will generate the tags file for @code{gnatfind} itself (if the sources
15506 are in the search path!).
15508 From @command{vi}, you can then use the command @samp{:tag @var{entity}}
15509 (replacing @var{entity} by whatever you are looking for), and vi will
15510 display a new file with the corresponding declaration of entity.
15513 @node Examples of gnatfind Usage
15514 @section Examples of @code{gnatfind} Usage
15518 @item gnatfind ^-f^/FULL_PATHNAME^ xyz:main.adb
15519 Find declarations for all entities xyz referenced at least once in
15520 main.adb. The references are search in every library file in the search
15523 The directories will be printed as well (as the @samp{^-f^/FULL_PATHNAME^}
15526 The output will look like:
15528 ^directory/^[directory]^main.ads:106:14: xyz <= declaration
15529 ^directory/^[directory]^main.adb:24:10: xyz <= body
15530 ^directory/^[directory]^foo.ads:45:23: xyz <= declaration
15534 that is to say, one of the entities xyz found in main.adb is declared at
15535 line 12 of main.ads (and its body is in main.adb), and another one is
15536 declared at line 45 of foo.ads
15538 @item gnatfind ^-fs^/FULL_PATHNAME/SOURCE_LINE^ xyz:main.adb
15539 This is the same command as the previous one, instead @code{gnatfind} will
15540 display the content of the Ada source file lines.
15542 The output will look like:
15545 ^directory/^[directory]^main.ads:106:14: xyz <= declaration
15547 ^directory/^[directory]^main.adb:24:10: xyz <= body
15549 ^directory/^[directory]^foo.ads:45:23: xyz <= declaration
15554 This can make it easier to find exactly the location your are looking
15557 @item gnatfind ^-r^/REFERENCES^ "*x*":main.ads:123 foo.adb
15558 Find references to all entities containing an x that are
15559 referenced on line 123 of main.ads.
15560 The references will be searched only in main.ads and foo.adb.
15562 @item gnatfind main.ads:123
15563 Find declarations and bodies for all entities that are referenced on
15564 line 123 of main.ads.
15566 This is the same as @code{gnatfind "*":main.adb:123}.
15568 @item gnatfind ^mydir/^[mydir]^main.adb:123:45
15569 Find the declaration for the entity referenced at column 45 in
15570 line 123 of file main.adb in directory mydir. Note that it
15571 is usual to omit the identifier name when the column is given,
15572 since the column position identifies a unique reference.
15574 The column has to be the beginning of the identifier, and should not
15575 point to any character in the middle of the identifier.
15579 @c *********************************
15580 @node The GNAT Pretty-Printer gnatpp
15581 @chapter The GNAT Pretty-Printer @command{gnatpp}
15583 @cindex Pretty-Printer
15586 ^The @command{gnatpp} tool^GNAT PRETTY^ is an ASIS-based utility
15587 for source reformatting / pretty-printing.
15588 It takes an Ada source file as input and generates a reformatted
15590 You can specify various style directives via switches; e.g.,
15591 identifier case conventions, rules of indentation, and comment layout.
15593 To produce a reformatted file, @command{gnatpp} generates and uses the ASIS
15594 tree for the input source and thus requires the input to be syntactically and
15595 semantically legal.
15596 If this condition is not met, @command{gnatpp} will terminate with an
15597 error message; no output file will be generated.
15599 If the source files presented to @command{gnatpp} contain
15600 preprocessing directives, then the output file will
15601 correspond to the generated source after all
15602 preprocessing is carried out. There is no way
15603 using @command{gnatpp} to obtain pretty printed files that
15604 include the preprocessing directives.
15606 If the compilation unit
15607 contained in the input source depends semantically upon units located
15608 outside the current directory, you have to provide the source search path
15609 when invoking @command{gnatpp}, if these units are contained in files with
15610 names that do not follow the GNAT file naming rules, you have to provide
15611 the configuration file describing the corresponding naming scheme;
15612 see the description of the @command{gnatpp}
15613 switches below. Another possibility is to use a project file and to
15614 call @command{gnatpp} through the @command{gnat} driver
15616 The @command{gnatpp} command has the form
15619 $ gnatpp @ovar{switches} @var{filename}
15626 @var{switches} is an optional sequence of switches defining such properties as
15627 the formatting rules, the source search path, and the destination for the
15631 @var{filename} is the name (including the extension) of the source file to
15632 reformat; ``wildcards'' or several file names on the same gnatpp command are
15633 allowed. The file name may contain path information; it does not have to
15634 follow the GNAT file naming rules
15638 * Switches for gnatpp::
15639 * Formatting Rules::
15642 @node Switches for gnatpp
15643 @section Switches for @command{gnatpp}
15646 The following subsections describe the various switches accepted by
15647 @command{gnatpp}, organized by category.
15650 You specify a switch by supplying a name and generally also a value.
15651 In many cases the values for a switch with a given name are incompatible with
15653 (for example the switch that controls the casing of a reserved word may have
15654 exactly one value: upper case, lower case, or
15655 mixed case) and thus exactly one such switch can be in effect for an
15656 invocation of @command{gnatpp}.
15657 If more than one is supplied, the last one is used.
15658 However, some values for the same switch are mutually compatible.
15659 You may supply several such switches to @command{gnatpp}, but then
15660 each must be specified in full, with both the name and the value.
15661 Abbreviated forms (the name appearing once, followed by each value) are
15663 For example, to set
15664 the alignment of the assignment delimiter both in declarations and in
15665 assignment statements, you must write @option{-A2A3}
15666 (or @option{-A2 -A3}), but not @option{-A23}.
15670 In many cases the set of options for a given qualifier are incompatible with
15671 each other (for example the qualifier that controls the casing of a reserved
15672 word may have exactly one option, which specifies either upper case, lower
15673 case, or mixed case), and thus exactly one such option can be in effect for
15674 an invocation of @command{gnatpp}.
15675 If more than one is supplied, the last one is used.
15676 However, some qualifiers have options that are mutually compatible,
15677 and then you may then supply several such options when invoking
15681 In most cases, it is obvious whether or not the
15682 ^values for a switch with a given name^options for a given qualifier^
15683 are compatible with each other.
15684 When the semantics might not be evident, the summaries below explicitly
15685 indicate the effect.
15688 * Alignment Control::
15690 * Construct Layout Control::
15691 * General Text Layout Control::
15692 * Other Formatting Options::
15693 * Setting the Source Search Path::
15694 * Output File Control::
15695 * Other gnatpp Switches::
15698 @node Alignment Control
15699 @subsection Alignment Control
15700 @cindex Alignment control in @command{gnatpp}
15703 Programs can be easier to read if certain constructs are vertically aligned.
15704 By default all alignments are set ON.
15705 Through the @option{^-A0^/ALIGN=OFF^} switch you may reset the default to
15706 OFF, and then use one or more of the other
15707 ^@option{-A@var{n}} switches^@option{/ALIGN} options^
15708 to activate alignment for specific constructs.
15711 @cindex @option{^-A@var{n}^/ALIGN^} (@command{gnatpp})
15715 Set all alignments to ON
15718 @item ^-A0^/ALIGN=OFF^
15719 Set all alignments to OFF
15721 @item ^-A1^/ALIGN=COLONS^
15722 Align @code{:} in declarations
15724 @item ^-A2^/ALIGN=DECLARATIONS^
15725 Align @code{:=} in initializations in declarations
15727 @item ^-A3^/ALIGN=STATEMENTS^
15728 Align @code{:=} in assignment statements
15730 @item ^-A4^/ALIGN=ARROWS^
15731 Align @code{=>} in associations
15733 @item ^-A5^/ALIGN=COMPONENT_CLAUSES^
15734 Align @code{at} keywords in the component clauses in record
15735 representation clauses
15739 The @option{^-A^/ALIGN^} switches are mutually compatible; any combination
15742 @node Casing Control
15743 @subsection Casing Control
15744 @cindex Casing control in @command{gnatpp}
15747 @command{gnatpp} allows you to specify the casing for reserved words,
15748 pragma names, attribute designators and identifiers.
15749 For identifiers you may define a
15750 general rule for name casing but also override this rule
15751 via a set of dictionary files.
15753 Three types of casing are supported: lower case, upper case, and mixed case.
15754 Lower and upper case are self-explanatory (but since some letters in
15755 Latin1 and other GNAT-supported character sets
15756 exist only in lower-case form, an upper case conversion will have no
15758 ``Mixed case'' means that the first letter, and also each letter immediately
15759 following an underscore, are converted to their uppercase forms;
15760 all the other letters are converted to their lowercase forms.
15763 @cindex @option{^-a@var{x}^/ATTRIBUTE^} (@command{gnatpp})
15764 @item ^-aL^/ATTRIBUTE_CASING=LOWER_CASE^
15765 Attribute designators are lower case
15767 @item ^-aU^/ATTRIBUTE_CASING=UPPER_CASE^
15768 Attribute designators are upper case
15770 @item ^-aM^/ATTRIBUTE_CASING=MIXED_CASE^
15771 Attribute designators are mixed case (this is the default)
15773 @cindex @option{^-k@var{x}^/KEYWORD_CASING^} (@command{gnatpp})
15774 @item ^-kL^/KEYWORD_CASING=LOWER_CASE^
15775 Keywords (technically, these are known in Ada as @emph{reserved words}) are
15776 lower case (this is the default)
15778 @item ^-kU^/KEYWORD_CASING=UPPER_CASE^
15779 Keywords are upper case
15781 @cindex @option{^-n@var{x}^/NAME_CASING^} (@command{gnatpp})
15782 @item ^-nD^/NAME_CASING=AS_DECLARED^
15783 Name casing for defining occurrences are as they appear in the source file
15784 (this is the default)
15786 @item ^-nU^/NAME_CASING=UPPER_CASE^
15787 Names are in upper case
15789 @item ^-nL^/NAME_CASING=LOWER_CASE^
15790 Names are in lower case
15792 @item ^-nM^/NAME_CASING=MIXED_CASE^
15793 Names are in mixed case
15795 @cindex @option{^-p@var{x}^/PRAGMA_CASING^} (@command{gnatpp})
15796 @item ^-pL^/PRAGMA_CASING=LOWER_CASE^
15797 Pragma names are lower case
15799 @item ^-pU^/PRAGMA_CASING=UPPER_CASE^
15800 Pragma names are upper case
15802 @item ^-pM^/PRAGMA_CASING=MIXED_CASE^
15803 Pragma names are mixed case (this is the default)
15805 @item ^-D@var{file}^/DICTIONARY=@var{file}^
15806 @cindex @option{^-D^/DICTIONARY^} (@command{gnatpp})
15807 Use @var{file} as a @emph{dictionary file} that defines
15808 the casing for a set of specified names,
15809 thereby overriding the effect on these names by
15810 any explicit or implicit
15811 ^-n^/NAME_CASING^ switch.
15812 To supply more than one dictionary file,
15813 use ^several @option{-D} switches^a list of files as options^.
15816 @option{gnatpp} implicitly uses a @emph{default dictionary file}
15817 to define the casing for the Ada predefined names and
15818 the names declared in the GNAT libraries.
15820 @item ^-D-^/SPECIFIC_CASING^
15821 @cindex @option{^-D-^/SPECIFIC_CASING^} (@command{gnatpp})
15822 Do not use the default dictionary file;
15823 instead, use the casing
15824 defined by a @option{^-n^/NAME_CASING^} switch and any explicit
15829 The structure of a dictionary file, and details on the conventions
15830 used in the default dictionary file, are defined in @ref{Name Casing}.
15832 The @option{^-D-^/SPECIFIC_CASING^} and
15833 @option{^-D@var{file}^/DICTIONARY=@var{file}^} switches are mutually
15836 @node Construct Layout Control
15837 @subsection Construct Layout Control
15838 @cindex Layout control in @command{gnatpp}
15841 This group of @command{gnatpp} switches controls the layout of comments and
15842 complex syntactic constructs. See @ref{Formatting Comments} for details
15846 @cindex @option{^-c@var{n}^/COMMENTS_LAYOUT^} (@command{gnatpp})
15847 @item ^-c0^/COMMENTS_LAYOUT=UNTOUCHED^
15848 All the comments remain unchanged
15850 @item ^-c1^/COMMENTS_LAYOUT=DEFAULT^
15851 GNAT-style comment line indentation (this is the default).
15853 @item ^-c2^/COMMENTS_LAYOUT=STANDARD_INDENT^
15854 Reference-manual comment line indentation.
15856 @item ^-c3^/COMMENTS_LAYOUT=GNAT_BEGINNING^
15857 GNAT-style comment beginning
15859 @item ^-c4^/COMMENTS_LAYOUT=REFORMAT^
15860 Reformat comment blocks
15862 @item ^-c5^/COMMENTS_LAYOUT=KEEP_SPECIAL^
15863 Keep unchanged special form comments
15865 Reformat comment blocks
15867 @cindex @option{^-l@var{n}^/CONSTRUCT_LAYOUT^} (@command{gnatpp})
15868 @item ^-l1^/CONSTRUCT_LAYOUT=GNAT^
15869 GNAT-style layout (this is the default)
15871 @item ^-l2^/CONSTRUCT_LAYOUT=COMPACT^
15874 @item ^-l3^/CONSTRUCT_LAYOUT=UNCOMPACT^
15877 @cindex @option{^-N^/NOTABS^} (@command{gnatpp})
15879 All the VT characters are removed from the comment text. All the HT characters
15880 are expanded with the sequences of space characters to get to the next tab
15883 @cindex @option{^--no-separate-is^/NO_SEPARATE_IS^} (@command{gnatpp})
15884 @item ^--no-separate-is^/NO_SEPARATE_IS^
15885 Do not place the keyword @code{is} on a separate line in a subprogram body in
15886 case if the spec occupies more then one line.
15888 @cindex @option{^--separate-loop-then^/SEPARATE_LOOP_THEN^} (@command{gnatpp})
15889 @item ^--separate-loop-then^/SEPARATE_LOOP_THEN^
15890 Place the keyword @code{loop} in FOR and WHILE loop statements and the
15891 keyword @code{then} in IF statements on a separate line.
15893 @cindex @option{^--no-separate-loop-then^/NO_SEPARATE_LOOP_THEN^} (@command{gnatpp})
15894 @item ^--no-separate-loop-then^/NO_SEPARATE_LOOP_THEN^
15895 Do not place the keyword @code{loop} in FOR and WHILE loop statements and the
15896 keyword @code{then} in IF statements on a separate line. This option is
15897 incompatible with @option{^--separate-loop-then^/SEPARATE_LOOP_THEN^} option.
15899 @cindex @option{^--use-on-new-line^/USE_ON_NEW_LINE^} (@command{gnatpp})
15900 @item ^--use-on-new-line^/USE_ON_NEW_LINE^
15901 Start each USE clause in a context clause from a separate line.
15903 @cindex @option{^--separate-stmt-name^/STMT_NAME_ON_NEW_LINE^} (@command{gnatpp})
15904 @item ^--separate-stmt-name^/STMT_NAME_ON_NEW_LINE^
15905 Use a separate line for a loop or block statement name, but do not use an extra
15906 indentation level for the statement itself.
15912 The @option{-c1} and @option{-c2} switches are incompatible.
15913 The @option{-c3} and @option{-c4} switches are compatible with each other and
15914 also with @option{-c1} and @option{-c2}. The @option{-c0} switch disables all
15915 the other comment formatting switches.
15917 The @option{-l1}, @option{-l2}, and @option{-l3} switches are incompatible.
15922 For the @option{/COMMENTS_LAYOUT} qualifier:
15925 The @option{DEFAULT} and @option{STANDARD_INDENT} options are incompatible.
15927 The @option{GNAT_BEGINNING} and @option{REFORMAT} options are compatible with
15928 each other and also with @option{DEFAULT} and @option{STANDARD_INDENT}.
15932 The @option{GNAT}, @option{COMPACT}, and @option{UNCOMPACT} options for the
15933 @option{/CONSTRUCT_LAYOUT} qualifier are incompatible.
15936 @node General Text Layout Control
15937 @subsection General Text Layout Control
15940 These switches allow control over line length and indentation.
15943 @item ^-M@var{nnn}^/LINE_LENGTH_MAX=@var{nnn}^
15944 @cindex @option{^-M^/LINE_LENGTH^} (@command{gnatpp})
15945 Maximum line length, @var{nnn} from 32@dots{}256, the default value is 79
15947 @item ^-i@var{nnn}^/INDENTATION_LEVEL=@var{nnn}^
15948 @cindex @option{^-i^/INDENTATION_LEVEL^} (@command{gnatpp})
15949 Indentation level, @var{nnn} from 1@dots{}9, the default value is 3
15951 @item ^-cl@var{nnn}^/CONTINUATION_INDENT=@var{nnn}^
15952 @cindex @option{^-cl^/CONTINUATION_INDENT^} (@command{gnatpp})
15953 Indentation level for continuation lines (relative to the line being
15954 continued), @var{nnn} from 1@dots{}9.
15956 value is one less then the (normal) indentation level, unless the
15957 indentation is set to 1 (in which case the default value for continuation
15958 line indentation is also 1)
15961 @node Other Formatting Options
15962 @subsection Other Formatting Options
15965 These switches control the inclusion of missing end/exit labels, and
15966 the indentation level in @b{case} statements.
15969 @item ^-e^/NO_MISSED_LABELS^
15970 @cindex @option{^-e^/NO_MISSED_LABELS^} (@command{gnatpp})
15971 Do not insert missing end/exit labels. An end label is the name of
15972 a construct that may optionally be repeated at the end of the
15973 construct's declaration;
15974 e.g., the names of packages, subprograms, and tasks.
15975 An exit label is the name of a loop that may appear as target
15976 of an exit statement within the loop.
15977 By default, @command{gnatpp} inserts these end/exit labels when
15978 they are absent from the original source. This option suppresses such
15979 insertion, so that the formatted source reflects the original.
15981 @item ^-ff^/FORM_FEED_AFTER_PRAGMA_PAGE^
15982 @cindex @option{^-ff^/FORM_FEED_AFTER_PRAGMA_PAGE^} (@command{gnatpp})
15983 Insert a Form Feed character after a pragma Page.
15985 @item ^-T@var{nnn}^/MAX_INDENT=@var{nnn}^
15986 @cindex @option{^-T^/MAX_INDENT^} (@command{gnatpp})
15987 Do not use an additional indentation level for @b{case} alternatives
15988 and variants if there are @var{nnn} or more (the default
15990 If @var{nnn} is 0, an additional indentation level is
15991 used for @b{case} alternatives and variants regardless of their number.
15994 @node Setting the Source Search Path
15995 @subsection Setting the Source Search Path
15998 To define the search path for the input source file, @command{gnatpp}
15999 uses the same switches as the GNAT compiler, with the same effects.
16002 @item ^-I^/SEARCH=^@var{dir}
16003 @cindex @option{^-I^/SEARCH^} (@code{gnatpp})
16004 The same as the corresponding gcc switch
16006 @item ^-I-^/NOCURRENT_DIRECTORY^
16007 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@code{gnatpp})
16008 The same as the corresponding gcc switch
16010 @item ^-gnatec^/CONFIGURATION_PRAGMAS_FILE^=@var{path}
16011 @cindex @option{^-gnatec^/CONFIGURATION_PRAGMAS_FILE^} (@code{gnatpp})
16012 The same as the corresponding gcc switch
16014 @item ^--RTS^/RUNTIME_SYSTEM^=@var{path}
16015 @cindex @option{^--RTS^/RUNTIME_SYSTEM^} (@code{gnatpp})
16016 The same as the corresponding gcc switch
16020 @node Output File Control
16021 @subsection Output File Control
16024 By default the output is sent to the file whose name is obtained by appending
16025 the ^@file{.pp}^@file{$PP}^ suffix to the name of the input file
16026 (if the file with this name already exists, it is unconditionally overwritten).
16027 Thus if the input file is @file{^my_ada_proc.adb^MY_ADA_PROC.ADB^} then
16028 @command{gnatpp} will produce @file{^my_ada_proc.adb.pp^MY_ADA_PROC.ADB$PP^}
16030 The output may be redirected by the following switches:
16033 @item ^-pipe^/STANDARD_OUTPUT^
16034 @cindex @option{^-pipe^/STANDARD_OUTPUT^} (@code{gnatpp})
16035 Send the output to @code{Standard_Output}
16037 @item ^-o @var{output_file}^/OUTPUT=@var{output_file}^
16038 @cindex @option{^-o^/OUTPUT^} (@code{gnatpp})
16039 Write the output into @var{output_file}.
16040 If @var{output_file} already exists, @command{gnatpp} terminates without
16041 reading or processing the input file.
16043 @item ^-of ^/FORCED_OUTPUT=^@var{output_file}
16044 @cindex @option{^-of^/FORCED_OUTPUT^} (@code{gnatpp})
16045 Write the output into @var{output_file}, overwriting the existing file
16046 (if one is present).
16048 @item ^-r^/REPLACE^
16049 @cindex @option{^-r^/REPLACE^} (@code{gnatpp})
16050 Replace the input source file with the reformatted output, and copy the
16051 original input source into the file whose name is obtained by appending the
16052 ^@file{.npp}^@file{$NPP}^ suffix to the name of the input file.
16053 If a file with this name already exists, @command{gnatpp} terminates without
16054 reading or processing the input file.
16056 @item ^-rf^/OVERRIDING_REPLACE^
16057 @cindex @option{^-rf^/OVERRIDING_REPLACE^} (@code{gnatpp})
16058 Like @option{^-r^/REPLACE^} except that if the file with the specified name
16059 already exists, it is overwritten.
16061 @item ^-rnb^/REPLACE_NO_BACKUP^
16062 @cindex @option{^-rnb^/REPLACE_NO_BACKUP^} (@code{gnatpp})
16063 Replace the input source file with the reformatted output without
16064 creating any backup copy of the input source.
16066 @item ^--eol=@var{xxx}^/END_OF_LINE=@var{xxx}^
16067 @cindex @option{^--eol^/END_OF_LINE^} (@code{gnatpp})
16068 Specifies the format of the reformatted output file. The @var{xxx}
16069 ^string specified with the switch^option^ may be either
16071 @item ``@option{^dos^DOS^}'' MS DOS style, lines end with CR LF characters
16072 @item ``@option{^crlf^CRLF^}''
16073 the same as @option{^crlf^CRLF^}
16074 @item ``@option{^unix^UNIX^}'' UNIX style, lines end with LF character
16075 @item ``@option{^lf^LF^}''
16076 the same as @option{^unix^UNIX^}
16079 @item ^-W^/RESULT_ENCODING=^@var{e}
16080 @cindex @option{^-W^/RESULT_ENCODING=^} (@command{gnatpp})
16081 Specify the wide character encoding method used to write the code in the
16083 @var{e} is one of the following:
16091 Upper half encoding
16093 @item ^s^SHIFT_JIS^
16103 Brackets encoding (default value)
16109 Options @option{^-pipe^/STANDARD_OUTPUT^},
16110 @option{^-o^/OUTPUT^} and
16111 @option{^-of^/FORCED_OUTPUT^} are allowed only if the call to gnatpp
16112 contains only one file to reformat.
16114 @option{^--eol^/END_OF_LINE^}
16116 @option{^-W^/RESULT_ENCODING^}
16117 cannot be used together
16118 with @option{^-pipe^/STANDARD_OUTPUT^} option.
16120 @node Other gnatpp Switches
16121 @subsection Other @code{gnatpp} Switches
16124 The additional @command{gnatpp} switches are defined in this subsection.
16127 @item ^-files @var{filename}^/FILES=@var{output_file}^
16128 @cindex @option{^-files^/FILES^} (@code{gnatpp})
16129 Take the argument source files from the specified file. This file should be an
16130 ordinary textual file containing file names separated by spaces or
16131 line breaks. You can use this switch more then once in the same call to
16132 @command{gnatpp}. You also can combine this switch with explicit list of
16135 @item ^-v^/VERBOSE^
16136 @cindex @option{^-v^/VERBOSE^} (@code{gnatpp})
16138 @command{gnatpp} generates version information and then
16139 a trace of the actions it takes to produce or obtain the ASIS tree.
16141 @item ^-w^/WARNINGS^
16142 @cindex @option{^-w^/WARNINGS^} (@code{gnatpp})
16144 @command{gnatpp} generates a warning whenever it cannot provide
16145 a required layout in the result source.
16148 @node Formatting Rules
16149 @section Formatting Rules
16152 The following subsections show how @command{gnatpp} treats ``white space'',
16153 comments, program layout, and name casing.
16154 They provide the detailed descriptions of the switches shown above.
16157 * White Space and Empty Lines::
16158 * Formatting Comments::
16159 * Construct Layout::
16163 @node White Space and Empty Lines
16164 @subsection White Space and Empty Lines
16167 @command{gnatpp} does not have an option to control space characters.
16168 It will add or remove spaces according to the style illustrated by the
16169 examples in the @cite{Ada Reference Manual}.
16171 The only format effectors
16172 (see @cite{Ada Reference Manual}, paragraph 2.1(13))
16173 that will appear in the output file are platform-specific line breaks,
16174 and also format effectors within (but not at the end of) comments.
16175 In particular, each horizontal tab character that is not inside
16176 a comment will be treated as a space and thus will appear in the
16177 output file as zero or more spaces depending on
16178 the reformatting of the line in which it appears.
16179 The only exception is a Form Feed character, which is inserted after a
16180 pragma @code{Page} when @option{-ff} is set.
16182 The output file will contain no lines with trailing ``white space'' (spaces,
16185 Empty lines in the original source are preserved
16186 only if they separate declarations or statements.
16187 In such contexts, a
16188 sequence of two or more empty lines is replaced by exactly one empty line.
16189 Note that a blank line will be removed if it separates two ``comment blocks''
16190 (a comment block is a sequence of whole-line comments).
16191 In order to preserve a visual separation between comment blocks, use an
16192 ``empty comment'' (a line comprising only hyphens) rather than an empty line.
16193 Likewise, if for some reason you wish to have a sequence of empty lines,
16194 use a sequence of empty comments instead.
16196 @node Formatting Comments
16197 @subsection Formatting Comments
16200 Comments in Ada code are of two kinds:
16203 a @emph{whole-line comment}, which appears by itself (possibly preceded by
16204 ``white space'') on a line
16207 an @emph{end-of-line comment}, which follows some other Ada lexical element
16212 The indentation of a whole-line comment is that of either
16213 the preceding or following line in
16214 the formatted source, depending on switch settings as will be described below.
16216 For an end-of-line comment, @command{gnatpp} leaves the same number of spaces
16217 between the end of the preceding Ada lexical element and the beginning
16218 of the comment as appear in the original source,
16219 unless either the comment has to be split to
16220 satisfy the line length limitation, or else the next line contains a
16221 whole line comment that is considered a continuation of this end-of-line
16222 comment (because it starts at the same position).
16224 cases, the start of the end-of-line comment is moved right to the nearest
16225 multiple of the indentation level.
16226 This may result in a ``line overflow'' (the right-shifted comment extending
16227 beyond the maximum line length), in which case the comment is split as
16230 There is a difference between @option{^-c1^/COMMENTS_LAYOUT=DEFAULT^}
16231 (GNAT-style comment line indentation)
16232 and @option{^-c2^/COMMENTS_LAYOUT=STANDARD_INDENT^}
16233 (reference-manual comment line indentation).
16234 With reference-manual style, a whole-line comment is indented as if it
16235 were a declaration or statement at the same place
16236 (i.e., according to the indentation of the preceding line(s)).
16237 With GNAT style, a whole-line comment that is immediately followed by an
16238 @b{if} or @b{case} statement alternative, a record variant, or the reserved
16239 word @b{begin}, is indented based on the construct that follows it.
16242 @smallexample @c ada
16254 Reference-manual indentation produces:
16256 @smallexample @c ada
16268 while GNAT-style indentation produces:
16270 @smallexample @c ada
16282 The @option{^-c3^/COMMENTS_LAYOUT=GNAT_BEGINNING^} switch
16283 (GNAT style comment beginning) has the following
16288 For each whole-line comment that does not end with two hyphens,
16289 @command{gnatpp} inserts spaces if necessary after the starting two hyphens
16290 to ensure that there are at least two spaces between these hyphens and the
16291 first non-blank character of the comment.
16295 For an end-of-line comment, if in the original source the next line is a
16296 whole-line comment that starts at the same position
16297 as the end-of-line comment,
16298 then the whole-line comment (and all whole-line comments
16299 that follow it and that start at the same position)
16300 will start at this position in the output file.
16303 That is, if in the original source we have:
16305 @smallexample @c ada
16308 A := B + C; -- B must be in the range Low1..High1
16309 -- C must be in the range Low2..High2
16310 --B+C will be in the range Low1+Low2..High1+High2
16316 Then in the formatted source we get
16318 @smallexample @c ada
16321 A := B + C; -- B must be in the range Low1..High1
16322 -- C must be in the range Low2..High2
16323 -- B+C will be in the range Low1+Low2..High1+High2
16329 A comment that exceeds the line length limit will be split.
16331 @option{^-c4^/COMMENTS_LAYOUT=REFORMAT^} (reformat comment blocks) is set and
16332 the line belongs to a reformattable block, splitting the line generates a
16333 @command{gnatpp} warning.
16334 The @option{^-c4^/COMMENTS_LAYOUT=REFORMAT^} switch specifies that whole-line
16335 comments may be reformatted in typical
16336 word processor style (that is, moving words between lines and putting as
16337 many words in a line as possible).
16340 The @option{^-c5^/COMMENTS_LAYOUT=KEEP_SPECIAL^} switch specifies, that comments
16341 that has a special format (that is, a character that is neither a letter nor digit
16342 not white space nor line break immediately following the leading @code{--} of
16343 the comment) should be without any change moved from the argument source
16344 into reformatted source. This switch allows to preserve comments that are used
16345 as a special marks in the code (e.g.@: SPARK annotation).
16347 @node Construct Layout
16348 @subsection Construct Layout
16351 In several cases the suggested layout in the Ada Reference Manual includes
16352 an extra level of indentation that many programmers prefer to avoid. The
16353 affected cases include:
16357 @item Record type declaration (RM 3.8)
16359 @item Record representation clause (RM 13.5.1)
16361 @item Loop statement in case if a loop has a statement identifier (RM 5.6)
16363 @item Block statement in case if a block has a statement identifier (RM 5.6)
16367 In compact mode (when GNAT style layout or compact layout is set),
16368 the pretty printer uses one level of indentation instead
16369 of two. This is achieved in the record definition and record representation
16370 clause cases by putting the @code{record} keyword on the same line as the
16371 start of the declaration or representation clause, and in the block and loop
16372 case by putting the block or loop header on the same line as the statement
16376 The difference between GNAT style @option{^-l1^/CONSTRUCT_LAYOUT=GNAT^}
16377 and compact @option{^-l2^/CONSTRUCT_LAYOUT=COMPACT^}
16378 layout on the one hand, and uncompact layout
16379 @option{^-l3^/CONSTRUCT_LAYOUT=UNCOMPACT^} on the other hand,
16380 can be illustrated by the following examples:
16384 @multitable @columnfractions .5 .5
16385 @item @i{GNAT style, compact layout} @tab @i{Uncompact layout}
16388 @smallexample @c ada
16395 @smallexample @c ada
16404 @smallexample @c ada
16406 a at 0 range 0 .. 31;
16407 b at 4 range 0 .. 31;
16411 @smallexample @c ada
16414 a at 0 range 0 .. 31;
16415 b at 4 range 0 .. 31;
16420 @smallexample @c ada
16428 @smallexample @c ada
16438 @smallexample @c ada
16439 Clear : for J in 1 .. 10 loop
16444 @smallexample @c ada
16446 for J in 1 .. 10 loop
16457 GNAT style, compact layout Uncompact layout
16459 type q is record type q is
16460 a : integer; record
16461 b : integer; a : integer;
16462 end record; b : integer;
16465 for q use record for q use
16466 a at 0 range 0 .. 31; record
16467 b at 4 range 0 .. 31; a at 0 range 0 .. 31;
16468 end record; b at 4 range 0 .. 31;
16471 Block : declare Block :
16472 A : Integer := 3; declare
16473 begin A : Integer := 3;
16475 end Block; Proc (A, A);
16478 Clear : for J in 1 .. 10 loop Clear :
16479 A (J) := 0; for J in 1 .. 10 loop
16480 end loop Clear; A (J) := 0;
16487 A further difference between GNAT style layout and compact layout is that
16488 GNAT style layout inserts empty lines as separation for
16489 compound statements, return statements and bodies.
16491 Note that the layout specified by
16492 @option{^--separate-stmt-name^/STMT_NAME_ON_NEW_LINE^}
16493 for named block and loop statements overrides the layout defined by these
16494 constructs by @option{^-l1^/CONSTRUCT_LAYOUT=GNAT^},
16495 @option{^-l2^/CONSTRUCT_LAYOUT=COMPACT^} or
16496 @option{^-l3^/CONSTRUCT_LAYOUT=UNCOMPACT^} option.
16499 @subsection Name Casing
16502 @command{gnatpp} always converts the usage occurrence of a (simple) name to
16503 the same casing as the corresponding defining identifier.
16505 You control the casing for defining occurrences via the
16506 @option{^-n^/NAME_CASING^} switch.
16508 With @option{-nD} (``as declared'', which is the default),
16511 With @option{/NAME_CASING=AS_DECLARED}, which is the default,
16513 defining occurrences appear exactly as in the source file
16514 where they are declared.
16515 The other ^values for this switch^options for this qualifier^ ---
16516 @option{^-nU^UPPER_CASE^},
16517 @option{^-nL^LOWER_CASE^},
16518 @option{^-nM^MIXED_CASE^} ---
16520 ^upper, lower, or mixed case, respectively^the corresponding casing^.
16521 If @command{gnatpp} changes the casing of a defining
16522 occurrence, it analogously changes the casing of all the
16523 usage occurrences of this name.
16525 If the defining occurrence of a name is not in the source compilation unit
16526 currently being processed by @command{gnatpp}, the casing of each reference to
16527 this name is changed according to the value of the @option{^-n^/NAME_CASING^}
16528 switch (subject to the dictionary file mechanism described below).
16529 Thus @command{gnatpp} acts as though the @option{^-n^/NAME_CASING^} switch
16531 casing for the defining occurrence of the name.
16533 Some names may need to be spelled with casing conventions that are not
16534 covered by the upper-, lower-, and mixed-case transformations.
16535 You can arrange correct casing by placing such names in a
16536 @emph{dictionary file},
16537 and then supplying a @option{^-D^/DICTIONARY^} switch.
16538 The casing of names from dictionary files overrides
16539 any @option{^-n^/NAME_CASING^} switch.
16541 To handle the casing of Ada predefined names and the names from GNAT libraries,
16542 @command{gnatpp} assumes a default dictionary file.
16543 The name of each predefined entity is spelled with the same casing as is used
16544 for the entity in the @cite{Ada Reference Manual}.
16545 The name of each entity in the GNAT libraries is spelled with the same casing
16546 as is used in the declaration of that entity.
16548 The @w{@option{^-D-^/SPECIFIC_CASING^}} switch suppresses the use of the
16549 default dictionary file.
16550 Instead, the casing for predefined and GNAT-defined names will be established
16551 by the @option{^-n^/NAME_CASING^} switch or explicit dictionary files.
16552 For example, by default the names @code{Ada.Text_IO} and @code{GNAT.OS_Lib}
16553 will appear as just shown,
16554 even in the presence of a @option{^-nU^/NAME_CASING=UPPER_CASE^} switch.
16555 To ensure that even such names are rendered in uppercase,
16556 additionally supply the @w{@option{^-D-^/SPECIFIC_CASING^}} switch
16557 (or else, less conveniently, place these names in upper case in a dictionary
16560 A dictionary file is
16561 a plain text file; each line in this file can be either a blank line
16562 (containing only space characters and ASCII.HT characters), an Ada comment
16563 line, or the specification of exactly one @emph{casing schema}.
16565 A casing schema is a string that has the following syntax:
16569 @var{casing_schema} ::= @var{identifier} | *@var{simple_identifier}*
16571 @var{simple_identifier} ::= @var{letter}@{@var{letter_or_digit}@}
16576 (See @cite{Ada Reference Manual}, Section 2.3) for the definition of the
16577 @var{identifier} lexical element and the @var{letter_or_digit} category.)
16579 The casing schema string can be followed by white space and/or an Ada-style
16580 comment; any amount of white space is allowed before the string.
16582 If a dictionary file is passed as
16584 the value of a @option{-D@var{file}} switch
16587 an option to the @option{/DICTIONARY} qualifier
16590 simple name and every identifier, @command{gnatpp} checks if the dictionary
16591 defines the casing for the name or for some of its parts (the term ``subword''
16592 is used below to denote the part of a name which is delimited by ``_'' or by
16593 the beginning or end of the word and which does not contain any ``_'' inside):
16597 if the whole name is in the dictionary, @command{gnatpp} uses for this name
16598 the casing defined by the dictionary; no subwords are checked for this word
16601 for every subword @command{gnatpp} checks if the dictionary contains the
16602 corresponding string of the form @code{*@var{simple_identifier}*},
16603 and if it does, the casing of this @var{simple_identifier} is used
16607 if the whole name does not contain any ``_'' inside, and if for this name
16608 the dictionary contains two entries - one of the form @var{identifier},
16609 and another - of the form *@var{simple_identifier}*, then the first one
16610 is applied to define the casing of this name
16613 if more than one dictionary file is passed as @command{gnatpp} switches, each
16614 dictionary adds new casing exceptions and overrides all the existing casing
16615 exceptions set by the previous dictionaries
16618 when @command{gnatpp} checks if the word or subword is in the dictionary,
16619 this check is not case sensitive
16623 For example, suppose we have the following source to reformat:
16625 @smallexample @c ada
16628 name1 : integer := 1;
16629 name4_name3_name2 : integer := 2;
16630 name2_name3_name4 : Boolean;
16633 name2_name3_name4 := name4_name3_name2 > name1;
16639 And suppose we have two dictionaries:
16656 If @command{gnatpp} is called with the following switches:
16660 @command{gnatpp -nM -D dict1 -D dict2 test.adb}
16663 @command{gnatpp test.adb /NAME_CASING=MIXED_CASE /DICTIONARY=(dict1, dict2)}
16668 then we will get the following name casing in the @command{gnatpp} output:
16670 @smallexample @c ada
16673 NAME1 : Integer := 1;
16674 Name4_NAME3_Name2 : Integer := 2;
16675 Name2_NAME3_Name4 : Boolean;
16678 Name2_NAME3_Name4 := Name4_NAME3_Name2 > NAME1;
16683 @c *********************************
16684 @node The GNAT Metric Tool gnatmetric
16685 @chapter The GNAT Metric Tool @command{gnatmetric}
16687 @cindex Metric tool
16690 ^The @command{gnatmetric} tool^@command{GNAT METRIC}^ is an ASIS-based utility
16691 for computing various program metrics.
16692 It takes an Ada source file as input and generates a file containing the
16693 metrics data as output. Various switches control which
16694 metrics are computed and output.
16696 @command{gnatmetric} generates and uses the ASIS
16697 tree for the input source and thus requires the input to be syntactically and
16698 semantically legal.
16699 If this condition is not met, @command{gnatmetric} will generate
16700 an error message; no metric information for this file will be
16701 computed and reported.
16703 If the compilation unit contained in the input source depends semantically
16704 upon units in files located outside the current directory, you have to provide
16705 the source search path when invoking @command{gnatmetric}.
16706 If it depends semantically upon units that are contained
16707 in files with names that do not follow the GNAT file naming rules, you have to
16708 provide the configuration file describing the corresponding naming scheme (see
16709 the description of the @command{gnatmetric} switches below.)
16710 Alternatively, you may use a project file and invoke @command{gnatmetric}
16711 through the @command{gnat} driver.
16713 The @command{gnatmetric} command has the form
16716 $ gnatmetric @ovar{switches} @{@var{filename}@} @r{[}-cargs @var{gcc_switches}@r{]}
16723 @var{switches} specify the metrics to compute and define the destination for
16727 Each @var{filename} is the name (including the extension) of a source
16728 file to process. ``Wildcards'' are allowed, and
16729 the file name may contain path information.
16730 If no @var{filename} is supplied, then the @var{switches} list must contain
16732 @option{-files} switch (@pxref{Other gnatmetric Switches}).
16733 Including both a @option{-files} switch and one or more
16734 @var{filename} arguments is permitted.
16737 @samp{-cargs @var{gcc_switches}} is a list of switches for
16738 @command{gcc}. They will be passed on to all compiler invocations made by
16739 @command{gnatmetric} to generate the ASIS trees. Here you can provide
16740 @option{^-I^/INCLUDE_DIRS=^} switches to form the source search path,
16741 and use the @option{-gnatec} switch to set the configuration file.
16745 * Switches for gnatmetric::
16748 @node Switches for gnatmetric
16749 @section Switches for @command{gnatmetric}
16752 The following subsections describe the various switches accepted by
16753 @command{gnatmetric}, organized by category.
16756 * Output Files Control::
16757 * Disable Metrics For Local Units::
16758 * Specifying a set of metrics to compute::
16759 * Other gnatmetric Switches::
16760 * Generate project-wide metrics::
16763 @node Output Files Control
16764 @subsection Output File Control
16765 @cindex Output file control in @command{gnatmetric}
16768 @command{gnatmetric} has two output formats. It can generate a
16769 textual (human-readable) form, and also XML. By default only textual
16770 output is generated.
16772 When generating the output in textual form, @command{gnatmetric} creates
16773 for each Ada source file a corresponding text file
16774 containing the computed metrics, except for the case when the set of metrics
16775 specified by gnatmetric parameters consists only of metrics that are computed
16776 for the whole set of analyzed sources, but not for each Ada source.
16777 By default, this file is placed in the same directory as where the source
16778 file is located, and its name is obtained
16779 by appending the ^@file{.metrix}^@file{$METRIX}^ suffix to the name of the
16782 All the output information generated in XML format is placed in a single
16783 file. By default this file is placed in the current directory and has the
16784 name ^@file{metrix.xml}^@file{METRIX$XML}^.
16786 Some of the computed metrics are summed over the units passed to
16787 @command{gnatmetric}; for example, the total number of lines of code.
16788 By default this information is sent to @file{stdout}, but a file
16789 can be specified with the @option{-og} switch.
16791 The following switches control the @command{gnatmetric} output:
16794 @cindex @option{^-x^/XML^} (@command{gnatmetric})
16796 Generate the XML output
16798 @cindex @option{^-nt^/NO_TEXT^} (@command{gnatmetric})
16799 @item ^-nt^/NO_TEXT^
16800 Do not generate the output in text form (implies @option{^-x^/XML^})
16802 @cindex @option{^-d^/DIRECTORY^} (@command{gnatmetric})
16803 @item ^-d @var{output_dir}^/DIRECTORY=@var{output_dir}^
16804 Put textual files with detailed metrics into @var{output_dir}
16806 @cindex @option{^-o^/SUFFIX_DETAILS^} (@command{gnatmetric})
16807 @item ^-o @var{file_suffix}^/SUFFIX_DETAILS=@var{file_suffix}^
16808 Use @var{file_suffix}, instead of ^@file{.metrix}^@file{$METRIX}^
16809 in the name of the output file.
16811 @cindex @option{^-og^/GLOBAL_OUTPUT^} (@command{gnatmetric})
16812 @item ^-og @var{file_name}^/GLOBAL_OUTPUT=@var{file_name}^
16813 Put global metrics into @var{file_name}
16815 @cindex @option{^-ox^/XML_OUTPUT^} (@command{gnatmetric})
16816 @item ^-ox @var{file_name}^/XML_OUTPUT=@var{file_name}^
16817 Put the XML output into @var{file_name} (also implies @option{^-x^/XML^})
16819 @cindex @option{^-sfn^/SHORT_SOURCE_FILE_NAME^} (@command{gnatmetric})
16820 @item ^-sfn^/SHORT_SOURCE_FILE_NAME^
16821 Use ``short'' source file names in the output. (The @command{gnatmetric}
16822 output includes the name(s) of the Ada source file(s) from which the metrics
16823 are computed. By default each name includes the absolute path. The
16824 @option{^-sfn^/SHORT_SOURCE_FILE_NAME^} switch causes @command{gnatmetric}
16825 to exclude all directory information from the file names that are output.)
16829 @node Disable Metrics For Local Units
16830 @subsection Disable Metrics For Local Units
16831 @cindex Disable Metrics For Local Units in @command{gnatmetric}
16834 @command{gnatmetric} relies on the GNAT compilation model @minus{}
16836 unit per one source file. It computes line metrics for the whole source
16837 file, and it also computes syntax
16838 and complexity metrics for the file's outermost unit.
16840 By default, @command{gnatmetric} will also compute all metrics for certain
16841 kinds of locally declared program units:
16845 subprogram (and generic subprogram) bodies;
16848 package (and generic package) specs and bodies;
16851 task object and type specifications and bodies;
16854 protected object and type specifications and bodies.
16858 These kinds of entities will be referred to as
16859 @emph{eligible local program units}, or simply @emph{eligible local units},
16860 @cindex Eligible local unit (for @command{gnatmetric})
16861 in the discussion below.
16863 Note that a subprogram declaration, generic instantiation,
16864 or renaming declaration only receives metrics
16865 computation when it appear as the outermost entity
16868 Suppression of metrics computation for eligible local units can be
16869 obtained via the following switch:
16872 @cindex @option{^-n@var{x}^/SUPPRESS^} (@command{gnatmetric})
16873 @item ^-nolocal^/SUPPRESS=LOCAL_DETAILS^
16874 Do not compute detailed metrics for eligible local program units
16878 @node Specifying a set of metrics to compute
16879 @subsection Specifying a set of metrics to compute
16882 By default all the metrics are computed and reported. The switches
16883 described in this subsection allow you to control, on an individual
16884 basis, whether metrics are computed and
16885 reported. If at least one positive metric
16886 switch is specified (that is, a switch that defines that a given
16887 metric or set of metrics is to be computed), then only
16888 explicitly specified metrics are reported.
16891 * Line Metrics Control::
16892 * Syntax Metrics Control::
16893 * Complexity Metrics Control::
16894 * Object-Oriented Metrics Control::
16897 @node Line Metrics Control
16898 @subsubsection Line Metrics Control
16899 @cindex Line metrics control in @command{gnatmetric}
16902 For any (legal) source file, and for each of its
16903 eligible local program units, @command{gnatmetric} computes the following
16908 the total number of lines;
16911 the total number of code lines (i.e., non-blank lines that are not comments)
16914 the number of comment lines
16917 the number of code lines containing end-of-line comments;
16920 the comment percentage: the ratio between the number of lines that contain
16921 comments and the number of all non-blank lines, expressed as a percentage;
16924 the number of empty lines and lines containing only space characters and/or
16925 format effectors (blank lines)
16928 the average number of code lines in subprogram bodies, task bodies, entry
16929 bodies and statement sequences in package bodies (this metric is only computed
16930 across the whole set of the analyzed units)
16935 @command{gnatmetric} sums the values of the line metrics for all the
16936 files being processed and then generates the cumulative results. The tool
16937 also computes for all the files being processed the average number of code
16940 You can use the following switches to select the specific line metrics
16941 to be computed and reported.
16944 @cindex @option{^--lines@var{x}^/LINE_COUNT_METRICS^} (@command{gnatmetric})
16947 @cindex @option{--no-lines@var{x}}
16950 @item ^--lines-all^/LINE_COUNT_METRICS=ALL_ON^
16951 Report all the line metrics
16953 @item ^--no-lines-all^/LINE_COUNT_METRICS=ALL_OFF^
16954 Do not report any of line metrics
16956 @item ^--lines^/LINE_COUNT_METRICS=ALL_LINES_ON^
16957 Report the number of all lines
16959 @item ^--no-lines^/LINE_COUNT_METRICS=ALL_LINES_OFF^
16960 Do not report the number of all lines
16962 @item ^--lines-code^/LINE_COUNT_METRICS=CODE_LINES_ON^
16963 Report the number of code lines
16965 @item ^--no-lines-code^/LINE_COUNT_METRICS=CODE_LINES_OFF^
16966 Do not report the number of code lines
16968 @item ^--lines-comment^/LINE_COUNT_METRICS=COMMENT_LINES_ON^
16969 Report the number of comment lines
16971 @item ^--no-lines-comment^/LINE_COUNT_METRICS=COMMENT_LINES_OFF^
16972 Do not report the number of comment lines
16974 @item ^--lines-eol-comment^/LINE_COUNT_METRICS=CODE_COMMENT_LINES_ON^
16975 Report the number of code lines containing
16976 end-of-line comments
16978 @item ^--no-lines-eol-comment^/LINE_COUNT_METRICS=CODE_COMMENT_LINES_OFF^
16979 Do not report the number of code lines containing
16980 end-of-line comments
16982 @item ^--lines-ratio^/LINE_COUNT_METRICS=COMMENT_PERCENTAGE_ON^
16983 Report the comment percentage in the program text
16985 @item ^--no-lines-ratio^/LINE_COUNT_METRICS=COMMENT_PERCENTAGE_OFF^
16986 Do not report the comment percentage in the program text
16988 @item ^--lines-blank^/LINE_COUNT_METRICS=BLANK_LINES_ON^
16989 Report the number of blank lines
16991 @item ^--no-lines-blank^/LINE_COUNT_METRICS=BLANK_LINES_OFF^
16992 Do not report the number of blank lines
16994 @item ^--lines-average^/LINE_COUNT_METRICS=AVERAGE_BODY_LINES_ON^
16995 Report the average number of code lines in subprogram bodies, task bodies,
16996 entry bodies and statement sequences in package bodies. The metric is computed
16997 and reported for the whole set of processed Ada sources only.
16999 @item ^--no-lines-average^/LINE_COUNT_METRICS=AVERAGE_BODY_LINES_OFF^
17000 Do not report the average number of code lines in subprogram bodies,
17001 task bodies, entry bodies and statement sequences in package bodies.
17005 @node Syntax Metrics Control
17006 @subsubsection Syntax Metrics Control
17007 @cindex Syntax metrics control in @command{gnatmetric}
17010 @command{gnatmetric} computes various syntactic metrics for the
17011 outermost unit and for each eligible local unit:
17014 @item LSLOC (``Logical Source Lines Of Code'')
17015 The total number of declarations and the total number of statements
17017 @item Maximal static nesting level of inner program units
17019 @cite{Ada Reference Manual}, 10.1(1), ``A program unit is either a
17020 package, a task unit, a protected unit, a
17021 protected entry, a generic unit, or an explicitly declared subprogram other
17022 than an enumeration literal.''
17024 @item Maximal nesting level of composite syntactic constructs
17025 This corresponds to the notion of the
17026 maximum nesting level in the GNAT built-in style checks
17027 (@pxref{Style Checking})
17031 For the outermost unit in the file, @command{gnatmetric} additionally computes
17032 the following metrics:
17035 @item Public subprograms
17036 This metric is computed for package specs. It is the
17037 number of subprograms and generic subprograms declared in the visible
17038 part (including the visible part of nested packages, protected objects, and
17041 @item All subprograms
17042 This metric is computed for bodies and subunits. The
17043 metric is equal to a total number of subprogram bodies in the compilation
17045 Neither generic instantiations nor renamings-as-a-body nor body stubs
17046 are counted. Any subprogram body is counted, independently of its nesting
17047 level and enclosing constructs. Generic bodies and bodies of protected
17048 subprograms are counted in the same way as ``usual'' subprogram bodies.
17051 This metric is computed for package specs and
17052 generic package declarations. It is the total number of types
17053 that can be referenced from outside this compilation unit, plus the
17054 number of types from all the visible parts of all the visible generic
17055 packages. Generic formal types are not counted. Only types, not subtypes,
17059 Along with the total number of public types, the following
17060 types are counted and reported separately:
17067 Root tagged types (abstract, non-abstract, private, non-private). Type
17068 extensions are @emph{not} counted
17071 Private types (including private extensions)
17082 This metric is computed for any compilation unit. It is equal to the total
17083 number of the declarations of different types given in the compilation unit.
17084 The private and the corresponding full type declaration are counted as one
17085 type declaration. Incomplete type declarations and generic formal types
17087 No distinction is made among different kinds of types (abstract,
17088 private etc.); the total number of types is computed and reported.
17093 By default, all the syntax metrics are computed and reported. You can use the
17094 following switches to select specific syntax metrics.
17098 @cindex @option{^--syntax@var{x}^/SYNTAX_METRICS^} (@command{gnatmetric})
17101 @cindex @option{--no-syntax@var{x}} (@command{gnatmetric})
17104 @item ^--syntax-all^/SYNTAX_METRICS=ALL_ON^
17105 Report all the syntax metrics
17107 @item ^--no-syntax-all^/ALL_OFF^
17108 Do not report any of syntax metrics
17110 @item ^--declarations^/SYNTAX_METRICS=DECLARATIONS_ON^
17111 Report the total number of declarations
17113 @item ^--no-declarations^/SYNTAX_METRICS=DECLARATIONS_OFF^
17114 Do not report the total number of declarations
17116 @item ^--statements^/SYNTAX_METRICS=STATEMENTS_ON^
17117 Report the total number of statements
17119 @item ^--no-statements^/SYNTAX_METRICS=STATEMENTS_OFF^
17120 Do not report the total number of statements
17122 @item ^--public-subprograms^/SYNTAX_METRICS=PUBLIC_SUBPROGRAMS_ON^
17123 Report the number of public subprograms in a compilation unit
17125 @item ^--no-public-subprograms^/SYNTAX_METRICS=PUBLIC_SUBPROGRAMS_OFF^
17126 Do not report the number of public subprograms in a compilation unit
17128 @item ^--all-subprograms^/SYNTAX_METRICS=ALL_SUBPROGRAMS_ON^
17129 Report the number of all the subprograms in a compilation unit
17131 @item ^--no-all-subprograms^/SYNTAX_METRICS=ALL_SUBPROGRAMS_OFF^
17132 Do not report the number of all the subprograms in a compilation unit
17134 @item ^--public-types^/SYNTAX_METRICS=PUBLIC_TYPES_ON^
17135 Report the number of public types in a compilation unit
17137 @item ^--no-public-types^/SYNTAX_METRICS=PUBLIC_TYPES_OFF^
17138 Do not report the number of public types in a compilation unit
17140 @item ^--all-types^/SYNTAX_METRICS=ALL_TYPES_ON^
17141 Report the number of all the types in a compilation unit
17143 @item ^--no-all-types^/SYNTAX_METRICS=ALL_TYPES_OFF^
17144 Do not report the number of all the types in a compilation unit
17146 @item ^--unit-nesting^/SYNTAX_METRICS=UNIT_NESTING_ON^
17147 Report the maximal program unit nesting level
17149 @item ^--no-unit-nesting^/SYNTAX_METRICS=UNIT_NESTING_OFF^
17150 Do not report the maximal program unit nesting level
17152 @item ^--construct-nesting^/SYNTAX_METRICS=CONSTRUCT_NESTING_ON^
17153 Report the maximal construct nesting level
17155 @item ^--no-construct-nesting^/SYNTAX_METRICS=CONSTRUCT_NESTING_OFF^
17156 Do not report the maximal construct nesting level
17160 @node Complexity Metrics Control
17161 @subsubsection Complexity Metrics Control
17162 @cindex Complexity metrics control in @command{gnatmetric}
17165 For a program unit that is an executable body (a subprogram body (including
17166 generic bodies), task body, entry body or a package body containing
17167 its own statement sequence) @command{gnatmetric} computes the following
17168 complexity metrics:
17172 McCabe cyclomatic complexity;
17175 McCabe essential complexity;
17178 maximal loop nesting level
17183 The McCabe complexity metrics are defined
17184 in @url{http://www.mccabe.com/pdf/nist235r.pdf}
17186 According to McCabe, both control statements and short-circuit control forms
17187 should be taken into account when computing cyclomatic complexity. For each
17188 body, we compute three metric values:
17192 the complexity introduced by control
17193 statements only, without taking into account short-circuit forms,
17196 the complexity introduced by short-circuit control forms only, and
17200 cyclomatic complexity, which is the sum of these two values.
17204 When computing cyclomatic and essential complexity, @command{gnatmetric} skips
17205 the code in the exception handlers and in all the nested program units.
17207 By default, all the complexity metrics are computed and reported.
17208 For more fine-grained control you can use
17209 the following switches:
17212 @cindex @option{^-complexity@var{x}^/COMPLEXITY_METRICS^} (@command{gnatmetric})
17215 @cindex @option{--no-complexity@var{x}}
17218 @item ^--complexity-all^/COMPLEXITY_METRICS=ALL_ON^
17219 Report all the complexity metrics
17221 @item ^--no-complexity-all^/COMPLEXITY_METRICS=ALL_OFF^
17222 Do not report any of complexity metrics
17224 @item ^--complexity-cyclomatic^/COMPLEXITY_METRICS=CYCLOMATIC_ON^
17225 Report the McCabe Cyclomatic Complexity
17227 @item ^--no-complexity-cyclomatic^/COMPLEXITY_METRICS=CYCLOMATIC_OFF^
17228 Do not report the McCabe Cyclomatic Complexity
17230 @item ^--complexity-essential^/COMPLEXITY_METRICS=ESSENTIAL_ON^
17231 Report the Essential Complexity
17233 @item ^--no-complexity-essential^/COMPLEXITY_METRICS=ESSENTIAL_OFF^
17234 Do not report the Essential Complexity
17236 @item ^--loop-nesting^/COMPLEXITY_METRICS=LOOP_NESTING_ON^
17237 Report maximal loop nesting level
17239 @item ^--no-loop-nesting^/COMPLEXITY_METRICS=LOOP_NESTING_OFF^
17240 Do not report maximal loop nesting level
17242 @item ^--complexity-average^/COMPLEXITY_METRICS=AVERAGE_COMPLEXITY_ON^
17243 Report the average McCabe Cyclomatic Complexity for all the subprogram bodies,
17244 task bodies, entry bodies and statement sequences in package bodies.
17245 The metric is computed and reported for whole set of processed Ada sources
17248 @item ^--no-complexity-average^/COMPLEXITY_METRICS=AVERAGE_COMPLEXITY_OFF^
17249 Do not report the average McCabe Cyclomatic Complexity for all the subprogram
17250 bodies, task bodies, entry bodies and statement sequences in package bodies
17252 @cindex @option{^-ne^/NO_EXITS_AS_GOTOS^} (@command{gnatmetric})
17253 @item ^-ne^/NO_EXITS_AS_GOTOS^
17254 Do not consider @code{exit} statements as @code{goto}s when
17255 computing Essential Complexity
17260 @node Object-Oriented Metrics Control
17261 @subsubsection Object-Oriented Metrics Control
17262 @cindex Object-Oriented metrics control in @command{gnatmetric}
17265 @cindex Coupling metrics (in in @command{gnatmetric})
17266 Coupling metrics are object-oriented metrics that measure the
17267 dependencies between a given class (or a group of classes) and the
17268 ``external world'' (that is, the other classes in the program). In this
17269 subsection the term ``class'' is used in its
17270 traditional object-oriented programming sense
17271 (an instantiable module that contains data and/or method members).
17272 A @emph{category} (of classes)
17273 is a group of closely related classes that are reused and/or
17276 A class @code{K}'s @emph{efferent coupling} is the number of classes
17277 that @code{K} depends upon.
17278 A category's efferent coupling is the number of classes outside the
17279 category that the classes inside the category depend upon.
17281 A class @code{K}'s @emph{afferent coupling} is the number of classes
17282 that depend upon @code{K}.
17283 A category's afferent coupling is the number of classes outside the
17284 category that depend on classes belonging to the category.
17286 Ada's implementation of the object-oriented paradigm does not use the
17287 traditional class notion, so the definition of the coupling
17288 metrics for Ada maps the class and class category notions
17289 onto Ada constructs.
17291 For the coupling metrics, several kinds of modules -- a library package,
17292 a library generic package, and a library generic package instantiation --
17293 that define a tagged type or an interface type are
17294 considered to be a class. A category consists of a library package (or
17295 a library generic package) that defines a tagged or an interface type,
17296 together with all its descendant (generic) packages that define tagged
17297 or interface types. For any package counted as a class,
17298 its body (if any) is considered
17299 together with its spec when counting the dependencies. For dependencies
17300 between classes, the Ada semantic dependencies are considered.
17301 For coupling metrics, only dependencies on units that are considered as
17302 classes, are considered.
17304 When computing coupling metrics, @command{gnatmetric} counts only
17305 dependencies between units that are arguments of the gnatmetric call.
17306 Coupling metrics are program-wide (or project-wide) metrics, so to
17307 get a valid result, you should call @command{gnatmetric} for
17308 the whole set of sources that make up your program. It can be done
17309 by calling @command{gnatmetric} from the GNAT driver with @option{-U}
17310 option (see See @ref{The GNAT Driver and Project Files} for details.
17312 By default, all the coupling metrics are disabled. You can use the following
17313 switches to specify the coupling metrics to be computed and reported:
17318 @cindex @option{--package@var{x}} (@command{gnatmetric})
17319 @cindex @option{--no-package@var{x}} (@command{gnatmetric})
17320 @cindex @option{--category@var{x}} (@command{gnatmetric})
17321 @cindex @option{--no-category@var{x}} (@command{gnatmetric})
17325 @cindex @option{/COUPLING_METRICS} (@command{gnatmetric})
17328 @item ^--coupling-all^/COUPLING_METRICS=ALL_ON^
17329 Report all the coupling metrics
17331 @item ^--no-coupling-all^/COUPLING_METRICS=ALL_OFF^
17332 Do not report any of metrics
17334 @item ^--package-efferent-coupling^/COUPLING_METRICS=PACKAGE_EFFERENT_ON^
17335 Report package efferent coupling
17337 @item ^--no-package-efferent-coupling^/COUPLING_METRICS=PACKAGE_EFFERENT_OFF^
17338 Do not report package efferent coupling
17340 @item ^--package-afferent-coupling^/COUPLING_METRICS=PACKAGE_AFFERENT_ON^
17341 Report package afferent coupling
17343 @item ^--no-package-afferent-coupling^/COUPLING_METRICS=PACKAGE_AFFERENT_OFF^
17344 Do not report package afferent coupling
17346 @item ^--category-efferent-coupling^/COUPLING_METRICS=CATEGORY_EFFERENT_ON^
17347 Report category efferent coupling
17349 @item ^--no-category-efferent-coupling^/COUPLING_METRICS=CATEGORY_EFFERENT_OFF^
17350 Do not report category efferent coupling
17352 @item ^--category-afferent-coupling^/COUPLING_METRICS=CATEGORY_AFFERENT_ON^
17353 Report category afferent coupling
17355 @item ^--no-category-afferent-coupling^/COUPLING_METRICS=CATEGORY_AFFERENT_OFF^
17356 Do not report category afferent coupling
17360 @node Other gnatmetric Switches
17361 @subsection Other @code{gnatmetric} Switches
17364 Additional @command{gnatmetric} switches are as follows:
17367 @item ^-files @var{filename}^/FILES=@var{filename}^
17368 @cindex @option{^-files^/FILES^} (@code{gnatmetric})
17369 Take the argument source files from the specified file. This file should be an
17370 ordinary text file containing file names separated by spaces or
17371 line breaks. You can use this switch more then once in the same call to
17372 @command{gnatmetric}. You also can combine this switch with
17373 an explicit list of files.
17375 @item ^-v^/VERBOSE^
17376 @cindex @option{^-v^/VERBOSE^} (@code{gnatmetric})
17378 @command{gnatmetric} generates version information and then
17379 a trace of sources being processed.
17381 @item ^-dv^/DEBUG_OUTPUT^
17382 @cindex @option{^-dv^/DEBUG_OUTPUT^} (@code{gnatmetric})
17384 @command{gnatmetric} generates various messages useful to understand what
17385 happens during the metrics computation
17388 @cindex @option{^-q^/QUIET^} (@code{gnatmetric})
17392 @node Generate project-wide metrics
17393 @subsection Generate project-wide metrics
17395 In order to compute metrics on all units of a given project, you can use
17396 the @command{gnat} driver along with the @option{-P} option:
17402 If the project @code{proj} depends upon other projects, you can compute
17403 the metrics on the project closure using the @option{-U} option:
17405 gnat metric -Pproj -U
17409 Finally, if not all the units are relevant to a particular main
17410 program in the project closure, you can generate metrics for the set
17411 of units needed to create a given main program (unit closure) using
17412 the @option{-U} option followed by the name of the main unit:
17414 gnat metric -Pproj -U main
17418 @c ***********************************
17419 @node File Name Krunching Using gnatkr
17420 @chapter File Name Krunching Using @code{gnatkr}
17424 This chapter discusses the method used by the compiler to shorten
17425 the default file names chosen for Ada units so that they do not
17426 exceed the maximum length permitted. It also describes the
17427 @code{gnatkr} utility that can be used to determine the result of
17428 applying this shortening.
17432 * Krunching Method::
17433 * Examples of gnatkr Usage::
17437 @section About @code{gnatkr}
17440 The default file naming rule in GNAT
17441 is that the file name must be derived from
17442 the unit name. The exact default rule is as follows:
17445 Take the unit name and replace all dots by hyphens.
17447 If such a replacement occurs in the
17448 second character position of a name, and the first character is
17449 ^@samp{a}, @samp{g}, @samp{s}, or @samp{i}, ^@samp{A}, @samp{G}, @samp{S}, or @samp{I},^
17450 then replace the dot by the character
17451 ^@samp{~} (tilde)^@samp{$} (dollar sign)^
17452 instead of a minus.
17454 The reason for this exception is to avoid clashes
17455 with the standard names for children of System, Ada, Interfaces,
17456 and GNAT, which use the prefixes
17457 ^@samp{s-}, @samp{a-}, @samp{i-}, and @samp{g-},^@samp{S-}, @samp{A-}, @samp{I-}, and @samp{G-},^
17460 The @option{^-gnatk^/FILE_NAME_MAX_LENGTH=^@var{nn}}
17461 switch of the compiler activates a ``krunching''
17462 circuit that limits file names to nn characters (where nn is a decimal
17463 integer). For example, using OpenVMS,
17464 where the maximum file name length is
17465 39, the value of nn is usually set to 39, but if you want to generate
17466 a set of files that would be usable if ported to a system with some
17467 different maximum file length, then a different value can be specified.
17468 The default value of 39 for OpenVMS need not be specified.
17470 The @code{gnatkr} utility can be used to determine the krunched name for
17471 a given file, when krunched to a specified maximum length.
17474 @section Using @code{gnatkr}
17477 The @code{gnatkr} command has the form
17481 $ gnatkr @var{name} @ovar{length}
17487 $ gnatkr @var{name} /COUNT=nn
17492 @var{name} is the uncrunched file name, derived from the name of the unit
17493 in the standard manner described in the previous section (i.e., in particular
17494 all dots are replaced by hyphens). The file name may or may not have an
17495 extension (defined as a suffix of the form period followed by arbitrary
17496 characters other than period). If an extension is present then it will
17497 be preserved in the output. For example, when krunching @file{hellofile.ads}
17498 to eight characters, the result will be hellofil.ads.
17500 Note: for compatibility with previous versions of @code{gnatkr} dots may
17501 appear in the name instead of hyphens, but the last dot will always be
17502 taken as the start of an extension. So if @code{gnatkr} is given an argument
17503 such as @file{Hello.World.adb} it will be treated exactly as if the first
17504 period had been a hyphen, and for example krunching to eight characters
17505 gives the result @file{hellworl.adb}.
17507 Note that the result is always all lower case (except on OpenVMS where it is
17508 all upper case). Characters of the other case are folded as required.
17510 @var{length} represents the length of the krunched name. The default
17511 when no argument is given is ^8^39^ characters. A length of zero stands for
17512 unlimited, in other words do not chop except for system files where the
17513 implied crunching length is always eight characters.
17516 The output is the krunched name. The output has an extension only if the
17517 original argument was a file name with an extension.
17519 @node Krunching Method
17520 @section Krunching Method
17523 The initial file name is determined by the name of the unit that the file
17524 contains. The name is formed by taking the full expanded name of the
17525 unit and replacing the separating dots with hyphens and
17526 using ^lowercase^uppercase^
17527 for all letters, except that a hyphen in the second character position is
17528 replaced by a ^tilde^dollar sign^ if the first character is
17529 ^@samp{a}, @samp{i}, @samp{g}, or @samp{s}^@samp{A}, @samp{I}, @samp{G}, or @samp{S}^.
17530 The extension is @code{.ads} for a
17531 spec and @code{.adb} for a body.
17532 Krunching does not affect the extension, but the file name is shortened to
17533 the specified length by following these rules:
17537 The name is divided into segments separated by hyphens, tildes or
17538 underscores and all hyphens, tildes, and underscores are
17539 eliminated. If this leaves the name short enough, we are done.
17542 If the name is too long, the longest segment is located (left-most
17543 if there are two of equal length), and shortened by dropping
17544 its last character. This is repeated until the name is short enough.
17546 As an example, consider the krunching of @*@file{our-strings-wide_fixed.adb}
17547 to fit the name into 8 characters as required by some operating systems.
17550 our-strings-wide_fixed 22
17551 our strings wide fixed 19
17552 our string wide fixed 18
17553 our strin wide fixed 17
17554 our stri wide fixed 16
17555 our stri wide fixe 15
17556 our str wide fixe 14
17557 our str wid fixe 13
17563 Final file name: oustwifi.adb
17567 The file names for all predefined units are always krunched to eight
17568 characters. The krunching of these predefined units uses the following
17569 special prefix replacements:
17573 replaced by @file{^a^A^-}
17576 replaced by @file{^g^G^-}
17579 replaced by @file{^i^I^-}
17582 replaced by @file{^s^S^-}
17585 These system files have a hyphen in the second character position. That
17586 is why normal user files replace such a character with a
17587 ^tilde^dollar sign^, to
17588 avoid confusion with system file names.
17590 As an example of this special rule, consider
17591 @*@file{ada-strings-wide_fixed.adb}, which gets krunched as follows:
17594 ada-strings-wide_fixed 22
17595 a- strings wide fixed 18
17596 a- string wide fixed 17
17597 a- strin wide fixed 16
17598 a- stri wide fixed 15
17599 a- stri wide fixe 14
17600 a- str wide fixe 13
17606 Final file name: a-stwifi.adb
17610 Of course no file shortening algorithm can guarantee uniqueness over all
17611 possible unit names, and if file name krunching is used then it is your
17612 responsibility to ensure that no name clashes occur. The utility
17613 program @code{gnatkr} is supplied for conveniently determining the
17614 krunched name of a file.
17616 @node Examples of gnatkr Usage
17617 @section Examples of @code{gnatkr} Usage
17624 $ gnatkr very_long_unit_name.ads --> velounna.ads
17625 $ gnatkr grandparent-parent-child.ads --> grparchi.ads
17626 $ gnatkr Grandparent.Parent.Child.ads --> grparchi.ads
17627 $ gnatkr grandparent-parent-child --> grparchi
17629 $ gnatkr very_long_unit_name.ads/count=6 --> vlunna.ads
17630 $ gnatkr very_long_unit_name.ads/count=0 --> very_long_unit_name.ads
17633 @node Preprocessing Using gnatprep
17634 @chapter Preprocessing Using @code{gnatprep}
17638 This chapter discusses how to use GNAT's @code{gnatprep} utility for simple
17640 Although designed for use with GNAT, @code{gnatprep} does not depend on any
17641 special GNAT features.
17642 For further discussion of conditional compilation in general, see
17643 @ref{Conditional Compilation}.
17646 * Preprocessing Symbols::
17648 * Switches for gnatprep::
17649 * Form of Definitions File::
17650 * Form of Input Text for gnatprep::
17653 @node Preprocessing Symbols
17654 @section Preprocessing Symbols
17657 Preprocessing symbols are defined in definition files and referred to in
17658 sources to be preprocessed. A Preprocessing symbol is an identifier, following
17659 normal Ada (case-insensitive) rules for its syntax, with the restriction that
17660 all characters need to be in the ASCII set (no accented letters).
17662 @node Using gnatprep
17663 @section Using @code{gnatprep}
17666 To call @code{gnatprep} use
17669 $ gnatprep @ovar{switches} @var{infile} @var{outfile} @ovar{deffile}
17676 is an optional sequence of switches as described in the next section.
17679 is the full name of the input file, which is an Ada source
17680 file containing preprocessor directives.
17683 is the full name of the output file, which is an Ada source
17684 in standard Ada form. When used with GNAT, this file name will
17685 normally have an ads or adb suffix.
17688 is the full name of a text file containing definitions of
17689 preprocessing symbols to be referenced by the preprocessor. This argument is
17690 optional, and can be replaced by the use of the @option{-D} switch.
17694 @node Switches for gnatprep
17695 @section Switches for @code{gnatprep}
17700 @item ^-b^/BLANK_LINES^
17701 @cindex @option{^-b^/BLANK_LINES^} (@command{gnatprep})
17702 Causes both preprocessor lines and the lines deleted by
17703 preprocessing to be replaced by blank lines in the output source file,
17704 preserving line numbers in the output file.
17706 @item ^-c^/COMMENTS^
17707 @cindex @option{^-c^/COMMENTS^} (@command{gnatprep})
17708 Causes both preprocessor lines and the lines deleted
17709 by preprocessing to be retained in the output source as comments marked
17710 with the special string @code{"--! "}. This option will result in line numbers
17711 being preserved in the output file.
17713 @item ^-C^/REPLACE_IN_COMMENTS^
17714 @cindex @option{^-C^/REPLACE_IN_COMMENTS^} (@command{gnatprep})
17715 Causes comments to be scanned. Normally comments are ignored by gnatprep.
17716 If this option is specified, then comments are scanned and any $symbol
17717 substitutions performed as in program text. This is particularly useful
17718 when structured comments are used (e.g., when writing programs in the
17719 SPARK dialect of Ada). Note that this switch is not available when
17720 doing integrated preprocessing (it would be useless in this context
17721 since comments are ignored by the compiler in any case).
17723 @item ^-Dsymbol=value^/ASSOCIATE="symbol=value"^
17724 @cindex @option{^-D^/ASSOCIATE^} (@command{gnatprep})
17725 Defines a new preprocessing symbol, associated with value. If no value is given
17726 on the command line, then symbol is considered to be @code{True}. This switch
17727 can be used in place of a definition file.
17731 @cindex @option{/REMOVE} (@command{gnatprep})
17732 This is the default setting which causes lines deleted by preprocessing
17733 to be entirely removed from the output file.
17736 @item ^-r^/REFERENCE^
17737 @cindex @option{^-r^/REFERENCE^} (@command{gnatprep})
17738 Causes a @code{Source_Reference} pragma to be generated that
17739 references the original input file, so that error messages will use
17740 the file name of this original file. The use of this switch implies
17741 that preprocessor lines are not to be removed from the file, so its
17742 use will force @option{^-b^/BLANK_LINES^} mode if
17743 @option{^-c^/COMMENTS^}
17744 has not been specified explicitly.
17746 Note that if the file to be preprocessed contains multiple units, then
17747 it will be necessary to @code{gnatchop} the output file from
17748 @code{gnatprep}. If a @code{Source_Reference} pragma is present
17749 in the preprocessed file, it will be respected by
17750 @code{gnatchop ^-r^/REFERENCE^}
17751 so that the final chopped files will correctly refer to the original
17752 input source file for @code{gnatprep}.
17754 @item ^-s^/SYMBOLS^
17755 @cindex @option{^-s^/SYMBOLS^} (@command{gnatprep})
17756 Causes a sorted list of symbol names and values to be
17757 listed on the standard output file.
17759 @item ^-u^/UNDEFINED^
17760 @cindex @option{^-u^/UNDEFINED^} (@command{gnatprep})
17761 Causes undefined symbols to be treated as having the value FALSE in the context
17762 of a preprocessor test. In the absence of this option, an undefined symbol in
17763 a @code{#if} or @code{#elsif} test will be treated as an error.
17769 Note: if neither @option{-b} nor @option{-c} is present,
17770 then preprocessor lines and
17771 deleted lines are completely removed from the output, unless -r is
17772 specified, in which case -b is assumed.
17775 @node Form of Definitions File
17776 @section Form of Definitions File
17779 The definitions file contains lines of the form
17786 where symbol is a preprocessing symbol, and value is one of the following:
17790 Empty, corresponding to a null substitution
17792 A string literal using normal Ada syntax
17794 Any sequence of characters from the set
17795 (letters, digits, period, underline).
17799 Comment lines may also appear in the definitions file, starting with
17800 the usual @code{--},
17801 and comments may be added to the definitions lines.
17803 @node Form of Input Text for gnatprep
17804 @section Form of Input Text for @code{gnatprep}
17807 The input text may contain preprocessor conditional inclusion lines,
17808 as well as general symbol substitution sequences.
17810 The preprocessor conditional inclusion commands have the form
17815 #if @i{expression} @r{[}then@r{]}
17817 #elsif @i{expression} @r{[}then@r{]}
17819 #elsif @i{expression} @r{[}then@r{]}
17830 In this example, @i{expression} is defined by the following grammar:
17832 @i{expression} ::= <symbol>
17833 @i{expression} ::= <symbol> = "<value>"
17834 @i{expression} ::= <symbol> = <symbol>
17835 @i{expression} ::= <symbol> 'Defined
17836 @i{expression} ::= not @i{expression}
17837 @i{expression} ::= @i{expression} and @i{expression}
17838 @i{expression} ::= @i{expression} or @i{expression}
17839 @i{expression} ::= @i{expression} and then @i{expression}
17840 @i{expression} ::= @i{expression} or else @i{expression}
17841 @i{expression} ::= ( @i{expression} )
17844 The following restriction exists: it is not allowed to have "and" or "or"
17845 following "not" in the same expression without parentheses. For example, this
17852 This should be one of the following:
17860 For the first test (@i{expression} ::= <symbol>) the symbol must have
17861 either the value true or false, that is to say the right-hand of the
17862 symbol definition must be one of the (case-insensitive) literals
17863 @code{True} or @code{False}. If the value is true, then the
17864 corresponding lines are included, and if the value is false, they are
17867 The test (@i{expression} ::= <symbol> @code{'Defined}) is true only if
17868 the symbol has been defined in the definition file or by a @option{-D}
17869 switch on the command line. Otherwise, the test is false.
17871 The equality tests are case insensitive, as are all the preprocessor lines.
17873 If the symbol referenced is not defined in the symbol definitions file,
17874 then the effect depends on whether or not switch @option{-u}
17875 is specified. If so, then the symbol is treated as if it had the value
17876 false and the test fails. If this switch is not specified, then
17877 it is an error to reference an undefined symbol. It is also an error to
17878 reference a symbol that is defined with a value other than @code{True}
17881 The use of the @code{not} operator inverts the sense of this logical test.
17882 The @code{not} operator cannot be combined with the @code{or} or @code{and}
17883 operators, without parentheses. For example, "if not X or Y then" is not
17884 allowed, but "if (not X) or Y then" and "if not (X or Y) then" are.
17886 The @code{then} keyword is optional as shown
17888 The @code{#} must be the first non-blank character on a line, but
17889 otherwise the format is free form. Spaces or tabs may appear between
17890 the @code{#} and the keyword. The keywords and the symbols are case
17891 insensitive as in normal Ada code. Comments may be used on a
17892 preprocessor line, but other than that, no other tokens may appear on a
17893 preprocessor line. Any number of @code{elsif} clauses can be present,
17894 including none at all. The @code{else} is optional, as in Ada.
17896 The @code{#} marking the start of a preprocessor line must be the first
17897 non-blank character on the line, i.e., it must be preceded only by
17898 spaces or horizontal tabs.
17900 Symbol substitution outside of preprocessor lines is obtained by using
17908 anywhere within a source line, except in a comment or within a
17909 string literal. The identifier
17910 following the @code{$} must match one of the symbols defined in the symbol
17911 definition file, and the result is to substitute the value of the
17912 symbol in place of @code{$symbol} in the output file.
17914 Note that although the substitution of strings within a string literal
17915 is not possible, it is possible to have a symbol whose defined value is
17916 a string literal. So instead of setting XYZ to @code{hello} and writing:
17919 Header : String := "$XYZ";
17923 you should set XYZ to @code{"hello"} and write:
17926 Header : String := $XYZ;
17930 and then the substitution will occur as desired.
17933 @node The GNAT Run-Time Library Builder gnatlbr
17934 @chapter The GNAT Run-Time Library Builder @code{gnatlbr}
17936 @cindex Library builder
17939 @code{gnatlbr} is a tool for rebuilding the GNAT run time with user
17940 supplied configuration pragmas.
17943 * Running gnatlbr::
17944 * Switches for gnatlbr::
17945 * Examples of gnatlbr Usage::
17948 @node Running gnatlbr
17949 @section Running @code{gnatlbr}
17952 The @code{gnatlbr} command has the form
17955 $ GNAT LIBRARY /@r{[}CREATE@r{|}SET@r{|}DELETE@r{]}=directory @r{[}/CONFIG=file@r{]}
17958 @node Switches for gnatlbr
17959 @section Switches for @code{gnatlbr}
17962 @code{gnatlbr} recognizes the following switches:
17966 @item /CREATE=directory
17967 @cindex @code{/CREATE} (@code{gnatlbr})
17968 Create the new run-time library in the specified directory.
17970 @item /SET=directory
17971 @cindex @code{/SET} (@code{gnatlbr})
17972 Make the library in the specified directory the current run-time library.
17974 @item /DELETE=directory
17975 @cindex @code{/DELETE} (@code{gnatlbr})
17976 Delete the run-time library in the specified directory.
17979 @cindex @code{/CONFIG} (@code{gnatlbr})
17980 With /CREATE: Use the configuration pragmas in the specified file when
17981 building the library.
17983 With /SET: Use the configuration pragmas in the specified file when
17988 @node Examples of gnatlbr Usage
17989 @section Example of @code{gnatlbr} Usage
17992 Contents of VAXFLOAT.ADC:
17993 pragma Float_Representation (VAX_Float);
17995 $ GNAT LIBRARY /CREATE=[.VAXFLOAT] /CONFIG=VAXFLOAT.ADC
17997 GNAT LIBRARY rebuilds the run-time library in directory [.VAXFLOAT]
18002 @node The GNAT Library Browser gnatls
18003 @chapter The GNAT Library Browser @code{gnatls}
18005 @cindex Library browser
18008 @code{gnatls} is a tool that outputs information about compiled
18009 units. It gives the relationship between objects, unit names and source
18010 files. It can also be used to check the source dependencies of a unit
18011 as well as various characteristics.
18013 Note: to invoke @code{gnatls} with a project file, use the @code{gnat}
18014 driver (see @ref{The GNAT Driver and Project Files}).
18018 * Switches for gnatls::
18019 * Examples of gnatls Usage::
18022 @node Running gnatls
18023 @section Running @code{gnatls}
18026 The @code{gnatls} command has the form
18029 $ gnatls switches @var{object_or_ali_file}
18033 The main argument is the list of object or @file{ali} files
18034 (@pxref{The Ada Library Information Files})
18035 for which information is requested.
18037 In normal mode, without additional option, @code{gnatls} produces a
18038 four-column listing. Each line represents information for a specific
18039 object. The first column gives the full path of the object, the second
18040 column gives the name of the principal unit in this object, the third
18041 column gives the status of the source and the fourth column gives the
18042 full path of the source representing this unit.
18043 Here is a simple example of use:
18047 ^./^[]^demo1.o demo1 DIF demo1.adb
18048 ^./^[]^demo2.o demo2 OK demo2.adb
18049 ^./^[]^hello.o h1 OK hello.adb
18050 ^./^[]^instr-child.o instr.child MOK instr-child.adb
18051 ^./^[]^instr.o instr OK instr.adb
18052 ^./^[]^tef.o tef DIF tef.adb
18053 ^./^[]^text_io_example.o text_io_example OK text_io_example.adb
18054 ^./^[]^tgef.o tgef DIF tgef.adb
18058 The first line can be interpreted as follows: the main unit which is
18060 object file @file{demo1.o} is demo1, whose main source is in
18061 @file{demo1.adb}. Furthermore, the version of the source used for the
18062 compilation of demo1 has been modified (DIF). Each source file has a status
18063 qualifier which can be:
18066 @item OK (unchanged)
18067 The version of the source file used for the compilation of the
18068 specified unit corresponds exactly to the actual source file.
18070 @item MOK (slightly modified)
18071 The version of the source file used for the compilation of the
18072 specified unit differs from the actual source file but not enough to
18073 require recompilation. If you use gnatmake with the qualifier
18074 @option{^-m (minimal recompilation)^/MINIMAL_RECOMPILATION^}, a file marked
18075 MOK will not be recompiled.
18077 @item DIF (modified)
18078 No version of the source found on the path corresponds to the source
18079 used to build this object.
18081 @item ??? (file not found)
18082 No source file was found for this unit.
18084 @item HID (hidden, unchanged version not first on PATH)
18085 The version of the source that corresponds exactly to the source used
18086 for compilation has been found on the path but it is hidden by another
18087 version of the same source that has been modified.
18091 @node Switches for gnatls
18092 @section Switches for @code{gnatls}
18095 @code{gnatls} recognizes the following switches:
18099 @cindex @option{--version} @command{gnatls}
18100 Display Copyright and version, then exit disregarding all other options.
18103 @cindex @option{--help} @command{gnatls}
18104 If @option{--version} was not used, display usage, then exit disregarding
18107 @item ^-a^/ALL_UNITS^
18108 @cindex @option{^-a^/ALL_UNITS^} (@code{gnatls})
18109 Consider all units, including those of the predefined Ada library.
18110 Especially useful with @option{^-d^/DEPENDENCIES^}.
18112 @item ^-d^/DEPENDENCIES^
18113 @cindex @option{^-d^/DEPENDENCIES^} (@code{gnatls})
18114 List sources from which specified units depend on.
18116 @item ^-h^/OUTPUT=OPTIONS^
18117 @cindex @option{^-h^/OUTPUT=OPTIONS^} (@code{gnatls})
18118 Output the list of options.
18120 @item ^-o^/OUTPUT=OBJECTS^
18121 @cindex @option{^-o^/OUTPUT=OBJECTS^} (@code{gnatls})
18122 Only output information about object files.
18124 @item ^-s^/OUTPUT=SOURCES^
18125 @cindex @option{^-s^/OUTPUT=SOURCES^} (@code{gnatls})
18126 Only output information about source files.
18128 @item ^-u^/OUTPUT=UNITS^
18129 @cindex @option{^-u^/OUTPUT=UNITS^} (@code{gnatls})
18130 Only output information about compilation units.
18132 @item ^-files^/FILES^=@var{file}
18133 @cindex @option{^-files^/FILES^} (@code{gnatls})
18134 Take as arguments the files listed in text file @var{file}.
18135 Text file @var{file} may contain empty lines that are ignored.
18136 Each nonempty line should contain the name of an existing file.
18137 Several such switches may be specified simultaneously.
18139 @item ^-aO^/OBJECT_SEARCH=^@var{dir}
18140 @itemx ^-aI^/SOURCE_SEARCH=^@var{dir}
18141 @itemx ^-I^/SEARCH=^@var{dir}
18142 @itemx ^-I-^/NOCURRENT_DIRECTORY^
18144 @cindex @option{^-aO^/OBJECT_SEARCH^} (@code{gnatls})
18145 @cindex @option{^-aI^/SOURCE_SEARCH^} (@code{gnatls})
18146 @cindex @option{^-I^/SEARCH^} (@code{gnatls})
18147 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@code{gnatls})
18148 Source path manipulation. Same meaning as the equivalent @command{gnatmake}
18149 flags (@pxref{Switches for gnatmake}).
18151 @item --RTS=@var{rts-path}
18152 @cindex @option{--RTS} (@code{gnatls})
18153 Specifies the default location of the runtime library. Same meaning as the
18154 equivalent @command{gnatmake} flag (@pxref{Switches for gnatmake}).
18156 @item ^-v^/OUTPUT=VERBOSE^
18157 @cindex @option{^-v^/OUTPUT=VERBOSE^} (@code{gnatls})
18158 Verbose mode. Output the complete source, object and project paths. Do not use
18159 the default column layout but instead use long format giving as much as
18160 information possible on each requested units, including special
18161 characteristics such as:
18164 @item Preelaborable
18165 The unit is preelaborable in the Ada sense.
18168 No elaboration code has been produced by the compiler for this unit.
18171 The unit is pure in the Ada sense.
18173 @item Elaborate_Body
18174 The unit contains a pragma Elaborate_Body.
18177 The unit contains a pragma Remote_Types.
18179 @item Shared_Passive
18180 The unit contains a pragma Shared_Passive.
18183 This unit is part of the predefined environment and cannot be modified
18186 @item Remote_Call_Interface
18187 The unit contains a pragma Remote_Call_Interface.
18193 @node Examples of gnatls Usage
18194 @section Example of @code{gnatls} Usage
18198 Example of using the verbose switch. Note how the source and
18199 object paths are affected by the -I switch.
18202 $ gnatls -v -I.. demo1.o
18204 GNATLS 5.03w (20041123-34)
18205 Copyright 1997-2004 Free Software Foundation, Inc.
18207 Source Search Path:
18208 <Current_Directory>
18210 /home/comar/local/adainclude/
18212 Object Search Path:
18213 <Current_Directory>
18215 /home/comar/local/lib/gcc-lib/x86-linux/3.4.3/adalib/
18217 Project Search Path:
18218 <Current_Directory>
18219 /home/comar/local/lib/gnat/
18224 Kind => subprogram body
18225 Flags => No_Elab_Code
18226 Source => demo1.adb modified
18230 The following is an example of use of the dependency list.
18231 Note the use of the -s switch
18232 which gives a straight list of source files. This can be useful for
18233 building specialized scripts.
18236 $ gnatls -d demo2.o
18237 ./demo2.o demo2 OK demo2.adb
18243 $ gnatls -d -s -a demo1.o
18245 /home/comar/local/adainclude/ada.ads
18246 /home/comar/local/adainclude/a-finali.ads
18247 /home/comar/local/adainclude/a-filico.ads
18248 /home/comar/local/adainclude/a-stream.ads
18249 /home/comar/local/adainclude/a-tags.ads
18252 /home/comar/local/adainclude/gnat.ads
18253 /home/comar/local/adainclude/g-io.ads
18255 /home/comar/local/adainclude/system.ads
18256 /home/comar/local/adainclude/s-exctab.ads
18257 /home/comar/local/adainclude/s-finimp.ads
18258 /home/comar/local/adainclude/s-finroo.ads
18259 /home/comar/local/adainclude/s-secsta.ads
18260 /home/comar/local/adainclude/s-stalib.ads
18261 /home/comar/local/adainclude/s-stoele.ads
18262 /home/comar/local/adainclude/s-stratt.ads
18263 /home/comar/local/adainclude/s-tasoli.ads
18264 /home/comar/local/adainclude/s-unstyp.ads
18265 /home/comar/local/adainclude/unchconv.ads
18271 GNAT LIST /DEPENDENCIES /OUTPUT=SOURCES /ALL_UNITS DEMO1.ADB
18273 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]ada.ads
18274 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-finali.ads
18275 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-filico.ads
18276 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-stream.ads
18277 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]a-tags.ads
18281 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]gnat.ads
18282 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]g-io.ads
18284 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]system.ads
18285 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-exctab.ads
18286 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-finimp.ads
18287 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-finroo.ads
18288 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-secsta.ads
18289 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-stalib.ads
18290 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-stoele.ads
18291 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-stratt.ads
18292 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-tasoli.ads
18293 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]s-unstyp.ads
18294 GNU:[LIB.OPENVMS7_1.2_8_1.ADALIB]unchconv.ads
18298 @node Cleaning Up Using gnatclean
18299 @chapter Cleaning Up Using @code{gnatclean}
18301 @cindex Cleaning tool
18304 @code{gnatclean} is a tool that allows the deletion of files produced by the
18305 compiler, binder and linker, including ALI files, object files, tree files,
18306 expanded source files, library files, interface copy source files, binder
18307 generated files and executable files.
18310 * Running gnatclean::
18311 * Switches for gnatclean::
18312 @c * Examples of gnatclean Usage::
18315 @node Running gnatclean
18316 @section Running @code{gnatclean}
18319 The @code{gnatclean} command has the form:
18322 $ gnatclean switches @var{names}
18326 @var{names} is a list of source file names. Suffixes @code{.^ads^ADS^} and
18327 @code{^adb^ADB^} may be omitted. If a project file is specified using switch
18328 @code{^-P^/PROJECT_FILE=^}, then @var{names} may be completely omitted.
18331 In normal mode, @code{gnatclean} delete the files produced by the compiler and,
18332 if switch @code{^-c^/COMPILER_FILES_ONLY^} is not specified, by the binder and
18333 the linker. In informative-only mode, specified by switch
18334 @code{^-n^/NODELETE^}, the list of files that would have been deleted in
18335 normal mode is listed, but no file is actually deleted.
18337 @node Switches for gnatclean
18338 @section Switches for @code{gnatclean}
18341 @code{gnatclean} recognizes the following switches:
18345 @cindex @option{--version} @command{gnatclean}
18346 Display Copyright and version, then exit disregarding all other options.
18349 @cindex @option{--help} @command{gnatclean}
18350 If @option{--version} was not used, display usage, then exit disregarding
18353 @item ^-c^/COMPILER_FILES_ONLY^
18354 @cindex @option{^-c^/COMPILER_FILES_ONLY^} (@code{gnatclean})
18355 Only attempt to delete the files produced by the compiler, not those produced
18356 by the binder or the linker. The files that are not to be deleted are library
18357 files, interface copy files, binder generated files and executable files.
18359 @item ^-D ^/DIRECTORY_OBJECTS=^@var{dir}
18360 @cindex @option{^-D^/DIRECTORY_OBJECTS^} (@code{gnatclean})
18361 Indicate that ALI and object files should normally be found in directory
18364 @item ^-F^/FULL_PATH_IN_BRIEF_MESSAGES^
18365 @cindex @option{^-F^/FULL_PATH_IN_BRIEF_MESSAGES^} (@code{gnatclean})
18366 When using project files, if some errors or warnings are detected during
18367 parsing and verbose mode is not in effect (no use of switch
18368 ^-v^/VERBOSE^), then error lines start with the full path name of the project
18369 file, rather than its simple file name.
18372 @cindex @option{^-h^/HELP^} (@code{gnatclean})
18373 Output a message explaining the usage of @code{^gnatclean^gnatclean^}.
18375 @item ^-n^/NODELETE^
18376 @cindex @option{^-n^/NODELETE^} (@code{gnatclean})
18377 Informative-only mode. Do not delete any files. Output the list of the files
18378 that would have been deleted if this switch was not specified.
18380 @item ^-P^/PROJECT_FILE=^@var{project}
18381 @cindex @option{^-P^/PROJECT_FILE^} (@code{gnatclean})
18382 Use project file @var{project}. Only one such switch can be used.
18383 When cleaning a project file, the files produced by the compilation of the
18384 immediate sources or inherited sources of the project files are to be
18385 deleted. This is not depending on the presence or not of executable names
18386 on the command line.
18389 @cindex @option{^-q^/QUIET^} (@code{gnatclean})
18390 Quiet output. If there are no errors, do not output anything, except in
18391 verbose mode (switch ^-v^/VERBOSE^) or in informative-only mode
18392 (switch ^-n^/NODELETE^).
18394 @item ^-r^/RECURSIVE^
18395 @cindex @option{^-r^/RECURSIVE^} (@code{gnatclean})
18396 When a project file is specified (using switch ^-P^/PROJECT_FILE=^),
18397 clean all imported and extended project files, recursively. If this switch
18398 is not specified, only the files related to the main project file are to be
18399 deleted. This switch has no effect if no project file is specified.
18401 @item ^-v^/VERBOSE^
18402 @cindex @option{^-v^/VERBOSE^} (@code{gnatclean})
18405 @item ^-vP^/MESSAGES_PROJECT_FILE=^@emph{x}
18406 @cindex @option{^-vP^/MESSAGES_PROJECT_FILE^} (@code{gnatclean})
18407 Indicates the verbosity of the parsing of GNAT project files.
18408 @xref{Switches Related to Project Files}.
18410 @item ^-X^/EXTERNAL_REFERENCE=^@var{name=value}
18411 @cindex @option{^-X^/EXTERNAL_REFERENCE^} (@code{gnatclean})
18412 Indicates that external variable @var{name} has the value @var{value}.
18413 The Project Manager will use this value for occurrences of
18414 @code{external(name)} when parsing the project file.
18415 @xref{Switches Related to Project Files}.
18417 @item ^-aO^/OBJECT_SEARCH=^@var{dir}
18418 @cindex @option{^-aO^/OBJECT_SEARCH^} (@code{gnatclean})
18419 When searching for ALI and object files, look in directory
18422 @item ^-I^/SEARCH=^@var{dir}
18423 @cindex @option{^-I^/SEARCH^} (@code{gnatclean})
18424 Equivalent to @option{^-aO^/OBJECT_SEARCH=^@var{dir}}.
18426 @item ^-I-^/NOCURRENT_DIRECTORY^
18427 @cindex @option{^-I-^/NOCURRENT_DIRECTORY^} (@code{gnatclean})
18428 @cindex Source files, suppressing search
18429 Do not look for ALI or object files in the directory
18430 where @code{gnatclean} was invoked.
18434 @c @node Examples of gnatclean Usage
18435 @c @section Examples of @code{gnatclean} Usage
18438 @node GNAT and Libraries
18439 @chapter GNAT and Libraries
18440 @cindex Library, building, installing, using
18443 This chapter describes how to build and use libraries with GNAT, and also shows
18444 how to recompile the GNAT run-time library. You should be familiar with the
18445 Project Manager facility (@pxref{GNAT Project Manager}) before reading this
18449 * Introduction to Libraries in GNAT::
18450 * General Ada Libraries::
18451 * Stand-alone Ada Libraries::
18452 * Rebuilding the GNAT Run-Time Library::
18455 @node Introduction to Libraries in GNAT
18456 @section Introduction to Libraries in GNAT
18459 A library is, conceptually, a collection of objects which does not have its
18460 own main thread of execution, but rather provides certain services to the
18461 applications that use it. A library can be either statically linked with the
18462 application, in which case its code is directly included in the application,
18463 or, on platforms that support it, be dynamically linked, in which case
18464 its code is shared by all applications making use of this library.
18466 GNAT supports both types of libraries.
18467 In the static case, the compiled code can be provided in different ways. The
18468 simplest approach is to provide directly the set of objects resulting from
18469 compilation of the library source files. Alternatively, you can group the
18470 objects into an archive using whatever commands are provided by the operating
18471 system. For the latter case, the objects are grouped into a shared library.
18473 In the GNAT environment, a library has three types of components:
18479 @xref{The Ada Library Information Files}.
18481 Object files, an archive or a shared library.
18485 A GNAT library may expose all its source files, which is useful for
18486 documentation purposes. Alternatively, it may expose only the units needed by
18487 an external user to make use of the library. That is to say, the specs
18488 reflecting the library services along with all the units needed to compile
18489 those specs, which can include generic bodies or any body implementing an
18490 inlined routine. In the case of @emph{stand-alone libraries} those exposed
18491 units are called @emph{interface units} (@pxref{Stand-alone Ada Libraries}).
18493 All compilation units comprising an application, including those in a library,
18494 need to be elaborated in an order partially defined by Ada's semantics. GNAT
18495 computes the elaboration order from the @file{ALI} files and this is why they
18496 constitute a mandatory part of GNAT libraries. Except in the case of
18497 @emph{stand-alone libraries}, where a specific library elaboration routine is
18498 produced independently of the application(s) using the library.
18500 @node General Ada Libraries
18501 @section General Ada Libraries
18504 * Building a library::
18505 * Installing a library::
18506 * Using a library::
18509 @node Building a library
18510 @subsection Building a library
18513 The easiest way to build a library is to use the Project Manager,
18514 which supports a special type of project called a @emph{Library Project}
18515 (@pxref{Library Projects}).
18517 A project is considered a library project, when two project-level attributes
18518 are defined in it: @code{Library_Name} and @code{Library_Dir}. In order to
18519 control different aspects of library configuration, additional optional
18520 project-level attributes can be specified:
18523 This attribute controls whether the library is to be static or dynamic
18525 @item Library_Version
18526 This attribute specifies the library version; this value is used
18527 during dynamic linking of shared libraries to determine if the currently
18528 installed versions of the binaries are compatible.
18530 @item Library_Options
18532 These attributes specify additional low-level options to be used during
18533 library generation, and redefine the actual application used to generate
18538 The GNAT Project Manager takes full care of the library maintenance task,
18539 including recompilation of the source files for which objects do not exist
18540 or are not up to date, assembly of the library archive, and installation of
18541 the library (i.e., copying associated source, object and @file{ALI} files
18542 to the specified location).
18544 Here is a simple library project file:
18545 @smallexample @c ada
18547 for Source_Dirs use ("src1", "src2");
18548 for Object_Dir use "obj";
18549 for Library_Name use "mylib";
18550 for Library_Dir use "lib";
18551 for Library_Kind use "dynamic";
18556 and the compilation command to build and install the library:
18558 @smallexample @c ada
18559 $ gnatmake -Pmy_lib
18563 It is not entirely trivial to perform manually all the steps required to
18564 produce a library. We recommend that you use the GNAT Project Manager
18565 for this task. In special cases where this is not desired, the necessary
18566 steps are discussed below.
18568 There are various possibilities for compiling the units that make up the
18569 library: for example with a Makefile (@pxref{Using the GNU make Utility}) or
18570 with a conventional script. For simple libraries, it is also possible to create
18571 a dummy main program which depends upon all the packages that comprise the
18572 interface of the library. This dummy main program can then be given to
18573 @command{gnatmake}, which will ensure that all necessary objects are built.
18575 After this task is accomplished, you should follow the standard procedure
18576 of the underlying operating system to produce the static or shared library.
18578 Here is an example of such a dummy program:
18579 @smallexample @c ada
18581 with My_Lib.Service1;
18582 with My_Lib.Service2;
18583 with My_Lib.Service3;
18584 procedure My_Lib_Dummy is
18592 Here are the generic commands that will build an archive or a shared library.
18595 # compiling the library
18596 $ gnatmake -c my_lib_dummy.adb
18598 # we don't need the dummy object itself
18599 $ rm my_lib_dummy.o my_lib_dummy.ali
18601 # create an archive with the remaining objects
18602 $ ar rc libmy_lib.a *.o
18603 # some systems may require "ranlib" to be run as well
18605 # or create a shared library
18606 $ gcc -shared -o libmy_lib.so *.o
18607 # some systems may require the code to have been compiled with -fPIC
18609 # remove the object files that are now in the library
18612 # Make the ALI files read-only so that gnatmake will not try to
18613 # regenerate the objects that are in the library
18618 Please note that the library must have a name of the form @file{lib@var{xxx}.a}
18619 or @file{lib@var{xxx}.so} (or @file{lib@var{xxx}.dll} on Windows) in order to
18620 be accessed by the directive @option{-l@var{xxx}} at link time.
18622 @node Installing a library
18623 @subsection Installing a library
18624 @cindex @code{ADA_PROJECT_PATH}
18627 If you use project files, library installation is part of the library build
18628 process. Thus no further action is needed in order to make use of the
18629 libraries that are built as part of the general application build. A usable
18630 version of the library is installed in the directory specified by the
18631 @code{Library_Dir} attribute of the library project file.
18633 You may want to install a library in a context different from where the library
18634 is built. This situation arises with third party suppliers, who may want
18635 to distribute a library in binary form where the user is not expected to be
18636 able to recompile the library. The simplest option in this case is to provide
18637 a project file slightly different from the one used to build the library, by
18638 using the @code{externally_built} attribute. For instance, the project
18639 file used to build the library in the previous section can be changed into the
18640 following one when the library is installed:
18642 @smallexample @c projectfile
18644 for Source_Dirs use ("src1", "src2");
18645 for Library_Name use "mylib";
18646 for Library_Dir use "lib";
18647 for Library_Kind use "dynamic";
18648 for Externally_Built use "true";
18653 This project file assumes that the directories @file{src1},
18654 @file{src2}, and @file{lib} exist in
18655 the directory containing the project file. The @code{externally_built}
18656 attribute makes it clear to the GNAT builder that it should not attempt to
18657 recompile any of the units from this library. It allows the library provider to
18658 restrict the source set to the minimum necessary for clients to make use of the
18659 library as described in the first section of this chapter. It is the
18660 responsibility of the library provider to install the necessary sources, ALI
18661 files and libraries in the directories mentioned in the project file. For
18662 convenience, the user's library project file should be installed in a location
18663 that will be searched automatically by the GNAT
18664 builder. These are the directories referenced in the @env{ADA_PROJECT_PATH}
18665 environment variable (@pxref{Importing Projects}), and also the default GNAT
18666 library location that can be queried with @command{gnatls -v} and is usually of
18667 the form $gnat_install_root/lib/gnat.
18669 When project files are not an option, it is also possible, but not recommended,
18670 to install the library so that the sources needed to use the library are on the
18671 Ada source path and the ALI files & libraries be on the Ada Object path (see
18672 @ref{Search Paths and the Run-Time Library (RTL)}. Alternatively, the system
18673 administrator can place general-purpose libraries in the default compiler
18674 paths, by specifying the libraries' location in the configuration files
18675 @file{ada_source_path} and @file{ada_object_path}. These configuration files
18676 must be located in the GNAT installation tree at the same place as the gcc spec
18677 file. The location of the gcc spec file can be determined as follows:
18683 The configuration files mentioned above have a simple format: each line
18684 must contain one unique directory name.
18685 Those names are added to the corresponding path
18686 in their order of appearance in the file. The names can be either absolute
18687 or relative; in the latter case, they are relative to where theses files
18690 The files @file{ada_source_path} and @file{ada_object_path} might not be
18692 GNAT installation, in which case, GNAT will look for its run-time library in
18693 the directories @file{adainclude} (for the sources) and @file{adalib} (for the
18694 objects and @file{ALI} files). When the files exist, the compiler does not
18695 look in @file{adainclude} and @file{adalib}, and thus the
18696 @file{ada_source_path} file
18697 must contain the location for the GNAT run-time sources (which can simply
18698 be @file{adainclude}). In the same way, the @file{ada_object_path} file must
18699 contain the location for the GNAT run-time objects (which can simply
18702 You can also specify a new default path to the run-time library at compilation
18703 time with the switch @option{--RTS=rts-path}. You can thus choose / change
18704 the run-time library you want your program to be compiled with. This switch is
18705 recognized by @command{gcc}, @command{gnatmake}, @command{gnatbind},
18706 @command{gnatls}, @command{gnatfind} and @command{gnatxref}.
18708 It is possible to install a library before or after the standard GNAT
18709 library, by reordering the lines in the configuration files. In general, a
18710 library must be installed before the GNAT library if it redefines
18713 @node Using a library
18714 @subsection Using a library
18716 @noindent Once again, the project facility greatly simplifies the use of
18717 libraries. In this context, using a library is just a matter of adding a
18718 @code{with} clause in the user project. For instance, to make use of the
18719 library @code{My_Lib} shown in examples in earlier sections, you can
18722 @smallexample @c projectfile
18729 Even if you have a third-party, non-Ada library, you can still use GNAT's
18730 Project Manager facility to provide a wrapper for it. For example, the
18731 following project, when @code{with}ed by your main project, will link with the
18732 third-party library @file{liba.a}:
18734 @smallexample @c projectfile
18737 for Externally_Built use "true";
18738 for Source_Files use ();
18739 for Library_Dir use "lib";
18740 for Library_Name use "a";
18741 for Library_Kind use "static";
18745 This is an alternative to the use of @code{pragma Linker_Options}. It is
18746 especially interesting in the context of systems with several interdependent
18747 static libraries where finding a proper linker order is not easy and best be
18748 left to the tools having visibility over project dependence information.
18751 In order to use an Ada library manually, you need to make sure that this
18752 library is on both your source and object path
18753 (see @ref{Search Paths and the Run-Time Library (RTL)}
18754 and @ref{Search Paths for gnatbind}). Furthermore, when the objects are grouped
18755 in an archive or a shared library, you need to specify the desired
18756 library at link time.
18758 For example, you can use the library @file{mylib} installed in
18759 @file{/dir/my_lib_src} and @file{/dir/my_lib_obj} with the following commands:
18762 $ gnatmake -aI/dir/my_lib_src -aO/dir/my_lib_obj my_appl \
18767 This can be expressed more simply:
18772 when the following conditions are met:
18775 @file{/dir/my_lib_src} has been added by the user to the environment
18776 variable @env{ADA_INCLUDE_PATH}, or by the administrator to the file
18777 @file{ada_source_path}
18779 @file{/dir/my_lib_obj} has been added by the user to the environment
18780 variable @env{ADA_OBJECTS_PATH}, or by the administrator to the file
18781 @file{ada_object_path}
18783 a pragma @code{Linker_Options} has been added to one of the sources.
18786 @smallexample @c ada
18787 pragma Linker_Options ("-lmy_lib");
18791 @node Stand-alone Ada Libraries
18792 @section Stand-alone Ada Libraries
18793 @cindex Stand-alone library, building, using
18796 * Introduction to Stand-alone Libraries::
18797 * Building a Stand-alone Library::
18798 * Creating a Stand-alone Library to be used in a non-Ada context::
18799 * Restrictions in Stand-alone Libraries::
18802 @node Introduction to Stand-alone Libraries
18803 @subsection Introduction to Stand-alone Libraries
18806 A Stand-alone Library (abbreviated ``SAL'') is a library that contains the
18808 elaborate the Ada units that are included in the library. In contrast with
18809 an ordinary library, which consists of all sources, objects and @file{ALI}
18811 library, a SAL may specify a restricted subset of compilation units
18812 to serve as a library interface. In this case, the fully
18813 self-sufficient set of files will normally consist of an objects
18814 archive, the sources of interface units' specs, and the @file{ALI}
18815 files of interface units.
18816 If an interface spec contains a generic unit or an inlined subprogram,
18818 source must also be provided; if the units that must be provided in the source
18819 form depend on other units, the source and @file{ALI} files of those must
18822 The main purpose of a SAL is to minimize the recompilation overhead of client
18823 applications when a new version of the library is installed. Specifically,
18824 if the interface sources have not changed, client applications do not need to
18825 be recompiled. If, furthermore, a SAL is provided in the shared form and its
18826 version, controlled by @code{Library_Version} attribute, is not changed,
18827 then the clients do not need to be relinked.
18829 SALs also allow the library providers to minimize the amount of library source
18830 text exposed to the clients. Such ``information hiding'' might be useful or
18831 necessary for various reasons.
18833 Stand-alone libraries are also well suited to be used in an executable whose
18834 main routine is not written in Ada.
18836 @node Building a Stand-alone Library
18837 @subsection Building a Stand-alone Library
18840 GNAT's Project facility provides a simple way of building and installing
18841 stand-alone libraries; see @ref{Stand-alone Library Projects}.
18842 To be a Stand-alone Library Project, in addition to the two attributes
18843 that make a project a Library Project (@code{Library_Name} and
18844 @code{Library_Dir}; see @ref{Library Projects}), the attribute
18845 @code{Library_Interface} must be defined. For example:
18847 @smallexample @c projectfile
18849 for Library_Dir use "lib_dir";
18850 for Library_Name use "dummy";
18851 for Library_Interface use ("int1", "int1.child");
18856 Attribute @code{Library_Interface} has a non-empty string list value,
18857 each string in the list designating a unit contained in an immediate source
18858 of the project file.
18860 When a Stand-alone Library is built, first the binder is invoked to build
18861 a package whose name depends on the library name
18862 (@file{^b~dummy.ads/b^B$DUMMY.ADS/B^} in the example above).
18863 This binder-generated package includes initialization and
18864 finalization procedures whose
18865 names depend on the library name (@code{dummyinit} and @code{dummyfinal}
18867 above). The object corresponding to this package is included in the library.
18869 You must ensure timely (e.g., prior to any use of interfaces in the SAL)
18870 calling of these procedures if a static SAL is built, or if a shared SAL
18872 with the project-level attribute @code{Library_Auto_Init} set to
18875 For a Stand-Alone Library, only the @file{ALI} files of the Interface Units
18876 (those that are listed in attribute @code{Library_Interface}) are copied to
18877 the Library Directory. As a consequence, only the Interface Units may be
18878 imported from Ada units outside of the library. If other units are imported,
18879 the binding phase will fail.
18881 The attribute @code{Library_Src_Dir} may be specified for a
18882 Stand-Alone Library. @code{Library_Src_Dir} is a simple attribute that has a
18883 single string value. Its value must be the path (absolute or relative to the
18884 project directory) of an existing directory. This directory cannot be the
18885 object directory or one of the source directories, but it can be the same as
18886 the library directory. The sources of the Interface
18887 Units of the library that are needed by an Ada client of the library will be
18888 copied to the designated directory, called the Interface Copy directory.
18889 These sources include the specs of the Interface Units, but they may also
18890 include bodies and subunits, when pragmas @code{Inline} or @code{Inline_Always}
18891 are used, or when there is a generic unit in the spec. Before the sources
18892 are copied to the Interface Copy directory, an attempt is made to delete all
18893 files in the Interface Copy directory.
18895 Building stand-alone libraries by hand is somewhat tedious, but for those
18896 occasions when it is necessary here are the steps that you need to perform:
18899 Compile all library sources.
18902 Invoke the binder with the switch @option{-n} (No Ada main program),
18903 with all the @file{ALI} files of the interfaces, and
18904 with the switch @option{-L} to give specific names to the @code{init}
18905 and @code{final} procedures. For example:
18907 gnatbind -n int1.ali int2.ali -Lsal1
18911 Compile the binder generated file:
18917 Link the dynamic library with all the necessary object files,
18918 indicating to the linker the names of the @code{init} (and possibly
18919 @code{final}) procedures for automatic initialization (and finalization).
18920 The built library should be placed in a directory different from
18921 the object directory.
18924 Copy the @code{ALI} files of the interface to the library directory,
18925 add in this copy an indication that it is an interface to a SAL
18926 (i.e., add a word @option{SL} on the line in the @file{ALI} file that starts
18927 with letter ``P'') and make the modified copy of the @file{ALI} file
18932 Using SALs is not different from using other libraries
18933 (see @ref{Using a library}).
18935 @node Creating a Stand-alone Library to be used in a non-Ada context
18936 @subsection Creating a Stand-alone Library to be used in a non-Ada context
18939 It is easy to adapt the SAL build procedure discussed above for use of a SAL in
18942 The only extra step required is to ensure that library interface subprograms
18943 are compatible with the main program, by means of @code{pragma Export}
18944 or @code{pragma Convention}.
18946 Here is an example of simple library interface for use with C main program:
18948 @smallexample @c ada
18949 package Interface is
18951 procedure Do_Something;
18952 pragma Export (C, Do_Something, "do_something");
18954 procedure Do_Something_Else;
18955 pragma Export (C, Do_Something_Else, "do_something_else");
18961 On the foreign language side, you must provide a ``foreign'' view of the
18962 library interface; remember that it should contain elaboration routines in
18963 addition to interface subprograms.
18965 The example below shows the content of @code{mylib_interface.h} (note
18966 that there is no rule for the naming of this file, any name can be used)
18968 /* the library elaboration procedure */
18969 extern void mylibinit (void);
18971 /* the library finalization procedure */
18972 extern void mylibfinal (void);
18974 /* the interface exported by the library */
18975 extern void do_something (void);
18976 extern void do_something_else (void);
18980 Libraries built as explained above can be used from any program, provided
18981 that the elaboration procedures (named @code{mylibinit} in the previous
18982 example) are called before the library services are used. Any number of
18983 libraries can be used simultaneously, as long as the elaboration
18984 procedure of each library is called.
18986 Below is an example of a C program that uses the @code{mylib} library.
18989 #include "mylib_interface.h"
18994 /* First, elaborate the library before using it */
18997 /* Main program, using the library exported entities */
18999 do_something_else ();
19001 /* Library finalization at the end of the program */
19008 Note that invoking any library finalization procedure generated by
19009 @code{gnatbind} shuts down the Ada run-time environment.
19011 finalization of all Ada libraries must be performed at the end of the program.
19012 No call to these libraries or to the Ada run-time library should be made
19013 after the finalization phase.
19015 @node Restrictions in Stand-alone Libraries
19016 @subsection Restrictions in Stand-alone Libraries
19019 The pragmas listed below should be used with caution inside libraries,
19020 as they can create incompatibilities with other Ada libraries:
19022 @item pragma @code{Locking_Policy}
19023 @item pragma @code{Queuing_Policy}
19024 @item pragma @code{Task_Dispatching_Policy}
19025 @item pragma @code{Unreserve_All_Interrupts}
19029 When using a library that contains such pragmas, the user must make sure
19030 that all libraries use the same pragmas with the same values. Otherwise,
19031 @code{Program_Error} will
19032 be raised during the elaboration of the conflicting
19033 libraries. The usage of these pragmas and its consequences for the user
19034 should therefore be well documented.
19036 Similarly, the traceback in the exception occurrence mechanism should be
19037 enabled or disabled in a consistent manner across all libraries.
19038 Otherwise, Program_Error will be raised during the elaboration of the
19039 conflicting libraries.
19041 If the @code{Version} or @code{Body_Version}
19042 attributes are used inside a library, then you need to
19043 perform a @code{gnatbind} step that specifies all @file{ALI} files in all
19044 libraries, so that version identifiers can be properly computed.
19045 In practice these attributes are rarely used, so this is unlikely
19046 to be a consideration.
19048 @node Rebuilding the GNAT Run-Time Library
19049 @section Rebuilding the GNAT Run-Time Library
19050 @cindex GNAT Run-Time Library, rebuilding
19051 @cindex Building the GNAT Run-Time Library
19052 @cindex Rebuilding the GNAT Run-Time Library
19053 @cindex Run-Time Library, rebuilding
19056 It may be useful to recompile the GNAT library in various contexts, the
19057 most important one being the use of partition-wide configuration pragmas
19058 such as @code{Normalize_Scalars}. A special Makefile called
19059 @code{Makefile.adalib} is provided to that effect and can be found in
19060 the directory containing the GNAT library. The location of this
19061 directory depends on the way the GNAT environment has been installed and can
19062 be determined by means of the command:
19069 The last entry in the object search path usually contains the
19070 gnat library. This Makefile contains its own documentation and in
19071 particular the set of instructions needed to rebuild a new library and
19074 @node Using the GNU make Utility
19075 @chapter Using the GNU @code{make} Utility
19079 This chapter offers some examples of makefiles that solve specific
19080 problems. It does not explain how to write a makefile (@pxref{Top,, GNU
19081 make, make, GNU @code{make}}), nor does it try to replace the
19082 @command{gnatmake} utility (@pxref{The GNAT Make Program gnatmake}).
19084 All the examples in this section are specific to the GNU version of
19085 make. Although @command{make} is a standard utility, and the basic language
19086 is the same, these examples use some advanced features found only in
19090 * Using gnatmake in a Makefile::
19091 * Automatically Creating a List of Directories::
19092 * Generating the Command Line Switches::
19093 * Overcoming Command Line Length Limits::
19096 @node Using gnatmake in a Makefile
19097 @section Using gnatmake in a Makefile
19102 Complex project organizations can be handled in a very powerful way by
19103 using GNU make combined with gnatmake. For instance, here is a Makefile
19104 which allows you to build each subsystem of a big project into a separate
19105 shared library. Such a makefile allows you to significantly reduce the link
19106 time of very big applications while maintaining full coherence at
19107 each step of the build process.
19109 The list of dependencies are handled automatically by
19110 @command{gnatmake}. The Makefile is simply used to call gnatmake in each of
19111 the appropriate directories.
19113 Note that you should also read the example on how to automatically
19114 create the list of directories
19115 (@pxref{Automatically Creating a List of Directories})
19116 which might help you in case your project has a lot of subdirectories.
19121 @font@heightrm=cmr8
19124 ## This Makefile is intended to be used with the following directory
19126 ## - The sources are split into a series of csc (computer software components)
19127 ## Each of these csc is put in its own directory.
19128 ## Their name are referenced by the directory names.
19129 ## They will be compiled into shared library (although this would also work
19130 ## with static libraries
19131 ## - The main program (and possibly other packages that do not belong to any
19132 ## csc is put in the top level directory (where the Makefile is).
19133 ## toplevel_dir __ first_csc (sources) __ lib (will contain the library)
19134 ## \_ second_csc (sources) __ lib (will contain the library)
19136 ## Although this Makefile is build for shared library, it is easy to modify
19137 ## to build partial link objects instead (modify the lines with -shared and
19140 ## With this makefile, you can change any file in the system or add any new
19141 ## file, and everything will be recompiled correctly (only the relevant shared
19142 ## objects will be recompiled, and the main program will be re-linked).
19144 # The list of computer software component for your project. This might be
19145 # generated automatically.
19148 # Name of the main program (no extension)
19151 # If we need to build objects with -fPIC, uncomment the following line
19154 # The following variable should give the directory containing libgnat.so
19155 # You can get this directory through 'gnatls -v'. This is usually the last
19156 # directory in the Object_Path.
19159 # The directories for the libraries
19160 # (This macro expands the list of CSC to the list of shared libraries, you
19161 # could simply use the expanded form:
19162 # LIB_DIR=aa/lib/libaa.so bb/lib/libbb.so cc/lib/libcc.so
19163 LIB_DIR=$@{foreach dir,$@{CSC_LIST@},$@{dir@}/lib/lib$@{dir@}.so@}
19165 $@{MAIN@}: objects $@{LIB_DIR@}
19166 gnatbind $@{MAIN@} $@{CSC_LIST:%=-aO%/lib@} -shared
19167 gnatlink $@{MAIN@} $@{CSC_LIST:%=-l%@}
19170 # recompile the sources
19171 gnatmake -c -i $@{MAIN@}.adb $@{NEED_FPIC@} $@{CSC_LIST:%=-I%@}
19173 # Note: In a future version of GNAT, the following commands will be simplified
19174 # by a new tool, gnatmlib
19176 mkdir -p $@{dir $@@ @}
19177 cd $@{dir $@@ @} && gcc -shared -o $@{notdir $@@ @} ../*.o -L$@{GLIB@} -lgnat
19178 cd $@{dir $@@ @} && cp -f ../*.ali .
19180 # The dependencies for the modules
19181 # Note that we have to force the expansion of *.o, since in some cases
19182 # make won't be able to do it itself.
19183 aa/lib/libaa.so: $@{wildcard aa/*.o@}
19184 bb/lib/libbb.so: $@{wildcard bb/*.o@}
19185 cc/lib/libcc.so: $@{wildcard cc/*.o@}
19187 # Make sure all of the shared libraries are in the path before starting the
19190 LD_LIBRARY_PATH=`pwd`/aa/lib:`pwd`/bb/lib:`pwd`/cc/lib ./$@{MAIN@}
19193 $@{RM@} -rf $@{CSC_LIST:%=%/lib@}
19194 $@{RM@} $@{CSC_LIST:%=%/*.ali@}
19195 $@{RM@} $@{CSC_LIST:%=%/*.o@}
19196 $@{RM@} *.o *.ali $@{MAIN@}
19199 @node Automatically Creating a List of Directories
19200 @section Automatically Creating a List of Directories
19203 In most makefiles, you will have to specify a list of directories, and
19204 store it in a variable. For small projects, it is often easier to
19205 specify each of them by hand, since you then have full control over what
19206 is the proper order for these directories, which ones should be
19209 However, in larger projects, which might involve hundreds of
19210 subdirectories, it might be more convenient to generate this list
19213 The example below presents two methods. The first one, although less
19214 general, gives you more control over the list. It involves wildcard
19215 characters, that are automatically expanded by @command{make}. Its
19216 shortcoming is that you need to explicitly specify some of the
19217 organization of your project, such as for instance the directory tree
19218 depth, whether some directories are found in a separate tree, @enddots{}
19220 The second method is the most general one. It requires an external
19221 program, called @command{find}, which is standard on all Unix systems. All
19222 the directories found under a given root directory will be added to the
19228 @font@heightrm=cmr8
19231 # The examples below are based on the following directory hierarchy:
19232 # All the directories can contain any number of files
19233 # ROOT_DIRECTORY -> a -> aa -> aaa
19236 # -> b -> ba -> baa
19239 # This Makefile creates a variable called DIRS, that can be reused any time
19240 # you need this list (see the other examples in this section)
19242 # The root of your project's directory hierarchy
19246 # First method: specify explicitly the list of directories
19247 # This allows you to specify any subset of all the directories you need.
19250 DIRS := a/aa/ a/ab/ b/ba/
19253 # Second method: use wildcards
19254 # Note that the argument(s) to wildcard below should end with a '/'.
19255 # Since wildcards also return file names, we have to filter them out
19256 # to avoid duplicate directory names.
19257 # We thus use make's @code{dir} and @code{sort} functions.
19258 # It sets DIRs to the following value (note that the directories aaa and baa
19259 # are not given, unless you change the arguments to wildcard).
19260 # DIRS= ./a/a/ ./b/ ./a/aa/ ./a/ab/ ./a/ac/ ./b/ba/ ./b/bb/ ./b/bc/
19263 DIRS := $@{sort $@{dir $@{wildcard $@{ROOT_DIRECTORY@}/*/
19264 $@{ROOT_DIRECTORY@}/*/*/@}@}@}
19267 # Third method: use an external program
19268 # This command is much faster if run on local disks, avoiding NFS slowdowns.
19269 # This is the most complete command: it sets DIRs to the following value:
19270 # DIRS= ./a ./a/aa ./a/aa/aaa ./a/ab ./a/ac ./b ./b/ba ./b/ba/baa ./b/bb ./b/bc
19273 DIRS := $@{shell find $@{ROOT_DIRECTORY@} -type d -print@}
19277 @node Generating the Command Line Switches
19278 @section Generating the Command Line Switches
19281 Once you have created the list of directories as explained in the
19282 previous section (@pxref{Automatically Creating a List of Directories}),
19283 you can easily generate the command line arguments to pass to gnatmake.
19285 For the sake of completeness, this example assumes that the source path
19286 is not the same as the object path, and that you have two separate lists
19290 # see "Automatically creating a list of directories" to create
19295 GNATMAKE_SWITCHES := $@{patsubst %,-aI%,$@{SOURCE_DIRS@}@}
19296 GNATMAKE_SWITCHES += $@{patsubst %,-aO%,$@{OBJECT_DIRS@}@}
19299 gnatmake $@{GNATMAKE_SWITCHES@} main_unit
19302 @node Overcoming Command Line Length Limits
19303 @section Overcoming Command Line Length Limits
19306 One problem that might be encountered on big projects is that many
19307 operating systems limit the length of the command line. It is thus hard to give
19308 gnatmake the list of source and object directories.
19310 This example shows how you can set up environment variables, which will
19311 make @command{gnatmake} behave exactly as if the directories had been
19312 specified on the command line, but have a much higher length limit (or
19313 even none on most systems).
19315 It assumes that you have created a list of directories in your Makefile,
19316 using one of the methods presented in
19317 @ref{Automatically Creating a List of Directories}.
19318 For the sake of completeness, we assume that the object
19319 path (where the ALI files are found) is different from the sources patch.
19321 Note a small trick in the Makefile below: for efficiency reasons, we
19322 create two temporary variables (SOURCE_LIST and OBJECT_LIST), that are
19323 expanded immediately by @code{make}. This way we overcome the standard
19324 make behavior which is to expand the variables only when they are
19327 On Windows, if you are using the standard Windows command shell, you must
19328 replace colons with semicolons in the assignments to these variables.
19333 @font@heightrm=cmr8
19336 # In this example, we create both ADA_INCLUDE_PATH and ADA_OBJECT_PATH.
19337 # This is the same thing as putting the -I arguments on the command line.
19338 # (the equivalent of using -aI on the command line would be to define
19339 # only ADA_INCLUDE_PATH, the equivalent of -aO is ADA_OBJECT_PATH).
19340 # You can of course have different values for these variables.
19342 # Note also that we need to keep the previous values of these variables, since
19343 # they might have been set before running 'make' to specify where the GNAT
19344 # library is installed.
19346 # see "Automatically creating a list of directories" to create these
19352 space:=$@{empty@} $@{empty@}
19353 SOURCE_LIST := $@{subst $@{space@},:,$@{SOURCE_DIRS@}@}
19354 OBJECT_LIST := $@{subst $@{space@},:,$@{OBJECT_DIRS@}@}
19355 ADA_INCLUDE_PATH += $@{SOURCE_LIST@}
19356 ADA_OBJECT_PATH += $@{OBJECT_LIST@}
19357 export ADA_INCLUDE_PATH
19358 export ADA_OBJECT_PATH
19365 @node Memory Management Issues
19366 @chapter Memory Management Issues
19369 This chapter describes some useful memory pools provided in the GNAT library
19370 and in particular the GNAT Debug Pool facility, which can be used to detect
19371 incorrect uses of access values (including ``dangling references'').
19373 It also describes the @command{gnatmem} tool, which can be used to track down
19378 * Some Useful Memory Pools::
19379 * The GNAT Debug Pool Facility::
19381 * The gnatmem Tool::
19385 @node Some Useful Memory Pools
19386 @section Some Useful Memory Pools
19387 @findex Memory Pool
19388 @cindex storage, pool
19391 The @code{System.Pool_Global} package offers the Unbounded_No_Reclaim_Pool
19392 storage pool. Allocations use the standard system call @code{malloc} while
19393 deallocations use the standard system call @code{free}. No reclamation is
19394 performed when the pool goes out of scope. For performance reasons, the
19395 standard default Ada allocators/deallocators do not use any explicit storage
19396 pools but if they did, they could use this storage pool without any change in
19397 behavior. That is why this storage pool is used when the user
19398 manages to make the default implicit allocator explicit as in this example:
19399 @smallexample @c ada
19400 type T1 is access Something;
19401 -- no Storage pool is defined for T2
19402 type T2 is access Something_Else;
19403 for T2'Storage_Pool use T1'Storage_Pool;
19404 -- the above is equivalent to
19405 for T2'Storage_Pool use System.Pool_Global.Global_Pool_Object;
19409 The @code{System.Pool_Local} package offers the Unbounded_Reclaim_Pool storage
19410 pool. The allocation strategy is similar to @code{Pool_Local}'s
19411 except that the all
19412 storage allocated with this pool is reclaimed when the pool object goes out of
19413 scope. This pool provides a explicit mechanism similar to the implicit one
19414 provided by several Ada 83 compilers for allocations performed through a local
19415 access type and whose purpose was to reclaim memory when exiting the
19416 scope of a given local access. As an example, the following program does not
19417 leak memory even though it does not perform explicit deallocation:
19419 @smallexample @c ada
19420 with System.Pool_Local;
19421 procedure Pooloc1 is
19422 procedure Internal is
19423 type A is access Integer;
19424 X : System.Pool_Local.Unbounded_Reclaim_Pool;
19425 for A'Storage_Pool use X;
19428 for I in 1 .. 50 loop
19433 for I in 1 .. 100 loop
19440 The @code{System.Pool_Size} package implements the Stack_Bounded_Pool used when
19441 @code{Storage_Size} is specified for an access type.
19442 The whole storage for the pool is
19443 allocated at once, usually on the stack at the point where the access type is
19444 elaborated. It is automatically reclaimed when exiting the scope where the
19445 access type is defined. This package is not intended to be used directly by the
19446 user and it is implicitly used for each such declaration:
19448 @smallexample @c ada
19449 type T1 is access Something;
19450 for T1'Storage_Size use 10_000;
19453 @node The GNAT Debug Pool Facility
19454 @section The GNAT Debug Pool Facility
19456 @cindex storage, pool, memory corruption
19459 The use of unchecked deallocation and unchecked conversion can easily
19460 lead to incorrect memory references. The problems generated by such
19461 references are usually difficult to tackle because the symptoms can be
19462 very remote from the origin of the problem. In such cases, it is
19463 very helpful to detect the problem as early as possible. This is the
19464 purpose of the Storage Pool provided by @code{GNAT.Debug_Pools}.
19466 In order to use the GNAT specific debugging pool, the user must
19467 associate a debug pool object with each of the access types that may be
19468 related to suspected memory problems. See Ada Reference Manual 13.11.
19469 @smallexample @c ada
19470 type Ptr is access Some_Type;
19471 Pool : GNAT.Debug_Pools.Debug_Pool;
19472 for Ptr'Storage_Pool use Pool;
19476 @code{GNAT.Debug_Pools} is derived from a GNAT-specific kind of
19477 pool: the @code{Checked_Pool}. Such pools, like standard Ada storage pools,
19478 allow the user to redefine allocation and deallocation strategies. They
19479 also provide a checkpoint for each dereference, through the use of
19480 the primitive operation @code{Dereference} which is implicitly called at
19481 each dereference of an access value.
19483 Once an access type has been associated with a debug pool, operations on
19484 values of the type may raise four distinct exceptions,
19485 which correspond to four potential kinds of memory corruption:
19488 @code{GNAT.Debug_Pools.Accessing_Not_Allocated_Storage}
19490 @code{GNAT.Debug_Pools.Accessing_Deallocated_Storage}
19492 @code{GNAT.Debug_Pools.Freeing_Not_Allocated_Storage}
19494 @code{GNAT.Debug_Pools.Freeing_Deallocated_Storage }
19498 For types associated with a Debug_Pool, dynamic allocation is performed using
19499 the standard GNAT allocation routine. References to all allocated chunks of
19500 memory are kept in an internal dictionary. Several deallocation strategies are
19501 provided, whereupon the user can choose to release the memory to the system,
19502 keep it allocated for further invalid access checks, or fill it with an easily
19503 recognizable pattern for debug sessions. The memory pattern is the old IBM
19504 hexadecimal convention: @code{16#DEADBEEF#}.
19506 See the documentation in the file g-debpoo.ads for more information on the
19507 various strategies.
19509 Upon each dereference, a check is made that the access value denotes a
19510 properly allocated memory location. Here is a complete example of use of
19511 @code{Debug_Pools}, that includes typical instances of memory corruption:
19512 @smallexample @c ada
19516 with Gnat.Io; use Gnat.Io;
19517 with Unchecked_Deallocation;
19518 with Unchecked_Conversion;
19519 with GNAT.Debug_Pools;
19520 with System.Storage_Elements;
19521 with Ada.Exceptions; use Ada.Exceptions;
19522 procedure Debug_Pool_Test is
19524 type T is access Integer;
19525 type U is access all T;
19527 P : GNAT.Debug_Pools.Debug_Pool;
19528 for T'Storage_Pool use P;
19530 procedure Free is new Unchecked_Deallocation (Integer, T);
19531 function UC is new Unchecked_Conversion (U, T);
19534 procedure Info is new GNAT.Debug_Pools.Print_Info(Put_Line);
19544 Put_Line (Integer'Image(B.all));
19546 when E : others => Put_Line ("raised: " & Exception_Name (E));
19551 when E : others => Put_Line ("raised: " & Exception_Name (E));
19555 Put_Line (Integer'Image(B.all));
19557 when E : others => Put_Line ("raised: " & Exception_Name (E));
19562 when E : others => Put_Line ("raised: " & Exception_Name (E));
19565 end Debug_Pool_Test;
19569 The debug pool mechanism provides the following precise diagnostics on the
19570 execution of this erroneous program:
19573 Total allocated bytes : 0
19574 Total deallocated bytes : 0
19575 Current Water Mark: 0
19579 Total allocated bytes : 8
19580 Total deallocated bytes : 0
19581 Current Water Mark: 8
19584 raised: GNAT.DEBUG_POOLS.ACCESSING_DEALLOCATED_STORAGE
19585 raised: GNAT.DEBUG_POOLS.FREEING_DEALLOCATED_STORAGE
19586 raised: GNAT.DEBUG_POOLS.ACCESSING_NOT_ALLOCATED_STORAGE
19587 raised: GNAT.DEBUG_POOLS.FREEING_NOT_ALLOCATED_STORAGE
19589 Total allocated bytes : 8
19590 Total deallocated bytes : 4
19591 Current Water Mark: 4
19596 @node The gnatmem Tool
19597 @section The @command{gnatmem} Tool
19601 The @code{gnatmem} utility monitors dynamic allocation and
19602 deallocation activity in a program, and displays information about
19603 incorrect deallocations and possible sources of memory leaks.
19604 It provides three type of information:
19607 General information concerning memory management, such as the total
19608 number of allocations and deallocations, the amount of allocated
19609 memory and the high water mark, i.e.@: the largest amount of allocated
19610 memory in the course of program execution.
19613 Backtraces for all incorrect deallocations, that is to say deallocations
19614 which do not correspond to a valid allocation.
19617 Information on each allocation that is potentially the origin of a memory
19622 * Running gnatmem::
19623 * Switches for gnatmem::
19624 * Example of gnatmem Usage::
19627 @node Running gnatmem
19628 @subsection Running @code{gnatmem}
19631 @code{gnatmem} makes use of the output created by the special version of
19632 allocation and deallocation routines that record call information. This
19633 allows to obtain accurate dynamic memory usage history at a minimal cost to
19634 the execution speed. Note however, that @code{gnatmem} is not supported on
19635 all platforms (currently, it is supported on AIX, HP-UX, GNU/Linux,
19636 Solaris and Windows NT/2000/XP (x86).
19639 The @code{gnatmem} command has the form
19642 $ gnatmem @ovar{switches} user_program
19646 The program must have been linked with the instrumented version of the
19647 allocation and deallocation routines. This is done by linking with the
19648 @file{libgmem.a} library. For correct symbolic backtrace information,
19649 the user program should be compiled with debugging options
19650 (see @ref{Switches for gcc}). For example to build @file{my_program}:
19653 $ gnatmake -g my_program -largs -lgmem
19657 As library @file{libgmem.a} contains an alternate body for package
19658 @code{System.Memory}, @file{s-memory.adb} should not be compiled and linked
19659 when an executable is linked with library @file{libgmem.a}. It is then not
19660 recommended to use @command{gnatmake} with switch @option{^-a^/ALL_FILES^}.
19663 When @file{my_program} is executed, the file @file{gmem.out} is produced.
19664 This file contains information about all allocations and deallocations
19665 performed by the program. It is produced by the instrumented allocations and
19666 deallocations routines and will be used by @code{gnatmem}.
19668 In order to produce symbolic backtrace information for allocations and
19669 deallocations performed by the GNAT run-time library, you need to use a
19670 version of that library that has been compiled with the @option{-g} switch
19671 (see @ref{Rebuilding the GNAT Run-Time Library}).
19673 Gnatmem must be supplied with the @file{gmem.out} file and the executable to
19674 examine. If the location of @file{gmem.out} file was not explicitly supplied by
19675 @option{-i} switch, gnatmem will assume that this file can be found in the
19676 current directory. For example, after you have executed @file{my_program},
19677 @file{gmem.out} can be analyzed by @code{gnatmem} using the command:
19680 $ gnatmem my_program
19684 This will produce the output with the following format:
19686 *************** debut cc
19688 $ gnatmem my_program
19692 Total number of allocations : 45
19693 Total number of deallocations : 6
19694 Final Water Mark (non freed mem) : 11.29 Kilobytes
19695 High Water Mark : 11.40 Kilobytes
19700 Allocation Root # 2
19701 -------------------
19702 Number of non freed allocations : 11
19703 Final Water Mark (non freed mem) : 1.16 Kilobytes
19704 High Water Mark : 1.27 Kilobytes
19706 my_program.adb:23 my_program.alloc
19712 The first block of output gives general information. In this case, the
19713 Ada construct ``@code{@b{new}}'' was executed 45 times, and only 6 calls to an
19714 Unchecked_Deallocation routine occurred.
19717 Subsequent paragraphs display information on all allocation roots.
19718 An allocation root is a specific point in the execution of the program
19719 that generates some dynamic allocation, such as a ``@code{@b{new}}''
19720 construct. This root is represented by an execution backtrace (or subprogram
19721 call stack). By default the backtrace depth for allocations roots is 1, so
19722 that a root corresponds exactly to a source location. The backtrace can
19723 be made deeper, to make the root more specific.
19725 @node Switches for gnatmem
19726 @subsection Switches for @code{gnatmem}
19729 @code{gnatmem} recognizes the following switches:
19734 @cindex @option{-q} (@code{gnatmem})
19735 Quiet. Gives the minimum output needed to identify the origin of the
19736 memory leaks. Omits statistical information.
19739 @cindex @var{N} (@code{gnatmem})
19740 N is an integer literal (usually between 1 and 10) which controls the
19741 depth of the backtraces defining allocation root. The default value for
19742 N is 1. The deeper the backtrace, the more precise the localization of
19743 the root. Note that the total number of roots can depend on this
19744 parameter. This parameter must be specified @emph{before} the name of the
19745 executable to be analyzed, to avoid ambiguity.
19748 @cindex @option{-b} (@code{gnatmem})
19749 This switch has the same effect as just depth parameter.
19751 @item -i @var{file}
19752 @cindex @option{-i} (@code{gnatmem})
19753 Do the @code{gnatmem} processing starting from @file{file}, rather than
19754 @file{gmem.out} in the current directory.
19757 @cindex @option{-m} (@code{gnatmem})
19758 This switch causes @code{gnatmem} to mask the allocation roots that have less
19759 than n leaks. The default value is 1. Specifying the value of 0 will allow to
19760 examine even the roots that didn't result in leaks.
19763 @cindex @option{-s} (@code{gnatmem})
19764 This switch causes @code{gnatmem} to sort the allocation roots according to the
19765 specified order of sort criteria, each identified by a single letter. The
19766 currently supported criteria are @code{n, h, w} standing respectively for
19767 number of unfreed allocations, high watermark, and final watermark
19768 corresponding to a specific root. The default order is @code{nwh}.
19772 @node Example of gnatmem Usage
19773 @subsection Example of @code{gnatmem} Usage
19776 The following example shows the use of @code{gnatmem}
19777 on a simple memory-leaking program.
19778 Suppose that we have the following Ada program:
19780 @smallexample @c ada
19783 with Unchecked_Deallocation;
19784 procedure Test_Gm is
19786 type T is array (1..1000) of Integer;
19787 type Ptr is access T;
19788 procedure Free is new Unchecked_Deallocation (T, Ptr);
19791 procedure My_Alloc is
19796 procedure My_DeAlloc is
19804 for I in 1 .. 5 loop
19805 for J in I .. 5 loop
19816 The program needs to be compiled with debugging option and linked with
19817 @code{gmem} library:
19820 $ gnatmake -g test_gm -largs -lgmem
19824 Then we execute the program as usual:
19831 Then @code{gnatmem} is invoked simply with
19837 which produces the following output (result may vary on different platforms):
19842 Total number of allocations : 18
19843 Total number of deallocations : 5
19844 Final Water Mark (non freed mem) : 53.00 Kilobytes
19845 High Water Mark : 56.90 Kilobytes
19847 Allocation Root # 1
19848 -------------------
19849 Number of non freed allocations : 11
19850 Final Water Mark (non freed mem) : 42.97 Kilobytes
19851 High Water Mark : 46.88 Kilobytes
19853 test_gm.adb:11 test_gm.my_alloc
19855 Allocation Root # 2
19856 -------------------
19857 Number of non freed allocations : 1
19858 Final Water Mark (non freed mem) : 10.02 Kilobytes
19859 High Water Mark : 10.02 Kilobytes
19861 s-secsta.adb:81 system.secondary_stack.ss_init
19863 Allocation Root # 3
19864 -------------------
19865 Number of non freed allocations : 1
19866 Final Water Mark (non freed mem) : 12 Bytes
19867 High Water Mark : 12 Bytes
19869 s-secsta.adb:181 system.secondary_stack.ss_init
19873 Note that the GNAT run time contains itself a certain number of
19874 allocations that have no corresponding deallocation,
19875 as shown here for root #2 and root
19876 #3. This is a normal behavior when the number of non-freed allocations
19877 is one, it allocates dynamic data structures that the run time needs for
19878 the complete lifetime of the program. Note also that there is only one
19879 allocation root in the user program with a single line back trace:
19880 test_gm.adb:11 test_gm.my_alloc, whereas a careful analysis of the
19881 program shows that 'My_Alloc' is called at 2 different points in the
19882 source (line 21 and line 24). If those two allocation roots need to be
19883 distinguished, the backtrace depth parameter can be used:
19886 $ gnatmem 3 test_gm
19890 which will give the following output:
19895 Total number of allocations : 18
19896 Total number of deallocations : 5
19897 Final Water Mark (non freed mem) : 53.00 Kilobytes
19898 High Water Mark : 56.90 Kilobytes
19900 Allocation Root # 1
19901 -------------------
19902 Number of non freed allocations : 10
19903 Final Water Mark (non freed mem) : 39.06 Kilobytes
19904 High Water Mark : 42.97 Kilobytes
19906 test_gm.adb:11 test_gm.my_alloc
19907 test_gm.adb:24 test_gm
19908 b_test_gm.c:52 main
19910 Allocation Root # 2
19911 -------------------
19912 Number of non freed allocations : 1
19913 Final Water Mark (non freed mem) : 10.02 Kilobytes
19914 High Water Mark : 10.02 Kilobytes
19916 s-secsta.adb:81 system.secondary_stack.ss_init
19917 s-secsta.adb:283 <system__secondary_stack___elabb>
19918 b_test_gm.c:33 adainit
19920 Allocation Root # 3
19921 -------------------
19922 Number of non freed allocations : 1
19923 Final Water Mark (non freed mem) : 3.91 Kilobytes
19924 High Water Mark : 3.91 Kilobytes
19926 test_gm.adb:11 test_gm.my_alloc
19927 test_gm.adb:21 test_gm
19928 b_test_gm.c:52 main
19930 Allocation Root # 4
19931 -------------------
19932 Number of non freed allocations : 1
19933 Final Water Mark (non freed mem) : 12 Bytes
19934 High Water Mark : 12 Bytes
19936 s-secsta.adb:181 system.secondary_stack.ss_init
19937 s-secsta.adb:283 <system__secondary_stack___elabb>
19938 b_test_gm.c:33 adainit
19942 The allocation root #1 of the first example has been split in 2 roots #1
19943 and #3 thanks to the more precise associated backtrace.
19947 @node Stack Related Facilities
19948 @chapter Stack Related Facilities
19951 This chapter describes some useful tools associated with stack
19952 checking and analysis. In
19953 particular, it deals with dynamic and static stack usage measurements.
19956 * Stack Overflow Checking::
19957 * Static Stack Usage Analysis::
19958 * Dynamic Stack Usage Analysis::
19961 @node Stack Overflow Checking
19962 @section Stack Overflow Checking
19963 @cindex Stack Overflow Checking
19964 @cindex -fstack-check
19967 For most operating systems, @command{gcc} does not perform stack overflow
19968 checking by default. This means that if the main environment task or
19969 some other task exceeds the available stack space, then unpredictable
19970 behavior will occur. Most native systems offer some level of protection by
19971 adding a guard page at the end of each task stack. This mechanism is usually
19972 not enough for dealing properly with stack overflow situations because
19973 a large local variable could ``jump'' above the guard page.
19974 Furthermore, when the
19975 guard page is hit, there may not be any space left on the stack for executing
19976 the exception propagation code. Enabling stack checking avoids
19979 To activate stack checking, compile all units with the gcc option
19980 @option{-fstack-check}. For example:
19983 gcc -c -fstack-check package1.adb
19987 Units compiled with this option will generate extra instructions to check
19988 that any use of the stack (for procedure calls or for declaring local
19989 variables in declare blocks) does not exceed the available stack space.
19990 If the space is exceeded, then a @code{Storage_Error} exception is raised.
19992 For declared tasks, the stack size is controlled by the size
19993 given in an applicable @code{Storage_Size} pragma or by the value specified
19994 at bind time with @option{-d} (@pxref{Switches for gnatbind}) or is set to
19995 the default size as defined in the GNAT runtime otherwise.
19997 For the environment task, the stack size depends on
19998 system defaults and is unknown to the compiler. Stack checking
19999 may still work correctly if a fixed
20000 size stack is allocated, but this cannot be guaranteed.
20002 To ensure that a clean exception is signalled for stack
20003 overflow, set the environment variable
20004 @env{GNAT_STACK_LIMIT} to indicate the maximum
20005 stack area that can be used, as in:
20006 @cindex GNAT_STACK_LIMIT
20009 SET GNAT_STACK_LIMIT 1600
20013 The limit is given in kilobytes, so the above declaration would
20014 set the stack limit of the environment task to 1.6 megabytes.
20015 Note that the only purpose of this usage is to limit the amount
20016 of stack used by the environment task. If it is necessary to
20017 increase the amount of stack for the environment task, then this
20018 is an operating systems issue, and must be addressed with the
20019 appropriate operating systems commands.
20022 To have a fixed size stack in the environment task, the stack must be put
20023 in the P0 address space and its size specified. Use these switches to
20027 gnatmake my_progs -largs "-Wl,--opt=STACK=4000,/p0image"
20031 The quotes are required to keep case. The number after @samp{STACK=} is the
20032 size of the environmental task stack in pagelets (512 bytes). In this example
20033 the stack size is about 2 megabytes.
20036 A consequence of the @option{/p0image} qualifier is also to makes RMS buffers
20037 be placed in P0 space. Refer to @cite{HP OpenVMS Linker Utility Manual} for
20038 more details about the @option{/p0image} qualifier and the @option{stack}
20042 @node Static Stack Usage Analysis
20043 @section Static Stack Usage Analysis
20044 @cindex Static Stack Usage Analysis
20045 @cindex -fstack-usage
20048 A unit compiled with @option{-fstack-usage} will generate an extra file
20050 the maximum amount of stack used, on a per-function basis.
20051 The file has the same
20052 basename as the target object file with a @file{.su} extension.
20053 Each line of this file is made up of three fields:
20057 The name of the function.
20061 One or more qualifiers: @code{static}, @code{dynamic}, @code{bounded}.
20064 The second field corresponds to the size of the known part of the function
20067 The qualifier @code{static} means that the function frame size
20069 It usually means that all local variables have a static size.
20070 In this case, the second field is a reliable measure of the function stack
20073 The qualifier @code{dynamic} means that the function frame size is not static.
20074 It happens mainly when some local variables have a dynamic size. When this
20075 qualifier appears alone, the second field is not a reliable measure
20076 of the function stack analysis. When it is qualified with @code{bounded}, it
20077 means that the second field is a reliable maximum of the function stack
20080 @node Dynamic Stack Usage Analysis
20081 @section Dynamic Stack Usage Analysis
20084 It is possible to measure the maximum amount of stack used by a task, by
20085 adding a switch to @command{gnatbind}, as:
20088 $ gnatbind -u0 file
20092 With this option, at each task termination, its stack usage is output on
20094 It is not always convenient to output the stack usage when the program
20095 is still running. Hence, it is possible to delay this output until program
20096 termination. for a given number of tasks specified as the argument of the
20097 @option{-u} option. For instance:
20100 $ gnatbind -u100 file
20104 will buffer the stack usage information of the first 100 tasks to terminate and
20105 output this info at program termination. Results are displayed in four
20109 Index | Task Name | Stack Size | Actual Use [min - max]
20116 is a number associated with each task.
20119 is the name of the task analyzed.
20122 is the maximum size for the stack.
20125 is the measure done by the stack analyzer. In order to prevent overflow,
20126 the stack is not entirely analyzed, and it's not possible to know exactly how
20127 much has actually been used. The real amount of stack used is between the min
20133 The environment task stack, e.g., the stack that contains the main unit, is
20134 only processed when the environment variable GNAT_STACK_LIMIT is set.
20137 @c *********************************
20139 @c *********************************
20140 @node Verifying Properties Using gnatcheck
20141 @chapter Verifying Properties Using @command{gnatcheck}
20143 @cindex @command{gnatcheck}
20146 The @command{gnatcheck} tool is an ASIS-based utility that checks properties
20147 of Ada source files according to a given set of semantic rules.
20150 In order to check compliance with a given rule, @command{gnatcheck} has to
20151 semantically analyze the Ada sources.
20152 Therefore, checks can only be performed on
20153 legal Ada units. Moreover, when a unit depends semantically upon units located
20154 outside the current directory, the source search path has to be provided when
20155 calling @command{gnatcheck}, either through a specified project file or
20156 through @command{gnatcheck} switches as described below.
20158 A number of rules are predefined in @command{gnatcheck} and are described
20159 later in this chapter.
20160 You can also add new rules, by modifying the @command{gnatcheck} code and
20161 rebuilding the tool. In order to add a simple rule making some local checks,
20162 a small amount of straightforward ASIS-based programming is usually needed.
20164 Project support for @command{gnatcheck} is provided by the GNAT
20165 driver (see @ref{The GNAT Driver and Project Files}).
20167 Invoking @command{gnatcheck} on the command line has the form:
20170 $ gnatcheck @ovar{switches} @{@var{filename}@}
20171 @r{[}^-files^/FILES^=@{@var{arg_list_filename}@}@r{]}
20172 @r{[}-cargs @var{gcc_switches}@r{]} @r{[}-rules @var{rule_options}@r{]}
20179 @var{switches} specify the general tool options
20182 Each @var{filename} is the name (including the extension) of a source
20183 file to process. ``Wildcards'' are allowed, and
20184 the file name may contain path information.
20187 Each @var{arg_list_filename} is the name (including the extension) of a text
20188 file containing the names of the source files to process, separated by spaces
20192 @var{gcc_switches} is a list of switches for
20193 @command{gcc}. They will be passed on to all compiler invocations made by
20194 @command{gnatcheck} to generate the ASIS trees. Here you can provide
20195 @option{^-I^/INCLUDE_DIRS=^} switches to form the source search path,
20196 and use the @option{-gnatec} switch to set the configuration file.
20199 @var{rule_options} is a list of options for controlling a set of
20200 rules to be checked by @command{gnatcheck} (@pxref{gnatcheck Rule Options}).
20204 Either a @file{@var{filename}} or an @file{@var{arg_list_filename}} must be supplied.
20207 * Format of the Report File::
20208 * General gnatcheck Switches::
20209 * gnatcheck Rule Options::
20210 * Adding the Results of Compiler Checks to gnatcheck Output::
20211 * Project-Wide Checks::
20212 * Predefined Rules::
20215 @node Format of the Report File
20216 @section Format of the Report File
20217 @cindex Report file (for @code{gnatcheck})
20220 The @command{gnatcheck} tool outputs on @file{stdout} all messages concerning
20222 It also creates, in the current
20223 directory, a text file named @file{^gnatcheck.out^GNATCHECK.OUT^} that
20224 contains the complete report of the last gnatcheck run. This report contains:
20226 @item a list of the Ada source files being checked,
20227 @item a list of enabled and disabled rules,
20228 @item a list of the diagnostic messages, ordered in three different ways
20229 and collected in three separate
20230 sections. Section 1 contains the raw list of diagnostic messages. It
20231 corresponds to the output going to @file{stdout}. Section 2 contains
20232 messages ordered by rules.
20233 Section 3 contains messages ordered by source files.
20236 @node General gnatcheck Switches
20237 @section General @command{gnatcheck} Switches
20240 The following switches control the general @command{gnatcheck} behavior
20244 @cindex @option{^-a^/ALL^} (@command{gnatcheck})
20246 Process all units including those with read-only ALI files such as
20247 those from GNAT Run-Time library.
20251 @cindex @option{-d} (@command{gnatcheck})
20256 @cindex @option{-dd} (@command{gnatcheck})
20258 Progress indicator mode (for use in GPS)
20261 @cindex @option{^-h^/HELP^} (@command{gnatcheck})
20263 List the predefined and user-defined rules. For more details see
20264 @ref{Predefined Rules}.
20266 @cindex @option{^-l^/LOCS^} (@command{gnatcheck})
20268 Use full source locations references in the report file. For a construct from
20269 a generic instantiation a full source location is a chain from the location
20270 of this construct in the generic unit to the place where this unit is
20273 @cindex @option{^-q^/QUIET^} (@command{gnatcheck})
20275 Quiet mode. All the diagnoses about rule violations are placed in the
20276 @command{gnatcheck} report file only, without duplicating in @file{stdout}.
20278 @cindex @option{^-s^/SHORT^} (@command{gnatcheck})
20280 Short format of the report file (no version information, no list of applied
20281 rules, no list of checked sources is included)
20283 @cindex @option{^-s1^/COMPILER_STYLE^} (@command{gnatcheck})
20284 @item ^-s1^/COMPILER_STYLE^
20285 Include the compiler-style section in the report file
20287 @cindex @option{^-s2^/BY_RULES^} (@command{gnatcheck})
20288 @item ^-s2^/BY_RULES^
20289 Include the section containing diagnoses ordered by rules in the report file
20291 @cindex @option{^-s3^/BY_FILES_BY_RULES^} (@command{gnatcheck})
20292 @item ^-s3^/BY_FILES_BY_RULES^
20293 Include the section containing diagnoses ordered by files and then by rules
20296 @cindex @option{^-v^/VERBOSE^} (@command{gnatcheck})
20297 @item ^-v^/VERBOSE^
20298 Verbose mode; @command{gnatcheck} generates version information and then
20299 a trace of sources being processed.
20304 Note that if any of the options @option{^-s1^/COMPILER_STYLE^},
20305 @option{^-s2^/BY_RULES^} or
20306 @option{^-s3^/BY_FILES_BY_RULES^} is specified,
20307 then the @command{gnatcheck} report file will only contain sections
20308 explicitly denoted by these options.
20310 @node gnatcheck Rule Options
20311 @section @command{gnatcheck} Rule Options
20314 The following options control the processing performed by
20315 @command{gnatcheck}.
20318 @cindex @option{+ALL} (@command{gnatcheck})
20320 Turn all the rule checks ON.
20322 @cindex @option{-ALL} (@command{gnatcheck})
20324 Turn all the rule checks OFF.
20326 @cindex @option{+R} (@command{gnatcheck})
20327 @item +R@var{rule_id}@r{[}:@var{param}@r{]}
20328 Turn on the check for a specified rule with the specified parameter, if any.
20329 @var{rule_id} must be the identifier of one of the currently implemented rules
20330 (use @option{^-h^/HELP^} for the list of implemented rules). Rule identifiers
20331 are not case-sensitive. The @var{param} item must
20332 be a string representing a valid parameter(s) for the specified rule.
20333 If it contains any space characters then this string must be enclosed in
20336 @cindex @option{-R} (@command{gnatcheck})
20337 @item -R@var{rule_id}@r{[}:@var{param}@r{]}
20338 Turn off the check for a specified rule with the specified parameter, if any.
20340 @cindex @option{-from} (@command{gnatcheck})
20341 @item -from=@var{rule_option_filename}
20342 Read the rule options from the text file @var{rule_option_filename}, referred as
20343 ``rule file'' below.
20348 The default behavior is that all the rule checks are enabled, except for
20349 the checks performed by the compiler.
20351 and the checks associated with the
20355 A rule file is a text file containing a set of rule options.
20356 @cindex Rule file (for @code{gnatcheck})
20357 The file may contain empty lines and Ada-style comments (comment
20358 lines and end-of-line comments). The rule file has free format; that is,
20359 you do not have to start a new rule option on a new line.
20361 A rule file may contain other @option{-from=@var{rule_option_filename}}
20362 options, each such option being replaced with the content of the
20363 corresponding rule file during the rule files processing. In case a
20364 cycle is detected (that is, @file{@var{rule_file_1}} reads rule options
20365 from @file{@var{rule_file_2}}, and @file{@var{rule_file_2}} reads
20366 (directly or indirectly) rule options from @file{@var{rule_file_1}}),
20367 the processing of rule files is interrupted and a part of their content
20371 @node Adding the Results of Compiler Checks to gnatcheck Output
20372 @section Adding the Results of Compiler Checks to @command{gnatcheck} Output
20375 The @command{gnatcheck} tool can include in the generated diagnostic messages
20377 the report file the results of the checks performed by the compiler. Though
20378 disabled by default, this effect may be obtained by using @option{+R} with
20379 the following rule identifiers and parameters:
20383 To record restrictions violations (that are performed by the compiler if the
20384 pragma @code{Restrictions} or @code{Restriction_Warnings} are given),
20386 @code{Restrictions} with the same parameters as pragma
20387 @code{Restrictions} or @code{Restriction_Warnings}.
20390 To record compiler style checks(@pxref{Style Checking}), use the rule named
20391 @code{Style_Checks}. A parameter of this rule can be either @code{All_Checks},
20392 which enables all the standard style checks that corresponds to @option{-gnatyy}
20393 GNAT style check option, or a string that has exactly the same
20394 structure and semantics as the @code{string_LITERAL} parameter of GNAT pragma
20395 @code{Style_Checks} (for further information about this pragma,
20396 @pxref{Pragma Style_Checks,,, gnat_rm, GNAT Reference Manual}).
20399 To record compiler warnings (@pxref{Warning Message Control}), use the rule
20400 named @code{Warnings} with a parameter that is a valid
20401 @i{static_string_expression} argument of GNAT pragma @code{Warnings}
20402 (for further information about this pragma, @pxref{Pragma Warnings,,,
20403 gnat_rm, GNAT Reference Manual}). Note, that in case of gnatcheck
20404 's' parameter, that corresponds to the GNAT @option{-gnatws} option, disables
20405 all the specific warnings, but not suppresses the warning mode,
20406 and 'e' parameter, corresponding to @option{-gnatwe} that means
20407 "treat warnings as errors", does not have any effect.
20411 To disable a specific restriction check, use @code{-RStyle_Checks} gnatcheck
20412 option with the corresponding restriction name as a parameter. @code{-R} is
20413 not available for @code{Style_Checks} and @code{Warnings} options, to disable
20414 warnings and style checks, use the corresponding warning and style options.
20416 @node Project-Wide Checks
20417 @section Project-Wide Checks
20418 @cindex Project-wide checks (for @command{gnatcheck})
20421 In order to perform checks on all units of a given project, you can use
20422 the GNAT driver along with the @option{-P} option:
20424 gnat check -Pproj -rules -from=my_rules
20428 If the project @code{proj} depends upon other projects, you can perform
20429 checks on the project closure using the @option{-U} option:
20431 gnat check -Pproj -U -rules -from=my_rules
20435 Finally, if not all the units are relevant to a particular main
20436 program in the project closure, you can perform checks for the set
20437 of units needed to create a given main program (unit closure) using
20438 the @option{-U} option followed by the name of the main unit:
20440 gnat check -Pproj -U main -rules -from=my_rules
20444 @node Predefined Rules
20445 @section Predefined Rules
20446 @cindex Predefined rules (for @command{gnatcheck})
20449 @c (Jan 2007) Since the global rules are still under development and are not
20450 @c documented, there is no point in explaining the difference between
20451 @c global and local rules
20453 A rule in @command{gnatcheck} is either local or global.
20454 A @emph{local rule} is a rule that applies to a well-defined section
20455 of a program and that can be checked by analyzing only this section.
20456 A @emph{global rule} requires analysis of some global properties of the
20457 whole program (mostly related to the program call graph).
20458 As of @value{NOW}, the implementation of global rules should be
20459 considered to be at a preliminary stage. You can use the
20460 @option{+GLOBAL} option to enable all the global rules, and the
20461 @option{-GLOBAL} rule option to disable all the global rules.
20463 All the global rules in the list below are
20464 so indicated by marking them ``GLOBAL''.
20465 This +GLOBAL and -GLOBAL options are not
20466 included in the list of gnatcheck options above, because at the moment they
20467 are considered as a temporary debug options.
20469 @command{gnatcheck} performs rule checks for generic
20470 instances only for global rules. This limitation may be relaxed in a later
20475 The following subsections document the rules implemented in
20476 @command{gnatcheck}.
20477 The subsection title is the same as the rule identifier, which may be
20478 used as a parameter of the @option{+R} or @option{-R} options.
20482 * Abstract_Type_Declarations::
20483 * Anonymous_Arrays::
20484 * Anonymous_Subtypes::
20486 * Boolean_Relational_Operators::
20488 * Ceiling_Violations::
20490 * Controlled_Type_Declarations::
20491 * Declarations_In_Blocks::
20492 * Default_Parameters::
20493 * Discriminated_Records::
20494 * Enumeration_Ranges_In_CASE_Statements::
20495 * Exceptions_As_Control_Flow::
20496 * EXIT_Statements_With_No_Loop_Name::
20497 * Expanded_Loop_Exit_Names::
20498 * Explicit_Full_Discrete_Ranges::
20499 * Float_Equality_Checks::
20500 * Forbidden_Pragmas::
20501 * Function_Style_Procedures::
20502 * Generics_In_Subprograms::
20503 * GOTO_Statements::
20504 * Implicit_IN_Mode_Parameters::
20505 * Implicit_SMALL_For_Fixed_Point_Types::
20506 * Improperly_Located_Instantiations::
20507 * Improper_Returns::
20508 * Library_Level_Subprograms::
20511 * Improperly_Called_Protected_Entries::
20513 * Metrics_Violation::
20514 * Misnamed_Identifiers::
20515 * Multiple_Entries_In_Protected_Definitions::
20517 * Non_Qualified_Aggregates::
20518 * Non_Short_Circuit_Operators::
20519 * Non_SPARK_Attributes::
20520 * Non_Tagged_Derived_Types::
20521 * Non_Visible_Exceptions::
20522 * Numeric_Literals::
20523 * OTHERS_In_Aggregates::
20524 * OTHERS_In_CASE_Statements::
20525 * OTHERS_In_Exception_Handlers::
20526 * Outer_Loop_Exits::
20527 * Overloaded_Operators::
20528 * Overly_Nested_Control_Structures::
20529 * Parameters_Out_Of_Order::
20530 * Positional_Actuals_For_Defaulted_Generic_Parameters::
20531 * Positional_Actuals_For_Defaulted_Parameters::
20532 * Positional_Components::
20533 * Positional_Generic_Parameters::
20534 * Positional_Parameters::
20535 * Predefined_Numeric_Types::
20536 * Raising_External_Exceptions::
20537 * Raising_Predefined_Exceptions::
20538 * Separate_Numeric_Error_Handlers::
20541 * Side_Effect_Functions::
20544 * Unassigned_OUT_Parameters::
20545 * Uncommented_BEGIN_In_Package_Bodies::
20546 * Unconstrained_Array_Returns::
20547 * Universal_Ranges::
20548 * Unnamed_Blocks_And_Loops::
20550 * Unused_Subprograms::
20552 * USE_PACKAGE_Clauses::
20553 * Volatile_Objects_Without_Address_Clauses::
20557 @node Abstract_Type_Declarations
20558 @subsection @code{Abstract_Type_Declarations}
20559 @cindex @code{Abstract_Type_Declarations} rule (for @command{gnatcheck})
20562 Flag all declarations of abstract types. For an abstract private
20563 type, both the private and full type declarations are flagged.
20565 This rule has no parameters.
20568 @node Anonymous_Arrays
20569 @subsection @code{Anonymous_Arrays}
20570 @cindex @code{Anonymous_Arrays} rule (for @command{gnatcheck})
20573 Flag all anonymous array type definitions (by Ada semantics these can only
20574 occur in object declarations).
20576 This rule has no parameters.
20578 @node Anonymous_Subtypes
20579 @subsection @code{Anonymous_Subtypes}
20580 @cindex @code{Anonymous_Subtypes} rule (for @command{gnatcheck})
20583 Flag all uses of anonymous subtypes. A use of an anonymous subtype is
20584 any instance of a subtype indication with a constraint, other than one
20585 that occurs immediately within a subtype declaration. Any use of a range
20586 other than as a constraint used immediately within a subtype declaration
20587 is considered as an anonymous subtype.
20589 An effect of this rule is that @code{for} loops such as the following are
20590 flagged (since @code{1..N} is formally a ``range''):
20592 @smallexample @c ada
20593 for I in 1 .. N loop
20599 Declaring an explicit subtype solves the problem:
20601 @smallexample @c ada
20602 subtype S is Integer range 1..N;
20610 This rule has no parameters.
20613 @subsection @code{Blocks}
20614 @cindex @code{Blocks} rule (for @command{gnatcheck})
20617 Flag each block statement.
20619 This rule has no parameters.
20621 @node Boolean_Relational_Operators
20622 @subsection @code{Boolean_Relational_Operators}
20623 @cindex @code{Boolean_Relational_Operators} rule (for @command{gnatcheck})
20626 Flag each call to a predefined relational operator (``<'', ``>'', ``<='',
20627 ``>='', ``='' and ``/='') for the predefined Boolean type.
20628 (This rule is useful in enforcing the SPARK language restrictions.)
20630 Calls to predefined relational operators of any type derived from
20631 @code{Standard.Boolean} are not detected. Calls to user-defined functions
20632 with these designators, and uses of operators that are renamings
20633 of the predefined relational operators for @code{Standard.Boolean},
20634 are likewise not detected.
20636 This rule has no parameters.
20639 @node Ceiling_Violations
20640 @subsection @code{Ceiling_Violations} (under construction, GLOBAL)
20641 @cindex @code{Ceiling_Violations} rule (for @command{gnatcheck})
20644 Flag invocations of a protected operation by a task whose priority exceeds
20645 the protected object's ceiling.
20647 As of @value{NOW}, this rule has the following limitations:
20652 We consider only pragmas Priority and Interrupt_Priority as means to define
20653 a task/protected operation priority. We do not consider the effect of using
20654 Ada.Dynamic_Priorities.Set_Priority procedure;
20657 We consider only base task priorities, and no priority inheritance. That is,
20658 we do not make a difference between calls issued during task activation and
20659 execution of the sequence of statements from task body;
20662 Any situation when the priority of protected operation caller is set by a
20663 dynamic expression (that is, the corresponding Priority or
20664 Interrupt_Priority pragma has a non-static expression as an argument) we
20665 treat as a priority inconsistency (and, therefore, detect this situation).
20669 At the moment the notion of the main subprogram is not implemented in
20670 gnatcheck, so any pragma Priority in a library level subprogram body (in case
20671 if this subprogram can be a main subprogram of a partition) changes the
20672 priority of an environment task. So if we have more then one such pragma in
20673 the set of processed sources, the pragma that is processed last, defines the
20674 priority of an environment task.
20676 This rule has no parameters.
20679 @node Controlled_Type_Declarations
20680 @subsection @code{Controlled_Type_Declarations}
20681 @cindex @code{Controlled_Type_Declarations} rule (for @command{gnatcheck})
20684 Flag all declarations of controlled types. A declaration of a private type
20685 is flagged if its full declaration declares a controlled type. A declaration
20686 of a derived type is flagged if its ancestor type is controlled. Subtype
20687 declarations are not checked. A declaration of a type that itself is not a
20688 descendant of a type declared in @code{Ada.Finalization} but has a controlled
20689 component is not checked.
20691 This rule has no parameters.
20695 @node Declarations_In_Blocks
20696 @subsection @code{Declarations_In_Blocks}
20697 @cindex @code{Declarations_In_Blocks} rule (for @command{gnatcheck})
20700 Flag all block statements containing local declarations. A @code{declare}
20701 block with an empty @i{declarative_part} or with a @i{declarative part}
20702 containing only pragmas and/or @code{use} clauses is not flagged.
20704 This rule has no parameters.
20707 @node Default_Parameters
20708 @subsection @code{Default_Parameters}
20709 @cindex @code{Default_Parameters} rule (for @command{gnatcheck})
20712 Flag all default expressions for subprogram parameters. Parameter
20713 declarations of formal and generic subprograms are also checked.
20715 This rule has no parameters.
20718 @node Discriminated_Records
20719 @subsection @code{Discriminated_Records}
20720 @cindex @code{Discriminated_Records} rule (for @command{gnatcheck})
20723 Flag all declarations of record types with discriminants. Only the
20724 declarations of record and record extension types are checked. Incomplete,
20725 formal, private, derived and private extension type declarations are not
20726 checked. Task and protected type declarations also are not checked.
20728 This rule has no parameters.
20731 @node Enumeration_Ranges_In_CASE_Statements
20732 @subsection @code{Enumeration_Ranges_In_CASE_Statements}
20733 @cindex @code{Enumeration_Ranges_In_CASE_Statements} (for @command{gnatcheck})
20736 Flag each use of a range of enumeration literals as a choice in a
20737 @code{case} statement.
20738 All forms for specifying a range (explicit ranges
20739 such as @code{A .. B}, subtype marks and @code{'Range} attributes) are flagged.
20740 An enumeration range is
20741 flagged even if contains exactly one enumeration value or no values at all. A
20742 type derived from an enumeration type is considered as an enumeration type.
20744 This rule helps prevent maintenance problems arising from adding an
20745 enumeration value to a type and having it implicitly handled by an existing
20746 @code{case} statement with an enumeration range that includes the new literal.
20748 This rule has no parameters.
20751 @node Exceptions_As_Control_Flow
20752 @subsection @code{Exceptions_As_Control_Flow}
20753 @cindex @code{Exceptions_As_Control_Flow} (for @command{gnatcheck})
20756 Flag each place where an exception is explicitly raised and handled in the
20757 same subprogram body. A @code{raise} statement in an exception handler,
20758 package body, task body or entry body is not flagged.
20760 The rule has no parameters.
20762 @node EXIT_Statements_With_No_Loop_Name
20763 @subsection @code{EXIT_Statements_With_No_Loop_Name}
20764 @cindex @code{EXIT_Statements_With_No_Loop_Name} (for @command{gnatcheck})
20767 Flag each @code{exit} statement that does not specify the name of the loop
20770 The rule has no parameters.
20773 @node Expanded_Loop_Exit_Names
20774 @subsection @code{Expanded_Loop_Exit_Names}
20775 @cindex @code{Expanded_Loop_Exit_Names} rule (for @command{gnatcheck})
20778 Flag all expanded loop names in @code{exit} statements.
20780 This rule has no parameters.
20782 @node Explicit_Full_Discrete_Ranges
20783 @subsection @code{Explicit_Full_Discrete_Ranges}
20784 @cindex @code{Explicit_Full_Discrete_Ranges} rule (for @command{gnatcheck})
20787 Flag each discrete range that has the form @code{A'First .. A'Last}.
20789 This rule has no parameters.
20791 @node Float_Equality_Checks
20792 @subsection @code{Float_Equality_Checks}
20793 @cindex @code{Float_Equality_Checks} rule (for @command{gnatcheck})
20796 Flag all calls to the predefined equality operations for floating-point types.
20797 Both ``@code{=}'' and ``@code{/=}'' operations are checked.
20798 User-defined equality operations are not flagged, nor are ``@code{=}''
20799 and ``@code{/=}'' operations for fixed-point types.
20801 This rule has no parameters.
20804 @node Forbidden_Pragmas
20805 @subsection @code{Forbidden_Pragmas}
20806 @cindex @code{Forbidden_Pragmas} rule (for @command{gnatcheck})
20809 Flag each use of the specified pragmas. The pragmas to be detected
20810 are named in the rule's parameters.
20812 This rule has the following parameters:
20815 @item For the @option{+R} option
20818 @item @emph{Pragma_Name}
20819 Adds the specified pragma to the set of pragmas to be
20820 checked and sets the checks for all the specified pragmas
20821 ON. @emph{Pragma_Name} is treated as a name of a pragma. If it
20822 does not correspond to any pragma name defined in the Ada
20823 standard or to the name of a GNAT-specific pragma defined
20824 in @ref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference
20825 Manual}, it is treated as the name of unknown pragma.
20828 All the GNAT-specific pragmas are detected; this sets
20829 the checks for all the specified pragmas ON.
20832 All pragmas are detected; this sets the rule ON.
20835 @item For the @option{-R} option
20837 @item @emph{Pragma_Name}
20838 Removes the specified pragma from the set of pragmas to be
20839 checked without affecting checks for
20840 other pragmas. @emph{Pragma_Name} is treated as a name
20841 of a pragma. If it does not correspond to any pragma
20842 defined in the Ada standard or to any name defined in
20843 @ref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference Manual},
20844 this option is treated as turning OFF detection of all unknown pragmas.
20847 Turn OFF detection of all GNAT-specific pragmas
20850 Clear the list of the pragmas to be detected and
20856 Parameters are not case sensitive. If @emph{Pragma_Name} does not have
20857 the syntax of an Ada identifier and therefore can not be considered
20858 as a pragma name, a diagnostic message is generated and the corresponding
20859 parameter is ignored.
20861 When more then one parameter is given in the same rule option, the parameters
20862 must be separated by a comma.
20864 If more then one option for this rule is specified for the @command{gnatcheck}
20865 call, a new option overrides the previous one(s).
20867 The @option{+R} option with no parameters turns the rule ON with the set of
20868 pragmas to be detected defined by the previous rule options.
20869 (By default this set is empty, so if the only option specified for the rule is
20870 @option{+RForbidden_Pragmas} (with
20871 no parameter), then the rule is enabled, but it does not detect anything).
20872 The @option{-R} option with no parameter turns the rule OFF, but it does not
20873 affect the set of pragmas to be detected.
20878 @node Function_Style_Procedures
20879 @subsection @code{Function_Style_Procedures}
20880 @cindex @code{Function_Style_Procedures} rule (for @command{gnatcheck})
20883 Flag each procedure that can be rewritten as a function. A procedure can be
20884 converted into a function if it has exactly one parameter of mode @code{out}
20885 and no parameters of mode @code{in out}. Procedure declarations,
20886 formal procedure declarations, and generic procedure declarations are always
20888 bodies and body stubs are flagged only if they do not have corresponding
20889 separate declarations. Procedure renamings and procedure instantiations are
20892 If a procedure can be rewritten as a function, but its @code{out} parameter is
20893 of a limited type, it is not flagged.
20895 Protected procedures are not flagged. Null procedures also are not flagged.
20897 This rule has no parameters.
20900 @node Generics_In_Subprograms
20901 @subsection @code{Generics_In_Subprograms}
20902 @cindex @code{Generics_In_Subprograms} rule (for @command{gnatcheck})
20905 Flag each declaration of a generic unit in a subprogram. Generic
20906 declarations in the bodies of generic subprograms are also flagged.
20907 A generic unit nested in another generic unit is not flagged.
20908 If a generic unit is
20909 declared in a local package that is declared in a subprogram body, the
20910 generic unit is flagged.
20912 This rule has no parameters.
20915 @node GOTO_Statements
20916 @subsection @code{GOTO_Statements}
20917 @cindex @code{GOTO_Statements} rule (for @command{gnatcheck})
20920 Flag each occurrence of a @code{goto} statement.
20922 This rule has no parameters.
20925 @node Implicit_IN_Mode_Parameters
20926 @subsection @code{Implicit_IN_Mode_Parameters}
20927 @cindex @code{Implicit_IN_Mode_Parameters} rule (for @command{gnatcheck})
20930 Flag each occurrence of a formal parameter with an implicit @code{in} mode.
20931 Note that @code{access} parameters, although they technically behave
20932 like @code{in} parameters, are not flagged.
20934 This rule has no parameters.
20937 @node Implicit_SMALL_For_Fixed_Point_Types
20938 @subsection @code{Implicit_SMALL_For_Fixed_Point_Types}
20939 @cindex @code{Implicit_SMALL_For_Fixed_Point_Types} rule (for @command{gnatcheck})
20942 Flag each fixed point type declaration that lacks an explicit
20943 representation clause to define its @code{'Small} value.
20944 Since @code{'Small} can be defined only for ordinary fixed point types,
20945 decimal fixed point type declarations are not checked.
20947 This rule has no parameters.
20950 @node Improperly_Located_Instantiations
20951 @subsection @code{Improperly_Located_Instantiations}
20952 @cindex @code{Improperly_Located_Instantiations} rule (for @command{gnatcheck})
20955 Flag all generic instantiations in library-level package specs
20956 (including library generic packages) and in all subprogram bodies.
20958 Instantiations in task and entry bodies are not flagged. Instantiations in the
20959 bodies of protected subprograms are flagged.
20961 This rule has no parameters.
20965 @node Improper_Returns
20966 @subsection @code{Improper_Returns}
20967 @cindex @code{Improper_Returns} rule (for @command{gnatcheck})
20970 Flag each explicit @code{return} statement in procedures, and
20971 multiple @code{return} statements in functions.
20972 Diagnostic messages are generated for all @code{return} statements
20973 in a procedure (thus each procedure must be written so that it
20974 returns implicitly at the end of its statement part),
20975 and for all @code{return} statements in a function after the first one.
20976 This rule supports the stylistic convention that each subprogram
20977 should have no more than one point of normal return.
20979 This rule has no parameters.
20982 @node Library_Level_Subprograms
20983 @subsection @code{Library_Level_Subprograms}
20984 @cindex @code{Library_Level_Subprograms} rule (for @command{gnatcheck})
20987 Flag all library-level subprograms (including generic subprogram instantiations).
20989 This rule has no parameters.
20992 @node Local_Packages
20993 @subsection @code{Local_Packages}
20994 @cindex @code{Local_Packages} rule (for @command{gnatcheck})
20997 Flag all local packages declared in package and generic package
20999 Local packages in bodies are not flagged.
21001 This rule has no parameters.
21004 @node Improperly_Called_Protected_Entries
21005 @subsection @code{Improperly_Called_Protected_Entries} (under construction, GLOBAL)
21006 @cindex @code{Improperly_Called_Protected_Entries} rule (for @command{gnatcheck})
21009 Flag each protected entry that can be called from more than one task.
21011 This rule has no parameters.
21014 @node Metrics_Violation
21015 @subsection @code{Metrics_Violation}
21016 @cindex @code{Metrics} rule (for @command{gnatcheck})
21019 This is an umbrella rule for a set of metrics-based checks. The parameters of
21020 the rule specify which metrics should be checked, and a bound (upper or lower,
21021 depending on the metric) for each specified metric. A construct is
21022 flagged if a specified metric can be computed for it, and the resulting value
21023 is higher then the upper bound (or less than the lower bound) specified.
21025 This rule has the following parameters:
21029 For the @option{+R} option:
21031 @item @i{Metric_Check_Name} < @i{LowerBound}
21032 Turns the check for the specified metric ON and specifies the lower bound
21033 for a given metric check
21035 @item @i{Metric_Check_Name} > @i{UpperBound}
21037 Turns the check for the specified metric ON and specifies the upper bound
21038 for a given metric check
21042 For the @option{-R} option:
21044 @item @i{Metric_Check_Name}
21045 Turns the check for the specified metric OFF
21050 Parameters are not case-sensitive. @i{Metric_Check_Name} must be
21051 the name of a metric supported by the @code{Metrics_Violation} rule
21052 (see the table below),
21053 otherwise the parameter is ignored. Whether the upper or lower bound
21054 is specified for a given check, depends on the metric. If a
21055 parameter for the @option{+R} option specifies an invalid limit, a
21056 warning is issued and the parameter is ignored.
21058 The @option{-R} option without parameters turns OFF all the previously enabled
21059 metric checks. the @option{+R} option without parameters turns ON all the
21060 metric checks that have been defined by previous @option{+R} options with
21061 valid parameters. @option{+R} option with a valid
21062 parameter also turns ON all the other metric checks that have been defined
21063 by previous @option{+R} options with valid parameters if they have been
21064 disabled by @option{-R} option without parameters.
21066 By default no metrics checks are ON, so the @option{+R} option without
21067 parameters actually does not specify any check.
21069 The following table shows the available metrics-based checks,
21070 including the constraint that must be satisfied by the bound that
21071 is specified for the check.
21073 @multitable {@code{Cyclomatic_Complexity}}{Cyclomatic complexity}{Positive integer}
21075 @headitem Check Name @tab Description @tab Bounds Value
21078 @item @b{Check Name} @tab @b{Description} @tab @b{Bounds Value}
21080 @c Above conditional code is workaround to bug in texi2html (Feb 2008)
21081 @item @code{Essential_Complexity} @tab Essential complexity @tab Positive integer
21082 @item @code{Cyclomatic_Complexity} @tab Cyclomatic complexity @tab Positive integer
21083 @item @code{LSLOC} @tab Logical Source Lines of Code @tab Positive integer
21087 The meaning and the computed values for all these metrics are exactly
21088 the same as for the corresponding metrics in @command{gnatmetric}.
21090 @emph{Example:} the rule
21092 +RMetrics_Violation: Cyclomatic_Complexity > 7
21095 means that all bodies with cyclomatic complexity exceeding 7 will be flagged.
21097 @node Misnamed_Identifiers
21098 @subsection @code{Misnamed_Identifiers}
21099 @cindex @code{Misnamed_Identifiers} rule (for @command{gnatcheck})
21102 Flag the declaration of each identifier that does not have a suffix
21103 corresponding to the kind of entity being declared.
21104 The following declarations are checked:
21111 constant declarations (but not number declarations)
21114 package renaming declarations (but not generic package renaming
21119 This rule may have parameters. When used without parameters, the rule enforces
21120 the following checks:
21124 type-defining names end with @code{_T}, unless the type is an access type,
21125 in which case the suffix must be @code{_A}
21127 constant names end with @code{_C}
21129 names defining package renamings end with @code{_R}
21133 For a private or incomplete type declaration the following checks are
21134 made for the defining name suffix:
21138 For an incomplete type declaration: if the corresponding full type
21139 declaration is available, the defining identifier from the full type
21140 declaration is checked, but the defining identifier from the incomplete type
21141 declaration is not; otherwise the defining identifier from the incomplete
21142 type declaration is checked against the suffix specified for type
21146 For a private type declaration (including private extensions), the defining
21147 identifier from the private type declaration is checked against the type
21148 suffix (even if the corresponding full declaration is an access type
21149 declaration), and the defining identifier from the corresponding full type
21150 declaration is not checked.
21154 For a deferred constant, the defining name in the corresponding full constant
21155 declaration is not checked.
21157 Defining names of formal types are not checked.
21159 The rule may have the following parameters:
21163 For the @option{+R} option:
21166 Sets the default listed above for all the names to be checked.
21168 @item Type_Suffix=@emph{string}
21169 Specifies the suffix for a type name.
21171 @item Access_Suffix=@emph{string}
21172 Specifies the suffix for an access type name. If
21173 this parameter is set, it overrides for access
21174 types the suffix set by the @code{Type_Suffix} parameter.
21176 @item Constant_Suffix=@emph{string}
21177 Specifies the suffix for a constant name.
21179 @item Renaming_Suffix=@emph{string}
21180 Specifies the suffix for a package renaming name.
21184 For the @option{-R} option:
21187 Remove all the suffixes specified for the
21188 identifier suffix checks, whether by default or
21189 as specified by other rule parameters. All the
21190 checks for this rule are disabled as a result.
21193 Removes the suffix specified for types. This
21194 disables checks for types but does not disable
21195 any other checks for this rule (including the
21196 check for access type names if @code{Access_Suffix} is
21199 @item Access_Suffix
21200 Removes the suffix specified for access types.
21201 This disables checks for access type names but
21202 does not disable any other checks for this rule.
21203 If @code{Type_Suffix} is set, access type names are
21204 checked as ordinary type names.
21206 @item Constant_Suffix
21207 Removes the suffix specified for constants. This
21208 disables checks for constant names but does not
21209 disable any other checks for this rule.
21211 @item Renaming_Suffix
21212 Removes the suffix specified for package
21213 renamings. This disables checks for package
21214 renamings but does not disable any other checks
21220 If more than one parameter is used, parameters must be separated by commas.
21222 If more than one option is specified for the @command{gnatcheck} invocation,
21223 a new option overrides the previous one(s).
21225 The @option{+RMisnamed_Identifiers} option (with no parameter) enables
21227 name suffixes specified by previous options used for this rule.
21229 The @option{-RMisnamed_Identifiers} option (with no parameter) disables
21230 all the checks but keeps
21231 all the suffixes specified by previous options used for this rule.
21233 The @emph{string} value must be a valid suffix for an Ada identifier (after
21234 trimming all the leading and trailing space characters, if any).
21235 Parameters are not case sensitive, except the @emph{string} part.
21237 If any error is detected in a rule parameter, the parameter is ignored.
21238 In such a case the options that are set for the rule are not
21243 @node Multiple_Entries_In_Protected_Definitions
21244 @subsection @code{Multiple_Entries_In_Protected_Definitions}
21245 @cindex @code{Multiple_Entries_In_Protected_Definitions} rule (for @command{gnatcheck})
21248 Flag each protected definition (i.e., each protected object/type declaration)
21249 that defines more than one entry.
21250 Diagnostic messages are generated for all the entry declarations
21251 except the first one. An entry family is counted as one entry. Entries from
21252 the private part of the protected definition are also checked.
21254 This rule has no parameters.
21257 @subsection @code{Name_Clashes}
21258 @cindex @code{Name_Clashes} rule (for @command{gnatcheck})
21261 Check that certain names are not used as defining identifiers. To activate
21262 this rule, you need to supply a reference to the dictionary file(s) as a rule
21263 parameter(s) (more then one dictionary file can be specified). If no
21264 dictionary file is set, this rule will not cause anything to be flagged.
21265 Only defining occurrences, not references, are checked.
21266 The check is not case-sensitive.
21268 This rule is enabled by default, but without setting any corresponding
21269 dictionary file(s); thus the default effect is to do no checks.
21271 A dictionary file is a plain text file. The maximum line length for this file
21272 is 1024 characters. If the line is longer then this limit, extra characters
21275 Each line can be either an empty line, a comment line, or a line containing
21276 a list of identifiers separated by space or HT characters.
21277 A comment is an Ada-style comment (from @code{--} to end-of-line).
21278 Identifiers must follow the Ada syntax for identifiers.
21279 A line containing one or more identifiers may end with a comment.
21281 @node Non_Qualified_Aggregates
21282 @subsection @code{Non_Qualified_Aggregates}
21283 @cindex @code{Non_Qualified_Aggregates} rule (for @command{gnatcheck})
21286 Flag each non-qualified aggregate.
21287 A non-qualified aggregate is an
21288 aggregate that is not the expression of a qualified expression. A
21289 string literal is not considered an aggregate, but an array
21290 aggregate of a string type is considered as a normal aggregate.
21291 Aggregates of anonymous array types are not flagged.
21293 This rule has no parameters.
21296 @node Non_Short_Circuit_Operators
21297 @subsection @code{Non_Short_Circuit_Operators}
21298 @cindex @code{Non_Short_Circuit_Operators} rule (for @command{gnatcheck})
21301 Flag all calls to predefined @code{and} and @code{or} operators for
21302 any boolean type. Calls to
21303 user-defined @code{and} and @code{or} and to operators defined by renaming
21304 declarations are not flagged. Calls to predefined @code{and} and @code{or}
21305 operators for modular types or boolean array types are not flagged.
21307 This rule has no parameters.
21311 @node Non_SPARK_Attributes
21312 @subsection @code{Non_SPARK_Attributes}
21313 @cindex @code{Non_SPARK_Attributes} rule (for @command{gnatcheck})
21316 The SPARK language defines the following subset of Ada 95 attribute
21317 designators as those that can be used in SPARK programs. The use of
21318 any other attribute is flagged.
21321 @item @code{'Adjacent}
21324 @item @code{'Ceiling}
21325 @item @code{'Component_Size}
21326 @item @code{'Compose}
21327 @item @code{'Copy_Sign}
21328 @item @code{'Delta}
21329 @item @code{'Denorm}
21330 @item @code{'Digits}
21331 @item @code{'Exponent}
21332 @item @code{'First}
21333 @item @code{'Floor}
21335 @item @code{'Fraction}
21337 @item @code{'Leading_Part}
21338 @item @code{'Length}
21339 @item @code{'Machine}
21340 @item @code{'Machine_Emax}
21341 @item @code{'Machine_Emin}
21342 @item @code{'Machine_Mantissa}
21343 @item @code{'Machine_Overflows}
21344 @item @code{'Machine_Radix}
21345 @item @code{'Machine_Rounds}
21348 @item @code{'Model}
21349 @item @code{'Model_Emin}
21350 @item @code{'Model_Epsilon}
21351 @item @code{'Model_Mantissa}
21352 @item @code{'Model_Small}
21353 @item @code{'Modulus}
21356 @item @code{'Range}
21357 @item @code{'Remainder}
21358 @item @code{'Rounding}
21359 @item @code{'Safe_First}
21360 @item @code{'Safe_Last}
21361 @item @code{'Scaling}
21362 @item @code{'Signed_Zeros}
21364 @item @code{'Small}
21366 @item @code{'Truncation}
21367 @item @code{'Unbiased_Rounding}
21369 @item @code{'Valid}
21373 This rule has no parameters.
21376 @node Non_Tagged_Derived_Types
21377 @subsection @code{Non_Tagged_Derived_Types}
21378 @cindex @code{Non_Tagged_Derived_Types} rule (for @command{gnatcheck})
21381 Flag all derived type declarations that do not have a record extension part.
21383 This rule has no parameters.
21387 @node Non_Visible_Exceptions
21388 @subsection @code{Non_Visible_Exceptions}
21389 @cindex @code{Non_Visible_Exceptions} rule (for @command{gnatcheck})
21392 Flag constructs leading to the possibility of propagating an exception
21393 out of the scope in which the exception is declared.
21394 Two cases are detected:
21398 An exception declaration in a subprogram body, task body or block
21399 statement is flagged if the body or statement does not contain a handler for
21400 that exception or a handler with an @code{others} choice.
21403 A @code{raise} statement in an exception handler of a subprogram body,
21404 task body or block statement is flagged if it (re)raises a locally
21405 declared exception. This may occur under the following circumstances:
21408 it explicitly raises a locally declared exception, or
21410 it does not specify an exception name (i.e., it is simply @code{raise;})
21411 and the enclosing handler contains a locally declared exception in its
21417 Renamings of local exceptions are not flagged.
21419 This rule has no parameters.
21422 @node Numeric_Literals
21423 @subsection @code{Numeric_Literals}
21424 @cindex @code{Numeric_Literals} rule (for @command{gnatcheck})
21427 Flag each use of a numeric literal in an index expression, and in any
21428 circumstance except for the following:
21432 a literal occurring in the initialization expression for a constant
21433 declaration or a named number declaration, or
21436 an integer literal that is less than or equal to a value
21437 specified by the @option{N} rule parameter.
21441 This rule may have the following parameters for the @option{+R} option:
21445 @emph{N} is an integer literal used as the maximal value that is not flagged
21446 (i.e., integer literals not exceeding this value are allowed)
21449 All integer literals are flagged
21453 If no parameters are set, the maximum unflagged value is 1.
21455 The last specified check limit (or the fact that there is no limit at
21456 all) is used when multiple @option{+R} options appear.
21458 The @option{-R} option for this rule has no parameters.
21459 It disables the rule but retains the last specified maximum unflagged value.
21460 If the @option{+R} option subsequently appears, this value is used as the
21461 threshold for the check.
21464 @node OTHERS_In_Aggregates
21465 @subsection @code{OTHERS_In_Aggregates}
21466 @cindex @code{OTHERS_In_Aggregates} rule (for @command{gnatcheck})
21469 Flag each use of an @code{others} choice in extension aggregates.
21470 In record and array aggregates, an @code{others} choice is flagged unless
21471 it is used to refer to all components, or to all but one component.
21473 If, in case of a named array aggregate, there are two associations, one
21474 with an @code{others} choice and another with a discrete range, the
21475 @code{others} choice is flagged even if the discrete range specifies
21476 exactly one component; for example, @code{(1..1 => 0, others => 1)}.
21478 This rule has no parameters.
21480 @node OTHERS_In_CASE_Statements
21481 @subsection @code{OTHERS_In_CASE_Statements}
21482 @cindex @code{OTHERS_In_CASE_Statements} rule (for @command{gnatcheck})
21485 Flag any use of an @code{others} choice in a @code{case} statement.
21487 This rule has no parameters.
21489 @node OTHERS_In_Exception_Handlers
21490 @subsection @code{OTHERS_In_Exception_Handlers}
21491 @cindex @code{OTHERS_In_Exception_Handlers} rule (for @command{gnatcheck})
21494 Flag any use of an @code{others} choice in an exception handler.
21496 This rule has no parameters.
21499 @node Outer_Loop_Exits
21500 @subsection @code{Outer_Loop_Exits}
21501 @cindex @code{Outer_Loop_Exits} rule (for @command{gnatcheck})
21504 Flag each @code{exit} statement containing a loop name that is not the name
21505 of the immediately enclosing @code{loop} statement.
21507 This rule has no parameters.
21510 @node Overloaded_Operators
21511 @subsection @code{Overloaded_Operators}
21512 @cindex @code{Overloaded_Operators} rule (for @command{gnatcheck})
21515 Flag each function declaration that overloads an operator symbol.
21516 A function body is checked only if the body does not have a
21517 separate spec. Formal functions are also checked. For a
21518 renaming declaration, only renaming-as-declaration is checked
21520 This rule has no parameters.
21523 @node Overly_Nested_Control_Structures
21524 @subsection @code{Overly_Nested_Control_Structures}
21525 @cindex @code{Overly_Nested_Control_Structures} rule (for @command{gnatcheck})
21528 Flag each control structure whose nesting level exceeds the value provided
21529 in the rule parameter.
21531 The control structures checked are the following:
21534 @item @code{if} statement
21535 @item @code{case} statement
21536 @item @code{loop} statement
21537 @item Selective accept statement
21538 @item Timed entry call statement
21539 @item Conditional entry call
21540 @item Asynchronous select statement
21544 The rule has the following parameter for the @option{+R} option:
21548 Positive integer specifying the maximal control structure nesting
21549 level that is not flagged
21553 If the parameter for the @option{+R} option is not specified or
21554 if it is not a positive integer, @option{+R} option is ignored.
21556 If more then one option is specified for the gnatcheck call, the later option and
21557 new parameter override the previous one(s).
21560 @node Parameters_Out_Of_Order
21561 @subsection @code{Parameters_Out_Of_Order}
21562 @cindex @code{Parameters_Out_Of_Order} rule (for @command{gnatcheck})
21565 Flag each subprogram and entry declaration whose formal parameters are not
21566 ordered according to the following scheme:
21570 @item @code{in} and @code{access} parameters first,
21571 then @code{in out} parameters,
21572 and then @code{out} parameters;
21574 @item for @code{in} mode, parameters with default initialization expressions
21579 Only the first violation of the described order is flagged.
21581 The following constructs are checked:
21584 @item subprogram declarations (including null procedures);
21585 @item generic subprogram declarations;
21586 @item formal subprogram declarations;
21587 @item entry declarations;
21588 @item subprogram bodies and subprogram body stubs that do not
21589 have separate specifications
21593 Subprogram renamings are not checked.
21595 This rule has no parameters.
21598 @node Positional_Actuals_For_Defaulted_Generic_Parameters
21599 @subsection @code{Positional_Actuals_For_Defaulted_Generic_Parameters}
21600 @cindex @code{Positional_Actuals_For_Defaulted_Generic_Parameters} rule (for @command{gnatcheck})
21603 Flag each generic actual parameter corresponding to a generic formal
21604 parameter with a default initialization, if positional notation is used.
21606 This rule has no parameters.
21608 @node Positional_Actuals_For_Defaulted_Parameters
21609 @subsection @code{Positional_Actuals_For_Defaulted_Parameters}
21610 @cindex @code{Positional_Actuals_For_Defaulted_Parameters} rule (for @command{gnatcheck})
21613 Flag each actual parameter to a subprogram or entry call where the
21614 corresponding formal parameter has a default expression, if positional
21617 This rule has no parameters.
21619 @node Positional_Components
21620 @subsection @code{Positional_Components}
21621 @cindex @code{Positional_Components} rule (for @command{gnatcheck})
21624 Flag each array, record and extension aggregate that includes positional
21627 This rule has no parameters.
21630 @node Positional_Generic_Parameters
21631 @subsection @code{Positional_Generic_Parameters}
21632 @cindex @code{Positional_Generic_Parameters} rule (for @command{gnatcheck})
21635 Flag each instantiation using positional parameter notation.
21637 This rule has no parameters.
21640 @node Positional_Parameters
21641 @subsection @code{Positional_Parameters}
21642 @cindex @code{Positional_Parameters} rule (for @command{gnatcheck})
21645 Flag each subprogram or entry call using positional parameter notation,
21646 except for the following:
21650 Invocations of prefix or infix operators are not flagged
21652 If the called subprogram or entry has only one formal parameter,
21653 the call is not flagged;
21655 If a subprogram call uses the @emph{Object.Operation} notation, then
21658 the first parameter (that is, @emph{Object}) is not flagged;
21660 if the called subprogram has only two parameters, the second parameter
21661 of the call is not flagged;
21666 This rule has no parameters.
21671 @node Predefined_Numeric_Types
21672 @subsection @code{Predefined_Numeric_Types}
21673 @cindex @code{Predefined_Numeric_Types} rule (for @command{gnatcheck})
21676 Flag each explicit use of the name of any numeric type or subtype defined
21677 in package @code{Standard}.
21679 The rationale for this rule is to detect when the
21680 program may depend on platform-specific characteristics of the implementation
21681 of the predefined numeric types. Note that this rule is over-pessimistic;
21682 for example, a program that uses @code{String} indexing
21683 likely needs a variable of type @code{Integer}.
21684 Another example is the flagging of predefined numeric types with explicit
21687 @smallexample @c ada
21688 subtype My_Integer is Integer range Left .. Right;
21689 Vy_Var : My_Integer;
21693 This rule detects only numeric types and subtypes defined in
21694 @code{Standard}. The use of numeric types and subtypes defined in other
21695 predefined packages (such as @code{System.Any_Priority} or
21696 @code{Ada.Text_IO.Count}) is not flagged
21698 This rule has no parameters.
21702 @node Raising_External_Exceptions
21703 @subsection @code{Raising_External_Exceptions}
21704 @cindex @code{Raising_External_Exceptions} rule (for @command{gnatcheck})
21707 Flag any @code{raise} statement, in a program unit declared in a library
21708 package or in a generic library package, for an exception that is
21709 neither a predefined exception nor an exception that is also declared (or
21710 renamed) in the visible part of the package.
21712 This rule has no parameters.
21716 @node Raising_Predefined_Exceptions
21717 @subsection @code{Raising_Predefined_Exceptions}
21718 @cindex @code{Raising_Predefined_Exceptions} rule (for @command{gnatcheck})
21721 Flag each @code{raise} statement that raises a predefined exception
21722 (i.e., one of the exceptions @code{Constraint_Error}, @code{Numeric_Error},
21723 @code{Program_Error}, @code{Storage_Error}, or @code{Tasking_Error}).
21725 This rule has no parameters.
21727 @node Separate_Numeric_Error_Handlers
21728 @subsection @code{Separate_Numeric_Error_Handlers}
21729 @cindex @code{Separate_Numeric_Error_Handlers} rule (for @command{gnatcheck})
21732 Flags each exception handler that contains a choice for
21733 the predefined @code{Constraint_Error} exception, but does not contain
21734 the choice for the predefined @code{Numeric_Error} exception, or
21735 that contains the choice for @code{Numeric_Error}, but does not contain the
21736 choice for @code{Constraint_Error}.
21738 This rule has no parameters.
21742 @subsection @code{Recursion} (under construction, GLOBAL)
21743 @cindex @code{Recursion} rule (for @command{gnatcheck})
21746 Flag recursive subprograms (cycles in the call graph). Declarations, and not
21747 calls, of recursive subprograms are detected.
21749 This rule has no parameters.
21753 @node Side_Effect_Functions
21754 @subsection @code{Side_Effect_Functions} (under construction, GLOBAL)
21755 @cindex @code{Side_Effect_Functions} rule (for @command{gnatcheck})
21758 Flag functions with side effects.
21760 We define a side effect as changing any data object that is not local for the
21761 body of this function.
21763 At the moment, we do NOT consider a side effect any input-output operations
21764 (changing a state or a content of any file).
21766 We do not consider protected functions for this rule (???)
21768 There are the following sources of side effect:
21771 @item Explicit (or direct) side-effect:
21775 direct assignment to a non-local variable;
21778 direct call to an entity that is known to change some data object that is
21779 not local for the body of this function (Note, that if F1 calls F2 and F2
21780 does have a side effect, this does not automatically mean that F1 also
21781 have a side effect, because it may be the case that F2 is declared in
21782 F1's body and it changes some data object that is global for F2, but
21786 @item Indirect side-effect:
21789 Subprogram calls implicitly issued by:
21792 computing initialization expressions from type declarations as a part
21793 of object elaboration or allocator evaluation;
21795 computing implicit parameters of subprogram or entry calls or generic
21800 activation of a task that change some non-local data object (directly or
21804 elaboration code of a package that is a result of a package instantiation;
21807 controlled objects;
21810 @item Situations when we can suspect a side-effect, but the full static check
21811 is either impossible or too hard:
21814 assignment to access variables or to the objects pointed by access
21818 call to a subprogram pointed by access-to-subprogram value
21826 This rule has no parameters.
21830 @subsection @code{Slices}
21831 @cindex @code{Slices} rule (for @command{gnatcheck})
21834 Flag all uses of array slicing
21836 This rule has no parameters.
21839 @node Unassigned_OUT_Parameters
21840 @subsection @code{Unassigned_OUT_Parameters}
21841 @cindex @code{Unassigned_OUT_Parameters} rule (for @command{gnatcheck})
21844 Flags procedures' @code{out} parameters that are not assigned, and
21845 identifies the contexts in which the assignments are missing.
21847 An @code{out} parameter is flagged in the statements in the procedure
21848 body's handled sequence of statements (before the procedure body's
21849 @code{exception} part, if any) if this sequence of statements contains
21850 no assignments to the parameter.
21852 An @code{out} parameter is flagged in an exception handler in the exception
21853 part of the procedure body's handled sequence of statements if the handler
21854 contains no assignment to the parameter.
21856 Bodies of generic procedures are also considered.
21858 The following are treated as assignments to an @code{out} parameter:
21862 an assignment statement, with the parameter or some component as the target;
21865 passing the parameter (or one of its components) as an @code{out} or
21866 @code{in out} parameter.
21870 This rule does not have any parameters.
21874 @node Uncommented_BEGIN_In_Package_Bodies
21875 @subsection @code{Uncommented_BEGIN_In_Package_Bodies}
21876 @cindex @code{Uncommented_BEGIN_In_Package_Bodies} rule (for @command{gnatcheck})
21879 Flags each package body with declarations and a statement part that does not
21880 include a trailing comment on the line containing the @code{begin} keyword;
21881 this trailing comment needs to specify the package name and nothing else.
21882 The @code{begin} is not flagged if the package body does not
21883 contain any declarations.
21885 If the @code{begin} keyword is placed on the
21886 same line as the last declaration or the first statement, it is flagged
21887 independently of whether the line contains a trailing comment. The
21888 diagnostic message is attached to the line containing the first statement.
21890 This rule has no parameters.
21893 @node Unconstrained_Array_Returns
21894 @subsection @code{Unconstrained_Array_Returns}
21895 @cindex @code{Unconstrained_Array_Returns} rule (for @command{gnatcheck})
21898 Flag each function returning an unconstrained array. Function declarations,
21899 function bodies (and body stubs) having no separate specifications,
21900 and generic function instantiations are checked.
21901 Generic function declarations, function calls and function renamings are
21904 This rule has no parameters.
21906 @node Universal_Ranges
21907 @subsection @code{Universal_Ranges}
21908 @cindex @code{Universal_Ranges} rule (for @command{gnatcheck})
21911 Flag discrete ranges that are a part of an index constraint, constrained
21912 array definition, or @code{for}-loop parameter specification, and whose bounds
21913 are both of type @i{universal_integer}. Ranges that have at least one
21914 bound of a specific type (such as @code{1 .. N}, where @code{N} is a variable
21915 or an expression of non-universal type) are not flagged.
21917 This rule has no parameters.
21920 @node Unnamed_Blocks_And_Loops
21921 @subsection @code{Unnamed_Blocks_And_Loops}
21922 @cindex @code{Unnamed_Blocks_And_Loops} rule (for @command{gnatcheck})
21925 Flag each unnamed block statement and loop statement.
21927 The rule has no parameters.
21932 @node Unused_Subprograms
21933 @subsection @code{Unused_Subprograms} (under construction, GLOBAL)
21934 @cindex @code{Unused_Subprograms} rule (for @command{gnatcheck})
21937 Flag all unused subprograms.
21939 This rule has no parameters.
21945 @node USE_PACKAGE_Clauses
21946 @subsection @code{USE_PACKAGE_Clauses}
21947 @cindex @code{USE_PACKAGE_Clauses} rule (for @command{gnatcheck})
21950 Flag all @code{use} clauses for packages; @code{use type} clauses are
21953 This rule has no parameters.
21957 @node Volatile_Objects_Without_Address_Clauses
21958 @subsection @code{Volatile_Objects_Without_Address_Clauses}
21959 @cindex @code{Volatile_Objects_Without_Address_Clauses} rule (for @command{gnatcheck})
21962 Flag each volatile object that does not have an address clause.
21964 The following check is made: if the pragma @code{Volatile} is applied to a
21965 data object or to its type, then an address clause must
21966 be supplied for this object.
21968 This rule does not check the components of data objects,
21969 array components that are volatile as a result of the pragma
21970 @code{Volatile_Components}, or objects that are volatile because
21971 they are atomic as a result of pragmas @code{Atomic} or
21972 @code{Atomic_Components}.
21974 Only variable declarations, and not constant declarations, are checked.
21976 This rule has no parameters.
21979 @c *********************************
21980 @node Creating Sample Bodies Using gnatstub
21981 @chapter Creating Sample Bodies Using @command{gnatstub}
21985 @command{gnatstub} creates body stubs, that is, empty but compilable bodies
21986 for library unit declarations.
21988 Note: to invoke @code{gnatstub} with a project file, use the @code{gnat}
21989 driver (see @ref{The GNAT Driver and Project Files}).
21991 To create a body stub, @command{gnatstub} has to compile the library
21992 unit declaration. Therefore, bodies can be created only for legal
21993 library units. Moreover, if a library unit depends semantically upon
21994 units located outside the current directory, you have to provide
21995 the source search path when calling @command{gnatstub}, see the description
21996 of @command{gnatstub} switches below.
21999 * Running gnatstub::
22000 * Switches for gnatstub::
22003 @node Running gnatstub
22004 @section Running @command{gnatstub}
22007 @command{gnatstub} has the command-line interface of the form
22010 $ gnatstub @ovar{switches} @var{filename} @ovar{directory}
22017 is the name of the source file that contains a library unit declaration
22018 for which a body must be created. The file name may contain the path
22020 The file name does not have to follow the GNAT file name conventions. If the
22022 does not follow GNAT file naming conventions, the name of the body file must
22024 explicitly as the value of the @option{^-o^/BODY=^@var{body-name}} option.
22025 If the file name follows the GNAT file naming
22026 conventions and the name of the body file is not provided,
22029 of the body file from the argument file name by replacing the @file{.ads}
22031 with the @file{.adb} suffix.
22034 indicates the directory in which the body stub is to be placed (the default
22039 is an optional sequence of switches as described in the next section
22042 @node Switches for gnatstub
22043 @section Switches for @command{gnatstub}
22049 @cindex @option{^-f^/FULL^} (@command{gnatstub})
22050 If the destination directory already contains a file with the name of the
22052 for the argument spec file, replace it with the generated body stub.
22054 @item ^-hs^/HEADER=SPEC^
22055 @cindex @option{^-hs^/HEADER=SPEC^} (@command{gnatstub})
22056 Put the comment header (i.e., all the comments preceding the
22057 compilation unit) from the source of the library unit declaration
22058 into the body stub.
22060 @item ^-hg^/HEADER=GENERAL^
22061 @cindex @option{^-hg^/HEADER=GENERAL^} (@command{gnatstub})
22062 Put a sample comment header into the body stub.
22064 @item ^--header-file=@var{filename}^/FROM_HEADER_FILE=@var{filename}^
22065 @cindex @option{^--header-file^/FROM_HEADER_FILE=^} (@command{gnatstub})
22066 Use the content of the file as the comment header for a generated body stub.
22070 @cindex @option{-IDIR} (@command{gnatstub})
22072 @cindex @option{-I-} (@command{gnatstub})
22075 @item /NOCURRENT_DIRECTORY
22076 @cindex @option{/NOCURRENT_DIRECTORY} (@command{gnatstub})
22078 ^These switches have ^This switch has^ the same meaning as in calls to
22080 ^They define ^It defines ^ the source search path in the call to
22081 @command{gcc} issued
22082 by @command{gnatstub} to compile an argument source file.
22084 @item ^-gnatec^/CONFIGURATION_PRAGMAS_FILE=^@var{PATH}
22085 @cindex @option{^-gnatec^/CONFIGURATION_PRAGMAS_FILE^} (@command{gnatstub})
22086 This switch has the same meaning as in calls to @command{gcc}.
22087 It defines the additional configuration file to be passed to the call to
22088 @command{gcc} issued
22089 by @command{gnatstub} to compile an argument source file.
22091 @item ^-gnatyM^/MAX_LINE_LENGTH=^@var{n}
22092 @cindex @option{^-gnatyM^/MAX_LINE_LENGTH^} (@command{gnatstub})
22093 (@var{n} is a non-negative integer). Set the maximum line length in the
22094 body stub to @var{n}; the default is 79. The maximum value that can be
22095 specified is 32767. Note that in the special case of configuration
22096 pragma files, the maximum is always 32767 regardless of whether or
22097 not this switch appears.
22099 @item ^-gnaty^/STYLE_CHECKS=^@var{n}
22100 @cindex @option{^-gnaty^/STYLE_CHECKS=^} (@command{gnatstub})
22101 (@var{n} is a non-negative integer from 1 to 9). Set the indentation level in
22102 the generated body sample to @var{n}.
22103 The default indentation is 3.
22105 @item ^-gnatyo^/ORDERED_SUBPROGRAMS^
22106 @cindex @option{^-gnato^/ORDERED_SUBPROGRAMS^} (@command{gnatstub})
22107 Order local bodies alphabetically. (By default local bodies are ordered
22108 in the same way as the corresponding local specs in the argument spec file.)
22110 @item ^-i^/INDENTATION=^@var{n}
22111 @cindex @option{^-i^/INDENTATION^} (@command{gnatstub})
22112 Same as @option{^-gnaty^/STYLE_CHECKS=^@var{n}}
22114 @item ^-k^/TREE_FILE=SAVE^
22115 @cindex @option{^-k^/TREE_FILE=SAVE^} (@command{gnatstub})
22116 Do not remove the tree file (i.e., the snapshot of the compiler internal
22117 structures used by @command{gnatstub}) after creating the body stub.
22119 @item ^-l^/LINE_LENGTH=^@var{n}
22120 @cindex @option{^-l^/LINE_LENGTH^} (@command{gnatstub})
22121 Same as @option{^-gnatyM^/MAX_LINE_LENGTH=^@var{n}}
22123 @item ^-o^/BODY=^@var{body-name}
22124 @cindex @option{^-o^/BODY^} (@command{gnatstub})
22125 Body file name. This should be set if the argument file name does not
22127 the GNAT file naming
22128 conventions. If this switch is omitted the default name for the body will be
22130 from the argument file name according to the GNAT file naming conventions.
22133 @cindex @option{^-q^/QUIET^} (@command{gnatstub})
22134 Quiet mode: do not generate a confirmation when a body is
22135 successfully created, and do not generate a message when a body is not
22139 @item ^-r^/TREE_FILE=REUSE^
22140 @cindex @option{^-r^/TREE_FILE=REUSE^} (@command{gnatstub})
22141 Reuse the tree file (if it exists) instead of creating it. Instead of
22142 creating the tree file for the library unit declaration, @command{gnatstub}
22143 tries to find it in the current directory and use it for creating
22144 a body. If the tree file is not found, no body is created. This option
22145 also implies @option{^-k^/SAVE^}, whether or not
22146 the latter is set explicitly.
22148 @item ^-t^/TREE_FILE=OVERWRITE^
22149 @cindex @option{^-t^/TREE_FILE=OVERWRITE^} (@command{gnatstub})
22150 Overwrite the existing tree file. If the current directory already
22151 contains the file which, according to the GNAT file naming rules should
22152 be considered as a tree file for the argument source file,
22154 will refuse to create the tree file needed to create a sample body
22155 unless this option is set.
22157 @item ^-v^/VERBOSE^
22158 @cindex @option{^-v^/VERBOSE^} (@command{gnatstub})
22159 Verbose mode: generate version information.
22163 @node Other Utility Programs
22164 @chapter Other Utility Programs
22167 This chapter discusses some other utility programs available in the Ada
22171 * Using Other Utility Programs with GNAT::
22172 * The External Symbol Naming Scheme of GNAT::
22173 * Converting Ada Files to html with gnathtml::
22174 * Installing gnathtml::
22181 @node Using Other Utility Programs with GNAT
22182 @section Using Other Utility Programs with GNAT
22185 The object files generated by GNAT are in standard system format and in
22186 particular the debugging information uses this format. This means
22187 programs generated by GNAT can be used with existing utilities that
22188 depend on these formats.
22191 In general, any utility program that works with C will also often work with
22192 Ada programs generated by GNAT. This includes software utilities such as
22193 gprof (a profiling program), @code{gdb} (the FSF debugger), and utilities such
22197 @node The External Symbol Naming Scheme of GNAT
22198 @section The External Symbol Naming Scheme of GNAT
22201 In order to interpret the output from GNAT, when using tools that are
22202 originally intended for use with other languages, it is useful to
22203 understand the conventions used to generate link names from the Ada
22206 All link names are in all lowercase letters. With the exception of library
22207 procedure names, the mechanism used is simply to use the full expanded
22208 Ada name with dots replaced by double underscores. For example, suppose
22209 we have the following package spec:
22211 @smallexample @c ada
22222 The variable @code{MN} has a full expanded Ada name of @code{QRS.MN}, so
22223 the corresponding link name is @code{qrs__mn}.
22225 Of course if a @code{pragma Export} is used this may be overridden:
22227 @smallexample @c ada
22232 pragma Export (Var1, C, External_Name => "var1_name");
22234 pragma Export (Var2, C, Link_Name => "var2_link_name");
22241 In this case, the link name for @var{Var1} is whatever link name the
22242 C compiler would assign for the C function @var{var1_name}. This typically
22243 would be either @var{var1_name} or @var{_var1_name}, depending on operating
22244 system conventions, but other possibilities exist. The link name for
22245 @var{Var2} is @var{var2_link_name}, and this is not operating system
22249 One exception occurs for library level procedures. A potential ambiguity
22250 arises between the required name @code{_main} for the C main program,
22251 and the name we would otherwise assign to an Ada library level procedure
22252 called @code{Main} (which might well not be the main program).
22254 To avoid this ambiguity, we attach the prefix @code{_ada_} to such
22255 names. So if we have a library level procedure such as
22257 @smallexample @c ada
22260 procedure Hello (S : String);
22266 the external name of this procedure will be @var{_ada_hello}.
22269 @node Converting Ada Files to html with gnathtml
22270 @section Converting Ada Files to HTML with @code{gnathtml}
22273 This @code{Perl} script allows Ada source files to be browsed using
22274 standard Web browsers. For installation procedure, see the section
22275 @xref{Installing gnathtml}.
22277 Ada reserved keywords are highlighted in a bold font and Ada comments in
22278 a blue font. Unless your program was compiled with the gcc @option{-gnatx}
22279 switch to suppress the generation of cross-referencing information, user
22280 defined variables and types will appear in a different color; you will
22281 be able to click on any identifier and go to its declaration.
22283 The command line is as follow:
22285 $ perl gnathtml.pl @ovar{^switches^options^} @var{ada-files}
22289 You can pass it as many Ada files as you want. @code{gnathtml} will generate
22290 an html file for every ada file, and a global file called @file{index.htm}.
22291 This file is an index of every identifier defined in the files.
22293 The available ^switches^options^ are the following ones:
22297 @cindex @option{-83} (@code{gnathtml})
22298 Only the Ada 83 subset of keywords will be highlighted.
22300 @item -cc @var{color}
22301 @cindex @option{-cc} (@code{gnathtml})
22302 This option allows you to change the color used for comments. The default
22303 value is green. The color argument can be any name accepted by html.
22306 @cindex @option{-d} (@code{gnathtml})
22307 If the Ada files depend on some other files (for instance through
22308 @code{with} clauses, the latter files will also be converted to html.
22309 Only the files in the user project will be converted to html, not the files
22310 in the run-time library itself.
22313 @cindex @option{-D} (@code{gnathtml})
22314 This command is the same as @option{-d} above, but @command{gnathtml} will
22315 also look for files in the run-time library, and generate html files for them.
22317 @item -ext @var{extension}
22318 @cindex @option{-ext} (@code{gnathtml})
22319 This option allows you to change the extension of the generated HTML files.
22320 If you do not specify an extension, it will default to @file{htm}.
22323 @cindex @option{-f} (@code{gnathtml})
22324 By default, gnathtml will generate html links only for global entities
22325 ('with'ed units, global variables and types,@dots{}). If you specify
22326 @option{-f} on the command line, then links will be generated for local
22329 @item -l @var{number}
22330 @cindex @option{-l} (@code{gnathtml})
22331 If this ^switch^option^ is provided and @var{number} is not 0, then
22332 @code{gnathtml} will number the html files every @var{number} line.
22335 @cindex @option{-I} (@code{gnathtml})
22336 Specify a directory to search for library files (@file{.ALI} files) and
22337 source files. You can provide several -I switches on the command line,
22338 and the directories will be parsed in the order of the command line.
22341 @cindex @option{-o} (@code{gnathtml})
22342 Specify the output directory for html files. By default, gnathtml will
22343 saved the generated html files in a subdirectory named @file{html/}.
22345 @item -p @var{file}
22346 @cindex @option{-p} (@code{gnathtml})
22347 If you are using Emacs and the most recent Emacs Ada mode, which provides
22348 a full Integrated Development Environment for compiling, checking,
22349 running and debugging applications, you may use @file{.gpr} files
22350 to give the directories where Emacs can find sources and object files.
22352 Using this ^switch^option^, you can tell gnathtml to use these files.
22353 This allows you to get an html version of your application, even if it
22354 is spread over multiple directories.
22356 @item -sc @var{color}
22357 @cindex @option{-sc} (@code{gnathtml})
22358 This ^switch^option^ allows you to change the color used for symbol
22360 The default value is red. The color argument can be any name accepted by html.
22362 @item -t @var{file}
22363 @cindex @option{-t} (@code{gnathtml})
22364 This ^switch^option^ provides the name of a file. This file contains a list of
22365 file names to be converted, and the effect is exactly as though they had
22366 appeared explicitly on the command line. This
22367 is the recommended way to work around the command line length limit on some
22372 @node Installing gnathtml
22373 @section Installing @code{gnathtml}
22376 @code{Perl} needs to be installed on your machine to run this script.
22377 @code{Perl} is freely available for almost every architecture and
22378 Operating System via the Internet.
22380 On Unix systems, you may want to modify the first line of the script
22381 @code{gnathtml}, to explicitly tell the Operating system where Perl
22382 is. The syntax of this line is:
22384 #!full_path_name_to_perl
22388 Alternatively, you may run the script using the following command line:
22391 $ perl gnathtml.pl @ovar{switches} @var{files}
22400 The GNAT distribution provides an Ada 95 template for the HP Language
22401 Sensitive Editor (LSE), a component of DECset. In order to
22402 access it, invoke LSE with the qualifier /ENVIRONMENT=GNU:[LIB]ADA95.ENV.
22409 GNAT supports The HP Performance Coverage Analyzer (PCA), a component
22410 of DECset. To use it proceed as outlined under ``HELP PCA'', except for running
22411 the collection phase with the /DEBUG qualifier.
22414 $ GNAT MAKE /DEBUG <PROGRAM_NAME>
22415 $ DEFINE LIB$DEBUG PCA$COLLECTOR
22416 $ RUN/DEBUG <PROGRAM_NAME>
22422 @c ******************************
22423 @node Code Coverage and Profiling
22424 @chapter Code Coverage and Profiling
22425 @cindex Code Coverage
22429 This chapter describes how to use @code{gcov} - coverage testing tool - and
22430 @code{gprof} - profiler tool - on your Ada programs.
22433 * Code Coverage of Ada Programs using gcov::
22434 * Profiling an Ada Program using gprof::
22437 @node Code Coverage of Ada Programs using gcov
22438 @section Code Coverage of Ada Programs using gcov
22440 @cindex -fprofile-arcs
22441 @cindex -ftest-coverage
22443 @cindex Code Coverage
22446 @code{gcov} is a test coverage program: it analyzes the execution of a given
22447 program on selected tests, to help you determine the portions of the program
22448 that are still untested.
22450 @code{gcov} is part of the GCC suite, and is described in detail in the GCC
22451 User's Guide. You can refer to this documentation for a more complete
22454 This chapter provides a quick startup guide, and
22455 details some Gnat-specific features.
22458 * Quick startup guide::
22462 @node Quick startup guide
22463 @subsection Quick startup guide
22465 In order to perform coverage analysis of a program using @code{gcov}, 3
22470 Code instrumentation during the compilation process
22472 Execution of the instrumented program
22474 Execution of the @code{gcov} tool to generate the result.
22477 The code instrumentation needed by gcov is created at the object level:
22478 The source code is not modified in any way, because the instrumentation code is
22479 inserted by gcc during the compilation process. To compile your code with code
22480 coverage activated, you need to recompile your whole project using the
22482 @code{-fprofile-arcs} and @code{-ftest-coverage}, and link it using
22483 @code{-fprofile-arcs}.
22486 $ gnatmake -P my_project.gpr -f -cargs -fprofile-arcs -ftest-coverage \
22487 -largs -fprofile-arcs
22490 This compilation process will create @file{.gcno} files together with
22491 the usual object files.
22493 Once the program is compiled with coverage instrumentation, you can
22494 run it as many times as needed - on portions of a test suite for
22495 example. The first execution will produce @file{.gcda} files at the
22496 same location as the @file{.gcno} files. The following executions
22497 will update those files, so that a cumulative result of the covered
22498 portions of the program is generated.
22500 Finally, you need to call the @code{gcov} tool. The different options of
22501 @code{gcov} are available in the GCC User's Guide, section 'Invoking gcov'.
22503 This will create annotated source files with a @file{.gcov} extension:
22504 @file{my_main.adb} file will be analysed in @file{my_main.adb.gcov}.
22506 @node Gnat specifics
22507 @subsection Gnat specifics
22509 Because Ada semantics, portions of the source code may be shared among
22510 several object files. This is the case for example when generics are
22511 involved, when inlining is active or when declarations generate initialisation
22512 calls. In order to take
22513 into account this shared code, you need to call @code{gcov} on all
22514 source files of the tested program at once.
22516 The list of source files might exceed the system's maximum command line
22517 length. In order to bypass this limitation, a new mechanism has been
22518 implemented in @code{gcov}: you can now list all your project's files into a
22519 text file, and provide this file to gcov as a parameter, preceded by a @@
22520 (e.g. @samp{gcov @@mysrclist.txt}).
22522 @node Profiling an Ada Program using gprof
22523 @section Profiling an Ada Program using gprof
22529 This section is not meant to be an exhaustive documentation of @code{gprof}.
22530 Full documentation for it can be found in the GNU Profiler User's Guide
22531 documentation that is part of this GNAT distribution.
22533 Profiling a program helps determine the parts of a program that are executed
22534 most often, and are therefore the most time-consuming.
22536 @code{gprof} is the standard GNU profiling tool; it has been enhanced to
22537 better handle Ada programs and multitasking.
22538 It is currently supported on the following platforms
22543 solaris sparc/sparc64/x86
22549 In order to profile a program using @code{gprof}, 3 steps are needed:
22553 Code instrumentation, requiring a full recompilation of the project with the
22556 Execution of the program under the analysis conditions, i.e. with the desired
22559 Analysis of the results using the @code{gprof} tool.
22563 The following sections detail the different steps, and indicate how
22564 to interpret the results:
22566 * Compilation for profiling::
22567 * Program execution::
22569 * Interpretation of profiling results::
22572 @node Compilation for profiling
22573 @subsection Compilation for profiling
22577 In order to profile a program the first step is to tell the compiler
22578 to generate the necessary profiling information. The compiler switch to be used
22579 is @code{-pg}, which must be added to other compilation switches. This
22580 switch needs to be specified both during compilation and link stages, and can
22581 be specified once when using gnatmake:
22584 gnatmake -f -pg -P my_project
22588 Note that only the objects that were compiled with the @samp{-pg} switch will be
22589 profiled; if you need to profile your whole project, use the
22590 @samp{-f} gnatmake switch to force full recompilation.
22592 @node Program execution
22593 @subsection Program execution
22596 Once the program has been compiled for profiling, you can run it as usual.
22598 The only constraint imposed by profiling is that the program must terminate
22599 normally. An interrupted program (via a Ctrl-C, kill, etc.) will not be
22602 Once the program completes execution, a data file called @file{gmon.out} is
22603 generated in the directory where the program was launched from. If this file
22604 already exists, it will be overwritten.
22606 @node Running gprof
22607 @subsection Running gprof
22610 The @code{gprof} tool is called as follow:
22613 gprof my_prog gmon.out
22624 The complete form of the gprof command line is the following:
22627 gprof [^switches^options^] [executable [data-file]]
22631 @code{gprof} supports numerous ^switch^options^. The order of these
22632 ^switch^options^ does not matter. The full list of options can be found in
22633 the GNU Profiler User's Guide documentation that comes with this documentation.
22635 The following is the subset of those switches that is most relevant:
22639 @item --demangle[=@var{style}]
22640 @itemx --no-demangle
22641 @cindex @option{--demangle} (@code{gprof})
22642 These options control whether symbol names should be demangled when
22643 printing output. The default is to demangle C++ symbols. The
22644 @code{--no-demangle} option may be used to turn off demangling. Different
22645 compilers have different mangling styles. The optional demangling style
22646 argument can be used to choose an appropriate demangling style for your
22647 compiler, in particular Ada symbols generated by GNAT can be demangled using
22648 @code{--demangle=gnat}.
22650 @item -e @var{function_name}
22651 @cindex @option{-e} (@code{gprof})
22652 The @samp{-e @var{function}} option tells @code{gprof} not to print
22653 information about the function @var{function_name} (and its
22654 children@dots{}) in the call graph. The function will still be listed
22655 as a child of any functions that call it, but its index number will be
22656 shown as @samp{[not printed]}. More than one @samp{-e} option may be
22657 given; only one @var{function_name} may be indicated with each @samp{-e}
22660 @item -E @var{function_name}
22661 @cindex @option{-E} (@code{gprof})
22662 The @code{-E @var{function}} option works like the @code{-e} option, but
22663 execution time spent in the function (and children who were not called from
22664 anywhere else), will not be used to compute the percentages-of-time for
22665 the call graph. More than one @samp{-E} option may be given; only one
22666 @var{function_name} may be indicated with each @samp{-E} option.
22668 @item -f @var{function_name}
22669 @cindex @option{-f} (@code{gprof})
22670 The @samp{-f @var{function}} option causes @code{gprof} to limit the
22671 call graph to the function @var{function_name} and its children (and
22672 their children@dots{}). More than one @samp{-f} option may be given;
22673 only one @var{function_name} may be indicated with each @samp{-f}
22676 @item -F @var{function_name}
22677 @cindex @option{-F} (@code{gprof})
22678 The @samp{-F @var{function}} option works like the @code{-f} option, but
22679 only time spent in the function and its children (and their
22680 children@dots{}) will be used to determine total-time and
22681 percentages-of-time for the call graph. More than one @samp{-F} option
22682 may be given; only one @var{function_name} may be indicated with each
22683 @samp{-F} option. The @samp{-F} option overrides the @samp{-E} option.
22687 @node Interpretation of profiling results
22688 @subsection Interpretation of profiling results
22692 The results of the profiling analysis are represented by two arrays: the
22693 'flat profile' and the 'call graph'. Full documentation of those outputs
22694 can be found in the GNU Profiler User's Guide.
22696 The flat profile shows the time spent in each function of the program, and how
22697 many time it has been called. This allows you to locate easily the most
22698 time-consuming functions.
22700 The call graph shows, for each subprogram, the subprograms that call it,
22701 and the subprograms that it calls. It also provides an estimate of the time
22702 spent in each of those callers/called subprograms.
22705 @c ******************************
22706 @node Running and Debugging Ada Programs
22707 @chapter Running and Debugging Ada Programs
22711 This chapter discusses how to debug Ada programs.
22713 It applies to GNAT on the Alpha OpenVMS platform;
22714 for I64 OpenVMS please refer to the @cite{OpenVMS Debugger Manual},
22715 since HP has implemented Ada support in the OpenVMS debugger on I64.
22718 An incorrect Ada program may be handled in three ways by the GNAT compiler:
22722 The illegality may be a violation of the static semantics of Ada. In
22723 that case GNAT diagnoses the constructs in the program that are illegal.
22724 It is then a straightforward matter for the user to modify those parts of
22728 The illegality may be a violation of the dynamic semantics of Ada. In
22729 that case the program compiles and executes, but may generate incorrect
22730 results, or may terminate abnormally with some exception.
22733 When presented with a program that contains convoluted errors, GNAT
22734 itself may terminate abnormally without providing full diagnostics on
22735 the incorrect user program.
22739 * The GNAT Debugger GDB::
22741 * Introduction to GDB Commands::
22742 * Using Ada Expressions::
22743 * Calling User-Defined Subprograms::
22744 * Using the Next Command in a Function::
22747 * Debugging Generic Units::
22748 * GNAT Abnormal Termination or Failure to Terminate::
22749 * Naming Conventions for GNAT Source Files::
22750 * Getting Internal Debugging Information::
22751 * Stack Traceback::
22757 @node The GNAT Debugger GDB
22758 @section The GNAT Debugger GDB
22761 @code{GDB} is a general purpose, platform-independent debugger that
22762 can be used to debug mixed-language programs compiled with @command{gcc},
22763 and in particular is capable of debugging Ada programs compiled with
22764 GNAT. The latest versions of @code{GDB} are Ada-aware and can handle
22765 complex Ada data structures.
22767 @xref{Top,, Debugging with GDB, gdb, Debugging with GDB},
22769 located in the GNU:[DOCS] directory,
22771 for full details on the usage of @code{GDB}, including a section on
22772 its usage on programs. This manual should be consulted for full
22773 details. The section that follows is a brief introduction to the
22774 philosophy and use of @code{GDB}.
22776 When GNAT programs are compiled, the compiler optionally writes debugging
22777 information into the generated object file, including information on
22778 line numbers, and on declared types and variables. This information is
22779 separate from the generated code. It makes the object files considerably
22780 larger, but it does not add to the size of the actual executable that
22781 will be loaded into memory, and has no impact on run-time performance. The
22782 generation of debug information is triggered by the use of the
22783 ^-g^/DEBUG^ switch in the @command{gcc} or @command{gnatmake} command
22784 used to carry out the compilations. It is important to emphasize that
22785 the use of these options does not change the generated code.
22787 The debugging information is written in standard system formats that
22788 are used by many tools, including debuggers and profilers. The format
22789 of the information is typically designed to describe C types and
22790 semantics, but GNAT implements a translation scheme which allows full
22791 details about Ada types and variables to be encoded into these
22792 standard C formats. Details of this encoding scheme may be found in
22793 the file exp_dbug.ads in the GNAT source distribution. However, the
22794 details of this encoding are, in general, of no interest to a user,
22795 since @code{GDB} automatically performs the necessary decoding.
22797 When a program is bound and linked, the debugging information is
22798 collected from the object files, and stored in the executable image of
22799 the program. Again, this process significantly increases the size of
22800 the generated executable file, but it does not increase the size of
22801 the executable program itself. Furthermore, if this program is run in
22802 the normal manner, it runs exactly as if the debug information were
22803 not present, and takes no more actual memory.
22805 However, if the program is run under control of @code{GDB}, the
22806 debugger is activated. The image of the program is loaded, at which
22807 point it is ready to run. If a run command is given, then the program
22808 will run exactly as it would have if @code{GDB} were not present. This
22809 is a crucial part of the @code{GDB} design philosophy. @code{GDB} is
22810 entirely non-intrusive until a breakpoint is encountered. If no
22811 breakpoint is ever hit, the program will run exactly as it would if no
22812 debugger were present. When a breakpoint is hit, @code{GDB} accesses
22813 the debugging information and can respond to user commands to inspect
22814 variables, and more generally to report on the state of execution.
22818 @section Running GDB
22821 This section describes how to initiate the debugger.
22822 @c The above sentence is really just filler, but it was otherwise
22823 @c clumsy to get the first paragraph nonindented given the conditional
22824 @c nature of the description
22827 The debugger can be launched from a @code{GPS} menu or
22828 directly from the command line. The description below covers the latter use.
22829 All the commands shown can be used in the @code{GPS} debug console window,
22830 but there are usually more GUI-based ways to achieve the same effect.
22833 The command to run @code{GDB} is
22836 $ ^gdb program^GDB PROGRAM^
22840 where @code{^program^PROGRAM^} is the name of the executable file. This
22841 activates the debugger and results in a prompt for debugger commands.
22842 The simplest command is simply @code{run}, which causes the program to run
22843 exactly as if the debugger were not present. The following section
22844 describes some of the additional commands that can be given to @code{GDB}.
22846 @c *******************************
22847 @node Introduction to GDB Commands
22848 @section Introduction to GDB Commands
22851 @code{GDB} contains a large repertoire of commands. @xref{Top,,
22852 Debugging with GDB, gdb, Debugging with GDB},
22854 located in the GNU:[DOCS] directory,
22856 for extensive documentation on the use
22857 of these commands, together with examples of their use. Furthermore,
22858 the command @command{help} invoked from within GDB activates a simple help
22859 facility which summarizes the available commands and their options.
22860 In this section we summarize a few of the most commonly
22861 used commands to give an idea of what @code{GDB} is about. You should create
22862 a simple program with debugging information and experiment with the use of
22863 these @code{GDB} commands on the program as you read through the
22867 @item set args @var{arguments}
22868 The @var{arguments} list above is a list of arguments to be passed to
22869 the program on a subsequent run command, just as though the arguments
22870 had been entered on a normal invocation of the program. The @code{set args}
22871 command is not needed if the program does not require arguments.
22874 The @code{run} command causes execution of the program to start from
22875 the beginning. If the program is already running, that is to say if
22876 you are currently positioned at a breakpoint, then a prompt will ask
22877 for confirmation that you want to abandon the current execution and
22880 @item breakpoint @var{location}
22881 The breakpoint command sets a breakpoint, that is to say a point at which
22882 execution will halt and @code{GDB} will await further
22883 commands. @var{location} is
22884 either a line number within a file, given in the format @code{file:linenumber},
22885 or it is the name of a subprogram. If you request that a breakpoint be set on
22886 a subprogram that is overloaded, a prompt will ask you to specify on which of
22887 those subprograms you want to breakpoint. You can also
22888 specify that all of them should be breakpointed. If the program is run
22889 and execution encounters the breakpoint, then the program
22890 stops and @code{GDB} signals that the breakpoint was encountered by
22891 printing the line of code before which the program is halted.
22893 @item breakpoint exception @var{name}
22894 A special form of the breakpoint command which breakpoints whenever
22895 exception @var{name} is raised.
22896 If @var{name} is omitted,
22897 then a breakpoint will occur when any exception is raised.
22899 @item print @var{expression}
22900 This will print the value of the given expression. Most simple
22901 Ada expression formats are properly handled by @code{GDB}, so the expression
22902 can contain function calls, variables, operators, and attribute references.
22905 Continues execution following a breakpoint, until the next breakpoint or the
22906 termination of the program.
22909 Executes a single line after a breakpoint. If the next statement
22910 is a subprogram call, execution continues into (the first statement of)
22911 the called subprogram.
22914 Executes a single line. If this line is a subprogram call, executes and
22915 returns from the call.
22918 Lists a few lines around the current source location. In practice, it
22919 is usually more convenient to have a separate edit window open with the
22920 relevant source file displayed. Successive applications of this command
22921 print subsequent lines. The command can be given an argument which is a
22922 line number, in which case it displays a few lines around the specified one.
22925 Displays a backtrace of the call chain. This command is typically
22926 used after a breakpoint has occurred, to examine the sequence of calls that
22927 leads to the current breakpoint. The display includes one line for each
22928 activation record (frame) corresponding to an active subprogram.
22931 At a breakpoint, @code{GDB} can display the values of variables local
22932 to the current frame. The command @code{up} can be used to
22933 examine the contents of other active frames, by moving the focus up
22934 the stack, that is to say from callee to caller, one frame at a time.
22937 Moves the focus of @code{GDB} down from the frame currently being
22938 examined to the frame of its callee (the reverse of the previous command),
22940 @item frame @var{n}
22941 Inspect the frame with the given number. The value 0 denotes the frame
22942 of the current breakpoint, that is to say the top of the call stack.
22947 The above list is a very short introduction to the commands that
22948 @code{GDB} provides. Important additional capabilities, including conditional
22949 breakpoints, the ability to execute command sequences on a breakpoint,
22950 the ability to debug at the machine instruction level and many other
22951 features are described in detail in @ref{Top,, Debugging with GDB, gdb,
22952 Debugging with GDB}. Note that most commands can be abbreviated
22953 (for example, c for continue, bt for backtrace).
22955 @node Using Ada Expressions
22956 @section Using Ada Expressions
22957 @cindex Ada expressions
22960 @code{GDB} supports a fairly large subset of Ada expression syntax, with some
22961 extensions. The philosophy behind the design of this subset is
22965 That @code{GDB} should provide basic literals and access to operations for
22966 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
22967 leaving more sophisticated computations to subprograms written into the
22968 program (which therefore may be called from @code{GDB}).
22971 That type safety and strict adherence to Ada language restrictions
22972 are not particularly important to the @code{GDB} user.
22975 That brevity is important to the @code{GDB} user.
22979 Thus, for brevity, the debugger acts as if there were
22980 implicit @code{with} and @code{use} clauses in effect for all user-written
22981 packages, thus making it unnecessary to fully qualify most names with
22982 their packages, regardless of context. Where this causes ambiguity,
22983 @code{GDB} asks the user's intent.
22985 For details on the supported Ada syntax, see @ref{Top,, Debugging with
22986 GDB, gdb, Debugging with GDB}.
22988 @node Calling User-Defined Subprograms
22989 @section Calling User-Defined Subprograms
22992 An important capability of @code{GDB} is the ability to call user-defined
22993 subprograms while debugging. This is achieved simply by entering
22994 a subprogram call statement in the form:
22997 call subprogram-name (parameters)
23001 The keyword @code{call} can be omitted in the normal case where the
23002 @code{subprogram-name} does not coincide with any of the predefined
23003 @code{GDB} commands.
23005 The effect is to invoke the given subprogram, passing it the
23006 list of parameters that is supplied. The parameters can be expressions and
23007 can include variables from the program being debugged. The
23008 subprogram must be defined
23009 at the library level within your program, and @code{GDB} will call the
23010 subprogram within the environment of your program execution (which
23011 means that the subprogram is free to access or even modify variables
23012 within your program).
23014 The most important use of this facility is in allowing the inclusion of
23015 debugging routines that are tailored to particular data structures
23016 in your program. Such debugging routines can be written to provide a suitably
23017 high-level description of an abstract type, rather than a low-level dump
23018 of its physical layout. After all, the standard
23019 @code{GDB print} command only knows the physical layout of your
23020 types, not their abstract meaning. Debugging routines can provide information
23021 at the desired semantic level and are thus enormously useful.
23023 For example, when debugging GNAT itself, it is crucial to have access to
23024 the contents of the tree nodes used to represent the program internally.
23025 But tree nodes are represented simply by an integer value (which in turn
23026 is an index into a table of nodes).
23027 Using the @code{print} command on a tree node would simply print this integer
23028 value, which is not very useful. But the PN routine (defined in file
23029 treepr.adb in the GNAT sources) takes a tree node as input, and displays
23030 a useful high level representation of the tree node, which includes the
23031 syntactic category of the node, its position in the source, the integers
23032 that denote descendant nodes and parent node, as well as varied
23033 semantic information. To study this example in more detail, you might want to
23034 look at the body of the PN procedure in the stated file.
23036 @node Using the Next Command in a Function
23037 @section Using the Next Command in a Function
23040 When you use the @code{next} command in a function, the current source
23041 location will advance to the next statement as usual. A special case
23042 arises in the case of a @code{return} statement.
23044 Part of the code for a return statement is the ``epilog'' of the function.
23045 This is the code that returns to the caller. There is only one copy of
23046 this epilog code, and it is typically associated with the last return
23047 statement in the function if there is more than one return. In some
23048 implementations, this epilog is associated with the first statement
23051 The result is that if you use the @code{next} command from a return
23052 statement that is not the last return statement of the function you
23053 may see a strange apparent jump to the last return statement or to
23054 the start of the function. You should simply ignore this odd jump.
23055 The value returned is always that from the first return statement
23056 that was stepped through.
23058 @node Ada Exceptions
23059 @section Breaking on Ada Exceptions
23063 You can set breakpoints that trip when your program raises
23064 selected exceptions.
23067 @item break exception
23068 Set a breakpoint that trips whenever (any task in the) program raises
23071 @item break exception @var{name}
23072 Set a breakpoint that trips whenever (any task in the) program raises
23073 the exception @var{name}.
23075 @item break exception unhandled
23076 Set a breakpoint that trips whenever (any task in the) program raises an
23077 exception for which there is no handler.
23079 @item info exceptions
23080 @itemx info exceptions @var{regexp}
23081 The @code{info exceptions} command permits the user to examine all defined
23082 exceptions within Ada programs. With a regular expression, @var{regexp}, as
23083 argument, prints out only those exceptions whose name matches @var{regexp}.
23091 @code{GDB} allows the following task-related commands:
23095 This command shows a list of current Ada tasks, as in the following example:
23102 ID TID P-ID Thread Pri State Name
23103 1 8088000 0 807e000 15 Child Activation Wait main_task
23104 2 80a4000 1 80ae000 15 Accept/Select Wait b
23105 3 809a800 1 80a4800 15 Child Activation Wait a
23106 * 4 80ae800 3 80b8000 15 Running c
23110 In this listing, the asterisk before the first task indicates it to be the
23111 currently running task. The first column lists the task ID that is used
23112 to refer to tasks in the following commands.
23114 @item break @var{linespec} task @var{taskid}
23115 @itemx break @var{linespec} task @var{taskid} if @dots{}
23116 @cindex Breakpoints and tasks
23117 These commands are like the @code{break @dots{} thread @dots{}}.
23118 @var{linespec} specifies source lines.
23120 Use the qualifier @samp{task @var{taskid}} with a breakpoint command
23121 to specify that you only want @code{GDB} to stop the program when a
23122 particular Ada task reaches this breakpoint. @var{taskid} is one of the
23123 numeric task identifiers assigned by @code{GDB}, shown in the first
23124 column of the @samp{info tasks} display.
23126 If you do not specify @samp{task @var{taskid}} when you set a
23127 breakpoint, the breakpoint applies to @emph{all} tasks of your
23130 You can use the @code{task} qualifier on conditional breakpoints as
23131 well; in this case, place @samp{task @var{taskid}} before the
23132 breakpoint condition (before the @code{if}).
23134 @item task @var{taskno}
23135 @cindex Task switching
23137 This command allows to switch to the task referred by @var{taskno}. In
23138 particular, This allows to browse the backtrace of the specified
23139 task. It is advised to switch back to the original task before
23140 continuing execution otherwise the scheduling of the program may be
23145 For more detailed information on the tasking support,
23146 see @ref{Top,, Debugging with GDB, gdb, Debugging with GDB}.
23148 @node Debugging Generic Units
23149 @section Debugging Generic Units
23150 @cindex Debugging Generic Units
23154 GNAT always uses code expansion for generic instantiation. This means that
23155 each time an instantiation occurs, a complete copy of the original code is
23156 made, with appropriate substitutions of formals by actuals.
23158 It is not possible to refer to the original generic entities in
23159 @code{GDB}, but it is always possible to debug a particular instance of
23160 a generic, by using the appropriate expanded names. For example, if we have
23162 @smallexample @c ada
23167 generic package k is
23168 procedure kp (v1 : in out integer);
23172 procedure kp (v1 : in out integer) is
23178 package k1 is new k;
23179 package k2 is new k;
23181 var : integer := 1;
23194 Then to break on a call to procedure kp in the k2 instance, simply
23198 (gdb) break g.k2.kp
23202 When the breakpoint occurs, you can step through the code of the
23203 instance in the normal manner and examine the values of local variables, as for
23206 @node GNAT Abnormal Termination or Failure to Terminate
23207 @section GNAT Abnormal Termination or Failure to Terminate
23208 @cindex GNAT Abnormal Termination or Failure to Terminate
23211 When presented with programs that contain serious errors in syntax
23213 GNAT may on rare occasions experience problems in operation, such
23215 segmentation fault or illegal memory access, raising an internal
23216 exception, terminating abnormally, or failing to terminate at all.
23217 In such cases, you can activate
23218 various features of GNAT that can help you pinpoint the construct in your
23219 program that is the likely source of the problem.
23221 The following strategies are presented in increasing order of
23222 difficulty, corresponding to your experience in using GNAT and your
23223 familiarity with compiler internals.
23227 Run @command{gcc} with the @option{-gnatf}. This first
23228 switch causes all errors on a given line to be reported. In its absence,
23229 only the first error on a line is displayed.
23231 The @option{-gnatdO} switch causes errors to be displayed as soon as they
23232 are encountered, rather than after compilation is terminated. If GNAT
23233 terminates prematurely or goes into an infinite loop, the last error
23234 message displayed may help to pinpoint the culprit.
23237 Run @command{gcc} with the @option{^-v (verbose)^/VERBOSE^} switch. In this
23238 mode, @command{gcc} produces ongoing information about the progress of the
23239 compilation and provides the name of each procedure as code is
23240 generated. This switch allows you to find which Ada procedure was being
23241 compiled when it encountered a code generation problem.
23244 @cindex @option{-gnatdc} switch
23245 Run @command{gcc} with the @option{-gnatdc} switch. This is a GNAT specific
23246 switch that does for the front-end what @option{^-v^VERBOSE^} does
23247 for the back end. The system prints the name of each unit,
23248 either a compilation unit or nested unit, as it is being analyzed.
23250 Finally, you can start
23251 @code{gdb} directly on the @code{gnat1} executable. @code{gnat1} is the
23252 front-end of GNAT, and can be run independently (normally it is just
23253 called from @command{gcc}). You can use @code{gdb} on @code{gnat1} as you
23254 would on a C program (but @pxref{The GNAT Debugger GDB} for caveats). The
23255 @code{where} command is the first line of attack; the variable
23256 @code{lineno} (seen by @code{print lineno}), used by the second phase of
23257 @code{gnat1} and by the @command{gcc} backend, indicates the source line at
23258 which the execution stopped, and @code{input_file name} indicates the name of
23262 @node Naming Conventions for GNAT Source Files
23263 @section Naming Conventions for GNAT Source Files
23266 In order to examine the workings of the GNAT system, the following
23267 brief description of its organization may be helpful:
23271 Files with prefix @file{^sc^SC^} contain the lexical scanner.
23274 All files prefixed with @file{^par^PAR^} are components of the parser. The
23275 numbers correspond to chapters of the Ada Reference Manual. For example,
23276 parsing of select statements can be found in @file{par-ch9.adb}.
23279 All files prefixed with @file{^sem^SEM^} perform semantic analysis. The
23280 numbers correspond to chapters of the Ada standard. For example, all
23281 issues involving context clauses can be found in @file{sem_ch10.adb}. In
23282 addition, some features of the language require sufficient special processing
23283 to justify their own semantic files: sem_aggr for aggregates, sem_disp for
23284 dynamic dispatching, etc.
23287 All files prefixed with @file{^exp^EXP^} perform normalization and
23288 expansion of the intermediate representation (abstract syntax tree, or AST).
23289 these files use the same numbering scheme as the parser and semantics files.
23290 For example, the construction of record initialization procedures is done in
23291 @file{exp_ch3.adb}.
23294 The files prefixed with @file{^bind^BIND^} implement the binder, which
23295 verifies the consistency of the compilation, determines an order of
23296 elaboration, and generates the bind file.
23299 The files @file{atree.ads} and @file{atree.adb} detail the low-level
23300 data structures used by the front-end.
23303 The files @file{sinfo.ads} and @file{sinfo.adb} detail the structure of
23304 the abstract syntax tree as produced by the parser.
23307 The files @file{einfo.ads} and @file{einfo.adb} detail the attributes of
23308 all entities, computed during semantic analysis.
23311 Library management issues are dealt with in files with prefix
23317 Ada files with the prefix @file{^a-^A-^} are children of @code{Ada}, as
23318 defined in Annex A.
23323 Files with prefix @file{^i-^I-^} are children of @code{Interfaces}, as
23324 defined in Annex B.
23328 Files with prefix @file{^s-^S-^} are children of @code{System}. This includes
23329 both language-defined children and GNAT run-time routines.
23333 Files with prefix @file{^g-^G-^} are children of @code{GNAT}. These are useful
23334 general-purpose packages, fully documented in their specs. All
23335 the other @file{.c} files are modifications of common @command{gcc} files.
23338 @node Getting Internal Debugging Information
23339 @section Getting Internal Debugging Information
23342 Most compilers have internal debugging switches and modes. GNAT
23343 does also, except GNAT internal debugging switches and modes are not
23344 secret. A summary and full description of all the compiler and binder
23345 debug flags are in the file @file{debug.adb}. You must obtain the
23346 sources of the compiler to see the full detailed effects of these flags.
23348 The switches that print the source of the program (reconstructed from
23349 the internal tree) are of general interest for user programs, as are the
23351 the full internal tree, and the entity table (the symbol table
23352 information). The reconstructed source provides a readable version of the
23353 program after the front-end has completed analysis and expansion,
23354 and is useful when studying the performance of specific constructs.
23355 For example, constraint checks are indicated, complex aggregates
23356 are replaced with loops and assignments, and tasking primitives
23357 are replaced with run-time calls.
23359 @node Stack Traceback
23360 @section Stack Traceback
23362 @cindex stack traceback
23363 @cindex stack unwinding
23366 Traceback is a mechanism to display the sequence of subprogram calls that
23367 leads to a specified execution point in a program. Often (but not always)
23368 the execution point is an instruction at which an exception has been raised.
23369 This mechanism is also known as @i{stack unwinding} because it obtains
23370 its information by scanning the run-time stack and recovering the activation
23371 records of all active subprograms. Stack unwinding is one of the most
23372 important tools for program debugging.
23374 The first entry stored in traceback corresponds to the deepest calling level,
23375 that is to say the subprogram currently executing the instruction
23376 from which we want to obtain the traceback.
23378 Note that there is no runtime performance penalty when stack traceback
23379 is enabled, and no exception is raised during program execution.
23382 * Non-Symbolic Traceback::
23383 * Symbolic Traceback::
23386 @node Non-Symbolic Traceback
23387 @subsection Non-Symbolic Traceback
23388 @cindex traceback, non-symbolic
23391 Note: this feature is not supported on all platforms. See
23392 @file{GNAT.Traceback spec in g-traceb.ads} for a complete list of supported
23396 * Tracebacks From an Unhandled Exception::
23397 * Tracebacks From Exception Occurrences (non-symbolic)::
23398 * Tracebacks From Anywhere in a Program (non-symbolic)::
23401 @node Tracebacks From an Unhandled Exception
23402 @subsubsection Tracebacks From an Unhandled Exception
23405 A runtime non-symbolic traceback is a list of addresses of call instructions.
23406 To enable this feature you must use the @option{-E}
23407 @code{gnatbind}'s option. With this option a stack traceback is stored as part
23408 of exception information. You can retrieve this information using the
23409 @code{addr2line} tool.
23411 Here is a simple example:
23413 @smallexample @c ada
23419 raise Constraint_Error;
23434 $ gnatmake stb -bargs -E
23437 Execution terminated by unhandled exception
23438 Exception name: CONSTRAINT_ERROR
23440 Call stack traceback locations:
23441 0x401373 0x40138b 0x40139c 0x401335 0x4011c4 0x4011f1 0x77e892a4
23445 As we see the traceback lists a sequence of addresses for the unhandled
23446 exception @code{CONSTRAINT_ERROR} raised in procedure P1. It is easy to
23447 guess that this exception come from procedure P1. To translate these
23448 addresses into the source lines where the calls appear, the
23449 @code{addr2line} tool, described below, is invaluable. The use of this tool
23450 requires the program to be compiled with debug information.
23453 $ gnatmake -g stb -bargs -E
23456 Execution terminated by unhandled exception
23457 Exception name: CONSTRAINT_ERROR
23459 Call stack traceback locations:
23460 0x401373 0x40138b 0x40139c 0x401335 0x4011c4 0x4011f1 0x77e892a4
23462 $ addr2line --exe=stb 0x401373 0x40138b 0x40139c 0x401335 0x4011c4
23463 0x4011f1 0x77e892a4
23465 00401373 at d:/stb/stb.adb:5
23466 0040138B at d:/stb/stb.adb:10
23467 0040139C at d:/stb/stb.adb:14
23468 00401335 at d:/stb/b~stb.adb:104
23469 004011C4 at /build/@dots{}/crt1.c:200
23470 004011F1 at /build/@dots{}/crt1.c:222
23471 77E892A4 in ?? at ??:0
23475 The @code{addr2line} tool has several other useful options:
23479 to get the function name corresponding to any location
23481 @item --demangle=gnat
23482 to use the gnat decoding mode for the function names. Note that
23483 for binutils version 2.9.x the option is simply @option{--demangle}.
23487 $ addr2line --exe=stb --functions --demangle=gnat 0x401373 0x40138b
23488 0x40139c 0x401335 0x4011c4 0x4011f1
23490 00401373 in stb.p1 at d:/stb/stb.adb:5
23491 0040138B in stb.p2 at d:/stb/stb.adb:10
23492 0040139C in stb at d:/stb/stb.adb:14
23493 00401335 in main at d:/stb/b~stb.adb:104
23494 004011C4 in <__mingw_CRTStartup> at /build/@dots{}/crt1.c:200
23495 004011F1 in <mainCRTStartup> at /build/@dots{}/crt1.c:222
23499 From this traceback we can see that the exception was raised in
23500 @file{stb.adb} at line 5, which was reached from a procedure call in
23501 @file{stb.adb} at line 10, and so on. The @file{b~std.adb} is the binder file,
23502 which contains the call to the main program.
23503 @xref{Running gnatbind}. The remaining entries are assorted runtime routines,
23504 and the output will vary from platform to platform.
23506 It is also possible to use @code{GDB} with these traceback addresses to debug
23507 the program. For example, we can break at a given code location, as reported
23508 in the stack traceback:
23514 Furthermore, this feature is not implemented inside Windows DLL. Only
23515 the non-symbolic traceback is reported in this case.
23518 (gdb) break *0x401373
23519 Breakpoint 1 at 0x401373: file stb.adb, line 5.
23523 It is important to note that the stack traceback addresses
23524 do not change when debug information is included. This is particularly useful
23525 because it makes it possible to release software without debug information (to
23526 minimize object size), get a field report that includes a stack traceback
23527 whenever an internal bug occurs, and then be able to retrieve the sequence
23528 of calls with the same program compiled with debug information.
23530 @node Tracebacks From Exception Occurrences (non-symbolic)
23531 @subsubsection Tracebacks From Exception Occurrences
23534 Non-symbolic tracebacks are obtained by using the @option{-E} binder argument.
23535 The stack traceback is attached to the exception information string, and can
23536 be retrieved in an exception handler within the Ada program, by means of the
23537 Ada facilities defined in @code{Ada.Exceptions}. Here is a simple example:
23539 @smallexample @c ada
23541 with Ada.Exceptions;
23546 use Ada.Exceptions;
23554 Text_IO.Put_Line (Exception_Information (E));
23568 This program will output:
23573 Exception name: CONSTRAINT_ERROR
23574 Message: stb.adb:12
23575 Call stack traceback locations:
23576 0x4015e4 0x401633 0x401644 0x401461 0x4011c4 0x4011f1 0x77e892a4
23579 @node Tracebacks From Anywhere in a Program (non-symbolic)
23580 @subsubsection Tracebacks From Anywhere in a Program
23583 It is also possible to retrieve a stack traceback from anywhere in a
23584 program. For this you need to
23585 use the @code{GNAT.Traceback} API. This package includes a procedure called
23586 @code{Call_Chain} that computes a complete stack traceback, as well as useful
23587 display procedures described below. It is not necessary to use the
23588 @option{-E gnatbind} option in this case, because the stack traceback mechanism
23589 is invoked explicitly.
23592 In the following example we compute a traceback at a specific location in
23593 the program, and we display it using @code{GNAT.Debug_Utilities.Image} to
23594 convert addresses to strings:
23596 @smallexample @c ada
23598 with GNAT.Traceback;
23599 with GNAT.Debug_Utilities;
23605 use GNAT.Traceback;
23608 TB : Tracebacks_Array (1 .. 10);
23609 -- We are asking for a maximum of 10 stack frames.
23611 -- Len will receive the actual number of stack frames returned.
23613 Call_Chain (TB, Len);
23615 Text_IO.Put ("In STB.P1 : ");
23617 for K in 1 .. Len loop
23618 Text_IO.Put (Debug_Utilities.Image (TB (K)));
23639 In STB.P1 : 16#0040_F1E4# 16#0040_14F2# 16#0040_170B# 16#0040_171C#
23640 16#0040_1461# 16#0040_11C4# 16#0040_11F1# 16#77E8_92A4#
23644 You can then get further information by invoking the @code{addr2line}
23645 tool as described earlier (note that the hexadecimal addresses
23646 need to be specified in C format, with a leading ``0x'').
23648 @node Symbolic Traceback
23649 @subsection Symbolic Traceback
23650 @cindex traceback, symbolic
23653 A symbolic traceback is a stack traceback in which procedure names are
23654 associated with each code location.
23657 Note that this feature is not supported on all platforms. See
23658 @file{GNAT.Traceback.Symbolic spec in g-trasym.ads} for a complete
23659 list of currently supported platforms.
23662 Note that the symbolic traceback requires that the program be compiled
23663 with debug information. If it is not compiled with debug information
23664 only the non-symbolic information will be valid.
23667 * Tracebacks From Exception Occurrences (symbolic)::
23668 * Tracebacks From Anywhere in a Program (symbolic)::
23671 @node Tracebacks From Exception Occurrences (symbolic)
23672 @subsubsection Tracebacks From Exception Occurrences
23674 @smallexample @c ada
23676 with GNAT.Traceback.Symbolic;
23682 raise Constraint_Error;
23699 Ada.Text_IO.Put_Line (GNAT.Traceback.Symbolic.Symbolic_Traceback (E));
23704 $ gnatmake -g .\stb -bargs -E -largs -lgnat -laddr2line -lintl
23707 0040149F in stb.p1 at stb.adb:8
23708 004014B7 in stb.p2 at stb.adb:13
23709 004014CF in stb.p3 at stb.adb:18
23710 004015DD in ada.stb at stb.adb:22
23711 00401461 in main at b~stb.adb:168
23712 004011C4 in __mingw_CRTStartup at crt1.c:200
23713 004011F1 in mainCRTStartup at crt1.c:222
23714 77E892A4 in ?? at ??:0
23718 In the above example the ``.\'' syntax in the @command{gnatmake} command
23719 is currently required by @command{addr2line} for files that are in
23720 the current working directory.
23721 Moreover, the exact sequence of linker options may vary from platform
23723 The above @option{-largs} section is for Windows platforms. By contrast,
23724 under Unix there is no need for the @option{-largs} section.
23725 Differences across platforms are due to details of linker implementation.
23727 @node Tracebacks From Anywhere in a Program (symbolic)
23728 @subsubsection Tracebacks From Anywhere in a Program
23731 It is possible to get a symbolic stack traceback
23732 from anywhere in a program, just as for non-symbolic tracebacks.
23733 The first step is to obtain a non-symbolic
23734 traceback, and then call @code{Symbolic_Traceback} to compute the symbolic
23735 information. Here is an example:
23737 @smallexample @c ada
23739 with GNAT.Traceback;
23740 with GNAT.Traceback.Symbolic;
23745 use GNAT.Traceback;
23746 use GNAT.Traceback.Symbolic;
23749 TB : Tracebacks_Array (1 .. 10);
23750 -- We are asking for a maximum of 10 stack frames.
23752 -- Len will receive the actual number of stack frames returned.
23754 Call_Chain (TB, Len);
23755 Text_IO.Put_Line (Symbolic_Traceback (TB (1 .. Len)));
23768 @c ******************************
23770 @node Compatibility with HP Ada
23771 @chapter Compatibility with HP Ada
23772 @cindex Compatibility
23777 @cindex Compatibility between GNAT and HP Ada
23778 This chapter compares HP Ada (formerly known as ``DEC Ada'')
23779 for OpenVMS Alpha and GNAT for OpenVMS for Alpha and for I64.
23780 GNAT is highly compatible
23781 with HP Ada, and it should generally be straightforward to port code
23782 from the HP Ada environment to GNAT. However, there are a few language
23783 and implementation differences of which the user must be aware. These
23784 differences are discussed in this chapter. In
23785 addition, the operating environment and command structure for the
23786 compiler are different, and these differences are also discussed.
23788 For further details on these and other compatibility issues,
23789 see Appendix E of the HP publication
23790 @cite{HP Ada, Technical Overview and Comparison on HP Platforms}.
23792 Except where otherwise indicated, the description of GNAT for OpenVMS
23793 applies to both the Alpha and I64 platforms.
23795 For information on porting Ada code from GNAT on Alpha OpenVMS to GNAT on
23796 I64 OpenVMS, see @ref{Transitioning to 64-Bit GNAT for OpenVMS}.
23798 The discussion in this chapter addresses specifically the implementation
23799 of Ada 83 for HP OpenVMS Alpha Systems. In cases where the implementation
23800 of HP Ada differs between OpenVMS Alpha Systems and OpenVMS VAX Systems,
23801 GNAT always follows the Alpha implementation.
23803 For GNAT running on other than VMS systems, all the HP Ada 83 pragmas and
23804 attributes are recognized, although only a subset of them can sensibly
23805 be implemented. The description of pragmas in
23806 @xref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference Manual}
23807 indicates whether or not they are applicable to non-VMS systems.
23810 * Ada Language Compatibility::
23811 * Differences in the Definition of Package System::
23812 * Language-Related Features::
23813 * The Package STANDARD::
23814 * The Package SYSTEM::
23815 * Tasking and Task-Related Features::
23816 * Pragmas and Pragma-Related Features::
23817 * Library of Predefined Units::
23819 * Main Program Definition::
23820 * Implementation-Defined Attributes::
23821 * Compiler and Run-Time Interfacing::
23822 * Program Compilation and Library Management::
23824 * Implementation Limits::
23825 * Tools and Utilities::
23828 @node Ada Language Compatibility
23829 @section Ada Language Compatibility
23832 GNAT handles Ada 95 and Ada 2005 as well as Ada 83, whereas HP Ada is only
23833 for Ada 83. Ada 95 and Ada 2005 are almost completely upwards compatible
23834 with Ada 83, and therefore Ada 83 programs will compile
23835 and run under GNAT with
23836 no changes or only minor changes. The @cite{Annotated Ada Reference Manual}
23837 provides details on specific incompatibilities.
23839 GNAT provides the switch @option{/83} on the @command{GNAT COMPILE} command,
23840 as well as the pragma @code{ADA_83}, to force the compiler to
23841 operate in Ada 83 mode. This mode does not guarantee complete
23842 conformance to Ada 83, but in practice is sufficient to
23843 eliminate most sources of incompatibilities.
23844 In particular, it eliminates the recognition of the
23845 additional Ada 95 and Ada 2005 keywords, so that their use as identifiers
23846 in Ada 83 programs is legal, and handles the cases of packages
23847 with optional bodies, and generics that instantiate unconstrained
23848 types without the use of @code{(<>)}.
23850 @node Differences in the Definition of Package System
23851 @section Differences in the Definition of Package @code{System}
23854 An Ada compiler is allowed to add
23855 implementation-dependent declarations to package @code{System}.
23857 GNAT does not take advantage of this permission, and the version of
23858 @code{System} provided by GNAT exactly matches that defined in the Ada
23861 However, HP Ada adds an extensive set of declarations to package
23863 as fully documented in the HP Ada manuals. To minimize changes required
23864 for programs that make use of these extensions, GNAT provides the pragma
23865 @code{Extend_System} for extending the definition of package System. By using:
23866 @cindex pragma @code{Extend_System}
23867 @cindex @code{Extend_System} pragma
23869 @smallexample @c ada
23872 pragma Extend_System (Aux_DEC);
23878 the set of definitions in @code{System} is extended to include those in
23879 package @code{System.Aux_DEC}.
23880 @cindex @code{System.Aux_DEC} package
23881 @cindex @code{Aux_DEC} package (child of @code{System})
23882 These definitions are incorporated directly into package @code{System},
23883 as though they had been declared there. For a
23884 list of the declarations added, see the spec of this package,
23885 which can be found in the file @file{s-auxdec.ads} in the GNAT library.
23886 @cindex @file{s-auxdec.ads} file
23887 The pragma @code{Extend_System} is a configuration pragma, which means that
23888 it can be placed in the file @file{gnat.adc}, so that it will automatically
23889 apply to all subsequent compilations. See @ref{Configuration Pragmas},
23890 for further details.
23892 An alternative approach that avoids the use of the non-standard
23893 @code{Extend_System} pragma is to add a context clause to the unit that
23894 references these facilities:
23896 @smallexample @c ada
23898 with System.Aux_DEC;
23899 use System.Aux_DEC;
23904 The effect is not quite semantically identical to incorporating
23905 the declarations directly into package @code{System},
23906 but most programs will not notice a difference
23907 unless they use prefix notation (e.g.@: @code{System.Integer_8})
23908 to reference the entities directly in package @code{System}.
23909 For units containing such references,
23910 the prefixes must either be removed, or the pragma @code{Extend_System}
23913 @node Language-Related Features
23914 @section Language-Related Features
23917 The following sections highlight differences in types,
23918 representations of types, operations, alignment, and
23922 * Integer Types and Representations::
23923 * Floating-Point Types and Representations::
23924 * Pragmas Float_Representation and Long_Float::
23925 * Fixed-Point Types and Representations::
23926 * Record and Array Component Alignment::
23927 * Address Clauses::
23928 * Other Representation Clauses::
23931 @node Integer Types and Representations
23932 @subsection Integer Types and Representations
23935 The set of predefined integer types is identical in HP Ada and GNAT.
23936 Furthermore the representation of these integer types is also identical,
23937 including the capability of size clauses forcing biased representation.
23940 HP Ada for OpenVMS Alpha systems has defined the
23941 following additional integer types in package @code{System}:
23958 @code{LARGEST_INTEGER}
23962 In GNAT, the first four of these types may be obtained from the
23963 standard Ada package @code{Interfaces}.
23964 Alternatively, by use of the pragma @code{Extend_System}, identical
23965 declarations can be referenced directly in package @code{System}.
23966 On both GNAT and HP Ada, the maximum integer size is 64 bits.
23968 @node Floating-Point Types and Representations
23969 @subsection Floating-Point Types and Representations
23970 @cindex Floating-Point types
23973 The set of predefined floating-point types is identical in HP Ada and GNAT.
23974 Furthermore the representation of these floating-point
23975 types is also identical. One important difference is that the default
23976 representation for HP Ada is @code{VAX_Float}, but the default representation
23979 Specific types may be declared to be @code{VAX_Float} or IEEE, using the
23980 pragma @code{Float_Representation} as described in the HP Ada
23982 For example, the declarations:
23984 @smallexample @c ada
23986 type F_Float is digits 6;
23987 pragma Float_Representation (VAX_Float, F_Float);
23992 declares a type @code{F_Float} that will be represented in @code{VAX_Float}
23994 This set of declarations actually appears in @code{System.Aux_DEC},
23996 the full set of additional floating-point declarations provided in
23997 the HP Ada version of package @code{System}.
23998 This and similar declarations may be accessed in a user program
23999 by using pragma @code{Extend_System}. The use of this
24000 pragma, and the related pragma @code{Long_Float} is described in further
24001 detail in the following section.
24003 @node Pragmas Float_Representation and Long_Float
24004 @subsection Pragmas @code{Float_Representation} and @code{Long_Float}
24007 HP Ada provides the pragma @code{Float_Representation}, which
24008 acts as a program library switch to allow control over
24009 the internal representation chosen for the predefined
24010 floating-point types declared in the package @code{Standard}.
24011 The format of this pragma is as follows:
24013 @smallexample @c ada
24015 pragma Float_Representation(VAX_Float | IEEE_Float);
24020 This pragma controls the representation of floating-point
24025 @code{VAX_Float} specifies that floating-point
24026 types are represented by default with the VAX system hardware types
24027 @code{F-floating}, @code{D-floating}, @code{G-floating}.
24028 Note that the @code{H-floating}
24029 type was available only on VAX systems, and is not available
24030 in either HP Ada or GNAT.
24033 @code{IEEE_Float} specifies that floating-point
24034 types are represented by default with the IEEE single and
24035 double floating-point types.
24039 GNAT provides an identical implementation of the pragma
24040 @code{Float_Representation}, except that it functions as a
24041 configuration pragma. Note that the
24042 notion of configuration pragma corresponds closely to the
24043 HP Ada notion of a program library switch.
24045 When no pragma is used in GNAT, the default is @code{IEEE_Float},
24047 from HP Ada 83, where the default is @code{VAX_Float}. In addition, the
24048 predefined libraries in GNAT are built using @code{IEEE_Float}, so it is not
24049 advisable to change the format of numbers passed to standard library
24050 routines, and if necessary explicit type conversions may be needed.
24052 The use of @code{IEEE_Float} is recommended in GNAT since it is more
24053 efficient, and (given that it conforms to an international standard)
24054 potentially more portable.
24055 The situation in which @code{VAX_Float} may be useful is in interfacing
24056 to existing code and data that expect the use of @code{VAX_Float}.
24057 In such a situation use the predefined @code{VAX_Float}
24058 types in package @code{System}, as extended by
24059 @code{Extend_System}. For example, use @code{System.F_Float}
24060 to specify the 32-bit @code{F-Float} format.
24063 On OpenVMS systems, HP Ada provides the pragma @code{Long_Float}
24064 to allow control over the internal representation chosen
24065 for the predefined type @code{Long_Float} and for floating-point
24066 type declarations with digits specified in the range 7 .. 15.
24067 The format of this pragma is as follows:
24069 @smallexample @c ada
24071 pragma Long_Float (D_FLOAT | G_FLOAT);
24075 @node Fixed-Point Types and Representations
24076 @subsection Fixed-Point Types and Representations
24079 On HP Ada for OpenVMS Alpha systems, rounding is
24080 away from zero for both positive and negative numbers.
24081 Therefore, @code{+0.5} rounds to @code{1},
24082 and @code{-0.5} rounds to @code{-1}.
24084 On GNAT the results of operations
24085 on fixed-point types are in accordance with the Ada
24086 rules. In particular, results of operations on decimal
24087 fixed-point types are truncated.
24089 @node Record and Array Component Alignment
24090 @subsection Record and Array Component Alignment
24093 On HP Ada for OpenVMS Alpha, all non-composite components
24094 are aligned on natural boundaries. For example, 1-byte
24095 components are aligned on byte boundaries, 2-byte
24096 components on 2-byte boundaries, 4-byte components on 4-byte
24097 byte boundaries, and so on. The OpenVMS Alpha hardware
24098 runs more efficiently with naturally aligned data.
24100 On GNAT, alignment rules are compatible
24101 with HP Ada for OpenVMS Alpha.
24103 @node Address Clauses
24104 @subsection Address Clauses
24107 In HP Ada and GNAT, address clauses are supported for
24108 objects and imported subprograms.
24109 The predefined type @code{System.Address} is a private type
24110 in both compilers on Alpha OpenVMS, with the same representation
24111 (it is simply a machine pointer). Addition, subtraction, and comparison
24112 operations are available in the standard Ada package
24113 @code{System.Storage_Elements}, or in package @code{System}
24114 if it is extended to include @code{System.Aux_DEC} using a
24115 pragma @code{Extend_System} as previously described.
24117 Note that code that @code{with}'s both this extended package @code{System}
24118 and the package @code{System.Storage_Elements} should not @code{use}
24119 both packages, or ambiguities will result. In general it is better
24120 not to mix these two sets of facilities. The Ada package was
24121 designed specifically to provide the kind of features that HP Ada
24122 adds directly to package @code{System}.
24124 The type @code{System.Address} is a 64-bit integer type in GNAT for
24125 I64 OpenVMS. For more information,
24126 see @ref{Transitioning to 64-Bit GNAT for OpenVMS}.
24128 GNAT is compatible with HP Ada in its handling of address
24129 clauses, except for some limitations in
24130 the form of address clauses for composite objects with
24131 initialization. Such address clauses are easily replaced
24132 by the use of an explicitly-defined constant as described
24133 in the Ada Reference Manual (13.1(22)). For example, the sequence
24136 @smallexample @c ada
24138 X, Y : Integer := Init_Func;
24139 Q : String (X .. Y) := "abc";
24141 for Q'Address use Compute_Address;
24146 will be rejected by GNAT, since the address cannot be computed at the time
24147 that @code{Q} is declared. To achieve the intended effect, write instead:
24149 @smallexample @c ada
24152 X, Y : Integer := Init_Func;
24153 Q_Address : constant Address := Compute_Address;
24154 Q : String (X .. Y) := "abc";
24156 for Q'Address use Q_Address;
24162 which will be accepted by GNAT (and other Ada compilers), and is also
24163 compatible with Ada 83. A fuller description of the restrictions
24164 on address specifications is found in @ref{Top, GNAT Reference Manual,
24165 About This Guide, gnat_rm, GNAT Reference Manual}.
24167 @node Other Representation Clauses
24168 @subsection Other Representation Clauses
24171 GNAT implements in a compatible manner all the representation
24172 clauses supported by HP Ada. In addition, GNAT
24173 implements the representation clause forms that were introduced in Ada 95,
24174 including @code{COMPONENT_SIZE} and @code{SIZE} clauses for objects.
24176 @node The Package STANDARD
24177 @section The Package @code{STANDARD}
24180 The package @code{STANDARD}, as implemented by HP Ada, is fully
24181 described in the @cite{Ada Reference Manual} and in the
24182 @cite{HP Ada Language Reference Manual}. As implemented by GNAT, the
24183 package @code{STANDARD} is described in the @cite{Ada Reference Manual}.
24185 In addition, HP Ada supports the Latin-1 character set in
24186 the type @code{CHARACTER}. GNAT supports the Latin-1 character set
24187 in the type @code{CHARACTER} and also Unicode (ISO 10646 BMP) in
24188 the type @code{WIDE_CHARACTER}.
24190 The floating-point types supported by GNAT are those
24191 supported by HP Ada, but the defaults are different, and are controlled by
24192 pragmas. See @ref{Floating-Point Types and Representations}, for details.
24194 @node The Package SYSTEM
24195 @section The Package @code{SYSTEM}
24198 HP Ada provides a specific version of the package
24199 @code{SYSTEM} for each platform on which the language is implemented.
24200 For the complete spec of the package @code{SYSTEM}, see
24201 Appendix F of the @cite{HP Ada Language Reference Manual}.
24203 On HP Ada, the package @code{SYSTEM} includes the following conversion
24206 @item @code{TO_ADDRESS(INTEGER)}
24208 @item @code{TO_ADDRESS(UNSIGNED_LONGWORD)}
24210 @item @code{TO_ADDRESS(}@i{universal_integer}@code{)}
24212 @item @code{TO_INTEGER(ADDRESS)}
24214 @item @code{TO_UNSIGNED_LONGWORD(ADDRESS)}
24216 @item Function @code{IMPORT_VALUE return UNSIGNED_LONGWORD} and the
24217 functions @code{IMPORT_ADDRESS} and @code{IMPORT_LARGEST_VALUE}
24221 By default, GNAT supplies a version of @code{SYSTEM} that matches
24222 the definition given in the @cite{Ada Reference Manual}.
24224 is a subset of the HP system definitions, which is as
24225 close as possible to the original definitions. The only difference
24226 is that the definition of @code{SYSTEM_NAME} is different:
24228 @smallexample @c ada
24230 type Name is (SYSTEM_NAME_GNAT);
24231 System_Name : constant Name := SYSTEM_NAME_GNAT;
24236 Also, GNAT adds the Ada declarations for
24237 @code{BIT_ORDER} and @code{DEFAULT_BIT_ORDER}.
24239 However, the use of the following pragma causes GNAT
24240 to extend the definition of package @code{SYSTEM} so that it
24241 encompasses the full set of HP-specific extensions,
24242 including the functions listed above:
24244 @smallexample @c ada
24246 pragma Extend_System (Aux_DEC);
24251 The pragma @code{Extend_System} is a configuration pragma that
24252 is most conveniently placed in the @file{gnat.adc} file. @xref{Pragma
24253 Extend_System,,, gnat_rm, GNAT Reference Manual} for further details.
24255 HP Ada does not allow the recompilation of the package
24256 @code{SYSTEM}. Instead HP Ada provides several pragmas
24257 (@code{SYSTEM_NAME}, @code{STORAGE_UNIT}, and @code{MEMORY_SIZE})
24258 to modify values in the package @code{SYSTEM}.
24259 On OpenVMS Alpha systems, the pragma
24260 @code{SYSTEM_NAME} takes the enumeration literal @code{OPENVMS_AXP} as
24261 its single argument.
24263 GNAT does permit the recompilation of package @code{SYSTEM} using
24264 the special switch @option{-gnatg}, and this switch can be used if
24265 it is necessary to modify the definitions in @code{SYSTEM}. GNAT does
24266 not permit the specification of @code{SYSTEM_NAME}, @code{STORAGE_UNIT}
24267 or @code{MEMORY_SIZE} by any other means.
24269 On GNAT systems, the pragma @code{SYSTEM_NAME} takes the
24270 enumeration literal @code{SYSTEM_NAME_GNAT}.
24272 The definitions provided by the use of
24274 @smallexample @c ada
24275 pragma Extend_System (AUX_Dec);
24279 are virtually identical to those provided by the HP Ada 83 package
24280 @code{SYSTEM}. One important difference is that the name of the
24282 function for type @code{UNSIGNED_LONGWORD} is changed to
24283 @code{TO_ADDRESS_LONG}.
24284 @xref{Address Clauses,,, gnat_rm, GNAT Reference Manual} for a
24285 discussion of why this change was necessary.
24288 The version of @code{TO_ADDRESS} taking a @i{universal_integer} argument
24290 an extension to Ada 83 not strictly compatible with the reference manual.
24291 GNAT, in order to be exactly compatible with the standard,
24292 does not provide this capability. In HP Ada 83, the
24293 point of this definition is to deal with a call like:
24295 @smallexample @c ada
24296 TO_ADDRESS (16#12777#);
24300 Normally, according to Ada 83 semantics, one would expect this to be
24301 ambiguous, since it matches both the @code{INTEGER} and
24302 @code{UNSIGNED_LONGWORD} forms of @code{TO_ADDRESS}.
24303 However, in HP Ada 83, there is no ambiguity, since the
24304 definition using @i{universal_integer} takes precedence.
24306 In GNAT, since the version with @i{universal_integer} cannot be supplied,
24308 not possible to be 100% compatible. Since there are many programs using
24309 numeric constants for the argument to @code{TO_ADDRESS}, the decision in
24311 to change the name of the function in the @code{UNSIGNED_LONGWORD} case,
24312 so the declarations provided in the GNAT version of @code{AUX_Dec} are:
24314 @smallexample @c ada
24315 function To_Address (X : Integer) return Address;
24316 pragma Pure_Function (To_Address);
24318 function To_Address_Long (X : Unsigned_Longword) return Address;
24319 pragma Pure_Function (To_Address_Long);
24323 This means that programs using @code{TO_ADDRESS} for
24324 @code{UNSIGNED_LONGWORD} must change the name to @code{TO_ADDRESS_LONG}.
24326 @node Tasking and Task-Related Features
24327 @section Tasking and Task-Related Features
24330 This section compares the treatment of tasking in GNAT
24331 and in HP Ada for OpenVMS Alpha.
24332 The GNAT description applies to both Alpha and I64 OpenVMS.
24333 For detailed information on tasking in
24334 HP Ada, see the @cite{HP Ada Language Reference Manual} and the
24335 relevant run-time reference manual.
24338 * Implementation of Tasks in HP Ada for OpenVMS Alpha Systems::
24339 * Assigning Task IDs::
24340 * Task IDs and Delays::
24341 * Task-Related Pragmas::
24342 * Scheduling and Task Priority::
24344 * External Interrupts::
24347 @node Implementation of Tasks in HP Ada for OpenVMS Alpha Systems
24348 @subsection Implementation of Tasks in HP Ada for OpenVMS Alpha Systems
24351 On OpenVMS Alpha systems, each Ada task (except a passive
24352 task) is implemented as a single stream of execution
24353 that is created and managed by the kernel. On these
24354 systems, HP Ada tasking support is based on DECthreads,
24355 an implementation of the POSIX standard for threads.
24357 Also, on OpenVMS Alpha systems, HP Ada tasks and foreign
24358 code that calls DECthreads routines can be used together.
24359 The interaction between Ada tasks and DECthreads routines
24360 can have some benefits. For example when on OpenVMS Alpha,
24361 HP Ada can call C code that is already threaded.
24363 GNAT uses the facilities of DECthreads,
24364 and Ada tasks are mapped to threads.
24366 @node Assigning Task IDs
24367 @subsection Assigning Task IDs
24370 The HP Ada Run-Time Library always assigns @code{%TASK 1} to
24371 the environment task that executes the main program. On
24372 OpenVMS Alpha systems, @code{%TASK 0} is often used for tasks
24373 that have been created but are not yet activated.
24375 On OpenVMS Alpha systems, task IDs are assigned at
24376 activation. On GNAT systems, task IDs are also assigned at
24377 task creation but do not have the same form or values as
24378 task ID values in HP Ada. There is no null task, and the
24379 environment task does not have a specific task ID value.
24381 @node Task IDs and Delays
24382 @subsection Task IDs and Delays
24385 On OpenVMS Alpha systems, tasking delays are implemented
24386 using Timer System Services. The Task ID is used for the
24387 identification of the timer request (the @code{REQIDT} parameter).
24388 If Timers are used in the application take care not to use
24389 @code{0} for the identification, because cancelling such a timer
24390 will cancel all timers and may lead to unpredictable results.
24392 @node Task-Related Pragmas
24393 @subsection Task-Related Pragmas
24396 Ada supplies the pragma @code{TASK_STORAGE}, which allows
24397 specification of the size of the guard area for a task
24398 stack. (The guard area forms an area of memory that has no
24399 read or write access and thus helps in the detection of
24400 stack overflow.) On OpenVMS Alpha systems, if the pragma
24401 @code{TASK_STORAGE} specifies a value of zero, a minimal guard
24402 area is created. In the absence of a pragma @code{TASK_STORAGE},
24403 a default guard area is created.
24405 GNAT supplies the following task-related pragmas:
24408 @item @code{TASK_INFO}
24410 This pragma appears within a task definition and
24411 applies to the task in which it appears. The argument
24412 must be of type @code{SYSTEM.TASK_INFO.TASK_INFO_TYPE}.
24414 @item @code{TASK_STORAGE}
24416 GNAT implements pragma @code{TASK_STORAGE} in the same way as HP Ada.
24417 Both HP Ada and GNAT supply the pragmas @code{PASSIVE},
24418 @code{SUPPRESS}, and @code{VOLATILE}.
24420 @node Scheduling and Task Priority
24421 @subsection Scheduling and Task Priority
24424 HP Ada implements the Ada language requirement that
24425 when two tasks are eligible for execution and they have
24426 different priorities, the lower priority task does not
24427 execute while the higher priority task is waiting. The HP
24428 Ada Run-Time Library keeps a task running until either the
24429 task is suspended or a higher priority task becomes ready.
24431 On OpenVMS Alpha systems, the default strategy is round-
24432 robin with preemption. Tasks of equal priority take turns
24433 at the processor. A task is run for a certain period of
24434 time and then placed at the tail of the ready queue for
24435 its priority level.
24437 HP Ada provides the implementation-defined pragma @code{TIME_SLICE},
24438 which can be used to enable or disable round-robin
24439 scheduling of tasks with the same priority.
24440 See the relevant HP Ada run-time reference manual for
24441 information on using the pragmas to control HP Ada task
24444 GNAT follows the scheduling rules of Annex D (Real-Time
24445 Annex) of the @cite{Ada Reference Manual}. In general, this
24446 scheduling strategy is fully compatible with HP Ada
24447 although it provides some additional constraints (as
24448 fully documented in Annex D).
24449 GNAT implements time slicing control in a manner compatible with
24450 HP Ada 83, by means of the pragma @code{Time_Slice}, whose semantics
24451 are identical to the HP Ada 83 pragma of the same name.
24452 Note that it is not possible to mix GNAT tasking and
24453 HP Ada 83 tasking in the same program, since the two run-time
24454 libraries are not compatible.
24456 @node The Task Stack
24457 @subsection The Task Stack
24460 In HP Ada, a task stack is allocated each time a
24461 non-passive task is activated. As soon as the task is
24462 terminated, the storage for the task stack is deallocated.
24463 If you specify a size of zero (bytes) with @code{T'STORAGE_SIZE},
24464 a default stack size is used. Also, regardless of the size
24465 specified, some additional space is allocated for task
24466 management purposes. On OpenVMS Alpha systems, at least
24467 one page is allocated.
24469 GNAT handles task stacks in a similar manner. In accordance with
24470 the Ada rules, it provides the pragma @code{STORAGE_SIZE} as
24471 an alternative method for controlling the task stack size.
24472 The specification of the attribute @code{T'STORAGE_SIZE} is also
24473 supported in a manner compatible with HP Ada.
24475 @node External Interrupts
24476 @subsection External Interrupts
24479 On HP Ada, external interrupts can be associated with task entries.
24480 GNAT is compatible with HP Ada in its handling of external interrupts.
24482 @node Pragmas and Pragma-Related Features
24483 @section Pragmas and Pragma-Related Features
24486 Both HP Ada and GNAT supply all language-defined pragmas
24487 as specified by the Ada 83 standard. GNAT also supplies all
24488 language-defined pragmas introduced by Ada 95 and Ada 2005.
24489 In addition, GNAT implements the implementation-defined pragmas
24493 @item @code{AST_ENTRY}
24495 @item @code{COMMON_OBJECT}
24497 @item @code{COMPONENT_ALIGNMENT}
24499 @item @code{EXPORT_EXCEPTION}
24501 @item @code{EXPORT_FUNCTION}
24503 @item @code{EXPORT_OBJECT}
24505 @item @code{EXPORT_PROCEDURE}
24507 @item @code{EXPORT_VALUED_PROCEDURE}
24509 @item @code{FLOAT_REPRESENTATION}
24513 @item @code{IMPORT_EXCEPTION}
24515 @item @code{IMPORT_FUNCTION}
24517 @item @code{IMPORT_OBJECT}
24519 @item @code{IMPORT_PROCEDURE}
24521 @item @code{IMPORT_VALUED_PROCEDURE}
24523 @item @code{INLINE_GENERIC}
24525 @item @code{INTERFACE_NAME}
24527 @item @code{LONG_FLOAT}
24529 @item @code{MAIN_STORAGE}
24531 @item @code{PASSIVE}
24533 @item @code{PSECT_OBJECT}
24535 @item @code{SHARE_GENERIC}
24537 @item @code{SUPPRESS_ALL}
24539 @item @code{TASK_STORAGE}
24541 @item @code{TIME_SLICE}
24547 These pragmas are all fully implemented, with the exception of @code{TITLE},
24548 @code{PASSIVE}, and @code{SHARE_GENERIC}, which are
24549 recognized, but which have no
24550 effect in GNAT. The effect of @code{PASSIVE} may be obtained by the
24551 use of Ada protected objects. In GNAT, all generics are inlined.
24553 Unlike HP Ada, the GNAT ``@code{EXPORT_}@i{subprogram}'' pragmas require
24554 a separate subprogram specification which must appear before the
24557 GNAT also supplies a number of implementation-defined pragmas as follows:
24559 @item @code{ABORT_DEFER}
24561 @item @code{ADA_83}
24563 @item @code{ADA_95}
24565 @item @code{ADA_05}
24567 @item @code{ANNOTATE}
24569 @item @code{ASSERT}
24571 @item @code{C_PASS_BY_COPY}
24573 @item @code{CPP_CLASS}
24575 @item @code{CPP_CONSTRUCTOR}
24577 @item @code{CPP_DESTRUCTOR}
24581 @item @code{EXTEND_SYSTEM}
24583 @item @code{LINKER_ALIAS}
24585 @item @code{LINKER_SECTION}
24587 @item @code{MACHINE_ATTRIBUTE}
24589 @item @code{NO_RETURN}
24591 @item @code{PURE_FUNCTION}
24593 @item @code{SOURCE_FILE_NAME}
24595 @item @code{SOURCE_REFERENCE}
24597 @item @code{TASK_INFO}
24599 @item @code{UNCHECKED_UNION}
24601 @item @code{UNIMPLEMENTED_UNIT}
24603 @item @code{UNIVERSAL_DATA}
24605 @item @code{UNSUPPRESS}
24607 @item @code{WARNINGS}
24609 @item @code{WEAK_EXTERNAL}
24613 For full details on these GNAT implementation-defined pragmas,
24614 see @ref{Implementation Defined Pragmas,,, gnat_rm, GNAT Reference
24618 * Restrictions on the Pragma INLINE::
24619 * Restrictions on the Pragma INTERFACE::
24620 * Restrictions on the Pragma SYSTEM_NAME::
24623 @node Restrictions on the Pragma INLINE
24624 @subsection Restrictions on Pragma @code{INLINE}
24627 HP Ada enforces the following restrictions on the pragma @code{INLINE}:
24629 @item Parameters cannot have a task type.
24631 @item Function results cannot be task types, unconstrained
24632 array types, or unconstrained types with discriminants.
24634 @item Bodies cannot declare the following:
24636 @item Subprogram body or stub (imported subprogram is allowed)
24640 @item Generic declarations
24642 @item Instantiations
24646 @item Access types (types derived from access types allowed)
24648 @item Array or record types
24650 @item Dependent tasks
24652 @item Direct recursive calls of subprogram or containing
24653 subprogram, directly or via a renaming
24659 In GNAT, the only restriction on pragma @code{INLINE} is that the
24660 body must occur before the call if both are in the same
24661 unit, and the size must be appropriately small. There are
24662 no other specific restrictions which cause subprograms to
24663 be incapable of being inlined.
24665 @node Restrictions on the Pragma INTERFACE
24666 @subsection Restrictions on Pragma @code{INTERFACE}
24669 The following restrictions on pragma @code{INTERFACE}
24670 are enforced by both HP Ada and GNAT:
24672 @item Languages accepted: Ada, Bliss, C, Fortran, Default.
24673 Default is the default on OpenVMS Alpha systems.
24675 @item Parameter passing: Language specifies default
24676 mechanisms but can be overridden with an @code{EXPORT} pragma.
24679 @item Ada: Use internal Ada rules.
24681 @item Bliss, C: Parameters must be mode @code{in}; cannot be
24682 record or task type. Result cannot be a string, an
24683 array, or a record.
24685 @item Fortran: Parameters cannot have a task type. Result cannot
24686 be a string, an array, or a record.
24691 GNAT is entirely upwards compatible with HP Ada, and in addition allows
24692 record parameters for all languages.
24694 @node Restrictions on the Pragma SYSTEM_NAME
24695 @subsection Restrictions on Pragma @code{SYSTEM_NAME}
24698 For HP Ada for OpenVMS Alpha, the enumeration literal
24699 for the type @code{NAME} is @code{OPENVMS_AXP}.
24700 In GNAT, the enumeration
24701 literal for the type @code{NAME} is @code{SYSTEM_NAME_GNAT}.
24703 @node Library of Predefined Units
24704 @section Library of Predefined Units
24707 A library of predefined units is provided as part of the
24708 HP Ada and GNAT implementations. HP Ada does not provide
24709 the package @code{MACHINE_CODE} but instead recommends importing
24712 The GNAT versions of the HP Ada Run-Time Library (@code{ADA$PREDEFINED:})
24713 units are taken from the OpenVMS Alpha version, not the OpenVMS VAX
24715 The HP Ada Predefined Library units are modified to remove post-Ada 83
24716 incompatibilities and to make them interoperable with GNAT
24717 (@pxref{Changes to DECLIB}, for details).
24718 The units are located in the @file{DECLIB} directory.
24720 The GNAT RTL is contained in
24721 the @file{ADALIB} directory, and
24722 the default search path is set up to find @code{DECLIB} units in preference
24723 to @code{ADALIB} units with the same name (@code{TEXT_IO},
24724 @code{SEQUENTIAL_IO}, and @code{DIRECT_IO}, for example).
24727 * Changes to DECLIB::
24730 @node Changes to DECLIB
24731 @subsection Changes to @code{DECLIB}
24734 The changes made to the HP Ada predefined library for GNAT and post-Ada 83
24735 compatibility are minor and include the following:
24738 @item Adjusting the location of pragmas and record representation
24739 clauses to obey Ada 95 (and thus Ada 2005) rules
24741 @item Adding the proper notation to generic formal parameters
24742 that take unconstrained types in instantiation
24744 @item Adding pragma @code{ELABORATE_BODY} to package specs
24745 that have package bodies not otherwise allowed
24747 @item Replacing occurrences of the identifier ``@code{PROTECTED}'' by
24748 ``@code{PROTECTD}''.
24749 Currently these are found only in the @code{STARLET} package spec.
24751 @item Changing @code{SYSTEM.ADDRESS} to @code{SYSTEM.SHORT_ADDRESS}
24752 where the address size is constrained to 32 bits.
24756 None of the above changes is visible to users.
24762 On OpenVMS Alpha, HP Ada provides the following strongly-typed bindings:
24765 @item Command Language Interpreter (CLI interface)
24767 @item DECtalk Run-Time Library (DTK interface)
24769 @item Librarian utility routines (LBR interface)
24771 @item General Purpose Run-Time Library (LIB interface)
24773 @item Math Run-Time Library (MTH interface)
24775 @item National Character Set Run-Time Library (NCS interface)
24777 @item Compiled Code Support Run-Time Library (OTS interface)
24779 @item Parallel Processing Run-Time Library (PPL interface)
24781 @item Screen Management Run-Time Library (SMG interface)
24783 @item Sort Run-Time Library (SOR interface)
24785 @item String Run-Time Library (STR interface)
24787 @item STARLET System Library
24790 @item X Window System Version 11R4 and 11R5 (X, XLIB interface)
24792 @item X Windows Toolkit (XT interface)
24794 @item X/Motif Version 1.1.3 and 1.2 (XM interface)
24798 GNAT provides implementations of these HP bindings in the @code{DECLIB}
24799 directory, on both the Alpha and I64 OpenVMS platforms.
24801 The X/Motif bindings used to build @code{DECLIB} are whatever versions are
24803 HP Ada @file{ADA$PREDEFINED} directory with extension @file{.ADC}.
24804 A pragma @code{Linker_Options} has been added to packages @code{Xm},
24805 @code{Xt}, and @code{X_Lib}
24806 causing the default X/Motif sharable image libraries to be linked in. This
24807 is done via options files named @file{xm.opt}, @file{xt.opt}, and
24808 @file{x_lib.opt} (also located in the @file{DECLIB} directory).
24810 It may be necessary to edit these options files to update or correct the
24811 library names if, for example, the newer X/Motif bindings from
24812 @file{ADA$EXAMPLES}
24813 had been (previous to installing GNAT) copied and renamed to supersede the
24814 default @file{ADA$PREDEFINED} versions.
24817 * Shared Libraries and Options Files::
24818 * Interfaces to C::
24821 @node Shared Libraries and Options Files
24822 @subsection Shared Libraries and Options Files
24825 When using the HP Ada
24826 predefined X and Motif bindings, the linking with their sharable images is
24827 done automatically by @command{GNAT LINK}.
24828 When using other X and Motif bindings, you need
24829 to add the corresponding sharable images to the command line for
24830 @code{GNAT LINK}. When linking with shared libraries, or with
24831 @file{.OPT} files, you must
24832 also add them to the command line for @command{GNAT LINK}.
24834 A shared library to be used with GNAT is built in the same way as other
24835 libraries under VMS. The VMS Link command can be used in standard fashion.
24837 @node Interfaces to C
24838 @subsection Interfaces to C
24842 provides the following Ada types and operations:
24845 @item C types package (@code{C_TYPES})
24847 @item C strings (@code{C_TYPES.NULL_TERMINATED})
24849 @item Other_types (@code{SHORT_INT})
24853 Interfacing to C with GNAT, you can use the above approach
24854 described for HP Ada or the facilities of Annex B of
24855 the @cite{Ada Reference Manual} (packages @code{INTERFACES.C},
24856 @code{INTERFACES.C.STRINGS} and @code{INTERFACES.C.POINTERS}). For more
24857 information, see @ref{Interfacing to C,,, gnat_rm, GNAT Reference Manual}.
24859 The @option{-gnatF} qualifier forces default and explicit
24860 @code{External_Name} parameters in pragmas @code{Import} and @code{Export}
24861 to be uppercased for compatibility with the default behavior
24862 of HP C. The qualifier has no effect on @code{Link_Name} parameters.
24864 @node Main Program Definition
24865 @section Main Program Definition
24868 The following section discusses differences in the
24869 definition of main programs on HP Ada and GNAT.
24870 On HP Ada, main programs are defined to meet the
24871 following conditions:
24873 @item Procedure with no formal parameters (returns @code{0} upon
24876 @item Procedure with no formal parameters (returns @code{42} when
24877 an unhandled exception is raised)
24879 @item Function with no formal parameters whose returned value
24880 is of a discrete type
24882 @item Procedure with one @code{out} formal of a discrete type for
24883 which a specification of pragma @code{EXPORT_VALUED_PROCEDURE} is given.
24888 When declared with the pragma @code{EXPORT_VALUED_PROCEDURE},
24889 a main function or main procedure returns a discrete
24890 value whose size is less than 64 bits (32 on VAX systems),
24891 the value is zero- or sign-extended as appropriate.
24892 On GNAT, main programs are defined as follows:
24894 @item Must be a non-generic, parameterless subprogram that
24895 is either a procedure or function returning an Ada
24896 @code{STANDARD.INTEGER} (the predefined type)
24898 @item Cannot be a generic subprogram or an instantiation of a
24902 @node Implementation-Defined Attributes
24903 @section Implementation-Defined Attributes
24906 GNAT provides all HP Ada implementation-defined
24909 @node Compiler and Run-Time Interfacing
24910 @section Compiler and Run-Time Interfacing
24913 HP Ada provides the following qualifiers to pass options to the linker
24916 @item @option{/WAIT} and @option{/SUBMIT}
24918 @item @option{/COMMAND}
24920 @item @option{/@r{[}NO@r{]}MAP}
24922 @item @option{/OUTPUT=@var{file-spec}}
24924 @item @option{/@r{[}NO@r{]}DEBUG} and @option{/@r{[}NO@r{]}TRACEBACK}
24928 To pass options to the linker, GNAT provides the following
24932 @item @option{/EXECUTABLE=@var{exec-name}}
24934 @item @option{/VERBOSE}
24936 @item @option{/@r{[}NO@r{]}DEBUG} and @option{/@r{[}NO@r{]}TRACEBACK}
24940 For more information on these switches, see
24941 @ref{Switches for gnatlink}.
24942 In HP Ada, the command-line switch @option{/OPTIMIZE} is available
24943 to control optimization. HP Ada also supplies the
24946 @item @code{OPTIMIZE}
24948 @item @code{INLINE}
24950 @item @code{INLINE_GENERIC}
24952 @item @code{SUPPRESS_ALL}
24954 @item @code{PASSIVE}
24958 In GNAT, optimization is controlled strictly by command
24959 line parameters, as described in the corresponding section of this guide.
24960 The HP pragmas for control of optimization are
24961 recognized but ignored.
24963 Note that in GNAT, the default is optimization off, whereas in HP Ada
24964 the default is that optimization is turned on.
24966 @node Program Compilation and Library Management
24967 @section Program Compilation and Library Management
24970 HP Ada and GNAT provide a comparable set of commands to
24971 build programs. HP Ada also provides a program library,
24972 which is a concept that does not exist on GNAT. Instead,
24973 GNAT provides directories of sources that are compiled as
24976 The following table summarizes
24977 the HP Ada commands and provides
24978 equivalent GNAT commands. In this table, some GNAT
24979 equivalents reflect the fact that GNAT does not use the
24980 concept of a program library. Instead, it uses a model
24981 in which collections of source and object files are used
24982 in a manner consistent with other languages like C and
24983 Fortran. Therefore, standard system file commands are used
24984 to manipulate these elements. Those GNAT commands are marked with
24986 Note that, unlike HP Ada, none of the GNAT commands accepts wild cards.
24989 @multitable @columnfractions .35 .65
24991 @item @emph{HP Ada Command}
24992 @tab @emph{GNAT Equivalent / Description}
24994 @item @command{ADA}
24995 @tab @command{GNAT COMPILE}@*
24996 Invokes the compiler to compile one or more Ada source files.
24998 @item @command{ACS ATTACH}@*
24999 @tab [No equivalent]@*
25000 Switches control of terminal from current process running the program
25003 @item @command{ACS CHECK}
25004 @tab @command{GNAT MAKE /DEPENDENCY_LIST}@*
25005 Forms the execution closure of one
25006 or more compiled units and checks completeness and currency.
25008 @item @command{ACS COMPILE}
25009 @tab @command{GNAT MAKE /ACTIONS=COMPILE}@*
25010 Forms the execution closure of one or
25011 more specified units, checks completeness and currency,
25012 identifies units that have revised source files, compiles same,
25013 and recompiles units that are or will become obsolete.
25014 Also completes incomplete generic instantiations.
25016 @item @command{ACS COPY FOREIGN}
25018 Copies a foreign object file into the program library as a
25021 @item @command{ACS COPY UNIT}
25023 Copies a compiled unit from one program library to another.
25025 @item @command{ACS CREATE LIBRARY}
25026 @tab Create /directory (*)@*
25027 Creates a program library.
25029 @item @command{ACS CREATE SUBLIBRARY}
25030 @tab Create /directory (*)@*
25031 Creates a program sublibrary.
25033 @item @command{ACS DELETE LIBRARY}
25035 Deletes a program library and its contents.
25037 @item @command{ACS DELETE SUBLIBRARY}
25039 Deletes a program sublibrary and its contents.
25041 @item @command{ACS DELETE UNIT}
25042 @tab Delete file (*)@*
25043 On OpenVMS systems, deletes one or more compiled units from
25044 the current program library.
25046 @item @command{ACS DIRECTORY}
25047 @tab Directory (*)@*
25048 On OpenVMS systems, lists units contained in the current
25051 @item @command{ACS ENTER FOREIGN}
25053 Allows the import of a foreign body as an Ada library
25054 spec and enters a reference to a pointer.
25056 @item @command{ACS ENTER UNIT}
25058 Enters a reference (pointer) from the current program library to
25059 a unit compiled into another program library.
25061 @item @command{ACS EXIT}
25062 @tab [No equivalent]@*
25063 Exits from the program library manager.
25065 @item @command{ACS EXPORT}
25067 Creates an object file that contains system-specific object code
25068 for one or more units. With GNAT, object files can simply be copied
25069 into the desired directory.
25071 @item @command{ACS EXTRACT SOURCE}
25073 Allows access to the copied source file for each Ada compilation unit
25075 @item @command{ACS HELP}
25076 @tab @command{HELP GNAT}@*
25077 Provides online help.
25079 @item @command{ACS LINK}
25080 @tab @command{GNAT LINK}@*
25081 Links an object file containing Ada units into an executable file.
25083 @item @command{ACS LOAD}
25085 Loads (partially compiles) Ada units into the program library.
25086 Allows loading a program from a collection of files into a library
25087 without knowing the relationship among units.
25089 @item @command{ACS MERGE}
25091 Merges into the current program library, one or more units from
25092 another library where they were modified.
25094 @item @command{ACS RECOMPILE}
25095 @tab @command{GNAT MAKE /ACTIONS=COMPILE}@*
25096 Recompiles from external or copied source files any obsolete
25097 unit in the closure. Also, completes any incomplete generic
25100 @item @command{ACS REENTER}
25101 @tab @command{GNAT MAKE}@*
25102 Reenters current references to units compiled after last entered
25103 with the @command{ACS ENTER UNIT} command.
25105 @item @command{ACS SET LIBRARY}
25106 @tab Set default (*)@*
25107 Defines a program library to be the compilation context as well
25108 as the target library for compiler output and commands in general.
25110 @item @command{ACS SET PRAGMA}
25111 @tab Edit @file{gnat.adc} (*)@*
25112 Redefines specified values of the library characteristics
25113 @code{LONG_ FLOAT}, @code{MEMORY_SIZE}, @code{SYSTEM_NAME},
25114 and @code{Float_Representation}.
25116 @item @command{ACS SET SOURCE}
25117 @tab Define @code{ADA_INCLUDE_PATH} path (*)@*
25118 Defines the source file search list for the @command{ACS COMPILE} command.
25120 @item @command{ACS SHOW LIBRARY}
25121 @tab Directory (*)@*
25122 Lists information about one or more program libraries.
25124 @item @command{ACS SHOW PROGRAM}
25125 @tab [No equivalent]@*
25126 Lists information about the execution closure of one or
25127 more units in the program library.
25129 @item @command{ACS SHOW SOURCE}
25130 @tab Show logical @code{ADA_INCLUDE_PATH}@*
25131 Shows the source file search used when compiling units.
25133 @item @command{ACS SHOW VERSION}
25134 @tab Compile with @option{VERBOSE} option
25135 Displays the version number of the compiler and program library
25138 @item @command{ACS SPAWN}
25139 @tab [No equivalent]@*
25140 Creates a subprocess of the current process (same as @command{DCL SPAWN}
25143 @item @command{ACS VERIFY}
25144 @tab [No equivalent]@*
25145 Performs a series of consistency checks on a program library to
25146 determine whether the library structure and library files are in
25153 @section Input-Output
25156 On OpenVMS Alpha systems, HP Ada uses OpenVMS Record
25157 Management Services (RMS) to perform operations on
25161 HP Ada and GNAT predefine an identical set of input-
25162 output packages. To make the use of the
25163 generic @code{TEXT_IO} operations more convenient, HP Ada
25164 provides predefined library packages that instantiate the
25165 integer and floating-point operations for the predefined
25166 integer and floating-point types as shown in the following table.
25168 @multitable @columnfractions .45 .55
25169 @item @emph{Package Name} @tab Instantiation
25171 @item @code{INTEGER_TEXT_IO}
25172 @tab @code{INTEGER_IO(INTEGER)}
25174 @item @code{SHORT_INTEGER_TEXT_IO}
25175 @tab @code{INTEGER_IO(SHORT_INTEGER)}
25177 @item @code{SHORT_SHORT_INTEGER_TEXT_IO}
25178 @tab @code{INTEGER_IO(SHORT_SHORT_INTEGER)}
25180 @item @code{FLOAT_TEXT_IO}
25181 @tab @code{FLOAT_IO(FLOAT)}
25183 @item @code{LONG_FLOAT_TEXT_IO}
25184 @tab @code{FLOAT_IO(LONG_FLOAT)}
25188 The HP Ada predefined packages and their operations
25189 are implemented using OpenVMS Alpha files and input-output
25190 facilities. HP Ada supports asynchronous input-output on OpenVMS Alpha.
25191 Familiarity with the following is recommended:
25193 @item RMS file organizations and access methods
25195 @item OpenVMS file specifications and directories
25197 @item OpenVMS File Definition Language (FDL)
25201 GNAT provides I/O facilities that are completely
25202 compatible with HP Ada. The distribution includes the
25203 standard HP Ada versions of all I/O packages, operating
25204 in a manner compatible with HP Ada. In particular, the
25205 following packages are by default the HP Ada (Ada 83)
25206 versions of these packages rather than the renamings
25207 suggested in Annex J of the Ada Reference Manual:
25209 @item @code{TEXT_IO}
25211 @item @code{SEQUENTIAL_IO}
25213 @item @code{DIRECT_IO}
25217 The use of the standard child package syntax (for
25218 example, @code{ADA.TEXT_IO}) retrieves the post-Ada 83 versions of these
25220 GNAT provides HP-compatible predefined instantiations
25221 of the @code{TEXT_IO} packages, and also
25222 provides the standard predefined instantiations required
25223 by the @cite{Ada Reference Manual}.
25225 For further information on how GNAT interfaces to the file
25226 system or how I/O is implemented in programs written in
25227 mixed languages, see @ref{Implementation of the Standard I/O,,,
25228 gnat_rm, GNAT Reference Manual}.
25229 This chapter covers the following:
25231 @item Standard I/O packages
25233 @item @code{FORM} strings
25235 @item @code{ADA.DIRECT_IO}
25237 @item @code{ADA.SEQUENTIAL_IO}
25239 @item @code{ADA.TEXT_IO}
25241 @item Stream pointer positioning
25243 @item Reading and writing non-regular files
25245 @item @code{GET_IMMEDIATE}
25247 @item Treating @code{TEXT_IO} files as streams
25254 @node Implementation Limits
25255 @section Implementation Limits
25258 The following table lists implementation limits for HP Ada
25260 @multitable @columnfractions .60 .20 .20
25262 @item @emph{Compilation Parameter}
25267 @item In a subprogram or entry declaration, maximum number of
25268 formal parameters that are of an unconstrained record type
25273 @item Maximum identifier length (number of characters)
25278 @item Maximum number of characters in a source line
25283 @item Maximum collection size (number of bytes)
25288 @item Maximum number of discriminants for a record type
25293 @item Maximum number of formal parameters in an entry or
25294 subprogram declaration
25299 @item Maximum number of dimensions in an array type
25304 @item Maximum number of library units and subunits in a compilation.
25309 @item Maximum number of library units and subunits in an execution.
25314 @item Maximum number of objects declared with the pragma @code{COMMON_OBJECT}
25315 or @code{PSECT_OBJECT}
25320 @item Maximum number of enumeration literals in an enumeration type
25326 @item Maximum number of lines in a source file
25331 @item Maximum number of bits in any object
25336 @item Maximum size of the static portion of a stack frame (approximate)
25341 @node Tools and Utilities
25342 @section Tools and Utilities
25345 The following table lists some of the OpenVMS development tools
25346 available for HP Ada, and the corresponding tools for
25347 use with @value{EDITION} on Alpha and I64 platforms.
25348 Aside from the debugger, all the OpenVMS tools identified are part
25349 of the DECset package.
25352 @c Specify table in TeX since Texinfo does a poor job
25356 \settabs\+Language-Sensitive Editor\quad
25357 &Product with HP Ada\quad
25360 &\it Product with HP Ada
25361 & \it Product with GNAT Pro\cr
25363 \+Code Management System
25367 \+Language-Sensitive Editor
25369 & emacs or HP LSE (Alpha)\cr
25379 & OpenVMS Debug (I64)\cr
25381 \+Source Code Analyzer /
25398 \+Coverage Analyzer
25402 \+Module Management
25404 & Not applicable\cr
25414 @c This is the Texinfo version of the table. It renders poorly in pdf, hence
25415 @c the TeX version above for the printed version
25417 @c @multitable @columnfractions .3 .4 .4
25418 @multitable {Source Code Analyzer /}{Tool with HP Ada}{Tool with GNAT Pro}
25420 @tab @i{Tool with HP Ada}
25421 @tab @i{Tool with @value{EDITION}}
25422 @item Code Management@*System
25425 @item Language-Sensitive@*Editor
25427 @tab emacs or HP LSE (Alpha)
25436 @tab OpenVMS Debug (I64)
25437 @item Source Code Analyzer /@*Cross Referencer
25441 @tab HP Digital Test@*Manager (DTM)
25443 @item Performance and@*Coverage Analyzer
25446 @item Module Management@*System
25448 @tab Not applicable
25455 @c **************************************
25456 @node Platform-Specific Information for the Run-Time Libraries
25457 @appendix Platform-Specific Information for the Run-Time Libraries
25458 @cindex Tasking and threads libraries
25459 @cindex Threads libraries and tasking
25460 @cindex Run-time libraries (platform-specific information)
25463 The GNAT run-time implementation may vary with respect to both the
25464 underlying threads library and the exception handling scheme.
25465 For threads support, one or more of the following are supplied:
25467 @item @b{native threads library}, a binding to the thread package from
25468 the underlying operating system
25470 @item @b{pthreads library} (Sparc Solaris only), a binding to the Solaris
25471 POSIX thread package
25475 For exception handling, either or both of two models are supplied:
25477 @item @b{Zero-Cost Exceptions} (``ZCX''),@footnote{
25478 Most programs should experience a substantial speed improvement by
25479 being compiled with a ZCX run-time.
25480 This is especially true for
25481 tasking applications or applications with many exception handlers.}
25482 @cindex Zero-Cost Exceptions
25483 @cindex ZCX (Zero-Cost Exceptions)
25484 which uses binder-generated tables that
25485 are interrogated at run time to locate a handler
25487 @item @b{setjmp / longjmp} (``SJLJ''),
25488 @cindex setjmp/longjmp Exception Model
25489 @cindex SJLJ (setjmp/longjmp Exception Model)
25490 which uses dynamically-set data to establish
25491 the set of handlers
25495 This appendix summarizes which combinations of threads and exception support
25496 are supplied on various GNAT platforms.
25497 It then shows how to select a particular library either
25498 permanently or temporarily,
25499 explains the properties of (and tradeoffs among) the various threads
25500 libraries, and provides some additional
25501 information about several specific platforms.
25504 * Summary of Run-Time Configurations::
25505 * Specifying a Run-Time Library::
25506 * Choosing the Scheduling Policy::
25507 * Solaris-Specific Considerations::
25508 * Linux-Specific Considerations::
25509 * AIX-Specific Considerations::
25510 * Irix-Specific Considerations::
25513 @node Summary of Run-Time Configurations
25514 @section Summary of Run-Time Configurations
25516 @multitable @columnfractions .30 .70
25517 @item @b{alpha-openvms}
25518 @item @code{@ @ }@i{rts-native (default)}
25519 @item @code{@ @ @ @ }Tasking @tab native VMS threads
25520 @item @code{@ @ @ @ }Exceptions @tab ZCX
25522 @item @b{alpha-tru64}
25523 @item @code{@ @ }@i{rts-native (default)}
25524 @item @code{@ @ @ @ }Tasking @tab native TRU64 threads
25525 @item @code{@ @ @ @ }Exceptions @tab ZCX
25527 @item @code{@ @ }@i{rts-sjlj}
25528 @item @code{@ @ @ @ }Tasking @tab native TRU64 threads
25529 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25531 @item @b{ia64-hp_linux}
25532 @item @code{@ @ }@i{rts-native (default)}
25533 @item @code{@ @ @ @ }Tasking @tab pthread library
25534 @item @code{@ @ @ @ }Exceptions @tab ZCX
25536 @item @b{ia64-hpux}
25537 @item @code{@ @ }@i{rts-native (default)}
25538 @item @code{@ @ @ @ }Tasking @tab native HP-UX threads
25539 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25541 @item @b{ia64-openvms}
25542 @item @code{@ @ }@i{rts-native (default)}
25543 @item @code{@ @ @ @ }Tasking @tab native VMS threads
25544 @item @code{@ @ @ @ }Exceptions @tab ZCX
25546 @item @b{ia64-sgi_linux}
25547 @item @code{@ @ }@i{rts-native (default)}
25548 @item @code{@ @ @ @ }Tasking @tab pthread library
25549 @item @code{@ @ @ @ }Exceptions @tab ZCX
25551 @item @b{mips-irix}
25552 @item @code{@ @ }@i{rts-native (default)}
25553 @item @code{@ @ @ @ }Tasking @tab native IRIX threads
25554 @item @code{@ @ @ @ }Exceptions @tab ZCX
25557 @item @code{@ @ }@i{rts-native (default)}
25558 @item @code{@ @ @ @ }Tasking @tab native HP-UX threads
25559 @item @code{@ @ @ @ }Exceptions @tab ZCX
25561 @item @code{@ @ }@i{rts-sjlj}
25562 @item @code{@ @ @ @ }Tasking @tab native HP-UX threads
25563 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25566 @item @code{@ @ }@i{rts-native (default)}
25567 @item @code{@ @ @ @ }Tasking @tab native AIX threads
25568 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25570 @item @b{ppc-darwin}
25571 @item @code{@ @ }@i{rts-native (default)}
25572 @item @code{@ @ @ @ }Tasking @tab native MacOS threads
25573 @item @code{@ @ @ @ }Exceptions @tab ZCX
25575 @item @b{sparc-solaris} @tab
25576 @item @code{@ @ }@i{rts-native (default)}
25577 @item @code{@ @ @ @ }Tasking @tab native Solaris threads library
25578 @item @code{@ @ @ @ }Exceptions @tab ZCX
25580 @item @code{@ @ }@i{rts-pthread}
25581 @item @code{@ @ @ @ }Tasking @tab pthread library
25582 @item @code{@ @ @ @ }Exceptions @tab ZCX
25584 @item @code{@ @ }@i{rts-sjlj}
25585 @item @code{@ @ @ @ }Tasking @tab native Solaris threads library
25586 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25588 @item @b{sparc64-solaris} @tab
25589 @item @code{@ @ }@i{rts-native (default)}
25590 @item @code{@ @ @ @ }Tasking @tab native Solaris threads library
25591 @item @code{@ @ @ @ }Exceptions @tab ZCX
25593 @item @b{x86-linux}
25594 @item @code{@ @ }@i{rts-native (default)}
25595 @item @code{@ @ @ @ }Tasking @tab pthread library
25596 @item @code{@ @ @ @ }Exceptions @tab ZCX
25598 @item @code{@ @ }@i{rts-sjlj}
25599 @item @code{@ @ @ @ }Tasking @tab pthread library
25600 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25603 @item @code{@ @ }@i{rts-native (default)}
25604 @item @code{@ @ @ @ }Tasking @tab native LynxOS threads
25605 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25607 @item @b{x86-solaris}
25608 @item @code{@ @ }@i{rts-native (default)}
25609 @item @code{@ @ @ @ }Tasking @tab native Solaris threads
25610 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25612 @item @b{x86-windows}
25613 @item @code{@ @ }@i{rts-native (default)}
25614 @item @code{@ @ @ @ }Tasking @tab native Win32 threads
25615 @item @code{@ @ @ @ }Exceptions @tab ZCX
25617 @item @code{@ @ }@i{rts-sjlj (default)}
25618 @item @code{@ @ @ @ }Tasking @tab native Win32 threads
25619 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25621 @item @b{x86_64-linux}
25622 @item @code{@ @ }@i{rts-native (default)}
25623 @item @code{@ @ @ @ }Tasking @tab pthread library
25624 @item @code{@ @ @ @ }Exceptions @tab ZCX
25626 @item @code{@ @ }@i{rts-sjlj}
25627 @item @code{@ @ @ @ }Tasking @tab pthread library
25628 @item @code{@ @ @ @ }Exceptions @tab SJLJ
25632 @node Specifying a Run-Time Library
25633 @section Specifying a Run-Time Library
25636 The @file{adainclude} subdirectory containing the sources of the GNAT
25637 run-time library, and the @file{adalib} subdirectory containing the
25638 @file{ALI} files and the static and/or shared GNAT library, are located
25639 in the gcc target-dependent area:
25642 target=$prefix/lib/gcc/gcc-@i{dumpmachine}/gcc-@i{dumpversion}/
25646 As indicated above, on some platforms several run-time libraries are supplied.
25647 These libraries are installed in the target dependent area and
25648 contain a complete source and binary subdirectory. The detailed description
25649 below explains the differences between the different libraries in terms of
25650 their thread support.
25652 The default run-time library (when GNAT is installed) is @emph{rts-native}.
25653 This default run time is selected by the means of soft links.
25654 For example on x86-linux:
25660 +--- adainclude----------+
25662 +--- adalib-----------+ |
25664 +--- rts-native | |
25666 | +--- adainclude <---+
25668 | +--- adalib <----+
25679 If the @i{rts-sjlj} library is to be selected on a permanent basis,
25680 these soft links can be modified with the following commands:
25684 $ rm -f adainclude adalib
25685 $ ln -s rts-sjlj/adainclude adainclude
25686 $ ln -s rts-sjlj/adalib adalib
25690 Alternatively, you can specify @file{rts-sjlj/adainclude} in the file
25691 @file{$target/ada_source_path} and @file{rts-sjlj/adalib} in
25692 @file{$target/ada_object_path}.
25694 Selecting another run-time library temporarily can be
25695 achieved by using the @option{--RTS} switch, e.g., @option{--RTS=sjlj}
25696 @cindex @option{--RTS} option
25698 @node Choosing the Scheduling Policy
25699 @section Choosing the Scheduling Policy
25702 When using a POSIX threads implementation, you have a choice of several
25703 scheduling policies: @code{SCHED_FIFO},
25704 @cindex @code{SCHED_FIFO} scheduling policy
25706 @cindex @code{SCHED_RR} scheduling policy
25707 and @code{SCHED_OTHER}.
25708 @cindex @code{SCHED_OTHER} scheduling policy
25709 Typically, the default is @code{SCHED_OTHER}, while using @code{SCHED_FIFO}
25710 or @code{SCHED_RR} requires special (e.g., root) privileges.
25712 By default, GNAT uses the @code{SCHED_OTHER} policy. To specify
25714 @cindex @code{SCHED_FIFO} scheduling policy
25715 you can use one of the following:
25719 @code{pragma Time_Slice (0.0)}
25720 @cindex pragma Time_Slice
25722 the corresponding binder option @option{-T0}
25723 @cindex @option{-T0} option
25725 @code{pragma Task_Dispatching_Policy (FIFO_Within_Priorities)}
25726 @cindex pragma Task_Dispatching_Policy
25730 To specify @code{SCHED_RR},
25731 @cindex @code{SCHED_RR} scheduling policy
25732 you should use @code{pragma Time_Slice} with a
25733 value greater than @code{0.0}, or else use the corresponding @option{-T}
25736 @node Solaris-Specific Considerations
25737 @section Solaris-Specific Considerations
25738 @cindex Solaris Sparc threads libraries
25741 This section addresses some topics related to the various threads libraries
25745 * Solaris Threads Issues::
25748 @node Solaris Threads Issues
25749 @subsection Solaris Threads Issues
25752 GNAT under Solaris/Sparc 32 bits comes with an alternate tasking run-time
25753 library based on POSIX threads --- @emph{rts-pthread}.
25754 @cindex rts-pthread threads library
25755 This run-time library has the advantage of being mostly shared across all
25756 POSIX-compliant thread implementations, and it also provides under
25757 @w{Solaris 8} the @code{PTHREAD_PRIO_INHERIT}
25758 @cindex @code{PTHREAD_PRIO_INHERIT} policy (under rts-pthread)
25759 and @code{PTHREAD_PRIO_PROTECT}
25760 @cindex @code{PTHREAD_PRIO_PROTECT} policy (under rts-pthread)
25761 semantics that can be selected using the predefined pragma
25762 @code{Locking_Policy}
25763 @cindex pragma Locking_Policy (under rts-pthread)
25765 @code{Inheritance_Locking} and @code{Ceiling_Locking} as the policy.
25766 @cindex @code{Inheritance_Locking} (under rts-pthread)
25767 @cindex @code{Ceiling_Locking} (under rts-pthread)
25769 As explained above, the native run-time library is based on the Solaris thread
25770 library (@code{libthread}) and is the default library.
25772 When the Solaris threads library is used (this is the default), programs
25773 compiled with GNAT can automatically take advantage of
25774 and can thus execute on multiple processors.
25775 The user can alternatively specify a processor on which the program should run
25776 to emulate a single-processor system. The multiprocessor / uniprocessor choice
25778 setting the environment variable @env{GNAT_PROCESSOR}
25779 @cindex @env{GNAT_PROCESSOR} environment variable (on Sparc Solaris)
25780 to one of the following:
25784 Use the default configuration (run the program on all
25785 available processors) - this is the same as having @code{GNAT_PROCESSOR}
25789 Let the run-time implementation choose one processor and run the program on
25792 @item 0 .. Last_Proc
25793 Run the program on the specified processor.
25794 @code{Last_Proc} is equal to @code{_SC_NPROCESSORS_CONF - 1}
25795 (where @code{_SC_NPROCESSORS_CONF} is a system variable).
25798 @node Linux-Specific Considerations
25799 @section Linux-Specific Considerations
25800 @cindex Linux threads libraries
25803 On GNU/Linux without NPTL support (usually system with GNU C Library
25804 older than 2.3), the signal model is not POSIX compliant, which means
25805 that to send a signal to the process, you need to send the signal to all
25806 threads, e.g.@: by using @code{killpg()}.
25808 @node AIX-Specific Considerations
25809 @section AIX-Specific Considerations
25810 @cindex AIX resolver library
25813 On AIX, the resolver library initializes some internal structure on
25814 the first call to @code{get*by*} functions, which are used to implement
25815 @code{GNAT.Sockets.Get_Host_By_Name} and
25816 @code{GNAT.Sockets.Get_Host_By_Address}.
25817 If such initialization occurs within an Ada task, and the stack size for
25818 the task is the default size, a stack overflow may occur.
25820 To avoid this overflow, the user should either ensure that the first call
25821 to @code{GNAT.Sockets.Get_Host_By_Name} or
25822 @code{GNAT.Sockets.Get_Host_By_Addrss}
25823 occurs in the environment task, or use @code{pragma Storage_Size} to
25824 specify a sufficiently large size for the stack of the task that contains
25827 @node Irix-Specific Considerations
25828 @section Irix-Specific Considerations
25829 @cindex Irix libraries
25832 The GCC support libraries coming with the Irix compiler have moved to
25833 their canonical place with respect to the general Irix ABI related
25834 conventions. Running applications built with the default shared GNAT
25835 run-time now requires the LD_LIBRARY_PATH environment variable to
25836 include this location. A possible way to achieve this is to issue the
25837 following command line on a bash prompt:
25841 $ LD_LIBRARY_PATH=$LD_LIBRARY_PATH:`dirname \`gcc --print-file-name=libgcc_s.so\``
25845 @c *******************************
25846 @node Example of Binder Output File
25847 @appendix Example of Binder Output File
25850 This Appendix displays the source code for @command{gnatbind}'s output
25851 file generated for a simple ``Hello World'' program.
25852 Comments have been added for clarification purposes.
25854 @smallexample @c adanocomment
25858 -- The package is called Ada_Main unless this name is actually used
25859 -- as a unit name in the partition, in which case some other unique
25863 package ada_main is
25865 Elab_Final_Code : Integer;
25866 pragma Import (C, Elab_Final_Code, "__gnat_inside_elab_final_code");
25868 -- The main program saves the parameters (argument count,
25869 -- argument values, environment pointer) in global variables
25870 -- for later access by other units including
25871 -- Ada.Command_Line.
25873 gnat_argc : Integer;
25874 gnat_argv : System.Address;
25875 gnat_envp : System.Address;
25877 -- The actual variables are stored in a library routine. This
25878 -- is useful for some shared library situations, where there
25879 -- are problems if variables are not in the library.
25881 pragma Import (C, gnat_argc);
25882 pragma Import (C, gnat_argv);
25883 pragma Import (C, gnat_envp);
25885 -- The exit status is similarly an external location
25887 gnat_exit_status : Integer;
25888 pragma Import (C, gnat_exit_status);
25890 GNAT_Version : constant String :=
25891 "GNAT Version: 6.0.0w (20061115)";
25892 pragma Export (C, GNAT_Version, "__gnat_version");
25894 -- This is the generated adafinal routine that performs
25895 -- finalization at the end of execution. In the case where
25896 -- Ada is the main program, this main program makes a call
25897 -- to adafinal at program termination.
25899 procedure adafinal;
25900 pragma Export (C, adafinal, "adafinal");
25902 -- This is the generated adainit routine that performs
25903 -- initialization at the start of execution. In the case
25904 -- where Ada is the main program, this main program makes
25905 -- a call to adainit at program startup.
25908 pragma Export (C, adainit, "adainit");
25910 -- This routine is called at the start of execution. It is
25911 -- a dummy routine that is used by the debugger to breakpoint
25912 -- at the start of execution.
25914 procedure Break_Start;
25915 pragma Import (C, Break_Start, "__gnat_break_start");
25917 -- This is the actual generated main program (it would be
25918 -- suppressed if the no main program switch were used). As
25919 -- required by standard system conventions, this program has
25920 -- the external name main.
25924 argv : System.Address;
25925 envp : System.Address)
25927 pragma Export (C, main, "main");
25929 -- The following set of constants give the version
25930 -- identification values for every unit in the bound
25931 -- partition. This identification is computed from all
25932 -- dependent semantic units, and corresponds to the
25933 -- string that would be returned by use of the
25934 -- Body_Version or Version attributes.
25936 type Version_32 is mod 2 ** 32;
25937 u00001 : constant Version_32 := 16#7880BEB3#;
25938 u00002 : constant Version_32 := 16#0D24CBD0#;
25939 u00003 : constant Version_32 := 16#3283DBEB#;
25940 u00004 : constant Version_32 := 16#2359F9ED#;
25941 u00005 : constant Version_32 := 16#664FB847#;
25942 u00006 : constant Version_32 := 16#68E803DF#;
25943 u00007 : constant Version_32 := 16#5572E604#;
25944 u00008 : constant Version_32 := 16#46B173D8#;
25945 u00009 : constant Version_32 := 16#156A40CF#;
25946 u00010 : constant Version_32 := 16#033DABE0#;
25947 u00011 : constant Version_32 := 16#6AB38FEA#;
25948 u00012 : constant Version_32 := 16#22B6217D#;
25949 u00013 : constant Version_32 := 16#68A22947#;
25950 u00014 : constant Version_32 := 16#18CC4A56#;
25951 u00015 : constant Version_32 := 16#08258E1B#;
25952 u00016 : constant Version_32 := 16#367D5222#;
25953 u00017 : constant Version_32 := 16#20C9ECA4#;
25954 u00018 : constant Version_32 := 16#50D32CB6#;
25955 u00019 : constant Version_32 := 16#39A8BB77#;
25956 u00020 : constant Version_32 := 16#5CF8FA2B#;
25957 u00021 : constant Version_32 := 16#2F1EB794#;
25958 u00022 : constant Version_32 := 16#31AB6444#;
25959 u00023 : constant Version_32 := 16#1574B6E9#;
25960 u00024 : constant Version_32 := 16#5109C189#;
25961 u00025 : constant Version_32 := 16#56D770CD#;
25962 u00026 : constant Version_32 := 16#02F9DE3D#;
25963 u00027 : constant Version_32 := 16#08AB6B2C#;
25964 u00028 : constant Version_32 := 16#3FA37670#;
25965 u00029 : constant Version_32 := 16#476457A0#;
25966 u00030 : constant Version_32 := 16#731E1B6E#;
25967 u00031 : constant Version_32 := 16#23C2E789#;
25968 u00032 : constant Version_32 := 16#0F1BD6A1#;
25969 u00033 : constant Version_32 := 16#7C25DE96#;
25970 u00034 : constant Version_32 := 16#39ADFFA2#;
25971 u00035 : constant Version_32 := 16#571DE3E7#;
25972 u00036 : constant Version_32 := 16#5EB646AB#;
25973 u00037 : constant Version_32 := 16#4249379B#;
25974 u00038 : constant Version_32 := 16#0357E00A#;
25975 u00039 : constant Version_32 := 16#3784FB72#;
25976 u00040 : constant Version_32 := 16#2E723019#;
25977 u00041 : constant Version_32 := 16#623358EA#;
25978 u00042 : constant Version_32 := 16#107F9465#;
25979 u00043 : constant Version_32 := 16#6843F68A#;
25980 u00044 : constant Version_32 := 16#63305874#;
25981 u00045 : constant Version_32 := 16#31E56CE1#;
25982 u00046 : constant Version_32 := 16#02917970#;
25983 u00047 : constant Version_32 := 16#6CCBA70E#;
25984 u00048 : constant Version_32 := 16#41CD4204#;
25985 u00049 : constant Version_32 := 16#572E3F58#;
25986 u00050 : constant Version_32 := 16#20729FF5#;
25987 u00051 : constant Version_32 := 16#1D4F93E8#;
25988 u00052 : constant Version_32 := 16#30B2EC3D#;
25989 u00053 : constant Version_32 := 16#34054F96#;
25990 u00054 : constant Version_32 := 16#5A199860#;
25991 u00055 : constant Version_32 := 16#0E7F912B#;
25992 u00056 : constant Version_32 := 16#5760634A#;
25993 u00057 : constant Version_32 := 16#5D851835#;
25995 -- The following Export pragmas export the version numbers
25996 -- with symbolic names ending in B (for body) or S
25997 -- (for spec) so that they can be located in a link. The
25998 -- information provided here is sufficient to track down
25999 -- the exact versions of units used in a given build.
26001 pragma Export (C, u00001, "helloB");
26002 pragma Export (C, u00002, "system__standard_libraryB");
26003 pragma Export (C, u00003, "system__standard_libraryS");
26004 pragma Export (C, u00004, "adaS");
26005 pragma Export (C, u00005, "ada__text_ioB");
26006 pragma Export (C, u00006, "ada__text_ioS");
26007 pragma Export (C, u00007, "ada__exceptionsB");
26008 pragma Export (C, u00008, "ada__exceptionsS");
26009 pragma Export (C, u00009, "gnatS");
26010 pragma Export (C, u00010, "gnat__heap_sort_aB");
26011 pragma Export (C, u00011, "gnat__heap_sort_aS");
26012 pragma Export (C, u00012, "systemS");
26013 pragma Export (C, u00013, "system__exception_tableB");
26014 pragma Export (C, u00014, "system__exception_tableS");
26015 pragma Export (C, u00015, "gnat__htableB");
26016 pragma Export (C, u00016, "gnat__htableS");
26017 pragma Export (C, u00017, "system__exceptionsS");
26018 pragma Export (C, u00018, "system__machine_state_operationsB");
26019 pragma Export (C, u00019, "system__machine_state_operationsS");
26020 pragma Export (C, u00020, "system__machine_codeS");
26021 pragma Export (C, u00021, "system__storage_elementsB");
26022 pragma Export (C, u00022, "system__storage_elementsS");
26023 pragma Export (C, u00023, "system__secondary_stackB");
26024 pragma Export (C, u00024, "system__secondary_stackS");
26025 pragma Export (C, u00025, "system__parametersB");
26026 pragma Export (C, u00026, "system__parametersS");
26027 pragma Export (C, u00027, "system__soft_linksB");
26028 pragma Export (C, u00028, "system__soft_linksS");
26029 pragma Export (C, u00029, "system__stack_checkingB");
26030 pragma Export (C, u00030, "system__stack_checkingS");
26031 pragma Export (C, u00031, "system__tracebackB");
26032 pragma Export (C, u00032, "system__tracebackS");
26033 pragma Export (C, u00033, "ada__streamsS");
26034 pragma Export (C, u00034, "ada__tagsB");
26035 pragma Export (C, u00035, "ada__tagsS");
26036 pragma Export (C, u00036, "system__string_opsB");
26037 pragma Export (C, u00037, "system__string_opsS");
26038 pragma Export (C, u00038, "interfacesS");
26039 pragma Export (C, u00039, "interfaces__c_streamsB");
26040 pragma Export (C, u00040, "interfaces__c_streamsS");
26041 pragma Export (C, u00041, "system__file_ioB");
26042 pragma Export (C, u00042, "system__file_ioS");
26043 pragma Export (C, u00043, "ada__finalizationB");
26044 pragma Export (C, u00044, "ada__finalizationS");
26045 pragma Export (C, u00045, "system__finalization_rootB");
26046 pragma Export (C, u00046, "system__finalization_rootS");
26047 pragma Export (C, u00047, "system__finalization_implementationB");
26048 pragma Export (C, u00048, "system__finalization_implementationS");
26049 pragma Export (C, u00049, "system__string_ops_concat_3B");
26050 pragma Export (C, u00050, "system__string_ops_concat_3S");
26051 pragma Export (C, u00051, "system__stream_attributesB");
26052 pragma Export (C, u00052, "system__stream_attributesS");
26053 pragma Export (C, u00053, "ada__io_exceptionsS");
26054 pragma Export (C, u00054, "system__unsigned_typesS");
26055 pragma Export (C, u00055, "system__file_control_blockS");
26056 pragma Export (C, u00056, "ada__finalization__list_controllerB");
26057 pragma Export (C, u00057, "ada__finalization__list_controllerS");
26059 -- BEGIN ELABORATION ORDER
26062 -- gnat.heap_sort_a (spec)
26063 -- gnat.heap_sort_a (body)
26064 -- gnat.htable (spec)
26065 -- gnat.htable (body)
26066 -- interfaces (spec)
26068 -- system.machine_code (spec)
26069 -- system.parameters (spec)
26070 -- system.parameters (body)
26071 -- interfaces.c_streams (spec)
26072 -- interfaces.c_streams (body)
26073 -- system.standard_library (spec)
26074 -- ada.exceptions (spec)
26075 -- system.exception_table (spec)
26076 -- system.exception_table (body)
26077 -- ada.io_exceptions (spec)
26078 -- system.exceptions (spec)
26079 -- system.storage_elements (spec)
26080 -- system.storage_elements (body)
26081 -- system.machine_state_operations (spec)
26082 -- system.machine_state_operations (body)
26083 -- system.secondary_stack (spec)
26084 -- system.stack_checking (spec)
26085 -- system.soft_links (spec)
26086 -- system.soft_links (body)
26087 -- system.stack_checking (body)
26088 -- system.secondary_stack (body)
26089 -- system.standard_library (body)
26090 -- system.string_ops (spec)
26091 -- system.string_ops (body)
26094 -- ada.streams (spec)
26095 -- system.finalization_root (spec)
26096 -- system.finalization_root (body)
26097 -- system.string_ops_concat_3 (spec)
26098 -- system.string_ops_concat_3 (body)
26099 -- system.traceback (spec)
26100 -- system.traceback (body)
26101 -- ada.exceptions (body)
26102 -- system.unsigned_types (spec)
26103 -- system.stream_attributes (spec)
26104 -- system.stream_attributes (body)
26105 -- system.finalization_implementation (spec)
26106 -- system.finalization_implementation (body)
26107 -- ada.finalization (spec)
26108 -- ada.finalization (body)
26109 -- ada.finalization.list_controller (spec)
26110 -- ada.finalization.list_controller (body)
26111 -- system.file_control_block (spec)
26112 -- system.file_io (spec)
26113 -- system.file_io (body)
26114 -- ada.text_io (spec)
26115 -- ada.text_io (body)
26117 -- END ELABORATION ORDER
26121 -- The following source file name pragmas allow the generated file
26122 -- names to be unique for different main programs. They are needed
26123 -- since the package name will always be Ada_Main.
26125 pragma Source_File_Name (ada_main, Spec_File_Name => "b~hello.ads");
26126 pragma Source_File_Name (ada_main, Body_File_Name => "b~hello.adb");
26128 -- Generated package body for Ada_Main starts here
26130 package body ada_main is
26132 -- The actual finalization is performed by calling the
26133 -- library routine in System.Standard_Library.Adafinal
26135 procedure Do_Finalize;
26136 pragma Import (C, Do_Finalize, "system__standard_library__adafinal");
26143 procedure adainit is
26145 -- These booleans are set to True once the associated unit has
26146 -- been elaborated. It is also used to avoid elaborating the
26147 -- same unit twice.
26150 pragma Import (Ada, E040, "interfaces__c_streams_E");
26153 pragma Import (Ada, E008, "ada__exceptions_E");
26156 pragma Import (Ada, E014, "system__exception_table_E");
26159 pragma Import (Ada, E053, "ada__io_exceptions_E");
26162 pragma Import (Ada, E017, "system__exceptions_E");
26165 pragma Import (Ada, E024, "system__secondary_stack_E");
26168 pragma Import (Ada, E030, "system__stack_checking_E");
26171 pragma Import (Ada, E028, "system__soft_links_E");
26174 pragma Import (Ada, E035, "ada__tags_E");
26177 pragma Import (Ada, E033, "ada__streams_E");
26180 pragma Import (Ada, E046, "system__finalization_root_E");
26183 pragma Import (Ada, E048, "system__finalization_implementation_E");
26186 pragma Import (Ada, E044, "ada__finalization_E");
26189 pragma Import (Ada, E057, "ada__finalization__list_controller_E");
26192 pragma Import (Ada, E055, "system__file_control_block_E");
26195 pragma Import (Ada, E042, "system__file_io_E");
26198 pragma Import (Ada, E006, "ada__text_io_E");
26200 -- Set_Globals is a library routine that stores away the
26201 -- value of the indicated set of global values in global
26202 -- variables within the library.
26204 procedure Set_Globals
26205 (Main_Priority : Integer;
26206 Time_Slice_Value : Integer;
26207 WC_Encoding : Character;
26208 Locking_Policy : Character;
26209 Queuing_Policy : Character;
26210 Task_Dispatching_Policy : Character;
26211 Adafinal : System.Address;
26212 Unreserve_All_Interrupts : Integer;
26213 Exception_Tracebacks : Integer);
26214 @findex __gnat_set_globals
26215 pragma Import (C, Set_Globals, "__gnat_set_globals");
26217 -- SDP_Table_Build is a library routine used to build the
26218 -- exception tables. See unit Ada.Exceptions in files
26219 -- a-except.ads/adb for full details of how zero cost
26220 -- exception handling works. This procedure, the call to
26221 -- it, and the two following tables are all omitted if the
26222 -- build is in longjmp/setjmp exception mode.
26224 @findex SDP_Table_Build
26225 @findex Zero Cost Exceptions
26226 procedure SDP_Table_Build
26227 (SDP_Addresses : System.Address;
26228 SDP_Count : Natural;
26229 Elab_Addresses : System.Address;
26230 Elab_Addr_Count : Natural);
26231 pragma Import (C, SDP_Table_Build, "__gnat_SDP_Table_Build");
26233 -- Table of Unit_Exception_Table addresses. Used for zero
26234 -- cost exception handling to build the top level table.
26236 ST : aliased constant array (1 .. 23) of System.Address := (
26238 Ada.Text_Io'UET_Address,
26239 Ada.Exceptions'UET_Address,
26240 Gnat.Heap_Sort_A'UET_Address,
26241 System.Exception_Table'UET_Address,
26242 System.Machine_State_Operations'UET_Address,
26243 System.Secondary_Stack'UET_Address,
26244 System.Parameters'UET_Address,
26245 System.Soft_Links'UET_Address,
26246 System.Stack_Checking'UET_Address,
26247 System.Traceback'UET_Address,
26248 Ada.Streams'UET_Address,
26249 Ada.Tags'UET_Address,
26250 System.String_Ops'UET_Address,
26251 Interfaces.C_Streams'UET_Address,
26252 System.File_Io'UET_Address,
26253 Ada.Finalization'UET_Address,
26254 System.Finalization_Root'UET_Address,
26255 System.Finalization_Implementation'UET_Address,
26256 System.String_Ops_Concat_3'UET_Address,
26257 System.Stream_Attributes'UET_Address,
26258 System.File_Control_Block'UET_Address,
26259 Ada.Finalization.List_Controller'UET_Address);
26261 -- Table of addresses of elaboration routines. Used for
26262 -- zero cost exception handling to make sure these
26263 -- addresses are included in the top level procedure
26266 EA : aliased constant array (1 .. 23) of System.Address := (
26267 adainit'Code_Address,
26268 Do_Finalize'Code_Address,
26269 Ada.Exceptions'Elab_Spec'Address,
26270 System.Exceptions'Elab_Spec'Address,
26271 Interfaces.C_Streams'Elab_Spec'Address,
26272 System.Exception_Table'Elab_Body'Address,
26273 Ada.Io_Exceptions'Elab_Spec'Address,
26274 System.Stack_Checking'Elab_Spec'Address,
26275 System.Soft_Links'Elab_Body'Address,
26276 System.Secondary_Stack'Elab_Body'Address,
26277 Ada.Tags'Elab_Spec'Address,
26278 Ada.Tags'Elab_Body'Address,
26279 Ada.Streams'Elab_Spec'Address,
26280 System.Finalization_Root'Elab_Spec'Address,
26281 Ada.Exceptions'Elab_Body'Address,
26282 System.Finalization_Implementation'Elab_Spec'Address,
26283 System.Finalization_Implementation'Elab_Body'Address,
26284 Ada.Finalization'Elab_Spec'Address,
26285 Ada.Finalization.List_Controller'Elab_Spec'Address,
26286 System.File_Control_Block'Elab_Spec'Address,
26287 System.File_Io'Elab_Body'Address,
26288 Ada.Text_Io'Elab_Spec'Address,
26289 Ada.Text_Io'Elab_Body'Address);
26291 -- Start of processing for adainit
26295 -- Call SDP_Table_Build to build the top level procedure
26296 -- table for zero cost exception handling (omitted in
26297 -- longjmp/setjmp mode).
26299 SDP_Table_Build (ST'Address, 23, EA'Address, 23);
26301 -- Call Set_Globals to record various information for
26302 -- this partition. The values are derived by the binder
26303 -- from information stored in the ali files by the compiler.
26305 @findex __gnat_set_globals
26307 (Main_Priority => -1,
26308 -- Priority of main program, -1 if no pragma Priority used
26310 Time_Slice_Value => -1,
26311 -- Time slice from Time_Slice pragma, -1 if none used
26313 WC_Encoding => 'b',
26314 -- Wide_Character encoding used, default is brackets
26316 Locking_Policy => ' ',
26317 -- Locking_Policy used, default of space means not
26318 -- specified, otherwise it is the first character of
26319 -- the policy name.
26321 Queuing_Policy => ' ',
26322 -- Queuing_Policy used, default of space means not
26323 -- specified, otherwise it is the first character of
26324 -- the policy name.
26326 Task_Dispatching_Policy => ' ',
26327 -- Task_Dispatching_Policy used, default of space means
26328 -- not specified, otherwise first character of the
26331 Adafinal => System.Null_Address,
26332 -- Address of Adafinal routine, not used anymore
26334 Unreserve_All_Interrupts => 0,
26335 -- Set true if pragma Unreserve_All_Interrupts was used
26337 Exception_Tracebacks => 0);
26338 -- Indicates if exception tracebacks are enabled
26340 Elab_Final_Code := 1;
26342 -- Now we have the elaboration calls for all units in the partition.
26343 -- The Elab_Spec and Elab_Body attributes generate references to the
26344 -- implicit elaboration procedures generated by the compiler for
26345 -- each unit that requires elaboration.
26348 Interfaces.C_Streams'Elab_Spec;
26352 Ada.Exceptions'Elab_Spec;
26355 System.Exception_Table'Elab_Body;
26359 Ada.Io_Exceptions'Elab_Spec;
26363 System.Exceptions'Elab_Spec;
26367 System.Stack_Checking'Elab_Spec;
26370 System.Soft_Links'Elab_Body;
26375 System.Secondary_Stack'Elab_Body;
26379 Ada.Tags'Elab_Spec;
26382 Ada.Tags'Elab_Body;
26386 Ada.Streams'Elab_Spec;
26390 System.Finalization_Root'Elab_Spec;
26394 Ada.Exceptions'Elab_Body;
26398 System.Finalization_Implementation'Elab_Spec;
26401 System.Finalization_Implementation'Elab_Body;
26405 Ada.Finalization'Elab_Spec;
26409 Ada.Finalization.List_Controller'Elab_Spec;
26413 System.File_Control_Block'Elab_Spec;
26417 System.File_Io'Elab_Body;
26421 Ada.Text_Io'Elab_Spec;
26424 Ada.Text_Io'Elab_Body;
26428 Elab_Final_Code := 0;
26436 procedure adafinal is
26445 -- main is actually a function, as in the ANSI C standard,
26446 -- defined to return the exit status. The three parameters
26447 -- are the argument count, argument values and environment
26450 @findex Main Program
26453 argv : System.Address;
26454 envp : System.Address)
26457 -- The initialize routine performs low level system
26458 -- initialization using a standard library routine which
26459 -- sets up signal handling and performs any other
26460 -- required setup. The routine can be found in file
26463 @findex __gnat_initialize
26464 procedure initialize;
26465 pragma Import (C, initialize, "__gnat_initialize");
26467 -- The finalize routine performs low level system
26468 -- finalization using a standard library routine. The
26469 -- routine is found in file a-final.c and in the standard
26470 -- distribution is a dummy routine that does nothing, so
26471 -- really this is a hook for special user finalization.
26473 @findex __gnat_finalize
26474 procedure finalize;
26475 pragma Import (C, finalize, "__gnat_finalize");
26477 -- We get to the main program of the partition by using
26478 -- pragma Import because if we try to with the unit and
26479 -- call it Ada style, then not only do we waste time
26480 -- recompiling it, but also, we don't really know the right
26481 -- switches (e.g.@: identifier character set) to be used
26484 procedure Ada_Main_Program;
26485 pragma Import (Ada, Ada_Main_Program, "_ada_hello");
26487 -- Start of processing for main
26490 -- Save global variables
26496 -- Call low level system initialization
26500 -- Call our generated Ada initialization routine
26504 -- This is the point at which we want the debugger to get
26509 -- Now we call the main program of the partition
26513 -- Perform Ada finalization
26517 -- Perform low level system finalization
26521 -- Return the proper exit status
26522 return (gnat_exit_status);
26525 -- This section is entirely comments, so it has no effect on the
26526 -- compilation of the Ada_Main package. It provides the list of
26527 -- object files and linker options, as well as some standard
26528 -- libraries needed for the link. The gnatlink utility parses
26529 -- this b~hello.adb file to read these comment lines to generate
26530 -- the appropriate command line arguments for the call to the
26531 -- system linker. The BEGIN/END lines are used for sentinels for
26532 -- this parsing operation.
26534 -- The exact file names will of course depend on the environment,
26535 -- host/target and location of files on the host system.
26537 @findex Object file list
26538 -- BEGIN Object file/option list
26541 -- -L/usr/local/gnat/lib/gcc-lib/i686-pc-linux-gnu/2.8.1/adalib/
26542 -- /usr/local/gnat/lib/gcc-lib/i686-pc-linux-gnu/2.8.1/adalib/libgnat.a
26543 -- END Object file/option list
26549 The Ada code in the above example is exactly what is generated by the
26550 binder. We have added comments to more clearly indicate the function
26551 of each part of the generated @code{Ada_Main} package.
26553 The code is standard Ada in all respects, and can be processed by any
26554 tools that handle Ada. In particular, it is possible to use the debugger
26555 in Ada mode to debug the generated @code{Ada_Main} package. For example,
26556 suppose that for reasons that you do not understand, your program is crashing
26557 during elaboration of the body of @code{Ada.Text_IO}. To locate this bug,
26558 you can place a breakpoint on the call:
26560 @smallexample @c ada
26561 Ada.Text_Io'Elab_Body;
26565 and trace the elaboration routine for this package to find out where
26566 the problem might be (more usually of course you would be debugging
26567 elaboration code in your own application).
26569 @node Elaboration Order Handling in GNAT
26570 @appendix Elaboration Order Handling in GNAT
26571 @cindex Order of elaboration
26572 @cindex Elaboration control
26575 * Elaboration Code::
26576 * Checking the Elaboration Order::
26577 * Controlling the Elaboration Order::
26578 * Controlling Elaboration in GNAT - Internal Calls::
26579 * Controlling Elaboration in GNAT - External Calls::
26580 * Default Behavior in GNAT - Ensuring Safety::
26581 * Treatment of Pragma Elaborate::
26582 * Elaboration Issues for Library Tasks::
26583 * Mixing Elaboration Models::
26584 * What to Do If the Default Elaboration Behavior Fails::
26585 * Elaboration for Access-to-Subprogram Values::
26586 * Summary of Procedures for Elaboration Control::
26587 * Other Elaboration Order Considerations::
26591 This chapter describes the handling of elaboration code in Ada and
26592 in GNAT, and discusses how the order of elaboration of program units can
26593 be controlled in GNAT, either automatically or with explicit programming
26596 @node Elaboration Code
26597 @section Elaboration Code
26600 Ada provides rather general mechanisms for executing code at elaboration
26601 time, that is to say before the main program starts executing. Such code arises
26605 @item Initializers for variables.
26606 Variables declared at the library level, in package specs or bodies, can
26607 require initialization that is performed at elaboration time, as in:
26608 @smallexample @c ada
26610 Sqrt_Half : Float := Sqrt (0.5);
26614 @item Package initialization code
26615 Code in a @code{BEGIN-END} section at the outer level of a package body is
26616 executed as part of the package body elaboration code.
26618 @item Library level task allocators
26619 Tasks that are declared using task allocators at the library level
26620 start executing immediately and hence can execute at elaboration time.
26624 Subprogram calls are possible in any of these contexts, which means that
26625 any arbitrary part of the program may be executed as part of the elaboration
26626 code. It is even possible to write a program which does all its work at
26627 elaboration time, with a null main program, although stylistically this
26628 would usually be considered an inappropriate way to structure
26631 An important concern arises in the context of elaboration code:
26632 we have to be sure that it is executed in an appropriate order. What we
26633 have is a series of elaboration code sections, potentially one section
26634 for each unit in the program. It is important that these execute
26635 in the correct order. Correctness here means that, taking the above
26636 example of the declaration of @code{Sqrt_Half},
26637 if some other piece of
26638 elaboration code references @code{Sqrt_Half},
26639 then it must run after the
26640 section of elaboration code that contains the declaration of
26643 There would never be any order of elaboration problem if we made a rule
26644 that whenever you @code{with} a unit, you must elaborate both the spec and body
26645 of that unit before elaborating the unit doing the @code{with}'ing:
26647 @smallexample @c ada
26651 package Unit_2 is @dots{}
26657 would require that both the body and spec of @code{Unit_1} be elaborated
26658 before the spec of @code{Unit_2}. However, a rule like that would be far too
26659 restrictive. In particular, it would make it impossible to have routines
26660 in separate packages that were mutually recursive.
26662 You might think that a clever enough compiler could look at the actual
26663 elaboration code and determine an appropriate correct order of elaboration,
26664 but in the general case, this is not possible. Consider the following
26667 In the body of @code{Unit_1}, we have a procedure @code{Func_1}
26669 the variable @code{Sqrt_1}, which is declared in the elaboration code
26670 of the body of @code{Unit_1}:
26672 @smallexample @c ada
26674 Sqrt_1 : Float := Sqrt (0.1);
26679 The elaboration code of the body of @code{Unit_1} also contains:
26681 @smallexample @c ada
26684 if expression_1 = 1 then
26685 Q := Unit_2.Func_2;
26692 @code{Unit_2} is exactly parallel,
26693 it has a procedure @code{Func_2} that references
26694 the variable @code{Sqrt_2}, which is declared in the elaboration code of
26695 the body @code{Unit_2}:
26697 @smallexample @c ada
26699 Sqrt_2 : Float := Sqrt (0.1);
26704 The elaboration code of the body of @code{Unit_2} also contains:
26706 @smallexample @c ada
26709 if expression_2 = 2 then
26710 Q := Unit_1.Func_1;
26717 Now the question is, which of the following orders of elaboration is
26742 If you carefully analyze the flow here, you will see that you cannot tell
26743 at compile time the answer to this question.
26744 If @code{expression_1} is not equal to 1,
26745 and @code{expression_2} is not equal to 2,
26746 then either order is acceptable, because neither of the function calls is
26747 executed. If both tests evaluate to true, then neither order is acceptable
26748 and in fact there is no correct order.
26750 If one of the two expressions is true, and the other is false, then one
26751 of the above orders is correct, and the other is incorrect. For example,
26752 if @code{expression_1} /= 1 and @code{expression_2} = 2,
26753 then the call to @code{Func_1}
26754 will occur, but not the call to @code{Func_2.}
26755 This means that it is essential
26756 to elaborate the body of @code{Unit_1} before
26757 the body of @code{Unit_2}, so the first
26758 order of elaboration is correct and the second is wrong.
26760 By making @code{expression_1} and @code{expression_2}
26761 depend on input data, or perhaps
26762 the time of day, we can make it impossible for the compiler or binder
26763 to figure out which of these expressions will be true, and hence it
26764 is impossible to guarantee a safe order of elaboration at run time.
26766 @node Checking the Elaboration Order
26767 @section Checking the Elaboration Order
26770 In some languages that involve the same kind of elaboration problems,
26771 e.g.@: Java and C++, the programmer is expected to worry about these
26772 ordering problems himself, and it is common to
26773 write a program in which an incorrect elaboration order gives
26774 surprising results, because it references variables before they
26776 Ada is designed to be a safe language, and a programmer-beware approach is
26777 clearly not sufficient. Consequently, the language provides three lines
26781 @item Standard rules
26782 Some standard rules restrict the possible choice of elaboration
26783 order. In particular, if you @code{with} a unit, then its spec is always
26784 elaborated before the unit doing the @code{with}. Similarly, a parent
26785 spec is always elaborated before the child spec, and finally
26786 a spec is always elaborated before its corresponding body.
26788 @item Dynamic elaboration checks
26789 @cindex Elaboration checks
26790 @cindex Checks, elaboration
26791 Dynamic checks are made at run time, so that if some entity is accessed
26792 before it is elaborated (typically by means of a subprogram call)
26793 then the exception (@code{Program_Error}) is raised.
26795 @item Elaboration control
26796 Facilities are provided for the programmer to specify the desired order
26800 Let's look at these facilities in more detail. First, the rules for
26801 dynamic checking. One possible rule would be simply to say that the
26802 exception is raised if you access a variable which has not yet been
26803 elaborated. The trouble with this approach is that it could require
26804 expensive checks on every variable reference. Instead Ada has two
26805 rules which are a little more restrictive, but easier to check, and
26809 @item Restrictions on calls
26810 A subprogram can only be called at elaboration time if its body
26811 has been elaborated. The rules for elaboration given above guarantee
26812 that the spec of the subprogram has been elaborated before the
26813 call, but not the body. If this rule is violated, then the
26814 exception @code{Program_Error} is raised.
26816 @item Restrictions on instantiations
26817 A generic unit can only be instantiated if the body of the generic
26818 unit has been elaborated. Again, the rules for elaboration given above
26819 guarantee that the spec of the generic unit has been elaborated
26820 before the instantiation, but not the body. If this rule is
26821 violated, then the exception @code{Program_Error} is raised.
26825 The idea is that if the body has been elaborated, then any variables
26826 it references must have been elaborated; by checking for the body being
26827 elaborated we guarantee that none of its references causes any
26828 trouble. As we noted above, this is a little too restrictive, because a
26829 subprogram that has no non-local references in its body may in fact be safe
26830 to call. However, it really would be unsafe to rely on this, because
26831 it would mean that the caller was aware of details of the implementation
26832 in the body. This goes against the basic tenets of Ada.
26834 A plausible implementation can be described as follows.
26835 A Boolean variable is associated with each subprogram
26836 and each generic unit. This variable is initialized to False, and is set to
26837 True at the point body is elaborated. Every call or instantiation checks the
26838 variable, and raises @code{Program_Error} if the variable is False.
26840 Note that one might think that it would be good enough to have one Boolean
26841 variable for each package, but that would not deal with cases of trying
26842 to call a body in the same package as the call
26843 that has not been elaborated yet.
26844 Of course a compiler may be able to do enough analysis to optimize away
26845 some of the Boolean variables as unnecessary, and @code{GNAT} indeed
26846 does such optimizations, but still the easiest conceptual model is to
26847 think of there being one variable per subprogram.
26849 @node Controlling the Elaboration Order
26850 @section Controlling the Elaboration Order
26853 In the previous section we discussed the rules in Ada which ensure
26854 that @code{Program_Error} is raised if an incorrect elaboration order is
26855 chosen. This prevents erroneous executions, but we need mechanisms to
26856 specify a correct execution and avoid the exception altogether.
26857 To achieve this, Ada provides a number of features for controlling
26858 the order of elaboration. We discuss these features in this section.
26860 First, there are several ways of indicating to the compiler that a given
26861 unit has no elaboration problems:
26864 @item packages that do not require a body
26865 A library package that does not require a body does not permit
26866 a body (this rule was introduced in Ada 95).
26867 Thus if we have a such a package, as in:
26869 @smallexample @c ada
26872 package Definitions is
26874 type m is new integer;
26876 type a is array (1 .. 10) of m;
26877 type b is array (1 .. 20) of m;
26885 A package that @code{with}'s @code{Definitions} may safely instantiate
26886 @code{Definitions.Subp} because the compiler can determine that there
26887 definitely is no package body to worry about in this case
26890 @cindex pragma Pure
26892 Places sufficient restrictions on a unit to guarantee that
26893 no call to any subprogram in the unit can result in an
26894 elaboration problem. This means that the compiler does not need
26895 to worry about the point of elaboration of such units, and in
26896 particular, does not need to check any calls to any subprograms
26899 @item pragma Preelaborate
26900 @findex Preelaborate
26901 @cindex pragma Preelaborate
26902 This pragma places slightly less stringent restrictions on a unit than
26904 but these restrictions are still sufficient to ensure that there
26905 are no elaboration problems with any calls to the unit.
26907 @item pragma Elaborate_Body
26908 @findex Elaborate_Body
26909 @cindex pragma Elaborate_Body
26910 This pragma requires that the body of a unit be elaborated immediately
26911 after its spec. Suppose a unit @code{A} has such a pragma,
26912 and unit @code{B} does
26913 a @code{with} of unit @code{A}. Recall that the standard rules require
26914 the spec of unit @code{A}
26915 to be elaborated before the @code{with}'ing unit; given the pragma in
26916 @code{A}, we also know that the body of @code{A}
26917 will be elaborated before @code{B}, so
26918 that calls to @code{A} are safe and do not need a check.
26923 unlike pragma @code{Pure} and pragma @code{Preelaborate},
26925 @code{Elaborate_Body} does not guarantee that the program is
26926 free of elaboration problems, because it may not be possible
26927 to satisfy the requested elaboration order.
26928 Let's go back to the example with @code{Unit_1} and @code{Unit_2}.
26930 marks @code{Unit_1} as @code{Elaborate_Body},
26931 and not @code{Unit_2,} then the order of
26932 elaboration will be:
26944 Now that means that the call to @code{Func_1} in @code{Unit_2}
26945 need not be checked,
26946 it must be safe. But the call to @code{Func_2} in
26947 @code{Unit_1} may still fail if
26948 @code{Expression_1} is equal to 1,
26949 and the programmer must still take
26950 responsibility for this not being the case.
26952 If all units carry a pragma @code{Elaborate_Body}, then all problems are
26953 eliminated, except for calls entirely within a body, which are
26954 in any case fully under programmer control. However, using the pragma
26955 everywhere is not always possible.
26956 In particular, for our @code{Unit_1}/@code{Unit_2} example, if
26957 we marked both of them as having pragma @code{Elaborate_Body}, then
26958 clearly there would be no possible elaboration order.
26960 The above pragmas allow a server to guarantee safe use by clients, and
26961 clearly this is the preferable approach. Consequently a good rule
26962 is to mark units as @code{Pure} or @code{Preelaborate} if possible,
26963 and if this is not possible,
26964 mark them as @code{Elaborate_Body} if possible.
26965 As we have seen, there are situations where neither of these
26966 three pragmas can be used.
26967 So we also provide methods for clients to control the
26968 order of elaboration of the servers on which they depend:
26971 @item pragma Elaborate (unit)
26973 @cindex pragma Elaborate
26974 This pragma is placed in the context clause, after a @code{with} clause,
26975 and it requires that the body of the named unit be elaborated before
26976 the unit in which the pragma occurs. The idea is to use this pragma
26977 if the current unit calls at elaboration time, directly or indirectly,
26978 some subprogram in the named unit.
26980 @item pragma Elaborate_All (unit)
26981 @findex Elaborate_All
26982 @cindex pragma Elaborate_All
26983 This is a stronger version of the Elaborate pragma. Consider the
26987 Unit A @code{with}'s unit B and calls B.Func in elab code
26988 Unit B @code{with}'s unit C, and B.Func calls C.Func
26992 Now if we put a pragma @code{Elaborate (B)}
26993 in unit @code{A}, this ensures that the
26994 body of @code{B} is elaborated before the call, but not the
26995 body of @code{C}, so
26996 the call to @code{C.Func} could still cause @code{Program_Error} to
26999 The effect of a pragma @code{Elaborate_All} is stronger, it requires
27000 not only that the body of the named unit be elaborated before the
27001 unit doing the @code{with}, but also the bodies of all units that the
27002 named unit uses, following @code{with} links transitively. For example,
27003 if we put a pragma @code{Elaborate_All (B)} in unit @code{A},
27005 not only that the body of @code{B} be elaborated before @code{A},
27007 body of @code{C}, because @code{B} @code{with}'s @code{C}.
27011 We are now in a position to give a usage rule in Ada for avoiding
27012 elaboration problems, at least if dynamic dispatching and access to
27013 subprogram values are not used. We will handle these cases separately
27016 The rule is simple. If a unit has elaboration code that can directly or
27017 indirectly make a call to a subprogram in a @code{with}'ed unit, or instantiate
27018 a generic package in a @code{with}'ed unit,
27019 then if the @code{with}'ed unit does not have
27020 pragma @code{Pure} or @code{Preelaborate}, then the client should have
27021 a pragma @code{Elaborate_All}
27022 for the @code{with}'ed unit. By following this rule a client is
27023 assured that calls can be made without risk of an exception.
27025 For generic subprogram instantiations, the rule can be relaxed to
27026 require only a pragma @code{Elaborate} since elaborating the body
27027 of a subprogram cannot cause any transitive elaboration (we are
27028 not calling the subprogram in this case, just elaborating its
27031 If this rule is not followed, then a program may be in one of four
27035 @item No order exists
27036 No order of elaboration exists which follows the rules, taking into
27037 account any @code{Elaborate}, @code{Elaborate_All},
27038 or @code{Elaborate_Body} pragmas. In
27039 this case, an Ada compiler must diagnose the situation at bind
27040 time, and refuse to build an executable program.
27042 @item One or more orders exist, all incorrect
27043 One or more acceptable elaboration orders exist, and all of them
27044 generate an elaboration order problem. In this case, the binder
27045 can build an executable program, but @code{Program_Error} will be raised
27046 when the program is run.
27048 @item Several orders exist, some right, some incorrect
27049 One or more acceptable elaboration orders exists, and some of them
27050 work, and some do not. The programmer has not controlled
27051 the order of elaboration, so the binder may or may not pick one of
27052 the correct orders, and the program may or may not raise an
27053 exception when it is run. This is the worst case, because it means
27054 that the program may fail when moved to another compiler, or even
27055 another version of the same compiler.
27057 @item One or more orders exists, all correct
27058 One ore more acceptable elaboration orders exist, and all of them
27059 work. In this case the program runs successfully. This state of
27060 affairs can be guaranteed by following the rule we gave above, but
27061 may be true even if the rule is not followed.
27065 Note that one additional advantage of following our rules on the use
27066 of @code{Elaborate} and @code{Elaborate_All}
27067 is that the program continues to stay in the ideal (all orders OK) state
27068 even if maintenance
27069 changes some bodies of some units. Conversely, if a program that does
27070 not follow this rule happens to be safe at some point, this state of affairs
27071 may deteriorate silently as a result of maintenance changes.
27073 You may have noticed that the above discussion did not mention
27074 the use of @code{Elaborate_Body}. This was a deliberate omission. If you
27075 @code{with} an @code{Elaborate_Body} unit, it still may be the case that
27076 code in the body makes calls to some other unit, so it is still necessary
27077 to use @code{Elaborate_All} on such units.
27079 @node Controlling Elaboration in GNAT - Internal Calls
27080 @section Controlling Elaboration in GNAT - Internal Calls
27083 In the case of internal calls, i.e., calls within a single package, the
27084 programmer has full control over the order of elaboration, and it is up
27085 to the programmer to elaborate declarations in an appropriate order. For
27088 @smallexample @c ada
27091 function One return Float;
27095 function One return Float is
27104 will obviously raise @code{Program_Error} at run time, because function
27105 One will be called before its body is elaborated. In this case GNAT will
27106 generate a warning that the call will raise @code{Program_Error}:
27112 2. function One return Float;
27114 4. Q : Float := One;
27116 >>> warning: cannot call "One" before body is elaborated
27117 >>> warning: Program_Error will be raised at run time
27120 6. function One return Float is
27133 Note that in this particular case, it is likely that the call is safe, because
27134 the function @code{One} does not access any global variables.
27135 Nevertheless in Ada, we do not want the validity of the check to depend on
27136 the contents of the body (think about the separate compilation case), so this
27137 is still wrong, as we discussed in the previous sections.
27139 The error is easily corrected by rearranging the declarations so that the
27140 body of @code{One} appears before the declaration containing the call
27141 (note that in Ada 95 and Ada 2005,
27142 declarations can appear in any order, so there is no restriction that
27143 would prevent this reordering, and if we write:
27145 @smallexample @c ada
27148 function One return Float;
27150 function One return Float is
27161 then all is well, no warning is generated, and no
27162 @code{Program_Error} exception
27164 Things are more complicated when a chain of subprograms is executed:
27166 @smallexample @c ada
27169 function A return Integer;
27170 function B return Integer;
27171 function C return Integer;
27173 function B return Integer is begin return A; end;
27174 function C return Integer is begin return B; end;
27178 function A return Integer is begin return 1; end;
27184 Now the call to @code{C}
27185 at elaboration time in the declaration of @code{X} is correct, because
27186 the body of @code{C} is already elaborated,
27187 and the call to @code{B} within the body of
27188 @code{C} is correct, but the call
27189 to @code{A} within the body of @code{B} is incorrect, because the body
27190 of @code{A} has not been elaborated, so @code{Program_Error}
27191 will be raised on the call to @code{A}.
27192 In this case GNAT will generate a
27193 warning that @code{Program_Error} may be
27194 raised at the point of the call. Let's look at the warning:
27200 2. function A return Integer;
27201 3. function B return Integer;
27202 4. function C return Integer;
27204 6. function B return Integer is begin return A; end;
27206 >>> warning: call to "A" before body is elaborated may
27207 raise Program_Error
27208 >>> warning: "B" called at line 7
27209 >>> warning: "C" called at line 9
27211 7. function C return Integer is begin return B; end;
27213 9. X : Integer := C;
27215 11. function A return Integer is begin return 1; end;
27225 Note that the message here says ``may raise'', instead of the direct case,
27226 where the message says ``will be raised''. That's because whether
27228 actually called depends in general on run-time flow of control.
27229 For example, if the body of @code{B} said
27231 @smallexample @c ada
27234 function B return Integer is
27236 if some-condition-depending-on-input-data then
27247 then we could not know until run time whether the incorrect call to A would
27248 actually occur, so @code{Program_Error} might
27249 or might not be raised. It is possible for a compiler to
27250 do a better job of analyzing bodies, to
27251 determine whether or not @code{Program_Error}
27252 might be raised, but it certainly
27253 couldn't do a perfect job (that would require solving the halting problem
27254 and is provably impossible), and because this is a warning anyway, it does
27255 not seem worth the effort to do the analysis. Cases in which it
27256 would be relevant are rare.
27258 In practice, warnings of either of the forms given
27259 above will usually correspond to
27260 real errors, and should be examined carefully and eliminated.
27261 In the rare case where a warning is bogus, it can be suppressed by any of
27262 the following methods:
27266 Compile with the @option{-gnatws} switch set
27269 Suppress @code{Elaboration_Check} for the called subprogram
27272 Use pragma @code{Warnings_Off} to turn warnings off for the call
27276 For the internal elaboration check case,
27277 GNAT by default generates the
27278 necessary run-time checks to ensure
27279 that @code{Program_Error} is raised if any
27280 call fails an elaboration check. Of course this can only happen if a
27281 warning has been issued as described above. The use of pragma
27282 @code{Suppress (Elaboration_Check)} may (but is not guaranteed to) suppress
27283 some of these checks, meaning that it may be possible (but is not
27284 guaranteed) for a program to be able to call a subprogram whose body
27285 is not yet elaborated, without raising a @code{Program_Error} exception.
27287 @node Controlling Elaboration in GNAT - External Calls
27288 @section Controlling Elaboration in GNAT - External Calls
27291 The previous section discussed the case in which the execution of a
27292 particular thread of elaboration code occurred entirely within a
27293 single unit. This is the easy case to handle, because a programmer
27294 has direct and total control over the order of elaboration, and
27295 furthermore, checks need only be generated in cases which are rare
27296 and which the compiler can easily detect.
27297 The situation is more complex when separate compilation is taken into account.
27298 Consider the following:
27300 @smallexample @c ada
27304 function Sqrt (Arg : Float) return Float;
27307 package body Math is
27308 function Sqrt (Arg : Float) return Float is
27317 X : Float := Math.Sqrt (0.5);
27330 where @code{Main} is the main program. When this program is executed, the
27331 elaboration code must first be executed, and one of the jobs of the
27332 binder is to determine the order in which the units of a program are
27333 to be elaborated. In this case we have four units: the spec and body
27335 the spec of @code{Stuff} and the body of @code{Main}).
27336 In what order should the four separate sections of elaboration code
27339 There are some restrictions in the order of elaboration that the binder
27340 can choose. In particular, if unit U has a @code{with}
27341 for a package @code{X}, then you
27342 are assured that the spec of @code{X}
27343 is elaborated before U , but you are
27344 not assured that the body of @code{X}
27345 is elaborated before U.
27346 This means that in the above case, the binder is allowed to choose the
27357 but that's not good, because now the call to @code{Math.Sqrt}
27358 that happens during
27359 the elaboration of the @code{Stuff}
27360 spec happens before the body of @code{Math.Sqrt} is
27361 elaborated, and hence causes @code{Program_Error} exception to be raised.
27362 At first glance, one might say that the binder is misbehaving, because
27363 obviously you want to elaborate the body of something you @code{with}
27365 that is not a general rule that can be followed in all cases. Consider
27367 @smallexample @c ada
27370 package X is @dots{}
27372 package Y is @dots{}
27375 package body Y is @dots{}
27378 package body X is @dots{}
27384 This is a common arrangement, and, apart from the order of elaboration
27385 problems that might arise in connection with elaboration code, this works fine.
27386 A rule that says that you must first elaborate the body of anything you
27387 @code{with} cannot work in this case:
27388 the body of @code{X} @code{with}'s @code{Y},
27389 which means you would have to
27390 elaborate the body of @code{Y} first, but that @code{with}'s @code{X},
27392 you have to elaborate the body of @code{X} first, but @dots{} and we have a
27393 loop that cannot be broken.
27395 It is true that the binder can in many cases guess an order of elaboration
27396 that is unlikely to cause a @code{Program_Error}
27397 exception to be raised, and it tries to do so (in the
27398 above example of @code{Math/Stuff/Spec}, the GNAT binder will
27400 elaborate the body of @code{Math} right after its spec, so all will be well).
27402 However, a program that blindly relies on the binder to be helpful can
27403 get into trouble, as we discussed in the previous sections, so
27405 provides a number of facilities for assisting the programmer in
27406 developing programs that are robust with respect to elaboration order.
27408 @node Default Behavior in GNAT - Ensuring Safety
27409 @section Default Behavior in GNAT - Ensuring Safety
27412 The default behavior in GNAT ensures elaboration safety. In its
27413 default mode GNAT implements the
27414 rule we previously described as the right approach. Let's restate it:
27418 @emph{If a unit has elaboration code that can directly or indirectly make a
27419 call to a subprogram in a @code{with}'ed unit, or instantiate a generic
27420 package in a @code{with}'ed unit, then if the @code{with}'ed unit
27421 does not have pragma @code{Pure} or
27422 @code{Preelaborate}, then the client should have an
27423 @code{Elaborate_All} pragma for the @code{with}'ed unit.}
27425 @emph{In the case of instantiating a generic subprogram, it is always
27426 sufficient to have only an @code{Elaborate} pragma for the
27427 @code{with}'ed unit.}
27431 By following this rule a client is assured that calls and instantiations
27432 can be made without risk of an exception.
27434 In this mode GNAT traces all calls that are potentially made from
27435 elaboration code, and puts in any missing implicit @code{Elaborate}
27436 and @code{Elaborate_All} pragmas.
27437 The advantage of this approach is that no elaboration problems
27438 are possible if the binder can find an elaboration order that is
27439 consistent with these implicit @code{Elaborate} and
27440 @code{Elaborate_All} pragmas. The
27441 disadvantage of this approach is that no such order may exist.
27443 If the binder does not generate any diagnostics, then it means that it has
27444 found an elaboration order that is guaranteed to be safe. However, the binder
27445 may still be relying on implicitly generated @code{Elaborate} and
27446 @code{Elaborate_All} pragmas so portability to other compilers than GNAT is not
27449 If it is important to guarantee portability, then the compilations should
27452 (warn on elaboration problems) switch. This will cause warning messages
27453 to be generated indicating the missing @code{Elaborate} and
27454 @code{Elaborate_All} pragmas.
27455 Consider the following source program:
27457 @smallexample @c ada
27462 m : integer := k.r;
27469 where it is clear that there
27470 should be a pragma @code{Elaborate_All}
27471 for unit @code{k}. An implicit pragma will be generated, and it is
27472 likely that the binder will be able to honor it. However, if you want
27473 to port this program to some other Ada compiler than GNAT.
27474 it is safer to include the pragma explicitly in the source. If this
27475 unit is compiled with the
27477 switch, then the compiler outputs a warning:
27484 3. m : integer := k.r;
27486 >>> warning: call to "r" may raise Program_Error
27487 >>> warning: missing pragma Elaborate_All for "k"
27495 and these warnings can be used as a guide for supplying manually
27496 the missing pragmas. It is usually a bad idea to use this warning
27497 option during development. That's because it will warn you when
27498 you need to put in a pragma, but cannot warn you when it is time
27499 to take it out. So the use of pragma @code{Elaborate_All} may lead to
27500 unnecessary dependencies and even false circularities.
27502 This default mode is more restrictive than the Ada Reference
27503 Manual, and it is possible to construct programs which will compile
27504 using the dynamic model described there, but will run into a
27505 circularity using the safer static model we have described.
27507 Of course any Ada compiler must be able to operate in a mode
27508 consistent with the requirements of the Ada Reference Manual,
27509 and in particular must have the capability of implementing the
27510 standard dynamic model of elaboration with run-time checks.
27512 In GNAT, this standard mode can be achieved either by the use of
27513 the @option{-gnatE} switch on the compiler (@command{gcc} or
27514 @command{gnatmake}) command, or by the use of the configuration pragma:
27516 @smallexample @c ada
27517 pragma Elaboration_Checks (RM);
27521 Either approach will cause the unit affected to be compiled using the
27522 standard dynamic run-time elaboration checks described in the Ada
27523 Reference Manual. The static model is generally preferable, since it
27524 is clearly safer to rely on compile and link time checks rather than
27525 run-time checks. However, in the case of legacy code, it may be
27526 difficult to meet the requirements of the static model. This
27527 issue is further discussed in
27528 @ref{What to Do If the Default Elaboration Behavior Fails}.
27530 Note that the static model provides a strict subset of the allowed
27531 behavior and programs of the Ada Reference Manual, so if you do
27532 adhere to the static model and no circularities exist,
27533 then you are assured that your program will
27534 work using the dynamic model, providing that you remove any
27535 pragma Elaborate statements from the source.
27537 @node Treatment of Pragma Elaborate
27538 @section Treatment of Pragma Elaborate
27539 @cindex Pragma Elaborate
27542 The use of @code{pragma Elaborate}
27543 should generally be avoided in Ada 95 and Ada 2005 programs,
27544 since there is no guarantee that transitive calls
27545 will be properly handled. Indeed at one point, this pragma was placed
27546 in Annex J (Obsolescent Features), on the grounds that it is never useful.
27548 Now that's a bit restrictive. In practice, the case in which
27549 @code{pragma Elaborate} is useful is when the caller knows that there
27550 are no transitive calls, or that the called unit contains all necessary
27551 transitive @code{pragma Elaborate} statements, and legacy code often
27552 contains such uses.
27554 Strictly speaking the static mode in GNAT should ignore such pragmas,
27555 since there is no assurance at compile time that the necessary safety
27556 conditions are met. In practice, this would cause GNAT to be incompatible
27557 with correctly written Ada 83 code that had all necessary
27558 @code{pragma Elaborate} statements in place. Consequently, we made the
27559 decision that GNAT in its default mode will believe that if it encounters
27560 a @code{pragma Elaborate} then the programmer knows what they are doing,
27561 and it will trust that no elaboration errors can occur.
27563 The result of this decision is two-fold. First to be safe using the
27564 static mode, you should remove all @code{pragma Elaborate} statements.
27565 Second, when fixing circularities in existing code, you can selectively
27566 use @code{pragma Elaborate} statements to convince the static mode of
27567 GNAT that it need not generate an implicit @code{pragma Elaborate_All}
27570 When using the static mode with @option{-gnatwl}, any use of
27571 @code{pragma Elaborate} will generate a warning about possible
27574 @node Elaboration Issues for Library Tasks
27575 @section Elaboration Issues for Library Tasks
27576 @cindex Library tasks, elaboration issues
27577 @cindex Elaboration of library tasks
27580 In this section we examine special elaboration issues that arise for
27581 programs that declare library level tasks.
27583 Generally the model of execution of an Ada program is that all units are
27584 elaborated, and then execution of the program starts. However, the
27585 declaration of library tasks definitely does not fit this model. The
27586 reason for this is that library tasks start as soon as they are declared
27587 (more precisely, as soon as the statement part of the enclosing package
27588 body is reached), that is to say before elaboration
27589 of the program is complete. This means that if such a task calls a
27590 subprogram, or an entry in another task, the callee may or may not be
27591 elaborated yet, and in the standard
27592 Reference Manual model of dynamic elaboration checks, you can even
27593 get timing dependent Program_Error exceptions, since there can be
27594 a race between the elaboration code and the task code.
27596 The static model of elaboration in GNAT seeks to avoid all such
27597 dynamic behavior, by being conservative, and the conservative
27598 approach in this particular case is to assume that all the code
27599 in a task body is potentially executed at elaboration time if
27600 a task is declared at the library level.
27602 This can definitely result in unexpected circularities. Consider
27603 the following example
27605 @smallexample @c ada
27611 type My_Int is new Integer;
27613 function Ident (M : My_Int) return My_Int;
27617 package body Decls is
27618 task body Lib_Task is
27624 function Ident (M : My_Int) return My_Int is
27632 procedure Put_Val (Arg : Decls.My_Int);
27636 package body Utils is
27637 procedure Put_Val (Arg : Decls.My_Int) is
27639 Text_IO.Put_Line (Decls.My_Int'Image (Decls.Ident (Arg)));
27646 Decls.Lib_Task.Start;
27651 If the above example is compiled in the default static elaboration
27652 mode, then a circularity occurs. The circularity comes from the call
27653 @code{Utils.Put_Val} in the task body of @code{Decls.Lib_Task}. Since
27654 this call occurs in elaboration code, we need an implicit pragma
27655 @code{Elaborate_All} for @code{Utils}. This means that not only must
27656 the spec and body of @code{Utils} be elaborated before the body
27657 of @code{Decls}, but also the spec and body of any unit that is
27658 @code{with'ed} by the body of @code{Utils} must also be elaborated before
27659 the body of @code{Decls}. This is the transitive implication of
27660 pragma @code{Elaborate_All} and it makes sense, because in general
27661 the body of @code{Put_Val} might have a call to something in a
27662 @code{with'ed} unit.
27664 In this case, the body of Utils (actually its spec) @code{with's}
27665 @code{Decls}. Unfortunately this means that the body of @code{Decls}
27666 must be elaborated before itself, in case there is a call from the
27667 body of @code{Utils}.
27669 Here is the exact chain of events we are worrying about:
27673 In the body of @code{Decls} a call is made from within the body of a library
27674 task to a subprogram in the package @code{Utils}. Since this call may
27675 occur at elaboration time (given that the task is activated at elaboration
27676 time), we have to assume the worst, i.e., that the
27677 call does happen at elaboration time.
27680 This means that the body and spec of @code{Util} must be elaborated before
27681 the body of @code{Decls} so that this call does not cause an access before
27685 Within the body of @code{Util}, specifically within the body of
27686 @code{Util.Put_Val} there may be calls to any unit @code{with}'ed
27690 One such @code{with}'ed package is package @code{Decls}, so there
27691 might be a call to a subprogram in @code{Decls} in @code{Put_Val}.
27692 In fact there is such a call in this example, but we would have to
27693 assume that there was such a call even if it were not there, since
27694 we are not supposed to write the body of @code{Decls} knowing what
27695 is in the body of @code{Utils}; certainly in the case of the
27696 static elaboration model, the compiler does not know what is in
27697 other bodies and must assume the worst.
27700 This means that the spec and body of @code{Decls} must also be
27701 elaborated before we elaborate the unit containing the call, but
27702 that unit is @code{Decls}! This means that the body of @code{Decls}
27703 must be elaborated before itself, and that's a circularity.
27707 Indeed, if you add an explicit pragma @code{Elaborate_All} for @code{Utils} in
27708 the body of @code{Decls} you will get a true Ada Reference Manual
27709 circularity that makes the program illegal.
27711 In practice, we have found that problems with the static model of
27712 elaboration in existing code often arise from library tasks, so
27713 we must address this particular situation.
27715 Note that if we compile and run the program above, using the dynamic model of
27716 elaboration (that is to say use the @option{-gnatE} switch),
27717 then it compiles, binds,
27718 links, and runs, printing the expected result of 2. Therefore in some sense
27719 the circularity here is only apparent, and we need to capture
27720 the properties of this program that distinguish it from other library-level
27721 tasks that have real elaboration problems.
27723 We have four possible answers to this question:
27728 Use the dynamic model of elaboration.
27730 If we use the @option{-gnatE} switch, then as noted above, the program works.
27731 Why is this? If we examine the task body, it is apparent that the task cannot
27733 @code{accept} statement until after elaboration has been completed, because
27734 the corresponding entry call comes from the main program, not earlier.
27735 This is why the dynamic model works here. But that's really giving
27736 up on a precise analysis, and we prefer to take this approach only if we cannot
27738 problem in any other manner. So let us examine two ways to reorganize
27739 the program to avoid the potential elaboration problem.
27742 Split library tasks into separate packages.
27744 Write separate packages, so that library tasks are isolated from
27745 other declarations as much as possible. Let us look at a variation on
27748 @smallexample @c ada
27756 package body Decls1 is
27757 task body Lib_Task is
27765 type My_Int is new Integer;
27766 function Ident (M : My_Int) return My_Int;
27770 package body Decls2 is
27771 function Ident (M : My_Int) return My_Int is
27779 procedure Put_Val (Arg : Decls2.My_Int);
27783 package body Utils is
27784 procedure Put_Val (Arg : Decls2.My_Int) is
27786 Text_IO.Put_Line (Decls2.My_Int'Image (Decls2.Ident (Arg)));
27793 Decls1.Lib_Task.Start;
27798 All we have done is to split @code{Decls} into two packages, one
27799 containing the library task, and one containing everything else. Now
27800 there is no cycle, and the program compiles, binds, links and executes
27801 using the default static model of elaboration.
27804 Declare separate task types.
27806 A significant part of the problem arises because of the use of the
27807 single task declaration form. This means that the elaboration of
27808 the task type, and the elaboration of the task itself (i.e.@: the
27809 creation of the task) happen at the same time. A good rule
27810 of style in Ada is to always create explicit task types. By
27811 following the additional step of placing task objects in separate
27812 packages from the task type declaration, many elaboration problems
27813 are avoided. Here is another modified example of the example program:
27815 @smallexample @c ada
27817 task type Lib_Task_Type is
27821 type My_Int is new Integer;
27823 function Ident (M : My_Int) return My_Int;
27827 package body Decls is
27828 task body Lib_Task_Type is
27834 function Ident (M : My_Int) return My_Int is
27842 procedure Put_Val (Arg : Decls.My_Int);
27846 package body Utils is
27847 procedure Put_Val (Arg : Decls.My_Int) is
27849 Text_IO.Put_Line (Decls.My_Int'Image (Decls.Ident (Arg)));
27855 Lib_Task : Decls.Lib_Task_Type;
27861 Declst.Lib_Task.Start;
27866 What we have done here is to replace the @code{task} declaration in
27867 package @code{Decls} with a @code{task type} declaration. Then we
27868 introduce a separate package @code{Declst} to contain the actual
27869 task object. This separates the elaboration issues for
27870 the @code{task type}
27871 declaration, which causes no trouble, from the elaboration issues
27872 of the task object, which is also unproblematic, since it is now independent
27873 of the elaboration of @code{Utils}.
27874 This separation of concerns also corresponds to
27875 a generally sound engineering principle of separating declarations
27876 from instances. This version of the program also compiles, binds, links,
27877 and executes, generating the expected output.
27880 Use No_Entry_Calls_In_Elaboration_Code restriction.
27881 @cindex No_Entry_Calls_In_Elaboration_Code
27883 The previous two approaches described how a program can be restructured
27884 to avoid the special problems caused by library task bodies. in practice,
27885 however, such restructuring may be difficult to apply to existing legacy code,
27886 so we must consider solutions that do not require massive rewriting.
27888 Let us consider more carefully why our original sample program works
27889 under the dynamic model of elaboration. The reason is that the code
27890 in the task body blocks immediately on the @code{accept}
27891 statement. Now of course there is nothing to prohibit elaboration
27892 code from making entry calls (for example from another library level task),
27893 so we cannot tell in isolation that
27894 the task will not execute the accept statement during elaboration.
27896 However, in practice it is very unusual to see elaboration code
27897 make any entry calls, and the pattern of tasks starting
27898 at elaboration time and then immediately blocking on @code{accept} or
27899 @code{select} statements is very common. What this means is that
27900 the compiler is being too pessimistic when it analyzes the
27901 whole package body as though it might be executed at elaboration
27904 If we know that the elaboration code contains no entry calls, (a very safe
27905 assumption most of the time, that could almost be made the default
27906 behavior), then we can compile all units of the program under control
27907 of the following configuration pragma:
27910 pragma Restrictions (No_Entry_Calls_In_Elaboration_Code);
27914 This pragma can be placed in the @file{gnat.adc} file in the usual
27915 manner. If we take our original unmodified program and compile it
27916 in the presence of a @file{gnat.adc} containing the above pragma,
27917 then once again, we can compile, bind, link, and execute, obtaining
27918 the expected result. In the presence of this pragma, the compiler does
27919 not trace calls in a task body, that appear after the first @code{accept}
27920 or @code{select} statement, and therefore does not report a potential
27921 circularity in the original program.
27923 The compiler will check to the extent it can that the above
27924 restriction is not violated, but it is not always possible to do a
27925 complete check at compile time, so it is important to use this
27926 pragma only if the stated restriction is in fact met, that is to say
27927 no task receives an entry call before elaboration of all units is completed.
27931 @node Mixing Elaboration Models
27932 @section Mixing Elaboration Models
27934 So far, we have assumed that the entire program is either compiled
27935 using the dynamic model or static model, ensuring consistency. It
27936 is possible to mix the two models, but rules have to be followed
27937 if this mixing is done to ensure that elaboration checks are not
27940 The basic rule is that @emph{a unit compiled with the static model cannot
27941 be @code{with'ed} by a unit compiled with the dynamic model}. The
27942 reason for this is that in the static model, a unit assumes that
27943 its clients guarantee to use (the equivalent of) pragma
27944 @code{Elaborate_All} so that no elaboration checks are required
27945 in inner subprograms, and this assumption is violated if the
27946 client is compiled with dynamic checks.
27948 The precise rule is as follows. A unit that is compiled with dynamic
27949 checks can only @code{with} a unit that meets at least one of the
27950 following criteria:
27955 The @code{with'ed} unit is itself compiled with dynamic elaboration
27956 checks (that is with the @option{-gnatE} switch.
27959 The @code{with'ed} unit is an internal GNAT implementation unit from
27960 the System, Interfaces, Ada, or GNAT hierarchies.
27963 The @code{with'ed} unit has pragma Preelaborate or pragma Pure.
27966 The @code{with'ing} unit (that is the client) has an explicit pragma
27967 @code{Elaborate_All} for the @code{with'ed} unit.
27972 If this rule is violated, that is if a unit with dynamic elaboration
27973 checks @code{with's} a unit that does not meet one of the above four
27974 criteria, then the binder (@code{gnatbind}) will issue a warning
27975 similar to that in the following example:
27978 warning: "x.ads" has dynamic elaboration checks and with's
27979 warning: "y.ads" which has static elaboration checks
27983 These warnings indicate that the rule has been violated, and that as a result
27984 elaboration checks may be missed in the resulting executable file.
27985 This warning may be suppressed using the @option{-ws} binder switch
27986 in the usual manner.
27988 One useful application of this mixing rule is in the case of a subsystem
27989 which does not itself @code{with} units from the remainder of the
27990 application. In this case, the entire subsystem can be compiled with
27991 dynamic checks to resolve a circularity in the subsystem, while
27992 allowing the main application that uses this subsystem to be compiled
27993 using the more reliable default static model.
27995 @node What to Do If the Default Elaboration Behavior Fails
27996 @section What to Do If the Default Elaboration Behavior Fails
27999 If the binder cannot find an acceptable order, it outputs detailed
28000 diagnostics. For example:
28006 error: elaboration circularity detected
28007 info: "proc (body)" must be elaborated before "pack (body)"
28008 info: reason: Elaborate_All probably needed in unit "pack (body)"
28009 info: recompile "pack (body)" with -gnatwl
28010 info: for full details
28011 info: "proc (body)"
28012 info: is needed by its spec:
28013 info: "proc (spec)"
28014 info: which is withed by:
28015 info: "pack (body)"
28016 info: "pack (body)" must be elaborated before "proc (body)"
28017 info: reason: pragma Elaborate in unit "proc (body)"
28023 In this case we have a cycle that the binder cannot break. On the one
28024 hand, there is an explicit pragma Elaborate in @code{proc} for
28025 @code{pack}. This means that the body of @code{pack} must be elaborated
28026 before the body of @code{proc}. On the other hand, there is elaboration
28027 code in @code{pack} that calls a subprogram in @code{proc}. This means
28028 that for maximum safety, there should really be a pragma
28029 Elaborate_All in @code{pack} for @code{proc} which would require that
28030 the body of @code{proc} be elaborated before the body of
28031 @code{pack}. Clearly both requirements cannot be satisfied.
28032 Faced with a circularity of this kind, you have three different options.
28035 @item Fix the program
28036 The most desirable option from the point of view of long-term maintenance
28037 is to rearrange the program so that the elaboration problems are avoided.
28038 One useful technique is to place the elaboration code into separate
28039 child packages. Another is to move some of the initialization code to
28040 explicitly called subprograms, where the program controls the order
28041 of initialization explicitly. Although this is the most desirable option,
28042 it may be impractical and involve too much modification, especially in
28043 the case of complex legacy code.
28045 @item Perform dynamic checks
28046 If the compilations are done using the
28048 (dynamic elaboration check) switch, then GNAT behaves in a quite different
28049 manner. Dynamic checks are generated for all calls that could possibly result
28050 in raising an exception. With this switch, the compiler does not generate
28051 implicit @code{Elaborate} or @code{Elaborate_All} pragmas. The behavior then is
28052 exactly as specified in the @cite{Ada Reference Manual}.
28053 The binder will generate
28054 an executable program that may or may not raise @code{Program_Error}, and then
28055 it is the programmer's job to ensure that it does not raise an exception. Note
28056 that it is important to compile all units with the switch, it cannot be used
28059 @item Suppress checks
28060 The drawback of dynamic checks is that they generate a
28061 significant overhead at run time, both in space and time. If you
28062 are absolutely sure that your program cannot raise any elaboration
28063 exceptions, and you still want to use the dynamic elaboration model,
28064 then you can use the configuration pragma
28065 @code{Suppress (Elaboration_Check)} to suppress all such checks. For
28066 example this pragma could be placed in the @file{gnat.adc} file.
28068 @item Suppress checks selectively
28069 When you know that certain calls or instantiations in elaboration code cannot
28070 possibly lead to an elaboration error, and the binder nevertheless complains
28071 about implicit @code{Elaborate} and @code{Elaborate_All} pragmas that lead to
28072 elaboration circularities, it is possible to remove those warnings locally and
28073 obtain a program that will bind. Clearly this can be unsafe, and it is the
28074 responsibility of the programmer to make sure that the resulting program has no
28075 elaboration anomalies. The pragma @code{Suppress (Elaboration_Check)} can be
28076 used with different granularity to suppress warnings and break elaboration
28081 Place the pragma that names the called subprogram in the declarative part
28082 that contains the call.
28085 Place the pragma in the declarative part, without naming an entity. This
28086 disables warnings on all calls in the corresponding declarative region.
28089 Place the pragma in the package spec that declares the called subprogram,
28090 and name the subprogram. This disables warnings on all elaboration calls to
28094 Place the pragma in the package spec that declares the called subprogram,
28095 without naming any entity. This disables warnings on all elaboration calls to
28096 all subprograms declared in this spec.
28098 @item Use Pragma Elaborate
28099 As previously described in section @xref{Treatment of Pragma Elaborate},
28100 GNAT in static mode assumes that a @code{pragma} Elaborate indicates correctly
28101 that no elaboration checks are required on calls to the designated unit.
28102 There may be cases in which the caller knows that no transitive calls
28103 can occur, so that a @code{pragma Elaborate} will be sufficient in a
28104 case where @code{pragma Elaborate_All} would cause a circularity.
28108 These five cases are listed in order of decreasing safety, and therefore
28109 require increasing programmer care in their application. Consider the
28112 @smallexample @c adanocomment
28114 function F1 return Integer;
28119 function F2 return Integer;
28120 function Pure (x : integer) return integer;
28121 -- pragma Suppress (Elaboration_Check, On => Pure); -- (3)
28122 -- pragma Suppress (Elaboration_Check); -- (4)
28126 package body Pack1 is
28127 function F1 return Integer is
28131 Val : integer := Pack2.Pure (11); -- Elab. call (1)
28134 -- pragma Suppress(Elaboration_Check, Pack2.F2); -- (1)
28135 -- pragma Suppress(Elaboration_Check); -- (2)
28137 X1 := Pack2.F2 + 1; -- Elab. call (2)
28142 package body Pack2 is
28143 function F2 return Integer is
28147 function Pure (x : integer) return integer is
28149 return x ** 3 - 3 * x;
28153 with Pack1, Ada.Text_IO;
28156 Ada.Text_IO.Put_Line(Pack1.X1'Img); -- 101
28159 In the absence of any pragmas, an attempt to bind this program produces
28160 the following diagnostics:
28166 error: elaboration circularity detected
28167 info: "pack1 (body)" must be elaborated before "pack1 (body)"
28168 info: reason: Elaborate_All probably needed in unit "pack1 (body)"
28169 info: recompile "pack1 (body)" with -gnatwl for full details
28170 info: "pack1 (body)"
28171 info: must be elaborated along with its spec:
28172 info: "pack1 (spec)"
28173 info: which is withed by:
28174 info: "pack2 (body)"
28175 info: which must be elaborated along with its spec:
28176 info: "pack2 (spec)"
28177 info: which is withed by:
28178 info: "pack1 (body)"
28181 The sources of the circularity are the two calls to @code{Pack2.Pure} and
28182 @code{Pack2.F2} in the body of @code{Pack1}. We can see that the call to
28183 F2 is safe, even though F2 calls F1, because the call appears after the
28184 elaboration of the body of F1. Therefore the pragma (1) is safe, and will
28185 remove the warning on the call. It is also possible to use pragma (2)
28186 because there are no other potentially unsafe calls in the block.
28189 The call to @code{Pure} is safe because this function does not depend on the
28190 state of @code{Pack2}. Therefore any call to this function is safe, and it
28191 is correct to place pragma (3) in the corresponding package spec.
28194 Finally, we could place pragma (4) in the spec of @code{Pack2} to disable
28195 warnings on all calls to functions declared therein. Note that this is not
28196 necessarily safe, and requires more detailed examination of the subprogram
28197 bodies involved. In particular, a call to @code{F2} requires that @code{F1}
28198 be already elaborated.
28202 It is hard to generalize on which of these four approaches should be
28203 taken. Obviously if it is possible to fix the program so that the default
28204 treatment works, this is preferable, but this may not always be practical.
28205 It is certainly simple enough to use
28207 but the danger in this case is that, even if the GNAT binder
28208 finds a correct elaboration order, it may not always do so,
28209 and certainly a binder from another Ada compiler might not. A
28210 combination of testing and analysis (for which the warnings generated
28213 switch can be useful) must be used to ensure that the program is free
28214 of errors. One switch that is useful in this testing is the
28215 @option{^-p (pessimistic elaboration order)^/PESSIMISTIC_ELABORATION_ORDER^}
28218 Normally the binder tries to find an order that has the best chance
28219 of avoiding elaboration problems. However, if this switch is used, the binder
28220 plays a devil's advocate role, and tries to choose the order that
28221 has the best chance of failing. If your program works even with this
28222 switch, then it has a better chance of being error free, but this is still
28225 For an example of this approach in action, consider the C-tests (executable
28226 tests) from the ACVC suite. If these are compiled and run with the default
28227 treatment, then all but one of them succeed without generating any error
28228 diagnostics from the binder. However, there is one test that fails, and
28229 this is not surprising, because the whole point of this test is to ensure
28230 that the compiler can handle cases where it is impossible to determine
28231 a correct order statically, and it checks that an exception is indeed
28232 raised at run time.
28234 This one test must be compiled and run using the
28236 switch, and then it passes. Alternatively, the entire suite can
28237 be run using this switch. It is never wrong to run with the dynamic
28238 elaboration switch if your code is correct, and we assume that the
28239 C-tests are indeed correct (it is less efficient, but efficiency is
28240 not a factor in running the ACVC tests.)
28242 @node Elaboration for Access-to-Subprogram Values
28243 @section Elaboration for Access-to-Subprogram Values
28244 @cindex Access-to-subprogram
28247 Access-to-subprogram types (introduced in Ada 95) complicate
28248 the handling of elaboration. The trouble is that it becomes
28249 impossible to tell at compile time which procedure
28250 is being called. This means that it is not possible for the binder
28251 to analyze the elaboration requirements in this case.
28253 If at the point at which the access value is created
28254 (i.e., the evaluation of @code{P'Access} for a subprogram @code{P}),
28255 the body of the subprogram is
28256 known to have been elaborated, then the access value is safe, and its use
28257 does not require a check. This may be achieved by appropriate arrangement
28258 of the order of declarations if the subprogram is in the current unit,
28259 or, if the subprogram is in another unit, by using pragma
28260 @code{Pure}, @code{Preelaborate}, or @code{Elaborate_Body}
28261 on the referenced unit.
28263 If the referenced body is not known to have been elaborated at the point
28264 the access value is created, then any use of the access value must do a
28265 dynamic check, and this dynamic check will fail and raise a
28266 @code{Program_Error} exception if the body has not been elaborated yet.
28267 GNAT will generate the necessary checks, and in addition, if the
28269 switch is set, will generate warnings that such checks are required.
28271 The use of dynamic dispatching for tagged types similarly generates
28272 a requirement for dynamic checks, and premature calls to any primitive
28273 operation of a tagged type before the body of the operation has been
28274 elaborated, will result in the raising of @code{Program_Error}.
28276 @node Summary of Procedures for Elaboration Control
28277 @section Summary of Procedures for Elaboration Control
28278 @cindex Elaboration control
28281 First, compile your program with the default options, using none of
28282 the special elaboration control switches. If the binder successfully
28283 binds your program, then you can be confident that, apart from issues
28284 raised by the use of access-to-subprogram types and dynamic dispatching,
28285 the program is free of elaboration errors. If it is important that the
28286 program be portable, then use the
28288 switch to generate warnings about missing @code{Elaborate} or
28289 @code{Elaborate_All} pragmas, and supply the missing pragmas.
28291 If the program fails to bind using the default static elaboration
28292 handling, then you can fix the program to eliminate the binder
28293 message, or recompile the entire program with the
28294 @option{-gnatE} switch to generate dynamic elaboration checks,
28295 and, if you are sure there really are no elaboration problems,
28296 use a global pragma @code{Suppress (Elaboration_Check)}.
28298 @node Other Elaboration Order Considerations
28299 @section Other Elaboration Order Considerations
28301 This section has been entirely concerned with the issue of finding a valid
28302 elaboration order, as defined by the Ada Reference Manual. In a case
28303 where several elaboration orders are valid, the task is to find one
28304 of the possible valid elaboration orders (and the static model in GNAT
28305 will ensure that this is achieved).
28307 The purpose of the elaboration rules in the Ada Reference Manual is to
28308 make sure that no entity is accessed before it has been elaborated. For
28309 a subprogram, this means that the spec and body must have been elaborated
28310 before the subprogram is called. For an object, this means that the object
28311 must have been elaborated before its value is read or written. A violation
28312 of either of these two requirements is an access before elaboration order,
28313 and this section has been all about avoiding such errors.
28315 In the case where more than one order of elaboration is possible, in the
28316 sense that access before elaboration errors are avoided, then any one of
28317 the orders is ``correct'' in the sense that it meets the requirements of
28318 the Ada Reference Manual, and no such error occurs.
28320 However, it may be the case for a given program, that there are
28321 constraints on the order of elaboration that come not from consideration
28322 of avoiding elaboration errors, but rather from extra-lingual logic
28323 requirements. Consider this example:
28325 @smallexample @c ada
28326 with Init_Constants;
28327 package Constants is
28332 package Init_Constants is
28333 procedure P; -- require a body
28334 end Init_Constants;
28337 package body Init_Constants is
28338 procedure P is begin null; end;
28342 end Init_Constants;
28346 Z : Integer := Constants.X + Constants.Y;
28350 with Text_IO; use Text_IO;
28353 Put_Line (Calc.Z'Img);
28358 In this example, there is more than one valid order of elaboration. For
28359 example both the following are correct orders:
28362 Init_Constants spec
28365 Init_Constants body
28370 Init_Constants spec
28371 Init_Constants body
28378 There is no language rule to prefer one or the other, both are correct
28379 from an order of elaboration point of view. But the programmatic effects
28380 of the two orders are very different. In the first, the elaboration routine
28381 of @code{Calc} initializes @code{Z} to zero, and then the main program
28382 runs with this value of zero. But in the second order, the elaboration
28383 routine of @code{Calc} runs after the body of Init_Constants has set
28384 @code{X} and @code{Y} and thus @code{Z} is set to 7 before @code{Main}
28387 One could perhaps by applying pretty clever non-artificial intelligence
28388 to the situation guess that it is more likely that the second order of
28389 elaboration is the one desired, but there is no formal linguistic reason
28390 to prefer one over the other. In fact in this particular case, GNAT will
28391 prefer the second order, because of the rule that bodies are elaborated
28392 as soon as possible, but it's just luck that this is what was wanted
28393 (if indeed the second order was preferred).
28395 If the program cares about the order of elaboration routines in a case like
28396 this, it is important to specify the order required. In this particular
28397 case, that could have been achieved by adding to the spec of Calc:
28399 @smallexample @c ada
28400 pragma Elaborate_All (Constants);
28404 which requires that the body (if any) and spec of @code{Constants},
28405 as well as the body and spec of any unit @code{with}'ed by
28406 @code{Constants} be elaborated before @code{Calc} is elaborated.
28408 Clearly no automatic method can always guess which alternative you require,
28409 and if you are working with legacy code that had constraints of this kind
28410 which were not properly specified by adding @code{Elaborate} or
28411 @code{Elaborate_All} pragmas, then indeed it is possible that two different
28412 compilers can choose different orders.
28414 However, GNAT does attempt to diagnose the common situation where there
28415 are uninitialized variables in the visible part of a package spec, and the
28416 corresponding package body has an elaboration block that directly or
28417 indirectly initialized one or more of these variables. This is the situation
28418 in which a pragma Elaborate_Body is usually desirable, and GNAT will generate
28419 a warning that suggests this addition if it detects this situation.
28421 The @code{gnatbind}
28422 @option{^-p^/PESSIMISTIC_ELABORATION^} switch may be useful in smoking
28423 out problems. This switch causes bodies to be elaborated as late as possible
28424 instead of as early as possible. In the example above, it would have forced
28425 the choice of the first elaboration order. If you get different results
28426 when using this switch, and particularly if one set of results is right,
28427 and one is wrong as far as you are concerned, it shows that you have some
28428 missing @code{Elaborate} pragmas. For the example above, we have the
28432 gnatmake -f -q main
28435 gnatmake -f -q main -bargs -p
28441 It is of course quite unlikely that both these results are correct, so
28442 it is up to you in a case like this to investigate the source of the
28443 difference, by looking at the two elaboration orders that are chosen,
28444 and figuring out which is correct, and then adding the necessary
28445 @code{Elaborate} or @code{Elaborate_All} pragmas to ensure the desired order.
28449 @c *******************************
28450 @node Conditional Compilation
28451 @appendix Conditional Compilation
28452 @c *******************************
28453 @cindex Conditional compilation
28456 It is often necessary to arrange for a single source program
28457 to serve multiple purposes, where it is compiled in different
28458 ways to achieve these different goals. Some examples of the
28459 need for this feature are
28462 @item Adapting a program to a different hardware environment
28463 @item Adapting a program to a different target architecture
28464 @item Turning debugging features on and off
28465 @item Arranging for a program to compile with different compilers
28469 In C, or C++, the typical approach would be to use the preprocessor
28470 that is defined as part of the language. The Ada language does not
28471 contain such a feature. This is not an oversight, but rather a very
28472 deliberate design decision, based on the experience that overuse of
28473 the preprocessing features in C and C++ can result in programs that
28474 are extremely difficult to maintain. For example, if we have ten
28475 switches that can be on or off, this means that there are a thousand
28476 separate programs, any one of which might not even be syntactically
28477 correct, and even if syntactically correct, the resulting program
28478 might not work correctly. Testing all combinations can quickly become
28481 Nevertheless, the need to tailor programs certainly exists, and in
28482 this Appendix we will discuss how this can
28483 be achieved using Ada in general, and GNAT in particular.
28486 * Use of Boolean Constants::
28487 * Debugging - A Special Case::
28488 * Conditionalizing Declarations::
28489 * Use of Alternative Implementations::
28493 @node Use of Boolean Constants
28494 @section Use of Boolean Constants
28497 In the case where the difference is simply which code
28498 sequence is executed, the cleanest solution is to use Boolean
28499 constants to control which code is executed.
28501 @smallexample @c ada
28503 FP_Initialize_Required : constant Boolean := True;
28505 if FP_Initialize_Required then
28512 Not only will the code inside the @code{if} statement not be executed if
28513 the constant Boolean is @code{False}, but it will also be completely
28514 deleted from the program.
28515 However, the code is only deleted after the @code{if} statement
28516 has been checked for syntactic and semantic correctness.
28517 (In contrast, with preprocessors the code is deleted before the
28518 compiler ever gets to see it, so it is not checked until the switch
28520 @cindex Preprocessors (contrasted with conditional compilation)
28522 Typically the Boolean constants will be in a separate package,
28525 @smallexample @c ada
28528 FP_Initialize_Required : constant Boolean := True;
28529 Reset_Available : constant Boolean := False;
28536 The @code{Config} package exists in multiple forms for the various targets,
28537 with an appropriate script selecting the version of @code{Config} needed.
28538 Then any other unit requiring conditional compilation can do a @code{with}
28539 of @code{Config} to make the constants visible.
28542 @node Debugging - A Special Case
28543 @section Debugging - A Special Case
28546 A common use of conditional code is to execute statements (for example
28547 dynamic checks, or output of intermediate results) under control of a
28548 debug switch, so that the debugging behavior can be turned on and off.
28549 This can be done using a Boolean constant to control whether the code
28552 @smallexample @c ada
28555 Put_Line ("got to the first stage!");
28563 @smallexample @c ada
28565 if Debugging and then Temperature > 999.0 then
28566 raise Temperature_Crazy;
28572 Since this is a common case, there are special features to deal with
28573 this in a convenient manner. For the case of tests, Ada 2005 has added
28574 a pragma @code{Assert} that can be used for such tests. This pragma is modeled
28575 @cindex pragma @code{Assert}
28576 on the @code{Assert} pragma that has always been available in GNAT, so this
28577 feature may be used with GNAT even if you are not using Ada 2005 features.
28578 The use of pragma @code{Assert} is described in
28579 @ref{Pragma Assert,,, gnat_rm, GNAT Reference Manual}, but as an
28580 example, the last test could be written:
28582 @smallexample @c ada
28583 pragma Assert (Temperature <= 999.0, "Temperature Crazy");
28589 @smallexample @c ada
28590 pragma Assert (Temperature <= 999.0);
28594 In both cases, if assertions are active and the temperature is excessive,
28595 the exception @code{Assert_Failure} will be raised, with the given string in
28596 the first case or a string indicating the location of the pragma in the second
28597 case used as the exception message.
28599 You can turn assertions on and off by using the @code{Assertion_Policy}
28601 @cindex pragma @code{Assertion_Policy}
28602 This is an Ada 2005 pragma which is implemented in all modes by
28603 GNAT, but only in the latest versions of GNAT which include Ada 2005
28604 capability. Alternatively, you can use the @option{-gnata} switch
28605 @cindex @option{-gnata} switch
28606 to enable assertions from the command line (this is recognized by all versions
28609 For the example above with the @code{Put_Line}, the GNAT-specific pragma
28610 @code{Debug} can be used:
28611 @cindex pragma @code{Debug}
28613 @smallexample @c ada
28614 pragma Debug (Put_Line ("got to the first stage!"));
28618 If debug pragmas are enabled, the argument, which must be of the form of
28619 a procedure call, is executed (in this case, @code{Put_Line} will be called).
28620 Only one call can be present, but of course a special debugging procedure
28621 containing any code you like can be included in the program and then
28622 called in a pragma @code{Debug} argument as needed.
28624 One advantage of pragma @code{Debug} over the @code{if Debugging then}
28625 construct is that pragma @code{Debug} can appear in declarative contexts,
28626 such as at the very beginning of a procedure, before local declarations have
28629 Debug pragmas are enabled using either the @option{-gnata} switch that also
28630 controls assertions, or with a separate Debug_Policy pragma.
28631 @cindex pragma @code{Debug_Policy}
28632 The latter pragma is new in the Ada 2005 versions of GNAT (but it can be used
28633 in Ada 95 and Ada 83 programs as well), and is analogous to
28634 pragma @code{Assertion_Policy} to control assertions.
28636 @code{Assertion_Policy} and @code{Debug_Policy} are configuration pragmas,
28637 and thus they can appear in @file{gnat.adc} if you are not using a
28638 project file, or in the file designated to contain configuration pragmas
28640 They then apply to all subsequent compilations. In practice the use of
28641 the @option{-gnata} switch is often the most convenient method of controlling
28642 the status of these pragmas.
28644 Note that a pragma is not a statement, so in contexts where a statement
28645 sequence is required, you can't just write a pragma on its own. You have
28646 to add a @code{null} statement.
28648 @smallexample @c ada
28651 @dots{} -- some statements
28653 pragma Assert (Num_Cases < 10);
28660 @node Conditionalizing Declarations
28661 @section Conditionalizing Declarations
28664 In some cases, it may be necessary to conditionalize declarations to meet
28665 different requirements. For example we might want a bit string whose length
28666 is set to meet some hardware message requirement.
28668 In some cases, it may be possible to do this using declare blocks controlled
28669 by conditional constants:
28671 @smallexample @c ada
28673 if Small_Machine then
28675 X : Bit_String (1 .. 10);
28681 X : Large_Bit_String (1 .. 1000);
28690 Note that in this approach, both declarations are analyzed by the
28691 compiler so this can only be used where both declarations are legal,
28692 even though one of them will not be used.
28694 Another approach is to define integer constants, e.g.@: @code{Bits_Per_Word}, or
28695 Boolean constants, e.g.@: @code{Little_Endian}, and then write declarations
28696 that are parameterized by these constants. For example
28698 @smallexample @c ada
28701 Field1 at 0 range Boolean'Pos (Little_Endian) * 10 .. Bits_Per_Word;
28707 If @code{Bits_Per_Word} is set to 32, this generates either
28709 @smallexample @c ada
28712 Field1 at 0 range 0 .. 32;
28718 for the big endian case, or
28720 @smallexample @c ada
28723 Field1 at 0 range 10 .. 32;
28729 for the little endian case. Since a powerful subset of Ada expression
28730 notation is usable for creating static constants, clever use of this
28731 feature can often solve quite difficult problems in conditionalizing
28732 compilation (note incidentally that in Ada 95, the little endian
28733 constant was introduced as @code{System.Default_Bit_Order}, so you do not
28734 need to define this one yourself).
28737 @node Use of Alternative Implementations
28738 @section Use of Alternative Implementations
28741 In some cases, none of the approaches described above are adequate. This
28742 can occur for example if the set of declarations required is radically
28743 different for two different configurations.
28745 In this situation, the official Ada way of dealing with conditionalizing
28746 such code is to write separate units for the different cases. As long as
28747 this does not result in excessive duplication of code, this can be done
28748 without creating maintenance problems. The approach is to share common
28749 code as far as possible, and then isolate the code and declarations
28750 that are different. Subunits are often a convenient method for breaking
28751 out a piece of a unit that is to be conditionalized, with separate files
28752 for different versions of the subunit for different targets, where the
28753 build script selects the right one to give to the compiler.
28754 @cindex Subunits (and conditional compilation)
28756 As an example, consider a situation where a new feature in Ada 2005
28757 allows something to be done in a really nice way. But your code must be able
28758 to compile with an Ada 95 compiler. Conceptually you want to say:
28760 @smallexample @c ada
28763 @dots{} neat Ada 2005 code
28765 @dots{} not quite as neat Ada 95 code
28771 where @code{Ada_2005} is a Boolean constant.
28773 But this won't work when @code{Ada_2005} is set to @code{False},
28774 since the @code{then} clause will be illegal for an Ada 95 compiler.
28775 (Recall that although such unreachable code would eventually be deleted
28776 by the compiler, it still needs to be legal. If it uses features
28777 introduced in Ada 2005, it will be illegal in Ada 95.)
28779 So instead we write
28781 @smallexample @c ada
28782 procedure Insert is separate;
28786 Then we have two files for the subunit @code{Insert}, with the two sets of
28788 If the package containing this is called @code{File_Queries}, then we might
28792 @item @file{file_queries-insert-2005.adb}
28793 @item @file{file_queries-insert-95.adb}
28797 and the build script renames the appropriate file to
28800 file_queries-insert.adb
28804 and then carries out the compilation.
28806 This can also be done with project files' naming schemes. For example:
28808 @smallexample @c project
28809 For Body ("File_Queries.Insert") use "file_queries-insert-2005.ada";
28813 Note also that with project files it is desirable to use a different extension
28814 than @file{ads} / @file{adb} for alternative versions. Otherwise a naming
28815 conflict may arise through another commonly used feature: to declare as part
28816 of the project a set of directories containing all the sources obeying the
28817 default naming scheme.
28819 The use of alternative units is certainly feasible in all situations,
28820 and for example the Ada part of the GNAT run-time is conditionalized
28821 based on the target architecture using this approach. As a specific example,
28822 consider the implementation of the AST feature in VMS. There is one
28830 which is the same for all architectures, and three bodies:
28834 used for all non-VMS operating systems
28835 @item s-asthan-vms-alpha.adb
28836 used for VMS on the Alpha
28837 @item s-asthan-vms-ia64.adb
28838 used for VMS on the ia64
28842 The dummy version @file{s-asthan.adb} simply raises exceptions noting that
28843 this operating system feature is not available, and the two remaining
28844 versions interface with the corresponding versions of VMS to provide
28845 VMS-compatible AST handling. The GNAT build script knows the architecture
28846 and operating system, and automatically selects the right version,
28847 renaming it if necessary to @file{s-asthan.adb} before the run-time build.
28849 Another style for arranging alternative implementations is through Ada's
28850 access-to-subprogram facility.
28851 In case some functionality is to be conditionally included,
28852 you can declare an access-to-procedure variable @code{Ref} that is initialized
28853 to designate a ``do nothing'' procedure, and then invoke @code{Ref.all}
28855 In some library package, set @code{Ref} to @code{Proc'Access} for some
28856 procedure @code{Proc} that performs the relevant processing.
28857 The initialization only occurs if the library package is included in the
28859 The same idea can also be implemented using tagged types and dispatching
28863 @node Preprocessing
28864 @section Preprocessing
28865 @cindex Preprocessing
28868 Although it is quite possible to conditionalize code without the use of
28869 C-style preprocessing, as described earlier in this section, it is
28870 nevertheless convenient in some cases to use the C approach. Moreover,
28871 older Ada compilers have often provided some preprocessing capability,
28872 so legacy code may depend on this approach, even though it is not
28875 To accommodate such use, GNAT provides a preprocessor (modeled to a large
28876 extent on the various preprocessors that have been used
28877 with legacy code on other compilers, to enable easier transition).
28879 The preprocessor may be used in two separate modes. It can be used quite
28880 separately from the compiler, to generate a separate output source file
28881 that is then fed to the compiler as a separate step. This is the
28882 @code{gnatprep} utility, whose use is fully described in
28883 @ref{Preprocessing Using gnatprep}.
28884 @cindex @code{gnatprep}
28886 The preprocessing language allows such constructs as
28890 #if DEBUG or PRIORITY > 4 then
28891 bunch of declarations
28893 completely different bunch of declarations
28899 The values of the symbols @code{DEBUG} and @code{PRIORITY} can be
28900 defined either on the command line or in a separate file.
28902 The other way of running the preprocessor is even closer to the C style and
28903 often more convenient. In this approach the preprocessing is integrated into
28904 the compilation process. The compiler is fed the preprocessor input which
28905 includes @code{#if} lines etc, and then the compiler carries out the
28906 preprocessing internally and processes the resulting output.
28907 For more details on this approach, see @ref{Integrated Preprocessing}.
28910 @c *******************************
28911 @node Inline Assembler
28912 @appendix Inline Assembler
28913 @c *******************************
28916 If you need to write low-level software that interacts directly
28917 with the hardware, Ada provides two ways to incorporate assembly
28918 language code into your program. First, you can import and invoke
28919 external routines written in assembly language, an Ada feature fully
28920 supported by GNAT@. However, for small sections of code it may be simpler
28921 or more efficient to include assembly language statements directly
28922 in your Ada source program, using the facilities of the implementation-defined
28923 package @code{System.Machine_Code}, which incorporates the gcc
28924 Inline Assembler. The Inline Assembler approach offers a number of advantages,
28925 including the following:
28928 @item No need to use non-Ada tools
28929 @item Consistent interface over different targets
28930 @item Automatic usage of the proper calling conventions
28931 @item Access to Ada constants and variables
28932 @item Definition of intrinsic routines
28933 @item Possibility of inlining a subprogram comprising assembler code
28934 @item Code optimizer can take Inline Assembler code into account
28937 This chapter presents a series of examples to show you how to use
28938 the Inline Assembler. Although it focuses on the Intel x86,
28939 the general approach applies also to other processors.
28940 It is assumed that you are familiar with Ada
28941 and with assembly language programming.
28944 * Basic Assembler Syntax::
28945 * A Simple Example of Inline Assembler::
28946 * Output Variables in Inline Assembler::
28947 * Input Variables in Inline Assembler::
28948 * Inlining Inline Assembler Code::
28949 * Other Asm Functionality::
28952 @c ---------------------------------------------------------------------------
28953 @node Basic Assembler Syntax
28954 @section Basic Assembler Syntax
28957 The assembler used by GNAT and gcc is based not on the Intel assembly
28958 language, but rather on a language that descends from the AT&T Unix
28959 assembler @emph{as} (and which is often referred to as ``AT&T syntax'').
28960 The following table summarizes the main features of @emph{as} syntax
28961 and points out the differences from the Intel conventions.
28962 See the gcc @emph{as} and @emph{gas} (an @emph{as} macro
28963 pre-processor) documentation for further information.
28966 @item Register names
28967 gcc / @emph{as}: Prefix with ``%''; for example @code{%eax}
28969 Intel: No extra punctuation; for example @code{eax}
28971 @item Immediate operand
28972 gcc / @emph{as}: Prefix with ``$''; for example @code{$4}
28974 Intel: No extra punctuation; for example @code{4}
28977 gcc / @emph{as}: Prefix with ``$''; for example @code{$loc}
28979 Intel: No extra punctuation; for example @code{loc}
28981 @item Memory contents
28982 gcc / @emph{as}: No extra punctuation; for example @code{loc}
28984 Intel: Square brackets; for example @code{[loc]}
28986 @item Register contents
28987 gcc / @emph{as}: Parentheses; for example @code{(%eax)}
28989 Intel: Square brackets; for example @code{[eax]}
28991 @item Hexadecimal numbers
28992 gcc / @emph{as}: Leading ``0x'' (C language syntax); for example @code{0xA0}
28994 Intel: Trailing ``h''; for example @code{A0h}
28997 gcc / @emph{as}: Explicit in op code; for example @code{movw} to move
29000 Intel: Implicit, deduced by assembler; for example @code{mov}
29002 @item Instruction repetition
29003 gcc / @emph{as}: Split into two lines; for example
29009 Intel: Keep on one line; for example @code{rep stosl}
29011 @item Order of operands
29012 gcc / @emph{as}: Source first; for example @code{movw $4, %eax}
29014 Intel: Destination first; for example @code{mov eax, 4}
29017 @c ---------------------------------------------------------------------------
29018 @node A Simple Example of Inline Assembler
29019 @section A Simple Example of Inline Assembler
29022 The following example will generate a single assembly language statement,
29023 @code{nop}, which does nothing. Despite its lack of run-time effect,
29024 the example will be useful in illustrating the basics of
29025 the Inline Assembler facility.
29027 @smallexample @c ada
29029 with System.Machine_Code; use System.Machine_Code;
29030 procedure Nothing is
29037 @code{Asm} is a procedure declared in package @code{System.Machine_Code};
29038 here it takes one parameter, a @emph{template string} that must be a static
29039 expression and that will form the generated instruction.
29040 @code{Asm} may be regarded as a compile-time procedure that parses
29041 the template string and additional parameters (none here),
29042 from which it generates a sequence of assembly language instructions.
29044 The examples in this chapter will illustrate several of the forms
29045 for invoking @code{Asm}; a complete specification of the syntax
29046 is found in @ref{Machine Code Insertions,,, gnat_rm, GNAT Reference
29049 Under the standard GNAT conventions, the @code{Nothing} procedure
29050 should be in a file named @file{nothing.adb}.
29051 You can build the executable in the usual way:
29055 However, the interesting aspect of this example is not its run-time behavior
29056 but rather the generated assembly code.
29057 To see this output, invoke the compiler as follows:
29059 gcc -c -S -fomit-frame-pointer -gnatp @file{nothing.adb}
29061 where the options are:
29065 compile only (no bind or link)
29067 generate assembler listing
29068 @item -fomit-frame-pointer
29069 do not set up separate stack frames
29071 do not add runtime checks
29074 This gives a human-readable assembler version of the code. The resulting
29075 file will have the same name as the Ada source file, but with a @code{.s}
29076 extension. In our example, the file @file{nothing.s} has the following
29081 .file "nothing.adb"
29083 ___gnu_compiled_ada:
29086 .globl __ada_nothing
29098 The assembly code you included is clearly indicated by
29099 the compiler, between the @code{#APP} and @code{#NO_APP}
29100 delimiters. The character before the 'APP' and 'NOAPP'
29101 can differ on different targets. For example, GNU/Linux uses '#APP' while
29102 on NT you will see '/APP'.
29104 If you make a mistake in your assembler code (such as using the
29105 wrong size modifier, or using a wrong operand for the instruction) GNAT
29106 will report this error in a temporary file, which will be deleted when
29107 the compilation is finished. Generating an assembler file will help
29108 in such cases, since you can assemble this file separately using the
29109 @emph{as} assembler that comes with gcc.
29111 Assembling the file using the command
29114 as @file{nothing.s}
29117 will give you error messages whose lines correspond to the assembler
29118 input file, so you can easily find and correct any mistakes you made.
29119 If there are no errors, @emph{as} will generate an object file
29120 @file{nothing.out}.
29122 @c ---------------------------------------------------------------------------
29123 @node Output Variables in Inline Assembler
29124 @section Output Variables in Inline Assembler
29127 The examples in this section, showing how to access the processor flags,
29128 illustrate how to specify the destination operands for assembly language
29131 @smallexample @c ada
29133 with Interfaces; use Interfaces;
29134 with Ada.Text_IO; use Ada.Text_IO;
29135 with System.Machine_Code; use System.Machine_Code;
29136 procedure Get_Flags is
29137 Flags : Unsigned_32;
29140 Asm ("pushfl" & LF & HT & -- push flags on stack
29141 "popl %%eax" & LF & HT & -- load eax with flags
29142 "movl %%eax, %0", -- store flags in variable
29143 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
29144 Put_Line ("Flags register:" & Flags'Img);
29149 In order to have a nicely aligned assembly listing, we have separated
29150 multiple assembler statements in the Asm template string with linefeed
29151 (ASCII.LF) and horizontal tab (ASCII.HT) characters.
29152 The resulting section of the assembly output file is:
29159 movl %eax, -40(%ebp)
29164 It would have been legal to write the Asm invocation as:
29167 Asm ("pushfl popl %%eax movl %%eax, %0")
29170 but in the generated assembler file, this would come out as:
29174 pushfl popl %eax movl %eax, -40(%ebp)
29178 which is not so convenient for the human reader.
29180 We use Ada comments
29181 at the end of each line to explain what the assembler instructions
29182 actually do. This is a useful convention.
29184 When writing Inline Assembler instructions, you need to precede each register
29185 and variable name with a percent sign. Since the assembler already requires
29186 a percent sign at the beginning of a register name, you need two consecutive
29187 percent signs for such names in the Asm template string, thus @code{%%eax}.
29188 In the generated assembly code, one of the percent signs will be stripped off.
29190 Names such as @code{%0}, @code{%1}, @code{%2}, etc., denote input or output
29191 variables: operands you later define using @code{Input} or @code{Output}
29192 parameters to @code{Asm}.
29193 An output variable is illustrated in
29194 the third statement in the Asm template string:
29198 The intent is to store the contents of the eax register in a variable that can
29199 be accessed in Ada. Simply writing @code{movl %%eax, Flags} would not
29200 necessarily work, since the compiler might optimize by using a register
29201 to hold Flags, and the expansion of the @code{movl} instruction would not be
29202 aware of this optimization. The solution is not to store the result directly
29203 but rather to advise the compiler to choose the correct operand form;
29204 that is the purpose of the @code{%0} output variable.
29206 Information about the output variable is supplied in the @code{Outputs}
29207 parameter to @code{Asm}:
29209 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
29212 The output is defined by the @code{Asm_Output} attribute of the target type;
29213 the general format is
29215 Type'Asm_Output (constraint_string, variable_name)
29218 The constraint string directs the compiler how
29219 to store/access the associated variable. In the example
29221 Unsigned_32'Asm_Output ("=m", Flags);
29223 the @code{"m"} (memory) constraint tells the compiler that the variable
29224 @code{Flags} should be stored in a memory variable, thus preventing
29225 the optimizer from keeping it in a register. In contrast,
29227 Unsigned_32'Asm_Output ("=r", Flags);
29229 uses the @code{"r"} (register) constraint, telling the compiler to
29230 store the variable in a register.
29232 If the constraint is preceded by the equal character (@strong{=}), it tells
29233 the compiler that the variable will be used to store data into it.
29235 In the @code{Get_Flags} example, we used the @code{"g"} (global) constraint,
29236 allowing the optimizer to choose whatever it deems best.
29238 There are a fairly large number of constraints, but the ones that are
29239 most useful (for the Intel x86 processor) are the following:
29245 global (i.e.@: can be stored anywhere)
29263 use one of eax, ebx, ecx or edx
29265 use one of eax, ebx, ecx, edx, esi or edi
29268 The full set of constraints is described in the gcc and @emph{as}
29269 documentation; note that it is possible to combine certain constraints
29270 in one constraint string.
29272 You specify the association of an output variable with an assembler operand
29273 through the @code{%}@emph{n} notation, where @emph{n} is a non-negative
29275 @smallexample @c ada
29277 Asm ("pushfl" & LF & HT & -- push flags on stack
29278 "popl %%eax" & LF & HT & -- load eax with flags
29279 "movl %%eax, %0", -- store flags in variable
29280 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
29284 @code{%0} will be replaced in the expanded code by the appropriate operand,
29286 the compiler decided for the @code{Flags} variable.
29288 In general, you may have any number of output variables:
29291 Count the operands starting at 0; thus @code{%0}, @code{%1}, etc.
29293 Specify the @code{Outputs} parameter as a parenthesized comma-separated list
29294 of @code{Asm_Output} attributes
29298 @smallexample @c ada
29300 Asm ("movl %%eax, %0" & LF & HT &
29301 "movl %%ebx, %1" & LF & HT &
29303 Outputs => (Unsigned_32'Asm_Output ("=g", Var_A), -- %0 = Var_A
29304 Unsigned_32'Asm_Output ("=g", Var_B), -- %1 = Var_B
29305 Unsigned_32'Asm_Output ("=g", Var_C))); -- %2 = Var_C
29309 where @code{Var_A}, @code{Var_B}, and @code{Var_C} are variables
29310 in the Ada program.
29312 As a variation on the @code{Get_Flags} example, we can use the constraints
29313 string to direct the compiler to store the eax register into the @code{Flags}
29314 variable, instead of including the store instruction explicitly in the
29315 @code{Asm} template string:
29317 @smallexample @c ada
29319 with Interfaces; use Interfaces;
29320 with Ada.Text_IO; use Ada.Text_IO;
29321 with System.Machine_Code; use System.Machine_Code;
29322 procedure Get_Flags_2 is
29323 Flags : Unsigned_32;
29326 Asm ("pushfl" & LF & HT & -- push flags on stack
29327 "popl %%eax", -- save flags in eax
29328 Outputs => Unsigned_32'Asm_Output ("=a", Flags));
29329 Put_Line ("Flags register:" & Flags'Img);
29335 The @code{"a"} constraint tells the compiler that the @code{Flags}
29336 variable will come from the eax register. Here is the resulting code:
29344 movl %eax,-40(%ebp)
29349 The compiler generated the store of eax into Flags after
29350 expanding the assembler code.
29352 Actually, there was no need to pop the flags into the eax register;
29353 more simply, we could just pop the flags directly into the program variable:
29355 @smallexample @c ada
29357 with Interfaces; use Interfaces;
29358 with Ada.Text_IO; use Ada.Text_IO;
29359 with System.Machine_Code; use System.Machine_Code;
29360 procedure Get_Flags_3 is
29361 Flags : Unsigned_32;
29364 Asm ("pushfl" & LF & HT & -- push flags on stack
29365 "pop %0", -- save flags in Flags
29366 Outputs => Unsigned_32'Asm_Output ("=g", Flags));
29367 Put_Line ("Flags register:" & Flags'Img);
29372 @c ---------------------------------------------------------------------------
29373 @node Input Variables in Inline Assembler
29374 @section Input Variables in Inline Assembler
29377 The example in this section illustrates how to specify the source operands
29378 for assembly language statements.
29379 The program simply increments its input value by 1:
29381 @smallexample @c ada
29383 with Interfaces; use Interfaces;
29384 with Ada.Text_IO; use Ada.Text_IO;
29385 with System.Machine_Code; use System.Machine_Code;
29386 procedure Increment is
29388 function Incr (Value : Unsigned_32) return Unsigned_32 is
29389 Result : Unsigned_32;
29392 Inputs => Unsigned_32'Asm_Input ("a", Value),
29393 Outputs => Unsigned_32'Asm_Output ("=a", Result));
29397 Value : Unsigned_32;
29401 Put_Line ("Value before is" & Value'Img);
29402 Value := Incr (Value);
29403 Put_Line ("Value after is" & Value'Img);
29408 The @code{Outputs} parameter to @code{Asm} specifies
29409 that the result will be in the eax register and that it is to be stored
29410 in the @code{Result} variable.
29412 The @code{Inputs} parameter looks much like the @code{Outputs} parameter,
29413 but with an @code{Asm_Input} attribute.
29414 The @code{"="} constraint, indicating an output value, is not present.
29416 You can have multiple input variables, in the same way that you can have more
29417 than one output variable.
29419 The parameter count (%0, %1) etc, now starts at the first input
29420 statement, and continues with the output statements.
29421 When both parameters use the same variable, the
29422 compiler will treat them as the same %n operand, which is the case here.
29424 Just as the @code{Outputs} parameter causes the register to be stored into the
29425 target variable after execution of the assembler statements, so does the
29426 @code{Inputs} parameter cause its variable to be loaded into the register
29427 before execution of the assembler statements.
29429 Thus the effect of the @code{Asm} invocation is:
29431 @item load the 32-bit value of @code{Value} into eax
29432 @item execute the @code{incl %eax} instruction
29433 @item store the contents of eax into the @code{Result} variable
29436 The resulting assembler file (with @option{-O2} optimization) contains:
29439 _increment__incr.1:
29452 @c ---------------------------------------------------------------------------
29453 @node Inlining Inline Assembler Code
29454 @section Inlining Inline Assembler Code
29457 For a short subprogram such as the @code{Incr} function in the previous
29458 section, the overhead of the call and return (creating / deleting the stack
29459 frame) can be significant, compared to the amount of code in the subprogram
29460 body. A solution is to apply Ada's @code{Inline} pragma to the subprogram,
29461 which directs the compiler to expand invocations of the subprogram at the
29462 point(s) of call, instead of setting up a stack frame for out-of-line calls.
29463 Here is the resulting program:
29465 @smallexample @c ada
29467 with Interfaces; use Interfaces;
29468 with Ada.Text_IO; use Ada.Text_IO;
29469 with System.Machine_Code; use System.Machine_Code;
29470 procedure Increment_2 is
29472 function Incr (Value : Unsigned_32) return Unsigned_32 is
29473 Result : Unsigned_32;
29476 Inputs => Unsigned_32'Asm_Input ("a", Value),
29477 Outputs => Unsigned_32'Asm_Output ("=a", Result));
29480 pragma Inline (Increment);
29482 Value : Unsigned_32;
29486 Put_Line ("Value before is" & Value'Img);
29487 Value := Increment (Value);
29488 Put_Line ("Value after is" & Value'Img);
29493 Compile the program with both optimization (@option{-O2}) and inlining
29494 (@option{-gnatn}) enabled.
29496 The @code{Incr} function is still compiled as usual, but at the
29497 point in @code{Increment} where our function used to be called:
29502 call _increment__incr.1
29507 the code for the function body directly appears:
29520 thus saving the overhead of stack frame setup and an out-of-line call.
29522 @c ---------------------------------------------------------------------------
29523 @node Other Asm Functionality
29524 @section Other @code{Asm} Functionality
29527 This section describes two important parameters to the @code{Asm}
29528 procedure: @code{Clobber}, which identifies register usage;
29529 and @code{Volatile}, which inhibits unwanted optimizations.
29532 * The Clobber Parameter::
29533 * The Volatile Parameter::
29536 @c ---------------------------------------------------------------------------
29537 @node The Clobber Parameter
29538 @subsection The @code{Clobber} Parameter
29541 One of the dangers of intermixing assembly language and a compiled language
29542 such as Ada is that the compiler needs to be aware of which registers are
29543 being used by the assembly code. In some cases, such as the earlier examples,
29544 the constraint string is sufficient to indicate register usage (e.g.,
29546 the eax register). But more generally, the compiler needs an explicit
29547 identification of the registers that are used by the Inline Assembly
29550 Using a register that the compiler doesn't know about
29551 could be a side effect of an instruction (like @code{mull}
29552 storing its result in both eax and edx).
29553 It can also arise from explicit register usage in your
29554 assembly code; for example:
29557 Asm ("movl %0, %%ebx" & LF & HT &
29559 Inputs => Unsigned_32'Asm_Input ("g", Var_In),
29560 Outputs => Unsigned_32'Asm_Output ("=g", Var_Out));
29564 where the compiler (since it does not analyze the @code{Asm} template string)
29565 does not know you are using the ebx register.
29567 In such cases you need to supply the @code{Clobber} parameter to @code{Asm},
29568 to identify the registers that will be used by your assembly code:
29572 Asm ("movl %0, %%ebx" & LF & HT &
29574 Inputs => Unsigned_32'Asm_Input ("g", Var_In),
29575 Outputs => Unsigned_32'Asm_Output ("=g", Var_Out),
29580 The Clobber parameter is a static string expression specifying the
29581 register(s) you are using. Note that register names are @emph{not} prefixed
29582 by a percent sign. Also, if more than one register is used then their names
29583 are separated by commas; e.g., @code{"eax, ebx"}
29585 The @code{Clobber} parameter has several additional uses:
29587 @item Use ``register'' name @code{cc} to indicate that flags might have changed
29588 @item Use ``register'' name @code{memory} if you changed a memory location
29591 @c ---------------------------------------------------------------------------
29592 @node The Volatile Parameter
29593 @subsection The @code{Volatile} Parameter
29594 @cindex Volatile parameter
29597 Compiler optimizations in the presence of Inline Assembler may sometimes have
29598 unwanted effects. For example, when an @code{Asm} invocation with an input
29599 variable is inside a loop, the compiler might move the loading of the input
29600 variable outside the loop, regarding it as a one-time initialization.
29602 If this effect is not desired, you can disable such optimizations by setting
29603 the @code{Volatile} parameter to @code{True}; for example:
29605 @smallexample @c ada
29607 Asm ("movl %0, %%ebx" & LF & HT &
29609 Inputs => Unsigned_32'Asm_Input ("g", Var_In),
29610 Outputs => Unsigned_32'Asm_Output ("=g", Var_Out),
29616 By default, @code{Volatile} is set to @code{False} unless there is no
29617 @code{Outputs} parameter.
29619 Although setting @code{Volatile} to @code{True} prevents unwanted
29620 optimizations, it will also disable other optimizations that might be
29621 important for efficiency. In general, you should set @code{Volatile}
29622 to @code{True} only if the compiler's optimizations have created
29624 @c END OF INLINE ASSEMBLER CHAPTER
29625 @c ===============================
29627 @c ***********************************
29628 @c * Compatibility and Porting Guide *
29629 @c ***********************************
29630 @node Compatibility and Porting Guide
29631 @appendix Compatibility and Porting Guide
29634 This chapter describes the compatibility issues that may arise between
29635 GNAT and other Ada compilation systems (including those for Ada 83),
29636 and shows how GNAT can expedite porting
29637 applications developed in other Ada environments.
29640 * Compatibility with Ada 83::
29641 * Compatibility between Ada 95 and Ada 2005::
29642 * Implementation-dependent characteristics::
29643 * Compatibility with Other Ada Systems::
29644 * Representation Clauses::
29646 @c Brief section is only in non-VMS version
29647 @c Full chapter is in VMS version
29648 * Compatibility with HP Ada 83::
29651 * Transitioning to 64-Bit GNAT for OpenVMS::
29655 @node Compatibility with Ada 83
29656 @section Compatibility with Ada 83
29657 @cindex Compatibility (between Ada 83 and Ada 95 / Ada 2005)
29660 Ada 95 and Ada 2005 are highly upwards compatible with Ada 83. In
29661 particular, the design intention was that the difficulties associated
29662 with moving from Ada 83 to Ada 95 or Ada 2005 should be no greater than those
29663 that occur when moving from one Ada 83 system to another.
29665 However, there are a number of points at which there are minor
29666 incompatibilities. The @cite{Ada 95 Annotated Reference Manual} contains
29667 full details of these issues,
29668 and should be consulted for a complete treatment.
29670 following subsections treat the most likely issues to be encountered.
29673 * Legal Ada 83 programs that are illegal in Ada 95::
29674 * More deterministic semantics::
29675 * Changed semantics::
29676 * Other language compatibility issues::
29679 @node Legal Ada 83 programs that are illegal in Ada 95
29680 @subsection Legal Ada 83 programs that are illegal in Ada 95
29682 Some legal Ada 83 programs are illegal (i.e., they will fail to compile) in
29683 Ada 95 and thus also in Ada 2005:
29686 @item Character literals
29687 Some uses of character literals are ambiguous. Since Ada 95 has introduced
29688 @code{Wide_Character} as a new predefined character type, some uses of
29689 character literals that were legal in Ada 83 are illegal in Ada 95.
29691 @smallexample @c ada
29692 for Char in 'A' .. 'Z' loop @dots{} end loop;
29696 The problem is that @code{'A'} and @code{'Z'} could be from either
29697 @code{Character} or @code{Wide_Character}. The simplest correction
29698 is to make the type explicit; e.g.:
29699 @smallexample @c ada
29700 for Char in Character range 'A' .. 'Z' loop @dots{} end loop;
29703 @item New reserved words
29704 The identifiers @code{abstract}, @code{aliased}, @code{protected},
29705 @code{requeue}, @code{tagged}, and @code{until} are reserved in Ada 95.
29706 Existing Ada 83 code using any of these identifiers must be edited to
29707 use some alternative name.
29709 @item Freezing rules
29710 The rules in Ada 95 are slightly different with regard to the point at
29711 which entities are frozen, and representation pragmas and clauses are
29712 not permitted past the freeze point. This shows up most typically in
29713 the form of an error message complaining that a representation item
29714 appears too late, and the appropriate corrective action is to move
29715 the item nearer to the declaration of the entity to which it refers.
29717 A particular case is that representation pragmas
29720 extended HP Ada 83 compatibility pragmas such as @code{Export_Procedure})
29722 cannot be applied to a subprogram body. If necessary, a separate subprogram
29723 declaration must be introduced to which the pragma can be applied.
29725 @item Optional bodies for library packages
29726 In Ada 83, a package that did not require a package body was nevertheless
29727 allowed to have one. This lead to certain surprises in compiling large
29728 systems (situations in which the body could be unexpectedly ignored by the
29729 binder). In Ada 95, if a package does not require a body then it is not
29730 permitted to have a body. To fix this problem, simply remove a redundant
29731 body if it is empty, or, if it is non-empty, introduce a dummy declaration
29732 into the spec that makes the body required. One approach is to add a private
29733 part to the package declaration (if necessary), and define a parameterless
29734 procedure called @code{Requires_Body}, which must then be given a dummy
29735 procedure body in the package body, which then becomes required.
29736 Another approach (assuming that this does not introduce elaboration
29737 circularities) is to add an @code{Elaborate_Body} pragma to the package spec,
29738 since one effect of this pragma is to require the presence of a package body.
29740 @item @code{Numeric_Error} is now the same as @code{Constraint_Error}
29741 In Ada 95, the exception @code{Numeric_Error} is a renaming of
29742 @code{Constraint_Error}.
29743 This means that it is illegal to have separate exception handlers for
29744 the two exceptions. The fix is simply to remove the handler for the
29745 @code{Numeric_Error} case (since even in Ada 83, a compiler was free to raise
29746 @code{Constraint_Error} in place of @code{Numeric_Error} in all cases).
29748 @item Indefinite subtypes in generics
29749 In Ada 83, it was permissible to pass an indefinite type (e.g.@: @code{String})
29750 as the actual for a generic formal private type, but then the instantiation
29751 would be illegal if there were any instances of declarations of variables
29752 of this type in the generic body. In Ada 95, to avoid this clear violation
29753 of the methodological principle known as the ``contract model'',
29754 the generic declaration explicitly indicates whether
29755 or not such instantiations are permitted. If a generic formal parameter
29756 has explicit unknown discriminants, indicated by using @code{(<>)} after the
29757 type name, then it can be instantiated with indefinite types, but no
29758 stand-alone variables can be declared of this type. Any attempt to declare
29759 such a variable will result in an illegality at the time the generic is
29760 declared. If the @code{(<>)} notation is not used, then it is illegal
29761 to instantiate the generic with an indefinite type.
29762 This is the potential incompatibility issue when porting Ada 83 code to Ada 95.
29763 It will show up as a compile time error, and
29764 the fix is usually simply to add the @code{(<>)} to the generic declaration.
29767 @node More deterministic semantics
29768 @subsection More deterministic semantics
29772 Conversions from real types to integer types round away from 0. In Ada 83
29773 the conversion Integer(2.5) could deliver either 2 or 3 as its value. This
29774 implementation freedom was intended to support unbiased rounding in
29775 statistical applications, but in practice it interfered with portability.
29776 In Ada 95 the conversion semantics are unambiguous, and rounding away from 0
29777 is required. Numeric code may be affected by this change in semantics.
29778 Note, though, that this issue is no worse than already existed in Ada 83
29779 when porting code from one vendor to another.
29782 The Real-Time Annex introduces a set of policies that define the behavior of
29783 features that were implementation dependent in Ada 83, such as the order in
29784 which open select branches are executed.
29787 @node Changed semantics
29788 @subsection Changed semantics
29791 The worst kind of incompatibility is one where a program that is legal in
29792 Ada 83 is also legal in Ada 95 but can have an effect in Ada 95 that was not
29793 possible in Ada 83. Fortunately this is extremely rare, but the one
29794 situation that you should be alert to is the change in the predefined type
29795 @code{Character} from 7-bit ASCII to 8-bit Latin-1.
29798 @item Range of type @code{Character}
29799 The range of @code{Standard.Character} is now the full 256 characters
29800 of Latin-1, whereas in most Ada 83 implementations it was restricted
29801 to 128 characters. Although some of the effects of
29802 this change will be manifest in compile-time rejection of legal
29803 Ada 83 programs it is possible for a working Ada 83 program to have
29804 a different effect in Ada 95, one that was not permitted in Ada 83.
29805 As an example, the expression
29806 @code{Character'Pos(Character'Last)} returned @code{127} in Ada 83 and now
29807 delivers @code{255} as its value.
29808 In general, you should look at the logic of any
29809 character-processing Ada 83 program and see whether it needs to be adapted
29810 to work correctly with Latin-1. Note that the predefined Ada 95 API has a
29811 character handling package that may be relevant if code needs to be adapted
29812 to account for the additional Latin-1 elements.
29813 The desirable fix is to
29814 modify the program to accommodate the full character set, but in some cases
29815 it may be convenient to define a subtype or derived type of Character that
29816 covers only the restricted range.
29820 @node Other language compatibility issues
29821 @subsection Other language compatibility issues
29824 @item @option{-gnat83} switch
29825 All implementations of GNAT provide a switch that causes GNAT to operate
29826 in Ada 83 mode. In this mode, some but not all compatibility problems
29827 of the type described above are handled automatically. For example, the
29828 new reserved words introduced in Ada 95 and Ada 2005 are treated simply
29829 as identifiers as in Ada 83.
29831 in practice, it is usually advisable to make the necessary modifications
29832 to the program to remove the need for using this switch.
29833 See @ref{Compiling Different Versions of Ada}.
29835 @item Support for removed Ada 83 pragmas and attributes
29836 A number of pragmas and attributes from Ada 83 were removed from Ada 95,
29837 generally because they were replaced by other mechanisms. Ada 95 and Ada 2005
29838 compilers are allowed, but not required, to implement these missing
29839 elements. In contrast with some other compilers, GNAT implements all
29840 such pragmas and attributes, eliminating this compatibility concern. These
29841 include @code{pragma Interface} and the floating point type attributes
29842 (@code{Emax}, @code{Mantissa}, etc.), among other items.
29846 @node Compatibility between Ada 95 and Ada 2005
29847 @section Compatibility between Ada 95 and Ada 2005
29848 @cindex Compatibility between Ada 95 and Ada 2005
29851 Although Ada 2005 was designed to be upwards compatible with Ada 95, there are
29852 a number of incompatibilities. Several are enumerated below;
29853 for a complete description please see the
29854 Annotated Ada 2005 Reference Manual, or section 9.1.1 in
29855 @cite{Rationale for Ada 2005}.
29858 @item New reserved words.
29859 The words @code{interface}, @code{overriding} and @code{synchronized} are
29860 reserved in Ada 2005.
29861 A pre-Ada 2005 program that uses any of these as an identifier will be
29864 @item New declarations in predefined packages.
29865 A number of packages in the predefined environment contain new declarations:
29866 @code{Ada.Exceptions}, @code{Ada.Real_Time}, @code{Ada.Strings},
29867 @code{Ada.Strings.Fixed}, @code{Ada.Strings.Bounded},
29868 @code{Ada.Strings.Unbounded}, @code{Ada.Strings.Wide_Fixed},
29869 @code{Ada.Strings.Wide_Bounded}, @code{Ada.Strings.Wide_Unbounded},
29870 @code{Ada.Tags}, @code{Ada.Text_IO}, and @code{Interfaces.C}.
29871 If an Ada 95 program does a @code{with} and @code{use} of any of these
29872 packages, the new declarations may cause name clashes.
29874 @item Access parameters.
29875 A nondispatching subprogram with an access parameter cannot be renamed
29876 as a dispatching operation. This was permitted in Ada 95.
29878 @item Access types, discriminants, and constraints.
29879 Rule changes in this area have led to some incompatibilities; for example,
29880 constrained subtypes of some access types are not permitted in Ada 2005.
29882 @item Aggregates for limited types.
29883 The allowance of aggregates for limited types in Ada 2005 raises the
29884 possibility of ambiguities in legal Ada 95 programs, since additional types
29885 now need to be considered in expression resolution.
29887 @item Fixed-point multiplication and division.
29888 Certain expressions involving ``*'' or ``/'' for a fixed-point type, which
29889 were legal in Ada 95 and invoked the predefined versions of these operations,
29891 The ambiguity may be resolved either by applying a type conversion to the
29892 expression, or by explicitly invoking the operation from package
29895 @item Return-by-reference types.
29896 The Ada 95 return-by-reference mechanism has been removed. Instead, the user
29897 can declare a function returning a value from an anonymous access type.
29901 @node Implementation-dependent characteristics
29902 @section Implementation-dependent characteristics
29904 Although the Ada language defines the semantics of each construct as
29905 precisely as practical, in some situations (for example for reasons of
29906 efficiency, or where the effect is heavily dependent on the host or target
29907 platform) the implementation is allowed some freedom. In porting Ada 83
29908 code to GNAT, you need to be aware of whether / how the existing code
29909 exercised such implementation dependencies. Such characteristics fall into
29910 several categories, and GNAT offers specific support in assisting the
29911 transition from certain Ada 83 compilers.
29914 * Implementation-defined pragmas::
29915 * Implementation-defined attributes::
29917 * Elaboration order::
29918 * Target-specific aspects::
29921 @node Implementation-defined pragmas
29922 @subsection Implementation-defined pragmas
29925 Ada compilers are allowed to supplement the language-defined pragmas, and
29926 these are a potential source of non-portability. All GNAT-defined pragmas
29927 are described in @ref{Implementation Defined Pragmas,,, gnat_rm, GNAT
29928 Reference Manual}, and these include several that are specifically
29929 intended to correspond to other vendors' Ada 83 pragmas.
29930 For migrating from VADS, the pragma @code{Use_VADS_Size} may be useful.
29931 For compatibility with HP Ada 83, GNAT supplies the pragmas
29932 @code{Extend_System}, @code{Ident}, @code{Inline_Generic},
29933 @code{Interface_Name}, @code{Passive}, @code{Suppress_All},
29934 and @code{Volatile}.
29935 Other relevant pragmas include @code{External} and @code{Link_With}.
29936 Some vendor-specific
29937 Ada 83 pragmas (@code{Share_Generic}, @code{Subtitle}, and @code{Title}) are
29939 avoiding compiler rejection of units that contain such pragmas; they are not
29940 relevant in a GNAT context and hence are not otherwise implemented.
29942 @node Implementation-defined attributes
29943 @subsection Implementation-defined attributes
29945 Analogous to pragmas, the set of attributes may be extended by an
29946 implementation. All GNAT-defined attributes are described in
29947 @ref{Implementation Defined Attributes,,, gnat_rm, GNAT Reference
29948 Manual}, and these include several that are specifically intended
29949 to correspond to other vendors' Ada 83 attributes. For migrating from VADS,
29950 the attribute @code{VADS_Size} may be useful. For compatibility with HP
29951 Ada 83, GNAT supplies the attributes @code{Bit}, @code{Machine_Size} and
29955 @subsection Libraries
29957 Vendors may supply libraries to supplement the standard Ada API. If Ada 83
29958 code uses vendor-specific libraries then there are several ways to manage
29959 this in Ada 95 or Ada 2005:
29962 If the source code for the libraries (specs and bodies) are
29963 available, then the libraries can be migrated in the same way as the
29966 If the source code for the specs but not the bodies are
29967 available, then you can reimplement the bodies.
29969 Some features introduced by Ada 95 obviate the need for library support. For
29970 example most Ada 83 vendors supplied a package for unsigned integers. The
29971 Ada 95 modular type feature is the preferred way to handle this need, so
29972 instead of migrating or reimplementing the unsigned integer package it may
29973 be preferable to retrofit the application using modular types.
29976 @node Elaboration order
29977 @subsection Elaboration order
29979 The implementation can choose any elaboration order consistent with the unit
29980 dependency relationship. This freedom means that some orders can result in
29981 Program_Error being raised due to an ``Access Before Elaboration'': an attempt
29982 to invoke a subprogram its body has been elaborated, or to instantiate a
29983 generic before the generic body has been elaborated. By default GNAT
29984 attempts to choose a safe order (one that will not encounter access before
29985 elaboration problems) by implicitly inserting @code{Elaborate} or
29986 @code{Elaborate_All} pragmas where
29987 needed. However, this can lead to the creation of elaboration circularities
29988 and a resulting rejection of the program by gnatbind. This issue is
29989 thoroughly described in @ref{Elaboration Order Handling in GNAT}.
29990 In brief, there are several
29991 ways to deal with this situation:
29995 Modify the program to eliminate the circularities, e.g.@: by moving
29996 elaboration-time code into explicitly-invoked procedures
29998 Constrain the elaboration order by including explicit @code{Elaborate_Body} or
29999 @code{Elaborate} pragmas, and then inhibit the generation of implicit
30000 @code{Elaborate_All}
30001 pragmas either globally (as an effect of the @option{-gnatE} switch) or locally
30002 (by selectively suppressing elaboration checks via pragma
30003 @code{Suppress(Elaboration_Check)} when it is safe to do so).
30006 @node Target-specific aspects
30007 @subsection Target-specific aspects
30009 Low-level applications need to deal with machine addresses, data
30010 representations, interfacing with assembler code, and similar issues. If
30011 such an Ada 83 application is being ported to different target hardware (for
30012 example where the byte endianness has changed) then you will need to
30013 carefully examine the program logic; the porting effort will heavily depend
30014 on the robustness of the original design. Moreover, Ada 95 (and thus
30015 Ada 2005) are sometimes
30016 incompatible with typical Ada 83 compiler practices regarding implicit
30017 packing, the meaning of the Size attribute, and the size of access values.
30018 GNAT's approach to these issues is described in @ref{Representation Clauses}.
30020 @node Compatibility with Other Ada Systems
30021 @section Compatibility with Other Ada Systems
30024 If programs avoid the use of implementation dependent and
30025 implementation defined features, as documented in the @cite{Ada
30026 Reference Manual}, there should be a high degree of portability between
30027 GNAT and other Ada systems. The following are specific items which
30028 have proved troublesome in moving Ada 95 programs from GNAT to other Ada 95
30029 compilers, but do not affect porting code to GNAT@.
30030 (As of @value{NOW}, GNAT is the only compiler available for Ada 2005;
30031 the following issues may or may not arise for Ada 2005 programs
30032 when other compilers appear.)
30035 @item Ada 83 Pragmas and Attributes
30036 Ada 95 compilers are allowed, but not required, to implement the missing
30037 Ada 83 pragmas and attributes that are no longer defined in Ada 95.
30038 GNAT implements all such pragmas and attributes, eliminating this as
30039 a compatibility concern, but some other Ada 95 compilers reject these
30040 pragmas and attributes.
30042 @item Specialized Needs Annexes
30043 GNAT implements the full set of special needs annexes. At the
30044 current time, it is the only Ada 95 compiler to do so. This means that
30045 programs making use of these features may not be portable to other Ada
30046 95 compilation systems.
30048 @item Representation Clauses
30049 Some other Ada 95 compilers implement only the minimal set of
30050 representation clauses required by the Ada 95 reference manual. GNAT goes
30051 far beyond this minimal set, as described in the next section.
30054 @node Representation Clauses
30055 @section Representation Clauses
30058 The Ada 83 reference manual was quite vague in describing both the minimal
30059 required implementation of representation clauses, and also their precise
30060 effects. Ada 95 (and thus also Ada 2005) are much more explicit, but the
30061 minimal set of capabilities required is still quite limited.
30063 GNAT implements the full required set of capabilities in
30064 Ada 95 and Ada 2005, but also goes much further, and in particular
30065 an effort has been made to be compatible with existing Ada 83 usage to the
30066 greatest extent possible.
30068 A few cases exist in which Ada 83 compiler behavior is incompatible with
30069 the requirements in Ada 95 (and thus also Ada 2005). These are instances of
30070 intentional or accidental dependence on specific implementation dependent
30071 characteristics of these Ada 83 compilers. The following is a list of
30072 the cases most likely to arise in existing Ada 83 code.
30075 @item Implicit Packing
30076 Some Ada 83 compilers allowed a Size specification to cause implicit
30077 packing of an array or record. This could cause expensive implicit
30078 conversions for change of representation in the presence of derived
30079 types, and the Ada design intends to avoid this possibility.
30080 Subsequent AI's were issued to make it clear that such implicit
30081 change of representation in response to a Size clause is inadvisable,
30082 and this recommendation is represented explicitly in the Ada 95 (and Ada 2005)
30083 Reference Manuals as implementation advice that is followed by GNAT@.
30084 The problem will show up as an error
30085 message rejecting the size clause. The fix is simply to provide
30086 the explicit pragma @code{Pack}, or for more fine tuned control, provide
30087 a Component_Size clause.
30089 @item Meaning of Size Attribute
30090 The Size attribute in Ada 95 (and Ada 2005) for discrete types is defined as
30091 the minimal number of bits required to hold values of the type. For example,
30092 on a 32-bit machine, the size of @code{Natural} will typically be 31 and not
30093 32 (since no sign bit is required). Some Ada 83 compilers gave 31, and
30094 some 32 in this situation. This problem will usually show up as a compile
30095 time error, but not always. It is a good idea to check all uses of the
30096 'Size attribute when porting Ada 83 code. The GNAT specific attribute
30097 Object_Size can provide a useful way of duplicating the behavior of
30098 some Ada 83 compiler systems.
30100 @item Size of Access Types
30101 A common assumption in Ada 83 code is that an access type is in fact a pointer,
30102 and that therefore it will be the same size as a System.Address value. This
30103 assumption is true for GNAT in most cases with one exception. For the case of
30104 a pointer to an unconstrained array type (where the bounds may vary from one
30105 value of the access type to another), the default is to use a ``fat pointer'',
30106 which is represented as two separate pointers, one to the bounds, and one to
30107 the array. This representation has a number of advantages, including improved
30108 efficiency. However, it may cause some difficulties in porting existing Ada 83
30109 code which makes the assumption that, for example, pointers fit in 32 bits on
30110 a machine with 32-bit addressing.
30112 To get around this problem, GNAT also permits the use of ``thin pointers'' for
30113 access types in this case (where the designated type is an unconstrained array
30114 type). These thin pointers are indeed the same size as a System.Address value.
30115 To specify a thin pointer, use a size clause for the type, for example:
30117 @smallexample @c ada
30118 type X is access all String;
30119 for X'Size use Standard'Address_Size;
30123 which will cause the type X to be represented using a single pointer.
30124 When using this representation, the bounds are right behind the array.
30125 This representation is slightly less efficient, and does not allow quite
30126 such flexibility in the use of foreign pointers or in using the
30127 Unrestricted_Access attribute to create pointers to non-aliased objects.
30128 But for any standard portable use of the access type it will work in
30129 a functionally correct manner and allow porting of existing code.
30130 Note that another way of forcing a thin pointer representation
30131 is to use a component size clause for the element size in an array,
30132 or a record representation clause for an access field in a record.
30136 @c This brief section is only in the non-VMS version
30137 @c The complete chapter on HP Ada is in the VMS version
30138 @node Compatibility with HP Ada 83
30139 @section Compatibility with HP Ada 83
30142 The VMS version of GNAT fully implements all the pragmas and attributes
30143 provided by HP Ada 83, as well as providing the standard HP Ada 83
30144 libraries, including Starlet. In addition, data layouts and parameter
30145 passing conventions are highly compatible. This means that porting
30146 existing HP Ada 83 code to GNAT in VMS systems should be easier than
30147 most other porting efforts. The following are some of the most
30148 significant differences between GNAT and HP Ada 83.
30151 @item Default floating-point representation
30152 In GNAT, the default floating-point format is IEEE, whereas in HP Ada 83,
30153 it is VMS format. GNAT does implement the necessary pragmas
30154 (Long_Float, Float_Representation) for changing this default.
30157 The package System in GNAT exactly corresponds to the definition in the
30158 Ada 95 reference manual, which means that it excludes many of the
30159 HP Ada 83 extensions. However, a separate package Aux_DEC is provided
30160 that contains the additional definitions, and a special pragma,
30161 Extend_System allows this package to be treated transparently as an
30162 extension of package System.
30165 The definitions provided by Aux_DEC are exactly compatible with those
30166 in the HP Ada 83 version of System, with one exception.
30167 HP Ada provides the following declarations:
30169 @smallexample @c ada
30170 TO_ADDRESS (INTEGER)
30171 TO_ADDRESS (UNSIGNED_LONGWORD)
30172 TO_ADDRESS (@i{universal_integer})
30176 The version of TO_ADDRESS taking a @i{universal integer} argument is in fact
30177 an extension to Ada 83 not strictly compatible with the reference manual.
30178 In GNAT, we are constrained to be exactly compatible with the standard,
30179 and this means we cannot provide this capability. In HP Ada 83, the
30180 point of this definition is to deal with a call like:
30182 @smallexample @c ada
30183 TO_ADDRESS (16#12777#);
30187 Normally, according to the Ada 83 standard, one would expect this to be
30188 ambiguous, since it matches both the INTEGER and UNSIGNED_LONGWORD forms
30189 of TO_ADDRESS@. However, in HP Ada 83, there is no ambiguity, since the
30190 definition using @i{universal_integer} takes precedence.
30192 In GNAT, since the version with @i{universal_integer} cannot be supplied, it
30193 is not possible to be 100% compatible. Since there are many programs using
30194 numeric constants for the argument to TO_ADDRESS, the decision in GNAT was
30195 to change the name of the function in the UNSIGNED_LONGWORD case, so the
30196 declarations provided in the GNAT version of AUX_Dec are:
30198 @smallexample @c ada
30199 function To_Address (X : Integer) return Address;
30200 pragma Pure_Function (To_Address);
30202 function To_Address_Long (X : Unsigned_Longword)
30204 pragma Pure_Function (To_Address_Long);
30208 This means that programs using TO_ADDRESS for UNSIGNED_LONGWORD must
30209 change the name to TO_ADDRESS_LONG@.
30211 @item Task_Id values
30212 The Task_Id values assigned will be different in the two systems, and GNAT
30213 does not provide a specified value for the Task_Id of the environment task,
30214 which in GNAT is treated like any other declared task.
30218 For full details on these and other less significant compatibility issues,
30219 see appendix E of the HP publication entitled @cite{HP Ada, Technical
30220 Overview and Comparison on HP Platforms}.
30222 For GNAT running on other than VMS systems, all the HP Ada 83 pragmas and
30223 attributes are recognized, although only a subset of them can sensibly
30224 be implemented. The description of pragmas in @ref{Implementation
30225 Defined Pragmas,,, gnat_rm, GNAT Reference Manual}
30226 indicates whether or not they are applicable to non-VMS systems.
30230 @node Transitioning to 64-Bit GNAT for OpenVMS
30231 @section Transitioning to 64-Bit @value{EDITION} for OpenVMS
30234 This section is meant to assist users of pre-2006 @value{EDITION}
30235 for Alpha OpenVMS who are transitioning to 64-bit @value{EDITION},
30236 the version of the GNAT technology supplied in 2006 and later for
30237 OpenVMS on both Alpha and I64.
30240 * Introduction to transitioning::
30241 * Migration of 32 bit code::
30242 * Taking advantage of 64 bit addressing::
30243 * Technical details::
30246 @node Introduction to transitioning
30247 @subsection Introduction
30250 64-bit @value{EDITION} for Open VMS has been designed to meet
30255 Providing a full conforming implementation of Ada 95 and Ada 2005
30258 Allowing maximum backward compatibility, thus easing migration of existing
30262 Supplying a path for exploiting the full 64-bit address range
30266 Ada's strong typing semantics has made it
30267 impractical to have different 32-bit and 64-bit modes. As soon as
30268 one object could possibly be outside the 32-bit address space, this
30269 would make it necessary for the @code{System.Address} type to be 64 bits.
30270 In particular, this would cause inconsistencies if 32-bit code is
30271 called from 64-bit code that raises an exception.
30273 This issue has been resolved by always using 64-bit addressing
30274 at the system level, but allowing for automatic conversions between
30275 32-bit and 64-bit addresses where required. Thus users who
30276 do not currently require 64-bit addressing capabilities, can
30277 recompile their code with only minimal changes (and indeed
30278 if the code is written in portable Ada, with no assumptions about
30279 the size of the @code{Address} type, then no changes at all are necessary).
30281 this approach provides a simple, gradual upgrade path to future
30282 use of larger memories than available for 32-bit systems.
30283 Also, newly written applications or libraries will by default
30284 be fully compatible with future systems exploiting 64-bit
30285 addressing capabilities.
30287 @ref{Migration of 32 bit code}, will focus on porting applications
30288 that do not require more than 2 GB of
30289 addressable memory. This code will be referred to as
30290 @emph{32-bit code}.
30291 For applications intending to exploit the full 64-bit address space,
30292 @ref{Taking advantage of 64 bit addressing},
30293 will consider further changes that may be required.
30294 Such code will be referred to below as @emph{64-bit code}.
30296 @node Migration of 32 bit code
30297 @subsection Migration of 32-bit code
30302 * Unchecked conversions::
30303 * Predefined constants::
30304 * Interfacing with C::
30305 * Experience with source compatibility::
30308 @node Address types
30309 @subsubsection Address types
30312 To solve the problem of mixing 64-bit and 32-bit addressing,
30313 while maintaining maximum backward compatibility, the following
30314 approach has been taken:
30318 @code{System.Address} always has a size of 64 bits
30321 @code{System.Short_Address} is a 32-bit subtype of @code{System.Address}
30325 Since @code{System.Short_Address} is a subtype of @code{System.Address},
30326 a @code{Short_Address}
30327 may be used where an @code{Address} is required, and vice versa, without
30328 needing explicit type conversions.
30329 By virtue of the Open VMS parameter passing conventions,
30331 and exported subprograms that have 32-bit address parameters are
30332 compatible with those that have 64-bit address parameters.
30333 (See @ref{Making code 64 bit clean} for details.)
30335 The areas that may need attention are those where record types have
30336 been defined that contain components of the type @code{System.Address}, and
30337 where objects of this type are passed to code expecting a record layout with
30340 Different compilers on different platforms cannot be
30341 expected to represent the same type in the same way,
30342 since alignment constraints
30343 and other system-dependent properties affect the compiler's decision.
30344 For that reason, Ada code
30345 generally uses representation clauses to specify the expected
30346 layout where required.
30348 If such a representation clause uses 32 bits for a component having
30349 the type @code{System.Address}, 64-bit @value{EDITION} for OpenVMS
30350 will detect that error and produce a specific diagnostic message.
30351 The developer should then determine whether the representation
30352 should be 64 bits or not and make either of two changes:
30353 change the size to 64 bits and leave the type as @code{System.Address}, or
30354 leave the size as 32 bits and change the type to @code{System.Short_Address}.
30355 Since @code{Short_Address} is a subtype of @code{Address}, no changes are
30356 required in any code setting or accessing the field; the compiler will
30357 automatically perform any needed conversions between address
30361 @subsubsection Access types
30364 By default, objects designated by access values are always
30365 allocated in the 32-bit
30366 address space. Thus legacy code will never contain
30367 any objects that are not addressable with 32-bit addresses, and
30368 the compiler will never raise exceptions as result of mixing
30369 32-bit and 64-bit addresses.
30371 However, the access values themselves are represented in 64 bits, for optimum
30372 performance and future compatibility with 64-bit code. As was
30373 the case with @code{System.Address}, the compiler will give an error message
30374 if an object or record component has a representation clause that
30375 requires the access value to fit in 32 bits. In such a situation,
30376 an explicit size clause for the access type, specifying 32 bits,
30377 will have the desired effect.
30379 General access types (declared with @code{access all}) can never be
30380 32 bits, as values of such types must be able to refer to any object
30381 of the designated type,
30382 including objects residing outside the 32-bit address range.
30383 Existing Ada 83 code will not contain such type definitions,
30384 however, since general access types were introduced in Ada 95.
30386 @node Unchecked conversions
30387 @subsubsection Unchecked conversions
30390 In the case of an @code{Unchecked_Conversion} where the source type is a
30391 64-bit access type or the type @code{System.Address}, and the target
30392 type is a 32-bit type, the compiler will generate a warning.
30393 Even though the generated code will still perform the required
30394 conversions, it is highly recommended in these cases to use
30395 respectively a 32-bit access type or @code{System.Short_Address}
30396 as the source type.
30398 @node Predefined constants
30399 @subsubsection Predefined constants
30402 The following table shows the correspondence between pre-2006 versions of
30403 @value{EDITION} on Alpha OpenVMS (``Old'') and 64-bit @value{EDITION}
30406 @multitable {@code{System.Short_Memory_Size}} {2**32} {2**64}
30407 @item @b{Constant} @tab @b{Old} @tab @b{New}
30408 @item @code{System.Word_Size} @tab 32 @tab 64
30409 @item @code{System.Memory_Size} @tab 2**32 @tab 2**64
30410 @item @code{System.Short_Memory_Size} @tab 2**32 @tab 2**32
30411 @item @code{System.Address_Size} @tab 32 @tab 64
30415 If you need to refer to the specific
30416 memory size of a 32-bit implementation, instead of the
30417 actual memory size, use @code{System.Short_Memory_Size}
30418 rather than @code{System.Memory_Size}.
30419 Similarly, references to @code{System.Address_Size} may need
30420 to be replaced by @code{System.Short_Address'Size}.
30421 The program @command{gnatfind} may be useful for locating
30422 references to the above constants, so that you can verify that they
30425 @node Interfacing with C
30426 @subsubsection Interfacing with C
30429 In order to minimize the impact of the transition to 64-bit addresses on
30430 legacy programs, some fundamental types in the @code{Interfaces.C}
30431 package hierarchy continue to be represented in 32 bits.
30432 These types are: @code{ptrdiff_t}, @code{size_t}, and @code{chars_ptr}.
30433 This eases integration with the default HP C layout choices, for example
30434 as found in the system routines in @code{DECC$SHR.EXE}.
30435 Because of this implementation choice, the type fully compatible with
30436 @code{chars_ptr} is now @code{Short_Address} and not @code{Address}.
30437 Depending on the context the compiler will issue a
30438 warning or an error when type @code{Address} is used, alerting the user to a
30439 potential problem. Otherwise 32-bit programs that use
30440 @code{Interfaces.C} should normally not require code modifications
30442 The other issue arising with C interfacing concerns pragma @code{Convention}.
30443 For VMS 64-bit systems, there is an issue of the appropriate default size
30444 of C convention pointers in the absence of an explicit size clause. The HP
30445 C compiler can choose either 32 or 64 bits depending on compiler options.
30446 GNAT chooses 32-bits rather than 64-bits in the default case where no size
30447 clause is given. This proves a better choice for porting 32-bit legacy
30448 applications. In order to have a 64-bit representation, it is necessary to
30449 specify a size representation clause. For example:
30451 @smallexample @c ada
30452 type int_star is access Interfaces.C.int;
30453 pragma Convention(C, int_star);
30454 for int_star'Size use 64; -- Necessary to get 64 and not 32 bits
30457 @node Experience with source compatibility
30458 @subsubsection Experience with source compatibility
30461 The Security Server and STARLET on I64 provide an interesting ``test case''
30462 for source compatibility issues, since it is in such system code
30463 where assumptions about @code{Address} size might be expected to occur.
30464 Indeed, there were a small number of occasions in the Security Server
30465 file @file{jibdef.ads}
30466 where a representation clause for a record type specified
30467 32 bits for a component of type @code{Address}.
30468 All of these errors were detected by the compiler.
30469 The repair was obvious and immediate; to simply replace @code{Address} by
30470 @code{Short_Address}.
30472 In the case of STARLET, there were several record types that should
30473 have had representation clauses but did not. In these record types
30474 there was an implicit assumption that an @code{Address} value occupied
30476 These compiled without error, but their usage resulted in run-time error
30477 returns from STARLET system calls.
30478 Future GNAT technology enhancements may include a tool that detects and flags
30479 these sorts of potential source code porting problems.
30481 @c ****************************************
30482 @node Taking advantage of 64 bit addressing
30483 @subsection Taking advantage of 64-bit addressing
30486 * Making code 64 bit clean::
30487 * Allocating memory from the 64 bit storage pool::
30488 * Restrictions on use of 64 bit objects::
30489 * Using 64 bit storage pools by default::
30490 * General access types::
30491 * STARLET and other predefined libraries::
30494 @node Making code 64 bit clean
30495 @subsubsection Making code 64-bit clean
30498 In order to prevent problems that may occur when (parts of) a
30499 system start using memory outside the 32-bit address range,
30500 we recommend some additional guidelines:
30504 For imported subprograms that take parameters of the
30505 type @code{System.Address}, ensure that these subprograms can
30506 indeed handle 64-bit addresses. If not, or when in doubt,
30507 change the subprogram declaration to specify
30508 @code{System.Short_Address} instead.
30511 Resolve all warnings related to size mismatches in
30512 unchecked conversions. Failing to do so causes
30513 erroneous execution if the source object is outside
30514 the 32-bit address space.
30517 (optional) Explicitly use the 32-bit storage pool
30518 for access types used in a 32-bit context, or use
30519 generic access types where possible
30520 (@pxref{Restrictions on use of 64 bit objects}).
30524 If these rules are followed, the compiler will automatically insert
30525 any necessary checks to ensure that no addresses or access values
30526 passed to 32-bit code ever refer to objects outside the 32-bit
30528 Any attempt to do this will raise @code{Constraint_Error}.
30530 @node Allocating memory from the 64 bit storage pool
30531 @subsubsection Allocating memory from the 64-bit storage pool
30534 For any access type @code{T} that potentially requires memory allocations
30535 beyond the 32-bit address space,
30536 use the following representation clause:
30538 @smallexample @c ada
30539 for T'Storage_Pool use System.Pool_64;
30542 @node Restrictions on use of 64 bit objects
30543 @subsubsection Restrictions on use of 64-bit objects
30546 Taking the address of an object allocated from a 64-bit storage pool,
30547 and then passing this address to a subprogram expecting
30548 @code{System.Short_Address},
30549 or assigning it to a variable of type @code{Short_Address}, will cause
30550 @code{Constraint_Error} to be raised. In case the code is not 64-bit clean
30551 (@pxref{Making code 64 bit clean}), or checks are suppressed,
30552 no exception is raised and execution
30553 will become erroneous.
30555 @node Using 64 bit storage pools by default
30556 @subsubsection Using 64-bit storage pools by default
30559 In some cases it may be desirable to have the compiler allocate
30560 from 64-bit storage pools by default. This may be the case for
30561 libraries that are 64-bit clean, but may be used in both 32-bit
30562 and 64-bit contexts. For these cases the following configuration
30563 pragma may be specified:
30565 @smallexample @c ada
30566 pragma Pool_64_Default;
30570 Any code compiled in the context of this pragma will by default
30571 use the @code{System.Pool_64} storage pool. This default may be overridden
30572 for a specific access type @code{T} by the representation clause:
30574 @smallexample @c ada
30575 for T'Storage_Pool use System.Pool_32;
30579 Any object whose address may be passed to a subprogram with a
30580 @code{Short_Address} argument, or assigned to a variable of type
30581 @code{Short_Address}, needs to be allocated from this pool.
30583 @node General access types
30584 @subsubsection General access types
30587 Objects designated by access values from a
30588 general access type (declared with @code{access all}) are never allocated
30589 from a 64-bit storage pool. Code that uses general access types will
30590 accept objects allocated in either 32-bit or 64-bit address spaces,
30591 but never allocate objects outside the 32-bit address space.
30592 Using general access types ensures maximum compatibility with both
30593 32-bit and 64-bit code.
30595 @node STARLET and other predefined libraries
30596 @subsubsection STARLET and other predefined libraries
30599 All code that comes as part of GNAT is 64-bit clean, but the
30600 restrictions given in @ref{Restrictions on use of 64 bit objects},
30601 still apply. Look at the package
30602 specs to see in which contexts objects allocated
30603 in 64-bit address space are acceptable.
30605 @node Technical details
30606 @subsection Technical details
30609 64-bit @value{EDITION} for Open VMS takes advantage of the freedom given in the
30610 Ada standard with respect to the type of @code{System.Address}. Previous
30611 versions of GNAT Pro have defined this type as private and implemented it as a
30614 In order to allow defining @code{System.Short_Address} as a proper subtype,
30615 and to match the implicit sign extension in parameter passing,
30616 in 64-bit @value{EDITION} for Open VMS, @code{System.Address} is defined as a
30617 visible (i.e., non-private) integer type.
30618 Standard operations on the type, such as the binary operators ``+'', ``-'',
30619 etc., that take @code{Address} operands and return an @code{Address} result,
30620 have been hidden by declaring these
30621 @code{abstract}, a feature introduced in Ada 95 that helps avoid the potential
30622 ambiguities that would otherwise result from overloading.
30623 (Note that, although @code{Address} is a visible integer type,
30624 good programming practice dictates against exploiting the type's
30625 integer properties such as literals, since this will compromise
30628 Defining @code{Address} as a visible integer type helps achieve
30629 maximum compatibility for existing Ada code,
30630 without sacrificing the capabilities of the 64-bit architecture.
30633 @c ************************************************
30635 @node Microsoft Windows Topics
30636 @appendix Microsoft Windows Topics
30642 This chapter describes topics that are specific to the Microsoft Windows
30643 platforms (NT, 2000, and XP Professional).
30646 * Using GNAT on Windows::
30647 * Using a network installation of GNAT::
30648 * CONSOLE and WINDOWS subsystems::
30649 * Temporary Files::
30650 * Mixed-Language Programming on Windows::
30651 * Windows Calling Conventions::
30652 * Introduction to Dynamic Link Libraries (DLLs)::
30653 * Using DLLs with GNAT::
30654 * Building DLLs with GNAT::
30655 * Building DLLs with GNAT Project files::
30656 * Building DLLs with gnatdll::
30657 * GNAT and Windows Resources::
30658 * Debugging a DLL::
30659 * Setting Stack Size from gnatlink::
30660 * Setting Heap Size from gnatlink::
30663 @node Using GNAT on Windows
30664 @section Using GNAT on Windows
30667 One of the strengths of the GNAT technology is that its tool set
30668 (@command{gcc}, @command{gnatbind}, @command{gnatlink}, @command{gnatmake}, the
30669 @code{gdb} debugger, etc.) is used in the same way regardless of the
30672 On Windows this tool set is complemented by a number of Microsoft-specific
30673 tools that have been provided to facilitate interoperability with Windows
30674 when this is required. With these tools:
30679 You can build applications using the @code{CONSOLE} or @code{WINDOWS}
30683 You can use any Dynamically Linked Library (DLL) in your Ada code (both
30684 relocatable and non-relocatable DLLs are supported).
30687 You can build Ada DLLs for use in other applications. These applications
30688 can be written in a language other than Ada (e.g., C, C++, etc). Again both
30689 relocatable and non-relocatable Ada DLLs are supported.
30692 You can include Windows resources in your Ada application.
30695 You can use or create COM/DCOM objects.
30699 Immediately below are listed all known general GNAT-for-Windows restrictions.
30700 Other restrictions about specific features like Windows Resources and DLLs
30701 are listed in separate sections below.
30706 It is not possible to use @code{GetLastError} and @code{SetLastError}
30707 when tasking, protected records, or exceptions are used. In these
30708 cases, in order to implement Ada semantics, the GNAT run-time system
30709 calls certain Win32 routines that set the last error variable to 0 upon
30710 success. It should be possible to use @code{GetLastError} and
30711 @code{SetLastError} when tasking, protected record, and exception
30712 features are not used, but it is not guaranteed to work.
30715 It is not possible to link against Microsoft libraries except for
30716 import libraries. The library must be built to be compatible with
30717 @file{MSVCRT.LIB} (/MD Microsoft compiler option), @file{LIBC.LIB} and
30718 @file{LIBCMT.LIB} (/ML or /MT Microsoft compiler options) are known to
30719 not be compatible with the GNAT runtime. Even if the library is
30720 compatible with @file{MSVCRT.LIB} it is not guaranteed to work.
30723 When the compilation environment is located on FAT32 drives, users may
30724 experience recompilations of the source files that have not changed if
30725 Daylight Saving Time (DST) state has changed since the last time files
30726 were compiled. NTFS drives do not have this problem.
30729 No components of the GNAT toolset use any entries in the Windows
30730 registry. The only entries that can be created are file associations and
30731 PATH settings, provided the user has chosen to create them at installation
30732 time, as well as some minimal book-keeping information needed to correctly
30733 uninstall or integrate different GNAT products.
30736 @node Using a network installation of GNAT
30737 @section Using a network installation of GNAT
30740 Make sure the system on which GNAT is installed is accessible from the
30741 current machine, i.e., the install location is shared over the network.
30742 Shared resources are accessed on Windows by means of UNC paths, which
30743 have the format @code{\\server\sharename\path}
30745 In order to use such a network installation, simply add the UNC path of the
30746 @file{bin} directory of your GNAT installation in front of your PATH. For
30747 example, if GNAT is installed in @file{\GNAT} directory of a share location
30748 called @file{c-drive} on a machine @file{LOKI}, the following command will
30751 @code{@ @ @ path \\loki\c-drive\gnat\bin;%path%}
30753 Be aware that every compilation using the network installation results in the
30754 transfer of large amounts of data across the network and will likely cause
30755 serious performance penalty.
30757 @node CONSOLE and WINDOWS subsystems
30758 @section CONSOLE and WINDOWS subsystems
30759 @cindex CONSOLE Subsystem
30760 @cindex WINDOWS Subsystem
30764 There are two main subsystems under Windows. The @code{CONSOLE} subsystem
30765 (which is the default subsystem) will always create a console when
30766 launching the application. This is not something desirable when the
30767 application has a Windows GUI. To get rid of this console the
30768 application must be using the @code{WINDOWS} subsystem. To do so
30769 the @option{-mwindows} linker option must be specified.
30772 $ gnatmake winprog -largs -mwindows
30775 @node Temporary Files
30776 @section Temporary Files
30777 @cindex Temporary files
30780 It is possible to control where temporary files gets created by setting
30781 the @env{TMP} environment variable. The file will be created:
30784 @item Under the directory pointed to by the @env{TMP} environment variable if
30785 this directory exists.
30787 @item Under @file{c:\temp}, if the @env{TMP} environment variable is not
30788 set (or not pointing to a directory) and if this directory exists.
30790 @item Under the current working directory otherwise.
30794 This allows you to determine exactly where the temporary
30795 file will be created. This is particularly useful in networked
30796 environments where you may not have write access to some
30799 @node Mixed-Language Programming on Windows
30800 @section Mixed-Language Programming on Windows
30803 Developing pure Ada applications on Windows is no different than on
30804 other GNAT-supported platforms. However, when developing or porting an
30805 application that contains a mix of Ada and C/C++, the choice of your
30806 Windows C/C++ development environment conditions your overall
30807 interoperability strategy.
30809 If you use @command{gcc} to compile the non-Ada part of your application,
30810 there are no Windows-specific restrictions that affect the overall
30811 interoperability with your Ada code. If you plan to use
30812 Microsoft tools (e.g.@: Microsoft Visual C/C++), you should be aware of
30813 the following limitations:
30817 You cannot link your Ada code with an object or library generated with
30818 Microsoft tools if these use the @code{.tls} section (Thread Local
30819 Storage section) since the GNAT linker does not yet support this section.
30822 You cannot link your Ada code with an object or library generated with
30823 Microsoft tools if these use I/O routines other than those provided in
30824 the Microsoft DLL: @code{msvcrt.dll}. This is because the GNAT run time
30825 uses the services of @code{msvcrt.dll} for its I/Os. Use of other I/O
30826 libraries can cause a conflict with @code{msvcrt.dll} services. For
30827 instance Visual C++ I/O stream routines conflict with those in
30832 If you do want to use the Microsoft tools for your non-Ada code and hit one
30833 of the above limitations, you have two choices:
30837 Encapsulate your non-Ada code in a DLL to be linked with your Ada
30838 application. In this case, use the Microsoft or whatever environment to
30839 build the DLL and use GNAT to build your executable
30840 (@pxref{Using DLLs with GNAT}).
30843 Or you can encapsulate your Ada code in a DLL to be linked with the
30844 other part of your application. In this case, use GNAT to build the DLL
30845 (@pxref{Building DLLs with GNAT}) and use the Microsoft or whatever
30846 environment to build your executable.
30849 @node Windows Calling Conventions
30850 @section Windows Calling Conventions
30855 * C Calling Convention::
30856 * Stdcall Calling Convention::
30857 * Win32 Calling Convention::
30858 * DLL Calling Convention::
30862 When a subprogram @code{F} (caller) calls a subprogram @code{G}
30863 (callee), there are several ways to push @code{G}'s parameters on the
30864 stack and there are several possible scenarios to clean up the stack
30865 upon @code{G}'s return. A calling convention is an agreed upon software
30866 protocol whereby the responsibilities between the caller (@code{F}) and
30867 the callee (@code{G}) are clearly defined. Several calling conventions
30868 are available for Windows:
30872 @code{C} (Microsoft defined)
30875 @code{Stdcall} (Microsoft defined)
30878 @code{Win32} (GNAT specific)
30881 @code{DLL} (GNAT specific)
30884 @node C Calling Convention
30885 @subsection @code{C} Calling Convention
30888 This is the default calling convention used when interfacing to C/C++
30889 routines compiled with either @command{gcc} or Microsoft Visual C++.
30891 In the @code{C} calling convention subprogram parameters are pushed on the
30892 stack by the caller from right to left. The caller itself is in charge of
30893 cleaning up the stack after the call. In addition, the name of a routine
30894 with @code{C} calling convention is mangled by adding a leading underscore.
30896 The name to use on the Ada side when importing (or exporting) a routine
30897 with @code{C} calling convention is the name of the routine. For
30898 instance the C function:
30901 int get_val (long);
30905 should be imported from Ada as follows:
30907 @smallexample @c ada
30909 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
30910 pragma Import (C, Get_Val, External_Name => "get_val");
30915 Note that in this particular case the @code{External_Name} parameter could
30916 have been omitted since, when missing, this parameter is taken to be the
30917 name of the Ada entity in lower case. When the @code{Link_Name} parameter
30918 is missing, as in the above example, this parameter is set to be the
30919 @code{External_Name} with a leading underscore.
30921 When importing a variable defined in C, you should always use the @code{C}
30922 calling convention unless the object containing the variable is part of a
30923 DLL (in which case you should use the @code{Stdcall} calling
30924 convention, @pxref{Stdcall Calling Convention}).
30926 @node Stdcall Calling Convention
30927 @subsection @code{Stdcall} Calling Convention
30930 This convention, which was the calling convention used for Pascal
30931 programs, is used by Microsoft for all the routines in the Win32 API for
30932 efficiency reasons. It must be used to import any routine for which this
30933 convention was specified.
30935 In the @code{Stdcall} calling convention subprogram parameters are pushed
30936 on the stack by the caller from right to left. The callee (and not the
30937 caller) is in charge of cleaning the stack on routine exit. In addition,
30938 the name of a routine with @code{Stdcall} calling convention is mangled by
30939 adding a leading underscore (as for the @code{C} calling convention) and a
30940 trailing @code{@@}@code{@var{nn}}, where @var{nn} is the overall size (in
30941 bytes) of the parameters passed to the routine.
30943 The name to use on the Ada side when importing a C routine with a
30944 @code{Stdcall} calling convention is the name of the C routine. The leading
30945 underscore and trailing @code{@@}@code{@var{nn}} are added automatically by
30946 the compiler. For instance the Win32 function:
30949 @b{APIENTRY} int get_val (long);
30953 should be imported from Ada as follows:
30955 @smallexample @c ada
30957 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
30958 pragma Import (Stdcall, Get_Val);
30959 -- On the x86 a long is 4 bytes, so the Link_Name is "_get_val@@4"
30964 As for the @code{C} calling convention, when the @code{External_Name}
30965 parameter is missing, it is taken to be the name of the Ada entity in lower
30966 case. If instead of writing the above import pragma you write:
30968 @smallexample @c ada
30970 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
30971 pragma Import (Stdcall, Get_Val, External_Name => "retrieve_val");
30976 then the imported routine is @code{_retrieve_val@@4}. However, if instead
30977 of specifying the @code{External_Name} parameter you specify the
30978 @code{Link_Name} as in the following example:
30980 @smallexample @c ada
30982 function Get_Val (V : Interfaces.C.long) return Interfaces.C.int;
30983 pragma Import (Stdcall, Get_Val, Link_Name => "retrieve_val");
30988 then the imported routine is @code{retrieve_val}, that is, there is no
30989 decoration at all. No leading underscore and no Stdcall suffix
30990 @code{@@}@code{@var{nn}}.
30993 This is especially important as in some special cases a DLL's entry
30994 point name lacks a trailing @code{@@}@code{@var{nn}} while the exported
30995 name generated for a call has it.
30998 It is also possible to import variables defined in a DLL by using an
30999 import pragma for a variable. As an example, if a DLL contains a
31000 variable defined as:
31007 then, to access this variable from Ada you should write:
31009 @smallexample @c ada
31011 My_Var : Interfaces.C.int;
31012 pragma Import (Stdcall, My_Var);
31017 Note that to ease building cross-platform bindings this convention
31018 will be handled as a @code{C} calling convention on non-Windows platforms.
31020 @node Win32 Calling Convention
31021 @subsection @code{Win32} Calling Convention
31024 This convention, which is GNAT-specific is fully equivalent to the
31025 @code{Stdcall} calling convention described above.
31027 @node DLL Calling Convention
31028 @subsection @code{DLL} Calling Convention
31031 This convention, which is GNAT-specific is fully equivalent to the
31032 @code{Stdcall} calling convention described above.
31034 @node Introduction to Dynamic Link Libraries (DLLs)
31035 @section Introduction to Dynamic Link Libraries (DLLs)
31039 A Dynamically Linked Library (DLL) is a library that can be shared by
31040 several applications running under Windows. A DLL can contain any number of
31041 routines and variables.
31043 One advantage of DLLs is that you can change and enhance them without
31044 forcing all the applications that depend on them to be relinked or
31045 recompiled. However, you should be aware than all calls to DLL routines are
31046 slower since, as you will understand below, such calls are indirect.
31048 To illustrate the remainder of this section, suppose that an application
31049 wants to use the services of a DLL @file{API.dll}. To use the services
31050 provided by @file{API.dll} you must statically link against the DLL or
31051 an import library which contains a jump table with an entry for each
31052 routine and variable exported by the DLL. In the Microsoft world this
31053 import library is called @file{API.lib}. When using GNAT this import
31054 library is called either @file{libAPI.dll.a}, @file{libapi.dll.a},
31055 @file{libAPI.a} or @file{libapi.a} (names are case insensitive).
31057 After you have linked your application with the DLL or the import library
31058 and you run your application, here is what happens:
31062 Your application is loaded into memory.
31065 The DLL @file{API.dll} is mapped into the address space of your
31066 application. This means that:
31070 The DLL will use the stack of the calling thread.
31073 The DLL will use the virtual address space of the calling process.
31076 The DLL will allocate memory from the virtual address space of the calling
31080 Handles (pointers) can be safely exchanged between routines in the DLL
31081 routines and routines in the application using the DLL.
31085 The entries in the jump table (from the import library @file{libAPI.dll.a}
31086 or @file{API.lib} or automatically created when linking against a DLL)
31087 which is part of your application are initialized with the addresses
31088 of the routines and variables in @file{API.dll}.
31091 If present in @file{API.dll}, routines @code{DllMain} or
31092 @code{DllMainCRTStartup} are invoked. These routines typically contain
31093 the initialization code needed for the well-being of the routines and
31094 variables exported by the DLL.
31098 There is an additional point which is worth mentioning. In the Windows
31099 world there are two kind of DLLs: relocatable and non-relocatable
31100 DLLs. Non-relocatable DLLs can only be loaded at a very specific address
31101 in the target application address space. If the addresses of two
31102 non-relocatable DLLs overlap and these happen to be used by the same
31103 application, a conflict will occur and the application will run
31104 incorrectly. Hence, when possible, it is always preferable to use and
31105 build relocatable DLLs. Both relocatable and non-relocatable DLLs are
31106 supported by GNAT. Note that the @option{-s} linker option (see GNU Linker
31107 User's Guide) removes the debugging symbols from the DLL but the DLL can
31108 still be relocated.
31110 As a side note, an interesting difference between Microsoft DLLs and
31111 Unix shared libraries, is the fact that on most Unix systems all public
31112 routines are exported by default in a Unix shared library, while under
31113 Windows it is possible (but not required) to list exported routines in
31114 a definition file (@pxref{The Definition File}).
31116 @node Using DLLs with GNAT
31117 @section Using DLLs with GNAT
31120 * Creating an Ada Spec for the DLL Services::
31121 * Creating an Import Library::
31125 To use the services of a DLL, say @file{API.dll}, in your Ada application
31130 The Ada spec for the routines and/or variables you want to access in
31131 @file{API.dll}. If not available this Ada spec must be built from the C/C++
31132 header files provided with the DLL.
31135 The import library (@file{libAPI.dll.a} or @file{API.lib}). As previously
31136 mentioned an import library is a statically linked library containing the
31137 import table which will be filled at load time to point to the actual
31138 @file{API.dll} routines. Sometimes you don't have an import library for the
31139 DLL you want to use. The following sections will explain how to build
31140 one. Note that this is optional.
31143 The actual DLL, @file{API.dll}.
31147 Once you have all the above, to compile an Ada application that uses the
31148 services of @file{API.dll} and whose main subprogram is @code{My_Ada_App},
31149 you simply issue the command
31152 $ gnatmake my_ada_app -largs -lAPI
31156 The argument @option{-largs -lAPI} at the end of the @command{gnatmake} command
31157 tells the GNAT linker to look first for a library named @file{API.lib}
31158 (Microsoft-style name) and if not found for a libraries named
31159 @file{libAPI.dll.a}, @file{API.dll.a} or @file{libAPI.a}.
31160 (GNAT-style name). Note that if the Ada package spec for @file{API.dll}
31161 contains the following pragma
31163 @smallexample @c ada
31164 pragma Linker_Options ("-lAPI");
31168 you do not have to add @option{-largs -lAPI} at the end of the
31169 @command{gnatmake} command.
31171 If any one of the items above is missing you will have to create it
31172 yourself. The following sections explain how to do so using as an
31173 example a fictitious DLL called @file{API.dll}.
31175 @node Creating an Ada Spec for the DLL Services
31176 @subsection Creating an Ada Spec for the DLL Services
31179 A DLL typically comes with a C/C++ header file which provides the
31180 definitions of the routines and variables exported by the DLL. The Ada
31181 equivalent of this header file is a package spec that contains definitions
31182 for the imported entities. If the DLL you intend to use does not come with
31183 an Ada spec you have to generate one such spec yourself. For example if
31184 the header file of @file{API.dll} is a file @file{api.h} containing the
31185 following two definitions:
31197 then the equivalent Ada spec could be:
31199 @smallexample @c ada
31202 with Interfaces.C.Strings;
31207 function Get (Str : C.Strings.Chars_Ptr) return C.int;
31210 pragma Import (C, Get);
31211 pragma Import (DLL, Some_Var);
31218 Note that a variable is
31219 @strong{always imported with a Stdcall convention}. A function
31220 can have @code{C} or @code{Stdcall} convention.
31221 (@pxref{Windows Calling Conventions}).
31223 @node Creating an Import Library
31224 @subsection Creating an Import Library
31225 @cindex Import library
31228 * The Definition File::
31229 * GNAT-Style Import Library::
31230 * Microsoft-Style Import Library::
31234 If a Microsoft-style import library @file{API.lib} or a GNAT-style
31235 import library @file{libAPI.dll.a} or @file{libAPI.a} is available
31236 with @file{API.dll} you can skip this section. You can also skip this
31237 section if @file{API.dll} or @file{libAPI.dll} is built with GNU tools
31238 as in this case it is possible to link directly against the
31239 DLL. Otherwise read on.
31241 @node The Definition File
31242 @subsubsection The Definition File
31243 @cindex Definition file
31247 As previously mentioned, and unlike Unix systems, the list of symbols
31248 that are exported from a DLL must be provided explicitly in Windows.
31249 The main goal of a definition file is precisely that: list the symbols
31250 exported by a DLL. A definition file (usually a file with a @code{.def}
31251 suffix) has the following structure:
31256 @r{[}LIBRARY @var{name}@r{]}
31257 @r{[}DESCRIPTION @var{string}@r{]}
31267 @item LIBRARY @var{name}
31268 This section, which is optional, gives the name of the DLL.
31270 @item DESCRIPTION @var{string}
31271 This section, which is optional, gives a description string that will be
31272 embedded in the import library.
31275 This section gives the list of exported symbols (procedures, functions or
31276 variables). For instance in the case of @file{API.dll} the @code{EXPORTS}
31277 section of @file{API.def} looks like:
31291 Note that you must specify the correct suffix (@code{@@}@code{@var{nn}})
31292 (@pxref{Windows Calling Conventions}) for a Stdcall
31293 calling convention function in the exported symbols list.
31296 There can actually be other sections in a definition file, but these
31297 sections are not relevant to the discussion at hand.
31299 @node GNAT-Style Import Library
31300 @subsubsection GNAT-Style Import Library
31303 To create a static import library from @file{API.dll} with the GNAT tools
31304 you should proceed as follows:
31308 Create the definition file @file{API.def} (@pxref{The Definition File}).
31309 For that use the @code{dll2def} tool as follows:
31312 $ dll2def API.dll > API.def
31316 @code{dll2def} is a very simple tool: it takes as input a DLL and prints
31317 to standard output the list of entry points in the DLL. Note that if
31318 some routines in the DLL have the @code{Stdcall} convention
31319 (@pxref{Windows Calling Conventions}) with stripped @code{@@}@var{nn}
31320 suffix then you'll have to edit @file{api.def} to add it, and specify
31321 @option{-k} to @command{gnatdll} when creating the import library.
31324 Here are some hints to find the right @code{@@}@var{nn} suffix.
31328 If you have the Microsoft import library (.lib), it is possible to get
31329 the right symbols by using Microsoft @code{dumpbin} tool (see the
31330 corresponding Microsoft documentation for further details).
31333 $ dumpbin /exports api.lib
31337 If you have a message about a missing symbol at link time the compiler
31338 tells you what symbol is expected. You just have to go back to the
31339 definition file and add the right suffix.
31343 Build the import library @code{libAPI.dll.a}, using @code{gnatdll}
31344 (@pxref{Using gnatdll}) as follows:
31347 $ gnatdll -e API.def -d API.dll
31351 @code{gnatdll} takes as input a definition file @file{API.def} and the
31352 name of the DLL containing the services listed in the definition file
31353 @file{API.dll}. The name of the static import library generated is
31354 computed from the name of the definition file as follows: if the
31355 definition file name is @var{xyz}@code{.def}, the import library name will
31356 be @code{lib}@var{xyz}@code{.a}. Note that in the previous example option
31357 @option{-e} could have been removed because the name of the definition
31358 file (before the ``@code{.def}'' suffix) is the same as the name of the
31359 DLL (@pxref{Using gnatdll} for more information about @code{gnatdll}).
31362 @node Microsoft-Style Import Library
31363 @subsubsection Microsoft-Style Import Library
31366 With GNAT you can either use a GNAT-style or Microsoft-style import
31367 library. A Microsoft import library is needed only if you plan to make an
31368 Ada DLL available to applications developed with Microsoft
31369 tools (@pxref{Mixed-Language Programming on Windows}).
31371 To create a Microsoft-style import library for @file{API.dll} you
31372 should proceed as follows:
31376 Create the definition file @file{API.def} from the DLL. For this use either
31377 the @code{dll2def} tool as described above or the Microsoft @code{dumpbin}
31378 tool (see the corresponding Microsoft documentation for further details).
31381 Build the actual import library using Microsoft's @code{lib} utility:
31384 $ lib -machine:IX86 -def:API.def -out:API.lib
31388 If you use the above command the definition file @file{API.def} must
31389 contain a line giving the name of the DLL:
31396 See the Microsoft documentation for further details about the usage of
31400 @node Building DLLs with GNAT
31401 @section Building DLLs with GNAT
31402 @cindex DLLs, building
31405 This section explain how to build DLLs using the GNAT built-in DLL
31406 support. With the following procedure it is straight forward to build
31407 and use DLLs with GNAT.
31411 @item building object files
31413 The first step is to build all objects files that are to be included
31414 into the DLL. This is done by using the standard @command{gnatmake} tool.
31416 @item building the DLL
31418 To build the DLL you must use @command{gcc}'s @option{-shared}
31419 option. It is quite simple to use this method:
31422 $ gcc -shared -o api.dll obj1.o obj2.o @dots{}
31425 It is important to note that in this case all symbols found in the
31426 object files are automatically exported. It is possible to restrict
31427 the set of symbols to export by passing to @command{gcc} a definition
31428 file, @pxref{The Definition File}. For example:
31431 $ gcc -shared -o api.dll api.def obj1.o obj2.o @dots{}
31434 If you use a definition file you must export the elaboration procedures
31435 for every package that required one. Elaboration procedures are named
31436 using the package name followed by "_E".
31438 @item preparing DLL to be used
31440 For the DLL to be used by client programs the bodies must be hidden
31441 from it and the .ali set with read-only attribute. This is very important
31442 otherwise GNAT will recompile all packages and will not actually use
31443 the code in the DLL. For example:
31447 $ copy *.ads *.ali api.dll apilib
31448 $ attrib +R apilib\*.ali
31453 At this point it is possible to use the DLL by directly linking
31454 against it. Note that you must use the GNAT shared runtime when using
31455 GNAT shared libraries. This is achieved by using @option{-shared} binder's
31459 $ gnatmake main -Iapilib -bargs -shared -largs -Lapilib -lAPI
31462 @node Building DLLs with GNAT Project files
31463 @section Building DLLs with GNAT Project files
31464 @cindex DLLs, building
31467 There is nothing specific to Windows in the build process.
31468 @pxref{Library Projects}.
31471 Due to a system limitation, it is not possible under Windows to create threads
31472 when inside the @code{DllMain} routine which is used for auto-initialization
31473 of shared libraries, so it is not possible to have library level tasks in SALs.
31475 @node Building DLLs with gnatdll
31476 @section Building DLLs with gnatdll
31477 @cindex DLLs, building
31480 * Limitations When Using Ada DLLs from Ada::
31481 * Exporting Ada Entities::
31482 * Ada DLLs and Elaboration::
31483 * Ada DLLs and Finalization::
31484 * Creating a Spec for Ada DLLs::
31485 * Creating the Definition File::
31490 Note that it is preferred to use the built-in GNAT DLL support
31491 (@pxref{Building DLLs with GNAT}) or GNAT Project files
31492 (@pxref{Building DLLs with GNAT Project files}) to build DLLs.
31494 This section explains how to build DLLs containing Ada code using
31495 @code{gnatdll}. These DLLs will be referred to as Ada DLLs in the
31496 remainder of this section.
31498 The steps required to build an Ada DLL that is to be used by Ada as well as
31499 non-Ada applications are as follows:
31503 You need to mark each Ada @i{entity} exported by the DLL with a @code{C} or
31504 @code{Stdcall} calling convention to avoid any Ada name mangling for the
31505 entities exported by the DLL (@pxref{Exporting Ada Entities}). You can
31506 skip this step if you plan to use the Ada DLL only from Ada applications.
31509 Your Ada code must export an initialization routine which calls the routine
31510 @code{adainit} generated by @command{gnatbind} to perform the elaboration of
31511 the Ada code in the DLL (@pxref{Ada DLLs and Elaboration}). The initialization
31512 routine exported by the Ada DLL must be invoked by the clients of the DLL
31513 to initialize the DLL.
31516 When useful, the DLL should also export a finalization routine which calls
31517 routine @code{adafinal} generated by @command{gnatbind} to perform the
31518 finalization of the Ada code in the DLL (@pxref{Ada DLLs and Finalization}).
31519 The finalization routine exported by the Ada DLL must be invoked by the
31520 clients of the DLL when the DLL services are no further needed.
31523 You must provide a spec for the services exported by the Ada DLL in each
31524 of the programming languages to which you plan to make the DLL available.
31527 You must provide a definition file listing the exported entities
31528 (@pxref{The Definition File}).
31531 Finally you must use @code{gnatdll} to produce the DLL and the import
31532 library (@pxref{Using gnatdll}).
31536 Note that a relocatable DLL stripped using the @code{strip}
31537 binutils tool will not be relocatable anymore. To build a DLL without
31538 debug information pass @code{-largs -s} to @code{gnatdll}. This
31539 restriction does not apply to a DLL built using a Library Project.
31540 @pxref{Library Projects}.
31542 @node Limitations When Using Ada DLLs from Ada
31543 @subsection Limitations When Using Ada DLLs from Ada
31546 When using Ada DLLs from Ada applications there is a limitation users
31547 should be aware of. Because on Windows the GNAT run time is not in a DLL of
31548 its own, each Ada DLL includes a part of the GNAT run time. Specifically,
31549 each Ada DLL includes the services of the GNAT run time that are necessary
31550 to the Ada code inside the DLL. As a result, when an Ada program uses an
31551 Ada DLL there are two independent GNAT run times: one in the Ada DLL and
31552 one in the main program.
31554 It is therefore not possible to exchange GNAT run-time objects between the
31555 Ada DLL and the main Ada program. Example of GNAT run-time objects are file
31556 handles (e.g.@: @code{Text_IO.File_Type}), tasks types, protected objects
31559 It is completely safe to exchange plain elementary, array or record types,
31560 Windows object handles, etc.
31562 @node Exporting Ada Entities
31563 @subsection Exporting Ada Entities
31564 @cindex Export table
31567 Building a DLL is a way to encapsulate a set of services usable from any
31568 application. As a result, the Ada entities exported by a DLL should be
31569 exported with the @code{C} or @code{Stdcall} calling conventions to avoid
31570 any Ada name mangling. As an example here is an Ada package
31571 @code{API}, spec and body, exporting two procedures, a function, and a
31574 @smallexample @c ada
31577 with Interfaces.C; use Interfaces;
31579 Count : C.int := 0;
31580 function Factorial (Val : C.int) return C.int;
31582 procedure Initialize_API;
31583 procedure Finalize_API;
31584 -- Initialization & Finalization routines. More in the next section.
31586 pragma Export (C, Initialize_API);
31587 pragma Export (C, Finalize_API);
31588 pragma Export (C, Count);
31589 pragma Export (C, Factorial);
31595 @smallexample @c ada
31598 package body API is
31599 function Factorial (Val : C.int) return C.int is
31602 Count := Count + 1;
31603 for K in 1 .. Val loop
31609 procedure Initialize_API is
31611 pragma Import (C, Adainit);
31614 end Initialize_API;
31616 procedure Finalize_API is
31617 procedure Adafinal;
31618 pragma Import (C, Adafinal);
31628 If the Ada DLL you are building will only be used by Ada applications
31629 you do not have to export Ada entities with a @code{C} or @code{Stdcall}
31630 convention. As an example, the previous package could be written as
31633 @smallexample @c ada
31637 Count : Integer := 0;
31638 function Factorial (Val : Integer) return Integer;
31640 procedure Initialize_API;
31641 procedure Finalize_API;
31642 -- Initialization and Finalization routines.
31648 @smallexample @c ada
31651 package body API is
31652 function Factorial (Val : Integer) return Integer is
31653 Fact : Integer := 1;
31655 Count := Count + 1;
31656 for K in 1 .. Val loop
31663 -- The remainder of this package body is unchanged.
31670 Note that if you do not export the Ada entities with a @code{C} or
31671 @code{Stdcall} convention you will have to provide the mangled Ada names
31672 in the definition file of the Ada DLL
31673 (@pxref{Creating the Definition File}).
31675 @node Ada DLLs and Elaboration
31676 @subsection Ada DLLs and Elaboration
31677 @cindex DLLs and elaboration
31680 The DLL that you are building contains your Ada code as well as all the
31681 routines in the Ada library that are needed by it. The first thing a
31682 user of your DLL must do is elaborate the Ada code
31683 (@pxref{Elaboration Order Handling in GNAT}).
31685 To achieve this you must export an initialization routine
31686 (@code{Initialize_API} in the previous example), which must be invoked
31687 before using any of the DLL services. This elaboration routine must call
31688 the Ada elaboration routine @code{adainit} generated by the GNAT binder
31689 (@pxref{Binding with Non-Ada Main Programs}). See the body of
31690 @code{Initialize_Api} for an example. Note that the GNAT binder is
31691 automatically invoked during the DLL build process by the @code{gnatdll}
31692 tool (@pxref{Using gnatdll}).
31694 When a DLL is loaded, Windows systematically invokes a routine called
31695 @code{DllMain}. It would therefore be possible to call @code{adainit}
31696 directly from @code{DllMain} without having to provide an explicit
31697 initialization routine. Unfortunately, it is not possible to call
31698 @code{adainit} from the @code{DllMain} if your program has library level
31699 tasks because access to the @code{DllMain} entry point is serialized by
31700 the system (that is, only a single thread can execute ``through'' it at a
31701 time), which means that the GNAT run time will deadlock waiting for the
31702 newly created task to complete its initialization.
31704 @node Ada DLLs and Finalization
31705 @subsection Ada DLLs and Finalization
31706 @cindex DLLs and finalization
31709 When the services of an Ada DLL are no longer needed, the client code should
31710 invoke the DLL finalization routine, if available. The DLL finalization
31711 routine is in charge of releasing all resources acquired by the DLL. In the
31712 case of the Ada code contained in the DLL, this is achieved by calling
31713 routine @code{adafinal} generated by the GNAT binder
31714 (@pxref{Binding with Non-Ada Main Programs}).
31715 See the body of @code{Finalize_Api} for an
31716 example. As already pointed out the GNAT binder is automatically invoked
31717 during the DLL build process by the @code{gnatdll} tool
31718 (@pxref{Using gnatdll}).
31720 @node Creating a Spec for Ada DLLs
31721 @subsection Creating a Spec for Ada DLLs
31724 To use the services exported by the Ada DLL from another programming
31725 language (e.g.@: C), you have to translate the specs of the exported Ada
31726 entities in that language. For instance in the case of @code{API.dll},
31727 the corresponding C header file could look like:
31732 extern int *_imp__count;
31733 #define count (*_imp__count)
31734 int factorial (int);
31740 It is important to understand that when building an Ada DLL to be used by
31741 other Ada applications, you need two different specs for the packages
31742 contained in the DLL: one for building the DLL and the other for using
31743 the DLL. This is because the @code{DLL} calling convention is needed to
31744 use a variable defined in a DLL, but when building the DLL, the variable
31745 must have either the @code{Ada} or @code{C} calling convention. As an
31746 example consider a DLL comprising the following package @code{API}:
31748 @smallexample @c ada
31752 Count : Integer := 0;
31754 -- Remainder of the package omitted.
31761 After producing a DLL containing package @code{API}, the spec that
31762 must be used to import @code{API.Count} from Ada code outside of the
31765 @smallexample @c ada
31770 pragma Import (DLL, Count);
31776 @node Creating the Definition File
31777 @subsection Creating the Definition File
31780 The definition file is the last file needed to build the DLL. It lists
31781 the exported symbols. As an example, the definition file for a DLL
31782 containing only package @code{API} (where all the entities are exported
31783 with a @code{C} calling convention) is:
31798 If the @code{C} calling convention is missing from package @code{API},
31799 then the definition file contains the mangled Ada names of the above
31800 entities, which in this case are:
31809 api__initialize_api
31814 @node Using gnatdll
31815 @subsection Using @code{gnatdll}
31819 * gnatdll Example::
31820 * gnatdll behind the Scenes::
31825 @code{gnatdll} is a tool to automate the DLL build process once all the Ada
31826 and non-Ada sources that make up your DLL have been compiled.
31827 @code{gnatdll} is actually in charge of two distinct tasks: build the
31828 static import library for the DLL and the actual DLL. The form of the
31829 @code{gnatdll} command is
31833 $ gnatdll @ovar{switches} @var{list-of-files} @r{[}-largs @var{opts}@r{]}
31838 where @var{list-of-files} is a list of ALI and object files. The object
31839 file list must be the exact list of objects corresponding to the non-Ada
31840 sources whose services are to be included in the DLL. The ALI file list
31841 must be the exact list of ALI files for the corresponding Ada sources
31842 whose services are to be included in the DLL. If @var{list-of-files} is
31843 missing, only the static import library is generated.
31846 You may specify any of the following switches to @code{gnatdll}:
31849 @item -a@ovar{address}
31850 @cindex @option{-a} (@code{gnatdll})
31851 Build a non-relocatable DLL at @var{address}. If @var{address} is not
31852 specified the default address @var{0x11000000} will be used. By default,
31853 when this switch is missing, @code{gnatdll} builds relocatable DLL. We
31854 advise the reader to build relocatable DLL.
31856 @item -b @var{address}
31857 @cindex @option{-b} (@code{gnatdll})
31858 Set the relocatable DLL base address. By default the address is
31861 @item -bargs @var{opts}
31862 @cindex @option{-bargs} (@code{gnatdll})
31863 Binder options. Pass @var{opts} to the binder.
31865 @item -d @var{dllfile}
31866 @cindex @option{-d} (@code{gnatdll})
31867 @var{dllfile} is the name of the DLL. This switch must be present for
31868 @code{gnatdll} to do anything. The name of the generated import library is
31869 obtained algorithmically from @var{dllfile} as shown in the following
31870 example: if @var{dllfile} is @code{xyz.dll}, the import library name is
31871 @code{libxyz.dll.a}. The name of the definition file to use (if not specified
31872 by option @option{-e}) is obtained algorithmically from @var{dllfile}
31873 as shown in the following example:
31874 if @var{dllfile} is @code{xyz.dll}, the definition
31875 file used is @code{xyz.def}.
31877 @item -e @var{deffile}
31878 @cindex @option{-e} (@code{gnatdll})
31879 @var{deffile} is the name of the definition file.
31882 @cindex @option{-g} (@code{gnatdll})
31883 Generate debugging information. This information is stored in the object
31884 file and copied from there to the final DLL file by the linker,
31885 where it can be read by the debugger. You must use the
31886 @option{-g} switch if you plan on using the debugger or the symbolic
31890 @cindex @option{-h} (@code{gnatdll})
31891 Help mode. Displays @code{gnatdll} switch usage information.
31894 @cindex @option{-I} (@code{gnatdll})
31895 Direct @code{gnatdll} to search the @var{dir} directory for source and
31896 object files needed to build the DLL.
31897 (@pxref{Search Paths and the Run-Time Library (RTL)}).
31900 @cindex @option{-k} (@code{gnatdll})
31901 Removes the @code{@@}@var{nn} suffix from the import library's exported
31902 names, but keeps them for the link names. You must specify this
31903 option if you want to use a @code{Stdcall} function in a DLL for which
31904 the @code{@@}@var{nn} suffix has been removed. This is the case for most
31905 of the Windows NT DLL for example. This option has no effect when
31906 @option{-n} option is specified.
31908 @item -l @var{file}
31909 @cindex @option{-l} (@code{gnatdll})
31910 The list of ALI and object files used to build the DLL are listed in
31911 @var{file}, instead of being given in the command line. Each line in
31912 @var{file} contains the name of an ALI or object file.
31915 @cindex @option{-n} (@code{gnatdll})
31916 No Import. Do not create the import library.
31919 @cindex @option{-q} (@code{gnatdll})
31920 Quiet mode. Do not display unnecessary messages.
31923 @cindex @option{-v} (@code{gnatdll})
31924 Verbose mode. Display extra information.
31926 @item -largs @var{opts}
31927 @cindex @option{-largs} (@code{gnatdll})
31928 Linker options. Pass @var{opts} to the linker.
31931 @node gnatdll Example
31932 @subsubsection @code{gnatdll} Example
31935 As an example the command to build a relocatable DLL from @file{api.adb}
31936 once @file{api.adb} has been compiled and @file{api.def} created is
31939 $ gnatdll -d api.dll api.ali
31943 The above command creates two files: @file{libapi.dll.a} (the import
31944 library) and @file{api.dll} (the actual DLL). If you want to create
31945 only the DLL, just type:
31948 $ gnatdll -d api.dll -n api.ali
31952 Alternatively if you want to create just the import library, type:
31955 $ gnatdll -d api.dll
31958 @node gnatdll behind the Scenes
31959 @subsubsection @code{gnatdll} behind the Scenes
31962 This section details the steps involved in creating a DLL. @code{gnatdll}
31963 does these steps for you. Unless you are interested in understanding what
31964 goes on behind the scenes, you should skip this section.
31966 We use the previous example of a DLL containing the Ada package @code{API},
31967 to illustrate the steps necessary to build a DLL. The starting point is a
31968 set of objects that will make up the DLL and the corresponding ALI
31969 files. In the case of this example this means that @file{api.o} and
31970 @file{api.ali} are available. To build a relocatable DLL, @code{gnatdll} does
31975 @code{gnatdll} builds the base file (@file{api.base}). A base file gives
31976 the information necessary to generate relocation information for the
31982 $ gnatlink api -o api.jnk -mdll -Wl,--base-file,api.base
31987 In addition to the base file, the @command{gnatlink} command generates an
31988 output file @file{api.jnk} which can be discarded. The @option{-mdll} switch
31989 asks @command{gnatlink} to generate the routines @code{DllMain} and
31990 @code{DllMainCRTStartup} that are called by the Windows loader when the DLL
31991 is loaded into memory.
31994 @code{gnatdll} uses @code{dlltool} (@pxref{Using dlltool}) to build the
31995 export table (@file{api.exp}). The export table contains the relocation
31996 information in a form which can be used during the final link to ensure
31997 that the Windows loader is able to place the DLL anywhere in memory.
32001 $ dlltool --dllname api.dll --def api.def --base-file api.base \
32002 --output-exp api.exp
32007 @code{gnatdll} builds the base file using the new export table. Note that
32008 @command{gnatbind} must be called once again since the binder generated file
32009 has been deleted during the previous call to @command{gnatlink}.
32014 $ gnatlink api -o api.jnk api.exp -mdll
32015 -Wl,--base-file,api.base
32020 @code{gnatdll} builds the new export table using the new base file and
32021 generates the DLL import library @file{libAPI.dll.a}.
32025 $ dlltool --dllname api.dll --def api.def --base-file api.base \
32026 --output-exp api.exp --output-lib libAPI.a
32031 Finally @code{gnatdll} builds the relocatable DLL using the final export
32037 $ gnatlink api api.exp -o api.dll -mdll
32042 @node Using dlltool
32043 @subsubsection Using @code{dlltool}
32046 @code{dlltool} is the low-level tool used by @code{gnatdll} to build
32047 DLLs and static import libraries. This section summarizes the most
32048 common @code{dlltool} switches. The form of the @code{dlltool} command
32052 $ dlltool @ovar{switches}
32056 @code{dlltool} switches include:
32059 @item --base-file @var{basefile}
32060 @cindex @option{--base-file} (@command{dlltool})
32061 Read the base file @var{basefile} generated by the linker. This switch
32062 is used to create a relocatable DLL.
32064 @item --def @var{deffile}
32065 @cindex @option{--def} (@command{dlltool})
32066 Read the definition file.
32068 @item --dllname @var{name}
32069 @cindex @option{--dllname} (@command{dlltool})
32070 Gives the name of the DLL. This switch is used to embed the name of the
32071 DLL in the static import library generated by @code{dlltool} with switch
32072 @option{--output-lib}.
32075 @cindex @option{-k} (@command{dlltool})
32076 Kill @code{@@}@var{nn} from exported names
32077 (@pxref{Windows Calling Conventions}
32078 for a discussion about @code{Stdcall}-style symbols.
32081 @cindex @option{--help} (@command{dlltool})
32082 Prints the @code{dlltool} switches with a concise description.
32084 @item --output-exp @var{exportfile}
32085 @cindex @option{--output-exp} (@command{dlltool})
32086 Generate an export file @var{exportfile}. The export file contains the
32087 export table (list of symbols in the DLL) and is used to create the DLL.
32089 @item --output-lib @var{libfile}
32090 @cindex @option{--output-lib} (@command{dlltool})
32091 Generate a static import library @var{libfile}.
32094 @cindex @option{-v} (@command{dlltool})
32097 @item --as @var{assembler-name}
32098 @cindex @option{--as} (@command{dlltool})
32099 Use @var{assembler-name} as the assembler. The default is @code{as}.
32102 @node GNAT and Windows Resources
32103 @section GNAT and Windows Resources
32104 @cindex Resources, windows
32107 * Building Resources::
32108 * Compiling Resources::
32109 * Using Resources::
32113 Resources are an easy way to add Windows specific objects to your
32114 application. The objects that can be added as resources include:
32143 This section explains how to build, compile and use resources.
32145 @node Building Resources
32146 @subsection Building Resources
32147 @cindex Resources, building
32150 A resource file is an ASCII file. By convention resource files have an
32151 @file{.rc} extension.
32152 The easiest way to build a resource file is to use Microsoft tools
32153 such as @code{imagedit.exe} to build bitmaps, icons and cursors and
32154 @code{dlgedit.exe} to build dialogs.
32155 It is always possible to build an @file{.rc} file yourself by writing a
32158 It is not our objective to explain how to write a resource file. A
32159 complete description of the resource script language can be found in the
32160 Microsoft documentation.
32162 @node Compiling Resources
32163 @subsection Compiling Resources
32166 @cindex Resources, compiling
32169 This section describes how to build a GNAT-compatible (COFF) object file
32170 containing the resources. This is done using the Resource Compiler
32171 @code{windres} as follows:
32174 $ windres -i myres.rc -o myres.o
32178 By default @code{windres} will run @command{gcc} to preprocess the @file{.rc}
32179 file. You can specify an alternate preprocessor (usually named
32180 @file{cpp.exe}) using the @code{windres} @option{--preprocessor}
32181 parameter. A list of all possible options may be obtained by entering
32182 the command @code{windres} @option{--help}.
32184 It is also possible to use the Microsoft resource compiler @code{rc.exe}
32185 to produce a @file{.res} file (binary resource file). See the
32186 corresponding Microsoft documentation for further details. In this case
32187 you need to use @code{windres} to translate the @file{.res} file to a
32188 GNAT-compatible object file as follows:
32191 $ windres -i myres.res -o myres.o
32194 @node Using Resources
32195 @subsection Using Resources
32196 @cindex Resources, using
32199 To include the resource file in your program just add the
32200 GNAT-compatible object file for the resource(s) to the linker
32201 arguments. With @command{gnatmake} this is done by using the @option{-largs}
32205 $ gnatmake myprog -largs myres.o
32208 @node Debugging a DLL
32209 @section Debugging a DLL
32210 @cindex DLL debugging
32213 * Program and DLL Both Built with GCC/GNAT::
32214 * Program Built with Foreign Tools and DLL Built with GCC/GNAT::
32218 Debugging a DLL is similar to debugging a standard program. But
32219 we have to deal with two different executable parts: the DLL and the
32220 program that uses it. We have the following four possibilities:
32224 The program and the DLL are built with @code{GCC/GNAT}.
32226 The program is built with foreign tools and the DLL is built with
32229 The program is built with @code{GCC/GNAT} and the DLL is built with
32235 In this section we address only cases one and two above.
32236 There is no point in trying to debug
32237 a DLL with @code{GNU/GDB}, if there is no GDB-compatible debugging
32238 information in it. To do so you must use a debugger compatible with the
32239 tools suite used to build the DLL.
32241 @node Program and DLL Both Built with GCC/GNAT
32242 @subsection Program and DLL Both Built with GCC/GNAT
32245 This is the simplest case. Both the DLL and the program have @code{GDB}
32246 compatible debugging information. It is then possible to break anywhere in
32247 the process. Let's suppose here that the main procedure is named
32248 @code{ada_main} and that in the DLL there is an entry point named
32252 The DLL (@pxref{Introduction to Dynamic Link Libraries (DLLs)}) and
32253 program must have been built with the debugging information (see GNAT -g
32254 switch). Here are the step-by-step instructions for debugging it:
32257 @item Launch @code{GDB} on the main program.
32263 @item Start the program and stop at the beginning of the main procedure
32270 This step is required to be able to set a breakpoint inside the DLL. As long
32271 as the program is not run, the DLL is not loaded. This has the
32272 consequence that the DLL debugging information is also not loaded, so it is not
32273 possible to set a breakpoint in the DLL.
32275 @item Set a breakpoint inside the DLL
32278 (gdb) break ada_dll
32285 At this stage a breakpoint is set inside the DLL. From there on
32286 you can use the standard approach to debug the whole program
32287 (@pxref{Running and Debugging Ada Programs}).
32290 @c This used to work, probably because the DLLs were non-relocatable
32291 @c keep this section around until the problem is sorted out.
32293 To break on the @code{DllMain} routine it is not possible to follow
32294 the procedure above. At the time the program stop on @code{ada_main}
32295 the @code{DllMain} routine as already been called. Either you can use
32296 the procedure below @pxref{Debugging the DLL Directly} or this procedure:
32299 @item Launch @code{GDB} on the main program.
32305 @item Load DLL symbols
32308 (gdb) add-sym api.dll
32311 @item Set a breakpoint inside the DLL
32314 (gdb) break ada_dll.adb:45
32317 Note that at this point it is not possible to break using the routine symbol
32318 directly as the program is not yet running. The solution is to break
32319 on the proper line (break in @file{ada_dll.adb} line 45).
32321 @item Start the program
32330 @node Program Built with Foreign Tools and DLL Built with GCC/GNAT
32331 @subsection Program Built with Foreign Tools and DLL Built with GCC/GNAT
32334 * Debugging the DLL Directly::
32335 * Attaching to a Running Process::
32339 In this case things are slightly more complex because it is not possible to
32340 start the main program and then break at the beginning to load the DLL and the
32341 associated DLL debugging information. It is not possible to break at the
32342 beginning of the program because there is no @code{GDB} debugging information,
32343 and therefore there is no direct way of getting initial control. This
32344 section addresses this issue by describing some methods that can be used
32345 to break somewhere in the DLL to debug it.
32348 First suppose that the main procedure is named @code{main} (this is for
32349 example some C code built with Microsoft Visual C) and that there is a
32350 DLL named @code{test.dll} containing an Ada entry point named
32354 The DLL (@pxref{Introduction to Dynamic Link Libraries (DLLs)}) must have
32355 been built with debugging information (see GNAT -g option).
32357 @node Debugging the DLL Directly
32358 @subsubsection Debugging the DLL Directly
32362 Find out the executable starting address
32365 $ objdump --file-header main.exe
32368 The starting address is reported on the last line. For example:
32371 main.exe: file format pei-i386
32372 architecture: i386, flags 0x0000010a:
32373 EXEC_P, HAS_DEBUG, D_PAGED
32374 start address 0x00401010
32378 Launch the debugger on the executable.
32385 Set a breakpoint at the starting address, and launch the program.
32388 $ (gdb) break *0x00401010
32392 The program will stop at the given address.
32395 Set a breakpoint on a DLL subroutine.
32398 (gdb) break ada_dll.adb:45
32401 Or if you want to break using a symbol on the DLL, you need first to
32402 select the Ada language (language used by the DLL).
32405 (gdb) set language ada
32406 (gdb) break ada_dll
32410 Continue the program.
32417 This will run the program until it reaches the breakpoint that has been
32418 set. From that point you can use the standard way to debug a program
32419 as described in (@pxref{Running and Debugging Ada Programs}).
32424 It is also possible to debug the DLL by attaching to a running process.
32426 @node Attaching to a Running Process
32427 @subsubsection Attaching to a Running Process
32428 @cindex DLL debugging, attach to process
32431 With @code{GDB} it is always possible to debug a running process by
32432 attaching to it. It is possible to debug a DLL this way. The limitation
32433 of this approach is that the DLL must run long enough to perform the
32434 attach operation. It may be useful for instance to insert a time wasting
32435 loop in the code of the DLL to meet this criterion.
32439 @item Launch the main program @file{main.exe}.
32445 @item Use the Windows @i{Task Manager} to find the process ID. Let's say
32446 that the process PID for @file{main.exe} is 208.
32454 @item Attach to the running process to be debugged.
32460 @item Load the process debugging information.
32463 (gdb) symbol-file main.exe
32466 @item Break somewhere in the DLL.
32469 (gdb) break ada_dll
32472 @item Continue process execution.
32481 This last step will resume the process execution, and stop at
32482 the breakpoint we have set. From there you can use the standard
32483 approach to debug a program as described in
32484 (@pxref{Running and Debugging Ada Programs}).
32486 @node Setting Stack Size from gnatlink
32487 @section Setting Stack Size from @command{gnatlink}
32490 It is possible to specify the program stack size at link time. On modern
32491 versions of Windows, starting with XP, this is mostly useful to set the size of
32492 the main stack (environment task). The other task stacks are set with pragma
32493 Storage_Size or with the @command{gnatbind -d} command.
32495 Since older versions of Windows (2000, NT4, etc.) do not allow setting the
32496 reserve size of individual tasks, the link-time stack size applies to all
32497 tasks, and pragma Storage_Size has no effect.
32498 In particular, Stack Overflow checks are made against this
32499 link-time specified size.
32501 This setting can be done with
32502 @command{gnatlink} using either:
32506 @item using @option{-Xlinker} linker option
32509 $ gnatlink hello -Xlinker --stack=0x10000,0x1000
32512 This sets the stack reserve size to 0x10000 bytes and the stack commit
32513 size to 0x1000 bytes.
32515 @item using @option{-Wl} linker option
32518 $ gnatlink hello -Wl,--stack=0x1000000
32521 This sets the stack reserve size to 0x1000000 bytes. Note that with
32522 @option{-Wl} option it is not possible to set the stack commit size
32523 because the coma is a separator for this option.
32527 @node Setting Heap Size from gnatlink
32528 @section Setting Heap Size from @command{gnatlink}
32531 Under Windows systems, it is possible to specify the program heap size from
32532 @command{gnatlink} using either:
32536 @item using @option{-Xlinker} linker option
32539 $ gnatlink hello -Xlinker --heap=0x10000,0x1000
32542 This sets the heap reserve size to 0x10000 bytes and the heap commit
32543 size to 0x1000 bytes.
32545 @item using @option{-Wl} linker option
32548 $ gnatlink hello -Wl,--heap=0x1000000
32551 This sets the heap reserve size to 0x1000000 bytes. Note that with
32552 @option{-Wl} option it is not possible to set the heap commit size
32553 because the coma is a separator for this option.
32559 @c **********************************
32560 @c * GNU Free Documentation License *
32561 @c **********************************
32563 @c GNU Free Documentation License
32565 @node Index,,GNU Free Documentation License, Top
32571 @c Put table of contents at end, otherwise it precedes the "title page" in
32572 @c the .txt version
32573 @c Edit the pdf file to move the contents to the beginning, after the title