1
\input texinfo @c -*-texinfo-*-
3
@setfilename gnat_rm.info
4
@documentencoding UTF-8
6
@*Generated by Sphinx 1.3b2.@*
8
@settitle GNAT Reference Manual
13
@dircategory GNU Ada Tools
15
* GNAT Reference Manual: (gnat_rm-6). Reference Manual for GNU Ada tools.
18
@definfoenclose strong,`,'
19
@definfoenclose emph,`,'
24
GNAT Reference Manual , November 18, 2015
28
Copyright @copyright{} 2008-2016, Free Software Foundation
34
@title GNAT Reference Manual
39
@c %** start of user preamble
41
@c %** end of user preamble
45
@top GNAT Reference Manual
50
@anchor{gnat_rm doc}@anchor{0}
51
@emph{GNAT, The GNU Ada Development Environment}
54
@include gcc-common.texi
55
GCC version @value{version-GCC}@*
58
Permission is granted to copy, distribute and/or modify this document
59
under the terms of the GNU Free Documentation License, Version 1.3 or
60
any later version published by the Free Software Foundation; with no
61
Invariant Sections, with the Front-Cover Texts being "GNAT Reference
62
Manual", and with no Back-Cover Texts. A copy of the license is
63
included in the section entitled @ref{1,,GNU Free Documentation License}.
67
* Implementation Defined Pragmas::
68
* Implementation Defined Aspects::
69
* Implementation Defined Attributes::
70
* Standard and Implementation Defined Restrictions::
71
* Implementation Advice::
72
* Implementation Defined Characteristics::
73
* Intrinsic Subprograms::
74
* Representation Clauses and Pragmas::
75
* Standard Library Routines::
76
* The Implementation of Standard I/O::
78
* Interfacing to Other Languages::
79
* Specialized Needs Annexes::
80
* Implementation of Specific Ada Features::
81
* Implementation of Ada 2012 Features::
82
* Obsolescent Features::
83
* Compatibility and Porting Guide::
84
* GNU Free Documentation License::
88
--- The Detailed Node Listing ---
92
* What This Reference Manual Contains::
94
* Related Information::
96
Implementation Defined Pragmas
98
* Pragma Abort_Defer::
99
* Pragma Abstract_State::
106
* Pragma Allow_Integer_Address::
109
* Pragma Assert_And_Cut::
110
* Pragma Assertion_Policy::
112
* Pragma Assume_No_Invalid_Values::
113
* Pragma Async_Readers::
114
* Pragma Async_Writers::
115
* Pragma Attribute_Definition::
116
* Pragma C_Pass_By_Copy::
118
* Pragma Check_Float_Overflow::
119
* Pragma Check_Name::
120
* Pragma Check_Policy::
122
* Pragma Common_Object::
123
* Pragma Compile_Time_Error::
124
* Pragma Compile_Time_Warning::
125
* Pragma Compiler_Unit::
126
* Pragma Compiler_Unit_Warning::
127
* Pragma Complete_Representation::
128
* Pragma Complex_Representation::
129
* Pragma Component_Alignment::
130
* Pragma Constant_After_Elaboration::
131
* Pragma Contract_Cases::
132
* Pragma Convention_Identifier::
134
* Pragma CPP_Constructor::
135
* Pragma CPP_Virtual::
136
* Pragma CPP_Vtable::
138
* Pragma Default_Initial_Condition::
140
* Pragma Debug_Policy::
141
* Pragma Default_Scalar_Storage_Order::
142
* Pragma Default_Storage_Pool::
144
* Pragma Detect_Blocking::
145
* Pragma Disable_Atomic_Synchronization::
146
* Pragma Dispatching_Domain::
147
* Pragma Effective_Reads::
148
* Pragma Effective_Writes::
149
* Pragma Elaboration_Checks::
151
* Pragma Enable_Atomic_Synchronization::
152
* Pragma Export_Function::
153
* Pragma Export_Object::
154
* Pragma Export_Procedure::
155
* Pragma Export_Value::
156
* Pragma Export_Valued_Procedure::
157
* Pragma Extend_System::
158
* Pragma Extensions_Allowed::
159
* Pragma Extensions_Visible::
161
* Pragma External_Name_Casing::
163
* Pragma Favor_Top_Level::
164
* Pragma Finalize_Storage_Only::
165
* Pragma Float_Representation::
169
* Pragma Ignore_Pragma::
170
* Pragma Implementation_Defined::
171
* Pragma Implemented::
172
* Pragma Implicit_Packing::
173
* Pragma Import_Function::
174
* Pragma Import_Object::
175
* Pragma Import_Procedure::
176
* Pragma Import_Valued_Procedure::
177
* Pragma Independent::
178
* Pragma Independent_Components::
179
* Pragma Initial_Condition::
180
* Pragma Initialize_Scalars::
181
* Pragma Initializes::
182
* Pragma Inline_Always::
183
* Pragma Inline_Generic::
185
* Pragma Interface_Name::
186
* Pragma Interrupt_Handler::
187
* Pragma Interrupt_State::
189
* Pragma Keep_Names::
192
* Pragma Linker_Alias::
193
* Pragma Linker_Constructor::
194
* Pragma Linker_Destructor::
195
* Pragma Linker_Section::
197
* Pragma Loop_Invariant::
198
* Pragma Loop_Optimize::
199
* Pragma Loop_Variant::
200
* Pragma Machine_Attribute::
202
* Pragma Main_Storage::
204
* Pragma No_Elaboration_Code_All::
207
* Pragma No_Run_Time::
208
* Pragma No_Strict_Aliasing::
209
* Pragma No_Tagged_Streams::
210
* Pragma Normalize_Scalars::
211
* Pragma Obsolescent::
212
* Pragma Optimize_Alignment::
214
* Pragma Overflow_Mode::
215
* Pragma Overriding_Renamings::
216
* Pragma Partition_Elaboration_Policy::
219
* Pragma Persistent_BSS::
222
* Pragma Postcondition::
223
* Pragma Post_Class::
225
* Pragma Precondition::
227
* Pragma Predicate_Failure::
228
* Pragma Preelaborable_Initialization::
229
* Pragma Prefix_Exception_Messages::
231
* Pragma Priority_Specific_Dispatching::
233
* Pragma Profile_Warnings::
234
* Pragma Propagate_Exceptions::
235
* Pragma Provide_Shift_Operators::
236
* Pragma Psect_Object::
237
* Pragma Pure_Function::
240
* Pragma Refined_Depends::
241
* Pragma Refined_Global::
242
* Pragma Refined_Post::
243
* Pragma Refined_State::
244
* Pragma Relative_Deadline::
245
* Pragma Remote_Access_Type::
246
* Pragma Restricted_Run_Time::
247
* Pragma Restriction_Warnings::
248
* Pragma Reviewable::
249
* Pragma Share_Generic::
251
* Pragma Short_Circuit_And_Or::
252
* Pragma Short_Descriptors::
253
* Pragma Simple_Storage_Pool_Type::
254
* Pragma Source_File_Name::
255
* Pragma Source_File_Name_Project::
256
* Pragma Source_Reference::
257
* Pragma SPARK_Mode::
258
* Pragma Static_Elaboration_Desired::
259
* Pragma Stream_Convert::
260
* Pragma Style_Checks::
263
* Pragma Suppress_All::
264
* Pragma Suppress_Debug_Info::
265
* Pragma Suppress_Exception_Locations::
266
* Pragma Suppress_Initialization::
268
* Pragma Task_Storage::
270
* Pragma Thread_Local_Storage::
271
* Pragma Time_Slice::
273
* Pragma Type_Invariant::
274
* Pragma Type_Invariant_Class::
275
* Pragma Unchecked_Union::
276
* Pragma Unevaluated_Use_Of_Old::
277
* Pragma Unimplemented_Unit::
278
* Pragma Universal_Aliasing::
279
* Pragma Universal_Data::
280
* Pragma Unmodified::
281
* Pragma Unreferenced::
282
* Pragma Unreferenced_Objects::
283
* Pragma Unreserve_All_Interrupts::
284
* Pragma Unsuppress::
285
* Pragma Use_VADS_Size::
286
* Pragma Validity_Checks::
288
* Pragma Volatile_Full_Access::
289
* Pragma Volatile_Function::
290
* Pragma Warning_As_Error::
292
* Pragma Weak_External::
293
* Pragma Wide_Character_Encoding::
295
Implementation Defined Aspects
297
* Aspect Abstract_State::
299
* Aspect Async_Readers::
300
* Aspect Async_Writers::
301
* Aspect Constant_After_Elaboration::
302
* Aspect Contract_Cases::
304
* Aspect Default_Initial_Condition::
306
* Aspect Dimension_System::
307
* Aspect Disable_Controlled::
308
* Aspect Effective_Reads::
309
* Aspect Effective_Writes::
310
* Aspect Extensions_Visible::
311
* Aspect Favor_Top_Level::
314
* Aspect Initial_Condition::
315
* Aspect Initializes::
316
* Aspect Inline_Always::
318
* Aspect Invariant'Class::
320
* Aspect Linker_Section::
322
* Aspect No_Elaboration_Code_All::
323
* Aspect No_Tagged_Streams::
324
* Aspect Object_Size::
325
* Aspect Obsolescent::
327
* Aspect Persistent_BSS::
329
* Aspect Pure_Function::
330
* Aspect Refined_Depends::
331
* Aspect Refined_Global::
332
* Aspect Refined_Post::
333
* Aspect Refined_State::
334
* Aspect Remote_Access_Type::
335
* Aspect Scalar_Storage_Order::
337
* Aspect Simple_Storage_Pool::
338
* Aspect Simple_Storage_Pool_Type::
339
* Aspect SPARK_Mode::
340
* Aspect Suppress_Debug_Info::
341
* Aspect Suppress_Initialization::
343
* Aspect Thread_Local_Storage::
344
* Aspect Universal_Aliasing::
345
* Aspect Universal_Data::
346
* Aspect Unmodified::
347
* Aspect Unreferenced::
348
* Aspect Unreferenced_Objects::
349
* Aspect Value_Size::
350
* Aspect Volatile_Full_Access::
351
* Aspect Volatile_Function::
354
Implementation Defined Attributes
356
* Attribute Abort_Signal::
357
* Attribute Address_Size::
358
* Attribute Asm_Input::
359
* Attribute Asm_Output::
360
* Attribute Atomic_Always_Lock_Free::
362
* Attribute Bit_Position::
363
* Attribute Code_Address::
364
* Attribute Compiler_Version::
365
* Attribute Constrained::
366
* Attribute Default_Bit_Order::
367
* Attribute Default_Scalar_Storage_Order::
369
* Attribute Descriptor_Size::
370
* Attribute Elaborated::
371
* Attribute Elab_Body::
372
* Attribute Elab_Spec::
373
* Attribute Elab_Subp_Body::
375
* Attribute Enabled::
376
* Attribute Enum_Rep::
377
* Attribute Enum_Val::
378
* Attribute Epsilon::
379
* Attribute Fast_Math::
380
* Attribute Fixed_Value::
381
* Attribute From_Any::
382
* Attribute Has_Access_Values::
383
* Attribute Has_Discriminants::
385
* Attribute Integer_Value::
386
* Attribute Invalid_Value::
387
* Attribute Iterable::
389
* Attribute Library_Level::
390
* Attribute Lock_Free::
391
* Attribute Loop_Entry::
392
* Attribute Machine_Size::
393
* Attribute Mantissa::
394
* Attribute Maximum_Alignment::
395
* Attribute Mechanism_Code::
396
* Attribute Null_Parameter::
397
* Attribute Object_Size::
399
* Attribute Passed_By_Reference::
400
* Attribute Pool_Address::
401
* Attribute Range_Length::
402
* Attribute Restriction_Set::
404
* Attribute Safe_Emax::
405
* Attribute Safe_Large::
406
* Attribute Safe_Small::
407
* Attribute Scalar_Storage_Order::
408
* Attribute Simple_Storage_Pool::
410
* Attribute Storage_Unit::
411
* Attribute Stub_Type::
412
* Attribute System_Allocator_Alignment::
413
* Attribute Target_Name::
414
* Attribute To_Address::
416
* Attribute Type_Class::
417
* Attribute Type_Key::
418
* Attribute TypeCode::
419
* Attribute Unconstrained_Array::
420
* Attribute Universal_Literal_String::
421
* Attribute Unrestricted_Access::
423
* Attribute Valid_Scalars::
424
* Attribute VADS_Size::
425
* Attribute Value_Size::
426
* Attribute Wchar_T_Size::
427
* Attribute Word_Size::
429
Standard and Implementation Defined Restrictions
431
* Partition-Wide Restrictions::
432
* Program Unit Level Restrictions::
434
Partition-Wide Restrictions
436
* Immediate_Reclamation::
437
* Max_Asynchronous_Select_Nesting::
438
* Max_Entry_Queue_Length::
439
* Max_Protected_Entries::
440
* Max_Select_Alternatives::
441
* Max_Storage_At_Blocking::
444
* No_Abort_Statements::
445
* No_Access_Parameter_Allocators::
446
* No_Access_Subprograms::
448
* No_Anonymous_Allocators::
449
* No_Asynchronous_Control::
452
* No_Default_Initialization::
455
* No_Direct_Boolean_Operators::
457
* No_Dispatching_Calls::
458
* No_Dynamic_Attachment::
459
* No_Dynamic_Priorities::
460
* No_Entry_Calls_In_Elaboration_Code::
461
* No_Enumeration_Maps::
462
* No_Exception_Handlers::
463
* No_Exception_Propagation::
464
* No_Exception_Registration::
468
* No_Floating_Point::
469
* No_Implicit_Conditionals::
470
* No_Implicit_Dynamic_Code::
471
* No_Implicit_Heap_Allocations::
472
* No_Implicit_Loops::
473
* No_Implicit_Protected_Object_Allocations::
474
* No_Implicit_Task_Allocations::
475
* No_Initialize_Scalars::
477
* No_Local_Allocators::
478
* No_Local_Protected_Objects::
479
* No_Local_Timing_Events::
480
* No_Long_Long_Integers::
481
* No_Multiple_Elaboration::
482
* No_Nested_Finalization::
483
* No_Protected_Type_Allocators::
484
* No_Protected_Types::
487
* No_Relative_Delay::
488
* No_Requeue_Statements::
489
* No_Secondary_Stack::
490
* No_Select_Statements::
491
* No_Specific_Termination_Handlers::
492
* No_Specification_of_Aspect::
493
* No_Standard_Allocators_After_Elaboration::
494
* No_Standard_Storage_Pools::
495
* No_Stream_Optimizations::
497
* No_Task_Allocators::
498
* No_Task_At_Interrupt_Priority::
499
* No_Task_Attributes_Package::
500
* No_Task_Hierarchy::
501
* No_Task_Termination::
503
* No_Terminate_Alternatives::
504
* No_Unchecked_Access::
505
* No_Unchecked_Conversion::
506
* No_Unchecked_Deallocation::
510
* Static_Priorities::
511
* Static_Storage_Size::
513
Program Unit Level Restrictions
515
* No_Elaboration_Code::
516
* No_Dynamic_Sized_Objects::
518
* No_Implementation_Aspect_Specifications::
519
* No_Implementation_Attributes::
520
* No_Implementation_Identifiers::
521
* No_Implementation_Pragmas::
522
* No_Implementation_Restrictions::
523
* No_Implementation_Units::
524
* No_Implicit_Aliasing::
525
* No_Obsolescent_Features::
526
* No_Wide_Characters::
529
Implementation Advice
531
* RM 1.1.3(20); Error Detection: RM 1 1 3 20 Error Detection.
532
* RM 1.1.3(31); Child Units: RM 1 1 3 31 Child Units.
533
* RM 1.1.5(12); Bounded Errors: RM 1 1 5 12 Bounded Errors.
534
* RM 2.8(16); Pragmas: RM 2 8 16 Pragmas.
535
* RM 2.8(17-19); Pragmas: RM 2 8 17-19 Pragmas.
536
* RM 3.5.2(5); Alternative Character Sets: RM 3 5 2 5 Alternative Character Sets.
537
* RM 3.5.4(28); Integer Types: RM 3 5 4 28 Integer Types.
538
* RM 3.5.4(29); Integer Types: RM 3 5 4 29 Integer Types.
539
* RM 3.5.5(8); Enumeration Values: RM 3 5 5 8 Enumeration Values.
540
* RM 3.5.7(17); Float Types: RM 3 5 7 17 Float Types.
541
* RM 3.6.2(11); Multidimensional Arrays: RM 3 6 2 11 Multidimensional Arrays.
542
* RM 9.6(30-31); Duration'Small: RM 9 6 30-31 Duration'Small.
543
* RM 10.2.1(12); Consistent Representation: RM 10 2 1 12 Consistent Representation.
544
* RM 11.4.1(19); Exception Information: RM 11 4 1 19 Exception Information.
545
* RM 11.5(28); Suppression of Checks: RM 11 5 28 Suppression of Checks.
546
* RM 13.1 (21-24); Representation Clauses: RM 13 1 21-24 Representation Clauses.
547
* RM 13.2(6-8); Packed Types: RM 13 2 6-8 Packed Types.
548
* RM 13.3(14-19); Address Clauses: RM 13 3 14-19 Address Clauses.
549
* RM 13.3(29-35); Alignment Clauses: RM 13 3 29-35 Alignment Clauses.
550
* RM 13.3(42-43); Size Clauses: RM 13 3 42-43 Size Clauses.
551
* RM 13.3(50-56); Size Clauses: RM 13 3 50-56 Size Clauses.
552
* RM 13.3(71-73); Component Size Clauses: RM 13 3 71-73 Component Size Clauses.
553
* RM 13.4(9-10); Enumeration Representation Clauses: RM 13 4 9-10 Enumeration Representation Clauses.
554
* RM 13.5.1(17-22); Record Representation Clauses: RM 13 5 1 17-22 Record Representation Clauses.
555
* RM 13.5.2(5); Storage Place Attributes: RM 13 5 2 5 Storage Place Attributes.
556
* RM 13.5.3(7-8); Bit Ordering: RM 13 5 3 7-8 Bit Ordering.
557
* RM 13.7(37); Address as Private: RM 13 7 37 Address as Private.
558
* RM 13.7.1(16); Address Operations: RM 13 7 1 16 Address Operations.
559
* RM 13.9(14-17); Unchecked Conversion: RM 13 9 14-17 Unchecked Conversion.
560
* RM 13.11(23-25); Implicit Heap Usage: RM 13 11 23-25 Implicit Heap Usage.
561
* RM 13.11.2(17); Unchecked Deallocation: RM 13 11 2 17 Unchecked Deallocation.
562
* RM 13.13.2(17); Stream Oriented Attributes: RM 13 13 2 17 Stream Oriented Attributes.
563
* RM A.1(52); Names of Predefined Numeric Types: RM A 1 52 Names of Predefined Numeric Types.
564
* RM A.3.2(49); Ada.Characters.Handling: RM A 3 2 49 Ada Characters Handling.
565
* RM A.4.4(106); Bounded-Length String Handling: RM A 4 4 106 Bounded-Length String Handling.
566
* RM A.5.2(46-47); Random Number Generation: RM A 5 2 46-47 Random Number Generation.
567
* RM A.10.7(23); Get_Immediate: RM A 10 7 23 Get_Immediate.
568
* RM B.1(39-41); Pragma Export: RM B 1 39-41 Pragma Export.
569
* RM B.2(12-13); Package Interfaces: RM B 2 12-13 Package Interfaces.
570
* RM B.3(63-71); Interfacing with C: RM B 3 63-71 Interfacing with C.
571
* RM B.4(95-98); Interfacing with COBOL: RM B 4 95-98 Interfacing with COBOL.
572
* RM B.5(22-26); Interfacing with Fortran: RM B 5 22-26 Interfacing with Fortran.
573
* RM C.1(3-5); Access to Machine Operations: RM C 1 3-5 Access to Machine Operations.
574
* RM C.1(10-16); Access to Machine Operations: RM C 1 10-16 Access to Machine Operations.
575
* RM C.3(28); Interrupt Support: RM C 3 28 Interrupt Support.
576
* RM C.3.1(20-21); Protected Procedure Handlers: RM C 3 1 20-21 Protected Procedure Handlers.
577
* RM C.3.2(25); Package Interrupts: RM C 3 2 25 Package Interrupts.
578
* RM C.4(14); Pre-elaboration Requirements: RM C 4 14 Pre-elaboration Requirements.
579
* RM C.5(8); Pragma Discard_Names: RM C 5 8 Pragma Discard_Names.
580
* RM C.7.2(30); The Package Task_Attributes: RM C 7 2 30 The Package Task_Attributes.
581
* RM D.3(17); Locking Policies: RM D 3 17 Locking Policies.
582
* RM D.4(16); Entry Queuing Policies: RM D 4 16 Entry Queuing Policies.
583
* RM D.6(9-10); Preemptive Abort: RM D 6 9-10 Preemptive Abort.
584
* RM D.7(21); Tasking Restrictions: RM D 7 21 Tasking Restrictions.
585
* RM D.8(47-49); Monotonic Time: RM D 8 47-49 Monotonic Time.
586
* RM E.5(28-29); Partition Communication Subsystem: RM E 5 28-29 Partition Communication Subsystem.
587
* RM F(7); COBOL Support: RM F 7 COBOL Support.
588
* RM F.1(2); Decimal Radix Support: RM F 1 2 Decimal Radix Support.
589
* RM G; Numerics: RM G Numerics.
590
* RM G.1.1(56-58); Complex Types: RM G 1 1 56-58 Complex Types.
591
* RM G.1.2(49); Complex Elementary Functions: RM G 1 2 49 Complex Elementary Functions.
592
* RM G.2.4(19); Accuracy Requirements: RM G 2 4 19 Accuracy Requirements.
593
* RM G.2.6(15); Complex Arithmetic Accuracy: RM G 2 6 15 Complex Arithmetic Accuracy.
594
* RM H.6(15/2); Pragma Partition_Elaboration_Policy: RM H 6 15/2 Pragma Partition_Elaboration_Policy.
596
Intrinsic Subprograms
598
* Intrinsic Operators::
602
* Exception_Information::
603
* Exception_Message::
607
* Shifts and Rotates::
610
Representation Clauses and Pragmas
612
* Alignment Clauses::
614
* Storage_Size Clauses::
615
* Size of Variant Record Objects::
616
* Biased Representation::
617
* Value_Size and Object_Size Clauses::
618
* Component_Size Clauses::
619
* Bit_Order Clauses::
620
* Effect of Bit_Order on Byte Ordering::
621
* Pragma Pack for Arrays::
622
* Pragma Pack for Records::
623
* Record Representation Clauses::
624
* Handling of Records with Holes::
625
* Enumeration Clauses::
627
* Use of Address Clauses for Memory-Mapped I/O::
628
* Effect of Convention on Representation::
629
* Conventions and Anonymous Access Types::
630
* Determining the Representations chosen by GNAT::
632
The Implementation of Standard I/O
634
* Standard I/O Packages::
640
* Wide_Wide_Text_IO::
644
* Filenames encoding::
645
* File content encoding::
647
* Operations on C Streams::
648
* Interfacing to C Streams::
652
* Stream Pointer Positioning::
653
* Reading and Writing Non-Regular Files::
655
* Treating Text_IO Files as Streams::
656
* Text_IO Extensions::
657
* Text_IO Facilities for Unbounded Strings::
661
* Stream Pointer Positioning: Stream Pointer Positioning<2>.
662
* Reading and Writing Non-Regular Files: Reading and Writing Non-Regular Files<2>.
666
* Stream Pointer Positioning: Stream Pointer Positioning<3>.
667
* Reading and Writing Non-Regular Files: Reading and Writing Non-Regular Files<3>.
671
* Ada.Characters.Latin_9 (a-chlat9.ads): Ada Characters Latin_9 a-chlat9 ads.
672
* Ada.Characters.Wide_Latin_1 (a-cwila1.ads): Ada Characters Wide_Latin_1 a-cwila1 ads.
673
* Ada.Characters.Wide_Latin_9 (a-cwila1.ads): Ada Characters Wide_Latin_9 a-cwila1 ads.
674
* Ada.Characters.Wide_Wide_Latin_1 (a-chzla1.ads): Ada Characters Wide_Wide_Latin_1 a-chzla1 ads.
675
* Ada.Characters.Wide_Wide_Latin_9 (a-chzla9.ads): Ada Characters Wide_Wide_Latin_9 a-chzla9 ads.
676
* Ada.Containers.Formal_Doubly_Linked_Lists (a-cfdlli.ads): Ada Containers Formal_Doubly_Linked_Lists a-cfdlli ads.
677
* Ada.Containers.Formal_Hashed_Maps (a-cfhama.ads): Ada Containers Formal_Hashed_Maps a-cfhama ads.
678
* Ada.Containers.Formal_Hashed_Sets (a-cfhase.ads): Ada Containers Formal_Hashed_Sets a-cfhase ads.
679
* Ada.Containers.Formal_Ordered_Maps (a-cforma.ads): Ada Containers Formal_Ordered_Maps a-cforma ads.
680
* Ada.Containers.Formal_Ordered_Sets (a-cforse.ads): Ada Containers Formal_Ordered_Sets a-cforse ads.
681
* Ada.Containers.Formal_Vectors (a-cofove.ads): Ada Containers Formal_Vectors a-cofove ads.
682
* Ada.Containers.Formal_Indefinite_Vectors (a-cfinve.ads): Ada Containers Formal_Indefinite_Vectors a-cfinve ads.
683
* Ada.Containers.Bounded_Holders (a-coboho.ads): Ada Containers Bounded_Holders a-coboho ads.
684
* Ada.Command_Line.Environment (a-colien.ads): Ada Command_Line Environment a-colien ads.
685
* Ada.Command_Line.Remove (a-colire.ads): Ada Command_Line Remove a-colire ads.
686
* Ada.Command_Line.Response_File (a-clrefi.ads): Ada Command_Line Response_File a-clrefi ads.
687
* Ada.Direct_IO.C_Streams (a-diocst.ads): Ada Direct_IO C_Streams a-diocst ads.
688
* Ada.Exceptions.Is_Null_Occurrence (a-einuoc.ads): Ada Exceptions Is_Null_Occurrence a-einuoc ads.
689
* Ada.Exceptions.Last_Chance_Handler (a-elchha.ads): Ada Exceptions Last_Chance_Handler a-elchha ads.
690
* Ada.Exceptions.Traceback (a-exctra.ads): Ada Exceptions Traceback a-exctra ads.
691
* Ada.Sequential_IO.C_Streams (a-siocst.ads): Ada Sequential_IO C_Streams a-siocst ads.
692
* Ada.Streams.Stream_IO.C_Streams (a-ssicst.ads): Ada Streams Stream_IO C_Streams a-ssicst ads.
693
* Ada.Strings.Unbounded.Text_IO (a-suteio.ads): Ada Strings Unbounded Text_IO a-suteio ads.
694
* Ada.Strings.Wide_Unbounded.Wide_Text_IO (a-swuwti.ads): Ada Strings Wide_Unbounded Wide_Text_IO a-swuwti ads.
695
* Ada.Strings.Wide_Wide_Unbounded.Wide_Wide_Text_IO (a-szuzti.ads): Ada Strings Wide_Wide_Unbounded Wide_Wide_Text_IO a-szuzti ads.
696
* Ada.Text_IO.C_Streams (a-tiocst.ads): Ada Text_IO C_Streams a-tiocst ads.
697
* Ada.Text_IO.Reset_Standard_Files (a-tirsfi.ads): Ada Text_IO Reset_Standard_Files a-tirsfi ads.
698
* Ada.Wide_Characters.Unicode (a-wichun.ads): Ada Wide_Characters Unicode a-wichun ads.
699
* Ada.Wide_Text_IO.C_Streams (a-wtcstr.ads): Ada Wide_Text_IO C_Streams a-wtcstr ads.
700
* Ada.Wide_Text_IO.Reset_Standard_Files (a-wrstfi.ads): Ada Wide_Text_IO Reset_Standard_Files a-wrstfi ads.
701
* Ada.Wide_Wide_Characters.Unicode (a-zchuni.ads): Ada Wide_Wide_Characters Unicode a-zchuni ads.
702
* Ada.Wide_Wide_Text_IO.C_Streams (a-ztcstr.ads): Ada Wide_Wide_Text_IO C_Streams a-ztcstr ads.
703
* Ada.Wide_Wide_Text_IO.Reset_Standard_Files (a-zrstfi.ads): Ada Wide_Wide_Text_IO Reset_Standard_Files a-zrstfi ads.
704
* GNAT.Altivec (g-altive.ads): GNAT Altivec g-altive ads.
705
* GNAT.Altivec.Conversions (g-altcon.ads): GNAT Altivec Conversions g-altcon ads.
706
* GNAT.Altivec.Vector_Operations (g-alveop.ads): GNAT Altivec Vector_Operations g-alveop ads.
707
* GNAT.Altivec.Vector_Types (g-alvety.ads): GNAT Altivec Vector_Types g-alvety ads.
708
* GNAT.Altivec.Vector_Views (g-alvevi.ads): GNAT Altivec Vector_Views g-alvevi ads.
709
* GNAT.Array_Split (g-arrspl.ads): GNAT Array_Split g-arrspl ads.
710
* GNAT.AWK (g-awk.ads): GNAT AWK g-awk ads.
711
* GNAT.Bind_Environment (g-binenv.ads): GNAT Bind_Environment g-binenv ads.
712
* GNAT.Bounded_Buffers (g-boubuf.ads): GNAT Bounded_Buffers g-boubuf ads.
713
* GNAT.Bounded_Mailboxes (g-boumai.ads): GNAT Bounded_Mailboxes g-boumai ads.
714
* GNAT.Bubble_Sort (g-bubsor.ads): GNAT Bubble_Sort g-bubsor ads.
715
* GNAT.Bubble_Sort_A (g-busora.ads): GNAT Bubble_Sort_A g-busora ads.
716
* GNAT.Bubble_Sort_G (g-busorg.ads): GNAT Bubble_Sort_G g-busorg ads.
717
* GNAT.Byte_Order_Mark (g-byorma.ads): GNAT Byte_Order_Mark g-byorma ads.
718
* GNAT.Byte_Swapping (g-bytswa.ads): GNAT Byte_Swapping g-bytswa ads.
719
* GNAT.Calendar (g-calend.ads): GNAT Calendar g-calend ads.
720
* GNAT.Calendar.Time_IO (g-catiio.ads): GNAT Calendar Time_IO g-catiio ads.
721
* GNAT.CRC32 (g-crc32.ads): GNAT CRC32 g-crc32 ads.
722
* GNAT.Case_Util (g-casuti.ads): GNAT Case_Util g-casuti ads.
723
* GNAT.CGI (g-cgi.ads): GNAT CGI g-cgi ads.
724
* GNAT.CGI.Cookie (g-cgicoo.ads): GNAT CGI Cookie g-cgicoo ads.
725
* GNAT.CGI.Debug (g-cgideb.ads): GNAT CGI Debug g-cgideb ads.
726
* GNAT.Command_Line (g-comlin.ads): GNAT Command_Line g-comlin ads.
727
* GNAT.Compiler_Version (g-comver.ads): GNAT Compiler_Version g-comver ads.
728
* GNAT.Ctrl_C (g-ctrl_c.ads): GNAT Ctrl_C g-ctrl_c ads.
729
* GNAT.Current_Exception (g-curexc.ads): GNAT Current_Exception g-curexc ads.
730
* GNAT.Debug_Pools (g-debpoo.ads): GNAT Debug_Pools g-debpoo ads.
731
* GNAT.Debug_Utilities (g-debuti.ads): GNAT Debug_Utilities g-debuti ads.
732
* GNAT.Decode_String (g-decstr.ads): GNAT Decode_String g-decstr ads.
733
* GNAT.Decode_UTF8_String (g-deutst.ads): GNAT Decode_UTF8_String g-deutst ads.
734
* GNAT.Directory_Operations (g-dirope.ads): GNAT Directory_Operations g-dirope ads.
735
* GNAT.Directory_Operations.Iteration (g-diopit.ads): GNAT Directory_Operations Iteration g-diopit ads.
736
* GNAT.Dynamic_HTables (g-dynhta.ads): GNAT Dynamic_HTables g-dynhta ads.
737
* GNAT.Dynamic_Tables (g-dyntab.ads): GNAT Dynamic_Tables g-dyntab ads.
738
* GNAT.Encode_String (g-encstr.ads): GNAT Encode_String g-encstr ads.
739
* GNAT.Encode_UTF8_String (g-enutst.ads): GNAT Encode_UTF8_String g-enutst ads.
740
* GNAT.Exception_Actions (g-excact.ads): GNAT Exception_Actions g-excact ads.
741
* GNAT.Exception_Traces (g-exctra.ads): GNAT Exception_Traces g-exctra ads.
742
* GNAT.Exceptions (g-expect.ads): GNAT Exceptions g-expect ads.
743
* GNAT.Expect (g-expect.ads): GNAT Expect g-expect ads.
744
* GNAT.Expect.TTY (g-exptty.ads): GNAT Expect TTY g-exptty ads.
745
* GNAT.Float_Control (g-flocon.ads): GNAT Float_Control g-flocon ads.
746
* GNAT.Formatted_String (g-forstr.ads): GNAT Formatted_String g-forstr ads.
747
* GNAT.Heap_Sort (g-heasor.ads): GNAT Heap_Sort g-heasor ads.
748
* GNAT.Heap_Sort_A (g-hesora.ads): GNAT Heap_Sort_A g-hesora ads.
749
* GNAT.Heap_Sort_G (g-hesorg.ads): GNAT Heap_Sort_G g-hesorg ads.
750
* GNAT.HTable (g-htable.ads): GNAT HTable g-htable ads.
751
* GNAT.IO (g-io.ads): GNAT IO g-io ads.
752
* GNAT.IO_Aux (g-io_aux.ads): GNAT IO_Aux g-io_aux ads.
753
* GNAT.Lock_Files (g-locfil.ads): GNAT Lock_Files g-locfil ads.
754
* GNAT.MBBS_Discrete_Random (g-mbdira.ads): GNAT MBBS_Discrete_Random g-mbdira ads.
755
* GNAT.MBBS_Float_Random (g-mbflra.ads): GNAT MBBS_Float_Random g-mbflra ads.
756
* GNAT.MD5 (g-md5.ads): GNAT MD5 g-md5 ads.
757
* GNAT.Memory_Dump (g-memdum.ads): GNAT Memory_Dump g-memdum ads.
758
* GNAT.Most_Recent_Exception (g-moreex.ads): GNAT Most_Recent_Exception g-moreex ads.
759
* GNAT.OS_Lib (g-os_lib.ads): GNAT OS_Lib g-os_lib ads.
760
* GNAT.Perfect_Hash_Generators (g-pehage.ads): GNAT Perfect_Hash_Generators g-pehage ads.
761
* GNAT.Random_Numbers (g-rannum.ads): GNAT Random_Numbers g-rannum ads.
762
* GNAT.Regexp (g-regexp.ads): GNAT Regexp g-regexp ads.
763
* GNAT.Registry (g-regist.ads): GNAT Registry g-regist ads.
764
* GNAT.Regpat (g-regpat.ads): GNAT Regpat g-regpat ads.
765
* GNAT.Rewrite_Data (g-rewdat.ads): GNAT Rewrite_Data g-rewdat ads.
766
* GNAT.Secondary_Stack_Info (g-sestin.ads): GNAT Secondary_Stack_Info g-sestin ads.
767
* GNAT.Semaphores (g-semaph.ads): GNAT Semaphores g-semaph ads.
768
* GNAT.Serial_Communications (g-sercom.ads): GNAT Serial_Communications g-sercom ads.
769
* GNAT.SHA1 (g-sha1.ads): GNAT SHA1 g-sha1 ads.
770
* GNAT.SHA224 (g-sha224.ads): GNAT SHA224 g-sha224 ads.
771
* GNAT.SHA256 (g-sha256.ads): GNAT SHA256 g-sha256 ads.
772
* GNAT.SHA384 (g-sha384.ads): GNAT SHA384 g-sha384 ads.
773
* GNAT.SHA512 (g-sha512.ads): GNAT SHA512 g-sha512 ads.
774
* GNAT.Signals (g-signal.ads): GNAT Signals g-signal ads.
775
* GNAT.Sockets (g-socket.ads): GNAT Sockets g-socket ads.
776
* GNAT.Source_Info (g-souinf.ads): GNAT Source_Info g-souinf ads.
777
* GNAT.Spelling_Checker (g-speche.ads): GNAT Spelling_Checker g-speche ads.
778
* GNAT.Spelling_Checker_Generic (g-spchge.ads): GNAT Spelling_Checker_Generic g-spchge ads.
779
* GNAT.Spitbol.Patterns (g-spipat.ads): GNAT Spitbol Patterns g-spipat ads.
780
* GNAT.Spitbol (g-spitbo.ads): GNAT Spitbol g-spitbo ads.
781
* GNAT.Spitbol.Table_Boolean (g-sptabo.ads): GNAT Spitbol Table_Boolean g-sptabo ads.
782
* GNAT.Spitbol.Table_Integer (g-sptain.ads): GNAT Spitbol Table_Integer g-sptain ads.
783
* GNAT.Spitbol.Table_VString (g-sptavs.ads): GNAT Spitbol Table_VString g-sptavs ads.
784
* GNAT.SSE (g-sse.ads): GNAT SSE g-sse ads.
785
* GNAT.SSE.Vector_Types (g-ssvety.ads): GNAT SSE Vector_Types g-ssvety ads.
786
* GNAT.Strings (g-string.ads): GNAT Strings g-string ads.
787
* GNAT.String_Split (g-strspl.ads): GNAT String_Split g-strspl ads.
788
* GNAT.Table (g-table.ads): GNAT Table g-table ads.
789
* GNAT.Task_Lock (g-tasloc.ads): GNAT Task_Lock g-tasloc ads.
790
* GNAT.Time_Stamp (g-timsta.ads): GNAT Time_Stamp g-timsta ads.
791
* GNAT.Threads (g-thread.ads): GNAT Threads g-thread ads.
792
* GNAT.Traceback (g-traceb.ads): GNAT Traceback g-traceb ads.
793
* GNAT.Traceback.Symbolic (g-trasym.ads): GNAT Traceback Symbolic g-trasym ads.
794
* GNAT.UTF_32 (g-table.ads): GNAT UTF_32 g-table ads.
795
* GNAT.Wide_Spelling_Checker (g-u3spch.ads): GNAT Wide_Spelling_Checker g-u3spch ads.
796
* GNAT.Wide_Spelling_Checker (g-wispch.ads): GNAT Wide_Spelling_Checker g-wispch ads.
797
* GNAT.Wide_String_Split (g-wistsp.ads): GNAT Wide_String_Split g-wistsp ads.
798
* GNAT.Wide_Wide_Spelling_Checker (g-zspche.ads): GNAT Wide_Wide_Spelling_Checker g-zspche ads.
799
* GNAT.Wide_Wide_String_Split (g-zistsp.ads): GNAT Wide_Wide_String_Split g-zistsp ads.
800
* Interfaces.C.Extensions (i-cexten.ads): Interfaces C Extensions i-cexten ads.
801
* Interfaces.C.Streams (i-cstrea.ads): Interfaces C Streams i-cstrea ads.
802
* Interfaces.Packed_Decimal (i-pacdec.ads): Interfaces Packed_Decimal i-pacdec ads.
803
* Interfaces.VxWorks (i-vxwork.ads): Interfaces VxWorks i-vxwork ads.
804
* Interfaces.VxWorks.IO (i-vxwoio.ads): Interfaces VxWorks IO i-vxwoio ads.
805
* System.Address_Image (s-addima.ads): System Address_Image s-addima ads.
806
* System.Assertions (s-assert.ads): System Assertions s-assert ads.
807
* System.Atomic_Counters (s-atocou.ads): System Atomic_Counters s-atocou ads.
808
* System.Memory (s-memory.ads): System Memory s-memory ads.
809
* System.Multiprocessors (s-multip.ads): System Multiprocessors s-multip ads.
810
* System.Multiprocessors.Dispatching_Domains (s-mudido.ads): System Multiprocessors Dispatching_Domains s-mudido ads.
811
* System.Partition_Interface (s-parint.ads): System Partition_Interface s-parint ads.
812
* System.Pool_Global (s-pooglo.ads): System Pool_Global s-pooglo ads.
813
* System.Pool_Local (s-pooloc.ads): System Pool_Local s-pooloc ads.
814
* System.Restrictions (s-restri.ads): System Restrictions s-restri ads.
815
* System.Rident (s-rident.ads): System Rident s-rident ads.
816
* System.Strings.Stream_Ops (s-ststop.ads): System Strings Stream_Ops s-ststop ads.
817
* System.Unsigned_Types (s-unstyp.ads): System Unsigned_Types s-unstyp ads.
818
* System.Wch_Cnv (s-wchcnv.ads): System Wch_Cnv s-wchcnv ads.
819
* System.Wch_Con (s-wchcon.ads): System Wch_Con s-wchcon ads.
821
Interfacing to Other Languages
824
* Interfacing to C++::
825
* Interfacing to COBOL::
826
* Interfacing to Fortran::
827
* Interfacing to non-GNAT Ada code::
829
Implementation of Specific Ada Features
831
* Machine Code Insertions::
832
* GNAT Implementation of Tasking::
833
* GNAT Implementation of Shared Passive Packages::
834
* Code Generation for Array Aggregates::
835
* The Size of Discriminated Records with Default Discriminants::
836
* Strict Conformance to the Ada Reference Manual::
838
GNAT Implementation of Tasking
840
* Mapping Ada Tasks onto the Underlying Kernel Threads::
841
* Ensuring Compliance with the Real-Time Annex::
843
Code Generation for Array Aggregates
845
* Static constant aggregates with static bounds::
846
* Constant aggregates with unconstrained nominal types::
847
* Aggregates with static bounds::
848
* Aggregates with nonstatic bounds::
849
* Aggregates in assignment statements::
853
* pragma No_Run_Time::
855
* pragma Restricted_Run_Time::
857
* package System.Task_Info (s-tasinf.ads): package System Task_Info s-tasinf ads.
859
Compatibility and Porting Guide
861
* Writing Portable Fixed-Point Declarations::
862
* Compatibility with Ada 83::
863
* Compatibility between Ada 95 and Ada 2005::
864
* Implementation-dependent characteristics::
865
* Compatibility with Other Ada Systems::
866
* Representation Clauses::
867
* Compatibility with HP Ada 83::
869
Compatibility with Ada 83
871
* Legal Ada 83 programs that are illegal in Ada 95::
872
* More deterministic semantics::
873
* Changed semantics::
874
* Other language compatibility issues::
876
Implementation-dependent characteristics
878
* Implementation-defined pragmas::
879
* Implementation-defined attributes::
881
* Elaboration order::
882
* Target-specific aspects::
887
@node About This Guide,Implementation Defined Pragmas,Top,Top
888
@anchor{gnat_rm/about_this_guide about-this-guide}@anchor{2}@anchor{gnat_rm/about_this_guide doc}@anchor{3}@anchor{gnat_rm/about_this_guide gnat-reference-manual}@anchor{4}@anchor{gnat_rm/about_this_guide id1}@anchor{5}
889
@chapter About This Guide
893
This manual contains useful information in writing programs using the
894
GNAT compiler. It includes information on implementation dependent
895
characteristics of GNAT, including all the information required by
896
Annex M of the Ada language standard.
898
GNAT implements Ada 95, Ada 2005 and Ada 2012, and it may also be
899
invoked in Ada 83 compatibility mode.
900
By default, GNAT assumes Ada 2012,
901
but you can override with a compiler switch
902
to explicitly specify the language version.
903
(Please refer to the @emph{GNAT User's Guide} for details on these switches.)
904
Throughout this manual, references to 'Ada' without a year suffix
905
apply to all the Ada versions of the language.
907
Ada is designed to be highly portable.
908
In general, a program will have the same effect even when compiled by
909
different compilers on different platforms.
910
However, since Ada is designed to be used in a
911
wide variety of applications, it also contains a number of system
912
dependent features to be used in interfacing to the external world.
914
@geindex Implementation-dependent features
918
Note: Any program that makes use of implementation-dependent features
919
may be non-portable. You should follow good programming practice and
920
isolate and clearly document any sections of your program that make use
921
of these features in a non-portable manner.
924
* What This Reference Manual Contains::
926
* Related Information::
930
@node What This Reference Manual Contains,Conventions,,About This Guide
931
@anchor{gnat_rm/about_this_guide what-this-reference-manual-contains}@anchor{6}
932
@section What This Reference Manual Contains
935
This reference manual contains the following chapters:
941
@ref{7,,Implementation Defined Pragmas}, lists GNAT implementation-dependent
942
pragmas, which can be used to extend and enhance the functionality of the
946
@ref{8,,Implementation Defined Attributes}, lists GNAT
947
implementation-dependent attributes, which can be used to extend and
948
enhance the functionality of the compiler.
951
@ref{9,,Standard and Implementation Defined Restrictions}, lists GNAT
952
implementation-dependent restrictions, which can be used to extend and
953
enhance the functionality of the compiler.
956
@ref{a,,Implementation Advice}, provides information on generally
957
desirable behavior which are not requirements that all compilers must
958
follow since it cannot be provided on all systems, or which may be
959
undesirable on some systems.
962
@ref{b,,Implementation Defined Characteristics}, provides a guide to
963
minimizing implementation dependent features.
966
@ref{c,,Intrinsic Subprograms}, describes the intrinsic subprograms
967
implemented by GNAT, and how they can be imported into user
968
application programs.
971
@ref{d,,Representation Clauses and Pragmas}, describes in detail the
972
way that GNAT represents data, and in particular the exact set
973
of representation clauses and pragmas that is accepted.
976
@ref{e,,Standard Library Routines}, provides a listing of packages and a
977
brief description of the functionality that is provided by Ada's
978
extensive set of standard library routines as implemented by GNAT.
981
@ref{f,,The Implementation of Standard I/O}, details how the GNAT
982
implementation of the input-output facilities.
985
@ref{10,,The GNAT Library}, is a catalog of packages that complement
986
the Ada predefined library.
989
@ref{11,,Interfacing to Other Languages}, describes how programs
990
written in Ada using GNAT can be interfaced to other programming
994
@ref{12,,Specialized Needs Annexes}, describes the GNAT implementation of all
995
of the specialized needs annexes.
998
@ref{13,,Implementation of Specific Ada Features}, discusses issues related
999
to GNAT's implementation of machine code insertions, tasking, and several
1003
@ref{14,,Implementation of Ada 2012 Features}, describes the status of the
1004
GNAT implementation of the Ada 2012 language standard.
1007
@ref{15,,Obsolescent Features} documents implementation dependent features,
1008
including pragmas and attributes, which are considered obsolescent, since
1009
there are other preferred ways of achieving the same results. These
1010
obsolescent forms are retained for backwards compatibility.
1013
@ref{16,,Compatibility and Porting Guide} presents some guidelines for
1014
developing portable Ada code, describes the compatibility issues that
1015
may arise between GNAT and other Ada compilation systems (including those
1016
for Ada 83), and shows how GNAT can expedite porting applications
1017
developed in other Ada environments.
1020
@ref{1,,GNU Free Documentation License} contains the license for this document.
1023
@geindex Ada 95 Language Reference Manual
1025
@geindex Ada 2005 Language Reference Manual
1027
This reference manual assumes a basic familiarity with the Ada 95 language, as
1029
@cite{International Standard ANSI/ISO/IEC-8652:1995}.
1030
It does not require knowledge of the new features introduced by Ada 2005 or
1032
All three reference manuals are included in the GNAT documentation
1035
@node Conventions,Related Information,What This Reference Manual Contains,About This Guide
1036
@anchor{gnat_rm/about_this_guide conventions}@anchor{17}
1037
@section Conventions
1040
@geindex Conventions
1041
@geindex typographical
1043
@geindex Typographical conventions
1045
Following are examples of the typographical and graphic conventions used
1052
@cite{Functions}, @cite{utility program names}, @cite{standard names},
1068
[optional information or parameters]
1071
Examples are described by text
1074
and then shown this way.
1078
Commands that are entered by the user are shown as preceded by a prompt string
1079
comprising the @code{$} character followed by a space.
1082
@node Related Information,,Conventions,About This Guide
1083
@anchor{gnat_rm/about_this_guide related-information}@anchor{18}
1084
@section Related Information
1087
See the following documents for further information on GNAT:
1093
@cite{GNAT User's Guide for Native Platforms},
1094
which provides information on how to use the
1095
GNAT development environment.
1098
@cite{Ada 95 Reference Manual}, the Ada 95 programming language standard.
1101
@cite{Ada 95 Annotated Reference Manual}, which is an annotated version
1102
of the Ada 95 standard. The annotations describe
1103
detailed aspects of the design decision, and in particular contain useful
1104
sections on Ada 83 compatibility.
1107
@cite{Ada 2005 Reference Manual}, the Ada 2005 programming language standard.
1110
@cite{Ada 2005 Annotated Reference Manual}, which is an annotated version
1111
of the Ada 2005 standard. The annotations describe
1112
detailed aspects of the design decision.
1115
@cite{Ada 2012 Reference Manual}, the Ada 2012 programming language standard.
1118
@cite{DEC Ada@comma{} Technical Overview and Comparison on DIGITAL Platforms},
1119
which contains specific information on compatibility between GNAT and
1123
@cite{DEC Ada@comma{} Language Reference Manual}, part number AA-PYZAB-TK, which
1124
describes in detail the pragmas and attributes provided by the DEC Ada 83
1128
@node Implementation Defined Pragmas,Implementation Defined Aspects,About This Guide,Top
1129
@anchor{gnat_rm/implementation_defined_pragmas implementation-defined-pragmas}@anchor{7}@anchor{gnat_rm/implementation_defined_pragmas doc}@anchor{19}@anchor{gnat_rm/implementation_defined_pragmas id1}@anchor{1a}
1130
@chapter Implementation Defined Pragmas
1133
Ada defines a set of pragmas that can be used to supply additional
1134
information to the compiler. These language defined pragmas are
1135
implemented in GNAT and work as described in the Ada Reference Manual.
1137
In addition, Ada allows implementations to define additional pragmas
1138
whose meaning is defined by the implementation. GNAT provides a number
1139
of these implementation-defined pragmas, which can be used to extend
1140
and enhance the functionality of the compiler. This section of the GNAT
1141
Reference Manual describes these additional pragmas.
1143
Note that any program using these pragmas might not be portable to other
1144
compilers (although GNAT implements this set of pragmas on all
1145
platforms). Therefore if portability to other compilers is an important
1146
consideration, the use of these pragmas should be minimized.
1149
* Pragma Abort_Defer::
1150
* Pragma Abstract_State::
1157
* Pragma Allow_Integer_Address::
1160
* Pragma Assert_And_Cut::
1161
* Pragma Assertion_Policy::
1163
* Pragma Assume_No_Invalid_Values::
1164
* Pragma Async_Readers::
1165
* Pragma Async_Writers::
1166
* Pragma Attribute_Definition::
1167
* Pragma C_Pass_By_Copy::
1169
* Pragma Check_Float_Overflow::
1170
* Pragma Check_Name::
1171
* Pragma Check_Policy::
1173
* Pragma Common_Object::
1174
* Pragma Compile_Time_Error::
1175
* Pragma Compile_Time_Warning::
1176
* Pragma Compiler_Unit::
1177
* Pragma Compiler_Unit_Warning::
1178
* Pragma Complete_Representation::
1179
* Pragma Complex_Representation::
1180
* Pragma Component_Alignment::
1181
* Pragma Constant_After_Elaboration::
1182
* Pragma Contract_Cases::
1183
* Pragma Convention_Identifier::
1184
* Pragma CPP_Class::
1185
* Pragma CPP_Constructor::
1186
* Pragma CPP_Virtual::
1187
* Pragma CPP_Vtable::
1189
* Pragma Default_Initial_Condition::
1191
* Pragma Debug_Policy::
1192
* Pragma Default_Scalar_Storage_Order::
1193
* Pragma Default_Storage_Pool::
1195
* Pragma Detect_Blocking::
1196
* Pragma Disable_Atomic_Synchronization::
1197
* Pragma Dispatching_Domain::
1198
* Pragma Effective_Reads::
1199
* Pragma Effective_Writes::
1200
* Pragma Elaboration_Checks::
1201
* Pragma Eliminate::
1202
* Pragma Enable_Atomic_Synchronization::
1203
* Pragma Export_Function::
1204
* Pragma Export_Object::
1205
* Pragma Export_Procedure::
1206
* Pragma Export_Value::
1207
* Pragma Export_Valued_Procedure::
1208
* Pragma Extend_System::
1209
* Pragma Extensions_Allowed::
1210
* Pragma Extensions_Visible::
1212
* Pragma External_Name_Casing::
1213
* Pragma Fast_Math::
1214
* Pragma Favor_Top_Level::
1215
* Pragma Finalize_Storage_Only::
1216
* Pragma Float_Representation::
1220
* Pragma Ignore_Pragma::
1221
* Pragma Implementation_Defined::
1222
* Pragma Implemented::
1223
* Pragma Implicit_Packing::
1224
* Pragma Import_Function::
1225
* Pragma Import_Object::
1226
* Pragma Import_Procedure::
1227
* Pragma Import_Valued_Procedure::
1228
* Pragma Independent::
1229
* Pragma Independent_Components::
1230
* Pragma Initial_Condition::
1231
* Pragma Initialize_Scalars::
1232
* Pragma Initializes::
1233
* Pragma Inline_Always::
1234
* Pragma Inline_Generic::
1235
* Pragma Interface::
1236
* Pragma Interface_Name::
1237
* Pragma Interrupt_Handler::
1238
* Pragma Interrupt_State::
1239
* Pragma Invariant::
1240
* Pragma Keep_Names::
1242
* Pragma Link_With::
1243
* Pragma Linker_Alias::
1244
* Pragma Linker_Constructor::
1245
* Pragma Linker_Destructor::
1246
* Pragma Linker_Section::
1247
* Pragma Lock_Free::
1248
* Pragma Loop_Invariant::
1249
* Pragma Loop_Optimize::
1250
* Pragma Loop_Variant::
1251
* Pragma Machine_Attribute::
1253
* Pragma Main_Storage::
1255
* Pragma No_Elaboration_Code_All::
1256
* Pragma No_Inline::
1257
* Pragma No_Return::
1258
* Pragma No_Run_Time::
1259
* Pragma No_Strict_Aliasing::
1260
* Pragma No_Tagged_Streams::
1261
* Pragma Normalize_Scalars::
1262
* Pragma Obsolescent::
1263
* Pragma Optimize_Alignment::
1265
* Pragma Overflow_Mode::
1266
* Pragma Overriding_Renamings::
1267
* Pragma Partition_Elaboration_Policy::
1270
* Pragma Persistent_BSS::
1273
* Pragma Postcondition::
1274
* Pragma Post_Class::
1276
* Pragma Precondition::
1277
* Pragma Predicate::
1278
* Pragma Predicate_Failure::
1279
* Pragma Preelaborable_Initialization::
1280
* Pragma Prefix_Exception_Messages::
1281
* Pragma Pre_Class::
1282
* Pragma Priority_Specific_Dispatching::
1284
* Pragma Profile_Warnings::
1285
* Pragma Propagate_Exceptions::
1286
* Pragma Provide_Shift_Operators::
1287
* Pragma Psect_Object::
1288
* Pragma Pure_Function::
1290
* Pragma Ravenscar::
1291
* Pragma Refined_Depends::
1292
* Pragma Refined_Global::
1293
* Pragma Refined_Post::
1294
* Pragma Refined_State::
1295
* Pragma Relative_Deadline::
1296
* Pragma Remote_Access_Type::
1297
* Pragma Restricted_Run_Time::
1298
* Pragma Restriction_Warnings::
1299
* Pragma Reviewable::
1300
* Pragma Share_Generic::
1302
* Pragma Short_Circuit_And_Or::
1303
* Pragma Short_Descriptors::
1304
* Pragma Simple_Storage_Pool_Type::
1305
* Pragma Source_File_Name::
1306
* Pragma Source_File_Name_Project::
1307
* Pragma Source_Reference::
1308
* Pragma SPARK_Mode::
1309
* Pragma Static_Elaboration_Desired::
1310
* Pragma Stream_Convert::
1311
* Pragma Style_Checks::
1314
* Pragma Suppress_All::
1315
* Pragma Suppress_Debug_Info::
1316
* Pragma Suppress_Exception_Locations::
1317
* Pragma Suppress_Initialization::
1318
* Pragma Task_Name::
1319
* Pragma Task_Storage::
1320
* Pragma Test_Case::
1321
* Pragma Thread_Local_Storage::
1322
* Pragma Time_Slice::
1324
* Pragma Type_Invariant::
1325
* Pragma Type_Invariant_Class::
1326
* Pragma Unchecked_Union::
1327
* Pragma Unevaluated_Use_Of_Old::
1328
* Pragma Unimplemented_Unit::
1329
* Pragma Universal_Aliasing::
1330
* Pragma Universal_Data::
1331
* Pragma Unmodified::
1332
* Pragma Unreferenced::
1333
* Pragma Unreferenced_Objects::
1334
* Pragma Unreserve_All_Interrupts::
1335
* Pragma Unsuppress::
1336
* Pragma Use_VADS_Size::
1337
* Pragma Validity_Checks::
1339
* Pragma Volatile_Full_Access::
1340
* Pragma Volatile_Function::
1341
* Pragma Warning_As_Error::
1343
* Pragma Weak_External::
1344
* Pragma Wide_Character_Encoding::
1348
@node Pragma Abort_Defer,Pragma Abstract_State,,Implementation Defined Pragmas
1349
@anchor{gnat_rm/implementation_defined_pragmas pragma-abort-defer}@anchor{1b}
1350
@section Pragma Abort_Defer
1353
@geindex Deferring aborts
1361
This pragma must appear at the start of the statement sequence of a
1362
handled sequence of statements (right after the @cite{begin}). It has
1363
the effect of deferring aborts for the sequence of statements (but not
1364
for the declarations or handlers, if any, associated with this statement
1367
@node Pragma Abstract_State,Pragma Ada_83,Pragma Abort_Defer,Implementation Defined Pragmas
1368
@anchor{gnat_rm/implementation_defined_pragmas pragma-abstract-state}@anchor{1c}
1369
@section Pragma Abstract_State
1375
pragma Abstract_State (ABSTRACT_STATE_LIST);
1377
ABSTRACT_STATE_LIST ::=
1379
| STATE_NAME_WITH_OPTIONS
1380
| (STATE_NAME_WITH_OPTIONS @{, STATE_NAME_WITH_OPTIONS@} )
1382
STATE_NAME_WITH_OPTIONS ::=
1384
| (STATE_NAME with OPTION_LIST)
1386
OPTION_LIST ::= OPTION @{, OPTION@}
1392
SIMPLE_OPTION ::= Ghost | Synchronous
1394
NAME_VALUE_OPTION ::=
1395
Part_Of => ABSTRACT_STATE
1396
| External [=> EXTERNAL_PROPERTY_LIST]
1398
EXTERNAL_PROPERTY_LIST ::=
1400
| (EXTERNAL_PROPERTY @{, EXTERNAL_PROPERTY@} )
1402
EXTERNAL_PROPERTY ::=
1403
Async_Readers [=> boolean_EXPRESSION]
1404
| Async_Writers [=> boolean_EXPRESSION]
1405
| Effective_Reads [=> boolean_EXPRESSION]
1406
| Effective_Writes [=> boolean_EXPRESSION]
1407
others => boolean_EXPRESSION
1409
STATE_NAME ::= defining_identifier
1411
ABSTRACT_STATE ::= name
1414
For the semantics of this pragma, see the entry for aspect @cite{Abstract_State} in
1415
the SPARK 2014 Reference Manual, section 7.1.4.
1417
@node Pragma Ada_83,Pragma Ada_95,Pragma Abstract_State,Implementation Defined Pragmas
1418
@anchor{gnat_rm/implementation_defined_pragmas pragma-ada-83}@anchor{1d}
1419
@section Pragma Ada_83
1428
A configuration pragma that establishes Ada 83 mode for the unit to
1429
which it applies, regardless of the mode set by the command line
1430
switches. In Ada 83 mode, GNAT attempts to be as compatible with
1431
the syntax and semantics of Ada 83, as defined in the original Ada
1432
83 Reference Manual as possible. In particular, the keywords added by Ada 95
1433
and Ada 2005 are not recognized, optional package bodies are allowed,
1434
and generics may name types with unknown discriminants without using
1435
the @cite{(<>)} notation. In addition, some but not all of the additional
1436
restrictions of Ada 83 are enforced.
1438
Ada 83 mode is intended for two purposes. Firstly, it allows existing
1439
Ada 83 code to be compiled and adapted to GNAT with less effort.
1440
Secondly, it aids in keeping code backwards compatible with Ada 83.
1441
However, there is no guarantee that code that is processed correctly
1442
by GNAT in Ada 83 mode will in fact compile and execute with an Ada
1443
83 compiler, since GNAT does not enforce all the additional checks
1446
@node Pragma Ada_95,Pragma Ada_05,Pragma Ada_83,Implementation Defined Pragmas
1447
@anchor{gnat_rm/implementation_defined_pragmas pragma-ada-95}@anchor{1e}
1448
@section Pragma Ada_95
1457
A configuration pragma that establishes Ada 95 mode for the unit to which
1458
it applies, regardless of the mode set by the command line switches.
1459
This mode is set automatically for the @cite{Ada} and @cite{System}
1460
packages and their children, so you need not specify it in these
1461
contexts. This pragma is useful when writing a reusable component that
1462
itself uses Ada 95 features, but which is intended to be usable from
1463
either Ada 83 or Ada 95 programs.
1465
@node Pragma Ada_05,Pragma Ada_2005,Pragma Ada_95,Implementation Defined Pragmas
1466
@anchor{gnat_rm/implementation_defined_pragmas pragma-ada-05}@anchor{1f}
1467
@section Pragma Ada_05
1474
pragma Ada_05 (local_NAME);
1477
A configuration pragma that establishes Ada 2005 mode for the unit to which
1478
it applies, regardless of the mode set by the command line switches.
1479
This pragma is useful when writing a reusable component that
1480
itself uses Ada 2005 features, but which is intended to be usable from
1481
either Ada 83 or Ada 95 programs.
1483
The one argument form (which is not a configuration pragma)
1484
is used for managing the transition from
1485
Ada 95 to Ada 2005 in the run-time library. If an entity is marked
1486
as Ada_2005 only, then referencing the entity in Ada_83 or Ada_95
1487
mode will generate a warning. In addition, in Ada_83 or Ada_95
1488
mode, a preference rule is established which does not choose
1489
such an entity unless it is unambiguously specified. This avoids
1490
extra subprograms marked this way from generating ambiguities in
1491
otherwise legal pre-Ada_2005 programs. The one argument form is
1492
intended for exclusive use in the GNAT run-time library.
1494
@node Pragma Ada_2005,Pragma Ada_12,Pragma Ada_05,Implementation Defined Pragmas
1495
@anchor{gnat_rm/implementation_defined_pragmas pragma-ada-2005}@anchor{20}
1496
@section Pragma Ada_2005
1505
This configuration pragma is a synonym for pragma Ada_05 and has the
1506
same syntax and effect.
1508
@node Pragma Ada_12,Pragma Ada_2012,Pragma Ada_2005,Implementation Defined Pragmas
1509
@anchor{gnat_rm/implementation_defined_pragmas pragma-ada-12}@anchor{21}
1510
@section Pragma Ada_12
1517
pragma Ada_12 (local_NAME);
1520
A configuration pragma that establishes Ada 2012 mode for the unit to which
1521
it applies, regardless of the mode set by the command line switches.
1522
This mode is set automatically for the @cite{Ada} and @cite{System}
1523
packages and their children, so you need not specify it in these
1524
contexts. This pragma is useful when writing a reusable component that
1525
itself uses Ada 2012 features, but which is intended to be usable from
1526
Ada 83, Ada 95, or Ada 2005 programs.
1528
The one argument form, which is not a configuration pragma,
1529
is used for managing the transition from Ada
1530
2005 to Ada 2012 in the run-time library. If an entity is marked
1531
as Ada_201 only, then referencing the entity in any pre-Ada_2012
1532
mode will generate a warning. In addition, in any pre-Ada_2012
1533
mode, a preference rule is established which does not choose
1534
such an entity unless it is unambiguously specified. This avoids
1535
extra subprograms marked this way from generating ambiguities in
1536
otherwise legal pre-Ada_2012 programs. The one argument form is
1537
intended for exclusive use in the GNAT run-time library.
1539
@node Pragma Ada_2012,Pragma Allow_Integer_Address,Pragma Ada_12,Implementation Defined Pragmas
1540
@anchor{gnat_rm/implementation_defined_pragmas pragma-ada-2012}@anchor{22}
1541
@section Pragma Ada_2012
1550
This configuration pragma is a synonym for pragma Ada_12 and has the
1551
same syntax and effect.
1553
@node Pragma Allow_Integer_Address,Pragma Annotate,Pragma Ada_2012,Implementation Defined Pragmas
1554
@anchor{gnat_rm/implementation_defined_pragmas pragma-allow-integer-address}@anchor{23}
1555
@section Pragma Allow_Integer_Address
1561
pragma Allow_Integer_Address;
1564
In almost all versions of GNAT, @cite{System.Address} is a private
1565
type in accordance with the implementation advice in the RM. This
1566
means that integer values,
1567
in particular integer literals, are not allowed as address values.
1568
If the configuration pragma
1569
@cite{Allow_Integer_Address} is given, then integer expressions may
1570
be used anywhere a value of type @cite{System.Address} is required.
1571
The effect is to introduce an implicit unchecked conversion from the
1572
integer value to type @cite{System.Address}. The reverse case of using
1573
an address where an integer type is required is handled analogously.
1574
The following example compiles without errors:
1577
pragma Allow_Integer_Address;
1578
with System; use System;
1579
package AddrAsInt is
1582
for X'Address use 16#1240#;
1583
for Y use at 16#3230#;
1584
m : Address := 16#4000#;
1585
n : constant Address := 4000;
1586
p : constant Address := Address (X + Y);
1587
v : Integer := y'Address;
1588
w : constant Integer := Integer (Y'Address);
1589
type R is new integer;
1592
for Z'Address use RR;
1596
Note that pragma @cite{Allow_Integer_Address} is ignored if @cite{System.Address}
1597
is not a private type. In implementations of @cite{GNAT} where
1598
System.Address is a visible integer type,
1599
this pragma serves no purpose but is ignored
1600
rather than rejected to allow common sets of sources to be used
1601
in the two situations.
1603
@node Pragma Annotate,Pragma Assert,Pragma Allow_Integer_Address,Implementation Defined Pragmas
1604
@anchor{gnat_rm/implementation_defined_pragmas pragma-annotate}@anchor{24}
1605
@section Pragma Annotate
1611
pragma Annotate (IDENTIFIER [, IDENTIFIER @{, ARG@}] [, entity => local_NAME]);
1613
ARG ::= NAME | EXPRESSION
1616
This pragma is used to annotate programs. @cite{identifier} identifies
1617
the type of annotation. GNAT verifies that it is an identifier, but does
1618
not otherwise analyze it. The second optional identifier is also left
1619
unanalyzed, and by convention is used to control the action of the tool to
1620
which the annotation is addressed. The remaining @cite{arg} arguments
1621
can be either string literals or more generally expressions.
1622
String literals are assumed to be either of type
1623
@cite{Standard.String} or else @cite{Wide_String} or @cite{Wide_Wide_String}
1624
depending on the character literals they contain.
1625
All other kinds of arguments are analyzed as expressions, and must be
1626
unambiguous. The last argument if present must have the identifier
1627
@cite{Entity} and GNAT verifies that a local name is given.
1629
The analyzed pragma is retained in the tree, but not otherwise processed
1630
by any part of the GNAT compiler, except to generate corresponding note
1631
lines in the generated ALI file. For the format of these note lines, see
1632
the compiler source file lib-writ.ads. This pragma is intended for use by
1633
external tools, including ASIS. The use of pragma Annotate does not
1634
affect the compilation process in any way. This pragma may be used as
1635
a configuration pragma.
1637
@node Pragma Assert,Pragma Assert_And_Cut,Pragma Annotate,Implementation Defined Pragmas
1638
@anchor{gnat_rm/implementation_defined_pragmas pragma-assert}@anchor{25}
1639
@section Pragma Assert
1647
[, string_EXPRESSION]);
1650
The effect of this pragma depends on whether the corresponding command
1651
line switch is set to activate assertions. The pragma expands into code
1652
equivalent to the following:
1655
if assertions-enabled then
1656
if not boolean_EXPRESSION then
1657
System.Assertions.Raise_Assert_Failure
1658
(string_EXPRESSION);
1663
The string argument, if given, is the message that will be associated
1664
with the exception occurrence if the exception is raised. If no second
1665
argument is given, the default message is @cite{file}:@cite{nnn},
1666
where @cite{file} is the name of the source file containing the assert,
1667
and @cite{nnn} is the line number of the assert.
1669
Note that, as with the @cite{if} statement to which it is equivalent, the
1670
type of the expression is either @cite{Standard.Boolean}, or any type derived
1671
from this standard type.
1673
Assert checks can be either checked or ignored. By default they are ignored.
1674
They will be checked if either the command line switch @emph{-gnata} is
1675
used, or if an @cite{Assertion_Policy} or @cite{Check_Policy} pragma is used
1676
to enable @cite{Assert_Checks}.
1678
If assertions are ignored, then there
1679
is no run-time effect (and in particular, any side effects from the
1680
expression will not occur at run time). (The expression is still
1681
analyzed at compile time, and may cause types to be frozen if they are
1682
mentioned here for the first time).
1684
If assertions are checked, then the given expression is tested, and if
1685
it is @cite{False} then @cite{System.Assertions.Raise_Assert_Failure} is called
1686
which results in the raising of @cite{Assert_Failure} with the given message.
1688
You should generally avoid side effects in the expression arguments of
1689
this pragma, because these side effects will turn on and off with the
1690
setting of the assertions mode, resulting in assertions that have an
1691
effect on the program. However, the expressions are analyzed for
1692
semantic correctness whether or not assertions are enabled, so turning
1693
assertions on and off cannot affect the legality of a program.
1695
Note that the implementation defined policy @cite{DISABLE}, given in a
1696
pragma @cite{Assertion_Policy}, can be used to suppress this semantic analysis.
1698
Note: this is a standard language-defined pragma in versions
1699
of Ada from 2005 on. In GNAT, it is implemented in all versions
1700
of Ada, and the DISABLE policy is an implementation-defined
1703
@node Pragma Assert_And_Cut,Pragma Assertion_Policy,Pragma Assert,Implementation Defined Pragmas
1704
@anchor{gnat_rm/implementation_defined_pragmas pragma-assert-and-cut}@anchor{26}
1705
@section Pragma Assert_And_Cut
1711
pragma Assert_And_Cut (
1713
[, string_EXPRESSION]);
1716
The effect of this pragma is identical to that of pragma @cite{Assert},
1717
except that in an @cite{Assertion_Policy} pragma, the identifier
1718
@cite{Assert_And_Cut} is used to control whether it is ignored or checked
1721
The intention is that this be used within a subprogram when the
1722
given test expresion sums up all the work done so far in the
1723
subprogram, so that the rest of the subprogram can be verified
1724
(informally or formally) using only the entry preconditions,
1725
and the expression in this pragma. This allows dividing up
1726
a subprogram into sections for the purposes of testing or
1727
formal verification. The pragma also serves as useful
1730
@node Pragma Assertion_Policy,Pragma Assume,Pragma Assert_And_Cut,Implementation Defined Pragmas
1731
@anchor{gnat_rm/implementation_defined_pragmas pragma-assertion-policy}@anchor{27}
1732
@section Pragma Assertion_Policy
1738
pragma Assertion_Policy (CHECK | DISABLE | IGNORE);
1740
pragma Assertion_Policy (
1741
ASSERTION_KIND => POLICY_IDENTIFIER
1742
@{, ASSERTION_KIND => POLICY_IDENTIFIER@});
1744
ASSERTION_KIND ::= RM_ASSERTION_KIND | ID_ASSERTION_KIND
1746
RM_ASSERTION_KIND ::= Assert |
1754
Type_Invariant'Class
1756
ID_ASSERTION_KIND ::= Assertions |
1769
Statement_Assertions
1771
POLICY_IDENTIFIER ::= Check | Disable | Ignore
1774
This is a standard Ada 2012 pragma that is available as an
1775
implementation-defined pragma in earlier versions of Ada.
1776
The assertion kinds @cite{RM_ASSERTION_KIND} are those defined in
1777
the Ada standard. The assertion kinds @cite{ID_ASSERTION_KIND}
1778
are implementation defined additions recognized by the GNAT compiler.
1780
The pragma applies in both cases to pragmas and aspects with matching
1781
names, e.g. @cite{Pre} applies to the Pre aspect, and @cite{Precondition}
1782
applies to both the @cite{Precondition} pragma
1783
and the aspect @cite{Precondition}. Note that the identifiers for
1784
pragmas Pre_Class and Post_Class are Pre'Class and Post'Class (not
1785
Pre_Class and Post_Class), since these pragmas are intended to be
1786
identical to the corresponding aspects).
1788
If the policy is @cite{CHECK}, then assertions are enabled, i.e.
1789
the corresponding pragma or aspect is activated.
1790
If the policy is @cite{IGNORE}, then assertions are ignored, i.e.
1791
the corresponding pragma or aspect is deactivated.
1792
This pragma overrides the effect of the @emph{-gnata} switch on the
1795
The implementation defined policy @cite{DISABLE} is like
1796
@cite{IGNORE} except that it completely disables semantic
1797
checking of the corresponding pragma or aspect. This is
1798
useful when the pragma or aspect argument references subprograms
1799
in a with'ed package which is replaced by a dummy package
1800
for the final build.
1802
The implementation defined assertion kind @cite{Assertions} applies to all
1803
assertion kinds. The form with no assertion kind given implies this
1804
choice, so it applies to all assertion kinds (RM defined, and
1805
implementation defined).
1807
The implementation defined assertion kind @cite{Statement_Assertions}
1808
applies to @cite{Assert}, @cite{Assert_And_Cut},
1809
@cite{Assume}, @cite{Loop_Invariant}, and @cite{Loop_Variant}.
1811
@node Pragma Assume,Pragma Assume_No_Invalid_Values,Pragma Assertion_Policy,Implementation Defined Pragmas
1812
@anchor{gnat_rm/implementation_defined_pragmas pragma-assume}@anchor{28}
1813
@section Pragma Assume
1821
[, string_EXPRESSION]);
1824
The effect of this pragma is identical to that of pragma @cite{Assert},
1825
except that in an @cite{Assertion_Policy} pragma, the identifier
1826
@cite{Assume} is used to control whether it is ignored or checked
1829
The intention is that this be used for assumptions about the
1830
external environment. So you cannot expect to verify formally
1831
or informally that the condition is met, this must be
1832
established by examining things outside the program itself.
1833
For example, we may have code that depends on the size of
1834
@cite{Long_Long_Integer} being at least 64. So we could write:
1837
pragma Assume (Long_Long_Integer'Size >= 64);
1840
This assumption cannot be proved from the program itself,
1841
but it acts as a useful run-time check that the assumption
1842
is met, and documents the need to ensure that it is met by
1843
reference to information outside the program.
1845
@node Pragma Assume_No_Invalid_Values,Pragma Async_Readers,Pragma Assume,Implementation Defined Pragmas
1846
@anchor{gnat_rm/implementation_defined_pragmas pragma-assume-no-invalid-values}@anchor{29}
1847
@section Pragma Assume_No_Invalid_Values
1850
@geindex Invalid representations
1852
@geindex Invalid values
1857
pragma Assume_No_Invalid_Values (On | Off);
1860
This is a configuration pragma that controls the assumptions made by the
1861
compiler about the occurrence of invalid representations (invalid values)
1864
The default behavior (corresponding to an Off argument for this pragma), is
1865
to assume that values may in general be invalid unless the compiler can
1866
prove they are valid. Consider the following example:
1869
V1 : Integer range 1 .. 10;
1870
V2 : Integer range 11 .. 20;
1872
for J in V2 .. V1 loop
1877
if V1 and V2 have valid values, then the loop is known at compile
1878
time not to execute since the lower bound must be greater than the
1879
upper bound. However in default mode, no such assumption is made,
1880
and the loop may execute. If @cite{Assume_No_Invalid_Values (On)}
1881
is given, the compiler will assume that any occurrence of a variable
1882
other than in an explicit @cite{'Valid} test always has a valid
1883
value, and the loop above will be optimized away.
1885
The use of @cite{Assume_No_Invalid_Values (On)} is appropriate if
1886
you know your code is free of uninitialized variables and other
1887
possible sources of invalid representations, and may result in
1888
more efficient code. A program that accesses an invalid representation
1889
with this pragma in effect is erroneous, so no guarantees can be made
1892
It is peculiar though permissible to use this pragma in conjunction
1893
with validity checking (-gnatVa). In such cases, accessing invalid
1894
values will generally give an exception, though formally the program
1895
is erroneous so there are no guarantees that this will always be the
1896
case, and it is recommended that these two options not be used together.
1898
@node Pragma Async_Readers,Pragma Async_Writers,Pragma Assume_No_Invalid_Values,Implementation Defined Pragmas
1899
@anchor{gnat_rm/implementation_defined_pragmas pragma-async-readers}@anchor{2a}
1900
@section Pragma Async_Readers
1906
pragma Asynch_Readers [ (boolean_EXPRESSION) ];
1909
For the semantics of this pragma, see the entry for aspect @cite{Async_Readers} in
1910
the SPARK 2014 Reference Manual, section 7.1.2.
1912
@node Pragma Async_Writers,Pragma Attribute_Definition,Pragma Async_Readers,Implementation Defined Pragmas
1913
@anchor{gnat_rm/implementation_defined_pragmas pragma-async-writers}@anchor{2b}
1914
@section Pragma Async_Writers
1920
pragma Asynch_Writers [ (boolean_EXPRESSION) ];
1923
For the semantics of this pragma, see the entry for aspect @cite{Async_Writers} in
1924
the SPARK 2014 Reference Manual, section 7.1.2.
1926
@node Pragma Attribute_Definition,Pragma C_Pass_By_Copy,Pragma Async_Writers,Implementation Defined Pragmas
1927
@anchor{gnat_rm/implementation_defined_pragmas pragma-attribute-definition}@anchor{2c}
1928
@section Pragma Attribute_Definition
1934
pragma Attribute_Definition
1935
([Attribute =>] ATTRIBUTE_DESIGNATOR,
1936
[Entity =>] LOCAL_NAME,
1937
[Expression =>] EXPRESSION | NAME);
1940
If @cite{Attribute} is a known attribute name, this pragma is equivalent to
1941
the attribute definition clause:
1944
for Entity'Attribute use Expression;
1947
If @cite{Attribute} is not a recognized attribute name, the pragma is
1948
ignored, and a warning is emitted. This allows source
1949
code to be written that takes advantage of some new attribute, while remaining
1950
compilable with earlier compilers.
1952
@node Pragma C_Pass_By_Copy,Pragma Check,Pragma Attribute_Definition,Implementation Defined Pragmas
1953
@anchor{gnat_rm/implementation_defined_pragmas pragma-c-pass-by-copy}@anchor{2d}
1954
@section Pragma C_Pass_By_Copy
1957
@geindex Passing by copy
1962
pragma C_Pass_By_Copy
1963
([Max_Size =>] static_integer_EXPRESSION);
1966
Normally the default mechanism for passing C convention records to C
1967
convention subprograms is to pass them by reference, as suggested by RM
1968
B.3(69). Use the configuration pragma @cite{C_Pass_By_Copy} to change
1969
this default, by requiring that record formal parameters be passed by
1970
copy if all of the following conditions are met:
1976
The size of the record type does not exceed the value specified for
1980
The record type has @cite{Convention C}.
1983
The formal parameter has this record type, and the subprogram has a
1984
foreign (non-Ada) convention.
1987
If these conditions are met the argument is passed by copy; i.e., in a
1988
manner consistent with what C expects if the corresponding formal in the
1989
C prototype is a struct (rather than a pointer to a struct).
1991
You can also pass records by copy by specifying the convention
1992
@cite{C_Pass_By_Copy} for the record type, or by using the extended
1993
@cite{Import} and @cite{Export} pragmas, which allow specification of
1994
passing mechanisms on a parameter by parameter basis.
1996
@node Pragma Check,Pragma Check_Float_Overflow,Pragma C_Pass_By_Copy,Implementation Defined Pragmas
1997
@anchor{gnat_rm/implementation_defined_pragmas pragma-check}@anchor{2e}
1998
@section Pragma Check
2003
@geindex Named assertions
2009
[Name =>] CHECK_KIND,
2010
[Check =>] Boolean_EXPRESSION
2011
[, [Message =>] string_EXPRESSION] );
2013
CHECK_KIND ::= IDENTIFIER |
2016
Type_Invariant'Class |
2020
This pragma is similar to the predefined pragma @cite{Assert} except that an
2021
extra identifier argument is present. In conjunction with pragma
2022
@cite{Check_Policy}, this can be used to define groups of assertions that can
2023
be independently controlled. The identifier @cite{Assertion} is special, it
2024
refers to the normal set of pragma @cite{Assert} statements.
2026
Checks introduced by this pragma are normally deactivated by default. They can
2027
be activated either by the command line option @emph{-gnata}, which turns on
2028
all checks, or individually controlled using pragma @cite{Check_Policy}.
2030
The identifiers @cite{Assertions} and @cite{Statement_Assertions} are not
2031
permitted as check kinds, since this would cause confusion with the use
2032
of these identifiers in @cite{Assertion_Policy} and @cite{Check_Policy}
2033
pragmas, where they are used to refer to sets of assertions.
2035
@node Pragma Check_Float_Overflow,Pragma Check_Name,Pragma Check,Implementation Defined Pragmas
2036
@anchor{gnat_rm/implementation_defined_pragmas pragma-check-float-overflow}@anchor{2f}
2037
@section Pragma Check_Float_Overflow
2040
@geindex Floating-point overflow
2045
pragma Check_Float_Overflow;
2048
In Ada, the predefined floating-point types (@cite{Short_Float},
2049
@cite{Float}, @cite{Long_Float}, @cite{Long_Long_Float}) are
2050
defined to be @emph{unconstrained}. This means that even though each
2051
has a well-defined base range, an operation that delivers a result
2052
outside this base range is not required to raise an exception.
2053
This implementation permission accommodates the notion
2054
of infinities in IEEE floating-point, and corresponds to the
2055
efficient execution mode on most machines. GNAT will not raise
2056
overflow exceptions on these machines; instead it will generate
2057
infinities and NaN's as defined in the IEEE standard.
2059
Generating infinities, although efficient, is not always desirable.
2060
Often the preferable approach is to check for overflow, even at the
2061
(perhaps considerable) expense of run-time performance.
2062
This can be accomplished by defining your own constrained floating-point subtypes -- i.e., by supplying explicit
2063
range constraints -- and indeed such a subtype
2064
can have the same base range as its base type. For example:
2067
subtype My_Float is Float range Float'Range;
2070
Here @cite{My_Float} has the same range as
2071
@cite{Float} but is constrained, so operations on
2072
@cite{My_Float} values will be checked for overflow
2075
This style will achieve the desired goal, but
2076
it is often more convenient to be able to simply use
2077
the standard predefined floating-point types as long
2078
as overflow checking could be guaranteed.
2079
The @cite{Check_Float_Overflow}
2080
configuration pragma achieves this effect. If a unit is compiled
2081
subject to this configuration pragma, then all operations
2082
on predefined floating-point types including operations on
2083
base types of these floating-point types will be treated as
2084
though those types were constrained, and overflow checks
2085
will be generated. The @cite{Constraint_Error}
2086
exception is raised if the result is out of range.
2088
This mode can also be set by use of the compiler
2089
switch @emph{-gnateF}.
2091
@node Pragma Check_Name,Pragma Check_Policy,Pragma Check_Float_Overflow,Implementation Defined Pragmas
2092
@anchor{gnat_rm/implementation_defined_pragmas pragma-check-name}@anchor{30}
2093
@section Pragma Check_Name
2096
@geindex Defining check names
2098
@geindex Check names
2104
pragma Check_Name (check_name_IDENTIFIER);
2107
This is a configuration pragma that defines a new implementation
2108
defined check name (unless IDENTIFIER matches one of the predefined
2109
check names, in which case the pragma has no effect). Check names
2110
are global to a partition, so if two or more configuration pragmas
2111
are present in a partition mentioning the same name, only one new
2112
check name is introduced.
2114
An implementation defined check name introduced with this pragma may
2115
be used in only three contexts: @cite{pragma Suppress},
2116
@cite{pragma Unsuppress},
2117
and as the prefix of a @cite{Check_Name'Enabled} attribute reference. For
2118
any of these three cases, the check name must be visible. A check
2119
name is visible if it is in the configuration pragmas applying to
2120
the current unit, or if it appears at the start of any unit that
2121
is part of the dependency set of the current unit (e.g., units that
2122
are mentioned in @cite{with} clauses).
2124
Check names introduced by this pragma are subject to control by compiler
2125
switches (in particular -gnatp) in the usual manner.
2127
@node Pragma Check_Policy,Pragma Comment,Pragma Check_Name,Implementation Defined Pragmas
2128
@anchor{gnat_rm/implementation_defined_pragmas pragma-check-policy}@anchor{31}
2129
@section Pragma Check_Policy
2132
@geindex Controlling assertions
2137
@geindex Check pragma control
2139
@geindex Named assertions
2145
([Name =>] CHECK_KIND,
2146
[Policy =>] POLICY_IDENTIFIER);
2148
pragma Check_Policy (
2149
CHECK_KIND => POLICY_IDENTIFIER
2150
@{, CHECK_KIND => POLICY_IDENTIFIER@});
2152
ASSERTION_KIND ::= RM_ASSERTION_KIND | ID_ASSERTION_KIND
2154
CHECK_KIND ::= IDENTIFIER |
2157
Type_Invariant'Class |
2160
The identifiers Name and Policy are not allowed as CHECK_KIND values. This
2161
avoids confusion between the two possible syntax forms for this pragma.
2163
POLICY_IDENTIFIER ::= ON | OFF | CHECK | DISABLE | IGNORE
2166
This pragma is used to set the checking policy for assertions (specified
2167
by aspects or pragmas), the @cite{Debug} pragma, or additional checks
2168
to be checked using the @cite{Check} pragma. It may appear either as
2169
a configuration pragma, or within a declarative part of package. In the
2170
latter case, it applies from the point where it appears to the end of
2171
the declarative region (like pragma @cite{Suppress}).
2173
The @cite{Check_Policy} pragma is similar to the
2174
predefined @cite{Assertion_Policy} pragma,
2175
and if the check kind corresponds to one of the assertion kinds that
2176
are allowed by @cite{Assertion_Policy}, then the effect is identical.
2178
If the first argument is Debug, then the policy applies to Debug pragmas,
2179
disabling their effect if the policy is @cite{OFF}, @cite{DISABLE}, or
2180
@cite{IGNORE}, and allowing them to execute with normal semantics if
2181
the policy is @cite{ON} or @cite{CHECK}. In addition if the policy is
2182
@cite{DISABLE}, then the procedure call in @cite{Debug} pragmas will
2183
be totally ignored and not analyzed semantically.
2185
Finally the first argument may be some other identifier than the above
2186
possibilities, in which case it controls a set of named assertions
2187
that can be checked using pragma @cite{Check}. For example, if the pragma:
2190
pragma Check_Policy (Critical_Error, OFF);
2193
is given, then subsequent @cite{Check} pragmas whose first argument is also
2194
@cite{Critical_Error} will be disabled.
2196
The check policy is @cite{OFF} to turn off corresponding checks, and @cite{ON}
2197
to turn on corresponding checks. The default for a set of checks for which no
2198
@cite{Check_Policy} is given is @cite{OFF} unless the compiler switch
2199
@emph{-gnata} is given, which turns on all checks by default.
2201
The check policy settings @cite{CHECK} and @cite{IGNORE} are recognized
2202
as synonyms for @cite{ON} and @cite{OFF}. These synonyms are provided for
2203
compatibility with the standard @cite{Assertion_Policy} pragma. The check
2204
policy setting @cite{DISABLE} causes the second argument of a corresponding
2205
@cite{Check} pragma to be completely ignored and not analyzed.
2207
@node Pragma Comment,Pragma Common_Object,Pragma Check_Policy,Implementation Defined Pragmas
2208
@anchor{gnat_rm/implementation_defined_pragmas pragma-comment}@anchor{32}
2209
@section Pragma Comment
2215
pragma Comment (static_string_EXPRESSION);
2218
This is almost identical in effect to pragma @cite{Ident}. It allows the
2219
placement of a comment into the object file and hence into the
2220
executable file if the operating system permits such usage. The
2221
difference is that @cite{Comment}, unlike @cite{Ident}, has
2222
no limitations on placement of the pragma (it can be placed
2223
anywhere in the main source unit), and if more than one pragma
2224
is used, all comments are retained.
2226
@node Pragma Common_Object,Pragma Compile_Time_Error,Pragma Comment,Implementation Defined Pragmas
2227
@anchor{gnat_rm/implementation_defined_pragmas pragma-common-object}@anchor{33}
2228
@section Pragma Common_Object
2234
pragma Common_Object (
2235
[Internal =>] LOCAL_NAME
2236
[, [External =>] EXTERNAL_SYMBOL]
2237
[, [Size =>] EXTERNAL_SYMBOL] );
2241
| static_string_EXPRESSION
2244
This pragma enables the shared use of variables stored in overlaid
2245
linker areas corresponding to the use of @cite{COMMON}
2246
in Fortran. The single
2247
object @cite{LOCAL_NAME} is assigned to the area designated by
2248
the @cite{External} argument.
2249
You may define a record to correspond to a series
2250
of fields. The @cite{Size} argument
2251
is syntax checked in GNAT, but otherwise ignored.
2253
@cite{Common_Object} is not supported on all platforms. If no
2254
support is available, then the code generator will issue a message
2255
indicating that the necessary attribute for implementation of this
2256
pragma is not available.
2258
@node Pragma Compile_Time_Error,Pragma Compile_Time_Warning,Pragma Common_Object,Implementation Defined Pragmas
2259
@anchor{gnat_rm/implementation_defined_pragmas pragma-compile-time-error}@anchor{34}
2260
@section Pragma Compile_Time_Error
2266
pragma Compile_Time_Error
2267
(boolean_EXPRESSION, static_string_EXPRESSION);
2270
This pragma can be used to generate additional compile time
2272
is particularly useful in generics, where errors can be issued for
2273
specific problematic instantiations. The first parameter is a boolean
2274
expression. The pragma is effective only if the value of this expression
2275
is known at compile time, and has the value True. The set of expressions
2276
whose values are known at compile time includes all static boolean
2277
expressions, and also other values which the compiler can determine
2278
at compile time (e.g., the size of a record type set by an explicit
2279
size representation clause, or the value of a variable which was
2280
initialized to a constant and is known not to have been modified).
2281
If these conditions are met, an error message is generated using
2282
the value given as the second argument. This string value may contain
2283
embedded ASCII.LF characters to break the message into multiple lines.
2285
@node Pragma Compile_Time_Warning,Pragma Compiler_Unit,Pragma Compile_Time_Error,Implementation Defined Pragmas
2286
@anchor{gnat_rm/implementation_defined_pragmas pragma-compile-time-warning}@anchor{35}
2287
@section Pragma Compile_Time_Warning
2293
pragma Compile_Time_Warning
2294
(boolean_EXPRESSION, static_string_EXPRESSION);
2297
Same as pragma Compile_Time_Error, except a warning is issued instead
2298
of an error message. Note that if this pragma is used in a package that
2299
is with'ed by a client, the client will get the warning even though it
2300
is issued by a with'ed package (normally warnings in with'ed units are
2301
suppressed, but this is a special exception to that rule).
2303
One typical use is within a generic where compile time known characteristics
2304
of formal parameters are tested, and warnings given appropriately. Another use
2305
with a first parameter of True is to warn a client about use of a package,
2306
for example that it is not fully implemented.
2308
@node Pragma Compiler_Unit,Pragma Compiler_Unit_Warning,Pragma Compile_Time_Warning,Implementation Defined Pragmas
2309
@anchor{gnat_rm/implementation_defined_pragmas pragma-compiler-unit}@anchor{36}
2310
@section Pragma Compiler_Unit
2316
pragma Compiler_Unit;
2319
This pragma is obsolete. It is equivalent to Compiler_Unit_Warning. It is
2320
retained so that old versions of the GNAT run-time that use this pragma can
2321
be compiled with newer versions of the compiler.
2323
@node Pragma Compiler_Unit_Warning,Pragma Complete_Representation,Pragma Compiler_Unit,Implementation Defined Pragmas
2324
@anchor{gnat_rm/implementation_defined_pragmas pragma-compiler-unit-warning}@anchor{37}
2325
@section Pragma Compiler_Unit_Warning
2331
pragma Compiler_Unit_Warning;
2334
This pragma is intended only for internal use in the GNAT run-time library.
2335
It indicates that the unit is used as part of the compiler build. The effect
2336
is to generate warnings for the use of constructs (for example, conditional
2337
expressions) that would cause trouble when bootstrapping using an older
2338
version of GNAT. For the exact list of restrictions, see the compiler sources
2339
and references to Check_Compiler_Unit.
2341
@node Pragma Complete_Representation,Pragma Complex_Representation,Pragma Compiler_Unit_Warning,Implementation Defined Pragmas
2342
@anchor{gnat_rm/implementation_defined_pragmas pragma-complete-representation}@anchor{38}
2343
@section Pragma Complete_Representation
2349
pragma Complete_Representation;
2352
This pragma must appear immediately within a record representation
2353
clause. Typical placements are before the first component clause
2354
or after the last component clause. The effect is to give an error
2355
message if any component is missing a component clause. This pragma
2356
may be used to ensure that a record representation clause is
2357
complete, and that this invariant is maintained if fields are
2358
added to the record in the future.
2360
@node Pragma Complex_Representation,Pragma Component_Alignment,Pragma Complete_Representation,Implementation Defined Pragmas
2361
@anchor{gnat_rm/implementation_defined_pragmas pragma-complex-representation}@anchor{39}
2362
@section Pragma Complex_Representation
2368
pragma Complex_Representation
2369
([Entity =>] LOCAL_NAME);
2372
The @cite{Entity} argument must be the name of a record type which has
2373
two fields of the same floating-point type. The effect of this pragma is
2374
to force gcc to use the special internal complex representation form for
2375
this record, which may be more efficient. Note that this may result in
2376
the code for this type not conforming to standard ABI (application
2377
binary interface) requirements for the handling of record types. For
2378
example, in some environments, there is a requirement for passing
2379
records by pointer, and the use of this pragma may result in passing
2380
this type in floating-point registers.
2382
@node Pragma Component_Alignment,Pragma Constant_After_Elaboration,Pragma Complex_Representation,Implementation Defined Pragmas
2383
@anchor{gnat_rm/implementation_defined_pragmas pragma-component-alignment}@anchor{3a}
2384
@section Pragma Component_Alignment
2387
@geindex Alignments of components
2389
@geindex Pragma Component_Alignment
2394
pragma Component_Alignment (
2395
[Form =>] ALIGNMENT_CHOICE
2396
[, [Name =>] type_LOCAL_NAME]);
2398
ALIGNMENT_CHOICE ::=
2405
Specifies the alignment of components in array or record types.
2406
The meaning of the @cite{Form} argument is as follows:
2410
@geindex Component_Size (in pragma Component_Alignment)
2416
@item @emph{Component_Size}
2418
Aligns scalar components and subcomponents of the array or record type
2419
on boundaries appropriate to their inherent size (naturally
2420
aligned). For example, 1-byte components are aligned on byte boundaries,
2421
2-byte integer components are aligned on 2-byte boundaries, 4-byte
2422
integer components are aligned on 4-byte boundaries and so on. These
2423
alignment rules correspond to the normal rules for C compilers on all
2424
machines except the VAX.
2426
@geindex Component_Size_4 (in pragma Component_Alignment)
2428
@item @emph{Component_Size_4}
2430
Naturally aligns components with a size of four or fewer
2431
bytes. Components that are larger than 4 bytes are placed on the next
2434
@geindex Storage_Unit (in pragma Component_Alignment)
2436
@item @emph{Storage_Unit}
2438
Specifies that array or record components are byte aligned, i.e.,
2439
aligned on boundaries determined by the value of the constant
2440
@cite{System.Storage_Unit}.
2442
@geindex Default (in pragma Component_Alignment)
2444
@item @emph{Default}
2446
Specifies that array or record components are aligned on default
2447
boundaries, appropriate to the underlying hardware or operating system or
2448
both. The @cite{Default} choice is the same as @cite{Component_Size} (natural
2452
If the @cite{Name} parameter is present, @cite{type_LOCAL_NAME} must
2453
refer to a local record or array type, and the specified alignment
2454
choice applies to the specified type. The use of
2455
@cite{Component_Alignment} together with a pragma @cite{Pack} causes the
2456
@cite{Component_Alignment} pragma to be ignored. The use of
2457
@cite{Component_Alignment} together with a record representation clause
2458
is only effective for fields not specified by the representation clause.
2460
If the @cite{Name} parameter is absent, the pragma can be used as either
2461
a configuration pragma, in which case it applies to one or more units in
2462
accordance with the normal rules for configuration pragmas, or it can be
2463
used within a declarative part, in which case it applies to types that
2464
are declared within this declarative part, or within any nested scope
2465
within this declarative part. In either case it specifies the alignment
2466
to be applied to any record or array type which has otherwise standard
2469
If the alignment for a record or array type is not specified (using
2470
pragma @cite{Pack}, pragma @cite{Component_Alignment}, or a record rep
2471
clause), the GNAT uses the default alignment as described previously.
2473
@node Pragma Constant_After_Elaboration,Pragma Contract_Cases,Pragma Component_Alignment,Implementation Defined Pragmas
2474
@anchor{gnat_rm/implementation_defined_pragmas pragma-constant-after-elaboration}@anchor{3b}
2475
@section Pragma Constant_After_Elaboration
2481
pragma Constant_After_Elaboration [ (boolean_EXPRESSION) ];
2484
For the semantics of this pragma, see the entry for aspect
2485
@cite{Constant_After_Elaboration} in the SPARK 2014 Reference Manual, section 3.3.1.
2487
@node Pragma Contract_Cases,Pragma Convention_Identifier,Pragma Constant_After_Elaboration,Implementation Defined Pragmas
2488
@anchor{gnat_rm/implementation_defined_pragmas pragma-contract-cases}@anchor{3c}
2489
@section Pragma Contract_Cases
2492
@geindex Contract cases
2497
pragma Contract_Cases ((CONTRACT_CASE @{, CONTRACT_CASE));
2499
CONTRACT_CASE ::= CASE_GUARD => CONSEQUENCE
2501
CASE_GUARD ::= boolean_EXPRESSION | others
2503
CONSEQUENCE ::= boolean_EXPRESSION
2506
The @cite{Contract_Cases} pragma allows defining fine-grain specifications
2507
that can complement or replace the contract given by a precondition and a
2508
postcondition. Additionally, the @cite{Contract_Cases} pragma can be used
2509
by testing and formal verification tools. The compiler checks its validity and,
2510
depending on the assertion policy at the point of declaration of the pragma,
2511
it may insert a check in the executable. For code generation, the contract
2515
pragma Contract_Cases (
2523
C1 : constant Boolean := Cond1; -- evaluated at subprogram entry
2524
C2 : constant Boolean := Cond2; -- evaluated at subprogram entry
2525
pragma Precondition ((C1 and not C2) or (C2 and not C1));
2526
pragma Postcondition (if C1 then Pred1);
2527
pragma Postcondition (if C2 then Pred2);
2530
The precondition ensures that one and only one of the conditions is
2531
satisfied on entry to the subprogram.
2532
The postcondition ensures that for the condition that was True on entry,
2533
the corrresponding consequence is True on exit. Other consequence expressions
2536
A precondition @cite{P} and postcondition @cite{Q} can also be
2537
expressed as contract cases:
2540
pragma Contract_Cases (P => Q);
2543
The placement and visibility rules for @cite{Contract_Cases} pragmas are
2544
identical to those described for preconditions and postconditions.
2546
The compiler checks that boolean expressions given in conditions and
2547
consequences are valid, where the rules for conditions are the same as
2548
the rule for an expression in @cite{Precondition} and the rules for
2549
consequences are the same as the rule for an expression in
2550
@cite{Postcondition}. In particular, attributes @cite{'Old} and
2551
@cite{'Result} can only be used within consequence expressions.
2552
The condition for the last contract case may be @cite{others}, to denote
2553
any case not captured by the previous cases. The
2554
following is an example of use within a package spec:
2557
package Math_Functions is
2559
function Sqrt (Arg : Float) return Float;
2560
pragma Contract_Cases ((Arg in 0 .. 99) => Sqrt'Result < 10,
2561
Arg >= 100 => Sqrt'Result >= 10,
2562
others => Sqrt'Result = 0);
2567
The meaning of contract cases is that only one case should apply at each
2568
call, as determined by the corresponding condition evaluating to True,
2569
and that the consequence for this case should hold when the subprogram
2572
@node Pragma Convention_Identifier,Pragma CPP_Class,Pragma Contract_Cases,Implementation Defined Pragmas
2573
@anchor{gnat_rm/implementation_defined_pragmas pragma-convention-identifier}@anchor{3d}
2574
@section Pragma Convention_Identifier
2577
@geindex Conventions
2583
pragma Convention_Identifier (
2584
[Name =>] IDENTIFIER,
2585
[Convention =>] convention_IDENTIFIER);
2588
This pragma provides a mechanism for supplying synonyms for existing
2589
convention identifiers. The @cite{Name} identifier can subsequently
2590
be used as a synonym for the given convention in other pragmas (including
2591
for example pragma @cite{Import} or another @cite{Convention_Identifier}
2592
pragma). As an example of the use of this, suppose you had legacy code
2593
which used Fortran77 as the identifier for Fortran. Then the pragma:
2596
pragma Convention_Identifier (Fortran77, Fortran);
2599
would allow the use of the convention identifier @cite{Fortran77} in
2600
subsequent code, avoiding the need to modify the sources. As another
2601
example, you could use this to parameterize convention requirements
2602
according to systems. Suppose you needed to use @cite{Stdcall} on
2603
windows systems, and @cite{C} on some other system, then you could
2604
define a convention identifier @cite{Library} and use a single
2605
@cite{Convention_Identifier} pragma to specify which convention
2606
would be used system-wide.
2608
@node Pragma CPP_Class,Pragma CPP_Constructor,Pragma Convention_Identifier,Implementation Defined Pragmas
2609
@anchor{gnat_rm/implementation_defined_pragmas pragma-cpp-class}@anchor{3e}
2610
@section Pragma CPP_Class
2613
@geindex Interfacing with C++
2618
pragma CPP_Class ([Entity =>] LOCAL_NAME);
2621
The argument denotes an entity in the current declarative region that is
2622
declared as a record type. It indicates that the type corresponds to an
2623
externally declared C++ class type, and is to be laid out the same way
2624
that C++ would lay out the type. If the C++ class has virtual primitives
2625
then the record must be declared as a tagged record type.
2627
Types for which @cite{CPP_Class} is specified do not have assignment or
2628
equality operators defined (such operations can be imported or declared
2629
as subprograms as required). Initialization is allowed only by constructor
2630
functions (see pragma @cite{CPP_Constructor}). Such types are implicitly
2631
limited if not explicitly declared as limited or derived from a limited
2632
type, and an error is issued in that case.
2634
See @ref{3f,,Interfacing to C++} for related information.
2636
Note: Pragma @cite{CPP_Class} is currently obsolete. It is supported
2637
for backward compatibility but its functionality is available
2638
using pragma @cite{Import} with @cite{Convention} = @cite{CPP}.
2640
@node Pragma CPP_Constructor,Pragma CPP_Virtual,Pragma CPP_Class,Implementation Defined Pragmas
2641
@anchor{gnat_rm/implementation_defined_pragmas pragma-cpp-constructor}@anchor{40}
2642
@section Pragma CPP_Constructor
2645
@geindex Interfacing with C++
2650
pragma CPP_Constructor ([Entity =>] LOCAL_NAME
2651
[, [External_Name =>] static_string_EXPRESSION ]
2652
[, [Link_Name =>] static_string_EXPRESSION ]);
2655
This pragma identifies an imported function (imported in the usual way
2656
with pragma @cite{Import}) as corresponding to a C++ constructor. If
2657
@cite{External_Name} and @cite{Link_Name} are not specified then the
2658
@cite{Entity} argument is a name that must have been previously mentioned
2659
in a pragma @cite{Import} with @cite{Convention} = @cite{CPP}. Such name
2660
must be of one of the following forms:
2666
@strong{function} @cite{Fname} @strong{return} T`
2669
@strong{function} @cite{Fname} @strong{return} T'Class
2672
@strong{function} @cite{Fname} (...) @strong{return} T`
2675
@strong{function} @cite{Fname} (...) @strong{return} T'Class
2678
where @cite{T} is a limited record type imported from C++ with pragma
2679
@cite{Import} and @cite{Convention} = @cite{CPP}.
2681
The first two forms import the default constructor, used when an object
2682
of type @cite{T} is created on the Ada side with no explicit constructor.
2683
The latter two forms cover all the non-default constructors of the type.
2684
See the GNAT User's Guide for details.
2686
If no constructors are imported, it is impossible to create any objects
2687
on the Ada side and the type is implicitly declared abstract.
2689
Pragma @cite{CPP_Constructor} is intended primarily for automatic generation
2690
using an automatic binding generator tool (such as the @cite{-fdump-ada-spec}
2692
See @ref{3f,,Interfacing to C++} for more related information.
2694
Note: The use of functions returning class-wide types for constructors is
2695
currently obsolete. They are supported for backward compatibility. The
2696
use of functions returning the type T leave the Ada sources more clear
2697
because the imported C++ constructors always return an object of type T;
2698
that is, they never return an object whose type is a descendant of type T.
2700
@node Pragma CPP_Virtual,Pragma CPP_Vtable,Pragma CPP_Constructor,Implementation Defined Pragmas
2701
@anchor{gnat_rm/implementation_defined_pragmas pragma-cpp-virtual}@anchor{41}
2702
@section Pragma CPP_Virtual
2705
@geindex Interfacing to C++
2707
This pragma is now obsolete and, other than generating a warning if warnings
2708
on obsolescent features are enabled, is completely ignored.
2709
It is retained for compatibility
2710
purposes. It used to be required to ensure compoatibility with C++, but
2711
is no longer required for that purpose because GNAT generates
2712
the same object layout as the G++ compiler by default.
2714
See @ref{3f,,Interfacing to C++} for related information.
2716
@node Pragma CPP_Vtable,Pragma CPU,Pragma CPP_Virtual,Implementation Defined Pragmas
2717
@anchor{gnat_rm/implementation_defined_pragmas pragma-cpp-vtable}@anchor{42}
2718
@section Pragma CPP_Vtable
2721
@geindex Interfacing with C++
2723
This pragma is now obsolete and, other than generating a warning if warnings
2724
on obsolescent features are enabled, is completely ignored.
2725
It used to be required to ensure compatibility with C++, but
2726
is no longer required for that purpose because GNAT generates
2727
the same object layout as the G++ compiler by default.
2729
See @ref{3f,,Interfacing to C++} for related information.
2731
@node Pragma CPU,Pragma Default_Initial_Condition,Pragma CPP_Vtable,Implementation Defined Pragmas
2732
@anchor{gnat_rm/implementation_defined_pragmas pragma-cpu}@anchor{43}
2739
pragma CPU (EXPRESSION);
2742
This pragma is standard in Ada 2012, but is available in all earlier
2743
versions of Ada as an implementation-defined pragma.
2744
See Ada 2012 Reference Manual for details.
2746
@node Pragma Default_Initial_Condition,Pragma Debug,Pragma CPU,Implementation Defined Pragmas
2747
@anchor{gnat_rm/implementation_defined_pragmas pragma-default-initial-condition}@anchor{44}
2748
@section Pragma Default_Initial_Condition
2754
pragma Default_Initial_Condition [ (null | boolean_EXPRESSION) ];
2757
For the semantics of this pragma, see the entry for aspect
2758
@cite{Default_Initial_Condition} in the SPARK 2014 Reference Manual, section 7.3.3.
2760
@node Pragma Debug,Pragma Debug_Policy,Pragma Default_Initial_Condition,Implementation Defined Pragmas
2761
@anchor{gnat_rm/implementation_defined_pragmas pragma-debug}@anchor{45}
2762
@section Pragma Debug
2768
pragma Debug ([CONDITION, ]PROCEDURE_CALL_WITHOUT_SEMICOLON);
2770
PROCEDURE_CALL_WITHOUT_SEMICOLON ::=
2772
| PROCEDURE_PREFIX ACTUAL_PARAMETER_PART
2775
The procedure call argument has the syntactic form of an expression, meeting
2776
the syntactic requirements for pragmas.
2778
If debug pragmas are not enabled or if the condition is present and evaluates
2779
to False, this pragma has no effect. If debug pragmas are enabled, the
2780
semantics of the pragma is exactly equivalent to the procedure call statement
2781
corresponding to the argument with a terminating semicolon. Pragmas are
2782
permitted in sequences of declarations, so you can use pragma @cite{Debug} to
2783
intersperse calls to debug procedures in the middle of declarations. Debug
2784
pragmas can be enabled either by use of the command line switch @emph{-gnata}
2785
or by use of the pragma @cite{Check_Policy} with a first argument of
2788
@node Pragma Debug_Policy,Pragma Default_Scalar_Storage_Order,Pragma Debug,Implementation Defined Pragmas
2789
@anchor{gnat_rm/implementation_defined_pragmas pragma-debug-policy}@anchor{46}
2790
@section Pragma Debug_Policy
2796
pragma Debug_Policy (CHECK | DISABLE | IGNORE | ON | OFF);
2799
This pragma is equivalent to a corresponding @cite{Check_Policy} pragma
2800
with a first argument of @cite{Debug}. It is retained for historical
2801
compatibility reasons.
2803
@node Pragma Default_Scalar_Storage_Order,Pragma Default_Storage_Pool,Pragma Debug_Policy,Implementation Defined Pragmas
2804
@anchor{gnat_rm/implementation_defined_pragmas pragma-default-scalar-storage-order}@anchor{47}
2805
@section Pragma Default_Scalar_Storage_Order
2808
@geindex Default_Scalar_Storage_Order
2810
@geindex Scalar_Storage_Order
2815
pragma Default_Scalar_Storage_Order (High_Order_First | Low_Order_First);
2818
Normally if no explicit @cite{Scalar_Storage_Order} is given for a record
2819
type or array type, then the scalar storage order defaults to the ordinary
2820
default for the target. But this default may be overridden using this pragma.
2821
The pragma may appear as a configuration pragma, or locally within a package
2822
spec or declarative part. In the latter case, it applies to all subsequent
2823
types declared within that package spec or declarative part.
2825
The following example shows the use of this pragma:
2828
pragma Default_Scalar_Storage_Order (High_Order_First);
2829
with System; use System;
2838
for L2'Scalar_Storage_Order use Low_Order_First;
2847
pragma Default_Scalar_Storage_Order (Low_Order_First);
2854
type H4a is new Inner.L4;
2862
In this example record types L.. have @cite{Low_Order_First} scalar
2863
storage order, and record types H.. have @cite{High_Order_First}.
2864
Note that in the case of @cite{H4a}, the order is not inherited
2865
from the parent type. Only an explicitly set @cite{Scalar_Storage_Order}
2866
gets inherited on type derivation.
2868
If this pragma is used as a configuration pragma which appears within a
2869
configuration pragma file (as opposed to appearing explicitly at the start
2870
of a single unit), then the binder will require that all units in a partition
2871
be compiled in a similar manner, other than run-time units, which are not
2872
affected by this pragma. Note that the use of this form is discouraged because
2873
it may significantly degrade the run-time performance of the software, instead
2874
the default scalar storage order ought to be changed only on a local basis.
2876
@node Pragma Default_Storage_Pool,Pragma Depends,Pragma Default_Scalar_Storage_Order,Implementation Defined Pragmas
2877
@anchor{gnat_rm/implementation_defined_pragmas pragma-default-storage-pool}@anchor{48}
2878
@section Pragma Default_Storage_Pool
2881
@geindex Default_Storage_Pool
2886
pragma Default_Storage_Pool (storage_pool_NAME | null);
2889
This pragma is standard in Ada 2012, but is available in all earlier
2890
versions of Ada as an implementation-defined pragma.
2891
See Ada 2012 Reference Manual for details.
2893
@node Pragma Depends,Pragma Detect_Blocking,Pragma Default_Storage_Pool,Implementation Defined Pragmas
2894
@anchor{gnat_rm/implementation_defined_pragmas pragma-depends}@anchor{49}
2895
@section Pragma Depends
2901
pragma Depends (DEPENDENCY_RELATION);
2903
DEPENDENCY_RELATION ::=
2905
| (DEPENDENCY_CLAUSE @{, DEPENDENCY_CLAUSE@})
2907
DEPENDENCY_CLAUSE ::=
2908
OUTPUT_LIST =>[+] INPUT_LIST
2909
| NULL_DEPENDENCY_CLAUSE
2911
NULL_DEPENDENCY_CLAUSE ::= null => INPUT_LIST
2913
OUTPUT_LIST ::= OUTPUT | (OUTPUT @{, OUTPUT@})
2915
INPUT_LIST ::= null | INPUT | (INPUT @{, INPUT@})
2917
OUTPUT ::= NAME | FUNCTION_RESULT
2920
where FUNCTION_RESULT is a function Result attribute_reference
2923
For the semantics of this pragma, see the entry for aspect @cite{Depends} in the
2924
SPARK 2014 Reference Manual, section 6.1.5.
2926
@node Pragma Detect_Blocking,Pragma Disable_Atomic_Synchronization,Pragma Depends,Implementation Defined Pragmas
2927
@anchor{gnat_rm/implementation_defined_pragmas pragma-detect-blocking}@anchor{4a}
2928
@section Pragma Detect_Blocking
2934
pragma Detect_Blocking;
2937
This is a standard pragma in Ada 2005, that is available in all earlier
2938
versions of Ada as an implementation-defined pragma.
2940
This is a configuration pragma that forces the detection of potentially
2941
blocking operations within a protected operation, and to raise Program_Error
2944
@node Pragma Disable_Atomic_Synchronization,Pragma Dispatching_Domain,Pragma Detect_Blocking,Implementation Defined Pragmas
2945
@anchor{gnat_rm/implementation_defined_pragmas pragma-disable-atomic-synchronization}@anchor{4b}
2946
@section Pragma Disable_Atomic_Synchronization
2949
@geindex Atomic Synchronization
2954
pragma Disable_Atomic_Synchronization [(Entity)];
2957
Ada requires that accesses (reads or writes) of an atomic variable be
2958
regarded as synchronization points in the case of multiple tasks.
2959
Particularly in the case of multi-processors this may require special
2960
handling, e.g. the generation of memory barriers. This capability may
2961
be turned off using this pragma in cases where it is known not to be
2964
The placement and scope rules for this pragma are the same as those
2965
for @cite{pragma Suppress}. In particular it can be used as a
2966
configuration pragma, or in a declaration sequence where it applies
2967
till the end of the scope. If an @cite{Entity} argument is present,
2968
the action applies only to that entity.
2970
@node Pragma Dispatching_Domain,Pragma Effective_Reads,Pragma Disable_Atomic_Synchronization,Implementation Defined Pragmas
2971
@anchor{gnat_rm/implementation_defined_pragmas pragma-dispatching-domain}@anchor{4c}
2972
@section Pragma Dispatching_Domain
2978
pragma Dispatching_Domain (EXPRESSION);
2981
This pragma is standard in Ada 2012, but is available in all earlier
2982
versions of Ada as an implementation-defined pragma.
2983
See Ada 2012 Reference Manual for details.
2985
@node Pragma Effective_Reads,Pragma Effective_Writes,Pragma Dispatching_Domain,Implementation Defined Pragmas
2986
@anchor{gnat_rm/implementation_defined_pragmas pragma-effective-reads}@anchor{4d}
2987
@section Pragma Effective_Reads
2993
pragma Effective_Reads [ (boolean_EXPRESSION) ];
2996
For the semantics of this pragma, see the entry for aspect @cite{Effective_Reads} in
2997
the SPARK 2014 Reference Manual, section 7.1.2.
2999
@node Pragma Effective_Writes,Pragma Elaboration_Checks,Pragma Effective_Reads,Implementation Defined Pragmas
3000
@anchor{gnat_rm/implementation_defined_pragmas pragma-effective-writes}@anchor{4e}
3001
@section Pragma Effective_Writes
3007
pragma Effective_Writes [ (boolean_EXPRESSION) ];
3010
For the semantics of this pragma, see the entry for aspect @cite{Effective_Writes}
3011
in the SPARK 2014 Reference Manual, section 7.1.2.
3013
@node Pragma Elaboration_Checks,Pragma Eliminate,Pragma Effective_Writes,Implementation Defined Pragmas
3014
@anchor{gnat_rm/implementation_defined_pragmas pragma-elaboration-checks}@anchor{4f}
3015
@section Pragma Elaboration_Checks
3018
@geindex Elaboration control
3023
pragma Elaboration_Checks (Dynamic | Static);
3026
This is a configuration pragma that provides control over the
3027
elaboration model used by the compilation affected by the
3028
pragma. If the parameter is @cite{Dynamic},
3029
then the dynamic elaboration
3030
model described in the Ada Reference Manual is used, as though
3031
the @emph{-gnatE} switch had been specified on the command
3032
line. If the parameter is @cite{Static}, then the default GNAT static
3033
model is used. This configuration pragma overrides the setting
3034
of the command line. For full details on the elaboration models
3035
used by the GNAT compiler, see the chapter on elaboration order handling
3036
in the @emph{GNAT User's Guide}.
3038
@node Pragma Eliminate,Pragma Enable_Atomic_Synchronization,Pragma Elaboration_Checks,Implementation Defined Pragmas
3039
@anchor{gnat_rm/implementation_defined_pragmas pragma-eliminate}@anchor{50}
3040
@section Pragma Eliminate
3043
@geindex Elimination of unused subprograms
3048
pragma Eliminate ([Entity =>] DEFINING_DESIGNATOR,
3049
[Source_Location =>] STRING_LITERAL);
3052
The string literal given for the source location is a string which
3053
specifies the line number of the occurrence of the entity, using
3054
the syntax for SOURCE_TRACE given below:
3057
SOURCE_TRACE ::= SOURCE_REFERENCE [LBRACKET SOURCE_TRACE RBRACKET]
3062
SOURCE_REFERENCE ::= FILE_NAME : LINE_NUMBER
3064
LINE_NUMBER ::= DIGIT @{DIGIT@}
3067
Spaces around the colon in a @cite{Source_Reference} are optional.
3069
The @cite{DEFINING_DESIGNATOR} matches the defining designator used in an
3070
explicit subprogram declaration, where the @cite{entity} name in this
3071
designator appears on the source line specified by the source location.
3073
The source trace that is given as the @cite{Source_Location} shall obey the
3074
following rules. The @cite{FILE_NAME} is the short name (with no directory
3075
information) of an Ada source file, given using exactly the required syntax
3076
for the underlying file system (e.g. case is important if the underlying
3077
operating system is case sensitive). @cite{LINE_NUMBER} gives the line
3078
number of the occurrence of the @cite{entity}
3079
as a decimal literal without an exponent or point. If an @cite{entity} is not
3080
declared in a generic instantiation (this includes generic subprogram
3081
instances), the source trace includes only one source reference. If an entity
3082
is declared inside a generic instantiation, its source trace (when parsing
3083
from left to right) starts with the source location of the declaration of the
3084
entity in the generic unit and ends with the source location of the
3085
instantiation (it is given in square brackets). This approach is recursively
3086
used in case of nested instantiations: the rightmost (nested most deeply in
3087
square brackets) element of the source trace is the location of the outermost
3088
instantiation, the next to left element is the location of the next (first
3089
nested) instantiation in the code of the corresponding generic unit, and so
3090
on, and the leftmost element (that is out of any square brackets) is the
3091
location of the declaration of the entity to eliminate in a generic unit.
3093
Note that the @cite{Source_Location} argument specifies which of a set of
3094
similarly named entities is being eliminated, dealing both with overloading,
3095
and also appearance of the same entity name in different scopes.
3097
This pragma indicates that the given entity is not used in the program to be
3098
compiled and built. The effect of the pragma is to allow the compiler to
3099
eliminate the code or data associated with the named entity. Any reference to
3100
an eliminated entity causes a compile-time or link-time error.
3102
The intention of pragma @cite{Eliminate} is to allow a program to be compiled
3103
in a system-independent manner, with unused entities eliminated, without
3104
needing to modify the source text. Normally the required set of
3105
@cite{Eliminate} pragmas is constructed automatically using the gnatelim tool.
3107
Any source file change that removes, splits, or
3108
adds lines may make the set of Eliminate pragmas invalid because their
3109
@cite{Source_Location} argument values may get out of date.
3111
Pragma @cite{Eliminate} may be used where the referenced entity is a dispatching
3112
operation. In this case all the subprograms to which the given operation can
3113
dispatch are considered to be unused (are never called as a result of a direct
3114
or a dispatching call).
3116
@node Pragma Enable_Atomic_Synchronization,Pragma Export_Function,Pragma Eliminate,Implementation Defined Pragmas
3117
@anchor{gnat_rm/implementation_defined_pragmas pragma-enable-atomic-synchronization}@anchor{51}
3118
@section Pragma Enable_Atomic_Synchronization
3121
@geindex Atomic Synchronization
3126
pragma Enable_Atomic_Synchronization [(Entity)];
3129
Ada requires that accesses (reads or writes) of an atomic variable be
3130
regarded as synchronization points in the case of multiple tasks.
3131
Particularly in the case of multi-processors this may require special
3132
handling, e.g. the generation of memory barriers. This synchronization
3133
is performed by default, but can be turned off using
3134
@cite{pragma Disable_Atomic_Synchronization}. The
3135
@cite{Enable_Atomic_Synchronization} pragma can be used to turn
3138
The placement and scope rules for this pragma are the same as those
3139
for @cite{pragma Unsuppress}. In particular it can be used as a
3140
configuration pragma, or in a declaration sequence where it applies
3141
till the end of the scope. If an @cite{Entity} argument is present,
3142
the action applies only to that entity.
3144
@node Pragma Export_Function,Pragma Export_Object,Pragma Enable_Atomic_Synchronization,Implementation Defined Pragmas
3145
@anchor{gnat_rm/implementation_defined_pragmas pragma-export-function}@anchor{52}
3146
@section Pragma Export_Function
3149
@geindex Argument passing mechanisms
3154
pragma Export_Function (
3155
[Internal =>] LOCAL_NAME
3156
[, [External =>] EXTERNAL_SYMBOL]
3157
[, [Parameter_Types =>] PARAMETER_TYPES]
3158
[, [Result_Type =>] result_SUBTYPE_MARK]
3159
[, [Mechanism =>] MECHANISM]
3160
[, [Result_Mechanism =>] MECHANISM_NAME]);
3164
| static_string_EXPRESSION
3169
| TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
3173
| subtype_Name ' Access
3177
| (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
3179
MECHANISM_ASSOCIATION ::=
3180
[formal_parameter_NAME =>] MECHANISM_NAME
3182
MECHANISM_NAME ::= Value | Reference
3185
Use this pragma to make a function externally callable and optionally
3186
provide information on mechanisms to be used for passing parameter and
3187
result values. We recommend, for the purposes of improving portability,
3188
this pragma always be used in conjunction with a separate pragma
3189
@cite{Export}, which must precede the pragma @cite{Export_Function}.
3190
GNAT does not require a separate pragma @cite{Export}, but if none is
3191
present, @cite{Convention Ada} is assumed, which is usually
3192
not what is wanted, so it is usually appropriate to use this
3193
pragma in conjunction with a @cite{Export} or @cite{Convention}
3194
pragma that specifies the desired foreign convention.
3195
Pragma @cite{Export_Function}
3196
(and @cite{Export}, if present) must appear in the same declarative
3197
region as the function to which they apply.
3199
@cite{internal_name} must uniquely designate the function to which the
3200
pragma applies. If more than one function name exists of this name in
3201
the declarative part you must use the @cite{Parameter_Types} and
3202
@cite{Result_Type} parameters is mandatory to achieve the required
3203
unique designation. @cite{subtype_mark`s in these parameters must exactly match the subtypes in the corresponding function specification@comma{} using positional notation to match parameters with subtype marks. The form with an `'Access} attribute can be used to match an
3204
anonymous access parameter.
3206
@geindex Suppressing external name
3208
Special treatment is given if the EXTERNAL is an explicit null
3209
string or a static string expressions that evaluates to the null
3210
string. In this case, no external name is generated. This form
3211
still allows the specification of parameter mechanisms.
3213
@node Pragma Export_Object,Pragma Export_Procedure,Pragma Export_Function,Implementation Defined Pragmas
3214
@anchor{gnat_rm/implementation_defined_pragmas pragma-export-object}@anchor{53}
3215
@section Pragma Export_Object
3221
pragma Export_Object
3222
[Internal =>] LOCAL_NAME
3223
[, [External =>] EXTERNAL_SYMBOL]
3224
[, [Size =>] EXTERNAL_SYMBOL]
3228
| static_string_EXPRESSION
3231
This pragma designates an object as exported, and apart from the
3232
extended rules for external symbols, is identical in effect to the use of
3233
the normal @cite{Export} pragma applied to an object. You may use a
3234
separate Export pragma (and you probably should from the point of view
3235
of portability), but it is not required. @cite{Size} is syntax checked,
3236
but otherwise ignored by GNAT.
3238
@node Pragma Export_Procedure,Pragma Export_Value,Pragma Export_Object,Implementation Defined Pragmas
3239
@anchor{gnat_rm/implementation_defined_pragmas pragma-export-procedure}@anchor{54}
3240
@section Pragma Export_Procedure
3246
pragma Export_Procedure (
3247
[Internal =>] LOCAL_NAME
3248
[, [External =>] EXTERNAL_SYMBOL]
3249
[, [Parameter_Types =>] PARAMETER_TYPES]
3250
[, [Mechanism =>] MECHANISM]);
3254
| static_string_EXPRESSION
3259
| TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
3263
| subtype_Name ' Access
3267
| (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
3269
MECHANISM_ASSOCIATION ::=
3270
[formal_parameter_NAME =>] MECHANISM_NAME
3272
MECHANISM_NAME ::= Value | Reference
3275
This pragma is identical to @cite{Export_Function} except that it
3276
applies to a procedure rather than a function and the parameters
3277
@cite{Result_Type} and @cite{Result_Mechanism} are not permitted.
3278
GNAT does not require a separate pragma @cite{Export}, but if none is
3279
present, @cite{Convention Ada} is assumed, which is usually
3280
not what is wanted, so it is usually appropriate to use this
3281
pragma in conjunction with a @cite{Export} or @cite{Convention}
3282
pragma that specifies the desired foreign convention.
3284
@geindex Suppressing external name
3286
Special treatment is given if the EXTERNAL is an explicit null
3287
string or a static string expressions that evaluates to the null
3288
string. In this case, no external name is generated. This form
3289
still allows the specification of parameter mechanisms.
3291
@node Pragma Export_Value,Pragma Export_Valued_Procedure,Pragma Export_Procedure,Implementation Defined Pragmas
3292
@anchor{gnat_rm/implementation_defined_pragmas pragma-export-value}@anchor{55}
3293
@section Pragma Export_Value
3299
pragma Export_Value (
3300
[Value =>] static_integer_EXPRESSION,
3301
[Link_Name =>] static_string_EXPRESSION);
3304
This pragma serves to export a static integer value for external use.
3305
The first argument specifies the value to be exported. The Link_Name
3306
argument specifies the symbolic name to be associated with the integer
3307
value. This pragma is useful for defining a named static value in Ada
3308
that can be referenced in assembly language units to be linked with
3309
the application. This pragma is currently supported only for the
3310
AAMP target and is ignored for other targets.
3312
@node Pragma Export_Valued_Procedure,Pragma Extend_System,Pragma Export_Value,Implementation Defined Pragmas
3313
@anchor{gnat_rm/implementation_defined_pragmas pragma-export-valued-procedure}@anchor{56}
3314
@section Pragma Export_Valued_Procedure
3320
pragma Export_Valued_Procedure (
3321
[Internal =>] LOCAL_NAME
3322
[, [External =>] EXTERNAL_SYMBOL]
3323
[, [Parameter_Types =>] PARAMETER_TYPES]
3324
[, [Mechanism =>] MECHANISM]);
3328
| static_string_EXPRESSION
3333
| TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
3337
| subtype_Name ' Access
3341
| (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
3343
MECHANISM_ASSOCIATION ::=
3344
[formal_parameter_NAME =>] MECHANISM_NAME
3346
MECHANISM_NAME ::= Value | Reference
3349
This pragma is identical to @cite{Export_Procedure} except that the
3350
first parameter of @cite{LOCAL_NAME}, which must be present, must be of
3351
mode @cite{OUT}, and externally the subprogram is treated as a function
3352
with this parameter as the result of the function. GNAT provides for
3353
this capability to allow the use of @cite{OUT} and @cite{IN OUT}
3354
parameters in interfacing to external functions (which are not permitted
3356
GNAT does not require a separate pragma @cite{Export}, but if none is
3357
present, @cite{Convention Ada} is assumed, which is almost certainly
3358
not what is wanted since the whole point of this pragma is to interface
3359
with foreign language functions, so it is usually appropriate to use this
3360
pragma in conjunction with a @cite{Export} or @cite{Convention}
3361
pragma that specifies the desired foreign convention.
3363
@geindex Suppressing external name
3365
Special treatment is given if the EXTERNAL is an explicit null
3366
string or a static string expressions that evaluates to the null
3367
string. In this case, no external name is generated. This form
3368
still allows the specification of parameter mechanisms.
3370
@node Pragma Extend_System,Pragma Extensions_Allowed,Pragma Export_Valued_Procedure,Implementation Defined Pragmas
3371
@anchor{gnat_rm/implementation_defined_pragmas pragma-extend-system}@anchor{57}
3372
@section Pragma Extend_System
3383
pragma Extend_System ([Name =>] IDENTIFIER);
3386
This pragma is used to provide backwards compatibility with other
3387
implementations that extend the facilities of package @cite{System}. In
3388
GNAT, @cite{System} contains only the definitions that are present in
3389
the Ada RM. However, other implementations, notably the DEC Ada 83
3390
implementation, provide many extensions to package @cite{System}.
3392
For each such implementation accommodated by this pragma, GNAT provides a
3393
package @cite{Aux_`xxx`}, e.g., @cite{Aux_DEC} for the DEC Ada 83
3394
implementation, which provides the required additional definitions. You
3395
can use this package in two ways. You can @cite{with} it in the normal
3396
way and access entities either by selection or using a @cite{use}
3397
clause. In this case no special processing is required.
3399
However, if existing code contains references such as
3400
@cite{System.`xxx`} where @cite{xxx} is an entity in the extended
3401
definitions provided in package @cite{System}, you may use this pragma
3402
to extend visibility in @cite{System} in a non-standard way that
3403
provides greater compatibility with the existing code. Pragma
3404
@cite{Extend_System} is a configuration pragma whose single argument is
3405
the name of the package containing the extended definition
3406
(e.g., @cite{Aux_DEC} for the DEC Ada case). A unit compiled under
3407
control of this pragma will be processed using special visibility
3408
processing that looks in package @cite{System.Aux_`xxx`} where
3409
@cite{Aux_`xxx`} is the pragma argument for any entity referenced in
3410
package @cite{System}, but not found in package @cite{System}.
3412
You can use this pragma either to access a predefined @cite{System}
3413
extension supplied with the compiler, for example @cite{Aux_DEC} or
3414
you can construct your own extension unit following the above
3415
definition. Note that such a package is a child of @cite{System}
3416
and thus is considered part of the implementation.
3417
To compile it you will have to use the @emph{-gnatg} switch
3418
for compiling System units, as explained in the
3421
@node Pragma Extensions_Allowed,Pragma Extensions_Visible,Pragma Extend_System,Implementation Defined Pragmas
3422
@anchor{gnat_rm/implementation_defined_pragmas pragma-extensions-allowed}@anchor{58}
3423
@section Pragma Extensions_Allowed
3426
@geindex Ada Extensions
3428
@geindex GNAT Extensions
3433
pragma Extensions_Allowed (On | Off);
3436
This configuration pragma enables or disables the implementation
3437
extension mode (the use of Off as a parameter cancels the effect
3438
of the @emph{-gnatX} command switch).
3440
In extension mode, the latest version of the Ada language is
3441
implemented (currently Ada 2012), and in addition a small number
3442
of GNAT specific extensions are recognized as follows:
3447
@item @emph{Constrained attribute for generic objects}
3449
The @cite{Constrained} attribute is permitted for objects of
3450
generic types. The result indicates if the corresponding actual
3454
@node Pragma Extensions_Visible,Pragma External,Pragma Extensions_Allowed,Implementation Defined Pragmas
3455
@anchor{gnat_rm/implementation_defined_pragmas pragma-extensions-visible}@anchor{59}
3456
@section Pragma Extensions_Visible
3462
pragma Extensions_Visible [ (boolean_EXPRESSION) ];
3465
For the semantics of this pragma, see the entry for aspect @cite{Extensions_Visible}
3466
in the SPARK 2014 Reference Manual, section 6.1.7.
3468
@node Pragma External,Pragma External_Name_Casing,Pragma Extensions_Visible,Implementation Defined Pragmas
3469
@anchor{gnat_rm/implementation_defined_pragmas pragma-external}@anchor{5a}
3470
@section Pragma External
3477
[ Convention =>] convention_IDENTIFIER,
3478
[ Entity =>] LOCAL_NAME
3479
[, [External_Name =>] static_string_EXPRESSION ]
3480
[, [Link_Name =>] static_string_EXPRESSION ]);
3483
This pragma is identical in syntax and semantics to pragma
3484
@cite{Export} as defined in the Ada Reference Manual. It is
3485
provided for compatibility with some Ada 83 compilers that
3486
used this pragma for exactly the same purposes as pragma
3487
@cite{Export} before the latter was standardized.
3489
@node Pragma External_Name_Casing,Pragma Fast_Math,Pragma External,Implementation Defined Pragmas
3490
@anchor{gnat_rm/implementation_defined_pragmas pragma-external-name-casing}@anchor{5b}
3491
@section Pragma External_Name_Casing
3494
@geindex Dec Ada 83 casing compatibility
3496
@geindex External Names
3499
@geindex Casing of External names
3504
pragma External_Name_Casing (
3505
Uppercase | Lowercase
3506
[, Uppercase | Lowercase | As_Is]);
3509
This pragma provides control over the casing of external names associated
3510
with Import and Export pragmas. There are two cases to consider:
3516
Implicit external names
3518
Implicit external names are derived from identifiers. The most common case
3519
arises when a standard Ada Import or Export pragma is used with only two
3523
pragma Import (C, C_Routine);
3526
Since Ada is a case-insensitive language, the spelling of the identifier in
3527
the Ada source program does not provide any information on the desired
3528
casing of the external name, and so a convention is needed. In GNAT the
3529
default treatment is that such names are converted to all lower case
3530
letters. This corresponds to the normal C style in many environments.
3531
The first argument of pragma @cite{External_Name_Casing} can be used to
3532
control this treatment. If @cite{Uppercase} is specified, then the name
3533
will be forced to all uppercase letters. If @cite{Lowercase} is specified,
3534
then the normal default of all lower case letters will be used.
3536
This same implicit treatment is also used in the case of extended DEC Ada 83
3537
compatible Import and Export pragmas where an external name is explicitly
3538
specified using an identifier rather than a string.
3541
Explicit external names
3543
Explicit external names are given as string literals. The most common case
3544
arises when a standard Ada Import or Export pragma is used with three
3548
pragma Import (C, C_Routine, "C_routine");
3551
In this case, the string literal normally provides the exact casing required
3552
for the external name. The second argument of pragma
3553
@cite{External_Name_Casing} may be used to modify this behavior.
3554
If @cite{Uppercase} is specified, then the name
3555
will be forced to all uppercase letters. If @cite{Lowercase} is specified,
3556
then the name will be forced to all lowercase letters. A specification of
3557
@cite{As_Is} provides the normal default behavior in which the casing is
3558
taken from the string provided.
3561
This pragma may appear anywhere that a pragma is valid. In particular, it
3562
can be used as a configuration pragma in the @code{gnat.adc} file, in which
3563
case it applies to all subsequent compilations, or it can be used as a program
3564
unit pragma, in which case it only applies to the current unit, or it can
3565
be used more locally to control individual Import/Export pragmas.
3567
It was primarily intended for use with OpenVMS systems, where many
3568
compilers convert all symbols to upper case by default. For interfacing to
3569
such compilers (e.g., the DEC C compiler), it may be convenient to use
3573
pragma External_Name_Casing (Uppercase, Uppercase);
3576
to enforce the upper casing of all external symbols.
3578
@node Pragma Fast_Math,Pragma Favor_Top_Level,Pragma External_Name_Casing,Implementation Defined Pragmas
3579
@anchor{gnat_rm/implementation_defined_pragmas pragma-fast-math}@anchor{5c}
3580
@section Pragma Fast_Math
3589
This is a configuration pragma which activates a mode in which speed is
3590
considered more important for floating-point operations than absolutely
3591
accurate adherence to the requirements of the standard. Currently the
3592
following operations are affected:
3597
@item @emph{Complex Multiplication}
3599
The normal simple formula for complex multiplication can result in intermediate
3600
overflows for numbers near the end of the range. The Ada standard requires that
3601
this situation be detected and corrected by scaling, but in Fast_Math mode such
3602
cases will simply result in overflow. Note that to take advantage of this you
3603
must instantiate your own version of @cite{Ada.Numerics.Generic_Complex_Types}
3604
under control of the pragma, rather than use the preinstantiated versions.
3607
@node Pragma Favor_Top_Level,Pragma Finalize_Storage_Only,Pragma Fast_Math,Implementation Defined Pragmas
3608
@anchor{gnat_rm/implementation_defined_pragmas pragma-favor-top-level}@anchor{5d}
3609
@section Pragma Favor_Top_Level
3615
pragma Favor_Top_Level (type_NAME);
3618
The named type must be an access-to-subprogram type. This pragma is an
3619
efficiency hint to the compiler, regarding the use of 'Access or
3620
'Unrestricted_Access on nested (non-library-level) subprograms. The
3621
pragma means that nested subprograms are not used with this type, or
3622
are rare, so that the generated code should be efficient in the
3623
top-level case. When this pragma is used, dynamically generated
3624
trampolines may be used on some targets for nested subprograms.
3625
See also the No_Implicit_Dynamic_Code restriction.
3627
@node Pragma Finalize_Storage_Only,Pragma Float_Representation,Pragma Favor_Top_Level,Implementation Defined Pragmas
3628
@anchor{gnat_rm/implementation_defined_pragmas pragma-finalize-storage-only}@anchor{5e}
3629
@section Pragma Finalize_Storage_Only
3635
pragma Finalize_Storage_Only (first_subtype_LOCAL_NAME);
3638
This pragma allows the compiler not to emit a Finalize call for objects
3639
defined at the library level. This is mostly useful for types where
3640
finalization is only used to deal with storage reclamation since in most
3641
environments it is not necessary to reclaim memory just before terminating
3642
execution, hence the name.
3644
@node Pragma Float_Representation,Pragma Ghost,Pragma Finalize_Storage_Only,Implementation Defined Pragmas
3645
@anchor{gnat_rm/implementation_defined_pragmas pragma-float-representation}@anchor{5f}
3646
@section Pragma Float_Representation
3652
pragma Float_Representation (FLOAT_REP[, float_type_LOCAL_NAME]);
3654
FLOAT_REP ::= VAX_Float | IEEE_Float
3657
In the one argument form, this pragma is a configuration pragma which
3658
allows control over the internal representation chosen for the predefined
3659
floating point types declared in the packages @cite{Standard} and
3660
@cite{System}. This pragma is only provided for compatibility and has no effect.
3662
The two argument form specifies the representation to be used for
3663
the specified floating-point type. The argument must
3664
be @cite{IEEE_Float} to specify the use of IEEE format, as follows:
3670
For a digits value of 6, 32-bit IEEE short format will be used.
3673
For a digits value of 15, 64-bit IEEE long format will be used.
3676
No other value of digits is permitted.
3679
@node Pragma Ghost,Pragma Global,Pragma Float_Representation,Implementation Defined Pragmas
3680
@anchor{gnat_rm/implementation_defined_pragmas pragma-ghost}@anchor{60}
3681
@section Pragma Ghost
3687
pragma Ghost [ (boolean_EXPRESSION) ];
3690
For the semantics of this pragma, see the entry for aspect @cite{Ghost} in the SPARK
3691
2014 Reference Manual, section 6.9.
3693
@node Pragma Global,Pragma Ident,Pragma Ghost,Implementation Defined Pragmas
3694
@anchor{gnat_rm/implementation_defined_pragmas pragma-global}@anchor{61}
3695
@section Pragma Global
3701
pragma Global (GLOBAL_SPECIFICATION);
3703
GLOBAL_SPECIFICATION ::=
3706
| (MODED_GLOBAL_LIST @{, MODED_GLOBAL_LIST@})
3708
MODED_GLOBAL_LIST ::= MODE_SELECTOR => GLOBAL_LIST
3710
MODE_SELECTOR ::= In_Out | Input | Output | Proof_In
3711
GLOBAL_LIST ::= GLOBAL_ITEM | (GLOBAL_ITEM @{, GLOBAL_ITEM@})
3712
GLOBAL_ITEM ::= NAME
3715
For the semantics of this pragma, see the entry for aspect @cite{Global} in the
3716
SPARK 2014 Reference Manual, section 6.1.4.
3718
@node Pragma Ident,Pragma Ignore_Pragma,Pragma Global,Implementation Defined Pragmas
3719
@anchor{gnat_rm/implementation_defined_pragmas pragma-ident}@anchor{62}
3720
@section Pragma Ident
3726
pragma Ident (static_string_EXPRESSION);
3729
This pragma is identical in effect to pragma @cite{Comment}. It is provided
3730
for compatibility with other Ada compilers providing this pragma.
3732
@node Pragma Ignore_Pragma,Pragma Implementation_Defined,Pragma Ident,Implementation Defined Pragmas
3733
@anchor{gnat_rm/implementation_defined_pragmas pragma-ignore-pragma}@anchor{63}
3734
@section Pragma Ignore_Pragma
3740
pragma Ignore_Pragma (pragma_IDENTIFIER);
3743
This is a configuration pragma
3744
that takes a single argument that is a simple identifier. Any subsequent
3745
use of a pragma whose pragma identifier matches this argument will be
3746
silently ignored. This may be useful when legacy code or code intended
3747
for compilation with some other compiler contains pragmas that match the
3748
name, but not the exact implementation, of a @cite{GNAT} pragma. The use of this
3749
pragma allows such pragmas to be ignored, which may be useful in @cite{CodePeer}
3750
mode, or during porting of legacy code.
3752
@node Pragma Implementation_Defined,Pragma Implemented,Pragma Ignore_Pragma,Implementation Defined Pragmas
3753
@anchor{gnat_rm/implementation_defined_pragmas pragma-implementation-defined}@anchor{64}
3754
@section Pragma Implementation_Defined
3760
pragma Implementation_Defined (local_NAME);
3763
This pragma marks a previously declared entioty as implementation-defined.
3764
For an overloaded entity, applies to the most recent homonym.
3767
pragma Implementation_Defined;
3770
The form with no arguments appears anywhere within a scope, most
3771
typically a package spec, and indicates that all entities that are
3772
defined within the package spec are Implementation_Defined.
3774
This pragma is used within the GNAT runtime library to identify
3775
implementation-defined entities introduced in language-defined units,
3776
for the purpose of implementing the No_Implementation_Identifiers
3779
@node Pragma Implemented,Pragma Implicit_Packing,Pragma Implementation_Defined,Implementation Defined Pragmas
3780
@anchor{gnat_rm/implementation_defined_pragmas pragma-implemented}@anchor{65}
3781
@section Pragma Implemented
3787
pragma Implemented (procedure_LOCAL_NAME, implementation_kind);
3789
implementation_kind ::= By_Entry | By_Protected_Procedure | By_Any
3792
This is an Ada 2012 representation pragma which applies to protected, task
3793
and synchronized interface primitives. The use of pragma Implemented provides
3794
a way to impose a static requirement on the overriding operation by adhering
3795
to one of the three implementation kinds: entry, protected procedure or any of
3796
the above. This pragma is available in all earlier versions of Ada as an
3797
implementation-defined pragma.
3800
type Synch_Iface is synchronized interface;
3801
procedure Prim_Op (Obj : in out Iface) is abstract;
3802
pragma Implemented (Prim_Op, By_Protected_Procedure);
3804
protected type Prot_1 is new Synch_Iface with
3805
procedure Prim_Op; -- Legal
3808
protected type Prot_2 is new Synch_Iface with
3809
entry Prim_Op; -- Illegal
3812
task type Task_Typ is new Synch_Iface with
3813
entry Prim_Op; -- Illegal
3817
When applied to the procedure_or_entry_NAME of a requeue statement, pragma
3818
Implemented determines the runtime behavior of the requeue. Implementation kind
3819
By_Entry guarantees that the action of requeueing will proceed from an entry to
3820
another entry. Implementation kind By_Protected_Procedure transforms the
3821
requeue into a dispatching call, thus eliminating the chance of blocking. Kind
3822
By_Any shares the behavior of By_Entry and By_Protected_Procedure depending on
3823
the target's overriding subprogram kind.
3825
@node Pragma Implicit_Packing,Pragma Import_Function,Pragma Implemented,Implementation Defined Pragmas
3826
@anchor{gnat_rm/implementation_defined_pragmas pragma-implicit-packing}@anchor{66}
3827
@section Pragma Implicit_Packing
3830
@geindex Rational Profile
3835
pragma Implicit_Packing;
3838
This is a configuration pragma that requests implicit packing for packed
3839
arrays for which a size clause is given but no explicit pragma Pack or
3840
specification of Component_Size is present. It also applies to records
3841
where no record representation clause is present. Consider this example:
3844
type R is array (0 .. 7) of Boolean;
3848
In accordance with the recommendation in the RM (RM 13.3(53)), a Size clause
3849
does not change the layout of a composite object. So the Size clause in the
3850
above example is normally rejected, since the default layout of the array uses
3851
8-bit components, and thus the array requires a minimum of 64 bits.
3853
If this declaration is compiled in a region of code covered by an occurrence
3854
of the configuration pragma Implicit_Packing, then the Size clause in this
3855
and similar examples will cause implicit packing and thus be accepted. For
3856
this implicit packing to occur, the type in question must be an array of small
3857
components whose size is known at compile time, and the Size clause must
3858
specify the exact size that corresponds to the number of elements in the array
3859
multiplied by the size in bits of the component type (both single and
3860
multi-dimensioned arrays can be controlled with this pragma).
3862
@geindex Array packing
3864
Similarly, the following example shows the use in the record case
3868
a, b, c, d, e, f, g, h : boolean;
3874
Without a pragma Pack, each Boolean field requires 8 bits, so the
3875
minimum size is 72 bits, but with a pragma Pack, 16 bits would be
3876
sufficient. The use of pragma Implicit_Packing allows this record
3877
declaration to compile without an explicit pragma Pack.
3879
@node Pragma Import_Function,Pragma Import_Object,Pragma Implicit_Packing,Implementation Defined Pragmas
3880
@anchor{gnat_rm/implementation_defined_pragmas pragma-import-function}@anchor{67}
3881
@section Pragma Import_Function
3887
pragma Import_Function (
3888
[Internal =>] LOCAL_NAME,
3889
[, [External =>] EXTERNAL_SYMBOL]
3890
[, [Parameter_Types =>] PARAMETER_TYPES]
3891
[, [Result_Type =>] SUBTYPE_MARK]
3892
[, [Mechanism =>] MECHANISM]
3893
[, [Result_Mechanism =>] MECHANISM_NAME]);
3897
| static_string_EXPRESSION
3901
| TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
3905
| subtype_Name ' Access
3909
| (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
3911
MECHANISM_ASSOCIATION ::=
3912
[formal_parameter_NAME =>] MECHANISM_NAME
3919
This pragma is used in conjunction with a pragma @cite{Import} to
3920
specify additional information for an imported function. The pragma
3921
@cite{Import} (or equivalent pragma @cite{Interface}) must precede the
3922
@cite{Import_Function} pragma and both must appear in the same
3923
declarative part as the function specification.
3925
The @cite{Internal} argument must uniquely designate
3926
the function to which the
3927
pragma applies. If more than one function name exists of this name in
3928
the declarative part you must use the @cite{Parameter_Types} and
3929
@cite{Result_Type} parameters to achieve the required unique
3930
designation. Subtype marks in these parameters must exactly match the
3931
subtypes in the corresponding function specification, using positional
3932
notation to match parameters with subtype marks.
3933
The form with an @cite{'Access} attribute can be used to match an
3934
anonymous access parameter.
3936
You may optionally use the @cite{Mechanism} and @cite{Result_Mechanism}
3937
parameters to specify passing mechanisms for the
3938
parameters and result. If you specify a single mechanism name, it
3939
applies to all parameters. Otherwise you may specify a mechanism on a
3940
parameter by parameter basis using either positional or named
3941
notation. If the mechanism is not specified, the default mechanism
3944
@node Pragma Import_Object,Pragma Import_Procedure,Pragma Import_Function,Implementation Defined Pragmas
3945
@anchor{gnat_rm/implementation_defined_pragmas pragma-import-object}@anchor{68}
3946
@section Pragma Import_Object
3952
pragma Import_Object
3953
[Internal =>] LOCAL_NAME
3954
[, [External =>] EXTERNAL_SYMBOL]
3955
[, [Size =>] EXTERNAL_SYMBOL]);
3959
| static_string_EXPRESSION
3962
This pragma designates an object as imported, and apart from the
3963
extended rules for external symbols, is identical in effect to the use of
3964
the normal @cite{Import} pragma applied to an object. Unlike the
3965
subprogram case, you need not use a separate @cite{Import} pragma,
3966
although you may do so (and probably should do so from a portability
3967
point of view). @cite{size} is syntax checked, but otherwise ignored by
3970
@node Pragma Import_Procedure,Pragma Import_Valued_Procedure,Pragma Import_Object,Implementation Defined Pragmas
3971
@anchor{gnat_rm/implementation_defined_pragmas pragma-import-procedure}@anchor{69}
3972
@section Pragma Import_Procedure
3978
pragma Import_Procedure (
3979
[Internal =>] LOCAL_NAME
3980
[, [External =>] EXTERNAL_SYMBOL]
3981
[, [Parameter_Types =>] PARAMETER_TYPES]
3982
[, [Mechanism =>] MECHANISM]);
3986
| static_string_EXPRESSION
3990
| TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
3994
| subtype_Name ' Access
3998
| (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
4000
MECHANISM_ASSOCIATION ::=
4001
[formal_parameter_NAME =>] MECHANISM_NAME
4003
MECHANISM_NAME ::= Value | Reference
4006
This pragma is identical to @cite{Import_Function} except that it
4007
applies to a procedure rather than a function and the parameters
4008
@cite{Result_Type} and @cite{Result_Mechanism} are not permitted.
4010
@node Pragma Import_Valued_Procedure,Pragma Independent,Pragma Import_Procedure,Implementation Defined Pragmas
4011
@anchor{gnat_rm/implementation_defined_pragmas pragma-import-valued-procedure}@anchor{6a}
4012
@section Pragma Import_Valued_Procedure
4018
pragma Import_Valued_Procedure (
4019
[Internal =>] LOCAL_NAME
4020
[, [External =>] EXTERNAL_SYMBOL]
4021
[, [Parameter_Types =>] PARAMETER_TYPES]
4022
[, [Mechanism =>] MECHANISM]);
4026
| static_string_EXPRESSION
4030
| TYPE_DESIGNATOR @{, TYPE_DESIGNATOR@}
4034
| subtype_Name ' Access
4038
| (MECHANISM_ASSOCIATION @{, MECHANISM_ASSOCIATION@})
4040
MECHANISM_ASSOCIATION ::=
4041
[formal_parameter_NAME =>] MECHANISM_NAME
4043
MECHANISM_NAME ::= Value | Reference
4046
This pragma is identical to @cite{Import_Procedure} except that the
4047
first parameter of @cite{LOCAL_NAME}, which must be present, must be of
4048
mode @cite{OUT}, and externally the subprogram is treated as a function
4049
with this parameter as the result of the function. The purpose of this
4050
capability is to allow the use of @cite{OUT} and @cite{IN OUT}
4051
parameters in interfacing to external functions (which are not permitted
4052
in Ada functions). You may optionally use the @cite{Mechanism}
4053
parameters to specify passing mechanisms for the parameters.
4054
If you specify a single mechanism name, it applies to all parameters.
4055
Otherwise you may specify a mechanism on a parameter by parameter
4056
basis using either positional or named notation. If the mechanism is not
4057
specified, the default mechanism is used.
4059
Note that it is important to use this pragma in conjunction with a separate
4060
pragma Import that specifies the desired convention, since otherwise the
4061
default convention is Ada, which is almost certainly not what is required.
4063
@node Pragma Independent,Pragma Independent_Components,Pragma Import_Valued_Procedure,Implementation Defined Pragmas
4064
@anchor{gnat_rm/implementation_defined_pragmas pragma-independent}@anchor{6b}
4065
@section Pragma Independent
4071
pragma Independent (Local_NAME);
4074
This pragma is standard in Ada 2012 mode (which also provides an aspect
4075
of the same name). It is also available as an implementation-defined
4076
pragma in all earlier versions. It specifies that the
4077
designated object or all objects of the designated type must be
4078
independently addressable. This means that separate tasks can safely
4079
manipulate such objects. For example, if two components of a record are
4080
independent, then two separate tasks may access these two components.
4082
constraints on the representation of the object (for instance prohibiting
4085
@node Pragma Independent_Components,Pragma Initial_Condition,Pragma Independent,Implementation Defined Pragmas
4086
@anchor{gnat_rm/implementation_defined_pragmas pragma-independent-components}@anchor{6c}
4087
@section Pragma Independent_Components
4093
pragma Independent_Components (Local_NAME);
4096
This pragma is standard in Ada 2012 mode (which also provides an aspect
4097
of the same name). It is also available as an implementation-defined
4098
pragma in all earlier versions. It specifies that the components of the
4099
designated object, or the components of each object of the designated
4101
independently addressable. This means that separate tasks can safely
4102
manipulate separate components in the composite object. This may place
4103
constraints on the representation of the object (for instance prohibiting
4106
@node Pragma Initial_Condition,Pragma Initialize_Scalars,Pragma Independent_Components,Implementation Defined Pragmas
4107
@anchor{gnat_rm/implementation_defined_pragmas pragma-initial-condition}@anchor{6d}
4108
@section Pragma Initial_Condition
4114
pragma Initial_Condition (boolean_EXPRESSION);
4117
For the semantics of this pragma, see the entry for aspect @cite{Initial_Condition}
4118
in the SPARK 2014 Reference Manual, section 7.1.6.
4120
@node Pragma Initialize_Scalars,Pragma Initializes,Pragma Initial_Condition,Implementation Defined Pragmas
4121
@anchor{gnat_rm/implementation_defined_pragmas pragma-initialize-scalars}@anchor{6e}
4122
@section Pragma Initialize_Scalars
4125
@geindex debugging with Initialize_Scalars
4130
pragma Initialize_Scalars;
4133
This pragma is similar to @cite{Normalize_Scalars} conceptually but has
4134
two important differences. First, there is no requirement for the pragma
4135
to be used uniformly in all units of a partition, in particular, it is fine
4136
to use this just for some or all of the application units of a partition,
4137
without needing to recompile the run-time library.
4139
In the case where some units are compiled with the pragma, and some without,
4140
then a declaration of a variable where the type is defined in package
4141
Standard or is locally declared will always be subject to initialization,
4142
as will any declaration of a scalar variable. For composite variables,
4143
whether the variable is initialized may also depend on whether the package
4144
in which the type of the variable is declared is compiled with the pragma.
4146
The other important difference is that you can control the value used
4147
for initializing scalar objects. At bind time, you can select several
4148
options for initialization. You can
4149
initialize with invalid values (similar to Normalize_Scalars, though for
4150
Initialize_Scalars it is not always possible to determine the invalid
4151
values in complex cases like signed component fields with non-standard
4152
sizes). You can also initialize with high or
4153
low values, or with a specified bit pattern. See the GNAT
4154
User's Guide for binder options for specifying these cases.
4156
This means that you can compile a program, and then without having to
4157
recompile the program, you can run it with different values being used
4158
for initializing otherwise uninitialized values, to test if your program
4159
behavior depends on the choice. Of course the behavior should not change,
4160
and if it does, then most likely you have an incorrect reference to an
4161
uninitialized value.
4163
It is even possible to change the value at execution time eliminating even
4164
the need to rebind with a different switch using an environment variable.
4165
See the GNAT User's Guide for details.
4167
Note that pragma @cite{Initialize_Scalars} is particularly useful in
4168
conjunction with the enhanced validity checking that is now provided
4169
in GNAT, which checks for invalid values under more conditions.
4170
Using this feature (see description of the @emph{-gnatV} flag in the
4171
GNAT User's Guide) in conjunction with
4172
pragma @cite{Initialize_Scalars}
4173
provides a powerful new tool to assist in the detection of problems
4174
caused by uninitialized variables.
4176
Note: the use of @cite{Initialize_Scalars} has a fairly extensive
4177
effect on the generated code. This may cause your code to be
4178
substantially larger. It may also cause an increase in the amount
4179
of stack required, so it is probably a good idea to turn on stack
4180
checking (see description of stack checking in the GNAT
4181
User's Guide) when using this pragma.
4183
@node Pragma Initializes,Pragma Inline_Always,Pragma Initialize_Scalars,Implementation Defined Pragmas
4184
@anchor{gnat_rm/implementation_defined_pragmas pragma-initializes}@anchor{6f}
4185
@section Pragma Initializes
4191
pragma Initializes (INITIALIZATION_LIST);
4193
INITIALIZATION_LIST ::=
4195
| (INITIALIZATION_ITEM @{, INITIALIZATION_ITEM@})
4197
INITIALIZATION_ITEM ::= name [=> INPUT_LIST]
4202
| (INPUT @{, INPUT@})
4207
For the semantics of this pragma, see the entry for aspect @cite{Initializes} in the
4208
SPARK 2014 Reference Manual, section 7.1.5.
4210
@node Pragma Inline_Always,Pragma Inline_Generic,Pragma Initializes,Implementation Defined Pragmas
4211
@anchor{gnat_rm/implementation_defined_pragmas pragma-inline-always}@anchor{70}
4212
@section Pragma Inline_Always
4218
pragma Inline_Always (NAME [, NAME]);
4221
Similar to pragma @cite{Inline} except that inlining is not subject to
4222
the use of option @emph{-gnatn} or @emph{-gnatN} and the inlining
4223
happens regardless of whether these options are used.
4225
@node Pragma Inline_Generic,Pragma Interface,Pragma Inline_Always,Implementation Defined Pragmas
4226
@anchor{gnat_rm/implementation_defined_pragmas pragma-inline-generic}@anchor{71}
4227
@section Pragma Inline_Generic
4233
pragma Inline_Generic (GNAME @{, GNAME@});
4235
GNAME ::= generic_unit_NAME | generic_instance_NAME
4238
This pragma is provided for compatibility with Dec Ada 83. It has
4239
no effect in @cite{GNAT} (which always inlines generics), other
4240
than to check that the given names are all names of generic units or
4243
@node Pragma Interface,Pragma Interface_Name,Pragma Inline_Generic,Implementation Defined Pragmas
4244
@anchor{gnat_rm/implementation_defined_pragmas pragma-interface}@anchor{72}
4245
@section Pragma Interface
4252
[Convention =>] convention_identifier,
4253
[Entity =>] local_NAME
4254
[, [External_Name =>] static_string_expression]
4255
[, [Link_Name =>] static_string_expression]);
4258
This pragma is identical in syntax and semantics to
4259
the standard Ada pragma @cite{Import}. It is provided for compatibility
4260
with Ada 83. The definition is upwards compatible both with pragma
4261
@cite{Interface} as defined in the Ada 83 Reference Manual, and also
4262
with some extended implementations of this pragma in certain Ada 83
4263
implementations. The only difference between pragma @cite{Interface}
4264
and pragma @cite{Import} is that there is special circuitry to allow
4265
both pragmas to appear for the same subprogram entity (normally it
4266
is illegal to have multiple @cite{Import} pragmas. This is useful in
4267
maintaining Ada 83/Ada 95 compatibility and is compatible with other
4270
@node Pragma Interface_Name,Pragma Interrupt_Handler,Pragma Interface,Implementation Defined Pragmas
4271
@anchor{gnat_rm/implementation_defined_pragmas pragma-interface-name}@anchor{73}
4272
@section Pragma Interface_Name
4278
pragma Interface_Name (
4279
[Entity =>] LOCAL_NAME
4280
[, [External_Name =>] static_string_EXPRESSION]
4281
[, [Link_Name =>] static_string_EXPRESSION]);
4284
This pragma provides an alternative way of specifying the interface name
4285
for an interfaced subprogram, and is provided for compatibility with Ada
4286
83 compilers that use the pragma for this purpose. You must provide at
4287
least one of @cite{External_Name} or @cite{Link_Name}.
4289
@node Pragma Interrupt_Handler,Pragma Interrupt_State,Pragma Interface_Name,Implementation Defined Pragmas
4290
@anchor{gnat_rm/implementation_defined_pragmas pragma-interrupt-handler}@anchor{74}
4291
@section Pragma Interrupt_Handler
4297
pragma Interrupt_Handler (procedure_LOCAL_NAME);
4300
This program unit pragma is supported for parameterless protected procedures
4301
as described in Annex C of the Ada Reference Manual. On the AAMP target
4302
the pragma can also be specified for nonprotected parameterless procedures
4303
that are declared at the library level (which includes procedures
4304
declared at the top level of a library package). In the case of AAMP,
4305
when this pragma is applied to a nonprotected procedure, the instruction
4306
@cite{IERET} is generated for returns from the procedure, enabling
4307
maskable interrupts, in place of the normal return instruction.
4309
@node Pragma Interrupt_State,Pragma Invariant,Pragma Interrupt_Handler,Implementation Defined Pragmas
4310
@anchor{gnat_rm/implementation_defined_pragmas pragma-interrupt-state}@anchor{75}
4311
@section Pragma Interrupt_State
4317
pragma Interrupt_State
4319
[State =>] SYSTEM | RUNTIME | USER);
4322
Normally certain interrupts are reserved to the implementation. Any attempt
4323
to attach an interrupt causes Program_Error to be raised, as described in
4324
RM C.3.2(22). A typical example is the @cite{SIGINT} interrupt used in
4325
many systems for an @code{Ctrl-C} interrupt. Normally this interrupt is
4326
reserved to the implementation, so that @code{Ctrl-C} can be used to
4327
interrupt execution. Additionally, signals such as @cite{SIGSEGV},
4328
@cite{SIGABRT}, @cite{SIGFPE} and @cite{SIGILL} are often mapped to specific
4329
Ada exceptions, or used to implement run-time functions such as the
4330
@cite{abort} statement and stack overflow checking.
4332
Pragma @cite{Interrupt_State} provides a general mechanism for overriding
4333
such uses of interrupts. It subsumes the functionality of pragma
4334
@cite{Unreserve_All_Interrupts}. Pragma @cite{Interrupt_State} is not
4335
available on Windows or VMS. On all other platforms than VxWorks,
4336
it applies to signals; on VxWorks, it applies to vectored hardware interrupts
4337
and may be used to mark interrupts required by the board support package
4340
Interrupts can be in one of three states:
4348
The interrupt is reserved (no Ada handler can be installed), and the
4349
Ada run-time may not install a handler. As a result you are guaranteed
4350
standard system default action if this interrupt is raised.
4355
The interrupt is reserved (no Ada handler can be installed). The run time
4356
is allowed to install a handler for internal control purposes, but is
4357
not required to do so.
4362
The interrupt is unreserved. The user may install a handler to provide
4366
These states are the allowed values of the @cite{State} parameter of the
4367
pragma. The @cite{Name} parameter is a value of the type
4368
@cite{Ada.Interrupts.Interrupt_ID}. Typically, it is a name declared in
4369
@cite{Ada.Interrupts.Names}.
4371
This is a configuration pragma, and the binder will check that there
4372
are no inconsistencies between different units in a partition in how a
4373
given interrupt is specified. It may appear anywhere a pragma is legal.
4375
The effect is to move the interrupt to the specified state.
4377
By declaring interrupts to be SYSTEM, you guarantee the standard system
4378
action, such as a core dump.
4380
By declaring interrupts to be USER, you guarantee that you can install
4383
Note that certain signals on many operating systems cannot be caught and
4384
handled by applications. In such cases, the pragma is ignored. See the
4385
operating system documentation, or the value of the array @cite{Reserved}
4386
declared in the spec of package @cite{System.OS_Interface}.
4388
Overriding the default state of signals used by the Ada runtime may interfere
4389
with an application's runtime behavior in the cases of the synchronous signals,
4390
and in the case of the signal used to implement the @cite{abort} statement.
4392
@node Pragma Invariant,Pragma Keep_Names,Pragma Interrupt_State,Implementation Defined Pragmas
4393
@anchor{gnat_rm/implementation_defined_pragmas pragma-invariant}@anchor{76}
4394
@section Pragma Invariant
4401
([Entity =>] private_type_LOCAL_NAME,
4402
[Check =>] EXPRESSION
4403
[,[Message =>] String_Expression]);
4406
This pragma provides exactly the same capabilities as the Type_Invariant aspect
4407
defined in AI05-0146-1, and in the Ada 2012 Reference Manual. The
4408
Type_Invariant aspect is fully implemented in Ada 2012 mode, but since it
4409
requires the use of the aspect syntax, which is not available except in 2012
4410
mode, it is not possible to use the Type_Invariant aspect in earlier versions
4411
of Ada. However the Invariant pragma may be used in any version of Ada. Also
4412
note that the aspect Invariant is a synonym in GNAT for the aspect
4413
Type_Invariant, but there is no pragma Type_Invariant.
4415
The pragma must appear within the visible part of the package specification,
4416
after the type to which its Entity argument appears. As with the Invariant
4417
aspect, the Check expression is not analyzed until the end of the visible
4418
part of the package, so it may contain forward references. The Message
4419
argument, if present, provides the exception message used if the invariant
4420
is violated. If no Message parameter is provided, a default message that
4421
identifies the line on which the pragma appears is used.
4423
It is permissible to have multiple Invariants for the same type entity, in
4424
which case they are and'ed together. It is permissible to use this pragma
4425
in Ada 2012 mode, but you cannot have both an invariant aspect and an
4426
invariant pragma for the same entity.
4428
For further details on the use of this pragma, see the Ada 2012 documentation
4429
of the Type_Invariant aspect.
4431
@node Pragma Keep_Names,Pragma License,Pragma Invariant,Implementation Defined Pragmas
4432
@anchor{gnat_rm/implementation_defined_pragmas pragma-keep-names}@anchor{77}
4433
@section Pragma Keep_Names
4439
pragma Keep_Names ([On =>] enumeration_first_subtype_LOCAL_NAME);
4442
The @cite{LOCAL_NAME} argument
4443
must refer to an enumeration first subtype
4444
in the current declarative part. The effect is to retain the enumeration
4445
literal names for use by @cite{Image} and @cite{Value} even if a global
4446
@cite{Discard_Names} pragma applies. This is useful when you want to
4447
generally suppress enumeration literal names and for example you therefore
4448
use a @cite{Discard_Names} pragma in the @code{gnat.adc} file, but you
4449
want to retain the names for specific enumeration types.
4451
@node Pragma License,Pragma Link_With,Pragma Keep_Names,Implementation Defined Pragmas
4452
@anchor{gnat_rm/implementation_defined_pragmas pragma-license}@anchor{78}
4453
@section Pragma License
4456
@geindex License checking
4461
pragma License (Unrestricted | GPL | Modified_GPL | Restricted);
4464
This pragma is provided to allow automated checking for appropriate license
4465
conditions with respect to the standard and modified GPL. A pragma
4466
@cite{License}, which is a configuration pragma that typically appears at
4467
the start of a source file or in a separate @code{gnat.adc} file, specifies
4468
the licensing conditions of a unit as follows:
4475
This is used for a unit that can be freely used with no license restrictions.
4476
Examples of such units are public domain units, and units from the Ada
4481
This is used for a unit that is licensed under the unmodified GPL, and which
4482
therefore cannot be @cite{with}'ed by a restricted unit.
4486
This is used for a unit licensed under the GNAT modified GPL that includes
4487
a special exception paragraph that specifically permits the inclusion of
4488
the unit in programs without requiring the entire program to be released
4493
This is used for a unit that is restricted in that it is not permitted to
4494
depend on units that are licensed under the GPL. Typical examples are
4495
proprietary code that is to be released under more restrictive license
4496
conditions. Note that restricted units are permitted to @cite{with} units
4497
which are licensed under the modified GPL (this is the whole point of the
4501
Normally a unit with no @cite{License} pragma is considered to have an
4502
unknown license, and no checking is done. However, standard GNAT headers
4503
are recognized, and license information is derived from them as follows.
4505
A GNAT license header starts with a line containing 78 hyphens. The following
4506
comment text is searched for the appearance of any of the following strings.
4508
If the string 'GNU General Public License' is found, then the unit is assumed
4509
to have GPL license, unless the string 'As a special exception' follows, in
4510
which case the license is assumed to be modified GPL.
4512
If one of the strings
4513
'This specification is adapted from the Ada Semantic Interface' or
4514
'This specification is derived from the Ada Reference Manual' is found
4515
then the unit is assumed to be unrestricted.
4517
These default actions means that a program with a restricted license pragma
4518
will automatically get warnings if a GPL unit is inappropriately
4519
@cite{with}'ed. For example, the program:
4524
procedure Secret_Stuff is
4529
if compiled with pragma @cite{License} (@cite{Restricted}) in a
4530
@code{gnat.adc} file will generate the warning:
4535
>>> license of withed unit "Sem_Ch3" is incompatible
4537
2. with GNAT.Sockets;
4538
3. procedure Secret_Stuff is
4541
Here we get a warning on @cite{Sem_Ch3} since it is part of the GNAT
4542
compiler and is licensed under the
4543
GPL, but no warning for @cite{GNAT.Sockets} which is part of the GNAT
4544
run time, and is therefore licensed under the modified GPL.
4546
@node Pragma Link_With,Pragma Linker_Alias,Pragma License,Implementation Defined Pragmas
4547
@anchor{gnat_rm/implementation_defined_pragmas pragma-link-with}@anchor{79}
4548
@section Pragma Link_With
4554
pragma Link_With (static_string_EXPRESSION @{,static_string_EXPRESSION@});
4557
This pragma is provided for compatibility with certain Ada 83 compilers.
4558
It has exactly the same effect as pragma @cite{Linker_Options} except
4559
that spaces occurring within one of the string expressions are treated
4560
as separators. For example, in the following case:
4563
pragma Link_With ("-labc -ldef");
4566
results in passing the strings @cite{-labc} and @cite{-ldef} as two
4567
separate arguments to the linker. In addition pragma Link_With allows
4568
multiple arguments, with the same effect as successive pragmas.
4570
@node Pragma Linker_Alias,Pragma Linker_Constructor,Pragma Link_With,Implementation Defined Pragmas
4571
@anchor{gnat_rm/implementation_defined_pragmas pragma-linker-alias}@anchor{7a}
4572
@section Pragma Linker_Alias
4578
pragma Linker_Alias (
4579
[Entity =>] LOCAL_NAME,
4580
[Target =>] static_string_EXPRESSION);
4583
@cite{LOCAL_NAME} must refer to an object that is declared at the library
4584
level. This pragma establishes the given entity as a linker alias for the
4585
given target. It is equivalent to @cite{__attribute__((alias))} in GNU C
4586
and causes @cite{LOCAL_NAME} to be emitted as an alias for the symbol
4587
@cite{static_string_EXPRESSION} in the object file, that is to say no space
4588
is reserved for @cite{LOCAL_NAME} by the assembler and it will be resolved
4589
to the same address as @cite{static_string_EXPRESSION} by the linker.
4591
The actual linker name for the target must be used (e.g., the fully
4592
encoded name with qualification in Ada, or the mangled name in C++),
4593
or it must be declared using the C convention with @cite{pragma Import}
4594
or @cite{pragma Export}.
4596
Not all target machines support this pragma. On some of them it is accepted
4597
only if @cite{pragma Weak_External} has been applied to @cite{LOCAL_NAME}.
4600
-- Example of the use of pragma Linker_Alias
4604
pragma Export (C, i);
4606
new_name_for_i : Integer;
4607
pragma Linker_Alias (new_name_for_i, "i");
4611
@node Pragma Linker_Constructor,Pragma Linker_Destructor,Pragma Linker_Alias,Implementation Defined Pragmas
4612
@anchor{gnat_rm/implementation_defined_pragmas pragma-linker-constructor}@anchor{7b}
4613
@section Pragma Linker_Constructor
4619
pragma Linker_Constructor (procedure_LOCAL_NAME);
4622
@cite{procedure_LOCAL_NAME} must refer to a parameterless procedure that
4623
is declared at the library level. A procedure to which this pragma is
4624
applied will be treated as an initialization routine by the linker.
4625
It is equivalent to @cite{__attribute__((constructor))} in GNU C and
4626
causes @cite{procedure_LOCAL_NAME} to be invoked before the entry point
4627
of the executable is called (or immediately after the shared library is
4628
loaded if the procedure is linked in a shared library), in particular
4629
before the Ada run-time environment is set up.
4631
Because of these specific contexts, the set of operations such a procedure
4632
can perform is very limited and the type of objects it can manipulate is
4633
essentially restricted to the elementary types. In particular, it must only
4634
contain code to which pragma Restrictions (No_Elaboration_Code) applies.
4636
This pragma is used by GNAT to implement auto-initialization of shared Stand
4637
Alone Libraries, which provides a related capability without the restrictions
4638
listed above. Where possible, the use of Stand Alone Libraries is preferable
4639
to the use of this pragma.
4641
@node Pragma Linker_Destructor,Pragma Linker_Section,Pragma Linker_Constructor,Implementation Defined Pragmas
4642
@anchor{gnat_rm/implementation_defined_pragmas pragma-linker-destructor}@anchor{7c}
4643
@section Pragma Linker_Destructor
4649
pragma Linker_Destructor (procedure_LOCAL_NAME);
4652
@cite{procedure_LOCAL_NAME} must refer to a parameterless procedure that
4653
is declared at the library level. A procedure to which this pragma is
4654
applied will be treated as a finalization routine by the linker.
4655
It is equivalent to @cite{__attribute__((destructor))} in GNU C and
4656
causes @cite{procedure_LOCAL_NAME} to be invoked after the entry point
4657
of the executable has exited (or immediately before the shared library
4658
is unloaded if the procedure is linked in a shared library), in particular
4659
after the Ada run-time environment is shut down.
4661
See @cite{pragma Linker_Constructor} for the set of restrictions that apply
4662
because of these specific contexts.
4664
@node Pragma Linker_Section,Pragma Lock_Free,Pragma Linker_Destructor,Implementation Defined Pragmas
4665
@anchor{gnat_rm/implementation_defined_pragmas pragma-linker-section}@anchor{7d}
4666
@section Pragma Linker_Section
4672
pragma Linker_Section (
4673
[Entity =>] LOCAL_NAME,
4674
[Section =>] static_string_EXPRESSION);
4677
@cite{LOCAL_NAME} must refer to an object, type, or subprogram that is
4678
declared at the library level. This pragma specifies the name of the
4679
linker section for the given entity. It is equivalent to
4680
@cite{__attribute__((section))} in GNU C and causes @cite{LOCAL_NAME} to
4681
be placed in the @cite{static_string_EXPRESSION} section of the
4682
executable (assuming the linker doesn't rename the section).
4683
GNAT also provides an implementation defined aspect of the same name.
4685
In the case of specifying this aspect for a type, the effect is to
4686
specify the corresponding for all library level objects of the type which
4687
do not have an explicit linker section set. Note that this only applies to
4688
whole objects, not to components of composite objects.
4690
In the case of a subprogram, the linker section applies to all previously
4691
declared matching overloaded subprograms in the current declarative part
4692
which do not already have a linker section assigned. The linker section
4693
aspect is useful in this case for specifying different linker sections
4694
for different elements of such an overloaded set.
4696
Note that an empty string specifies that no linker section is specified.
4697
This is not quite the same as omitting the pragma or aspect, since it
4698
can be used to specify that one element of an overloaded set of subprograms
4699
has the default linker section, or that one object of a type for which a
4700
linker section is specified should has the default linker section.
4702
The compiler normally places library-level entities in standard sections
4703
depending on the class: procedures and functions generally go in the
4704
@cite{.text} section, initialized variables in the @cite{.data} section
4705
and uninitialized variables in the @cite{.bss} section.
4707
Other, special sections may exist on given target machines to map special
4708
hardware, for example I/O ports or flash memory. This pragma is a means to
4709
defer the final layout of the executable to the linker, thus fully working
4710
at the symbolic level with the compiler.
4712
Some file formats do not support arbitrary sections so not all target
4713
machines support this pragma. The use of this pragma may cause a program
4714
execution to be erroneous if it is used to place an entity into an
4715
inappropriate section (e.g., a modified variable into the @cite{.text}
4716
section). See also @cite{pragma Persistent_BSS}.
4719
-- Example of the use of pragma Linker_Section
4723
pragma Volatile (Port_A);
4724
pragma Linker_Section (Port_A, ".bss.port_a");
4727
pragma Volatile (Port_B);
4728
pragma Linker_Section (Port_B, ".bss.port_b");
4730
type Port_Type is new Integer with Linker_Section => ".bss";
4731
PA : Port_Type with Linker_Section => ".bss.PA";
4732
PB : Port_Type; -- ends up in linker section ".bss"
4734
procedure Q with Linker_Section => "Qsection";
4738
@node Pragma Lock_Free,Pragma Loop_Invariant,Pragma Linker_Section,Implementation Defined Pragmas
4739
@anchor{gnat_rm/implementation_defined_pragmas pragma-lock-free}@anchor{7e}
4740
@section Pragma Lock_Free
4744
This pragma may be specified for protected types or objects. It specifies that
4745
the implementation of protected operations must be implemented without locks.
4746
Compilation fails if the compiler cannot generate lock-free code for the
4749
@node Pragma Loop_Invariant,Pragma Loop_Optimize,Pragma Lock_Free,Implementation Defined Pragmas
4750
@anchor{gnat_rm/implementation_defined_pragmas pragma-loop-invariant}@anchor{7f}
4751
@section Pragma Loop_Invariant
4757
pragma Loop_Invariant ( boolean_EXPRESSION );
4760
The effect of this pragma is similar to that of pragma @cite{Assert},
4761
except that in an @cite{Assertion_Policy} pragma, the identifier
4762
@cite{Loop_Invariant} is used to control whether it is ignored or checked
4765
@cite{Loop_Invariant} can only appear as one of the items in the sequence
4766
of statements of a loop body, or nested inside block statements that
4767
appear in the sequence of statements of a loop body.
4768
The intention is that it be used to
4769
represent a "loop invariant" assertion, i.e. something that is true each
4770
time through the loop, and which can be used to show that the loop is
4771
achieving its purpose.
4773
Multiple @cite{Loop_Invariant} and @cite{Loop_Variant} pragmas that
4774
apply to the same loop should be grouped in the same sequence of
4777
To aid in writing such invariants, the special attribute @cite{Loop_Entry}
4778
may be used to refer to the value of an expression on entry to the loop. This
4779
attribute can only be used within the expression of a @cite{Loop_Invariant}
4780
pragma. For full details, see documentation of attribute @cite{Loop_Entry}.
4782
@node Pragma Loop_Optimize,Pragma Loop_Variant,Pragma Loop_Invariant,Implementation Defined Pragmas
4783
@anchor{gnat_rm/implementation_defined_pragmas pragma-loop-optimize}@anchor{80}
4784
@section Pragma Loop_Optimize
4790
pragma Loop_Optimize (OPTIMIZATION_HINT @{, OPTIMIZATION_HINT@});
4792
OPTIMIZATION_HINT ::= Ivdep | No_Unroll | Unroll | No_Vector | Vector
4795
This pragma must appear immediately within a loop statement. It allows the
4796
programmer to specify optimization hints for the enclosing loop. The hints
4797
are not mutually exclusive and can be freely mixed, but not all combinations
4798
will yield a sensible outcome.
4800
There are five supported optimization hints for a loop:
4808
The programmer asserts that there are no loop-carried dependencies
4809
which would prevent consecutive iterations of the loop from being
4810
executed simultaneously.
4815
The loop must not be unrolled. This is a strong hint: the compiler will not
4816
unroll a loop marked with this hint.
4821
The loop should be unrolled. This is a weak hint: the compiler will try to
4822
apply unrolling to this loop preferably to other optimizations, notably
4823
vectorization, but there is no guarantee that the loop will be unrolled.
4828
The loop must not be vectorized. This is a strong hint: the compiler will not
4829
vectorize a loop marked with this hint.
4834
The loop should be vectorized. This is a weak hint: the compiler will try to
4835
apply vectorization to this loop preferably to other optimizations, notably
4836
unrolling, but there is no guarantee that the loop will be vectorized.
4839
These hints do not remove the need to pass the appropriate switches to the
4840
compiler in order to enable the relevant optimizations, that is to say
4841
@emph{-funroll-loops} for unrolling and @emph{-ftree-vectorize} for
4844
@node Pragma Loop_Variant,Pragma Machine_Attribute,Pragma Loop_Optimize,Implementation Defined Pragmas
4845
@anchor{gnat_rm/implementation_defined_pragmas pragma-loop-variant}@anchor{81}
4846
@section Pragma Loop_Variant
4852
pragma Loop_Variant ( LOOP_VARIANT_ITEM @{, LOOP_VARIANT_ITEM @} );
4853
LOOP_VARIANT_ITEM ::= CHANGE_DIRECTION => discrete_EXPRESSION
4854
CHANGE_DIRECTION ::= Increases | Decreases
4857
@cite{Loop_Variant} can only appear as one of the items in the sequence
4858
of statements of a loop body, or nested inside block statements that
4859
appear in the sequence of statements of a loop body.
4860
It allows the specification of quantities which must always
4861
decrease or increase in successive iterations of the loop. In its simplest
4862
form, just one expression is specified, whose value must increase or decrease
4863
on each iteration of the loop.
4865
In a more complex form, multiple arguments can be given which are intepreted
4866
in a nesting lexicographic manner. For example:
4869
pragma Loop_Variant (Increases => X, Decreases => Y);
4872
specifies that each time through the loop either X increases, or X stays
4873
the same and Y decreases. A @cite{Loop_Variant} pragma ensures that the
4874
loop is making progress. It can be useful in helping to show informally
4875
or prove formally that the loop always terminates.
4877
@cite{Loop_Variant} is an assertion whose effect can be controlled using
4878
an @cite{Assertion_Policy} with a check name of @cite{Loop_Variant}. The
4879
policy can be @cite{Check} to enable the loop variant check, @cite{Ignore}
4880
to ignore the check (in which case the pragma has no effect on the program),
4881
or @cite{Disable} in which case the pragma is not even checked for correct
4884
Multiple @cite{Loop_Invariant} and @cite{Loop_Variant} pragmas that
4885
apply to the same loop should be grouped in the same sequence of
4888
The @cite{Loop_Entry} attribute may be used within the expressions of the
4889
@cite{Loop_Variant} pragma to refer to values on entry to the loop.
4891
@node Pragma Machine_Attribute,Pragma Main,Pragma Loop_Variant,Implementation Defined Pragmas
4892
@anchor{gnat_rm/implementation_defined_pragmas pragma-machine-attribute}@anchor{82}
4893
@section Pragma Machine_Attribute
4899
pragma Machine_Attribute (
4900
[Entity =>] LOCAL_NAME,
4901
[Attribute_Name =>] static_string_EXPRESSION
4902
[, [Info =>] static_EXPRESSION] );
4905
Machine-dependent attributes can be specified for types and/or
4906
declarations. This pragma is semantically equivalent to
4907
@cite{__attribute__((`attribute_name}))` (if @cite{info} is not
4908
specified) or @cite{__attribute__((`attribute_name`(`info})))
4909
in GNU C, where @code{attribute_name} is recognized by the
4910
compiler middle-end or the @cite{TARGET_ATTRIBUTE_TABLE} machine
4911
specific macro. A string literal for the optional parameter @cite{info}
4912
is transformed into an identifier, which may make this pragma unusable
4913
for some attributes.
4914
For further information see @cite{GNU Compiler Collection (GCC) Internals}.
4916
@node Pragma Main,Pragma Main_Storage,Pragma Machine_Attribute,Implementation Defined Pragmas
4917
@anchor{gnat_rm/implementation_defined_pragmas pragma-main}@anchor{83}
4918
@section Pragma Main
4925
(MAIN_OPTION [, MAIN_OPTION]);
4928
[Stack_Size =>] static_integer_EXPRESSION
4929
| [Task_Stack_Size_Default =>] static_integer_EXPRESSION
4930
| [Time_Slicing_Enabled =>] static_boolean_EXPRESSION
4933
This pragma is provided for compatibility with OpenVMS VAX Systems. It has
4934
no effect in GNAT, other than being syntax checked.
4936
@node Pragma Main_Storage,Pragma No_Body,Pragma Main,Implementation Defined Pragmas
4937
@anchor{gnat_rm/implementation_defined_pragmas pragma-main-storage}@anchor{84}
4938
@section Pragma Main_Storage
4945
(MAIN_STORAGE_OPTION [, MAIN_STORAGE_OPTION]);
4947
MAIN_STORAGE_OPTION ::=
4948
[WORKING_STORAGE =>] static_SIMPLE_EXPRESSION
4949
| [TOP_GUARD =>] static_SIMPLE_EXPRESSION
4952
This pragma is provided for compatibility with OpenVMS VAX Systems. It has
4953
no effect in GNAT, other than being syntax checked.
4955
@node Pragma No_Body,Pragma No_Elaboration_Code_All,Pragma Main_Storage,Implementation Defined Pragmas
4956
@anchor{gnat_rm/implementation_defined_pragmas pragma-no-body}@anchor{85}
4957
@section Pragma No_Body
4966
There are a number of cases in which a package spec does not require a body,
4967
and in fact a body is not permitted. GNAT will not permit the spec to be
4968
compiled if there is a body around. The pragma No_Body allows you to provide
4969
a body file, even in a case where no body is allowed. The body file must
4970
contain only comments and a single No_Body pragma. This is recognized by
4971
the compiler as indicating that no body is logically present.
4973
This is particularly useful during maintenance when a package is modified in
4974
such a way that a body needed before is no longer needed. The provision of a
4975
dummy body with a No_Body pragma ensures that there is no interference from
4976
earlier versions of the package body.
4978
@node Pragma No_Elaboration_Code_All,Pragma No_Inline,Pragma No_Body,Implementation Defined Pragmas
4979
@anchor{gnat_rm/implementation_defined_pragmas pragma-no-elaboration-code-all}@anchor{86}
4980
@section Pragma No_Elaboration_Code_All
4986
pragma No_Elaboration_Code_All [(program_unit_NAME)];
4989
This is a program unit pragma (there is also an equivalent aspect of the
4990
same name) that establishes the restriction @cite{No_Elaboration_Code} for
4991
the current unit and any extended main source units (body and subunits.
4992
It also has has the effect of enforcing a transitive application of this
4993
aspect, so that if any unit is implicitly or explicitly WITH'ed by the
4994
current unit, it must also have the No_Elaboration_Code_All aspect set.
4995
It may be applied to package or subprogram specs or their generic versions.
4997
@node Pragma No_Inline,Pragma No_Return,Pragma No_Elaboration_Code_All,Implementation Defined Pragmas
4998
@anchor{gnat_rm/implementation_defined_pragmas pragma-no-inline}@anchor{87}
4999
@section Pragma No_Inline
5005
pragma No_Inline (NAME @{, NAME@});
5008
This pragma suppresses inlining for the callable entity or the instances of
5009
the generic subprogram designated by @cite{NAME}, including inlining that
5010
results from the use of pragma @cite{Inline}. This pragma is always active,
5011
in particular it is not subject to the use of option @emph{-gnatn} or
5012
@emph{-gnatN}. It is illegal to specify both pragma @cite{No_Inline} and
5013
pragma @cite{Inline_Always} for the same @cite{NAME}.
5015
@node Pragma No_Return,Pragma No_Run_Time,Pragma No_Inline,Implementation Defined Pragmas
5016
@anchor{gnat_rm/implementation_defined_pragmas pragma-no-return}@anchor{88}
5017
@section Pragma No_Return
5023
pragma No_Return (procedure_LOCAL_NAME @{, procedure_LOCAL_NAME@});
5026
Each @cite{procedure_LOCAL_NAME} argument must refer to one or more procedure
5027
declarations in the current declarative part. A procedure to which this
5028
pragma is applied may not contain any explicit @cite{return} statements.
5029
In addition, if the procedure contains any implicit returns from falling
5030
off the end of a statement sequence, then execution of that implicit
5031
return will cause Program_Error to be raised.
5033
One use of this pragma is to identify procedures whose only purpose is to raise
5034
an exception. Another use of this pragma is to suppress incorrect warnings
5035
about missing returns in functions, where the last statement of a function
5036
statement sequence is a call to such a procedure.
5038
Note that in Ada 2005 mode, this pragma is part of the language. It is
5039
available in all earlier versions of Ada as an implementation-defined
5042
@node Pragma No_Run_Time,Pragma No_Strict_Aliasing,Pragma No_Return,Implementation Defined Pragmas
5043
@anchor{gnat_rm/implementation_defined_pragmas pragma-no-run-time}@anchor{89}
5044
@section Pragma No_Run_Time
5053
This is an obsolete configuration pragma that historically was used to
5054
set up a runtime library with no object code. It is now used only for
5055
internal testing. The pragma has been superseded by the reconfigurable
5056
runtime capability of @cite{GNAT}.
5058
@node Pragma No_Strict_Aliasing,Pragma No_Tagged_Streams,Pragma No_Run_Time,Implementation Defined Pragmas
5059
@anchor{gnat_rm/implementation_defined_pragmas pragma-no-strict-aliasing}@anchor{8a}
5060
@section Pragma No_Strict_Aliasing
5066
pragma No_Strict_Aliasing [([Entity =>] type_LOCAL_NAME)];
5069
@cite{type_LOCAL_NAME} must refer to an access type
5070
declaration in the current declarative part. The effect is to inhibit
5071
strict aliasing optimization for the given type. The form with no
5072
arguments is a configuration pragma which applies to all access types
5073
declared in units to which the pragma applies. For a detailed
5074
description of the strict aliasing optimization, and the situations
5075
in which it must be suppressed, see the section on Optimization and Strict Aliasing
5076
in the @cite{GNAT User's Guide}.
5078
This pragma currently has no effects on access to unconstrained array types.
5080
@node Pragma No_Tagged_Streams,Pragma Normalize_Scalars,Pragma No_Strict_Aliasing,Implementation Defined Pragmas
5081
@anchor{gnat_rm/implementation_defined_pragmas pragma-no-tagged-streams}@anchor{8b}
5082
@section Pragma No_Tagged_Streams
5088
pragma No_Tagged_Streams;
5089
pragma No_Tagged_Streams [([Entity =>] tagged_type_LOCAL_NAME)];
5092
Normally when a tagged type is introduced using a full type declaration,
5093
part of the processing includes generating stream access routines to be
5094
used by stream attributes referencing the type (or one of its subtypes
5095
or derived types). This can involve the generation of significant amounts
5096
of code which is wasted space if stream routines are not needed for the
5099
The @cite{No_Tagged_Streams} pragma causes the generation of these stream
5100
routines to be skipped, and any attempt to use stream operations on
5101
types subject to this pragma will be statically rejected as illegal.
5103
There are two forms of the pragma. The form with no arguments must appear
5104
in a declarative sequence or in the declarations of a package spec. This
5105
pragma affects all subsequent root tagged types declared in the declaration
5106
sequence, and specifies that no stream routines be generated. The form with
5107
an argument (for which there is also a corresponding aspect) specifies a
5108
single root tagged type for which stream routines are not to be generated.
5110
Once the pragma has been given for a particular root tagged type, all subtypes
5111
and derived types of this type inherit the pragma automatically, so the effect
5112
applies to a complete hierarchy (this is necessary to deal with the class-wide
5113
dispatching versions of the stream routines).
5115
@node Pragma Normalize_Scalars,Pragma Obsolescent,Pragma No_Tagged_Streams,Implementation Defined Pragmas
5116
@anchor{gnat_rm/implementation_defined_pragmas pragma-normalize-scalars}@anchor{8c}
5117
@section Pragma Normalize_Scalars
5123
pragma Normalize_Scalars;
5126
This is a language defined pragma which is fully implemented in GNAT. The
5127
effect is to cause all scalar objects that are not otherwise initialized
5128
to be initialized. The initial values are implementation dependent and
5134
@item @emph{Standard.Character}
5136
Objects whose root type is Standard.Character are initialized to
5137
Character'Last unless the subtype range excludes NUL (in which case
5138
NUL is used). This choice will always generate an invalid value if
5141
@item @emph{Standard.Wide_Character}
5143
Objects whose root type is Standard.Wide_Character are initialized to
5144
Wide_Character'Last unless the subtype range excludes NUL (in which case
5145
NUL is used). This choice will always generate an invalid value if
5148
@item @emph{Standard.Wide_Wide_Character}
5150
Objects whose root type is Standard.Wide_Wide_Character are initialized to
5151
the invalid value 16#FFFF_FFFF# unless the subtype range excludes NUL (in
5152
which case NUL is used). This choice will always generate an invalid value if
5155
@item @emph{Integer types}
5157
Objects of an integer type are treated differently depending on whether
5158
negative values are present in the subtype. If no negative values are
5159
present, then all one bits is used as the initial value except in the
5160
special case where zero is excluded from the subtype, in which case
5161
all zero bits are used. This choice will always generate an invalid
5162
value if one exists.
5164
For subtypes with negative values present, the largest negative number
5165
is used, except in the unusual case where this largest negative number
5166
is in the subtype, and the largest positive number is not, in which case
5167
the largest positive value is used. This choice will always generate
5168
an invalid value if one exists.
5170
@item @emph{Floating-Point Types}
5172
Objects of all floating-point types are initialized to all 1-bits. For
5173
standard IEEE format, this corresponds to a NaN (not a number) which is
5174
indeed an invalid value.
5176
@item @emph{Fixed-Point Types}
5178
Objects of all fixed-point types are treated as described above for integers,
5179
with the rules applying to the underlying integer value used to represent
5180
the fixed-point value.
5182
@item @emph{Modular types}
5184
Objects of a modular type are initialized to all one bits, except in
5185
the special case where zero is excluded from the subtype, in which
5186
case all zero bits are used. This choice will always generate an
5187
invalid value if one exists.
5189
@item @emph{Enumeration types}
5191
Objects of an enumeration type are initialized to all one-bits, i.e., to
5192
the value @cite{2 ** typ'Size - 1} unless the subtype excludes the literal
5193
whose Pos value is zero, in which case a code of zero is used. This choice
5194
will always generate an invalid value if one exists.
5197
@node Pragma Obsolescent,Pragma Optimize_Alignment,Pragma Normalize_Scalars,Implementation Defined Pragmas
5198
@anchor{gnat_rm/implementation_defined_pragmas pragma-obsolescent}@anchor{8d}@anchor{gnat_rm/implementation_defined_pragmas id2}@anchor{8e}
5199
@section Pragma Obsolescent
5207
pragma Obsolescent (
5208
[Message =>] static_string_EXPRESSION
5209
[,[Version =>] Ada_05]]);
5211
pragma Obsolescent (
5213
[,[Message =>] static_string_EXPRESSION
5214
[,[Version =>] Ada_05]] );
5217
This pragma can occur immediately following a declaration of an entity,
5218
including the case of a record component. If no Entity argument is present,
5219
then this declaration is the one to which the pragma applies. If an Entity
5220
parameter is present, it must either match the name of the entity in this
5221
declaration, or alternatively, the pragma can immediately follow an enumeration
5222
type declaration, where the Entity argument names one of the enumeration
5225
This pragma is used to indicate that the named entity
5226
is considered obsolescent and should not be used. Typically this is
5227
used when an API must be modified by eventually removing or modifying
5228
existing subprograms or other entities. The pragma can be used at an
5229
intermediate stage when the entity is still present, but will be
5232
The effect of this pragma is to output a warning message on a reference to
5233
an entity thus marked that the subprogram is obsolescent if the appropriate
5234
warning option in the compiler is activated. If the Message parameter is
5235
present, then a second warning message is given containing this text. In
5236
addition, a reference to the entity is considered to be a violation of pragma
5237
Restrictions (No_Obsolescent_Features).
5239
This pragma can also be used as a program unit pragma for a package,
5240
in which case the entity name is the name of the package, and the
5241
pragma indicates that the entire package is considered
5242
obsolescent. In this case a client @cite{with}'ing such a package
5243
violates the restriction, and the @cite{with} statement is
5244
flagged with warnings if the warning option is set.
5246
If the Version parameter is present (which must be exactly
5247
the identifier Ada_05, no other argument is allowed), then the
5248
indication of obsolescence applies only when compiling in Ada 2005
5249
mode. This is primarily intended for dealing with the situations
5250
in the predefined library where subprograms or packages
5251
have become defined as obsolescent in Ada 2005
5252
(e.g., in Ada.Characters.Handling), but may be used anywhere.
5254
The following examples show typical uses of this pragma:
5258
pragma Obsolescent (p, Message => "use pp instead of p");
5263
pragma Obsolescent ("use q2new instead");
5265
type R is new integer;
5268
Message => "use RR in Ada 2005",
5278
type E is (a, bc, 'd', quack);
5279
pragma Obsolescent (Entity => bc)
5280
pragma Obsolescent (Entity => 'd')
5283
(a, b : character) return character;
5284
pragma Obsolescent (Entity => "+");
5288
Note that, as for all pragmas, if you use a pragma argument identifier,
5289
then all subsequent parameters must also use a pragma argument identifier.
5290
So if you specify "Entity =>" for the Entity argument, and a Message
5291
argument is present, it must be preceded by "Message =>".
5293
@node Pragma Optimize_Alignment,Pragma Ordered,Pragma Obsolescent,Implementation Defined Pragmas
5294
@anchor{gnat_rm/implementation_defined_pragmas pragma-optimize-alignment}@anchor{8f}
5295
@section Pragma Optimize_Alignment
5299
@geindex default settings
5304
pragma Optimize_Alignment (TIME | SPACE | OFF);
5307
This is a configuration pragma which affects the choice of default alignments
5308
for types and objects where no alignment is explicitly specified. There is a
5309
time/space trade-off in the selection of these values. Large alignments result
5310
in more efficient code, at the expense of larger data space, since sizes have
5311
to be increased to match these alignments. Smaller alignments save space, but
5312
the access code is slower. The normal choice of default alignments for types
5313
and individual alignment promotions for objects (which is what you get if you
5314
do not use this pragma, or if you use an argument of OFF), tries to balance
5315
these two requirements.
5317
Specifying SPACE causes smaller default alignments to be chosen in two cases.
5318
First any packed record is given an alignment of 1. Second, if a size is given
5319
for the type, then the alignment is chosen to avoid increasing this size. For
5331
In the default mode, this type gets an alignment of 4, so that access to the
5332
Integer field X are efficient. But this means that objects of the type end up
5333
with a size of 8 bytes. This is a valid choice, since sizes of objects are
5334
allowed to be bigger than the size of the type, but it can waste space if for
5335
example fields of type R appear in an enclosing record. If the above type is
5336
compiled in @cite{Optimize_Alignment (Space)} mode, the alignment is set to 1.
5338
However, there is one case in which SPACE is ignored. If a variable length
5339
record (that is a discriminated record with a component which is an array
5340
whose length depends on a discriminant), has a pragma Pack, then it is not
5341
in general possible to set the alignment of such a record to one, so the
5342
pragma is ignored in this case (with a warning).
5344
Specifying SPACE also disables alignment promotions for standalone objects,
5345
which occur when the compiler increases the alignment of a specific object
5346
without changing the alignment of its type.
5348
Specifying TIME causes larger default alignments to be chosen in the case of
5349
small types with sizes that are not a power of 2. For example, consider:
5362
The default alignment for this record is normally 1, but if this type is
5363
compiled in @cite{Optimize_Alignment (Time)} mode, then the alignment is set
5364
to 4, which wastes space for objects of the type, since they are now 4 bytes
5365
long, but results in more efficient access when the whole record is referenced.
5367
As noted above, this is a configuration pragma, and there is a requirement
5368
that all units in a partition be compiled with a consistent setting of the
5369
optimization setting. This would normally be achieved by use of a configuration
5370
pragma file containing the appropriate setting. The exception to this rule is
5371
that units with an explicit configuration pragma in the same file as the source
5372
unit are excluded from the consistency check, as are all predefined units. The
5373
latter are compiled by default in pragma Optimize_Alignment (Off) mode if no
5374
pragma appears at the start of the file.
5376
@node Pragma Ordered,Pragma Overflow_Mode,Pragma Optimize_Alignment,Implementation Defined Pragmas
5377
@anchor{gnat_rm/implementation_defined_pragmas pragma-ordered}@anchor{90}
5378
@section Pragma Ordered
5384
pragma Ordered (enumeration_first_subtype_LOCAL_NAME);
5387
Most enumeration types are from a conceptual point of view unordered.
5388
For example, consider:
5391
type Color is (Red, Blue, Green, Yellow);
5394
By Ada semantics @cite{Blue > Red} and @cite{Green > Blue},
5395
but really these relations make no sense; the enumeration type merely
5396
specifies a set of possible colors, and the order is unimportant.
5398
For unordered enumeration types, it is generally a good idea if
5399
clients avoid comparisons (other than equality or inequality) and
5400
explicit ranges. (A @emph{client} is a unit where the type is referenced,
5401
other than the unit where the type is declared, its body, and its subunits.)
5402
For example, if code buried in some client says:
5405
if Current_Color < Yellow then ...
5406
if Current_Color in Blue .. Green then ...
5409
then the client code is relying on the order, which is undesirable.
5410
It makes the code hard to read and creates maintenance difficulties if
5411
entries have to be added to the enumeration type. Instead,
5412
the code in the client should list the possibilities, or an
5413
appropriate subtype should be declared in the unit that declares
5414
the original enumeration type. E.g., the following subtype could
5415
be declared along with the type @cite{Color}:
5418
subtype RBG is Color range Red .. Green;
5421
and then the client could write:
5424
if Current_Color in RBG then ...
5425
if Current_Color = Blue or Current_Color = Green then ...
5428
However, some enumeration types are legitimately ordered from a conceptual
5429
point of view. For example, if you declare:
5432
type Day is (Mon, Tue, Wed, Thu, Fri, Sat, Sun);
5435
then the ordering imposed by the language is reasonable, and
5436
clients can depend on it, writing for example:
5439
if D in Mon .. Fri then ...
5443
The pragma @emph{Ordered} is provided to mark enumeration types that
5444
are conceptually ordered, alerting the reader that clients may depend
5445
on the ordering. GNAT provides a pragma to mark enumerations as ordered
5446
rather than one to mark them as unordered, since in our experience,
5447
the great majority of enumeration types are conceptually unordered.
5449
The types @cite{Boolean}, @cite{Character}, @cite{Wide_Character},
5450
and @cite{Wide_Wide_Character}
5451
are considered to be ordered types, so each is declared with a
5452
pragma @cite{Ordered} in package @cite{Standard}.
5454
Normally pragma @cite{Ordered} serves only as documentation and a guide for
5455
coding standards, but GNAT provides a warning switch @emph{-gnatw.u} that
5456
requests warnings for inappropriate uses (comparisons and explicit
5457
subranges) for unordered types. If this switch is used, then any
5458
enumeration type not marked with pragma @cite{Ordered} will be considered
5459
as unordered, and will generate warnings for inappropriate uses.
5461
Note that generic types are not considered ordered or unordered (since the
5462
template can be instantiated for both cases), so we never generate warnings
5463
for the case of generic enumerated types.
5465
For additional information please refer to the description of the
5466
@emph{-gnatw.u} switch in the GNAT User's Guide.
5468
@node Pragma Overflow_Mode,Pragma Overriding_Renamings,Pragma Ordered,Implementation Defined Pragmas
5469
@anchor{gnat_rm/implementation_defined_pragmas pragma-overflow-mode}@anchor{91}
5470
@section Pragma Overflow_Mode
5476
pragma Overflow_Mode
5478
[,[Assertions =>] MODE]);
5480
MODE ::= STRICT | MINIMIZED | ELIMINATED
5483
This pragma sets the current overflow mode to the given setting. For details
5484
of the meaning of these modes, please refer to the
5485
'Overflow Check Handling in GNAT' appendix in the
5486
GNAT User's Guide. If only the @cite{General} parameter is present,
5487
the given mode applies to all expressions. If both parameters are present,
5488
the @cite{General} mode applies to expressions outside assertions, and
5489
the @cite{Eliminated} mode applies to expressions within assertions.
5491
The case of the @cite{MODE} parameter is ignored,
5492
so @cite{MINIMIZED}, @cite{Minimized} and
5493
@cite{minimized} all have the same effect.
5495
The @cite{Overflow_Mode} pragma has the same scoping and placement
5496
rules as pragma @cite{Suppress}, so it can occur either as a
5497
configuration pragma, specifying a default for the whole
5498
program, or in a declarative scope, where it applies to the
5499
remaining declarations and statements in that scope.
5501
The pragma @cite{Suppress (Overflow_Check)} suppresses
5502
overflow checking, but does not affect the overflow mode.
5504
The pragma @cite{Unsuppress (Overflow_Check)} unsuppresses (enables)
5505
overflow checking, but does not affect the overflow mode.
5507
@node Pragma Overriding_Renamings,Pragma Partition_Elaboration_Policy,Pragma Overflow_Mode,Implementation Defined Pragmas
5508
@anchor{gnat_rm/implementation_defined_pragmas pragma-overriding-renamings}@anchor{92}
5509
@section Pragma Overriding_Renamings
5512
@geindex Rational profile
5514
@geindex Rational compatibility
5519
pragma Overriding_Renamings;
5522
This is a GNAT configuration pragma to simplify porting
5523
legacy code accepted by the Rational
5524
Ada compiler. In the presence of this pragma, a renaming declaration that
5525
renames an inherited operation declared in the same scope is legal if selected
5526
notation is used as in:
5529
pragma Overriding_Renamings;
5534
function F (..) renames R.F;
5539
RM 8.3 (15) stipulates that an overridden operation is not visible within the
5540
declaration of the overriding operation.
5542
@node Pragma Partition_Elaboration_Policy,Pragma Part_Of,Pragma Overriding_Renamings,Implementation Defined Pragmas
5543
@anchor{gnat_rm/implementation_defined_pragmas pragma-partition-elaboration-policy}@anchor{93}
5544
@section Pragma Partition_Elaboration_Policy
5550
pragma Partition_Elaboration_Policy (POLICY_IDENTIFIER);
5552
POLICY_IDENTIFIER ::= Concurrent | Sequential
5555
This pragma is standard in Ada 2005, but is available in all earlier
5556
versions of Ada as an implementation-defined pragma.
5557
See Ada 2012 Reference Manual for details.
5559
@node Pragma Part_Of,Pragma Passive,Pragma Partition_Elaboration_Policy,Implementation Defined Pragmas
5560
@anchor{gnat_rm/implementation_defined_pragmas pragma-part-of}@anchor{94}
5561
@section Pragma Part_Of
5567
pragma Part_Of (ABSTRACT_STATE);
5569
ABSTRACT_STATE ::= NAME
5572
For the semantics of this pragma, see the entry for aspect @cite{Part_Of} in the
5573
SPARK 2014 Reference Manual, section 7.2.6.
5575
@node Pragma Passive,Pragma Persistent_BSS,Pragma Part_Of,Implementation Defined Pragmas
5576
@anchor{gnat_rm/implementation_defined_pragmas pragma-passive}@anchor{95}
5577
@section Pragma Passive
5583
pragma Passive [(Semaphore | No)];
5586
Syntax checked, but otherwise ignored by GNAT. This is recognized for
5587
compatibility with DEC Ada 83 implementations, where it is used within a
5588
task definition to request that a task be made passive. If the argument
5589
@cite{Semaphore} is present, or the argument is omitted, then DEC Ada 83
5590
treats the pragma as an assertion that the containing task is passive
5591
and that optimization of context switch with this task is permitted and
5592
desired. If the argument @cite{No} is present, the task must not be
5593
optimized. GNAT does not attempt to optimize any tasks in this manner
5594
(since protected objects are available in place of passive tasks).
5596
For more information on the subject of passive tasks, see the section
5597
'Passive Task Optimization' in the GNAT Users Guide.
5599
@node Pragma Persistent_BSS,Pragma Polling,Pragma Passive,Implementation Defined Pragmas
5600
@anchor{gnat_rm/implementation_defined_pragmas pragma-persistent-bss}@anchor{96}
5601
@section Pragma Persistent_BSS
5607
pragma Persistent_BSS [(LOCAL_NAME)]
5610
This pragma allows selected objects to be placed in the @cite{.persistent_bss}
5611
section. On some targets the linker and loader provide for special
5612
treatment of this section, allowing a program to be reloaded without
5613
affecting the contents of this data (hence the name persistent).
5615
There are two forms of usage. If an argument is given, it must be the
5616
local name of a library level object, with no explicit initialization
5617
and whose type is potentially persistent. If no argument is given, then
5618
the pragma is a configuration pragma, and applies to all library level
5619
objects with no explicit initialization of potentially persistent types.
5621
A potentially persistent type is a scalar type, or an untagged,
5622
non-discriminated record, all of whose components have no explicit
5623
initialization and are themselves of a potentially persistent type,
5624
or an array, all of whose constraints are static, and whose component
5625
type is potentially persistent.
5627
If this pragma is used on a target where this feature is not supported,
5628
then the pragma will be ignored. See also @cite{pragma Linker_Section}.
5630
@node Pragma Polling,Pragma Post,Pragma Persistent_BSS,Implementation Defined Pragmas
5631
@anchor{gnat_rm/implementation_defined_pragmas pragma-polling}@anchor{97}
5632
@section Pragma Polling
5638
pragma Polling (ON | OFF);
5641
This pragma controls the generation of polling code. This is normally off.
5642
If @cite{pragma Polling (ON)} is used then periodic calls are generated to
5643
the routine @cite{Ada.Exceptions.Poll}. This routine is a separate unit in the
5644
runtime library, and can be found in file @code{a-excpol.adb}.
5646
Pragma @cite{Polling} can appear as a configuration pragma (for example it
5647
can be placed in the @code{gnat.adc} file) to enable polling globally, or it
5648
can be used in the statement or declaration sequence to control polling
5651
A call to the polling routine is generated at the start of every loop and
5652
at the start of every subprogram call. This guarantees that the @cite{Poll}
5653
routine is called frequently, and places an upper bound (determined by
5654
the complexity of the code) on the period between two @cite{Poll} calls.
5656
The primary purpose of the polling interface is to enable asynchronous
5657
aborts on targets that cannot otherwise support it (for example Windows
5658
NT), but it may be used for any other purpose requiring periodic polling.
5659
The standard version is null, and can be replaced by a user program. This
5660
will require re-compilation of the @cite{Ada.Exceptions} package that can
5661
be found in files @code{a-except.ads} and @code{a-except.adb}.
5663
A standard alternative unit (in file @code{4wexcpol.adb} in the standard GNAT
5664
distribution) is used to enable the asynchronous abort capability on
5665
targets that do not normally support the capability. The version of
5666
@cite{Poll} in this file makes a call to the appropriate runtime routine
5667
to test for an abort condition.
5669
Note that polling can also be enabled by use of the @emph{-gnatP} switch.
5670
See the section on switches for gcc in the @cite{GNAT User's Guide}.
5672
@node Pragma Post,Pragma Postcondition,Pragma Polling,Implementation Defined Pragmas
5673
@anchor{gnat_rm/implementation_defined_pragmas pragma-post}@anchor{98}
5674
@section Pragma Post
5680
@geindex postconditions
5685
pragma Post (Boolean_Expression);
5688
The @cite{Post} pragma is intended to be an exact replacement for
5689
the language-defined
5690
@cite{Post} aspect, and shares its restrictions and semantics.
5691
It must appear either immediately following the corresponding
5692
subprogram declaration (only other pragmas may intervene), or
5693
if there is no separate subprogram declaration, then it can
5694
appear at the start of the declarations in a subprogram body
5695
(preceded only by other pragmas).
5697
@node Pragma Postcondition,Pragma Post_Class,Pragma Post,Implementation Defined Pragmas
5698
@anchor{gnat_rm/implementation_defined_pragmas pragma-postcondition}@anchor{99}
5699
@section Pragma Postcondition
5702
@geindex Postcondition
5705
@geindex postconditions
5710
pragma Postcondition (
5711
[Check =>] Boolean_Expression
5712
[,[Message =>] String_Expression]);
5715
The @cite{Postcondition} pragma allows specification of automatic
5716
postcondition checks for subprograms. These checks are similar to
5717
assertions, but are automatically inserted just prior to the return
5718
statements of the subprogram with which they are associated (including
5719
implicit returns at the end of procedure bodies and associated
5720
exception handlers).
5722
In addition, the boolean expression which is the condition which
5723
must be true may contain references to function'Result in the case
5724
of a function to refer to the returned value.
5726
@cite{Postcondition} pragmas may appear either immediately following the
5727
(separate) declaration of a subprogram, or at the start of the
5728
declarations of a subprogram body. Only other pragmas may intervene
5729
(that is appear between the subprogram declaration and its
5730
postconditions, or appear before the postcondition in the
5731
declaration sequence in a subprogram body). In the case of a
5732
postcondition appearing after a subprogram declaration, the
5733
formal arguments of the subprogram are visible, and can be
5734
referenced in the postcondition expressions.
5736
The postconditions are collected and automatically tested just
5737
before any return (implicit or explicit) in the subprogram body.
5738
A postcondition is only recognized if postconditions are active
5739
at the time the pragma is encountered. The compiler switch @emph{gnata}
5740
turns on all postconditions by default, and pragma @cite{Check_Policy}
5741
with an identifier of @cite{Postcondition} can also be used to
5742
control whether postconditions are active.
5744
The general approach is that postconditions are placed in the spec
5745
if they represent functional aspects which make sense to the client.
5746
For example we might have:
5749
function Direction return Integer;
5750
pragma Postcondition
5751
(Direction'Result = +1
5753
Direction'Result = -1);
5756
which serves to document that the result must be +1 or -1, and
5757
will test that this is the case at run time if postcondition
5760
Postconditions within the subprogram body can be used to
5761
check that some internal aspect of the implementation,
5762
not visible to the client, is operating as expected.
5763
For instance if a square root routine keeps an internal
5764
counter of the number of times it is called, then we
5765
might have the following postcondition:
5768
Sqrt_Calls : Natural := 0;
5770
function Sqrt (Arg : Float) return Float is
5771
pragma Postcondition
5772
(Sqrt_Calls = Sqrt_Calls'Old + 1);
5777
As this example, shows, the use of the @cite{Old} attribute
5778
is often useful in postconditions to refer to the state on
5779
entry to the subprogram.
5781
Note that postconditions are only checked on normal returns
5782
from the subprogram. If an abnormal return results from
5783
raising an exception, then the postconditions are not checked.
5785
If a postcondition fails, then the exception
5786
@cite{System.Assertions.Assert_Failure} is raised. If
5787
a message argument was supplied, then the given string
5788
will be used as the exception message. If no message
5789
argument was supplied, then the default message has
5790
the form "Postcondition failed at file_name:line". The
5791
exception is raised in the context of the subprogram
5792
body, so it is possible to catch postcondition failures
5793
within the subprogram body itself.
5795
Within a package spec, normal visibility rules
5796
in Ada would prevent forward references within a
5797
postcondition pragma to functions defined later in
5798
the same package. This would introduce undesirable
5799
ordering constraints. To avoid this problem, all
5800
postcondition pragmas are analyzed at the end of
5801
the package spec, allowing forward references.
5803
The following example shows that this even allows
5804
mutually recursive postconditions as in:
5807
package Parity_Functions is
5808
function Odd (X : Natural) return Boolean;
5809
pragma Postcondition
5813
(x /= 0 and then Even (X - 1))));
5815
function Even (X : Natural) return Boolean;
5816
pragma Postcondition
5820
(x /= 1 and then Odd (X - 1))));
5822
end Parity_Functions;
5825
There are no restrictions on the complexity or form of
5826
conditions used within @cite{Postcondition} pragmas.
5827
The following example shows that it is even possible
5828
to verify performance behavior.
5833
Performance : constant Float;
5834
-- Performance constant set by implementation
5835
-- to match target architecture behavior.
5837
procedure Treesort (Arg : String);
5838
-- Sorts characters of argument using N*logN sort
5839
pragma Postcondition
5840
(Float (Clock - Clock'Old) <=
5841
Float (Arg'Length) *
5842
log (Float (Arg'Length)) *
5847
Note: postcondition pragmas associated with subprograms that are
5848
marked as Inline_Always, or those marked as Inline with front-end
5849
inlining (-gnatN option set) are accepted and legality-checked
5850
by the compiler, but are ignored at run-time even if postcondition
5851
checking is enabled.
5853
Note that pragma @cite{Postcondition} differs from the language-defined
5854
@cite{Post} aspect (and corresponding @cite{Post} pragma) in allowing
5855
multiple occurrences, allowing occurences in the body even if there
5856
is a separate spec, and allowing a second string parameter, and the
5857
use of the pragma identifier @cite{Check}. Historically, pragma
5858
@cite{Postcondition} was implemented prior to the development of
5859
Ada 2012, and has been retained in its original form for
5860
compatibility purposes.
5862
@node Pragma Post_Class,Pragma Pre,Pragma Postcondition,Implementation Defined Pragmas
5863
@anchor{gnat_rm/implementation_defined_pragmas pragma-post-class}@anchor{9a}
5864
@section Pragma Post_Class
5870
@geindex postconditions
5875
pragma Post_Class (Boolean_Expression);
5878
The @cite{Post_Class} pragma is intended to be an exact replacement for
5879
the language-defined
5880
@cite{Post'Class} aspect, and shares its restrictions and semantics.
5881
It must appear either immediately following the corresponding
5882
subprogram declaration (only other pragmas may intervene), or
5883
if there is no separate subprogram declaration, then it can
5884
appear at the start of the declarations in a subprogram body
5885
(preceded only by other pragmas).
5887
Note: This pragma is called @cite{Post_Class} rather than
5888
@cite{Post'Class} because the latter would not be strictly
5889
conforming to the allowed syntax for pragmas. The motivation
5890
for provinding pragmas equivalent to the aspects is to allow a program
5891
to be written using the pragmas, and then compiled if necessary
5892
using an Ada compiler that does not recognize the pragmas or
5893
aspects, but is prepared to ignore the pragmas. The assertion
5894
policy that controls this pragma is @cite{Post'Class}, not
5897
@node Pragma Pre,Pragma Precondition,Pragma Post_Class,Implementation Defined Pragmas
5898
@anchor{gnat_rm/implementation_defined_pragmas pragma-pre}@anchor{9b}
5905
@geindex preconditions
5910
pragma Pre (Boolean_Expression);
5913
The @cite{Pre} pragma is intended to be an exact replacement for
5914
the language-defined
5915
@cite{Pre} aspect, and shares its restrictions and semantics.
5916
It must appear either immediately following the corresponding
5917
subprogram declaration (only other pragmas may intervene), or
5918
if there is no separate subprogram declaration, then it can
5919
appear at the start of the declarations in a subprogram body
5920
(preceded only by other pragmas).
5922
@node Pragma Precondition,Pragma Predicate,Pragma Pre,Implementation Defined Pragmas
5923
@anchor{gnat_rm/implementation_defined_pragmas pragma-precondition}@anchor{9c}
5924
@section Pragma Precondition
5927
@geindex Preconditions
5930
@geindex preconditions
5935
pragma Precondition (
5936
[Check =>] Boolean_Expression
5937
[,[Message =>] String_Expression]);
5940
The @cite{Precondition} pragma is similar to @cite{Postcondition}
5941
except that the corresponding checks take place immediately upon
5942
entry to the subprogram, and if a precondition fails, the exception
5943
is raised in the context of the caller, and the attribute 'Result
5944
cannot be used within the precondition expression.
5946
Otherwise, the placement and visibility rules are identical to those
5947
described for postconditions. The following is an example of use
5948
within a package spec:
5951
package Math_Functions is
5953
function Sqrt (Arg : Float) return Float;
5954
pragma Precondition (Arg >= 0.0)
5959
@cite{Precondition} pragmas may appear either immediately following the
5960
(separate) declaration of a subprogram, or at the start of the
5961
declarations of a subprogram body. Only other pragmas may intervene
5962
(that is appear between the subprogram declaration and its
5963
postconditions, or appear before the postcondition in the
5964
declaration sequence in a subprogram body).
5966
Note: precondition pragmas associated with subprograms that are
5967
marked as Inline_Always, or those marked as Inline with front-end
5968
inlining (-gnatN option set) are accepted and legality-checked
5969
by the compiler, but are ignored at run-time even if precondition
5970
checking is enabled.
5972
Note that pragma @cite{Precondition} differs from the language-defined
5973
@cite{Pre} aspect (and corresponding @cite{Pre} pragma) in allowing
5974
multiple occurrences, allowing occurences in the body even if there
5975
is a separate spec, and allowing a second string parameter, and the
5976
use of the pragma identifier @cite{Check}. Historically, pragma
5977
@cite{Precondition} was implemented prior to the development of
5978
Ada 2012, and has been retained in its original form for
5979
compatibility purposes.
5981
@node Pragma Predicate,Pragma Predicate_Failure,Pragma Precondition,Implementation Defined Pragmas
5982
@anchor{gnat_rm/implementation_defined_pragmas pragma-predicate}@anchor{9d}
5983
@section Pragma Predicate
5990
([Entity =>] type_LOCAL_NAME,
5991
[Check =>] EXPRESSION);
5994
This pragma (available in all versions of Ada in GNAT) encompasses both
5995
the @cite{Static_Predicate} and @cite{Dynamic_Predicate} aspects in
5996
Ada 2012. A predicate is regarded as static if it has an allowed form
5997
for @cite{Static_Predicate} and is otherwise treated as a
5998
@cite{Dynamic_Predicate}. Otherwise, predicates specified by this
5999
pragma behave exactly as described in the Ada 2012 reference manual.
6000
For example, if we have
6003
type R is range 1 .. 10;
6005
pragma Predicate (Entity => S, Check => S not in 4 .. 6);
6007
pragma Predicate (Entity => Q, Check => F(Q) or G(Q));
6010
the effect is identical to the following Ada 2012 code:
6013
type R is range 1 .. 10;
6015
Static_Predicate => S not in 4 .. 6;
6017
Dynamic_Predicate => F(Q) or G(Q);
6020
Note that there are no pragmas @cite{Dynamic_Predicate}
6021
or @cite{Static_Predicate}. That is
6022
because these pragmas would affect legality and semantics of
6023
the program and thus do not have a neutral effect if ignored.
6024
The motivation behind providing pragmas equivalent to
6025
corresponding aspects is to allow a program to be written
6026
using the pragmas, and then compiled with a compiler that
6027
will ignore the pragmas. That doesn't work in the case of
6028
static and dynamic predicates, since if the corresponding
6029
pragmas are ignored, then the behavior of the program is
6030
fundamentally changed (for example a membership test
6031
@cite{A in B} would not take into account a predicate
6032
defined for subtype B). When following this approach, the
6033
use of predicates should be avoided.
6035
@node Pragma Predicate_Failure,Pragma Preelaborable_Initialization,Pragma Predicate,Implementation Defined Pragmas
6036
@anchor{gnat_rm/implementation_defined_pragmas pragma-predicate-failure}@anchor{9e}
6037
@section Pragma Predicate_Failure
6043
pragma Predicate_Failure
6044
([Entity =>] type_LOCAL_NAME,
6045
[Message =>] String_Expression);
6048
The @cite{Predicate_Failure} pragma is intended to be an exact replacement for
6049
the language-defined
6050
@cite{Predicate_Failure} aspect, and shares its restrictions and semantics.
6052
@node Pragma Preelaborable_Initialization,Pragma Prefix_Exception_Messages,Pragma Predicate_Failure,Implementation Defined Pragmas
6053
@anchor{gnat_rm/implementation_defined_pragmas pragma-preelaborable-initialization}@anchor{9f}
6054
@section Pragma Preelaborable_Initialization
6060
pragma Preelaborable_Initialization (DIRECT_NAME);
6063
This pragma is standard in Ada 2005, but is available in all earlier
6064
versions of Ada as an implementation-defined pragma.
6065
See Ada 2012 Reference Manual for details.
6067
@node Pragma Prefix_Exception_Messages,Pragma Pre_Class,Pragma Preelaborable_Initialization,Implementation Defined Pragmas
6068
@anchor{gnat_rm/implementation_defined_pragmas pragma-prefix-exception-messages}@anchor{a0}
6069
@section Pragma Prefix_Exception_Messages
6072
@geindex Prefix_Exception_Messages
6076
@geindex Exception_Message
6081
pragma Prefix_Exception_Messages;
6084
This is an implementation-defined configuration pragma that affects the
6085
behavior of raise statements with a message given as a static string
6086
constant (typically a string literal). In such cases, the string will
6087
be automatically prefixed by the name of the enclosing entity (giving
6088
the package and subprogram containing the raise statement). This helps
6089
to identify where messages are coming from, and this mode is automatic
6090
for the run-time library.
6092
The pragma has no effect if the message is computed with an expression other
6093
than a static string constant, since the assumption in this case is that
6094
the program computes exactly the string it wants. If you still want the
6095
prefixing in this case, you can always call
6096
@cite{GNAT.Source_Info.Enclosing_Entity} and prepend the string manually.
6098
@node Pragma Pre_Class,Pragma Priority_Specific_Dispatching,Pragma Prefix_Exception_Messages,Implementation Defined Pragmas
6099
@anchor{gnat_rm/implementation_defined_pragmas pragma-pre-class}@anchor{a1}
6100
@section Pragma Pre_Class
6106
@geindex preconditions
6111
pragma Pre_Class (Boolean_Expression);
6114
The @cite{Pre_Class} pragma is intended to be an exact replacement for
6115
the language-defined
6116
@cite{Pre'Class} aspect, and shares its restrictions and semantics.
6117
It must appear either immediately following the corresponding
6118
subprogram declaration (only other pragmas may intervene), or
6119
if there is no separate subprogram declaration, then it can
6120
appear at the start of the declarations in a subprogram body
6121
(preceded only by other pragmas).
6123
Note: This pragma is called @cite{Pre_Class} rather than
6124
@cite{Pre'Class} because the latter would not be strictly
6125
conforming to the allowed syntax for pragmas. The motivation
6126
for providing pragmas equivalent to the aspects is to allow a program
6127
to be written using the pragmas, and then compiled if necessary
6128
using an Ada compiler that does not recognize the pragmas or
6129
aspects, but is prepared to ignore the pragmas. The assertion
6130
policy that controls this pragma is @cite{Pre'Class}, not
6133
@node Pragma Priority_Specific_Dispatching,Pragma Profile,Pragma Pre_Class,Implementation Defined Pragmas
6134
@anchor{gnat_rm/implementation_defined_pragmas pragma-priority-specific-dispatching}@anchor{a2}
6135
@section Pragma Priority_Specific_Dispatching
6141
pragma Priority_Specific_Dispatching (
6143
first_priority_EXPRESSION,
6144
last_priority_EXPRESSION)
6146
POLICY_IDENTIFIER ::=
6147
EDF_Across_Priorities |
6148
FIFO_Within_Priorities |
6149
Non_Preemptive_Within_Priorities |
6150
Round_Robin_Within_Priorities
6153
This pragma is standard in Ada 2005, but is available in all earlier
6154
versions of Ada as an implementation-defined pragma.
6155
See Ada 2012 Reference Manual for details.
6157
@node Pragma Profile,Pragma Profile_Warnings,Pragma Priority_Specific_Dispatching,Implementation Defined Pragmas
6158
@anchor{gnat_rm/implementation_defined_pragmas pragma-profile}@anchor{a3}
6159
@section Pragma Profile
6165
pragma Profile (Ravenscar | Restricted | Rational | GNAT_Extended_Ravenscar);
6168
This pragma is standard in Ada 2005, but is available in all earlier
6169
versions of Ada as an implementation-defined pragma. This is a
6170
configuration pragma that establishes a set of configuration pragmas
6171
that depend on the argument. @cite{Ravenscar} is standard in Ada 2005.
6172
The other possibilities (@cite{Restricted}, @cite{Rational}, @cite{GNAT_Extended_Ravenscar})
6173
are implementation-defined. The set of configuration pragmas
6174
is defined in the following sections.
6180
Pragma Profile (Ravenscar)
6182
The @cite{Ravenscar} profile is standard in Ada 2005,
6183
but is available in all earlier
6184
versions of Ada as an implementation-defined pragma. This profile
6185
establishes the following set of configuration pragmas:
6191
@code{Task_Dispatching_Policy (FIFO_Within_Priorities)}
6193
[RM D.2.2] Tasks are dispatched following a preemptive
6194
priority-ordered scheduling policy.
6197
@code{Locking_Policy (Ceiling_Locking)}
6199
[RM D.3] While tasks and interrupts execute a protected action, they inherit
6200
the ceiling priority of the corresponding protected object.
6203
@code{Detect_Blocking}
6205
This pragma forces the detection of potentially blocking operations within a
6206
protected operation, and to raise Program_Error if that happens.
6209
plus the following set of restrictions:
6215
@code{Max_Entry_Queue_Length => 1}
6217
No task can be queued on a protected entry.
6220
@code{Max_Protected_Entries => 1}
6223
@code{Max_Task_Entries => 0}
6225
No rendezvous statements are allowed.
6228
@code{No_Abort_Statements}
6231
@code{No_Dynamic_Attachment}
6234
@code{No_Dynamic_Priorities}
6237
@code{No_Implicit_Heap_Allocations}
6240
@code{No_Local_Protected_Objects}
6243
@code{No_Local_Timing_Events}
6246
@code{No_Protected_Type_Allocators}
6249
@code{No_Relative_Delay}
6252
@code{No_Requeue_Statements}
6255
@code{No_Select_Statements}
6258
@code{No_Specific_Termination_Handlers}
6261
@code{No_Task_Allocators}
6264
@code{No_Task_Hierarchy}
6267
@code{No_Task_Termination}
6270
@code{Simple_Barriers}
6273
The Ravenscar profile also includes the following restrictions that specify
6274
that there are no semantic dependences on the corresponding predefined
6281
@code{No_Dependence => Ada.Asynchronous_Task_Control}
6284
@code{No_Dependence => Ada.Calendar}
6287
@code{No_Dependence => Ada.Execution_Time.Group_Budget}
6290
@code{No_Dependence => Ada.Execution_Time.Timers}
6293
@code{No_Dependence => Ada.Task_Attributes}
6296
@code{No_Dependence => System.Multiprocessors.Dispatching_Domains}
6299
This set of configuration pragmas and restrictions correspond to the
6300
definition of the 'Ravenscar Profile' for limited tasking, devised and
6301
published by the @cite{International Real-Time Ada Workshop@comma{} 1997}.
6302
A description is also available at
6303
@indicateurl{http://www-users.cs.york.ac.uk/~burns/ravenscar.ps}.
6305
The original definition of the profile was revised at subsequent IRTAW
6306
meetings. It has been included in the ISO
6307
@cite{Guide for the Use of the Ada Programming Language in High Integrity Systems},
6308
and was made part of the Ada 2005 standard.
6309
The formal definition given by
6310
the Ada Rapporteur Group (ARG) can be found in two Ada Issues (AI-249 and
6311
AI-305) available at
6312
@indicateurl{http://www.ada-auth.org/cgi-bin/cvsweb.cgi/ais/ai-00249.txt} and
6313
@indicateurl{http://www.ada-auth.org/cgi-bin/cvsweb.cgi/ais/ai-00305.txt}.
6315
The above set is a superset of the restrictions provided by pragma
6316
@code{Profile (Restricted)}, it includes six additional restrictions
6317
(@code{Simple_Barriers}, @code{No_Select_Statements},
6318
@code{No_Calendar}, @code{No_Implicit_Heap_Allocations},
6319
@code{No_Relative_Delay} and @code{No_Task_Termination}). This means
6320
that pragma @code{Profile (Ravenscar)}, like the pragma
6321
@code{Profile (Restricted)},
6322
automatically causes the use of a simplified,
6323
more efficient version of the tasking run-time library.
6326
Pragma Profile (GNAT_Extended_Ravenscar)
6328
This profile corresponds to a GNAT specific extension of the
6329
Ravenscar profile. The profile may change in the future although
6330
only in a compatible way: some restrictions may be removed or
6331
relaxed. It is defined as a variation of the Ravenscar profile.
6333
The @code{No_Implicit_Heap_Allocations} restriction has been replaced
6334
by @code{No_Implicit_Task_Allocations} and
6335
@code{No_Implicit_Protected_Object_Allocations}.
6337
The @code{Simple_Barriers} restriction has been replaced by
6338
@code{Pure_Barriers}.
6341
Pragma Profile (Restricted)
6343
This profile corresponds to the GNAT restricted run time. It
6344
establishes the following set of restrictions:
6350
@code{No_Abort_Statements}
6353
@code{No_Entry_Queue}
6356
@code{No_Task_Hierarchy}
6359
@code{No_Task_Allocators}
6362
@code{No_Dynamic_Priorities}
6365
@code{No_Terminate_Alternatives}
6368
@code{No_Dynamic_Attachment}
6371
@code{No_Protected_Type_Allocators}
6374
@code{No_Local_Protected_Objects}
6377
@code{No_Requeue_Statements}
6380
@code{No_Task_Attributes_Package}
6383
@code{Max_Asynchronous_Select_Nesting = 0}
6386
@code{Max_Task_Entries = 0}
6389
@code{Max_Protected_Entries = 1}
6392
@code{Max_Select_Alternatives = 0}
6395
This set of restrictions causes the automatic selection of a simplified
6396
version of the run time that provides improved performance for the
6397
limited set of tasking functionality permitted by this set of restrictions.
6400
Pragma Profile (Rational)
6402
The Rational profile is intended to facilitate porting legacy code that
6403
compiles with the Rational APEX compiler, even when the code includes non-
6404
conforming Ada constructs. The profile enables the following three pragmas:
6410
@code{pragma Implicit_Packing}
6413
@code{pragma Overriding_Renamings}
6416
@code{pragma Use_VADS_Size}
6420
@node Pragma Profile_Warnings,Pragma Propagate_Exceptions,Pragma Profile,Implementation Defined Pragmas
6421
@anchor{gnat_rm/implementation_defined_pragmas pragma-profile-warnings}@anchor{a4}
6422
@section Pragma Profile_Warnings
6428
pragma Profile_Warnings (Ravenscar | Restricted | Rational);
6431
This is an implementation-defined pragma that is similar in
6432
effect to @cite{pragma Profile} except that instead of
6433
generating @cite{Restrictions} pragmas, it generates
6434
@cite{Restriction_Warnings} pragmas. The result is that
6435
violations of the profile generate warning messages instead
6438
@node Pragma Propagate_Exceptions,Pragma Provide_Shift_Operators,Pragma Profile_Warnings,Implementation Defined Pragmas
6439
@anchor{gnat_rm/implementation_defined_pragmas pragma-propagate-exceptions}@anchor{a5}
6440
@section Pragma Propagate_Exceptions
6443
@geindex Interfacing to C++
6448
pragma Propagate_Exceptions;
6451
This pragma is now obsolete and, other than generating a warning if warnings
6452
on obsolescent features are enabled, is ignored.
6453
It is retained for compatibility
6454
purposes. It used to be used in connection with optimization of
6455
a now-obsolete mechanism for implementation of exceptions.
6457
@node Pragma Provide_Shift_Operators,Pragma Psect_Object,Pragma Propagate_Exceptions,Implementation Defined Pragmas
6458
@anchor{gnat_rm/implementation_defined_pragmas pragma-provide-shift-operators}@anchor{a6}
6459
@section Pragma Provide_Shift_Operators
6462
@geindex Shift operators
6467
pragma Provide_Shift_Operators (integer_first_subtype_LOCAL_NAME);
6470
This pragma can be applied to a first subtype local name that specifies
6471
either an unsigned or signed type. It has the effect of providing the
6472
five shift operators (Shift_Left, Shift_Right, Shift_Right_Arithmetic,
6473
Rotate_Left and Rotate_Right) for the given type. It is similar to
6474
including the function declarations for these five operators, together
6475
with the pragma Import (Intrinsic, ...) statements.
6477
@node Pragma Psect_Object,Pragma Pure_Function,Pragma Provide_Shift_Operators,Implementation Defined Pragmas
6478
@anchor{gnat_rm/implementation_defined_pragmas pragma-psect-object}@anchor{a7}
6479
@section Pragma Psect_Object
6485
pragma Psect_Object (
6486
[Internal =>] LOCAL_NAME,
6487
[, [External =>] EXTERNAL_SYMBOL]
6488
[, [Size =>] EXTERNAL_SYMBOL]);
6492
| static_string_EXPRESSION
6495
This pragma is identical in effect to pragma @cite{Common_Object}.
6497
@node Pragma Pure_Function,Pragma Rational,Pragma Psect_Object,Implementation Defined Pragmas
6498
@anchor{gnat_rm/implementation_defined_pragmas pragma-pure-function}@anchor{a8}
6499
@section Pragma Pure_Function
6505
pragma Pure_Function ([Entity =>] function_LOCAL_NAME);
6508
This pragma appears in the same declarative part as a function
6509
declaration (or a set of function declarations if more than one
6510
overloaded declaration exists, in which case the pragma applies
6511
to all entities). It specifies that the function @cite{Entity} is
6512
to be considered pure for the purposes of code generation. This means
6513
that the compiler can assume that there are no side effects, and
6514
in particular that two calls with identical arguments produce the
6515
same result. It also means that the function can be used in an
6518
Note that, quite deliberately, there are no static checks to try
6519
to ensure that this promise is met, so @cite{Pure_Function} can be used
6520
with functions that are conceptually pure, even if they do modify
6521
global variables. For example, a square root function that is
6522
instrumented to count the number of times it is called is still
6523
conceptually pure, and can still be optimized, even though it
6524
modifies a global variable (the count). Memo functions are another
6525
example (where a table of previous calls is kept and consulted to
6526
avoid re-computation).
6528
Note also that the normal rules excluding optimization of subprograms
6529
in pure units (when parameter types are descended from System.Address,
6530
or when the full view of a parameter type is limited), do not apply
6531
for the Pure_Function case. If you explicitly specify Pure_Function,
6532
the compiler may optimize away calls with identical arguments, and
6533
if that results in unexpected behavior, the proper action is not to
6534
use the pragma for subprograms that are not (conceptually) pure.
6536
Note: Most functions in a @cite{Pure} package are automatically pure, and
6537
there is no need to use pragma @cite{Pure_Function} for such functions. One
6538
exception is any function that has at least one formal of type
6539
@cite{System.Address} or a type derived from it. Such functions are not
6540
considered pure by default, since the compiler assumes that the
6541
@cite{Address} parameter may be functioning as a pointer and that the
6542
referenced data may change even if the address value does not.
6543
Similarly, imported functions are not considered to be pure by default,
6544
since there is no way of checking that they are in fact pure. The use
6545
of pragma @cite{Pure_Function} for such a function will override these default
6546
assumption, and cause the compiler to treat a designated subprogram as pure
6549
Note: If pragma @cite{Pure_Function} is applied to a renamed function, it
6550
applies to the underlying renamed function. This can be used to
6551
disambiguate cases of overloading where some but not all functions
6552
in a set of overloaded functions are to be designated as pure.
6554
If pragma @cite{Pure_Function} is applied to a library level function, the
6555
function is also considered pure from an optimization point of view, but the
6556
unit is not a Pure unit in the categorization sense. So for example, a function
6557
thus marked is free to @cite{with} non-pure units.
6559
@node Pragma Rational,Pragma Ravenscar,Pragma Pure_Function,Implementation Defined Pragmas
6560
@anchor{gnat_rm/implementation_defined_pragmas pragma-rational}@anchor{a9}
6561
@section Pragma Rational
6570
This pragma is considered obsolescent, but is retained for
6571
compatibility purposes. It is equivalent to:
6574
pragma Profile (Rational);
6577
@node Pragma Ravenscar,Pragma Refined_Depends,Pragma Rational,Implementation Defined Pragmas
6578
@anchor{gnat_rm/implementation_defined_pragmas pragma-ravenscar}@anchor{aa}
6579
@section Pragma Ravenscar
6588
This pragma is considered obsolescent, but is retained for
6589
compatibility purposes. It is equivalent to:
6592
pragma Profile (Ravenscar);
6595
which is the preferred method of setting the @cite{Ravenscar} profile.
6597
@node Pragma Refined_Depends,Pragma Refined_Global,Pragma Ravenscar,Implementation Defined Pragmas
6598
@anchor{gnat_rm/implementation_defined_pragmas pragma-refined-depends}@anchor{ab}
6599
@section Pragma Refined_Depends
6605
pragma Refined_Depends (DEPENDENCY_RELATION);
6607
DEPENDENCY_RELATION ::=
6609
| (DEPENDENCY_CLAUSE @{, DEPENDENCY_CLAUSE@})
6611
DEPENDENCY_CLAUSE ::=
6612
OUTPUT_LIST =>[+] INPUT_LIST
6613
| NULL_DEPENDENCY_CLAUSE
6615
NULL_DEPENDENCY_CLAUSE ::= null => INPUT_LIST
6617
OUTPUT_LIST ::= OUTPUT | (OUTPUT @{, OUTPUT@})
6619
INPUT_LIST ::= null | INPUT | (INPUT @{, INPUT@})
6621
OUTPUT ::= NAME | FUNCTION_RESULT
6624
where FUNCTION_RESULT is a function Result attribute_reference
6627
For the semantics of this pragma, see the entry for aspect @cite{Refined_Depends} in
6628
the SPARK 2014 Reference Manual, section 6.1.5.
6630
@node Pragma Refined_Global,Pragma Refined_Post,Pragma Refined_Depends,Implementation Defined Pragmas
6631
@anchor{gnat_rm/implementation_defined_pragmas pragma-refined-global}@anchor{ac}
6632
@section Pragma Refined_Global
6638
pragma Refined_Global (GLOBAL_SPECIFICATION);
6640
GLOBAL_SPECIFICATION ::=
6643
| (MODED_GLOBAL_LIST @{, MODED_GLOBAL_LIST@})
6645
MODED_GLOBAL_LIST ::= MODE_SELECTOR => GLOBAL_LIST
6647
MODE_SELECTOR ::= In_Out | Input | Output | Proof_In
6648
GLOBAL_LIST ::= GLOBAL_ITEM | (GLOBAL_ITEM @{, GLOBAL_ITEM@})
6649
GLOBAL_ITEM ::= NAME
6652
For the semantics of this pragma, see the entry for aspect @cite{Refined_Global} in
6653
the SPARK 2014 Reference Manual, section 6.1.4.
6655
@node Pragma Refined_Post,Pragma Refined_State,Pragma Refined_Global,Implementation Defined Pragmas
6656
@anchor{gnat_rm/implementation_defined_pragmas pragma-refined-post}@anchor{ad}
6657
@section Pragma Refined_Post
6663
pragma Refined_Post (boolean_EXPRESSION);
6666
For the semantics of this pragma, see the entry for aspect @cite{Refined_Post} in
6667
the SPARK 2014 Reference Manual, section 7.2.7.
6669
@node Pragma Refined_State,Pragma Relative_Deadline,Pragma Refined_Post,Implementation Defined Pragmas
6670
@anchor{gnat_rm/implementation_defined_pragmas pragma-refined-state}@anchor{ae}
6671
@section Pragma Refined_State
6677
pragma Refined_State (REFINEMENT_LIST);
6680
(REFINEMENT_CLAUSE @{, REFINEMENT_CLAUSE@})
6682
REFINEMENT_CLAUSE ::= state_NAME => CONSTITUENT_LIST
6684
CONSTITUENT_LIST ::=
6687
| (CONSTITUENT @{, CONSTITUENT@})
6689
CONSTITUENT ::= object_NAME | state_NAME
6692
For the semantics of this pragma, see the entry for aspect @cite{Refined_State} in
6693
the SPARK 2014 Reference Manual, section 7.2.2.
6695
@node Pragma Relative_Deadline,Pragma Remote_Access_Type,Pragma Refined_State,Implementation Defined Pragmas
6696
@anchor{gnat_rm/implementation_defined_pragmas pragma-relative-deadline}@anchor{af}
6697
@section Pragma Relative_Deadline
6703
pragma Relative_Deadline (time_span_EXPRESSION);
6706
This pragma is standard in Ada 2005, but is available in all earlier
6707
versions of Ada as an implementation-defined pragma.
6708
See Ada 2012 Reference Manual for details.
6710
@node Pragma Remote_Access_Type,Pragma Restricted_Run_Time,Pragma Relative_Deadline,Implementation Defined Pragmas
6711
@anchor{gnat_rm/implementation_defined_pragmas pragma-remote-access-type}@anchor{b0}
6712
@section Pragma Remote_Access_Type
6718
pragma Remote_Access_Type ([Entity =>] formal_access_type_LOCAL_NAME);
6721
This pragma appears in the formal part of a generic declaration.
6722
It specifies an exception to the RM rule from E.2.2(17/2), which forbids
6723
the use of a remote access to class-wide type as actual for a formal
6726
When this pragma applies to a formal access type @cite{Entity}, that
6727
type is treated as a remote access to class-wide type in the generic.
6728
It must be a formal general access type, and its designated type must
6729
be the class-wide type of a formal tagged limited private type from the
6730
same generic declaration.
6732
In the generic unit, the formal type is subject to all restrictions
6733
pertaining to remote access to class-wide types. At instantiation, the
6734
actual type must be a remote access to class-wide type.
6736
@node Pragma Restricted_Run_Time,Pragma Restriction_Warnings,Pragma Remote_Access_Type,Implementation Defined Pragmas
6737
@anchor{gnat_rm/implementation_defined_pragmas pragma-restricted-run-time}@anchor{b1}
6738
@section Pragma Restricted_Run_Time
6744
pragma Restricted_Run_Time;
6747
This pragma is considered obsolescent, but is retained for
6748
compatibility purposes. It is equivalent to:
6751
pragma Profile (Restricted);
6754
which is the preferred method of setting the restricted run time
6757
@node Pragma Restriction_Warnings,Pragma Reviewable,Pragma Restricted_Run_Time,Implementation Defined Pragmas
6758
@anchor{gnat_rm/implementation_defined_pragmas pragma-restriction-warnings}@anchor{b2}
6759
@section Pragma Restriction_Warnings
6765
pragma Restriction_Warnings
6766
(restriction_IDENTIFIER @{, restriction_IDENTIFIER@});
6769
This pragma allows a series of restriction identifiers to be
6770
specified (the list of allowed identifiers is the same as for
6771
pragma @cite{Restrictions}). For each of these identifiers
6772
the compiler checks for violations of the restriction, but
6773
generates a warning message rather than an error message
6774
if the restriction is violated.
6776
One use of this is in situations where you want to know
6777
about violations of a restriction, but you want to ignore some of
6778
these violations. Consider this example, where you want to set
6779
Ada_95 mode and enable style checks, but you want to know about
6780
any other use of implementation pragmas:
6783
pragma Restriction_Warnings (No_Implementation_Pragmas);
6784
pragma Warnings (Off, "violation of No_Implementation_Pragmas");
6786
pragma Style_Checks ("2bfhkM160");
6787
pragma Warnings (On, "violation of No_Implementation_Pragmas");
6790
By including the above lines in a configuration pragmas file,
6791
the Ada_95 and Style_Checks pragmas are accepted without
6792
generating a warning, but any other use of implementation
6793
defined pragmas will cause a warning to be generated.
6795
@node Pragma Reviewable,Pragma Share_Generic,Pragma Restriction_Warnings,Implementation Defined Pragmas
6796
@anchor{gnat_rm/implementation_defined_pragmas pragma-reviewable}@anchor{b3}
6797
@section Pragma Reviewable
6806
This pragma is an RM-defined standard pragma, but has no effect on the
6807
program being compiled, or on the code generated for the program.
6809
To obtain the required output specified in RM H.3.1, the compiler must be
6810
run with various special switches as follows:
6816
@emph{Where compiler-generated run-time checks remain}
6818
The switch @emph{-gnatGL}
6819
may be used to list the expanded code in pseudo-Ada form.
6820
Runtime checks show up in the listing either as explicit
6821
checks or operators marked with @{@} to indicate a check is present.
6824
@emph{An identification of known exceptions at compile time}
6826
If the program is compiled with @emph{-gnatwa},
6827
the compiler warning messages will indicate all cases where the compiler
6828
detects that an exception is certain to occur at run time.
6831
@emph{Possible reads of uninitialized variables}
6833
The compiler warns of many such cases, but its output is incomplete.
6837
A supplemental static analysis tool
6838
may be used to obtain a comprehensive list of all
6839
possible points at which uninitialized data may be read.
6845
@emph{Where run-time support routines are implicitly invoked}
6847
In the output from @emph{-gnatGL},
6848
run-time calls are explicitly listed as calls to the relevant
6852
@emph{Object code listing}
6854
This may be obtained either by using the @emph{-S} switch,
6855
or the objdump utility.
6858
@emph{Constructs known to be erroneous at compile time}
6860
These are identified by warnings issued by the compiler (use @emph{-gnatwa}).
6863
@emph{Stack usage information}
6865
Static stack usage data (maximum per-subprogram) can be obtained via the
6866
@emph{-fstack-usage} switch to the compiler.
6867
Dynamic stack usage data (per task) can be obtained via the @emph{-u} switch
6876
@emph{Object code listing of entire partition}
6878
This can be obtained by compiling the partition with @emph{-S},
6879
or by applying objdump
6880
to all the object files that are part of the partition.
6883
@emph{A description of the run-time model}
6885
The full sources of the run-time are available, and the documentation of
6886
these routines describes how these run-time routines interface to the
6887
underlying operating system facilities.
6890
@emph{Control and data-flow information}
6894
A supplemental static analysis tool
6895
may be used to obtain complete control and data-flow information, as well as
6896
comprehensive messages identifying possible problems based on this
6899
@node Pragma Share_Generic,Pragma Shared,Pragma Reviewable,Implementation Defined Pragmas
6900
@anchor{gnat_rm/implementation_defined_pragmas pragma-share-generic}@anchor{b4}
6901
@section Pragma Share_Generic
6907
pragma Share_Generic (GNAME @{, GNAME@});
6909
GNAME ::= generic_unit_NAME | generic_instance_NAME
6912
This pragma is provided for compatibility with Dec Ada 83. It has
6913
no effect in @cite{GNAT} (which does not implement shared generics), other
6914
than to check that the given names are all names of generic units or
6917
@node Pragma Shared,Pragma Short_Circuit_And_Or,Pragma Share_Generic,Implementation Defined Pragmas
6918
@anchor{gnat_rm/implementation_defined_pragmas pragma-shared}@anchor{b5}
6919
@section Pragma Shared
6922
This pragma is provided for compatibility with Ada 83. The syntax and
6923
semantics are identical to pragma Atomic.
6925
@node Pragma Short_Circuit_And_Or,Pragma Short_Descriptors,Pragma Shared,Implementation Defined Pragmas
6926
@anchor{gnat_rm/implementation_defined_pragmas pragma-short-circuit-and-or}@anchor{b6}
6927
@section Pragma Short_Circuit_And_Or
6933
pragma Short_Circuit_And_Or;
6936
This configuration pragma causes any occurrence of the AND operator applied to
6937
operands of type Standard.Boolean to be short-circuited (i.e. the AND operator
6938
is treated as if it were AND THEN). Or is similarly treated as OR ELSE. This
6939
may be useful in the context of certification protocols requiring the use of
6940
short-circuited logical operators. If this configuration pragma occurs locally
6941
within the file being compiled, it applies only to the file being compiled.
6942
There is no requirement that all units in a partition use this option.
6944
@node Pragma Short_Descriptors,Pragma Simple_Storage_Pool_Type,Pragma Short_Circuit_And_Or,Implementation Defined Pragmas
6945
@anchor{gnat_rm/implementation_defined_pragmas pragma-short-descriptors}@anchor{b7}
6946
@section Pragma Short_Descriptors
6952
pragma Short_Descriptors
6955
This pragma is provided for compatibility with other Ada implementations. It
6956
is recognized but ignored by all current versions of GNAT.
6958
@node Pragma Simple_Storage_Pool_Type,Pragma Source_File_Name,Pragma Short_Descriptors,Implementation Defined Pragmas
6959
@anchor{gnat_rm/implementation_defined_pragmas pragma-simple-storage-pool-type}@anchor{b8}
6960
@section Pragma Simple_Storage_Pool_Type
6963
@geindex Storage pool
6966
@geindex Simple storage pool
6971
pragma Simple_Storage_Pool_Type (type_LOCAL_NAME);
6974
A type can be established as a 'simple storage pool type' by applying
6975
the representation pragma @cite{Simple_Storage_Pool_Type} to the type.
6976
A type named in the pragma must be a library-level immutably limited record
6977
type or limited tagged type declared immediately within a package declaration.
6978
The type can also be a limited private type whose full type is allowed as
6979
a simple storage pool type.
6981
For a simple storage pool type @cite{SSP}, nonabstract primitive subprograms
6982
@cite{Allocate}, @cite{Deallocate}, and @cite{Storage_Size} can be declared that
6983
are subtype conformant with the following subprogram declarations:
6988
Storage_Address : out System.Address;
6989
Size_In_Storage_Elements : System.Storage_Elements.Storage_Count;
6990
Alignment : System.Storage_Elements.Storage_Count);
6992
procedure Deallocate
6994
Storage_Address : System.Address;
6995
Size_In_Storage_Elements : System.Storage_Elements.Storage_Count;
6996
Alignment : System.Storage_Elements.Storage_Count);
6998
function Storage_Size (Pool : SSP)
6999
return System.Storage_Elements.Storage_Count;
7002
Procedure @cite{Allocate} must be declared, whereas @cite{Deallocate} and
7003
@cite{Storage_Size} are optional. If @cite{Deallocate} is not declared, then
7004
applying an unchecked deallocation has no effect other than to set its actual
7005
parameter to null. If @cite{Storage_Size} is not declared, then the
7006
@cite{Storage_Size} attribute applied to an access type associated with
7007
a pool object of type SSP returns zero. Additional operations can be declared
7008
for a simple storage pool type (such as for supporting a mark/release
7009
storage-management discipline).
7011
An object of a simple storage pool type can be associated with an access
7012
type by specifying the attribute
7013
@ref{b9,,Simple_Storage_Pool}. For example:
7016
My_Pool : My_Simple_Storage_Pool_Type;
7018
type Acc is access My_Data_Type;
7020
for Acc'Simple_Storage_Pool use My_Pool;
7023
See attribute @ref{b9,,Simple_Storage_Pool}
7024
for further details.
7026
@node Pragma Source_File_Name,Pragma Source_File_Name_Project,Pragma Simple_Storage_Pool_Type,Implementation Defined Pragmas
7027
@anchor{gnat_rm/implementation_defined_pragmas pragma-source-file-name}@anchor{ba}@anchor{gnat_rm/implementation_defined_pragmas id3}@anchor{bb}
7028
@section Pragma Source_File_Name
7034
pragma Source_File_Name (
7035
[Unit_Name =>] unit_NAME,
7036
Spec_File_Name => STRING_LITERAL,
7037
[Index => INTEGER_LITERAL]);
7039
pragma Source_File_Name (
7040
[Unit_Name =>] unit_NAME,
7041
Body_File_Name => STRING_LITERAL,
7042
[Index => INTEGER_LITERAL]);
7045
Use this to override the normal naming convention. It is a configuration
7046
pragma, and so has the usual applicability of configuration pragmas
7047
(i.e., it applies to either an entire partition, or to all units in a
7048
compilation, or to a single unit, depending on how it is used.
7049
@cite{unit_name} is mapped to @cite{file_name_literal}. The identifier for
7050
the second argument is required, and indicates whether this is the file
7051
name for the spec or for the body.
7053
The optional Index argument should be used when a file contains multiple
7054
units, and when you do not want to use @cite{gnatchop} to separate then
7055
into multiple files (which is the recommended procedure to limit the
7056
number of recompilations that are needed when some sources change).
7057
For instance, if the source file @code{source.ada} contains
7071
you could use the following configuration pragmas:
7074
pragma Source_File_Name
7075
(B, Spec_File_Name => "source.ada", Index => 1);
7076
pragma Source_File_Name
7077
(A, Body_File_Name => "source.ada", Index => 2);
7080
Note that the @cite{gnatname} utility can also be used to generate those
7081
configuration pragmas.
7083
Another form of the @cite{Source_File_Name} pragma allows
7084
the specification of patterns defining alternative file naming schemes
7085
to apply to all files.
7088
pragma Source_File_Name
7089
( [Spec_File_Name =>] STRING_LITERAL
7090
[,[Casing =>] CASING_SPEC]
7091
[,[Dot_Replacement =>] STRING_LITERAL]);
7093
pragma Source_File_Name
7094
( [Body_File_Name =>] STRING_LITERAL
7095
[,[Casing =>] CASING_SPEC]
7096
[,[Dot_Replacement =>] STRING_LITERAL]);
7098
pragma Source_File_Name
7099
( [Subunit_File_Name =>] STRING_LITERAL
7100
[,[Casing =>] CASING_SPEC]
7101
[,[Dot_Replacement =>] STRING_LITERAL]);
7103
CASING_SPEC ::= Lowercase | Uppercase | Mixedcase
7106
The first argument is a pattern that contains a single asterisk indicating
7107
the point at which the unit name is to be inserted in the pattern string
7108
to form the file name. The second argument is optional. If present it
7109
specifies the casing of the unit name in the resulting file name string.
7110
The default is lower case. Finally the third argument allows for systematic
7111
replacement of any dots in the unit name by the specified string literal.
7113
Note that Source_File_Name pragmas should not be used if you are using
7114
project files. The reason for this rule is that the project manager is not
7115
aware of these pragmas, and so other tools that use the projet file would not
7116
be aware of the intended naming conventions. If you are using project files,
7117
file naming is controlled by Source_File_Name_Project pragmas, which are
7118
usually supplied automatically by the project manager. A pragma
7119
Source_File_Name cannot appear after a @ref{bc,,Pragma Source_File_Name_Project}.
7121
For more details on the use of the @cite{Source_File_Name} pragma, see the
7122
sections on @cite{Using Other File Names} and @cite{Alternative File Naming Schemes' in the :title:`GNAT User's Guide}.
7124
@node Pragma Source_File_Name_Project,Pragma Source_Reference,Pragma Source_File_Name,Implementation Defined Pragmas
7125
@anchor{gnat_rm/implementation_defined_pragmas id4}@anchor{bd}@anchor{gnat_rm/implementation_defined_pragmas pragma-source-file-name-project}@anchor{bc}
7126
@section Pragma Source_File_Name_Project
7129
This pragma has the same syntax and semantics as pragma Source_File_Name.
7130
It is only allowed as a stand-alone configuration pragma.
7131
It cannot appear after a @ref{ba,,Pragma Source_File_Name}, and
7132
most importantly, once pragma Source_File_Name_Project appears,
7133
no further Source_File_Name pragmas are allowed.
7135
The intention is that Source_File_Name_Project pragmas are always
7136
generated by the Project Manager in a manner consistent with the naming
7137
specified in a project file, and when naming is controlled in this manner,
7138
it is not permissible to attempt to modify this naming scheme using
7139
Source_File_Name or Source_File_Name_Project pragmas (which would not be
7140
known to the project manager).
7142
@node Pragma Source_Reference,Pragma SPARK_Mode,Pragma Source_File_Name_Project,Implementation Defined Pragmas
7143
@anchor{gnat_rm/implementation_defined_pragmas pragma-source-reference}@anchor{be}
7144
@section Pragma Source_Reference
7150
pragma Source_Reference (INTEGER_LITERAL, STRING_LITERAL);
7153
This pragma must appear as the first line of a source file.
7154
@cite{integer_literal} is the logical line number of the line following
7155
the pragma line (for use in error messages and debugging
7156
information). @cite{string_literal} is a static string constant that
7157
specifies the file name to be used in error messages and debugging
7158
information. This is most notably used for the output of @cite{gnatchop}
7159
with the @emph{-r} switch, to make sure that the original unchopped
7160
source file is the one referred to.
7162
The second argument must be a string literal, it cannot be a static
7163
string expression other than a string literal. This is because its value
7164
is needed for error messages issued by all phases of the compiler.
7166
@node Pragma SPARK_Mode,Pragma Static_Elaboration_Desired,Pragma Source_Reference,Implementation Defined Pragmas
7167
@anchor{gnat_rm/implementation_defined_pragmas pragma-spark-mode}@anchor{bf}
7168
@section Pragma SPARK_Mode
7174
pragma SPARK_Mode [(On | Off)] ;
7177
In general a program can have some parts that are in SPARK 2014 (and
7178
follow all the rules in the SPARK Reference Manual), and some parts
7179
that are full Ada 2012.
7181
The SPARK_Mode pragma is used to identify which parts are in SPARK
7182
2014 (by default programs are in full Ada). The SPARK_Mode pragma can
7183
be used in the following places:
7189
As a configuration pragma, in which case it sets the default mode for
7190
all units compiled with this pragma.
7193
Immediately following a library-level subprogram spec
7196
Immediately within a library-level package body
7199
Immediately following the @cite{private} keyword of a library-level
7203
Immediately following the @cite{begin} keyword of a library-level
7207
Immediately within a library-level subprogram body
7210
Normally a subprogram or package spec/body inherits the current mode
7211
that is active at the point it is declared. But this can be overridden
7212
by pragma within the spec or body as above.
7214
The basic consistency rule is that you can't turn SPARK_Mode back
7215
@cite{On}, once you have explicitly (with a pragma) turned if
7216
@cite{Off}. So the following rules apply:
7218
If a subprogram spec has SPARK_Mode @cite{Off}, then the body must
7219
also have SPARK_Mode @cite{Off}.
7221
For a package, we have four parts:
7227
the package public declarations
7230
the package private part
7233
the body of the package
7236
the elaboration code after @cite{begin}
7239
For a package, the rule is that if you explicitly turn SPARK_Mode
7240
@cite{Off} for any part, then all the following parts must have
7241
SPARK_Mode @cite{Off}. Note that this may require repeating a pragma
7242
SPARK_Mode (@cite{Off}) in the body. For example, if we have a
7243
configuration pragma SPARK_Mode (@cite{On}) that turns the mode on by
7244
default everywhere, and one particular package spec has pragma
7245
SPARK_Mode (@cite{Off}), then that pragma will need to be repeated in
7248
@node Pragma Static_Elaboration_Desired,Pragma Stream_Convert,Pragma SPARK_Mode,Implementation Defined Pragmas
7249
@anchor{gnat_rm/implementation_defined_pragmas pragma-static-elaboration-desired}@anchor{c0}
7250
@section Pragma Static_Elaboration_Desired
7256
pragma Static_Elaboration_Desired;
7259
This pragma is used to indicate that the compiler should attempt to initialize
7260
statically the objects declared in the library unit to which the pragma applies,
7261
when these objects are initialized (explicitly or implicitly) by an aggregate.
7262
In the absence of this pragma, aggregates in object declarations are expanded
7263
into assignments and loops, even when the aggregate components are static
7264
constants. When the aggregate is present the compiler builds a static expression
7265
that requires no run-time code, so that the initialized object can be placed in
7266
read-only data space. If the components are not static, or the aggregate has
7267
more that 100 components, the compiler emits a warning that the pragma cannot
7268
be obeyed. (See also the restriction No_Implicit_Loops, which supports static
7269
construction of larger aggregates with static components that include an others
7272
@node Pragma Stream_Convert,Pragma Style_Checks,Pragma Static_Elaboration_Desired,Implementation Defined Pragmas
7273
@anchor{gnat_rm/implementation_defined_pragmas pragma-stream-convert}@anchor{c1}
7274
@section Pragma Stream_Convert
7280
pragma Stream_Convert (
7281
[Entity =>] type_LOCAL_NAME,
7282
[Read =>] function_NAME,
7283
[Write =>] function_NAME);
7286
This pragma provides an efficient way of providing user-defined stream
7287
attributes. Not only is it simpler to use than specifying the attributes
7288
directly, but more importantly, it allows the specification to be made in such
7289
a way that the predefined unit Ada.Streams is not loaded unless it is actually
7290
needed (i.e. unless the stream attributes are actually used); the use of
7291
the Stream_Convert pragma adds no overhead at all, unless the stream
7292
attributes are actually used on the designated type.
7294
The first argument specifies the type for which stream functions are
7295
provided. The second parameter provides a function used to read values
7296
of this type. It must name a function whose argument type may be any
7297
subtype, and whose returned type must be the type given as the first
7298
argument to the pragma.
7300
The meaning of the @cite{Read} parameter is that if a stream attribute directly
7301
or indirectly specifies reading of the type given as the first parameter,
7302
then a value of the type given as the argument to the Read function is
7303
read from the stream, and then the Read function is used to convert this
7304
to the required target type.
7306
Similarly the @cite{Write} parameter specifies how to treat write attributes
7307
that directly or indirectly apply to the type given as the first parameter.
7308
It must have an input parameter of the type specified by the first parameter,
7309
and the return type must be the same as the input type of the Read function.
7310
The effect is to first call the Write function to convert to the given stream
7311
type, and then write the result type to the stream.
7313
The Read and Write functions must not be overloaded subprograms. If necessary
7314
renamings can be supplied to meet this requirement.
7315
The usage of this attribute is best illustrated by a simple example, taken
7316
from the GNAT implementation of package Ada.Strings.Unbounded:
7319
function To_Unbounded (S : String) return Unbounded_String
7320
renames To_Unbounded_String;
7322
pragma Stream_Convert
7323
(Unbounded_String, To_Unbounded, To_String);
7326
The specifications of the referenced functions, as given in the Ada
7327
Reference Manual are:
7330
function To_Unbounded_String (Source : String)
7331
return Unbounded_String;
7333
function To_String (Source : Unbounded_String)
7337
The effect is that if the value of an unbounded string is written to a stream,
7338
then the representation of the item in the stream is in the same format that
7339
would be used for @cite{Standard.String'Output}, and this same representation
7340
is expected when a value of this type is read from the stream. Note that the
7341
value written always includes the bounds, even for Unbounded_String'Write,
7342
since Unbounded_String is not an array type.
7344
Note that the @cite{Stream_Convert} pragma is not effective in the case of
7345
a derived type of a non-limited tagged type. If such a type is specified then
7346
the pragma is silently ignored, and the default implementation of the stream
7347
attributes is used instead.
7349
@node Pragma Style_Checks,Pragma Subtitle,Pragma Stream_Convert,Implementation Defined Pragmas
7350
@anchor{gnat_rm/implementation_defined_pragmas pragma-style-checks}@anchor{c2}
7351
@section Pragma Style_Checks
7357
pragma Style_Checks (string_LITERAL | ALL_CHECKS |
7358
On | Off [, LOCAL_NAME]);
7361
This pragma is used in conjunction with compiler switches to control the
7362
built in style checking provided by GNAT. The compiler switches, if set,
7363
provide an initial setting for the switches, and this pragma may be used
7364
to modify these settings, or the settings may be provided entirely by
7365
the use of the pragma. This pragma can be used anywhere that a pragma
7366
is legal, including use as a configuration pragma (including use in
7367
the @code{gnat.adc} file).
7369
The form with a string literal specifies which style options are to be
7370
activated. These are additive, so they apply in addition to any previously
7371
set style check options. The codes for the options are the same as those
7372
used in the @emph{-gnaty} switch to @emph{gcc} or @emph{gnatmake}.
7373
For example the following two methods can be used to enable
7381
pragma Style_Checks ("l");
7390
The form ALL_CHECKS activates all standard checks (its use is equivalent
7391
to the use of the @cite{gnaty} switch with no options.
7392
See the @cite{GNAT User's Guide} for details.)
7394
Note: the behavior is slightly different in GNAT mode (@emph{-gnatg} used).
7395
In this case, ALL_CHECKS implies the standard set of GNAT mode style check
7396
options (i.e. equivalent to @emph{-gnatyg}).
7398
The forms with @cite{Off} and @cite{On}
7399
can be used to temporarily disable style checks
7400
as shown in the following example:
7403
pragma Style_Checks ("k"); -- requires keywords in lower case
7404
pragma Style_Checks (Off); -- turn off style checks
7405
NULL; -- this will not generate an error message
7406
pragma Style_Checks (On); -- turn style checks back on
7407
NULL; -- this will generate an error message
7410
Finally the two argument form is allowed only if the first argument is
7411
@cite{On} or @cite{Off}. The effect is to turn of semantic style checks
7412
for the specified entity, as shown in the following example:
7415
pragma Style_Checks ("r"); -- require consistency of identifier casing
7417
Rf1 : Integer := ARG; -- incorrect, wrong case
7418
pragma Style_Checks (Off, Arg);
7419
Rf2 : Integer := ARG; -- OK, no error
7422
@node Pragma Subtitle,Pragma Suppress,Pragma Style_Checks,Implementation Defined Pragmas
7423
@anchor{gnat_rm/implementation_defined_pragmas pragma-subtitle}@anchor{c3}
7424
@section Pragma Subtitle
7430
pragma Subtitle ([Subtitle =>] STRING_LITERAL);
7433
This pragma is recognized for compatibility with other Ada compilers
7434
but is ignored by GNAT.
7436
@node Pragma Suppress,Pragma Suppress_All,Pragma Subtitle,Implementation Defined Pragmas
7437
@anchor{gnat_rm/implementation_defined_pragmas pragma-suppress}@anchor{c4}
7438
@section Pragma Suppress
7444
pragma Suppress (Identifier [, [On =>] Name]);
7447
This is a standard pragma, and supports all the check names required in
7448
the RM. It is included here because GNAT recognizes some additional check
7449
names that are implementation defined (as permitted by the RM):
7455
@cite{Alignment_Check} can be used to suppress alignment checks
7456
on addresses used in address clauses. Such checks can also be suppressed
7457
by suppressing range checks, but the specific use of @cite{Alignment_Check}
7458
allows suppression of alignment checks without suppressing other range checks.
7459
Note that @cite{Alignment_Check} is suppressed by default on machines (such as
7460
the x86) with non-strict alignment.
7463
@cite{Atomic_Synchronization} can be used to suppress the special memory
7464
synchronization instructions that are normally generated for access to
7465
@cite{Atomic} variables to ensure correct synchronization between tasks
7466
that use such variables for synchronization purposes.
7469
@cite{Duplicated_Tag_Check} Can be used to suppress the check that is generated
7470
for a duplicated tag value when a tagged type is declared.
7473
@cite{Container_Checks} Can be used to suppress all checks within Ada.Containers
7474
and instances of its children, including Tampering_Check.
7477
@cite{Tampering_Check} Can be used to suppress tampering check in the containers.
7480
@cite{Predicate_Check} can be used to control whether predicate checks are
7481
active. It is applicable only to predicates for which the policy is
7482
@cite{Check}. Unlike @cite{Assertion_Policy}, which determines if a given
7483
predicate is ignored or checked for the whole program, the use of
7484
@cite{Suppress} and @cite{Unsuppress} with this check name allows a given
7485
predicate to be turned on and off at specific points in the program.
7488
@cite{Validity_Check} can be used specifically to control validity checks.
7489
If @cite{Suppress} is used to suppress validity checks, then no validity
7490
checks are performed, including those specified by the appropriate compiler
7491
switch or the @cite{Validity_Checks} pragma.
7494
Additional check names previously introduced by use of the @cite{Check_Name}
7495
pragma are also allowed.
7498
Note that pragma Suppress gives the compiler permission to omit
7499
checks, but does not require the compiler to omit checks. The compiler
7500
will generate checks if they are essentially free, even when they are
7501
suppressed. In particular, if the compiler can prove that a certain
7502
check will necessarily fail, it will generate code to do an
7503
unconditional 'raise', even if checks are suppressed. The compiler
7506
Of course, run-time checks are omitted whenever the compiler can prove
7507
that they will not fail, whether or not checks are suppressed.
7509
@node Pragma Suppress_All,Pragma Suppress_Debug_Info,Pragma Suppress,Implementation Defined Pragmas
7510
@anchor{gnat_rm/implementation_defined_pragmas pragma-suppress-all}@anchor{c5}
7511
@section Pragma Suppress_All
7517
pragma Suppress_All;
7520
This pragma can appear anywhere within a unit.
7521
The effect is to apply @cite{Suppress (All_Checks)} to the unit
7522
in which it appears. This pragma is implemented for compatibility with DEC
7523
Ada 83 usage where it appears at the end of a unit, and for compatibility
7524
with Rational Ada, where it appears as a program unit pragma.
7525
The use of the standard Ada pragma @cite{Suppress (All_Checks)}
7526
as a normal configuration pragma is the preferred usage in GNAT.
7528
@node Pragma Suppress_Debug_Info,Pragma Suppress_Exception_Locations,Pragma Suppress_All,Implementation Defined Pragmas
7529
@anchor{gnat_rm/implementation_defined_pragmas pragma-suppress-debug-info}@anchor{c6}
7530
@section Pragma Suppress_Debug_Info
7536
pragma Suppress_Debug_Info ([Entity =>] LOCAL_NAME);
7539
This pragma can be used to suppress generation of debug information
7540
for the specified entity. It is intended primarily for use in debugging
7541
the debugger, and navigating around debugger problems.
7543
@node Pragma Suppress_Exception_Locations,Pragma Suppress_Initialization,Pragma Suppress_Debug_Info,Implementation Defined Pragmas
7544
@anchor{gnat_rm/implementation_defined_pragmas pragma-suppress-exception-locations}@anchor{c7}
7545
@section Pragma Suppress_Exception_Locations
7551
pragma Suppress_Exception_Locations;
7554
In normal mode, a raise statement for an exception by default generates
7555
an exception message giving the file name and line number for the location
7556
of the raise. This is useful for debugging and logging purposes, but this
7557
entails extra space for the strings for the messages. The configuration
7558
pragma @cite{Suppress_Exception_Locations} can be used to suppress the
7559
generation of these strings, with the result that space is saved, but the
7560
exception message for such raises is null. This configuration pragma may
7561
appear in a global configuration pragma file, or in a specific unit as
7562
usual. It is not required that this pragma be used consistently within
7563
a partition, so it is fine to have some units within a partition compiled
7564
with this pragma and others compiled in normal mode without it.
7566
@node Pragma Suppress_Initialization,Pragma Task_Name,Pragma Suppress_Exception_Locations,Implementation Defined Pragmas
7567
@anchor{gnat_rm/implementation_defined_pragmas pragma-suppress-initialization}@anchor{c8}
7568
@section Pragma Suppress_Initialization
7571
@geindex Suppressing initialization
7573
@geindex Initialization
7574
@geindex suppression of
7579
pragma Suppress_Initialization ([Entity =>] variable_or_subtype_Name);
7582
Here variable_or_subtype_Name is the name introduced by a type declaration
7583
or subtype declaration or the name of a variable introduced by an
7586
In the case of a type or subtype
7587
this pragma suppresses any implicit or explicit initialization
7588
for all variables of the given type or subtype,
7589
including initialization resulting from the use of pragmas
7590
Normalize_Scalars or Initialize_Scalars.
7592
This is considered a representation item, so it cannot be given after
7593
the type is frozen. It applies to all subsequent object declarations,
7594
and also any allocator that creates objects of the type.
7596
If the pragma is given for the first subtype, then it is considered
7597
to apply to the base type and all its subtypes. If the pragma is given
7598
for other than a first subtype, then it applies only to the given subtype.
7599
The pragma may not be given after the type is frozen.
7601
Note that this includes eliminating initialization of discriminants
7602
for discriminated types, and tags for tagged types. In these cases,
7603
you will have to use some non-portable mechanism (e.g. address
7604
overlays or unchecked conversion) to achieve required initialization
7605
of these fields before accessing any object of the corresponding type.
7607
For the variable case, implicit initialization for the named variable
7608
is suppressed, just as though its subtype had been given in a pragma
7609
Suppress_Initialization, as described above.
7611
@node Pragma Task_Name,Pragma Task_Storage,Pragma Suppress_Initialization,Implementation Defined Pragmas
7612
@anchor{gnat_rm/implementation_defined_pragmas pragma-task-name}@anchor{c9}
7613
@section Pragma Task_Name
7619
pragma Task_Name (string_EXPRESSION);
7622
This pragma appears within a task definition (like pragma
7623
@cite{Priority}) and applies to the task in which it appears. The
7624
argument must be of type String, and provides a name to be used for
7625
the task instance when the task is created. Note that this expression
7626
is not required to be static, and in particular, it can contain
7627
references to task discriminants. This facility can be used to
7628
provide different names for different tasks as they are created,
7629
as illustrated in the example below.
7631
The task name is recorded internally in the run-time structures
7632
and is accessible to tools like the debugger. In addition the
7633
routine @cite{Ada.Task_Identification.Image} will return this
7634
string, with a unique task address appended.
7637
-- Example of the use of pragma Task_Name
7639
with Ada.Task_Identification;
7640
use Ada.Task_Identification;
7641
with Text_IO; use Text_IO;
7644
type Astring is access String;
7646
task type Task_Typ (Name : access String) is
7647
pragma Task_Name (Name.all);
7650
task body Task_Typ is
7651
Nam : constant String := Image (Current_Task);
7653
Put_Line ("-->" & Nam (1 .. 14) & "<--");
7656
type Ptr_Task is access Task_Typ;
7657
Task_Var : Ptr_Task;
7661
new Task_Typ (new String'("This is task 1"));
7663
new Task_Typ (new String'("This is task 2"));
7667
@node Pragma Task_Storage,Pragma Test_Case,Pragma Task_Name,Implementation Defined Pragmas
7668
@anchor{gnat_rm/implementation_defined_pragmas pragma-task-storage}@anchor{ca}
7669
@section Pragma Task_Storage
7675
pragma Task_Storage (
7676
[Task_Type =>] LOCAL_NAME,
7677
[Top_Guard =>] static_integer_EXPRESSION);
7680
This pragma specifies the length of the guard area for tasks. The guard
7681
area is an additional storage area allocated to a task. A value of zero
7682
means that either no guard area is created or a minimal guard area is
7683
created, depending on the target. This pragma can appear anywhere a
7684
@cite{Storage_Size} attribute definition clause is allowed for a task
7687
@node Pragma Test_Case,Pragma Thread_Local_Storage,Pragma Task_Storage,Implementation Defined Pragmas
7688
@anchor{gnat_rm/implementation_defined_pragmas pragma-test-case}@anchor{cb}
7689
@section Pragma Test_Case
7698
[Name =>] static_string_Expression
7699
,[Mode =>] (Nominal | Robustness)
7700
[, Requires => Boolean_Expression]
7701
[, Ensures => Boolean_Expression]);
7704
The @cite{Test_Case} pragma allows defining fine-grain specifications
7705
for use by testing tools.
7706
The compiler checks the validity of the @cite{Test_Case} pragma, but its
7707
presence does not lead to any modification of the code generated by the
7710
@cite{Test_Case} pragmas may only appear immediately following the
7711
(separate) declaration of a subprogram in a package declaration, inside
7712
a package spec unit. Only other pragmas may intervene (that is appear
7713
between the subprogram declaration and a test case).
7715
The compiler checks that boolean expressions given in @cite{Requires} and
7716
@cite{Ensures} are valid, where the rules for @cite{Requires} are the
7717
same as the rule for an expression in @cite{Precondition} and the rules
7718
for @cite{Ensures} are the same as the rule for an expression in
7719
@cite{Postcondition}. In particular, attributes @cite{'Old} and
7720
@cite{'Result} can only be used within the @cite{Ensures}
7721
expression. The following is an example of use within a package spec:
7724
package Math_Functions is
7726
function Sqrt (Arg : Float) return Float;
7727
pragma Test_Case (Name => "Test 1",
7729
Requires => Arg < 10000,
7730
Ensures => Sqrt'Result < 10);
7735
The meaning of a test case is that there is at least one context where
7736
@cite{Requires} holds such that, if the associated subprogram is executed in
7737
that context, then @cite{Ensures} holds when the subprogram returns.
7738
Mode @cite{Nominal} indicates that the input context should also satisfy the
7739
precondition of the subprogram, and the output context should also satisfy its
7740
postcondition. Mode @cite{Robustness} indicates that the precondition and
7741
postcondition of the subprogram should be ignored for this test case.
7743
@node Pragma Thread_Local_Storage,Pragma Time_Slice,Pragma Test_Case,Implementation Defined Pragmas
7744
@anchor{gnat_rm/implementation_defined_pragmas pragma-thread-local-storage}@anchor{cc}
7745
@section Pragma Thread_Local_Storage
7748
@geindex Task specific storage
7750
@geindex TLS (Thread Local Storage)
7752
@geindex Task_Attributes
7757
pragma Thread_Local_Storage ([Entity =>] LOCAL_NAME);
7760
This pragma specifies that the specified entity, which must be
7761
a variable declared in a library level package, is to be marked as
7762
"Thread Local Storage" (@cite{TLS}). On systems supporting this (which
7763
include Windows, Solaris, GNU/Linux and VxWorks 6), this causes each
7764
thread (and hence each Ada task) to see a distinct copy of the variable.
7766
The variable may not have default initialization, and if there is
7767
an explicit initialization, it must be either @cite{null} for an
7768
access variable, or a static expression for a scalar variable.
7769
This provides a low level mechanism similar to that provided by
7770
the @cite{Ada.Task_Attributes} package, but much more efficient
7771
and is also useful in writing interface code that will interact
7772
with foreign threads.
7774
If this pragma is used on a system where @cite{TLS} is not supported,
7775
then an error message will be generated and the program will be rejected.
7777
@node Pragma Time_Slice,Pragma Title,Pragma Thread_Local_Storage,Implementation Defined Pragmas
7778
@anchor{gnat_rm/implementation_defined_pragmas pragma-time-slice}@anchor{cd}
7779
@section Pragma Time_Slice
7785
pragma Time_Slice (static_duration_EXPRESSION);
7788
For implementations of GNAT on operating systems where it is possible
7789
to supply a time slice value, this pragma may be used for this purpose.
7790
It is ignored if it is used in a system that does not allow this control,
7791
or if it appears in other than the main program unit.
7793
@node Pragma Title,Pragma Type_Invariant,Pragma Time_Slice,Implementation Defined Pragmas
7794
@anchor{gnat_rm/implementation_defined_pragmas pragma-title}@anchor{ce}
7795
@section Pragma Title
7801
pragma Title (TITLING_OPTION [, TITLING OPTION]);
7804
[Title =>] STRING_LITERAL,
7805
| [Subtitle =>] STRING_LITERAL
7808
Syntax checked but otherwise ignored by GNAT. This is a listing control
7809
pragma used in DEC Ada 83 implementations to provide a title and/or
7810
subtitle for the program listing. The program listing generated by GNAT
7811
does not have titles or subtitles.
7813
Unlike other pragmas, the full flexibility of named notation is allowed
7814
for this pragma, i.e., the parameters may be given in any order if named
7815
notation is used, and named and positional notation can be mixed
7816
following the normal rules for procedure calls in Ada.
7818
@node Pragma Type_Invariant,Pragma Type_Invariant_Class,Pragma Title,Implementation Defined Pragmas
7819
@anchor{gnat_rm/implementation_defined_pragmas pragma-type-invariant}@anchor{cf}
7820
@section Pragma Type_Invariant
7826
pragma Type_Invariant
7827
([Entity =>] type_LOCAL_NAME,
7828
[Check =>] EXPRESSION);
7831
The @cite{Type_Invariant} pragma is intended to be an exact
7832
replacement for the language-defined @cite{Type_Invariant}
7833
aspect, and shares its restrictions and semantics. It differs
7834
from the language defined @cite{Invariant} pragma in that it
7835
does not permit a string parameter, and it is
7836
controlled by the assertion identifier @cite{Type_Invariant}
7837
rather than @cite{Invariant}.
7839
@node Pragma Type_Invariant_Class,Pragma Unchecked_Union,Pragma Type_Invariant,Implementation Defined Pragmas
7840
@anchor{gnat_rm/implementation_defined_pragmas pragma-type-invariant-class}@anchor{d0}
7841
@section Pragma Type_Invariant_Class
7847
pragma Type_Invariant_Class
7848
([Entity =>] type_LOCAL_NAME,
7849
[Check =>] EXPRESSION);
7852
The @cite{Type_Invariant_Class} pragma is intended to be an exact
7853
replacement for the language-defined @cite{Type_Invariant'Class}
7854
aspect, and shares its restrictions and semantics.
7856
Note: This pragma is called @cite{Type_Invariant_Class} rather than
7857
@cite{Type_Invariant'Class} because the latter would not be strictly
7858
conforming to the allowed syntax for pragmas. The motivation
7859
for providing pragmas equivalent to the aspects is to allow a program
7860
to be written using the pragmas, and then compiled if necessary
7861
using an Ada compiler that does not recognize the pragmas or
7862
aspects, but is prepared to ignore the pragmas. The assertion
7863
policy that controls this pragma is @cite{Type_Invariant'Class},
7864
not @cite{Type_Invariant_Class}.
7866
@node Pragma Unchecked_Union,Pragma Unevaluated_Use_Of_Old,Pragma Type_Invariant_Class,Implementation Defined Pragmas
7867
@anchor{gnat_rm/implementation_defined_pragmas pragma-unchecked-union}@anchor{d1}
7868
@section Pragma Unchecked_Union
7871
@geindex Unions in C
7876
pragma Unchecked_Union (first_subtype_LOCAL_NAME);
7879
This pragma is used to specify a representation of a record type that is
7880
equivalent to a C union. It was introduced as a GNAT implementation defined
7881
pragma in the GNAT Ada 95 mode. Ada 2005 includes an extended version of this
7882
pragma, making it language defined, and GNAT fully implements this extended
7883
version in all language modes (Ada 83, Ada 95, and Ada 2005). For full
7884
details, consult the Ada 2012 Reference Manual, section B.3.3.
7886
@node Pragma Unevaluated_Use_Of_Old,Pragma Unimplemented_Unit,Pragma Unchecked_Union,Implementation Defined Pragmas
7887
@anchor{gnat_rm/implementation_defined_pragmas pragma-unevaluated-use-of-old}@anchor{d2}
7888
@section Pragma Unevaluated_Use_Of_Old
7891
@geindex Attribute Old
7893
@geindex Attribute Loop_Entry
7895
@geindex Unevaluated_Use_Of_Old
7900
pragma Unevaluated_Use_Of_Old (Error | Warn | Allow);
7903
This pragma controls the processing of attributes Old and Loop_Entry.
7904
If either of these attributes is used in a potentially unevaluated
7905
expression (e.g. the then or else parts of an if expression), then
7906
normally this usage is considered illegal if the prefix of the attribute
7907
is other than an entity name. The language requires this
7908
behavior for Old, and GNAT copies the same rule for Loop_Entry.
7910
The reason for this rule is that otherwise, we can have a situation
7911
where we save the Old value, and this results in an exception, even
7912
though we might not evaluate the attribute. Consider this example:
7915
package UnevalOld is
7917
procedure U (A : String; C : Boolean) -- ERROR
7918
with Post => (if C then A(1)'Old = K else True);
7922
If procedure U is called with a string with a lower bound of 2, and
7923
C false, then an exception would be raised trying to evaluate A(1)
7924
on entry even though the value would not be actually used.
7926
Although the rule guarantees against this possibility, it is sometimes
7927
too restrictive. For example if we know that the string has a lower
7928
bound of 1, then we will never raise an exception.
7929
The pragma @cite{Unevaluated_Use_Of_Old} can be
7930
used to modify this behavior. If the argument is @cite{Error} then an
7931
error is given (this is the default RM behavior). If the argument is
7932
@cite{Warn} then the usage is allowed as legal but with a warning
7933
that an exception might be raised. If the argument is @cite{Allow}
7934
then the usage is allowed as legal without generating a warning.
7936
This pragma may appear as a configuration pragma, or in a declarative
7937
part or package specification. In the latter case it applies to
7938
uses up to the end of the corresponding statement sequence or
7939
sequence of package declarations.
7941
@node Pragma Unimplemented_Unit,Pragma Universal_Aliasing,Pragma Unevaluated_Use_Of_Old,Implementation Defined Pragmas
7942
@anchor{gnat_rm/implementation_defined_pragmas pragma-unimplemented-unit}@anchor{d3}
7943
@section Pragma Unimplemented_Unit
7949
pragma Unimplemented_Unit;
7952
If this pragma occurs in a unit that is processed by the compiler, GNAT
7953
aborts with the message @code{xxx not implemented}, where
7954
@cite{xxx} is the name of the current compilation unit. This pragma is
7955
intended to allow the compiler to handle unimplemented library units in
7958
The abort only happens if code is being generated. Thus you can use
7959
specs of unimplemented packages in syntax or semantic checking mode.
7961
@node Pragma Universal_Aliasing,Pragma Universal_Data,Pragma Unimplemented_Unit,Implementation Defined Pragmas
7962
@anchor{gnat_rm/implementation_defined_pragmas pragma-universal-aliasing}@anchor{d4}
7963
@section Pragma Universal_Aliasing
7969
pragma Universal_Aliasing [([Entity =>] type_LOCAL_NAME)];
7972
@cite{type_LOCAL_NAME} must refer to a type declaration in the current
7973
declarative part. The effect is to inhibit strict type-based aliasing
7974
optimization for the given type. In other words, the effect is as though
7975
access types designating this type were subject to pragma No_Strict_Aliasing.
7976
For a detailed description of the strict aliasing optimization, and the
7977
situations in which it must be suppressed, see the section on
7978
@cite{Optimization and Strict Aliasing} in the @cite{GNAT User's Guide}.
7980
@node Pragma Universal_Data,Pragma Unmodified,Pragma Universal_Aliasing,Implementation Defined Pragmas
7981
@anchor{gnat_rm/implementation_defined_pragmas pragma-universal-data}@anchor{d5}
7982
@section Pragma Universal_Data
7988
pragma Universal_Data [(library_unit_Name)];
7991
This pragma is supported only for the AAMP target and is ignored for
7992
other targets. The pragma specifies that all library-level objects
7993
(Counter 0 data) associated with the library unit are to be accessed
7994
and updated using universal addressing (24-bit addresses for AAMP5)
7995
rather than the default of 16-bit Data Environment (DENV) addressing.
7996
Use of this pragma will generally result in less efficient code for
7997
references to global data associated with the library unit, but
7998
allows such data to be located anywhere in memory. This pragma is
7999
a library unit pragma, but can also be used as a configuration pragma
8000
(including use in the @code{gnat.adc} file). The functionality
8001
of this pragma is also available by applying the -univ switch on the
8002
compilations of units where universal addressing of the data is desired.
8004
@node Pragma Unmodified,Pragma Unreferenced,Pragma Universal_Data,Implementation Defined Pragmas
8005
@anchor{gnat_rm/implementation_defined_pragmas pragma-unmodified}@anchor{d6}
8006
@section Pragma Unmodified
8015
pragma Unmodified (LOCAL_NAME @{, LOCAL_NAME@});
8018
This pragma signals that the assignable entities (variables,
8019
@cite{out} parameters, @cite{in out} parameters) whose names are listed are
8020
deliberately not assigned in the current source unit. This
8021
suppresses warnings about the
8022
entities being referenced but not assigned, and in addition a warning will be
8023
generated if one of these entities is in fact assigned in the
8024
same unit as the pragma (or in the corresponding body, or one
8027
This is particularly useful for clearly signaling that a particular
8028
parameter is not modified, even though the spec suggests that it might
8031
For the variable case, warnings are never given for unreferenced variables
8032
whose name contains one of the substrings
8033
@cite{DISCARD@comma{} DUMMY@comma{} IGNORE@comma{} JUNK@comma{} UNUSED} in any casing. Such names
8034
are typically to be used in cases where such warnings are expected.
8035
Thus it is never necessary to use @cite{pragma Unmodified} for such
8036
variables, though it is harmless to do so.
8038
@node Pragma Unreferenced,Pragma Unreferenced_Objects,Pragma Unmodified,Implementation Defined Pragmas
8039
@anchor{gnat_rm/implementation_defined_pragmas pragma-unreferenced}@anchor{d7}
8040
@section Pragma Unreferenced
8044
@geindex unreferenced
8049
pragma Unreferenced (LOCAL_NAME @{, LOCAL_NAME@});
8050
pragma Unreferenced (library_unit_NAME @{, library_unit_NAME@});
8053
This pragma signals that the entities whose names are listed are
8054
deliberately not referenced in the current source unit after the
8055
occurrence of the pragma. This
8056
suppresses warnings about the
8057
entities being unreferenced, and in addition a warning will be
8058
generated if one of these entities is in fact subsequently referenced in the
8059
same unit as the pragma (or in the corresponding body, or one
8062
This is particularly useful for clearly signaling that a particular
8063
parameter is not referenced in some particular subprogram implementation
8064
and that this is deliberate. It can also be useful in the case of
8065
objects declared only for their initialization or finalization side
8068
If @cite{LOCAL_NAME} identifies more than one matching homonym in the
8069
current scope, then the entity most recently declared is the one to which
8070
the pragma applies. Note that in the case of accept formals, the pragma
8071
Unreferenced may appear immediately after the keyword @cite{do} which
8072
allows the indication of whether or not accept formals are referenced
8073
or not to be given individually for each accept statement.
8075
The left hand side of an assignment does not count as a reference for the
8076
purpose of this pragma. Thus it is fine to assign to an entity for which
8077
pragma Unreferenced is given.
8079
Note that if a warning is desired for all calls to a given subprogram,
8080
regardless of whether they occur in the same unit as the subprogram
8081
declaration, then this pragma should not be used (calls from another
8082
unit would not be flagged); pragma Obsolescent can be used instead
8083
for this purpose, see @ref{8d,,Pragma Obsolescent}.
8085
The second form of pragma @cite{Unreferenced} is used within a context
8086
clause. In this case the arguments must be unit names of units previously
8087
mentioned in @cite{with} clauses (similar to the usage of pragma
8088
@cite{Elaborate_All}. The effect is to suppress warnings about unreferenced
8089
units and unreferenced entities within these units.
8091
For the variable case, warnings are never given for unreferenced variables
8092
whose name contains one of the substrings
8093
@cite{DISCARD@comma{} DUMMY@comma{} IGNORE@comma{} JUNK@comma{} UNUSED} in any casing. Such names
8094
are typically to be used in cases where such warnings are expected.
8095
Thus it is never necessary to use @cite{pragma Unreferenced} for such
8096
variables, though it is harmless to do so.
8098
@node Pragma Unreferenced_Objects,Pragma Unreserve_All_Interrupts,Pragma Unreferenced,Implementation Defined Pragmas
8099
@anchor{gnat_rm/implementation_defined_pragmas pragma-unreferenced-objects}@anchor{d8}
8100
@section Pragma Unreferenced_Objects
8104
@geindex unreferenced
8109
pragma Unreferenced_Objects (local_subtype_NAME @{, local_subtype_NAME@});
8112
This pragma signals that for the types or subtypes whose names are
8113
listed, objects which are declared with one of these types or subtypes may
8114
not be referenced, and if no references appear, no warnings are given.
8116
This is particularly useful for objects which are declared solely for their
8117
initialization and finalization effect. Such variables are sometimes referred
8118
to as RAII variables (Resource Acquisition Is Initialization). Using this
8119
pragma on the relevant type (most typically a limited controlled type), the
8120
compiler will automatically suppress unwanted warnings about these variables
8121
not being referenced.
8123
@node Pragma Unreserve_All_Interrupts,Pragma Unsuppress,Pragma Unreferenced_Objects,Implementation Defined Pragmas
8124
@anchor{gnat_rm/implementation_defined_pragmas pragma-unreserve-all-interrupts}@anchor{d9}
8125
@section Pragma Unreserve_All_Interrupts
8131
pragma Unreserve_All_Interrupts;
8134
Normally certain interrupts are reserved to the implementation. Any attempt
8135
to attach an interrupt causes Program_Error to be raised, as described in
8136
RM C.3.2(22). A typical example is the @cite{SIGINT} interrupt used in
8137
many systems for a @code{Ctrl-C} interrupt. Normally this interrupt is
8138
reserved to the implementation, so that @code{Ctrl-C} can be used to
8139
interrupt execution.
8141
If the pragma @cite{Unreserve_All_Interrupts} appears anywhere in any unit in
8142
a program, then all such interrupts are unreserved. This allows the
8143
program to handle these interrupts, but disables their standard
8144
functions. For example, if this pragma is used, then pressing
8145
@code{Ctrl-C} will not automatically interrupt execution. However,
8146
a program can then handle the @cite{SIGINT} interrupt as it chooses.
8148
For a full list of the interrupts handled in a specific implementation,
8149
see the source code for the spec of @cite{Ada.Interrupts.Names} in
8150
file @code{a-intnam.ads}. This is a target dependent file that contains the
8151
list of interrupts recognized for a given target. The documentation in
8152
this file also specifies what interrupts are affected by the use of
8153
the @cite{Unreserve_All_Interrupts} pragma.
8155
For a more general facility for controlling what interrupts can be
8156
handled, see pragma @cite{Interrupt_State}, which subsumes the functionality
8157
of the @cite{Unreserve_All_Interrupts} pragma.
8159
@node Pragma Unsuppress,Pragma Use_VADS_Size,Pragma Unreserve_All_Interrupts,Implementation Defined Pragmas
8160
@anchor{gnat_rm/implementation_defined_pragmas pragma-unsuppress}@anchor{da}
8161
@section Pragma Unsuppress
8167
pragma Unsuppress (IDENTIFIER [, [On =>] NAME]);
8170
This pragma undoes the effect of a previous pragma @cite{Suppress}. If
8171
there is no corresponding pragma @cite{Suppress} in effect, it has no
8172
effect. The range of the effect is the same as for pragma
8173
@cite{Suppress}. The meaning of the arguments is identical to that used
8174
in pragma @cite{Suppress}.
8176
One important application is to ensure that checks are on in cases where
8177
code depends on the checks for its correct functioning, so that the code
8178
will compile correctly even if the compiler switches are set to suppress
8179
checks. For example, in a program that depends on external names of tagged
8180
types and wants to ensure that the duplicated tag check occurs even if all
8181
run-time checks are suppressed by a compiler switch, the following
8182
configuration pragma will ensure this test is not suppressed:
8185
pragma Unsuppress (Duplicated_Tag_Check);
8188
This pragma is standard in Ada 2005. It is available in all earlier versions
8189
of Ada as an implementation-defined pragma.
8191
Note that in addition to the checks defined in the Ada RM, GNAT recogizes a
8192
number of implementation-defined check names. See the description of pragma
8193
@cite{Suppress} for full details.
8195
@node Pragma Use_VADS_Size,Pragma Validity_Checks,Pragma Unsuppress,Implementation Defined Pragmas
8196
@anchor{gnat_rm/implementation_defined_pragmas pragma-use-vads-size}@anchor{db}
8197
@section Pragma Use_VADS_Size
8201
@geindex VADS compatibility
8203
@geindex Rational profile
8208
pragma Use_VADS_Size;
8211
This is a configuration pragma. In a unit to which it applies, any use
8212
of the 'Size attribute is automatically interpreted as a use of the
8213
'VADS_Size attribute. Note that this may result in incorrect semantic
8214
processing of valid Ada 95 or Ada 2005 programs. This is intended to aid in
8215
the handling of existing code which depends on the interpretation of Size
8216
as implemented in the VADS compiler. See description of the VADS_Size
8217
attribute for further details.
8219
@node Pragma Validity_Checks,Pragma Volatile,Pragma Use_VADS_Size,Implementation Defined Pragmas
8220
@anchor{gnat_rm/implementation_defined_pragmas pragma-validity-checks}@anchor{dc}
8221
@section Pragma Validity_Checks
8227
pragma Validity_Checks (string_LITERAL | ALL_CHECKS | On | Off);
8230
This pragma is used in conjunction with compiler switches to control the
8231
built-in validity checking provided by GNAT. The compiler switches, if set
8232
provide an initial setting for the switches, and this pragma may be used
8233
to modify these settings, or the settings may be provided entirely by
8234
the use of the pragma. This pragma can be used anywhere that a pragma
8235
is legal, including use as a configuration pragma (including use in
8236
the @code{gnat.adc} file).
8238
The form with a string literal specifies which validity options are to be
8239
activated. The validity checks are first set to include only the default
8240
reference manual settings, and then a string of letters in the string
8241
specifies the exact set of options required. The form of this string
8242
is exactly as described for the @emph{-gnatVx} compiler switch (see the
8243
GNAT User's Guide for details). For example the following two
8244
methods can be used to enable validity checking for mode @cite{in} and
8245
@cite{in out} subprogram parameters:
8252
pragma Validity_Checks ("im");
8257
$ gcc -c -gnatVim ...
8261
The form ALL_CHECKS activates all standard checks (its use is equivalent
8262
to the use of the @cite{gnatva} switch.
8264
The forms with @cite{Off} and @cite{On}
8265
can be used to temporarily disable validity checks
8266
as shown in the following example:
8269
pragma Validity_Checks ("c"); -- validity checks for copies
8270
pragma Validity_Checks (Off); -- turn off validity checks
8271
A := B; -- B will not be validity checked
8272
pragma Validity_Checks (On); -- turn validity checks back on
8273
A := C; -- C will be validity checked
8276
@node Pragma Volatile,Pragma Volatile_Full_Access,Pragma Validity_Checks,Implementation Defined Pragmas
8277
@anchor{gnat_rm/implementation_defined_pragmas pragma-volatile}@anchor{dd}
8278
@section Pragma Volatile
8284
pragma Volatile (LOCAL_NAME);
8287
This pragma is defined by the Ada Reference Manual, and the GNAT
8288
implementation is fully conformant with this definition. The reason it
8289
is mentioned in this section is that a pragma of the same name was supplied
8290
in some Ada 83 compilers, including DEC Ada 83. The Ada 95 / Ada 2005
8291
implementation of pragma Volatile is upwards compatible with the
8292
implementation in DEC Ada 83.
8294
@node Pragma Volatile_Full_Access,Pragma Volatile_Function,Pragma Volatile,Implementation Defined Pragmas
8295
@anchor{gnat_rm/implementation_defined_pragmas pragma-volatile-full-access}@anchor{de}
8296
@section Pragma Volatile_Full_Access
8302
pragma Volatile_Full_Access (LOCAL_NAME);
8305
This is similar in effect to pragma Volatile, except that any reference to the
8306
object is guaranteed to be done only with instructions that read or write all
8307
the bits of the object. Furthermore, if the object is of a composite type,
8308
then any reference to a component of the object is guaranteed to read and/or
8309
write all the bits of the object.
8311
The intention is that this be suitable for use with memory-mapped I/O devices
8312
on some machines. Note that there are two important respects in which this is
8313
different from @cite{pragma Atomic}. First a reference to a @cite{Volatile_Full_Access}
8314
object is not a sequential action in the RM 9.10 sense and, therefore, does
8315
not create a synchronization point. Second, in the case of @cite{pragma Atomic},
8316
there is no guarantee that all the bits will be accessed if the reference
8317
is not to the whole object; the compiler is allowed (and generally will)
8318
access only part of the object in this case.
8320
It is not permissible to specify @cite{Atomic} and @cite{Volatile_Full_Access} for
8323
It is not permissible to specify @cite{Volatile_Full_Access} for a composite
8324
(record or array) type or object that has at least one @cite{Aliased} component.
8326
@node Pragma Volatile_Function,Pragma Warning_As_Error,Pragma Volatile_Full_Access,Implementation Defined Pragmas
8327
@anchor{gnat_rm/implementation_defined_pragmas pragma-volatile-function}@anchor{df}
8328
@section Pragma Volatile_Function
8334
pragma Volatile_Function [ (boolean_EXPRESSION) ];
8337
For the semantics of this pragma, see the entry for aspect @cite{Volatile_Function}
8338
in the SPARK 2014 Reference Manual, section 7.1.2.
8340
@node Pragma Warning_As_Error,Pragma Warnings,Pragma Volatile_Function,Implementation Defined Pragmas
8341
@anchor{gnat_rm/implementation_defined_pragmas pragma-warning-as-error}@anchor{e0}
8342
@section Pragma Warning_As_Error
8348
pragma Warning_As_Error (static_string_EXPRESSION);
8351
This configuration pragma allows the programmer to specify a set
8352
of warnings that will be treated as errors. Any warning which
8353
matches the pattern given by the pragma argument will be treated
8354
as an error. This gives much more precise control that -gnatwe
8355
which treats all warnings as errors.
8357
The pattern may contain asterisks, which match zero or more characters in
8358
the message. For example, you can use
8359
@cite{pragma Warning_As_Error ("bits of*unused")} to treat the warning
8360
message @cite{warning: 960 bits of "a" unused} as an error. No other regular
8361
expression notations are permitted. All characters other than asterisk in
8362
these three specific cases are treated as literal characters in the match.
8363
The match is case insensitive, for example XYZ matches xyz.
8365
Note that the pattern matches if it occurs anywhere within the warning
8366
message string (it is not necessary to put an asterisk at the start and
8367
the end of the message, since this is implied).
8369
Another possibility for the static_string_EXPRESSION which works whether
8370
or not error tags are enabled (@emph{-gnatw.d}) is to use the
8371
@emph{-gnatw} tag string, enclosed in brackets,
8372
as shown in the example below, to treat a class of warnings as errors.
8374
The above use of patterns to match the message applies only to warning
8375
messages generated by the front end. This pragma can also be applied to
8376
warnings provided by the back end and mentioned in @ref{e1,,Pragma Warnings}.
8377
By using a single full @emph{-Wxxx} switch in the pragma, such warnings
8378
can also be treated as errors.
8380
The pragma can appear either in a global configuration pragma file
8381
(e.g. @code{gnat.adc}), or at the start of a file. Given a global
8382
configuration pragma file containing:
8385
pragma Warning_As_Error ("[-gnatwj]");
8388
which will treat all obsolescent feature warnings as errors, the
8389
following program compiles as shown (compile options here are
8390
@emph{-gnatwa.d -gnatl -gnatj55}).
8393
1. pragma Warning_As_Error ("*never assigned*");
8394
2. function Warnerr return String is
8397
>>> error: variable "X" is never read and
8398
never assigned [-gnatwv] [warning-as-error]
8402
>>> warning: variable "Y" is assigned but
8403
never read [-gnatwu]
8409
>>> error: use of "%" is an obsolescent
8410
feature (RM J.2(4)), use """ instead
8411
[-gnatwj] [warning-as-error]
8415
8 lines: No errors, 3 warnings (2 treated as errors)
8418
Note that this pragma does not affect the set of warnings issued in
8419
any way, it merely changes the effect of a matching warning if one
8420
is produced as a result of other warnings options. As shown in this
8421
example, if the pragma results in a warning being treated as an error,
8422
the tag is changed from "warning:" to "error:" and the string
8423
"[warning-as-error]" is appended to the end of the message.
8425
@node Pragma Warnings,Pragma Weak_External,Pragma Warning_As_Error,Implementation Defined Pragmas
8426
@anchor{gnat_rm/implementation_defined_pragmas id5}@anchor{e2}@anchor{gnat_rm/implementation_defined_pragmas pragma-warnings}@anchor{e1}
8427
@section Pragma Warnings
8433
pragma Warnings ([TOOL_NAME,] DETAILS [, REASON]);
8435
DETAILS ::= On | Off
8436
DETAILS ::= On | Off, local_NAME
8437
DETAILS ::= static_string_EXPRESSION
8438
DETAILS ::= On | Off, static_string_EXPRESSION
8440
TOOL_NAME ::= GNAT | GNATProve
8442
REASON ::= Reason => STRING_LITERAL @{& STRING_LITERAL@}
8445
Note: in Ada 83 mode, a string literal may be used in place of a static string
8446
expression (which does not exist in Ada 83).
8448
Note if the second argument of @cite{DETAILS} is a @cite{local_NAME} then the
8449
second form is always understood. If the intention is to use
8450
the fourth form, then you can write @cite{NAME & ""} to force the
8451
intepretation as a @cite{static_string_EXPRESSION}.
8453
Note: if the first argument is a valid @cite{TOOL_NAME}, it will be interpreted
8454
that way. The use of the @cite{TOOL_NAME} argument is relevant only to users
8455
of SPARK and GNATprove, see last part of this section for details.
8457
Normally warnings are enabled, with the output being controlled by
8458
the command line switch. Warnings (@cite{Off}) turns off generation of
8459
warnings until a Warnings (@cite{On}) is encountered or the end of the
8460
current unit. If generation of warnings is turned off using this
8461
pragma, then some or all of the warning messages are suppressed,
8462
regardless of the setting of the command line switches.
8464
The @cite{Reason} parameter may optionally appear as the last argument
8465
in any of the forms of this pragma. It is intended purely for the
8466
purposes of documenting the reason for the @cite{Warnings} pragma.
8467
The compiler will check that the argument is a static string but
8468
otherwise ignore this argument. Other tools may provide specialized
8469
processing for this string.
8471
The form with a single argument (or two arguments if Reason present),
8472
where the first argument is @cite{ON} or @cite{OFF}
8473
may be used as a configuration pragma.
8475
If the @cite{LOCAL_NAME} parameter is present, warnings are suppressed for
8476
the specified entity. This suppression is effective from the point where
8477
it occurs till the end of the extended scope of the variable (similar to
8478
the scope of @cite{Suppress}). This form cannot be used as a configuration
8481
In the case where the first argument is other than @cite{ON} or
8483
the third form with a single static_string_EXPRESSION argument (and possible
8484
reason) provides more precise
8485
control over which warnings are active. The string is a list of letters
8486
specifying which warnings are to be activated and which deactivated. The
8487
code for these letters is the same as the string used in the command
8488
line switch controlling warnings. For a brief summary, use the gnatmake
8489
command with no arguments, which will generate usage information containing
8490
the list of warnings switches supported. For
8491
full details see the section on @cite{Warning Message Control} in the
8492
@cite{GNAT User's Guide}.
8493
This form can also be used as a configuration pragma.
8495
The warnings controlled by the @emph{-gnatw} switch are generated by the
8496
front end of the compiler. The GCC back end can provide additional warnings
8497
and they are controlled by the @emph{-W} switch. Such warnings can be
8498
identified by the appearance of a string of the form @cite{[-Wxxx]} in the
8499
message which designates the @emph{-Wxxx} switch that controls the message.
8500
The form with a single static_string_EXPRESSION argument also works for these
8501
warnings, but the string must be a single full @emph{-Wxxx} switch in this
8502
case. The above reference lists a few examples of these additional warnings.
8504
The specified warnings will be in effect until the end of the program
8505
or another pragma Warnings is encountered. The effect of the pragma is
8506
cumulative. Initially the set of warnings is the standard default set
8507
as possibly modified by compiler switches. Then each pragma Warning
8508
modifies this set of warnings as specified. This form of the pragma may
8509
also be used as a configuration pragma.
8511
The fourth form, with an @cite{On|Off} parameter and a string, is used to
8512
control individual messages, based on their text. The string argument
8513
is a pattern that is used to match against the text of individual
8514
warning messages (not including the initial "warning: " tag).
8516
The pattern may contain asterisks, which match zero or more characters in
8517
the message. For example, you can use
8518
@cite{pragma Warnings (Off@comma{} "bits of*unused")} to suppress the warning
8519
message @cite{warning: 960 bits of "a" unused}. No other regular
8520
expression notations are permitted. All characters other than asterisk in
8521
these three specific cases are treated as literal characters in the match.
8522
The match is case insensitive, for example XYZ matches xyz.
8524
Note that the pattern matches if it occurs anywhere within the warning
8525
message string (it is not necessary to put an asterisk at the start and
8526
the end of the message, since this is implied).
8528
The above use of patterns to match the message applies only to warning
8529
messages generated by the front end. This form of the pragma with a string
8530
argument can also be used to control warnings provided by the back end and
8531
mentioned above. By using a single full @emph{-Wxxx} switch in the pragma,
8532
such warnings can be turned on and off.
8534
There are two ways to use the pragma in this form. The OFF form can be used
8535
as a configuration pragma. The effect is to suppress all warnings (if any)
8536
that match the pattern string throughout the compilation (or match the
8537
-W switch in the back end case).
8539
The second usage is to suppress a warning locally, and in this case, two
8540
pragmas must appear in sequence:
8543
pragma Warnings (Off, Pattern);
8544
... code where given warning is to be suppressed
8545
pragma Warnings (On, Pattern);
8548
In this usage, the pattern string must match in the Off and On
8549
pragmas, and (if @emph{-gnatw.w} is given) at least one matching
8550
warning must be suppressed.
8552
Note: to write a string that will match any warning, use the string
8553
@cite{"***"}. It will not work to use a single asterisk or two
8554
asterisks since this looks like an operator name. This form with three
8555
asterisks is similar in effect to specifying @cite{pragma Warnings (Off)} except (if @emph{-gnatw.w} is given) that a matching
8556
@cite{pragma Warnings (On@comma{} "***")} will be required. This can be
8557
helpful in avoiding forgetting to turn warnings back on.
8559
Note: the debug flag -gnatd.i (@cite{/NOWARNINGS_PRAGMAS} in VMS) can be
8560
used to cause the compiler to entirely ignore all WARNINGS pragmas. This can
8561
be useful in checking whether obsolete pragmas in existing programs are hiding
8564
Note: pragma Warnings does not affect the processing of style messages. See
8565
separate entry for pragma Style_Checks for control of style messages.
8567
Users of the formal verification tool GNATprove for the SPARK subset of Ada may
8568
use the version of the pragma with a @cite{TOOL_NAME} parameter.
8570
If present, @cite{TOOL_NAME} is the name of a tool, currently either @cite{GNAT} for the
8571
compiler or @cite{GNATprove} for the formal verification tool. A given tool only
8572
takes into account pragma Warnings that do not specify a tool name, or that
8573
specify the matching tool name. This makes it possible to disable warnings
8574
selectively for each tool, and as a consequence to detect useless pragma
8575
Warnings with switch @cite{-gnatw.w}.
8577
@node Pragma Weak_External,Pragma Wide_Character_Encoding,Pragma Warnings,Implementation Defined Pragmas
8578
@anchor{gnat_rm/implementation_defined_pragmas pragma-weak-external}@anchor{e3}
8579
@section Pragma Weak_External
8585
pragma Weak_External ([Entity =>] LOCAL_NAME);
8588
@cite{LOCAL_NAME} must refer to an object that is declared at the library
8589
level. This pragma specifies that the given entity should be marked as a
8590
weak symbol for the linker. It is equivalent to @cite{__attribute__((weak))}
8591
in GNU C and causes @cite{LOCAL_NAME} to be emitted as a weak symbol instead
8592
of a regular symbol, that is to say a symbol that does not have to be
8593
resolved by the linker if used in conjunction with a pragma Import.
8595
When a weak symbol is not resolved by the linker, its address is set to
8596
zero. This is useful in writing interfaces to external modules that may
8597
or may not be linked in the final executable, for example depending on
8598
configuration settings.
8600
If a program references at run time an entity to which this pragma has been
8601
applied, and the corresponding symbol was not resolved at link time, then
8602
the execution of the program is erroneous. It is not erroneous to take the
8603
Address of such an entity, for example to guard potential references,
8604
as shown in the example below.
8606
Some file formats do not support weak symbols so not all target machines
8607
support this pragma.
8610
-- Example of the use of pragma Weak_External
8612
package External_Module is
8614
pragma Import (C, key);
8615
pragma Weak_External (key);
8616
function Present return boolean;
8617
end External_Module;
8619
with System; use System;
8620
package body External_Module is
8621
function Present return boolean is
8623
return key'Address /= System.Null_Address;
8625
end External_Module;
8628
@node Pragma Wide_Character_Encoding,,Pragma Weak_External,Implementation Defined Pragmas
8629
@anchor{gnat_rm/implementation_defined_pragmas pragma-wide-character-encoding}@anchor{e4}
8630
@section Pragma Wide_Character_Encoding
8636
pragma Wide_Character_Encoding (IDENTIFIER | CHARACTER_LITERAL);
8639
This pragma specifies the wide character encoding to be used in program
8640
source text appearing subsequently. It is a configuration pragma, but may
8641
also be used at any point that a pragma is allowed, and it is permissible
8642
to have more than one such pragma in a file, allowing multiple encodings
8643
to appear within the same file.
8645
The argument can be an identifier or a character literal. In the identifier
8646
case, it is one of @cite{HEX}, @cite{UPPER}, @cite{SHIFT_JIS},
8647
@cite{EUC}, @cite{UTF8}, or @cite{BRACKETS}. In the character literal
8648
case it is correspondingly one of the characters @code{h}, @code{u},
8649
@code{s}, @code{e}, @code{8}, or @code{b}.
8651
Note that when the pragma is used within a file, it affects only the
8652
encoding within that file, and does not affect withed units, specs,
8655
@node Implementation Defined Aspects,Implementation Defined Attributes,Implementation Defined Pragmas,Top
8656
@anchor{gnat_rm/implementation_defined_aspects implementation-defined-aspects}@anchor{e5}@anchor{gnat_rm/implementation_defined_aspects doc}@anchor{e6}@anchor{gnat_rm/implementation_defined_aspects id1}@anchor{e7}
8657
@chapter Implementation Defined Aspects
8660
Ada defines (throughout the Ada 2012 reference manual, summarized
8661
in Annex K) a set of aspects that can be specified for certain entities.
8662
These language defined aspects are implemented in GNAT in Ada 2012 mode
8663
and work as described in the Ada 2012 Reference Manual.
8665
In addition, Ada 2012 allows implementations to define additional aspects
8666
whose meaning is defined by the implementation. GNAT provides
8667
a number of these implementation-defined aspects which can be used
8668
to extend and enhance the functionality of the compiler. This section of
8669
the GNAT reference manual describes these additional aspects.
8671
Note that any program using these aspects may not be portable to
8672
other compilers (although GNAT implements this set of aspects on all
8673
platforms). Therefore if portability to other compilers is an important
8674
consideration, you should minimize the use of these aspects.
8676
Note that for many of these aspects, the effect is essentially similar
8677
to the use of a pragma or attribute specification with the same name
8678
applied to the entity. For example, if we write:
8681
type R is range 1 .. 100
8682
with Value_Size => 10;
8685
then the effect is the same as:
8688
type R is range 1 .. 100;
8689
for R'Value_Size use 10;
8695
type R is new Integer
8696
with Shared => True;
8699
then the effect is the same as:
8702
type R is new Integer;
8706
In the documentation below, such cases are simply marked
8707
as being boolean aspects equivalent to the corresponding pragma
8708
or attribute definition clause.
8711
* Aspect Abstract_State::
8713
* Aspect Async_Readers::
8714
* Aspect Async_Writers::
8715
* Aspect Constant_After_Elaboration::
8716
* Aspect Contract_Cases::
8718
* Aspect Default_Initial_Condition::
8719
* Aspect Dimension::
8720
* Aspect Dimension_System::
8721
* Aspect Disable_Controlled::
8722
* Aspect Effective_Reads::
8723
* Aspect Effective_Writes::
8724
* Aspect Extensions_Visible::
8725
* Aspect Favor_Top_Level::
8728
* Aspect Initial_Condition::
8729
* Aspect Initializes::
8730
* Aspect Inline_Always::
8731
* Aspect Invariant::
8732
* Aspect Invariant'Class::
8734
* Aspect Linker_Section::
8735
* Aspect Lock_Free::
8736
* Aspect No_Elaboration_Code_All::
8737
* Aspect No_Tagged_Streams::
8738
* Aspect Object_Size::
8739
* Aspect Obsolescent::
8741
* Aspect Persistent_BSS::
8742
* Aspect Predicate::
8743
* Aspect Pure_Function::
8744
* Aspect Refined_Depends::
8745
* Aspect Refined_Global::
8746
* Aspect Refined_Post::
8747
* Aspect Refined_State::
8748
* Aspect Remote_Access_Type::
8749
* Aspect Scalar_Storage_Order::
8751
* Aspect Simple_Storage_Pool::
8752
* Aspect Simple_Storage_Pool_Type::
8753
* Aspect SPARK_Mode::
8754
* Aspect Suppress_Debug_Info::
8755
* Aspect Suppress_Initialization::
8756
* Aspect Test_Case::
8757
* Aspect Thread_Local_Storage::
8758
* Aspect Universal_Aliasing::
8759
* Aspect Universal_Data::
8760
* Aspect Unmodified::
8761
* Aspect Unreferenced::
8762
* Aspect Unreferenced_Objects::
8763
* Aspect Value_Size::
8764
* Aspect Volatile_Full_Access::
8765
* Aspect Volatile_Function::
8770
@node Aspect Abstract_State,Annotate,,Implementation Defined Aspects
8771
@anchor{gnat_rm/implementation_defined_aspects aspect-abstract-state}@anchor{e8}
8772
@section Aspect Abstract_State
8775
@geindex Abstract_State
8777
This aspect is equivalent to pragma @cite{Abstract_State}.
8779
@node Annotate,Aspect Async_Readers,Aspect Abstract_State,Implementation Defined Aspects
8780
@anchor{gnat_rm/implementation_defined_aspects annotate}@anchor{e9}
8786
There are three forms of this aspect (where ID is an identifier,
8787
and ARG is a general expression).
8792
@item @emph{Annotate => ID}
8794
Equivalent to @cite{pragma Annotate (ID@comma{} Entity => Name);}
8796
@item @emph{Annotate => (ID)}
8798
Equivalent to @cite{pragma Annotate (ID@comma{} Entity => Name);}
8800
@item @emph{Annotate => (ID ,ID @{, ARG@})}
8802
Equivalent to @cite{pragma Annotate (ID@comma{} ID @{@comma{} ARG@}@comma{} Entity => Name);}
8805
@node Aspect Async_Readers,Aspect Async_Writers,Annotate,Implementation Defined Aspects
8806
@anchor{gnat_rm/implementation_defined_aspects aspect-async-readers}@anchor{ea}
8807
@section Aspect Async_Readers
8810
@geindex Async_Readers
8812
This boolean aspect is equivalent to pragma @cite{Async_Readers}.
8814
@node Aspect Async_Writers,Aspect Constant_After_Elaboration,Aspect Async_Readers,Implementation Defined Aspects
8815
@anchor{gnat_rm/implementation_defined_aspects aspect-async-writers}@anchor{eb}
8816
@section Aspect Async_Writers
8819
@geindex Async_Writers
8821
This boolean aspect is equivalent to pragma @cite{Async_Writers}.
8823
@node Aspect Constant_After_Elaboration,Aspect Contract_Cases,Aspect Async_Writers,Implementation Defined Aspects
8824
@anchor{gnat_rm/implementation_defined_aspects aspect-constant-after-elaboration}@anchor{ec}
8825
@section Aspect Constant_After_Elaboration
8828
@geindex Constant_After_Elaboration
8830
This aspect is equivalent to pragma @cite{Constant_After_Elaboration}.
8832
@node Aspect Contract_Cases,Aspect Depends,Aspect Constant_After_Elaboration,Implementation Defined Aspects
8833
@anchor{gnat_rm/implementation_defined_aspects aspect-contract-cases}@anchor{ed}
8834
@section Aspect Contract_Cases
8837
@geindex Contract_Cases
8839
This aspect is equivalent to pragma @cite{Contract_Cases}, the sequence
8840
of clauses being enclosed in parentheses so that syntactically it is an
8843
@node Aspect Depends,Aspect Default_Initial_Condition,Aspect Contract_Cases,Implementation Defined Aspects
8844
@anchor{gnat_rm/implementation_defined_aspects aspect-depends}@anchor{ee}
8845
@section Aspect Depends
8850
This aspect is equivalent to pragma @cite{Depends}.
8852
@node Aspect Default_Initial_Condition,Aspect Dimension,Aspect Depends,Implementation Defined Aspects
8853
@anchor{gnat_rm/implementation_defined_aspects aspect-default-initial-condition}@anchor{ef}
8854
@section Aspect Default_Initial_Condition
8857
@geindex Default_Initial_Condition
8859
This aspect is equivalent to pragma @cite{Default_Initial_Condition}.
8861
@node Aspect Dimension,Aspect Dimension_System,Aspect Default_Initial_Condition,Implementation Defined Aspects
8862
@anchor{gnat_rm/implementation_defined_aspects aspect-dimension}@anchor{f0}
8863
@section Aspect Dimension
8868
The @cite{Dimension} aspect is used to specify the dimensions of a given
8869
subtype of a dimensioned numeric type. The aspect also specifies a symbol
8870
used when doing formatted output of dimensioned quantities. The syntax is:
8874
([Symbol =>] SYMBOL, DIMENSION_VALUE @{, DIMENSION_Value@})
8876
SYMBOL ::= STRING_LITERAL | CHARACTER_LITERAL
8880
| others => RATIONAL
8881
| DISCRETE_CHOICE_LIST => RATIONAL
8883
RATIONAL ::= [-] NUMERIC_LITERAL [/ NUMERIC_LITERAL]
8886
This aspect can only be applied to a subtype whose parent type has
8887
a @cite{Dimension_Systen} aspect. The aspect must specify values for
8888
all dimensions of the system. The rational values are the powers of the
8889
corresponding dimensions that are used by the compiler to verify that
8890
physical (numeric) computations are dimensionally consistent. For example,
8891
the computation of a force must result in dimensions (L => 1, M => 1, T => -2).
8892
For further examples of the usage
8893
of this aspect, see package @cite{System.Dim.Mks}.
8894
Note that when the dimensioned type is an integer type, then any
8895
dimension value must be an integer literal.
8897
@node Aspect Dimension_System,Aspect Disable_Controlled,Aspect Dimension,Implementation Defined Aspects
8898
@anchor{gnat_rm/implementation_defined_aspects aspect-dimension-system}@anchor{f1}
8899
@section Aspect Dimension_System
8902
@geindex Dimension_System
8904
The @cite{Dimension_System} aspect is used to define a system of
8905
dimensions that will be used in subsequent subtype declarations with
8906
@cite{Dimension} aspects that reference this system. The syntax is:
8909
with Dimension_System => (DIMENSION @{, DIMENSION@});
8911
DIMENSION ::= ([Unit_Name =>] IDENTIFIER,
8912
[Unit_Symbol =>] SYMBOL,
8913
[Dim_Symbol =>] SYMBOL)
8915
SYMBOL ::= CHARACTER_LITERAL | STRING_LITERAL
8918
This aspect is applied to a type, which must be a numeric derived type
8919
(typically a floating-point type), that
8920
will represent values within the dimension system. Each @cite{DIMENSION}
8921
corresponds to one particular dimension. A maximum of 7 dimensions may
8922
be specified. @cite{Unit_Name} is the name of the dimension (for example
8923
@cite{Meter}). @cite{Unit_Symbol} is the shorthand used for quantities
8924
of this dimension (for example @cite{m} for @cite{Meter}).
8925
@cite{Dim_Symbol} gives
8926
the identification within the dimension system (typically this is a
8927
single letter, e.g. @cite{L} standing for length for unit name @cite{Meter}).
8928
The @cite{Unit_Symbol} is used in formatted output of dimensioned quantities.
8929
The @cite{Dim_Symbol} is used in error messages when numeric operations have
8930
inconsistent dimensions.
8932
GNAT provides the standard definition of the International MKS system in
8933
the run-time package @cite{System.Dim.Mks}. You can easily define
8934
similar packages for cgs units or British units, and define conversion factors
8935
between values in different systems. The MKS system is characterized by the
8939
type Mks_Type is new Long_Long_Float with
8940
Dimension_System => (
8941
(Unit_Name => Meter, Unit_Symbol => 'm', Dim_Symbol => 'L'),
8942
(Unit_Name => Kilogram, Unit_Symbol => "kg", Dim_Symbol => 'M'),
8943
(Unit_Name => Second, Unit_Symbol => 's', Dim_Symbol => 'T'),
8944
(Unit_Name => Ampere, Unit_Symbol => 'A', Dim_Symbol => 'I'),
8945
(Unit_Name => Kelvin, Unit_Symbol => 'K', Dim_Symbol => '@@'),
8946
(Unit_Name => Mole, Unit_Symbol => "mol", Dim_Symbol => 'N'),
8947
(Unit_Name => Candela, Unit_Symbol => "cd", Dim_Symbol => 'J'));
8950
Note that in the above type definition, we use the @cite{at} symbol (@code{@@}) to
8951
represent a theta character (avoiding the use of extended Latin-1
8952
characters in this context).
8954
See section 'Performing Dimensionality Analysis in GNAT' in the GNAT Users
8955
Guide for detailed examples of use of the dimension system.
8957
@node Aspect Disable_Controlled,Aspect Effective_Reads,Aspect Dimension_System,Implementation Defined Aspects
8958
@anchor{gnat_rm/implementation_defined_aspects aspect-disable-controlled}@anchor{f2}
8959
@section Aspect Disable_Controlled
8962
@geindex Disable_Controlled
8964
The aspect @cite{Disable_Controlled} is defined for controlled record types. If
8965
active, this aspect causes suppression of all related calls to @cite{Initialize},
8966
@cite{Adjust}, and @cite{Finalize}. The intended use is for conditional compilation,
8967
where for example you might want a record to be controlled or not depending on
8968
whether some run-time check is enabled or suppressed.
8970
@node Aspect Effective_Reads,Aspect Effective_Writes,Aspect Disable_Controlled,Implementation Defined Aspects
8971
@anchor{gnat_rm/implementation_defined_aspects aspect-effective-reads}@anchor{f3}
8972
@section Aspect Effective_Reads
8975
@geindex Effective_Reads
8977
This aspect is equivalent to pragma @cite{Effective_Reads}.
8979
@node Aspect Effective_Writes,Aspect Extensions_Visible,Aspect Effective_Reads,Implementation Defined Aspects
8980
@anchor{gnat_rm/implementation_defined_aspects aspect-effective-writes}@anchor{f4}
8981
@section Aspect Effective_Writes
8984
@geindex Effective_Writes
8986
This aspect is equivalent to pragma @cite{Effective_Writes}.
8988
@node Aspect Extensions_Visible,Aspect Favor_Top_Level,Aspect Effective_Writes,Implementation Defined Aspects
8989
@anchor{gnat_rm/implementation_defined_aspects aspect-extensions-visible}@anchor{f5}
8990
@section Aspect Extensions_Visible
8993
@geindex Extensions_Visible
8995
This aspect is equivalent to pragma @cite{Extensions_Visible}.
8997
@node Aspect Favor_Top_Level,Aspect Ghost,Aspect Extensions_Visible,Implementation Defined Aspects
8998
@anchor{gnat_rm/implementation_defined_aspects aspect-favor-top-level}@anchor{f6}
8999
@section Aspect Favor_Top_Level
9002
@geindex Favor_Top_Level
9004
This boolean aspect is equivalent to pragma @cite{Favor_Top_Level}.
9006
@node Aspect Ghost,Aspect Global,Aspect Favor_Top_Level,Implementation Defined Aspects
9007
@anchor{gnat_rm/implementation_defined_aspects aspect-ghost}@anchor{f7}
9008
@section Aspect Ghost
9013
This aspect is equivalent to pragma @cite{Ghost}.
9015
@node Aspect Global,Aspect Initial_Condition,Aspect Ghost,Implementation Defined Aspects
9016
@anchor{gnat_rm/implementation_defined_aspects aspect-global}@anchor{f8}
9017
@section Aspect Global
9022
This aspect is equivalent to pragma @cite{Global}.
9024
@node Aspect Initial_Condition,Aspect Initializes,Aspect Global,Implementation Defined Aspects
9025
@anchor{gnat_rm/implementation_defined_aspects aspect-initial-condition}@anchor{f9}
9026
@section Aspect Initial_Condition
9029
@geindex Initial_Condition
9031
This aspect is equivalent to pragma @cite{Initial_Condition}.
9033
@node Aspect Initializes,Aspect Inline_Always,Aspect Initial_Condition,Implementation Defined Aspects
9034
@anchor{gnat_rm/implementation_defined_aspects aspect-initializes}@anchor{fa}
9035
@section Aspect Initializes
9038
@geindex Initializes
9040
This aspect is equivalent to pragma @cite{Initializes}.
9042
@node Aspect Inline_Always,Aspect Invariant,Aspect Initializes,Implementation Defined Aspects
9043
@anchor{gnat_rm/implementation_defined_aspects aspect-inline-always}@anchor{fb}
9044
@section Aspect Inline_Always
9047
@geindex Inline_Always
9049
This boolean aspect is equivalent to pragma @cite{Inline_Always}.
9051
@node Aspect Invariant,Aspect Invariant'Class,Aspect Inline_Always,Implementation Defined Aspects
9052
@anchor{gnat_rm/implementation_defined_aspects aspect-invariant}@anchor{fc}
9053
@section Aspect Invariant
9058
This aspect is equivalent to pragma @cite{Invariant}. It is a
9059
synonym for the language defined aspect @cite{Type_Invariant} except
9060
that it is separately controllable using pragma @cite{Assertion_Policy}.
9062
@node Aspect Invariant'Class,Aspect Iterable,Aspect Invariant,Implementation Defined Aspects
9063
@anchor{gnat_rm/implementation_defined_aspects aspect-invariant-class}@anchor{fd}
9064
@section Aspect Invariant'Class
9067
@geindex Invariant'Class
9069
This aspect is equivalent to pragma @cite{Type_Invariant_Class}. It is a
9070
synonym for the language defined aspect @cite{Type_Invariant'Class} except
9071
that it is separately controllable using pragma @cite{Assertion_Policy}.
9073
@node Aspect Iterable,Aspect Linker_Section,Aspect Invariant'Class,Implementation Defined Aspects
9074
@anchor{gnat_rm/implementation_defined_aspects aspect-iterable}@anchor{fe}
9075
@section Aspect Iterable
9080
This aspect provides a light-weight mechanism for loops and quantified
9081
expressions over container types, without the overhead imposed by the tampering
9082
checks of standard Ada 2012 iterators. The value of the aspect is an aggregate
9083
with four named components: @cite{First}, @cite{Next}, @cite{Has_Element}, and @cite{Element} (the
9084
last one being optional). When only 3 components are specified, only the
9085
@cite{for .. in} form of iteration over cursors is available. When all 4 components
9086
are specified, both this form and the @cite{for .. of} form of iteration over
9087
elements are available. The following is a typical example of use:
9090
type List is private with
9091
Iterable => (First => First_Cursor,
9093
Has_Element => Cursor_Has_Element,
9094
[Element => Get_Element]);
9101
The value denoted by @cite{First} must denote a primitive operation of the
9102
container type that returns a @cite{Cursor}, which must a be a type declared in
9103
the container package or visible from it. For example:
9107
function First_Cursor (Cont : Container) return Cursor;
9114
The value of @cite{Next} is a primitive operation of the container type that takes
9115
both a container and a cursor and yields a cursor. For example:
9119
function Advance (Cont : Container; Position : Cursor) return Cursor;
9126
The value of @cite{Has_Element} is a primitive operation of the container type
9127
that takes both a container and a cursor and yields a boolean. For example:
9131
function Cursor_Has_Element (Cont : Container; Position : Cursor) return Boolean;
9138
The value of @cite{Element} is a primitive operation of the container type that
9139
takes both a container and a cursor and yields an @cite{Element_Type}, which must
9140
be a type declared in the container package or visible from it. For example:
9144
function Get_Element (Cont : Container; Position : Cursor) return Element_Type;
9147
This aspect is used in the GNAT-defined formal container packages.
9149
@node Aspect Linker_Section,Aspect Lock_Free,Aspect Iterable,Implementation Defined Aspects
9150
@anchor{gnat_rm/implementation_defined_aspects aspect-linker-section}@anchor{ff}
9151
@section Aspect Linker_Section
9154
@geindex Linker_Section
9156
This aspect is equivalent to an @cite{Linker_Section} pragma.
9158
@node Aspect Lock_Free,Aspect No_Elaboration_Code_All,Aspect Linker_Section,Implementation Defined Aspects
9159
@anchor{gnat_rm/implementation_defined_aspects aspect-lock-free}@anchor{100}
9160
@section Aspect Lock_Free
9165
This boolean aspect is equivalent to pragma @cite{Lock_Free}.
9167
@node Aspect No_Elaboration_Code_All,Aspect No_Tagged_Streams,Aspect Lock_Free,Implementation Defined Aspects
9168
@anchor{gnat_rm/implementation_defined_aspects aspect-no-elaboration-code-all}@anchor{101}
9169
@section Aspect No_Elaboration_Code_All
9172
@geindex No_Elaboration_Code_All
9174
This aspect is equivalent to a @cite{pragma No_Elaboration_Code_All}
9175
statement for a program unit.
9177
@node Aspect No_Tagged_Streams,Aspect Object_Size,Aspect No_Elaboration_Code_All,Implementation Defined Aspects
9178
@anchor{gnat_rm/implementation_defined_aspects aspect-no-tagged-streams}@anchor{102}
9179
@section Aspect No_Tagged_Streams
9182
@geindex No_Tagged_Streams
9184
This aspect is equivalent to a @cite{pragma No_Tagged_Streams} with an
9185
argument specifying a root tagged type (thus this aspect can only be
9186
applied to such a type).
9188
@node Aspect Object_Size,Aspect Obsolescent,Aspect No_Tagged_Streams,Implementation Defined Aspects
9189
@anchor{gnat_rm/implementation_defined_aspects aspect-object-size}@anchor{103}
9190
@section Aspect Object_Size
9193
@geindex Object_Size
9195
This aspect is equivalent to an @cite{Object_Size} attribute definition
9198
@node Aspect Obsolescent,Aspect Part_Of,Aspect Object_Size,Implementation Defined Aspects
9199
@anchor{gnat_rm/implementation_defined_aspects aspect-obsolescent}@anchor{104}
9200
@section Aspect Obsolescent
9203
@geindex Obsolsecent
9205
This aspect is equivalent to an @cite{Obsolescent} pragma. Note that the
9206
evaluation of this aspect happens at the point of occurrence, it is not
9207
delayed until the freeze point.
9209
@node Aspect Part_Of,Aspect Persistent_BSS,Aspect Obsolescent,Implementation Defined Aspects
9210
@anchor{gnat_rm/implementation_defined_aspects aspect-part-of}@anchor{105}
9211
@section Aspect Part_Of
9216
This aspect is equivalent to pragma @cite{Part_Of}.
9218
@node Aspect Persistent_BSS,Aspect Predicate,Aspect Part_Of,Implementation Defined Aspects
9219
@anchor{gnat_rm/implementation_defined_aspects aspect-persistent-bss}@anchor{106}
9220
@section Aspect Persistent_BSS
9223
@geindex Persistent_BSS
9225
This boolean aspect is equivalent to pragma @cite{Persistent_BSS}.
9227
@node Aspect Predicate,Aspect Pure_Function,Aspect Persistent_BSS,Implementation Defined Aspects
9228
@anchor{gnat_rm/implementation_defined_aspects aspect-predicate}@anchor{107}
9229
@section Aspect Predicate
9234
This aspect is equivalent to pragma @cite{Predicate}. It is thus
9235
similar to the language defined aspects @cite{Dynamic_Predicate}
9236
and @cite{Static_Predicate} except that whether the resulting
9237
predicate is static or dynamic is controlled by the form of the
9238
expression. It is also separately controllable using pragma
9239
@cite{Assertion_Policy}.
9241
@node Aspect Pure_Function,Aspect Refined_Depends,Aspect Predicate,Implementation Defined Aspects
9242
@anchor{gnat_rm/implementation_defined_aspects aspect-pure-function}@anchor{108}
9243
@section Aspect Pure_Function
9246
@geindex Pure_Function
9248
This boolean aspect is equivalent to pragma @cite{Pure_Function}.
9250
@node Aspect Refined_Depends,Aspect Refined_Global,Aspect Pure_Function,Implementation Defined Aspects
9251
@anchor{gnat_rm/implementation_defined_aspects aspect-refined-depends}@anchor{109}
9252
@section Aspect Refined_Depends
9255
@geindex Refined_Depends
9257
This aspect is equivalent to pragma @cite{Refined_Depends}.
9259
@node Aspect Refined_Global,Aspect Refined_Post,Aspect Refined_Depends,Implementation Defined Aspects
9260
@anchor{gnat_rm/implementation_defined_aspects aspect-refined-global}@anchor{10a}
9261
@section Aspect Refined_Global
9264
@geindex Refined_Global
9266
This aspect is equivalent to pragma @cite{Refined_Global}.
9268
@node Aspect Refined_Post,Aspect Refined_State,Aspect Refined_Global,Implementation Defined Aspects
9269
@anchor{gnat_rm/implementation_defined_aspects aspect-refined-post}@anchor{10b}
9270
@section Aspect Refined_Post
9273
@geindex Refined_Post
9275
This aspect is equivalent to pragma @cite{Refined_Post}.
9277
@node Aspect Refined_State,Aspect Remote_Access_Type,Aspect Refined_Post,Implementation Defined Aspects
9278
@anchor{gnat_rm/implementation_defined_aspects aspect-refined-state}@anchor{10c}
9279
@section Aspect Refined_State
9282
@geindex Refined_State
9284
This aspect is equivalent to pragma @cite{Refined_State}.
9286
@node Aspect Remote_Access_Type,Aspect Scalar_Storage_Order,Aspect Refined_State,Implementation Defined Aspects
9287
@anchor{gnat_rm/implementation_defined_aspects aspect-remote-access-type}@anchor{10d}
9288
@section Aspect Remote_Access_Type
9291
@geindex Remote_Access_Type
9293
This aspect is equivalent to pragma @cite{Remote_Access_Type}.
9295
@node Aspect Scalar_Storage_Order,Aspect Shared,Aspect Remote_Access_Type,Implementation Defined Aspects
9296
@anchor{gnat_rm/implementation_defined_aspects aspect-scalar-storage-order}@anchor{10e}
9297
@section Aspect Scalar_Storage_Order
9300
@geindex Scalar_Storage_Order
9302
This aspect is equivalent to a @cite{Scalar_Storage_Order}
9303
attribute definition clause.
9305
@node Aspect Shared,Aspect Simple_Storage_Pool,Aspect Scalar_Storage_Order,Implementation Defined Aspects
9306
@anchor{gnat_rm/implementation_defined_aspects aspect-shared}@anchor{10f}
9307
@section Aspect Shared
9312
This boolean aspect is equivalent to pragma @cite{Shared},
9313
and is thus a synonym for aspect @cite{Atomic}.
9315
@node Aspect Simple_Storage_Pool,Aspect Simple_Storage_Pool_Type,Aspect Shared,Implementation Defined Aspects
9316
@anchor{gnat_rm/implementation_defined_aspects aspect-simple-storage-pool}@anchor{110}
9317
@section Aspect Simple_Storage_Pool
9320
@geindex Simple_Storage_Pool
9322
This aspect is equivalent to a @cite{Simple_Storage_Pool}
9323
attribute definition clause.
9325
@node Aspect Simple_Storage_Pool_Type,Aspect SPARK_Mode,Aspect Simple_Storage_Pool,Implementation Defined Aspects
9326
@anchor{gnat_rm/implementation_defined_aspects aspect-simple-storage-pool-type}@anchor{111}
9327
@section Aspect Simple_Storage_Pool_Type
9330
@geindex Simple_Storage_Pool_Type
9332
This boolean aspect is equivalent to pragma @cite{Simple_Storage_Pool_Type}.
9334
@node Aspect SPARK_Mode,Aspect Suppress_Debug_Info,Aspect Simple_Storage_Pool_Type,Implementation Defined Aspects
9335
@anchor{gnat_rm/implementation_defined_aspects aspect-spark-mode}@anchor{112}
9336
@section Aspect SPARK_Mode
9341
This aspect is equivalent to pragma @cite{SPARK_Mode} and
9342
may be specified for either or both of the specification and body
9343
of a subprogram or package.
9345
@node Aspect Suppress_Debug_Info,Aspect Suppress_Initialization,Aspect SPARK_Mode,Implementation Defined Aspects
9346
@anchor{gnat_rm/implementation_defined_aspects aspect-suppress-debug-info}@anchor{113}
9347
@section Aspect Suppress_Debug_Info
9350
@geindex Suppress_Debug_Info
9352
This boolean aspect is equivalent to pragma @cite{Suppress_Debug_Info}.
9354
@node Aspect Suppress_Initialization,Aspect Test_Case,Aspect Suppress_Debug_Info,Implementation Defined Aspects
9355
@anchor{gnat_rm/implementation_defined_aspects aspect-suppress-initialization}@anchor{114}
9356
@section Aspect Suppress_Initialization
9359
@geindex Suppress_Initialization
9361
This boolean aspect is equivalent to pragma @cite{Suppress_Initialization}.
9363
@node Aspect Test_Case,Aspect Thread_Local_Storage,Aspect Suppress_Initialization,Implementation Defined Aspects
9364
@anchor{gnat_rm/implementation_defined_aspects aspect-test-case}@anchor{115}
9365
@section Aspect Test_Case
9370
This aspect is equivalent to pragma @cite{Test_Case}.
9372
@node Aspect Thread_Local_Storage,Aspect Universal_Aliasing,Aspect Test_Case,Implementation Defined Aspects
9373
@anchor{gnat_rm/implementation_defined_aspects aspect-thread-local-storage}@anchor{116}
9374
@section Aspect Thread_Local_Storage
9377
@geindex Thread_Local_Storage
9379
This boolean aspect is equivalent to pragma @cite{Thread_Local_Storage}.
9381
@node Aspect Universal_Aliasing,Aspect Universal_Data,Aspect Thread_Local_Storage,Implementation Defined Aspects
9382
@anchor{gnat_rm/implementation_defined_aspects aspect-universal-aliasing}@anchor{117}
9383
@section Aspect Universal_Aliasing
9386
@geindex Universal_Aliasing
9388
This boolean aspect is equivalent to pragma @cite{Universal_Aliasing}.
9390
@node Aspect Universal_Data,Aspect Unmodified,Aspect Universal_Aliasing,Implementation Defined Aspects
9391
@anchor{gnat_rm/implementation_defined_aspects aspect-universal-data}@anchor{118}
9392
@section Aspect Universal_Data
9395
@geindex Universal_Data
9397
This aspect is equivalent to pragma @cite{Universal_Data}.
9399
@node Aspect Unmodified,Aspect Unreferenced,Aspect Universal_Data,Implementation Defined Aspects
9400
@anchor{gnat_rm/implementation_defined_aspects aspect-unmodified}@anchor{119}
9401
@section Aspect Unmodified
9406
This boolean aspect is equivalent to pragma @cite{Unmodified}.
9408
@node Aspect Unreferenced,Aspect Unreferenced_Objects,Aspect Unmodified,Implementation Defined Aspects
9409
@anchor{gnat_rm/implementation_defined_aspects aspect-unreferenced}@anchor{11a}
9410
@section Aspect Unreferenced
9413
@geindex Unreferenced
9415
This boolean aspect is equivalent to pragma @cite{Unreferenced}. Note that
9416
in the case of formal parameters, it is not permitted to have aspects for
9417
a formal parameter, so in this case the pragma form must be used.
9419
@node Aspect Unreferenced_Objects,Aspect Value_Size,Aspect Unreferenced,Implementation Defined Aspects
9420
@anchor{gnat_rm/implementation_defined_aspects aspect-unreferenced-objects}@anchor{11b}
9421
@section Aspect Unreferenced_Objects
9424
@geindex Unreferenced_Objects
9426
This boolean aspect is equivalent to pragma @cite{Unreferenced_Objects}.
9428
@node Aspect Value_Size,Aspect Volatile_Full_Access,Aspect Unreferenced_Objects,Implementation Defined Aspects
9429
@anchor{gnat_rm/implementation_defined_aspects aspect-value-size}@anchor{11c}
9430
@section Aspect Value_Size
9435
This aspect is equivalent to a @cite{Value_Size}
9436
attribute definition clause.
9438
@node Aspect Volatile_Full_Access,Aspect Volatile_Function,Aspect Value_Size,Implementation Defined Aspects
9439
@anchor{gnat_rm/implementation_defined_aspects aspect-volatile-full-access}@anchor{11d}
9440
@section Aspect Volatile_Full_Access
9443
@geindex Volatile_Full_Access
9445
This boolean aspect is equivalent to pragma @cite{Volatile_Full_Access}.
9447
@node Aspect Volatile_Function,Aspect Warnings,Aspect Volatile_Full_Access,Implementation Defined Aspects
9448
@anchor{gnat_rm/implementation_defined_aspects aspect-volatile-function}@anchor{11e}
9449
@section Aspect Volatile_Function
9452
@geindex Volatile_Function
9454
This boolean aspect is equivalent to pragma @cite{Volatile_Function}.
9456
@node Aspect Warnings,,Aspect Volatile_Function,Implementation Defined Aspects
9457
@anchor{gnat_rm/implementation_defined_aspects aspect-warnings}@anchor{11f}
9458
@section Aspect Warnings
9463
This aspect is equivalent to the two argument form of pragma @cite{Warnings},
9464
where the first argument is @cite{ON} or @cite{OFF} and the second argument
9467
@node Implementation Defined Attributes,Standard and Implementation Defined Restrictions,Implementation Defined Aspects,Top
9468
@anchor{gnat_rm/implementation_defined_attributes doc}@anchor{120}@anchor{gnat_rm/implementation_defined_attributes implementation-defined-attributes}@anchor{8}@anchor{gnat_rm/implementation_defined_attributes id1}@anchor{121}
9469
@chapter Implementation Defined Attributes
9472
Ada defines (throughout the Ada reference manual,
9473
summarized in Annex K),
9474
a set of attributes that provide useful additional functionality in all
9475
areas of the language. These language defined attributes are implemented
9476
in GNAT and work as described in the Ada Reference Manual.
9478
In addition, Ada allows implementations to define additional
9479
attributes whose meaning is defined by the implementation. GNAT provides
9480
a number of these implementation-dependent attributes which can be used
9481
to extend and enhance the functionality of the compiler. This section of
9482
the GNAT reference manual describes these additional attributes. It also
9483
describes additional implementation-dependent features of standard
9484
language-defined attributes.
9486
Note that any program using these attributes may not be portable to
9487
other compilers (although GNAT implements this set of attributes on all
9488
platforms). Therefore if portability to other compilers is an important
9489
consideration, you should minimize the use of these attributes.
9492
* Attribute Abort_Signal::
9493
* Attribute Address_Size::
9494
* Attribute Asm_Input::
9495
* Attribute Asm_Output::
9496
* Attribute Atomic_Always_Lock_Free::
9498
* Attribute Bit_Position::
9499
* Attribute Code_Address::
9500
* Attribute Compiler_Version::
9501
* Attribute Constrained::
9502
* Attribute Default_Bit_Order::
9503
* Attribute Default_Scalar_Storage_Order::
9505
* Attribute Descriptor_Size::
9506
* Attribute Elaborated::
9507
* Attribute Elab_Body::
9508
* Attribute Elab_Spec::
9509
* Attribute Elab_Subp_Body::
9511
* Attribute Enabled::
9512
* Attribute Enum_Rep::
9513
* Attribute Enum_Val::
9514
* Attribute Epsilon::
9515
* Attribute Fast_Math::
9516
* Attribute Fixed_Value::
9517
* Attribute From_Any::
9518
* Attribute Has_Access_Values::
9519
* Attribute Has_Discriminants::
9521
* Attribute Integer_Value::
9522
* Attribute Invalid_Value::
9523
* Attribute Iterable::
9525
* Attribute Library_Level::
9526
* Attribute Lock_Free::
9527
* Attribute Loop_Entry::
9528
* Attribute Machine_Size::
9529
* Attribute Mantissa::
9530
* Attribute Maximum_Alignment::
9531
* Attribute Mechanism_Code::
9532
* Attribute Null_Parameter::
9533
* Attribute Object_Size::
9535
* Attribute Passed_By_Reference::
9536
* Attribute Pool_Address::
9537
* Attribute Range_Length::
9538
* Attribute Restriction_Set::
9539
* Attribute Result::
9540
* Attribute Safe_Emax::
9541
* Attribute Safe_Large::
9542
* Attribute Safe_Small::
9543
* Attribute Scalar_Storage_Order::
9544
* Attribute Simple_Storage_Pool::
9546
* Attribute Storage_Unit::
9547
* Attribute Stub_Type::
9548
* Attribute System_Allocator_Alignment::
9549
* Attribute Target_Name::
9550
* Attribute To_Address::
9551
* Attribute To_Any::
9552
* Attribute Type_Class::
9553
* Attribute Type_Key::
9554
* Attribute TypeCode::
9555
* Attribute Unconstrained_Array::
9556
* Attribute Universal_Literal_String::
9557
* Attribute Unrestricted_Access::
9558
* Attribute Update::
9559
* Attribute Valid_Scalars::
9560
* Attribute VADS_Size::
9561
* Attribute Value_Size::
9562
* Attribute Wchar_T_Size::
9563
* Attribute Word_Size::
9567
@node Attribute Abort_Signal,Attribute Address_Size,,Implementation Defined Attributes
9568
@anchor{gnat_rm/implementation_defined_attributes attribute-abort-signal}@anchor{122}
9569
@section Attribute Abort_Signal
9572
@geindex Abort_Signal
9574
@cite{Standard'Abort_Signal} (@cite{Standard} is the only allowed
9575
prefix) provides the entity for the special exception used to signal
9576
task abort or asynchronous transfer of control. Normally this attribute
9577
should only be used in the tasking runtime (it is highly peculiar, and
9578
completely outside the normal semantics of Ada, for a user program to
9579
intercept the abort exception).
9581
@node Attribute Address_Size,Attribute Asm_Input,Attribute Abort_Signal,Implementation Defined Attributes
9582
@anchor{gnat_rm/implementation_defined_attributes attribute-address-size}@anchor{123}
9583
@section Attribute Address_Size
9586
@geindex Size of `Address`
9588
@geindex Address_Size
9590
@cite{Standard'Address_Size} (@cite{Standard} is the only allowed
9591
prefix) is a static constant giving the number of bits in an
9592
@cite{Address}. It is the same value as System.Address'Size,
9593
but has the advantage of being static, while a direct
9594
reference to System.Address'Size is nonstatic because Address
9597
@node Attribute Asm_Input,Attribute Asm_Output,Attribute Address_Size,Implementation Defined Attributes
9598
@anchor{gnat_rm/implementation_defined_attributes attribute-asm-input}@anchor{124}
9599
@section Attribute Asm_Input
9604
The @cite{Asm_Input} attribute denotes a function that takes two
9605
parameters. The first is a string, the second is an expression of the
9606
type designated by the prefix. The first (string) argument is required
9607
to be a static expression, and is the constraint for the parameter,
9608
(e.g., what kind of register is required). The second argument is the
9609
value to be used as the input argument. The possible values for the
9610
constant are the same as those used in the RTL, and are dependent on
9611
the configuration file used to built the GCC back end.
9612
@ref{125,,Machine Code Insertions}
9614
@node Attribute Asm_Output,Attribute Atomic_Always_Lock_Free,Attribute Asm_Input,Implementation Defined Attributes
9615
@anchor{gnat_rm/implementation_defined_attributes attribute-asm-output}@anchor{126}
9616
@section Attribute Asm_Output
9621
The @cite{Asm_Output} attribute denotes a function that takes two
9622
parameters. The first is a string, the second is the name of a variable
9623
of the type designated by the attribute prefix. The first (string)
9624
argument is required to be a static expression and designates the
9625
constraint for the parameter (e.g., what kind of register is
9626
required). The second argument is the variable to be updated with the
9627
result. The possible values for constraint are the same as those used in
9628
the RTL, and are dependent on the configuration file used to build the
9629
GCC back end. If there are no output operands, then this argument may
9630
either be omitted, or explicitly given as @cite{No_Output_Operands}.
9631
@ref{125,,Machine Code Insertions}
9633
@node Attribute Atomic_Always_Lock_Free,Attribute Bit,Attribute Asm_Output,Implementation Defined Attributes
9634
@anchor{gnat_rm/implementation_defined_attributes attribute-atomic-always-lock-free}@anchor{127}
9635
@section Attribute Atomic_Always_Lock_Free
9638
@geindex Atomic_Always_Lock_Free
9640
The prefix of the @cite{Atomic_Always_Lock_Free} attribute is a type.
9641
The result is a Boolean value which is True if the type has discriminants,
9642
and False otherwise. The result indicate whether atomic operations are
9643
supported by the target for the given type.
9645
@node Attribute Bit,Attribute Bit_Position,Attribute Atomic_Always_Lock_Free,Implementation Defined Attributes
9646
@anchor{gnat_rm/implementation_defined_attributes attribute-bit}@anchor{128}
9647
@section Attribute Bit
9652
@code{obj'Bit}, where @cite{obj} is any object, yields the bit
9653
offset within the storage unit (byte) that contains the first bit of
9654
storage allocated for the object. The value of this attribute is of the
9655
type @cite{Universal_Integer}, and is always a non-negative number not
9656
exceeding the value of @cite{System.Storage_Unit}.
9658
For an object that is a variable or a constant allocated in a register,
9659
the value is zero. (The use of this attribute does not force the
9660
allocation of a variable to memory).
9662
For an object that is a formal parameter, this attribute applies
9663
to either the matching actual parameter or to a copy of the
9664
matching actual parameter.
9666
For an access object the value is zero. Note that
9667
@code{obj.all'Bit} is subject to an @cite{Access_Check} for the
9668
designated object. Similarly for a record component
9669
@code{X.C'Bit} is subject to a discriminant check and
9670
@code{X(I).Bit} and @code{X(I1..I2)'Bit}
9671
are subject to index checks.
9673
This attribute is designed to be compatible with the DEC Ada 83 definition
9674
and implementation of the @cite{Bit} attribute.
9676
@node Attribute Bit_Position,Attribute Code_Address,Attribute Bit,Implementation Defined Attributes
9677
@anchor{gnat_rm/implementation_defined_attributes attribute-bit-position}@anchor{129}
9678
@section Attribute Bit_Position
9681
@geindex Bit_Position
9683
@code{R.C'Bit_Position}, where @cite{R} is a record object and @cite{C} is one
9684
of the fields of the record type, yields the bit
9685
offset within the record contains the first bit of
9686
storage allocated for the object. The value of this attribute is of the
9687
type @cite{Universal_Integer}. The value depends only on the field
9688
@cite{C} and is independent of the alignment of
9689
the containing record @cite{R}.
9691
@node Attribute Code_Address,Attribute Compiler_Version,Attribute Bit_Position,Implementation Defined Attributes
9692
@anchor{gnat_rm/implementation_defined_attributes attribute-code-address}@anchor{12a}
9693
@section Attribute Code_Address
9696
@geindex Code_Address
9698
@geindex Subprogram address
9700
@geindex Address of subprogram code
9703
attribute may be applied to subprograms in Ada 95 and Ada 2005, but the
9704
intended effect seems to be to provide
9705
an address value which can be used to call the subprogram by means of
9706
an address clause as in the following example:
9712
for L'Address use K'Address;
9713
pragma Import (Ada, L);
9716
A call to @cite{L} is then expected to result in a call to @cite{K}.
9717
In Ada 83, where there were no access-to-subprogram values, this was
9718
a common work-around for getting the effect of an indirect call.
9719
GNAT implements the above use of @cite{Address} and the technique
9720
illustrated by the example code works correctly.
9722
However, for some purposes, it is useful to have the address of the start
9723
of the generated code for the subprogram. On some architectures, this is
9724
not necessarily the same as the @cite{Address} value described above.
9725
For example, the @cite{Address} value may reference a subprogram
9726
descriptor rather than the subprogram itself.
9728
The @cite{'Code_Address} attribute, which can only be applied to
9729
subprogram entities, always returns the address of the start of the
9730
generated code of the specified subprogram, which may or may not be
9731
the same value as is returned by the corresponding @cite{'Address}
9734
@node Attribute Compiler_Version,Attribute Constrained,Attribute Code_Address,Implementation Defined Attributes
9735
@anchor{gnat_rm/implementation_defined_attributes attribute-compiler-version}@anchor{12b}
9736
@section Attribute Compiler_Version
9739
@geindex Compiler_Version
9741
@cite{Standard'Compiler_Version} (@cite{Standard} is the only allowed
9742
prefix) yields a static string identifying the version of the compiler
9743
being used to compile the unit containing the attribute reference.
9745
@node Attribute Constrained,Attribute Default_Bit_Order,Attribute Compiler_Version,Implementation Defined Attributes
9746
@anchor{gnat_rm/implementation_defined_attributes attribute-constrained}@anchor{12c}
9747
@section Attribute Constrained
9750
@geindex Constrained
9752
In addition to the usage of this attribute in the Ada RM, @cite{GNAT}
9753
also permits the use of the @cite{'Constrained} attribute
9754
in a generic template
9755
for any type, including types without discriminants. The value of this
9756
attribute in the generic instance when applied to a scalar type or a
9757
record type without discriminants is always @cite{True}. This usage is
9758
compatible with older Ada compilers, including notably DEC Ada.
9760
@node Attribute Default_Bit_Order,Attribute Default_Scalar_Storage_Order,Attribute Constrained,Implementation Defined Attributes
9761
@anchor{gnat_rm/implementation_defined_attributes attribute-default-bit-order}@anchor{12d}
9762
@section Attribute Default_Bit_Order
9767
@geindex Little endian
9769
@geindex Default_Bit_Order
9771
@cite{Standard'Default_Bit_Order} (@cite{Standard} is the only
9772
permissible prefix), provides the value @cite{System.Default_Bit_Order}
9773
as a @cite{Pos} value (0 for @cite{High_Order_First}, 1 for
9774
@cite{Low_Order_First}). This is used to construct the definition of
9775
@cite{Default_Bit_Order} in package @cite{System}.
9777
@node Attribute Default_Scalar_Storage_Order,Attribute Deref,Attribute Default_Bit_Order,Implementation Defined Attributes
9778
@anchor{gnat_rm/implementation_defined_attributes attribute-default-scalar-storage-order}@anchor{12e}
9779
@section Attribute Default_Scalar_Storage_Order
9784
@geindex Little endian
9786
@geindex Default_Scalar_Storage_Order
9788
@cite{Standard'Default_Scalar_Storage_Order} (@cite{Standard} is the only
9789
permissible prefix), provides the current value of the default scalar storage
9790
order (as specified using pragma @cite{Default_Scalar_Storage_Order}, or
9791
equal to @cite{Default_Bit_Order} if unspecified) as a
9792
@cite{System.Bit_Order} value. This is a static attribute.
9794
@node Attribute Deref,Attribute Descriptor_Size,Attribute Default_Scalar_Storage_Order,Implementation Defined Attributes
9795
@anchor{gnat_rm/implementation_defined_attributes attribute-deref}@anchor{12f}
9796
@section Attribute Deref
9801
The attribute @cite{typ'Deref(expr)} where @cite{expr} is of type @cite{System.Address} yields
9802
the variable of type @cite{typ} that is located at the given address. It is similar
9803
to @cite{(totyp (expr).all)}, where @cite{totyp} is an unchecked conversion from address to
9804
a named access-to-@cite{typ} type, except that it yields a variable, so it can be
9805
used on the left side of an assignment.
9807
@node Attribute Descriptor_Size,Attribute Elaborated,Attribute Deref,Implementation Defined Attributes
9808
@anchor{gnat_rm/implementation_defined_attributes attribute-descriptor-size}@anchor{130}
9809
@section Attribute Descriptor_Size
9814
@geindex Dope vector
9816
@geindex Descriptor_Size
9818
Nonstatic attribute @cite{Descriptor_Size} returns the size in bits of the
9819
descriptor allocated for a type. The result is non-zero only for unconstrained
9820
array types and the returned value is of type universal integer. In GNAT, an
9821
array descriptor contains bounds information and is located immediately before
9822
the first element of the array.
9825
type Unconstr_Array is array (Positive range <>) of Boolean;
9826
Put_Line ("Descriptor size = " & Unconstr_Array'Descriptor_Size'Img);
9829
The attribute takes into account any additional padding due to type alignment.
9830
In the example above, the descriptor contains two values of type
9831
@cite{Positive} representing the low and high bound. Since @cite{Positive} has
9832
a size of 31 bits and an alignment of 4, the descriptor size is @cite{2 * Positive'Size + 2} or 64 bits.
9834
@node Attribute Elaborated,Attribute Elab_Body,Attribute Descriptor_Size,Implementation Defined Attributes
9835
@anchor{gnat_rm/implementation_defined_attributes attribute-elaborated}@anchor{131}
9836
@section Attribute Elaborated
9841
The prefix of the @cite{'Elaborated} attribute must be a unit name. The
9842
value is a Boolean which indicates whether or not the given unit has been
9843
elaborated. This attribute is primarily intended for internal use by the
9844
generated code for dynamic elaboration checking, but it can also be used
9845
in user programs. The value will always be True once elaboration of all
9846
units has been completed. An exception is for units which need no
9847
elaboration, the value is always False for such units.
9849
@node Attribute Elab_Body,Attribute Elab_Spec,Attribute Elaborated,Implementation Defined Attributes
9850
@anchor{gnat_rm/implementation_defined_attributes attribute-elab-body}@anchor{132}
9851
@section Attribute Elab_Body
9856
This attribute can only be applied to a program unit name. It returns
9857
the entity for the corresponding elaboration procedure for elaborating
9858
the body of the referenced unit. This is used in the main generated
9859
elaboration procedure by the binder and is not normally used in any
9860
other context. However, there may be specialized situations in which it
9861
is useful to be able to call this elaboration procedure from Ada code,
9862
e.g., if it is necessary to do selective re-elaboration to fix some
9865
@node Attribute Elab_Spec,Attribute Elab_Subp_Body,Attribute Elab_Body,Implementation Defined Attributes
9866
@anchor{gnat_rm/implementation_defined_attributes attribute-elab-spec}@anchor{133}
9867
@section Attribute Elab_Spec
9872
This attribute can only be applied to a program unit name. It returns
9873
the entity for the corresponding elaboration procedure for elaborating
9874
the spec of the referenced unit. This is used in the main
9875
generated elaboration procedure by the binder and is not normally used
9876
in any other context. However, there may be specialized situations in
9877
which it is useful to be able to call this elaboration procedure from
9878
Ada code, e.g., if it is necessary to do selective re-elaboration to fix
9881
@node Attribute Elab_Subp_Body,Attribute Emax,Attribute Elab_Spec,Implementation Defined Attributes
9882
@anchor{gnat_rm/implementation_defined_attributes attribute-elab-subp-body}@anchor{134}
9883
@section Attribute Elab_Subp_Body
9886
@geindex Elab_Subp_Body
9888
This attribute can only be applied to a library level subprogram
9889
name and is only allowed in CodePeer mode. It returns the entity
9890
for the corresponding elaboration procedure for elaborating the body
9891
of the referenced subprogram unit. This is used in the main generated
9892
elaboration procedure by the binder in CodePeer mode only and is unrecognized
9895
@node Attribute Emax,Attribute Enabled,Attribute Elab_Subp_Body,Implementation Defined Attributes
9896
@anchor{gnat_rm/implementation_defined_attributes attribute-emax}@anchor{135}
9897
@section Attribute Emax
9900
@geindex Ada 83 attributes
9904
The @cite{Emax} attribute is provided for compatibility with Ada 83. See
9905
the Ada 83 reference manual for an exact description of the semantics of
9908
@node Attribute Enabled,Attribute Enum_Rep,Attribute Emax,Implementation Defined Attributes
9909
@anchor{gnat_rm/implementation_defined_attributes attribute-enabled}@anchor{136}
9910
@section Attribute Enabled
9915
The @cite{Enabled} attribute allows an application program to check at compile
9916
time to see if the designated check is currently enabled. The prefix is a
9917
simple identifier, referencing any predefined check name (other than
9918
@cite{All_Checks}) or a check name introduced by pragma Check_Name. If
9919
no argument is given for the attribute, the check is for the general state
9920
of the check, if an argument is given, then it is an entity name, and the
9921
check indicates whether an @cite{Suppress} or @cite{Unsuppress} has been
9922
given naming the entity (if not, then the argument is ignored).
9924
Note that instantiations inherit the check status at the point of the
9925
instantiation, so a useful idiom is to have a library package that
9926
introduces a check name with @cite{pragma Check_Name}, and then contains
9927
generic packages or subprograms which use the @cite{Enabled} attribute
9928
to see if the check is enabled. A user of this package can then issue
9929
a @cite{pragma Suppress} or @cite{pragma Unsuppress} before instantiating
9930
the package or subprogram, controlling whether the check will be present.
9932
@node Attribute Enum_Rep,Attribute Enum_Val,Attribute Enabled,Implementation Defined Attributes
9933
@anchor{gnat_rm/implementation_defined_attributes attribute-enum-rep}@anchor{137}
9934
@section Attribute Enum_Rep
9937
@geindex Representation of enums
9941
For every enumeration subtype @cite{S}, @code{S'Enum_Rep} denotes a
9942
function with the following spec:
9945
function S'Enum_Rep (Arg : S'Base) return <Universal_Integer>;
9948
It is also allowable to apply @cite{Enum_Rep} directly to an object of an
9949
enumeration type or to a non-overloaded enumeration
9950
literal. In this case @code{S'Enum_Rep} is equivalent to
9951
@code{typ'Enum_Rep(S)} where @cite{typ} is the type of the
9952
enumeration literal or object.
9954
The function returns the representation value for the given enumeration
9955
value. This will be equal to value of the @cite{Pos} attribute in the
9956
absence of an enumeration representation clause. This is a static
9957
attribute (i.e.,:the result is static if the argument is static).
9959
@code{S'Enum_Rep} can also be used with integer types and objects,
9960
in which case it simply returns the integer value. The reason for this
9961
is to allow it to be used for @cite{(<>)} discrete formal arguments in
9962
a generic unit that can be instantiated with either enumeration types
9963
or integer types. Note that if @cite{Enum_Rep} is used on a modular
9964
type whose upper bound exceeds the upper bound of the largest signed
9965
integer type, and the argument is a variable, so that the universal
9966
integer calculation is done at run time, then the call to @cite{Enum_Rep}
9967
may raise @cite{Constraint_Error}.
9969
@node Attribute Enum_Val,Attribute Epsilon,Attribute Enum_Rep,Implementation Defined Attributes
9970
@anchor{gnat_rm/implementation_defined_attributes attribute-enum-val}@anchor{138}
9971
@section Attribute Enum_Val
9974
@geindex Representation of enums
9978
For every enumeration subtype @cite{S}, @code{S'Enum_Val} denotes a
9979
function with the following spec:
9982
function S'Enum_Val (Arg : <Universal_Integer>) return S'Base;
9985
The function returns the enumeration value whose representation matches the
9986
argument, or raises Constraint_Error if no enumeration literal of the type
9987
has the matching value.
9988
This will be equal to value of the @cite{Val} attribute in the
9989
absence of an enumeration representation clause. This is a static
9990
attribute (i.e., the result is static if the argument is static).
9992
@node Attribute Epsilon,Attribute Fast_Math,Attribute Enum_Val,Implementation Defined Attributes
9993
@anchor{gnat_rm/implementation_defined_attributes attribute-epsilon}@anchor{139}
9994
@section Attribute Epsilon
9997
@geindex Ada 83 attributes
10001
The @cite{Epsilon} attribute is provided for compatibility with Ada 83. See
10002
the Ada 83 reference manual for an exact description of the semantics of
10005
@node Attribute Fast_Math,Attribute Fixed_Value,Attribute Epsilon,Implementation Defined Attributes
10006
@anchor{gnat_rm/implementation_defined_attributes attribute-fast-math}@anchor{13a}
10007
@section Attribute Fast_Math
10012
@cite{Standard'Fast_Math} (@cite{Standard} is the only allowed
10013
prefix) yields a static Boolean value that is True if pragma
10014
@cite{Fast_Math} is active, and False otherwise.
10016
@node Attribute Fixed_Value,Attribute From_Any,Attribute Fast_Math,Implementation Defined Attributes
10017
@anchor{gnat_rm/implementation_defined_attributes attribute-fixed-value}@anchor{13b}
10018
@section Attribute Fixed_Value
10021
@geindex Fixed_Value
10023
For every fixed-point type @cite{S}, @code{S'Fixed_Value} denotes a
10024
function with the following specification:
10027
function S'Fixed_Value (Arg : <Universal_Integer>) return S;
10030
The value returned is the fixed-point value @cite{V} such that:
10036
The effect is thus similar to first converting the argument to the
10037
integer type used to represent @cite{S}, and then doing an unchecked
10038
conversion to the fixed-point type. The difference is
10039
that there are full range checks, to ensure that the result is in range.
10040
This attribute is primarily intended for use in implementation of the
10041
input-output functions for fixed-point values.
10043
@node Attribute From_Any,Attribute Has_Access_Values,Attribute Fixed_Value,Implementation Defined Attributes
10044
@anchor{gnat_rm/implementation_defined_attributes attribute-from-any}@anchor{13c}
10045
@section Attribute From_Any
10050
This internal attribute is used for the generation of remote subprogram
10051
stubs in the context of the Distributed Systems Annex.
10053
@node Attribute Has_Access_Values,Attribute Has_Discriminants,Attribute From_Any,Implementation Defined Attributes
10054
@anchor{gnat_rm/implementation_defined_attributes attribute-has-access-values}@anchor{13d}
10055
@section Attribute Has_Access_Values
10058
@geindex Access values
10059
@geindex testing for
10061
@geindex Has_Access_Values
10063
The prefix of the @cite{Has_Access_Values} attribute is a type. The result
10064
is a Boolean value which is True if the is an access type, or is a composite
10065
type with a component (at any nesting depth) that is an access type, and is
10067
The intended use of this attribute is in conjunction with generic
10068
definitions. If the attribute is applied to a generic private type, it
10069
indicates whether or not the corresponding actual type has access values.
10071
@node Attribute Has_Discriminants,Attribute Img,Attribute Has_Access_Values,Implementation Defined Attributes
10072
@anchor{gnat_rm/implementation_defined_attributes attribute-has-discriminants}@anchor{13e}
10073
@section Attribute Has_Discriminants
10076
@geindex Discriminants
10077
@geindex testing for
10079
@geindex Has_Discriminants
10081
The prefix of the @cite{Has_Discriminants} attribute is a type. The result
10082
is a Boolean value which is True if the type has discriminants, and False
10083
otherwise. The intended use of this attribute is in conjunction with generic
10084
definitions. If the attribute is applied to a generic private type, it
10085
indicates whether or not the corresponding actual type has discriminants.
10087
@node Attribute Img,Attribute Integer_Value,Attribute Has_Discriminants,Implementation Defined Attributes
10088
@anchor{gnat_rm/implementation_defined_attributes attribute-img}@anchor{13f}
10089
@section Attribute Img
10094
The @cite{Img} attribute differs from @cite{Image} in that it is applied
10095
directly to an object, and yields the same result as
10096
@cite{Image} for the subtype of the object. This is convenient for
10100
Put_Line ("X = " & X'Img);
10103
has the same meaning as the more verbose:
10106
Put_Line ("X = " & T'Image (X));
10109
where @cite{T} is the (sub)type of the object @cite{X}.
10111
Note that technically, in analogy to @cite{Image},
10112
@cite{X'Img} returns a parameterless function
10113
that returns the appropriate string when called. This means that
10114
@cite{X'Img} can be renamed as a function-returning-string, or used
10115
in an instantiation as a function parameter.
10117
@node Attribute Integer_Value,Attribute Invalid_Value,Attribute Img,Implementation Defined Attributes
10118
@anchor{gnat_rm/implementation_defined_attributes attribute-integer-value}@anchor{140}
10119
@section Attribute Integer_Value
10122
@geindex Integer_Value
10124
For every integer type @cite{S}, @code{S'Integer_Value} denotes a
10125
function with the following spec:
10128
function S'Integer_Value (Arg : <Universal_Fixed>) return S;
10131
The value returned is the integer value @cite{V}, such that:
10137
where @cite{T} is the type of @cite{Arg}.
10138
The effect is thus similar to first doing an unchecked conversion from
10139
the fixed-point type to its corresponding implementation type, and then
10140
converting the result to the target integer type. The difference is
10141
that there are full range checks, to ensure that the result is in range.
10142
This attribute is primarily intended for use in implementation of the
10143
standard input-output functions for fixed-point values.
10145
@node Attribute Invalid_Value,Attribute Iterable,Attribute Integer_Value,Implementation Defined Attributes
10146
@anchor{gnat_rm/implementation_defined_attributes attribute-invalid-value}@anchor{141}
10147
@section Attribute Invalid_Value
10150
@geindex Invalid_Value
10152
For every scalar type S, S'Invalid_Value returns an undefined value of the
10153
type. If possible this value is an invalid representation for the type. The
10154
value returned is identical to the value used to initialize an otherwise
10155
uninitialized value of the type if pragma Initialize_Scalars is used,
10156
including the ability to modify the value with the binder -Sxx flag and
10157
relevant environment variables at run time.
10159
@node Attribute Iterable,Attribute Large,Attribute Invalid_Value,Implementation Defined Attributes
10160
@anchor{gnat_rm/implementation_defined_attributes attribute-iterable}@anchor{142}
10161
@section Attribute Iterable
10166
Equivalent to Aspect Iterable.
10168
@node Attribute Large,Attribute Library_Level,Attribute Iterable,Implementation Defined Attributes
10169
@anchor{gnat_rm/implementation_defined_attributes attribute-large}@anchor{143}
10170
@section Attribute Large
10173
@geindex Ada 83 attributes
10177
The @cite{Large} attribute is provided for compatibility with Ada 83. See
10178
the Ada 83 reference manual for an exact description of the semantics of
10181
@node Attribute Library_Level,Attribute Lock_Free,Attribute Large,Implementation Defined Attributes
10182
@anchor{gnat_rm/implementation_defined_attributes attribute-library-level}@anchor{144}
10183
@section Attribute Library_Level
10186
@geindex Library_Level
10188
@cite{P'Library_Level}, where P is an entity name,
10189
returns a Boolean value which is True if the entity is declared
10190
at the library level, and False otherwise. Note that within a
10191
generic instantition, the name of the generic unit denotes the
10192
instance, which means that this attribute can be used to test
10193
if a generic is instantiated at the library level, as shown
10200
pragma Compile_Time_Error
10201
(not Gen'Library_Level,
10202
"Gen can only be instantiated at library level");
10207
@node Attribute Lock_Free,Attribute Loop_Entry,Attribute Library_Level,Implementation Defined Attributes
10208
@anchor{gnat_rm/implementation_defined_attributes attribute-lock-free}@anchor{145}
10209
@section Attribute Lock_Free
10214
@cite{P'Lock_Free}, where P is a protected object, returns True if a
10215
pragma @cite{Lock_Free} applies to P.
10217
@node Attribute Loop_Entry,Attribute Machine_Size,Attribute Lock_Free,Implementation Defined Attributes
10218
@anchor{gnat_rm/implementation_defined_attributes attribute-loop-entry}@anchor{146}
10219
@section Attribute Loop_Entry
10222
@geindex Loop_Entry
10227
X'Loop_Entry [(loop_name)]
10230
The @cite{Loop_Entry} attribute is used to refer to the value that an
10231
expression had upon entry to a given loop in much the same way that the
10232
@cite{Old} attribute in a subprogram postcondition can be used to refer
10233
to the value an expression had upon entry to the subprogram. The
10234
relevant loop is either identified by the given loop name, or it is the
10235
innermost enclosing loop when no loop name is given.
10237
A @cite{Loop_Entry} attribute can only occur within a
10238
@cite{Loop_Variant} or @cite{Loop_Invariant} pragma. A common use of
10239
@cite{Loop_Entry} is to compare the current value of objects with their
10240
initial value at loop entry, in a @cite{Loop_Invariant} pragma.
10242
The effect of using @cite{X'Loop_Entry} is the same as declaring
10243
a constant initialized with the initial value of @cite{X} at loop
10244
entry. This copy is not performed if the loop is not entered, or if the
10245
corresponding pragmas are ignored or disabled.
10247
@node Attribute Machine_Size,Attribute Mantissa,Attribute Loop_Entry,Implementation Defined Attributes
10248
@anchor{gnat_rm/implementation_defined_attributes attribute-machine-size}@anchor{147}
10249
@section Attribute Machine_Size
10252
@geindex Machine_Size
10254
This attribute is identical to the @cite{Object_Size} attribute. It is
10255
provided for compatibility with the DEC Ada 83 attribute of this name.
10257
@node Attribute Mantissa,Attribute Maximum_Alignment,Attribute Machine_Size,Implementation Defined Attributes
10258
@anchor{gnat_rm/implementation_defined_attributes attribute-mantissa}@anchor{148}
10259
@section Attribute Mantissa
10262
@geindex Ada 83 attributes
10266
The @cite{Mantissa} attribute is provided for compatibility with Ada 83. See
10267
the Ada 83 reference manual for an exact description of the semantics of
10270
@node Attribute Maximum_Alignment,Attribute Mechanism_Code,Attribute Mantissa,Implementation Defined Attributes
10271
@anchor{gnat_rm/implementation_defined_attributes attribute-maximum-alignment}@anchor{149}@anchor{gnat_rm/implementation_defined_attributes id2}@anchor{14a}
10272
@section Attribute Maximum_Alignment
10278
@geindex Maximum_Alignment
10280
@cite{Standard'Maximum_Alignment} (@cite{Standard} is the only
10281
permissible prefix) provides the maximum useful alignment value for the
10282
target. This is a static value that can be used to specify the alignment
10283
for an object, guaranteeing that it is properly aligned in all
10286
@node Attribute Mechanism_Code,Attribute Null_Parameter,Attribute Maximum_Alignment,Implementation Defined Attributes
10287
@anchor{gnat_rm/implementation_defined_attributes attribute-mechanism-code}@anchor{14b}
10288
@section Attribute Mechanism_Code
10291
@geindex Return values
10292
@geindex passing mechanism
10294
@geindex Parameters
10295
@geindex passing mechanism
10297
@geindex Mechanism_Code
10299
@code{function'Mechanism_Code} yields an integer code for the
10300
mechanism used for the result of function, and
10301
@code{subprogram'Mechanism_Code (n)} yields the mechanism
10302
used for formal parameter number @cite{n} (a static integer value with 1
10303
meaning the first parameter) of @cite{subprogram}. The code returned is:
10317
@node Attribute Null_Parameter,Attribute Object_Size,Attribute Mechanism_Code,Implementation Defined Attributes
10318
@anchor{gnat_rm/implementation_defined_attributes attribute-null-parameter}@anchor{14c}
10319
@section Attribute Null_Parameter
10322
@geindex Zero address
10325
@geindex Null_Parameter
10327
A reference @code{T'Null_Parameter} denotes an imaginary object of
10328
type or subtype @cite{T} allocated at machine address zero. The attribute
10329
is allowed only as the default expression of a formal parameter, or as
10330
an actual expression of a subprogram call. In either case, the
10331
subprogram must be imported.
10333
The identity of the object is represented by the address zero in the
10334
argument list, independent of the passing mechanism (explicit or
10337
This capability is needed to specify that a zero address should be
10338
passed for a record or other composite object passed by reference.
10339
There is no way of indicating this without the @cite{Null_Parameter}
10342
@node Attribute Object_Size,Attribute Old,Attribute Null_Parameter,Implementation Defined Attributes
10343
@anchor{gnat_rm/implementation_defined_attributes attribute-object-size}@anchor{14d}
10344
@section Attribute Object_Size
10348
@geindex used for objects
10350
@geindex Object_Size
10352
The size of an object is not necessarily the same as the size of the type
10353
of an object. This is because by default object sizes are increased to be
10354
a multiple of the alignment of the object. For example,
10355
@cite{Natural'Size} is
10356
31, but by default objects of type @cite{Natural} will have a size of 32 bits.
10357
Similarly, a record containing an integer and a character:
10366
will have a size of 40 (that is @cite{Rec'Size} will be 40). The
10367
alignment will be 4, because of the
10368
integer field, and so the default size of record objects for this type
10369
will be 64 (8 bytes).
10371
If the alignment of the above record is specified to be 1, then the
10372
object size will be 40 (5 bytes). This is true by default, and also
10373
an object size of 40 can be explicitly specified in this case.
10375
A consequence of this capability is that different object sizes can be
10376
given to subtypes that would otherwise be considered in Ada to be
10377
statically matching. But it makes no sense to consider such subtypes
10378
as statically matching. Consequently, in @cite{GNAT} we add a rule
10379
to the static matching rules that requires object sizes to match.
10380
Consider this example:
10383
1. procedure BadAVConvert is
10384
2. type R is new Integer;
10385
3. subtype R1 is R range 1 .. 10;
10386
4. subtype R2 is R range 1 .. 10;
10387
5. for R1'Object_Size use 8;
10388
6. for R2'Object_Size use 16;
10389
7. type R1P is access all R1;
10390
8. type R2P is access all R2;
10391
9. R1PV : R1P := new R1'(4);
10394
12. R2PV := R2P (R1PV);
10396
>>> target designated subtype not compatible with
10397
type "R1" defined at line 3
10402
In the absence of lines 5 and 6,
10403
types @cite{R1} and @cite{R2} statically match and
10404
hence the conversion on line 12 is legal. But since lines 5 and 6
10405
cause the object sizes to differ, @cite{GNAT} considers that types
10406
@cite{R1} and @cite{R2} are not statically matching, and line 12
10407
generates the diagnostic shown above.
10409
Similar additional checks are performed in other contexts requiring
10410
statically matching subtypes.
10412
@node Attribute Old,Attribute Passed_By_Reference,Attribute Object_Size,Implementation Defined Attributes
10413
@anchor{gnat_rm/implementation_defined_attributes attribute-old}@anchor{14e}
10414
@section Attribute Old
10419
In addition to the usage of @cite{Old} defined in the Ada 2012 RM (usage
10420
within @cite{Post} aspect), GNAT also permits the use of this attribute
10421
in implementation defined pragmas @cite{Postcondition},
10422
@cite{Contract_Cases} and @cite{Test_Case}. Also usages of
10423
@cite{Old} which would be illegal according to the Ada 2012 RM
10424
definition are allowed under control of
10425
implementation defined pragma @cite{Unevaluated_Use_Of_Old}.
10427
@node Attribute Passed_By_Reference,Attribute Pool_Address,Attribute Old,Implementation Defined Attributes
10428
@anchor{gnat_rm/implementation_defined_attributes attribute-passed-by-reference}@anchor{14f}
10429
@section Attribute Passed_By_Reference
10432
@geindex Parameters
10433
@geindex when passed by reference
10435
@geindex Passed_By_Reference
10437
@code{type'Passed_By_Reference} for any subtype @cite{type} returns
10438
a value of type @cite{Boolean} value that is @cite{True} if the type is
10439
normally passed by reference and @cite{False} if the type is normally
10440
passed by copy in calls. For scalar types, the result is always @cite{False}
10441
and is static. For non-scalar types, the result is nonstatic.
10443
@node Attribute Pool_Address,Attribute Range_Length,Attribute Passed_By_Reference,Implementation Defined Attributes
10444
@anchor{gnat_rm/implementation_defined_attributes attribute-pool-address}@anchor{150}
10445
@section Attribute Pool_Address
10448
@geindex Parameters
10449
@geindex when passed by reference
10451
@geindex Pool_Address
10453
@code{X'Pool_Address} for any object @cite{X} returns the address
10454
of X within its storage pool. This is the same as
10455
@code{X'Address}, except that for an unconstrained array whose
10456
bounds are allocated just before the first component,
10457
@code{X'Pool_Address} returns the address of those bounds,
10458
whereas @code{X'Address} returns the address of the first
10461
Here, we are interpreting 'storage pool' broadly to mean
10462
@code{wherever the object is allocated}, which could be a
10463
user-defined storage pool,
10464
the global heap, on the stack, or in a static memory area.
10465
For an object created by @cite{new}, @code{Ptr.all'Pool_Address} is
10466
what is passed to @cite{Allocate} and returned from @cite{Deallocate}.
10468
@node Attribute Range_Length,Attribute Restriction_Set,Attribute Pool_Address,Implementation Defined Attributes
10469
@anchor{gnat_rm/implementation_defined_attributes attribute-range-length}@anchor{151}
10470
@section Attribute Range_Length
10473
@geindex Range_Length
10475
@code{type'Range_Length} for any discrete type @cite{type} yields
10476
the number of values represented by the subtype (zero for a null
10477
range). The result is static for static subtypes. @cite{Range_Length}
10478
applied to the index subtype of a one dimensional array always gives the
10479
same result as @cite{Length} applied to the array itself.
10481
@node Attribute Restriction_Set,Attribute Result,Attribute Range_Length,Implementation Defined Attributes
10482
@anchor{gnat_rm/implementation_defined_attributes attribute-restriction-set}@anchor{152}
10483
@section Attribute Restriction_Set
10486
@geindex Restriction_Set
10488
@geindex Restrictions
10490
This attribute allows compile time testing of restrictions that
10491
are currently in effect. It is primarily intended for specializing
10492
code in the run-time based on restrictions that are active (e.g.
10493
don't need to save fpt registers if restriction No_Floating_Point
10494
is known to be in effect), but can be used anywhere.
10496
There are two forms:
10499
System'Restriction_Set (partition_boolean_restriction_NAME)
10500
System'Restriction_Set (No_Dependence => library_unit_NAME);
10503
In the case of the first form, the only restriction names
10504
allowed are parameterless restrictions that are checked
10505
for consistency at bind time. For a complete list see the
10506
subtype @cite{System.Rident.Partition_Boolean_Restrictions}.
10508
The result returned is True if the restriction is known to
10509
be in effect, and False if the restriction is known not to
10510
be in effect. An important guarantee is that the value of
10511
a Restriction_Set attribute is known to be consistent throughout
10512
all the code of a partition.
10514
This is trivially achieved if the entire partition is compiled
10515
with a consistent set of restriction pragmas. However, the
10516
compilation model does not require this. It is possible to
10517
compile one set of units with one set of pragmas, and another
10518
set of units with another set of pragmas. It is even possible
10519
to compile a spec with one set of pragmas, and then WITH the
10520
same spec with a different set of pragmas. Inconsistencies
10521
in the actual use of the restriction are checked at bind time.
10523
In order to achieve the guarantee of consistency for the
10524
Restriction_Set pragma, we consider that a use of the pragma
10525
that yields False is equivalent to a violation of the
10528
So for example if you write
10531
if System'Restriction_Set (No_Floating_Point) then
10538
And the result is False, so that the else branch is executed,
10539
you can assume that this restriction is not set for any unit
10540
in the partition. This is checked by considering this use of
10541
the restriction pragma to be a violation of the restriction
10542
No_Floating_Point. This means that no other unit can attempt
10543
to set this restriction (if some unit does attempt to set it,
10544
the binder will refuse to bind the partition).
10546
Technical note: The restriction name and the unit name are
10547
intepreted entirely syntactically, as in the corresponding
10548
Restrictions pragma, they are not analyzed semantically,
10549
so they do not have a type.
10551
@node Attribute Result,Attribute Safe_Emax,Attribute Restriction_Set,Implementation Defined Attributes
10552
@anchor{gnat_rm/implementation_defined_attributes attribute-result}@anchor{153}
10553
@section Attribute Result
10558
@code{function'Result} can only be used with in a Postcondition pragma
10559
for a function. The prefix must be the name of the corresponding function. This
10560
is used to refer to the result of the function in the postcondition expression.
10561
For a further discussion of the use of this attribute and examples of its use,
10562
see the description of pragma Postcondition.
10564
@node Attribute Safe_Emax,Attribute Safe_Large,Attribute Result,Implementation Defined Attributes
10565
@anchor{gnat_rm/implementation_defined_attributes attribute-safe-emax}@anchor{154}
10566
@section Attribute Safe_Emax
10569
@geindex Ada 83 attributes
10573
The @cite{Safe_Emax} attribute is provided for compatibility with Ada 83. See
10574
the Ada 83 reference manual for an exact description of the semantics of
10577
@node Attribute Safe_Large,Attribute Safe_Small,Attribute Safe_Emax,Implementation Defined Attributes
10578
@anchor{gnat_rm/implementation_defined_attributes attribute-safe-large}@anchor{155}
10579
@section Attribute Safe_Large
10582
@geindex Ada 83 attributes
10584
@geindex Safe_Large
10586
The @cite{Safe_Large} attribute is provided for compatibility with Ada 83. See
10587
the Ada 83 reference manual for an exact description of the semantics of
10590
@node Attribute Safe_Small,Attribute Scalar_Storage_Order,Attribute Safe_Large,Implementation Defined Attributes
10591
@anchor{gnat_rm/implementation_defined_attributes attribute-safe-small}@anchor{156}
10592
@section Attribute Safe_Small
10595
@geindex Ada 83 attributes
10597
@geindex Safe_Small
10599
The @cite{Safe_Small} attribute is provided for compatibility with Ada 83. See
10600
the Ada 83 reference manual for an exact description of the semantics of
10603
@node Attribute Scalar_Storage_Order,Attribute Simple_Storage_Pool,Attribute Safe_Small,Implementation Defined Attributes
10604
@anchor{gnat_rm/implementation_defined_attributes attribute-scalar-storage-order}@anchor{157}
10605
@section Attribute Scalar_Storage_Order
10608
@geindex Endianness
10610
@geindex Scalar storage order
10612
@geindex Scalar_Storage_Order
10614
For every array or record type @cite{S}, the representation attribute
10615
@cite{Scalar_Storage_Order} denotes the order in which storage elements
10616
that make up scalar components are ordered within S. The value given must
10617
be a static expression of type System.Bit_Order. The following is an example
10618
of the use of this feature:
10621
-- Component type definitions
10623
subtype Yr_Type is Natural range 0 .. 127;
10624
subtype Mo_Type is Natural range 1 .. 12;
10625
subtype Da_Type is Natural range 1 .. 31;
10627
-- Record declaration
10629
type Date is record
10630
Years_Since_1980 : Yr_Type;
10632
Day_Of_Month : Da_Type;
10635
-- Record representation clause
10637
for Date use record
10638
Years_Since_1980 at 0 range 0 .. 6;
10639
Month at 0 range 7 .. 10;
10640
Day_Of_Month at 0 range 11 .. 15;
10643
-- Attribute definition clauses
10645
for Date'Bit_Order use System.High_Order_First;
10646
for Date'Scalar_Storage_Order use System.High_Order_First;
10647
-- If Scalar_Storage_Order is specified, it must be consistent with
10648
-- Bit_Order, so it's best to always define the latter explicitly if
10649
-- the former is used.
10652
Other properties are as for standard representation attribute @cite{Bit_Order},
10653
as defined by Ada RM 13.5.3(4). The default is @cite{System.Default_Bit_Order}.
10655
For a record type @cite{T}, if @code{T'Scalar_Storage_Order} is
10656
specified explicitly, it shall be equal to @code{T'Bit_Order}. Note:
10657
this means that if a @cite{Scalar_Storage_Order} attribute definition
10658
clause is not confirming, then the type's @cite{Bit_Order} shall be
10659
specified explicitly and set to the same value.
10661
Derived types inherit an explicitly set scalar storage order from their parent
10662
types. This may be overridden for the derived type by giving an explicit scalar
10663
storage order for the derived type. For a record extension, the derived type
10664
must have the same scalar storage order as the parent type.
10666
If a component of @cite{T} is of a record or array type, then that type must
10667
also have a @cite{Scalar_Storage_Order} attribute definition clause.
10669
A component of a record or array type that is a packed array, or that
10670
does not start on a byte boundary, must have the same scalar storage order
10671
as the enclosing record or array type.
10673
No component of a type that has an explicit @cite{Scalar_Storage_Order}
10674
attribute definition may be aliased.
10676
A confirming @cite{Scalar_Storage_Order} attribute definition clause (i.e.
10677
with a value equal to @cite{System.Default_Bit_Order}) has no effect.
10679
If the opposite storage order is specified, then whenever the value of
10680
a scalar component of an object of type @cite{S} is read, the storage
10681
elements of the enclosing machine scalar are first reversed (before
10682
retrieving the component value, possibly applying some shift and mask
10683
operatings on the enclosing machine scalar), and the opposite operation
10684
is done for writes.
10686
In that case, the restrictions set forth in 13.5.1(10.3/2) for scalar components
10687
are relaxed. Instead, the following rules apply:
10693
the underlying storage elements are those at positions
10694
@cite{(position + first_bit / storage_element_size) .. (position + (last_bit + storage_element_size - 1) / storage_element_size)}
10697
the sequence of underlying storage elements shall have
10698
a size no greater than the largest machine scalar
10701
the enclosing machine scalar is defined as the smallest machine
10702
scalar starting at a position no greater than
10703
@cite{position + first_bit / storage_element_size} and covering
10704
storage elements at least up to @cite{position + (last_bit + storage_element_size - 1) / storage_element_size}
10707
the position of the component is interpreted relative to that machine
10711
If no scalar storage order is specified for a type (either directly, or by
10712
inheritance in the case of a derived type), then the default is normally
10713
the native ordering of the target, but this default can be overridden using
10714
pragma @cite{Default_Scalar_Storage_Order}.
10716
Note that the scalar storage order only affects the in-memory data
10717
representation. It has no effect on the representation used by stream
10720
@node Attribute Simple_Storage_Pool,Attribute Small,Attribute Scalar_Storage_Order,Implementation Defined Attributes
10721
@anchor{gnat_rm/implementation_defined_attributes attribute-simple-storage-pool}@anchor{b9}@anchor{gnat_rm/implementation_defined_attributes id3}@anchor{158}
10722
@section Attribute Simple_Storage_Pool
10725
@geindex Storage pool
10728
@geindex Simple storage pool
10730
@geindex Simple_Storage_Pool
10732
For every nonformal, nonderived access-to-object type @cite{Acc}, the
10733
representation attribute @cite{Simple_Storage_Pool} may be specified
10734
via an attribute_definition_clause (or by specifying the equivalent aspect):
10737
My_Pool : My_Simple_Storage_Pool_Type;
10739
type Acc is access My_Data_Type;
10741
for Acc'Simple_Storage_Pool use My_Pool;
10744
The name given in an attribute_definition_clause for the
10745
@cite{Simple_Storage_Pool} attribute shall denote a variable of
10746
a 'simple storage pool type' (see pragma @cite{Simple_Storage_Pool_Type}).
10748
The use of this attribute is only allowed for a prefix denoting a type
10749
for which it has been specified. The type of the attribute is the type
10750
of the variable specified as the simple storage pool of the access type,
10751
and the attribute denotes that variable.
10753
It is illegal to specify both @cite{Storage_Pool} and @cite{Simple_Storage_Pool}
10754
for the same access type.
10756
If the @cite{Simple_Storage_Pool} attribute has been specified for an access
10757
type, then applying the @cite{Storage_Pool} attribute to the type is flagged
10758
with a warning and its evaluation raises the exception @cite{Program_Error}.
10760
If the Simple_Storage_Pool attribute has been specified for an access
10761
type @cite{S}, then the evaluation of the attribute @code{S'Storage_Size}
10762
returns the result of calling @code{Storage_Size (S'Simple_Storage_Pool)},
10763
which is intended to indicate the number of storage elements reserved for
10764
the simple storage pool. If the Storage_Size function has not been defined
10765
for the simple storage pool type, then this attribute returns zero.
10767
If an access type @cite{S} has a specified simple storage pool of type
10768
@cite{SSP}, then the evaluation of an allocator for that access type calls
10769
the primitive @cite{Allocate} procedure for type @cite{SSP}, passing
10770
@code{S'Simple_Storage_Pool} as the pool parameter. The detailed
10771
semantics of such allocators is the same as those defined for allocators
10772
in section 13.11 of the @cite{Ada Reference Manual}, with the term
10773
@cite{simple storage pool} substituted for @cite{storage pool}.
10775
If an access type @cite{S} has a specified simple storage pool of type
10776
@cite{SSP}, then a call to an instance of the @cite{Ada.Unchecked_Deallocation}
10777
for that access type invokes the primitive @cite{Deallocate} procedure
10778
for type @cite{SSP}, passing @code{S'Simple_Storage_Pool} as the pool
10779
parameter. The detailed semantics of such unchecked deallocations is the same
10780
as defined in section 13.11.2 of the Ada Reference Manual, except that the
10781
term 'simple storage pool' is substituted for 'storage pool'.
10783
@node Attribute Small,Attribute Storage_Unit,Attribute Simple_Storage_Pool,Implementation Defined Attributes
10784
@anchor{gnat_rm/implementation_defined_attributes attribute-small}@anchor{159}
10785
@section Attribute Small
10788
@geindex Ada 83 attributes
10792
The @cite{Small} attribute is defined in Ada 95 (and Ada 2005) only for
10794
GNAT also allows this attribute to be applied to floating-point types
10795
for compatibility with Ada 83. See
10796
the Ada 83 reference manual for an exact description of the semantics of
10797
this attribute when applied to floating-point types.
10799
@node Attribute Storage_Unit,Attribute Stub_Type,Attribute Small,Implementation Defined Attributes
10800
@anchor{gnat_rm/implementation_defined_attributes attribute-storage-unit}@anchor{15a}
10801
@section Attribute Storage_Unit
10804
@geindex Storage_Unit
10806
@cite{Standard'Storage_Unit} (@cite{Standard} is the only permissible
10807
prefix) provides the same value as @cite{System.Storage_Unit}.
10809
@node Attribute Stub_Type,Attribute System_Allocator_Alignment,Attribute Storage_Unit,Implementation Defined Attributes
10810
@anchor{gnat_rm/implementation_defined_attributes attribute-stub-type}@anchor{15b}
10811
@section Attribute Stub_Type
10816
The GNAT implementation of remote access-to-classwide types is
10817
organized as described in AARM section E.4 (20.t): a value of an RACW type
10818
(designating a remote object) is represented as a normal access
10819
value, pointing to a "stub" object which in turn contains the
10820
necessary information to contact the designated remote object. A
10821
call on any dispatching operation of such a stub object does the
10822
remote call, if necessary, using the information in the stub object
10823
to locate the target partition, etc.
10825
For a prefix @cite{T} that denotes a remote access-to-classwide type,
10826
@cite{T'Stub_Type} denotes the type of the corresponding stub objects.
10828
By construction, the layout of @cite{T'Stub_Type} is identical to that of
10829
type @cite{RACW_Stub_Type} declared in the internal implementation-defined
10830
unit @cite{System.Partition_Interface}. Use of this attribute will create
10831
an implicit dependency on this unit.
10833
@node Attribute System_Allocator_Alignment,Attribute Target_Name,Attribute Stub_Type,Implementation Defined Attributes
10834
@anchor{gnat_rm/implementation_defined_attributes attribute-system-allocator-alignment}@anchor{15c}
10835
@section Attribute System_Allocator_Alignment
10841
@geindex System_Allocator_Alignment
10843
@cite{Standard'System_Allocator_Alignment} (@cite{Standard} is the only
10844
permissible prefix) provides the observable guaranted to be honored by
10845
the system allocator (malloc). This is a static value that can be used
10846
in user storage pools based on malloc either to reject allocation
10847
with alignment too large or to enable a realignment circuitry if the
10848
alignment request is larger than this value.
10850
@node Attribute Target_Name,Attribute To_Address,Attribute System_Allocator_Alignment,Implementation Defined Attributes
10851
@anchor{gnat_rm/implementation_defined_attributes attribute-target-name}@anchor{15d}
10852
@section Attribute Target_Name
10855
@geindex Target_Name
10857
@cite{Standard'Target_Name} (@cite{Standard} is the only permissible
10858
prefix) provides a static string value that identifies the target
10859
for the current compilation. For GCC implementations, this is the
10860
standard gcc target name without the terminating slash (for
10861
example, GNAT 5.0 on windows yields "i586-pc-mingw32msv").
10863
@node Attribute To_Address,Attribute To_Any,Attribute Target_Name,Implementation Defined Attributes
10864
@anchor{gnat_rm/implementation_defined_attributes attribute-to-address}@anchor{15e}
10865
@section Attribute To_Address
10868
@geindex To_Address
10870
The @cite{System'To_Address}
10871
(@cite{System} is the only permissible prefix)
10872
denotes a function identical to
10873
@cite{System.Storage_Elements.To_Address} except that
10874
it is a static attribute. This means that if its argument is
10875
a static expression, then the result of the attribute is a
10876
static expression. This means that such an expression can be
10877
used in contexts (e.g., preelaborable packages) which require a
10878
static expression and where the function call could not be used
10879
(since the function call is always nonstatic, even if its
10880
argument is static). The argument must be in the range
10881
-(2**(m-1) .. 2**m-1, where m is the memory size
10882
(typically 32 or 64). Negative values are intepreted in a
10883
modular manner (e.g., -1 means the same as 16#FFFF_FFFF# on
10884
a 32 bits machine).
10886
@node Attribute To_Any,Attribute Type_Class,Attribute To_Address,Implementation Defined Attributes
10887
@anchor{gnat_rm/implementation_defined_attributes attribute-to-any}@anchor{15f}
10888
@section Attribute To_Any
10893
This internal attribute is used for the generation of remote subprogram
10894
stubs in the context of the Distributed Systems Annex.
10896
@node Attribute Type_Class,Attribute Type_Key,Attribute To_Any,Implementation Defined Attributes
10897
@anchor{gnat_rm/implementation_defined_attributes attribute-type-class}@anchor{160}
10898
@section Attribute Type_Class
10901
@geindex Type_Class
10903
@code{type'Type_Class} for any type or subtype @cite{type} yields
10904
the value of the type class for the full type of @cite{type}. If
10905
@cite{type} is a generic formal type, the value is the value for the
10906
corresponding actual subtype. The value of this attribute is of type
10907
@code{System.Aux_DEC.Type_Class}, which has the following definition:
10911
(Type_Class_Enumeration,
10912
Type_Class_Integer,
10913
Type_Class_Fixed_Point,
10914
Type_Class_Floating_Point,
10919
Type_Class_Address);
10922
Protected types yield the value @cite{Type_Class_Task}, which thus
10923
applies to all concurrent types. This attribute is designed to
10924
be compatible with the DEC Ada 83 attribute of the same name.
10926
@node Attribute Type_Key,Attribute TypeCode,Attribute Type_Class,Implementation Defined Attributes
10927
@anchor{gnat_rm/implementation_defined_attributes attribute-type-key}@anchor{161}
10928
@section Attribute Type_Key
10933
The @cite{Type_Key} attribute is applicable to a type or subtype and
10934
yields a value of type Standard.String containing encoded information
10935
about the type or subtype. This provides improved compatibility with
10936
other implementations that support this attribute.
10938
@node Attribute TypeCode,Attribute Unconstrained_Array,Attribute Type_Key,Implementation Defined Attributes
10939
@anchor{gnat_rm/implementation_defined_attributes attribute-typecode}@anchor{162}
10940
@section Attribute TypeCode
10945
This internal attribute is used for the generation of remote subprogram
10946
stubs in the context of the Distributed Systems Annex.
10948
@node Attribute Unconstrained_Array,Attribute Universal_Literal_String,Attribute TypeCode,Implementation Defined Attributes
10949
@anchor{gnat_rm/implementation_defined_attributes attribute-unconstrained-array}@anchor{163}
10950
@section Attribute Unconstrained_Array
10953
@geindex Unconstrained_Array
10955
The @cite{Unconstrained_Array} attribute can be used with a prefix that
10956
denotes any type or subtype. It is a static attribute that yields
10957
@cite{True} if the prefix designates an unconstrained array,
10958
and @cite{False} otherwise. In a generic instance, the result is
10959
still static, and yields the result of applying this test to the
10962
@node Attribute Universal_Literal_String,Attribute Unrestricted_Access,Attribute Unconstrained_Array,Implementation Defined Attributes
10963
@anchor{gnat_rm/implementation_defined_attributes attribute-universal-literal-string}@anchor{164}
10964
@section Attribute Universal_Literal_String
10967
@geindex Named numbers
10968
@geindex representation of
10970
@geindex Universal_Literal_String
10972
The prefix of @cite{Universal_Literal_String} must be a named
10973
number. The static result is the string consisting of the characters of
10974
the number as defined in the original source. This allows the user
10975
program to access the actual text of named numbers without intermediate
10976
conversions and without the need to enclose the strings in quotes (which
10977
would preclude their use as numbers).
10979
For example, the following program prints the first 50 digits of pi:
10982
with Text_IO; use Text_IO;
10986
Put (Ada.Numerics.Pi'Universal_Literal_String);
10990
@node Attribute Unrestricted_Access,Attribute Update,Attribute Universal_Literal_String,Implementation Defined Attributes
10991
@anchor{gnat_rm/implementation_defined_attributes attribute-unrestricted-access}@anchor{165}
10992
@section Attribute Unrestricted_Access
10996
@geindex unrestricted
10998
@geindex Unrestricted_Access
11000
The @cite{Unrestricted_Access} attribute is similar to @cite{Access}
11001
except that all accessibility and aliased view checks are omitted. This
11002
is a user-beware attribute.
11004
For objects, it is similar to @cite{Address}, for which it is a
11005
desirable replacement where the value desired is an access type.
11006
In other words, its effect is similar to first applying the
11007
@cite{Address} attribute and then doing an unchecked conversion to a
11008
desired access type.
11010
For subprograms, @cite{P'Unrestricted_Access} may be used where
11011
@cite{P'Access} would be illegal, to construct a value of a
11012
less-nested named access type that designates a more-nested
11013
subprogram. This value may be used in indirect calls, so long as the
11014
more-nested subprogram still exists; once the subprogram containing it
11015
has returned, such calls are erroneous. For example:
11020
type Less_Nested is not null access procedure;
11021
Global : Less_Nested;
11029
Local_Var : Integer;
11031
procedure More_Nested is
11036
Global := More_Nested'Unrestricted_Access;
11043
When P1 is called from P2, the call via Global is OK, but if P1 were
11044
called after P2 returns, it would be an erroneous use of a dangling
11047
For objects, it is possible to use @cite{Unrestricted_Access} for any
11048
type. However, if the result is of an access-to-unconstrained array
11049
subtype, then the resulting pointer has the same scope as the context
11050
of the attribute, and must not be returned to some enclosing scope.
11051
For instance, if a function uses @cite{Unrestricted_Access} to create
11052
an access-to-unconstrained-array and returns that value to the caller,
11053
the result will involve dangling pointers. In addition, it is only
11054
valid to create pointers to unconstrained arrays using this attribute
11055
if the pointer has the normal default 'fat' representation where a
11056
pointer has two components, one points to the array and one points to
11057
the bounds. If a size clause is used to force 'thin' representation
11058
for a pointer to unconstrained where there is only space for a single
11059
pointer, then the resulting pointer is not usable.
11061
In the simple case where a direct use of Unrestricted_Access attempts
11062
to make a thin pointer for a non-aliased object, the compiler will
11063
reject the use as illegal, as shown in the following example:
11066
with System; use System;
11067
procedure SliceUA2 is
11068
type A is access all String;
11069
for A'Size use Standard'Address_Size;
11071
procedure P (Arg : A) is
11076
X : String := "hello world!";
11077
X2 : aliased String := "hello world!";
11079
AV : A := X'Unrestricted_Access; -- ERROR
11081
>>> illegal use of Unrestricted_Access attribute
11082
>>> attempt to generate thin pointer to unaliased object
11085
P (X'Unrestricted_Access); -- ERROR
11087
>>> illegal use of Unrestricted_Access attribute
11088
>>> attempt to generate thin pointer to unaliased object
11090
P (X(7 .. 12)'Unrestricted_Access); -- ERROR
11092
>>> illegal use of Unrestricted_Access attribute
11093
>>> attempt to generate thin pointer to unaliased object
11095
P (X2'Unrestricted_Access); -- OK
11099
but other cases cannot be detected by the compiler, and are
11100
considered to be erroneous. Consider the following example:
11103
with System; use System;
11104
with System; use System;
11105
procedure SliceUA is
11106
type AF is access all String;
11108
type A is access all String;
11109
for A'Size use Standard'Address_Size;
11111
procedure P (Arg : A) is
11113
if Arg'Length /= 6 then
11114
raise Program_Error;
11118
X : String := "hello world!";
11119
Y : AF := X (7 .. 12)'Unrestricted_Access;
11126
A normal unconstrained array value
11127
or a constrained array object marked as aliased has the bounds in memory
11128
just before the array, so a thin pointer can retrieve both the data and
11129
the bounds. But in this case, the non-aliased object @cite{X} does not have the
11130
bounds before the string. If the size clause for type @cite{A}
11131
were not present, then the pointer
11132
would be a fat pointer, where one component is a pointer to the bounds,
11133
and all would be well. But with the size clause present, the conversion from
11134
fat pointer to thin pointer in the call loses the bounds, and so this
11135
is erroneous, and the program likely raises a @cite{Program_Error} exception.
11137
In general, it is advisable to completely
11138
avoid mixing the use of thin pointers and the use of
11139
@cite{Unrestricted_Access} where the designated type is an
11140
unconstrained array. The use of thin pointers should be restricted to
11141
cases of porting legacy code that implicitly assumes the size of pointers,
11142
and such code should not in any case be using this attribute.
11144
Another erroneous situation arises if the attribute is
11145
applied to a constant. The resulting pointer can be used to access the
11146
constant, but the effect of trying to modify a constant in this manner
11147
is not well-defined. Consider this example:
11150
P : constant Integer := 4;
11151
type R is access all Integer;
11152
RV : R := P'Unrestricted_Access;
11157
Here we attempt to modify the constant P from 4 to 3, but the compiler may
11158
or may not notice this attempt, and subsequent references to P may yield
11159
either the value 3 or the value 4 or the assignment may blow up if the
11160
compiler decides to put P in read-only memory. One particular case where
11161
@cite{Unrestricted_Access} can be used in this way is to modify the
11162
value of an @cite{IN} parameter:
11165
procedure K (S : in String) is
11166
type R is access all Character;
11167
RV : R := S (3)'Unrestricted_Access;
11173
In general this is a risky approach. It may appear to "work" but such uses of
11174
@cite{Unrestricted_Access} are potentially non-portable, even from one version
11175
of @cite{GNAT} to another, so are best avoided if possible.
11177
@node Attribute Update,Attribute Valid_Scalars,Attribute Unrestricted_Access,Implementation Defined Attributes
11178
@anchor{gnat_rm/implementation_defined_attributes attribute-update}@anchor{166}
11179
@section Attribute Update
11184
The @cite{Update} attribute creates a copy of an array or record value
11185
with one or more modified components. The syntax is:
11188
PREFIX'Update ( RECORD_COMPONENT_ASSOCIATION_LIST )
11189
PREFIX'Update ( ARRAY_COMPONENT_ASSOCIATION @{, ARRAY_COMPONENT_ASSOCIATION @} )
11190
PREFIX'Update ( MULTIDIMENSIONAL_ARRAY_COMPONENT_ASSOCIATION
11191
@{, MULTIDIMENSIONAL_ARRAY_COMPONENT_ASSOCIATION @} )
11193
MULTIDIMENSIONAL_ARRAY_COMPONENT_ASSOCIATION ::= INDEX_EXPRESSION_LIST_LIST => EXPRESSION
11194
INDEX_EXPRESSION_LIST_LIST ::= INDEX_EXPRESSION_LIST @{| INDEX_EXPRESSION_LIST @}
11195
INDEX_EXPRESSION_LIST ::= ( EXPRESSION @{, EXPRESSION @} )
11198
where @cite{PREFIX} is the name of an array or record object, the
11199
association list in parentheses does not contain an @cite{others}
11200
choice and the box symbol @cite{<>} may not appear in any
11201
expression. The effect is to yield a copy of the array or record value
11202
which is unchanged apart from the components mentioned in the
11203
association list, which are changed to the indicated value. The
11204
original value of the array or record value is not affected. For
11208
type Arr is Array (1 .. 5) of Integer;
11210
Avar1 : Arr := (1,2,3,4,5);
11211
Avar2 : Arr := Avar1'Update (2 => 10, 3 .. 4 => 20);
11214
yields a value for @cite{Avar2} of 1,10,20,20,5 with @cite{Avar1}
11215
begin unmodified. Similarly:
11218
type Rec is A, B, C : Integer;
11220
Rvar1 : Rec := (A => 1, B => 2, C => 3);
11221
Rvar2 : Rec := Rvar1'Update (B => 20);
11224
yields a value for @cite{Rvar2} of (A => 1, B => 20, C => 3),
11225
with @cite{Rvar1} being unmodifed.
11226
Note that the value of the attribute reference is computed
11227
completely before it is used. This means that if you write:
11230
Avar1 := Avar1'Update (1 => 10, 2 => Function_Call);
11233
then the value of @cite{Avar1} is not modified if @cite{Function_Call}
11234
raises an exception, unlike the effect of a series of direct assignments
11235
to elements of @cite{Avar1}. In general this requires that
11236
two extra complete copies of the object are required, which should be
11237
kept in mind when considering efficiency.
11239
The @cite{Update} attribute cannot be applied to prefixes of a limited
11240
type, and cannot reference discriminants in the case of a record type.
11241
The accessibility level of an Update attribute result object is defined
11242
as for an aggregate.
11244
In the record case, no component can be mentioned more than once. In
11245
the array case, two overlapping ranges can appear in the association list,
11246
in which case the modifications are processed left to right.
11248
Multi-dimensional arrays can be modified, as shown by this example:
11251
A : array (1 .. 10, 1 .. 10) of Integer;
11253
A := A'Update ((1, 2) => 20, (3, 4) => 30);
11256
which changes element (1,2) to 20 and (3,4) to 30.
11258
@node Attribute Valid_Scalars,Attribute VADS_Size,Attribute Update,Implementation Defined Attributes
11259
@anchor{gnat_rm/implementation_defined_attributes attribute-valid-scalars}@anchor{167}
11260
@section Attribute Valid_Scalars
11263
@geindex Valid_Scalars
11265
The @cite{'Valid_Scalars} attribute is intended to make it easier to
11266
check the validity of scalar subcomponents of composite objects. It
11267
is defined for any prefix @cite{X} that denotes an object.
11268
The value of this attribute is of the predefined type Boolean.
11269
@cite{X'Valid_Scalars} yields True if and only if evaluation of
11270
@cite{P'Valid} yields True for every scalar part P of X or if X has
11271
no scalar parts. It is not specified in what order the scalar parts
11272
are checked, nor whether any more are checked after any one of them
11273
is determined to be invalid. If the prefix @cite{X} is of a class-wide
11274
type @cite{T'Class} (where @cite{T} is the associated specific type),
11275
or if the prefix @cite{X} is of a specific tagged type @cite{T}, then
11276
only the scalar parts of components of @cite{T} are traversed; in other
11277
words, components of extensions of @cite{T} are not traversed even if
11278
@cite{T'Class (X)'Tag /= T'Tag} . The compiler will issue a warning if it can
11279
be determined at compile time that the prefix of the attribute has no
11280
scalar parts (e.g., if the prefix is of an access type, an interface type,
11281
an undiscriminated task type, or an undiscriminated protected type).
11283
For scalar types, @cite{Valid_Scalars} is equivalent to @cite{Valid}. The use
11284
of this attribute is not permitted for @cite{Unchecked_Union} types for which
11285
in general it is not possible to determine the values of the discriminants.
11287
Note: @cite{Valid_Scalars} can generate a lot of code, especially in the case
11288
of a large variant record. If the attribute is called in many places in the
11289
same program applied to objects of the same type, it can reduce program size
11290
to write a function with a single use of the attribute, and then call that
11291
function from multiple places.
11293
@node Attribute VADS_Size,Attribute Value_Size,Attribute Valid_Scalars,Implementation Defined Attributes
11294
@anchor{gnat_rm/implementation_defined_attributes attribute-vads-size}@anchor{168}
11295
@section Attribute VADS_Size
11299
@geindex VADS compatibility
11303
The @cite{'VADS_Size} attribute is intended to make it easier to port
11304
legacy code which relies on the semantics of @cite{'Size} as implemented
11305
by the VADS Ada 83 compiler. GNAT makes a best effort at duplicating the
11306
same semantic interpretation. In particular, @cite{'VADS_Size} applied
11307
to a predefined or other primitive type with no Size clause yields the
11308
Object_Size (for example, @cite{Natural'Size} is 32 rather than 31 on
11309
typical machines). In addition @cite{'VADS_Size} applied to an object
11310
gives the result that would be obtained by applying the attribute to
11311
the corresponding type.
11313
@node Attribute Value_Size,Attribute Wchar_T_Size,Attribute VADS_Size,Implementation Defined Attributes
11314
@anchor{gnat_rm/implementation_defined_attributes attribute-value-size}@anchor{169}
11315
@section Attribute Value_Size
11319
@geindex setting for not-first subtype
11321
@geindex Value_Size
11323
@code{type'Value_Size} is the number of bits required to represent
11324
a value of the given subtype. It is the same as @code{type'Size},
11325
but, unlike @cite{Size}, may be set for non-first subtypes.
11327
@node Attribute Wchar_T_Size,Attribute Word_Size,Attribute Value_Size,Implementation Defined Attributes
11328
@anchor{gnat_rm/implementation_defined_attributes attribute-wchar-t-size}@anchor{16a}
11329
@section Attribute Wchar_T_Size
11332
@geindex Wchar_T_Size
11334
@cite{Standard'Wchar_T_Size} (@cite{Standard} is the only permissible
11335
prefix) provides the size in bits of the C @cite{wchar_t} type
11336
primarily for constructing the definition of this type in
11337
package @cite{Interfaces.C}. The result is a static constant.
11339
@node Attribute Word_Size,,Attribute Wchar_T_Size,Implementation Defined Attributes
11340
@anchor{gnat_rm/implementation_defined_attributes attribute-word-size}@anchor{16b}
11341
@section Attribute Word_Size
11346
@cite{Standard'Word_Size} (@cite{Standard} is the only permissible
11347
prefix) provides the value @cite{System.Word_Size}. The result is
11350
@node Standard and Implementation Defined Restrictions,Implementation Advice,Implementation Defined Attributes,Top
11351
@anchor{gnat_rm/standard_and_implementation_defined_restrictions standard-and-implementation-defined-restrictions}@anchor{9}@anchor{gnat_rm/standard_and_implementation_defined_restrictions doc}@anchor{16c}@anchor{gnat_rm/standard_and_implementation_defined_restrictions id1}@anchor{16d}
11352
@chapter Standard and Implementation Defined Restrictions
11355
All Ada Reference Manual-defined Restriction identifiers are implemented:
11361
language-defined restrictions (see 13.12.1)
11364
tasking restrictions (see D.7)
11367
high integrity restrictions (see H.4)
11370
GNAT implements additional restriction identifiers. All restrictions, whether
11371
language defined or GNAT-specific, are listed in the following.
11374
* Partition-Wide Restrictions::
11375
* Program Unit Level Restrictions::
11379
@node Partition-Wide Restrictions,Program Unit Level Restrictions,,Standard and Implementation Defined Restrictions
11380
@anchor{gnat_rm/standard_and_implementation_defined_restrictions partition-wide-restrictions}@anchor{16e}@anchor{gnat_rm/standard_and_implementation_defined_restrictions id2}@anchor{16f}
11381
@section Partition-Wide Restrictions
11384
There are two separate lists of restriction identifiers. The first
11385
set requires consistency throughout a partition (in other words, if the
11386
restriction identifier is used for any compilation unit in the partition,
11387
then all compilation units in the partition must obey the restriction).
11390
* Immediate_Reclamation::
11391
* Max_Asynchronous_Select_Nesting::
11392
* Max_Entry_Queue_Length::
11393
* Max_Protected_Entries::
11394
* Max_Select_Alternatives::
11395
* Max_Storage_At_Blocking::
11396
* Max_Task_Entries::
11398
* No_Abort_Statements::
11399
* No_Access_Parameter_Allocators::
11400
* No_Access_Subprograms::
11402
* No_Anonymous_Allocators::
11403
* No_Asynchronous_Control::
11405
* No_Coextensions::
11406
* No_Default_Initialization::
11409
* No_Direct_Boolean_Operators::
11411
* No_Dispatching_Calls::
11412
* No_Dynamic_Attachment::
11413
* No_Dynamic_Priorities::
11414
* No_Entry_Calls_In_Elaboration_Code::
11415
* No_Enumeration_Maps::
11416
* No_Exception_Handlers::
11417
* No_Exception_Propagation::
11418
* No_Exception_Registration::
11420
* No_Finalization::
11422
* No_Floating_Point::
11423
* No_Implicit_Conditionals::
11424
* No_Implicit_Dynamic_Code::
11425
* No_Implicit_Heap_Allocations::
11426
* No_Implicit_Loops::
11427
* No_Implicit_Protected_Object_Allocations::
11428
* No_Implicit_Task_Allocations::
11429
* No_Initialize_Scalars::
11431
* No_Local_Allocators::
11432
* No_Local_Protected_Objects::
11433
* No_Local_Timing_Events::
11434
* No_Long_Long_Integers::
11435
* No_Multiple_Elaboration::
11436
* No_Nested_Finalization::
11437
* No_Protected_Type_Allocators::
11438
* No_Protected_Types::
11441
* No_Relative_Delay::
11442
* No_Requeue_Statements::
11443
* No_Secondary_Stack::
11444
* No_Select_Statements::
11445
* No_Specific_Termination_Handlers::
11446
* No_Specification_of_Aspect::
11447
* No_Standard_Allocators_After_Elaboration::
11448
* No_Standard_Storage_Pools::
11449
* No_Stream_Optimizations::
11451
* No_Task_Allocators::
11452
* No_Task_At_Interrupt_Priority::
11453
* No_Task_Attributes_Package::
11454
* No_Task_Hierarchy::
11455
* No_Task_Termination::
11457
* No_Terminate_Alternatives::
11458
* No_Unchecked_Access::
11459
* No_Unchecked_Conversion::
11460
* No_Unchecked_Deallocation::
11461
* No_Use_Of_Entity::
11463
* Simple_Barriers::
11464
* Static_Priorities::
11465
* Static_Storage_Size::
11469
@node Immediate_Reclamation,Max_Asynchronous_Select_Nesting,,Partition-Wide Restrictions
11470
@anchor{gnat_rm/standard_and_implementation_defined_restrictions immediate-reclamation}@anchor{170}
11471
@subsection Immediate_Reclamation
11474
@geindex Immediate_Reclamation
11476
[RM H.4] This restriction ensures that, except for storage occupied by
11477
objects created by allocators and not deallocated via unchecked
11478
deallocation, any storage reserved at run time for an object is
11479
immediately reclaimed when the object no longer exists.
11481
@node Max_Asynchronous_Select_Nesting,Max_Entry_Queue_Length,Immediate_Reclamation,Partition-Wide Restrictions
11482
@anchor{gnat_rm/standard_and_implementation_defined_restrictions max-asynchronous-select-nesting}@anchor{171}
11483
@subsection Max_Asynchronous_Select_Nesting
11486
@geindex Max_Asynchronous_Select_Nesting
11488
[RM D.7] Specifies the maximum dynamic nesting level of asynchronous
11489
selects. Violations of this restriction with a value of zero are
11490
detected at compile time. Violations of this restriction with values
11491
other than zero cause Storage_Error to be raised.
11493
@node Max_Entry_Queue_Length,Max_Protected_Entries,Max_Asynchronous_Select_Nesting,Partition-Wide Restrictions
11494
@anchor{gnat_rm/standard_and_implementation_defined_restrictions max-entry-queue-length}@anchor{172}
11495
@subsection Max_Entry_Queue_Length
11498
@geindex Max_Entry_Queue_Length
11500
[RM D.7] This restriction is a declaration that any protected entry compiled in
11501
the scope of the restriction has at most the specified number of
11502
tasks waiting on the entry at any one time, and so no queue is required.
11503
Note that this restriction is checked at run time. Violation of this
11504
restriction results in the raising of Program_Error exception at the point of
11507
@geindex Max_Entry_Queue_Depth
11509
The restriction @cite{Max_Entry_Queue_Depth} is recognized as a
11510
synonym for @cite{Max_Entry_Queue_Length}. This is retained for historical
11511
compatibility purposes (and a warning will be generated for its use if
11512
warnings on obsolescent features are activated).
11514
@node Max_Protected_Entries,Max_Select_Alternatives,Max_Entry_Queue_Length,Partition-Wide Restrictions
11515
@anchor{gnat_rm/standard_and_implementation_defined_restrictions max-protected-entries}@anchor{173}
11516
@subsection Max_Protected_Entries
11519
@geindex Max_Protected_Entries
11521
[RM D.7] Specifies the maximum number of entries per protected type. The
11522
bounds of every entry family of a protected unit shall be static, or shall be
11523
defined by a discriminant of a subtype whose corresponding bound is static.
11525
@node Max_Select_Alternatives,Max_Storage_At_Blocking,Max_Protected_Entries,Partition-Wide Restrictions
11526
@anchor{gnat_rm/standard_and_implementation_defined_restrictions max-select-alternatives}@anchor{174}
11527
@subsection Max_Select_Alternatives
11530
@geindex Max_Select_Alternatives
11532
[RM D.7] Specifies the maximum number of alternatives in a selective accept.
11534
@node Max_Storage_At_Blocking,Max_Task_Entries,Max_Select_Alternatives,Partition-Wide Restrictions
11535
@anchor{gnat_rm/standard_and_implementation_defined_restrictions max-storage-at-blocking}@anchor{175}
11536
@subsection Max_Storage_At_Blocking
11539
@geindex Max_Storage_At_Blocking
11541
[RM D.7] Specifies the maximum portion (in storage elements) of a task's
11542
Storage_Size that can be retained by a blocked task. A violation of this
11543
restriction causes Storage_Error to be raised.
11545
@node Max_Task_Entries,Max_Tasks,Max_Storage_At_Blocking,Partition-Wide Restrictions
11546
@anchor{gnat_rm/standard_and_implementation_defined_restrictions max-task-entries}@anchor{176}
11547
@subsection Max_Task_Entries
11550
@geindex Max_Task_Entries
11552
[RM D.7] Specifies the maximum number of entries
11553
per task. The bounds of every entry family
11554
of a task unit shall be static, or shall be
11555
defined by a discriminant of a subtype whose
11556
corresponding bound is static.
11558
@node Max_Tasks,No_Abort_Statements,Max_Task_Entries,Partition-Wide Restrictions
11559
@anchor{gnat_rm/standard_and_implementation_defined_restrictions max-tasks}@anchor{177}
11560
@subsection Max_Tasks
11565
[RM D.7] Specifies the maximum number of task that may be created, not
11566
counting the creation of the environment task. Violations of this
11567
restriction with a value of zero are detected at compile
11568
time. Violations of this restriction with values other than zero cause
11569
Storage_Error to be raised.
11571
@node No_Abort_Statements,No_Access_Parameter_Allocators,Max_Tasks,Partition-Wide Restrictions
11572
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-abort-statements}@anchor{178}
11573
@subsection No_Abort_Statements
11576
@geindex No_Abort_Statements
11578
[RM D.7] There are no abort_statements, and there are
11579
no calls to Task_Identification.Abort_Task.
11581
@node No_Access_Parameter_Allocators,No_Access_Subprograms,No_Abort_Statements,Partition-Wide Restrictions
11582
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-access-parameter-allocators}@anchor{179}
11583
@subsection No_Access_Parameter_Allocators
11586
@geindex No_Access_Parameter_Allocators
11588
[RM H.4] This restriction ensures at compile time that there are no
11589
occurrences of an allocator as the actual parameter to an access
11592
@node No_Access_Subprograms,No_Allocators,No_Access_Parameter_Allocators,Partition-Wide Restrictions
11593
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-access-subprograms}@anchor{17a}
11594
@subsection No_Access_Subprograms
11597
@geindex No_Access_Subprograms
11599
[RM H.4] This restriction ensures at compile time that there are no
11600
declarations of access-to-subprogram types.
11602
@node No_Allocators,No_Anonymous_Allocators,No_Access_Subprograms,Partition-Wide Restrictions
11603
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-allocators}@anchor{17b}
11604
@subsection No_Allocators
11607
@geindex No_Allocators
11609
[RM H.4] This restriction ensures at compile time that there are no
11610
occurrences of an allocator.
11612
@node No_Anonymous_Allocators,No_Asynchronous_Control,No_Allocators,Partition-Wide Restrictions
11613
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-anonymous-allocators}@anchor{17c}
11614
@subsection No_Anonymous_Allocators
11617
@geindex No_Anonymous_Allocators
11619
[RM H.4] This restriction ensures at compile time that there are no
11620
occurrences of an allocator of anonymous access type.
11622
@node No_Asynchronous_Control,No_Calendar,No_Anonymous_Allocators,Partition-Wide Restrictions
11623
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-asynchronous-control}@anchor{17d}
11624
@subsection No_Asynchronous_Control
11627
@geindex No_Asynchronous_Control
11629
[RM J.13] This restriction ensures at compile time that there are no semantic
11630
dependences on the predefined package Asynchronous_Task_Control.
11632
@node No_Calendar,No_Coextensions,No_Asynchronous_Control,Partition-Wide Restrictions
11633
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-calendar}@anchor{17e}
11634
@subsection No_Calendar
11637
@geindex No_Calendar
11639
[GNAT] This restriction ensures at compile time that there are no semantic
11640
dependences on package Calendar.
11642
@node No_Coextensions,No_Default_Initialization,No_Calendar,Partition-Wide Restrictions
11643
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-coextensions}@anchor{17f}
11644
@subsection No_Coextensions
11647
@geindex No_Coextensions
11649
[RM H.4] This restriction ensures at compile time that there are no
11650
coextensions. See 3.10.2.
11652
@node No_Default_Initialization,No_Delay,No_Coextensions,Partition-Wide Restrictions
11653
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-default-initialization}@anchor{180}
11654
@subsection No_Default_Initialization
11657
@geindex No_Default_Initialization
11659
[GNAT] This restriction prohibits any instance of default initialization
11660
of variables. The binder implements a consistency rule which prevents
11661
any unit compiled without the restriction from with'ing a unit with the
11662
restriction (this allows the generation of initialization procedures to
11663
be skipped, since you can be sure that no call is ever generated to an
11664
initialization procedure in a unit with the restriction active). If used
11665
in conjunction with Initialize_Scalars or Normalize_Scalars, the effect
11666
is to prohibit all cases of variables declared without a specific
11667
initializer (including the case of OUT scalar parameters).
11669
@node No_Delay,No_Dependence,No_Default_Initialization,Partition-Wide Restrictions
11670
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-delay}@anchor{181}
11671
@subsection No_Delay
11676
[RM H.4] This restriction ensures at compile time that there are no
11677
delay statements and no semantic dependences on package Calendar.
11679
@node No_Dependence,No_Direct_Boolean_Operators,No_Delay,Partition-Wide Restrictions
11680
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-dependence}@anchor{182}
11681
@subsection No_Dependence
11684
@geindex No_Dependence
11686
[RM 13.12.1] This restriction ensures at compile time that there are no
11687
dependences on a library unit.
11689
@node No_Direct_Boolean_Operators,No_Dispatch,No_Dependence,Partition-Wide Restrictions
11690
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-direct-boolean-operators}@anchor{183}
11691
@subsection No_Direct_Boolean_Operators
11694
@geindex No_Direct_Boolean_Operators
11696
[GNAT] This restriction ensures that no logical operators (and/or/xor)
11697
are used on operands of type Boolean (or any type derived from Boolean).
11698
This is intended for use in safety critical programs where the certification
11699
protocol requires the use of short-circuit (and then, or else) forms for all
11700
composite boolean operations.
11702
@node No_Dispatch,No_Dispatching_Calls,No_Direct_Boolean_Operators,Partition-Wide Restrictions
11703
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-dispatch}@anchor{184}
11704
@subsection No_Dispatch
11707
@geindex No_Dispatch
11709
[RM H.4] This restriction ensures at compile time that there are no
11710
occurrences of @cite{T'Class}, for any (tagged) subtype @cite{T}.
11712
@node No_Dispatching_Calls,No_Dynamic_Attachment,No_Dispatch,Partition-Wide Restrictions
11713
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-dispatching-calls}@anchor{185}
11714
@subsection No_Dispatching_Calls
11717
@geindex No_Dispatching_Calls
11719
[GNAT] This restriction ensures at compile time that the code generated by the
11720
compiler involves no dispatching calls. The use of this restriction allows the
11721
safe use of record extensions, classwide membership tests and other classwide
11722
features not involving implicit dispatching. This restriction ensures that
11723
the code contains no indirect calls through a dispatching mechanism. Note that
11724
this includes internally-generated calls created by the compiler, for example
11725
in the implementation of class-wide objects assignments. The
11726
membership test is allowed in the presence of this restriction, because its
11727
implementation requires no dispatching.
11728
This restriction is comparable to the official Ada restriction
11729
@cite{No_Dispatch} except that it is a bit less restrictive in that it allows
11730
all classwide constructs that do not imply dispatching.
11731
The following example indicates constructs that violate this restriction.
11735
type T is tagged record
11738
procedure P (X : T);
11740
type DT is new T with record
11741
More_Data : Natural;
11743
procedure Q (X : DT);
11747
procedure Example is
11748
procedure Test (O : T'Class) is
11749
N : Natural := O'Size;-- Error: Dispatching call
11750
C : T'Class := O; -- Error: implicit Dispatching Call
11752
if O in DT'Class then -- OK : Membership test
11753
Q (DT (O)); -- OK : Type conversion plus direct call
11755
P (O); -- Error: Dispatching call
11761
P (Obj); -- OK : Direct call
11762
P (T (Obj)); -- OK : Type conversion plus direct call
11763
P (T'Class (Obj)); -- Error: Dispatching call
11765
Test (Obj); -- OK : Type conversion
11767
if Obj in T'Class then -- OK : Membership test
11773
@node No_Dynamic_Attachment,No_Dynamic_Priorities,No_Dispatching_Calls,Partition-Wide Restrictions
11774
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-dynamic-attachment}@anchor{186}
11775
@subsection No_Dynamic_Attachment
11778
@geindex No_Dynamic_Attachment
11780
[RM D.7] This restriction ensures that there is no call to any of the
11781
operations defined in package Ada.Interrupts
11782
(Is_Reserved, Is_Attached, Current_Handler, Attach_Handler, Exchange_Handler,
11783
Detach_Handler, and Reference).
11785
@geindex No_Dynamic_Interrupts
11787
The restriction @cite{No_Dynamic_Interrupts} is recognized as a
11788
synonym for @cite{No_Dynamic_Attachment}. This is retained for historical
11789
compatibility purposes (and a warning will be generated for its use if
11790
warnings on obsolescent features are activated).
11792
@node No_Dynamic_Priorities,No_Entry_Calls_In_Elaboration_Code,No_Dynamic_Attachment,Partition-Wide Restrictions
11793
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-dynamic-priorities}@anchor{187}
11794
@subsection No_Dynamic_Priorities
11797
@geindex No_Dynamic_Priorities
11799
[RM D.7] There are no semantic dependencies on the package Dynamic_Priorities.
11801
@node No_Entry_Calls_In_Elaboration_Code,No_Enumeration_Maps,No_Dynamic_Priorities,Partition-Wide Restrictions
11802
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-entry-calls-in-elaboration-code}@anchor{188}
11803
@subsection No_Entry_Calls_In_Elaboration_Code
11806
@geindex No_Entry_Calls_In_Elaboration_Code
11808
[GNAT] This restriction ensures at compile time that no task or protected entry
11809
calls are made during elaboration code. As a result of the use of this
11810
restriction, the compiler can assume that no code past an accept statement
11811
in a task can be executed at elaboration time.
11813
@node No_Enumeration_Maps,No_Exception_Handlers,No_Entry_Calls_In_Elaboration_Code,Partition-Wide Restrictions
11814
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-enumeration-maps}@anchor{189}
11815
@subsection No_Enumeration_Maps
11818
@geindex No_Enumeration_Maps
11820
[GNAT] This restriction ensures at compile time that no operations requiring
11821
enumeration maps are used (that is Image and Value attributes applied
11822
to enumeration types).
11824
@node No_Exception_Handlers,No_Exception_Propagation,No_Enumeration_Maps,Partition-Wide Restrictions
11825
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-exception-handlers}@anchor{18a}
11826
@subsection No_Exception_Handlers
11829
@geindex No_Exception_Handlers
11831
[GNAT] This restriction ensures at compile time that there are no explicit
11832
exception handlers. It also indicates that no exception propagation will
11833
be provided. In this mode, exceptions may be raised but will result in
11834
an immediate call to the last chance handler, a routine that the user
11835
must define with the following profile:
11838
procedure Last_Chance_Handler
11839
(Source_Location : System.Address; Line : Integer);
11840
pragma Export (C, Last_Chance_Handler,
11841
"__gnat_last_chance_handler");
11844
The parameter is a C null-terminated string representing a message to be
11845
associated with the exception (typically the source location of the raise
11846
statement generated by the compiler). The Line parameter when nonzero
11847
represents the line number in the source program where the raise occurs.
11849
@node No_Exception_Propagation,No_Exception_Registration,No_Exception_Handlers,Partition-Wide Restrictions
11850
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-exception-propagation}@anchor{18b}
11851
@subsection No_Exception_Propagation
11854
@geindex No_Exception_Propagation
11856
[GNAT] This restriction guarantees that exceptions are never propagated
11857
to an outer subprogram scope. The only case in which an exception may
11858
be raised is when the handler is statically in the same subprogram, so
11859
that the effect of a raise is essentially like a goto statement. Any
11860
other raise statement (implicit or explicit) will be considered
11861
unhandled. Exception handlers are allowed, but may not contain an
11862
exception occurrence identifier (exception choice). In addition, use of
11863
the package GNAT.Current_Exception is not permitted, and reraise
11864
statements (raise with no operand) are not permitted.
11866
@node No_Exception_Registration,No_Exceptions,No_Exception_Propagation,Partition-Wide Restrictions
11867
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-exception-registration}@anchor{18c}
11868
@subsection No_Exception_Registration
11871
@geindex No_Exception_Registration
11873
[GNAT] This restriction ensures at compile time that no stream operations for
11874
types Exception_Id or Exception_Occurrence are used. This also makes it
11875
impossible to pass exceptions to or from a partition with this restriction
11876
in a distributed environment. If this restriction is active, the generated
11877
code is simplified by omitting the otherwise-required global registration
11878
of exceptions when they are declared.
11880
@node No_Exceptions,No_Finalization,No_Exception_Registration,Partition-Wide Restrictions
11881
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-exceptions}@anchor{18d}
11882
@subsection No_Exceptions
11885
@geindex No_Exceptions
11887
[RM H.4] This restriction ensures at compile time that there are no
11888
raise statements and no exception handlers.
11890
@node No_Finalization,No_Fixed_Point,No_Exceptions,Partition-Wide Restrictions
11891
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-finalization}@anchor{18e}
11892
@subsection No_Finalization
11895
@geindex No_Finalization
11897
[GNAT] This restriction disables the language features described in
11898
chapter 7.6 of the Ada 2005 RM as well as all form of code generation
11899
performed by the compiler to support these features. The following types
11900
are no longer considered controlled when this restriction is in effect:
11906
@cite{Ada.Finalization.Controlled}
11909
@cite{Ada.Finalization.Limited_Controlled}
11912
Derivations from @cite{Controlled} or @cite{Limited_Controlled}
11924
Array and record types with controlled components
11927
The compiler no longer generates code to initialize, finalize or adjust an
11928
object or a nested component, either declared on the stack or on the heap. The
11929
deallocation of a controlled object no longer finalizes its contents.
11931
@node No_Fixed_Point,No_Floating_Point,No_Finalization,Partition-Wide Restrictions
11932
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-fixed-point}@anchor{18f}
11933
@subsection No_Fixed_Point
11936
@geindex No_Fixed_Point
11938
[RM H.4] This restriction ensures at compile time that there are no
11939
occurrences of fixed point types and operations.
11941
@node No_Floating_Point,No_Implicit_Conditionals,No_Fixed_Point,Partition-Wide Restrictions
11942
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-floating-point}@anchor{190}
11943
@subsection No_Floating_Point
11946
@geindex No_Floating_Point
11948
[RM H.4] This restriction ensures at compile time that there are no
11949
occurrences of floating point types and operations.
11951
@node No_Implicit_Conditionals,No_Implicit_Dynamic_Code,No_Floating_Point,Partition-Wide Restrictions
11952
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implicit-conditionals}@anchor{191}
11953
@subsection No_Implicit_Conditionals
11956
@geindex No_Implicit_Conditionals
11958
[GNAT] This restriction ensures that the generated code does not contain any
11959
implicit conditionals, either by modifying the generated code where possible,
11960
or by rejecting any construct that would otherwise generate an implicit
11961
conditional. Note that this check does not include run time constraint
11962
checks, which on some targets may generate implicit conditionals as
11963
well. To control the latter, constraint checks can be suppressed in the
11964
normal manner. Constructs generating implicit conditionals include comparisons
11965
of composite objects and the Max/Min attributes.
11967
@node No_Implicit_Dynamic_Code,No_Implicit_Heap_Allocations,No_Implicit_Conditionals,Partition-Wide Restrictions
11968
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implicit-dynamic-code}@anchor{192}
11969
@subsection No_Implicit_Dynamic_Code
11972
@geindex No_Implicit_Dynamic_Code
11974
@geindex trampoline
11976
[GNAT] This restriction prevents the compiler from building 'trampolines'.
11977
This is a structure that is built on the stack and contains dynamic
11978
code to be executed at run time. On some targets, a trampoline is
11979
built for the following features: @cite{Access},
11980
@cite{Unrestricted_Access}, or @cite{Address} of a nested subprogram;
11981
nested task bodies; primitive operations of nested tagged types.
11982
Trampolines do not work on machines that prevent execution of stack
11983
data. For example, on windows systems, enabling DEP (data execution
11984
protection) will cause trampolines to raise an exception.
11985
Trampolines are also quite slow at run time.
11987
On many targets, trampolines have been largely eliminated. Look at the
11988
version of system.ads for your target --- if it has
11989
Always_Compatible_Rep equal to False, then trampolines are largely
11990
eliminated. In particular, a trampoline is built for the following
11991
features: @cite{Address} of a nested subprogram;
11992
@cite{Access} or @cite{Unrestricted_Access} of a nested subprogram,
11993
but only if pragma Favor_Top_Level applies, or the access type has a
11994
foreign-language convention; primitive operations of nested tagged
11997
@node No_Implicit_Heap_Allocations,No_Implicit_Loops,No_Implicit_Dynamic_Code,Partition-Wide Restrictions
11998
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implicit-heap-allocations}@anchor{193}
11999
@subsection No_Implicit_Heap_Allocations
12002
@geindex No_Implicit_Heap_Allocations
12004
[RM D.7] No constructs are allowed to cause implicit heap allocation.
12006
@node No_Implicit_Loops,No_Implicit_Protected_Object_Allocations,No_Implicit_Heap_Allocations,Partition-Wide Restrictions
12007
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implicit-loops}@anchor{194}
12008
@subsection No_Implicit_Loops
12011
@geindex No_Implicit_Loops
12013
[GNAT] This restriction ensures that the generated code does not contain any
12014
implicit @cite{for} loops, either by modifying
12015
the generated code where possible,
12016
or by rejecting any construct that would otherwise generate an implicit
12017
@cite{for} loop. If this restriction is active, it is possible to build
12018
large array aggregates with all static components without generating an
12019
intermediate temporary, and without generating a loop to initialize individual
12020
components. Otherwise, a loop is created for arrays larger than about 5000
12023
@node No_Implicit_Protected_Object_Allocations,No_Implicit_Task_Allocations,No_Implicit_Loops,Partition-Wide Restrictions
12024
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implicit-protected-object-allocations}@anchor{195}
12025
@subsection No_Implicit_Protected_Object_Allocations
12028
@geindex No_Implicit_Protected_Object_Allocations
12030
[GNAT] No constructs are allowed to cause implicit heap allocation of a
12033
@node No_Implicit_Task_Allocations,No_Initialize_Scalars,No_Implicit_Protected_Object_Allocations,Partition-Wide Restrictions
12034
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implicit-task-allocations}@anchor{196}
12035
@subsection No_Implicit_Task_Allocations
12038
@geindex No_Implicit_Task_Allocations
12040
[GNAT] No constructs are allowed to cause implicit heap allocation of a task.
12042
@node No_Initialize_Scalars,No_IO,No_Implicit_Task_Allocations,Partition-Wide Restrictions
12043
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-initialize-scalars}@anchor{197}
12044
@subsection No_Initialize_Scalars
12047
@geindex No_Initialize_Scalars
12049
[GNAT] This restriction ensures that no unit in the partition is compiled with
12050
pragma Initialize_Scalars. This allows the generation of more efficient
12051
code, and in particular eliminates dummy null initialization routines that
12052
are otherwise generated for some record and array types.
12054
@node No_IO,No_Local_Allocators,No_Initialize_Scalars,Partition-Wide Restrictions
12055
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-io}@anchor{198}
12061
[RM H.4] This restriction ensures at compile time that there are no
12062
dependences on any of the library units Sequential_IO, Direct_IO,
12063
Text_IO, Wide_Text_IO, Wide_Wide_Text_IO, or Stream_IO.
12065
@node No_Local_Allocators,No_Local_Protected_Objects,No_IO,Partition-Wide Restrictions
12066
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-local-allocators}@anchor{199}
12067
@subsection No_Local_Allocators
12070
@geindex No_Local_Allocators
12072
[RM H.4] This restriction ensures at compile time that there are no
12073
occurrences of an allocator in subprograms, generic subprograms, tasks,
12076
@node No_Local_Protected_Objects,No_Local_Timing_Events,No_Local_Allocators,Partition-Wide Restrictions
12077
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-local-protected-objects}@anchor{19a}
12078
@subsection No_Local_Protected_Objects
12081
@geindex No_Local_Protected_Objects
12083
[RM D.7] This restriction ensures at compile time that protected objects are
12084
only declared at the library level.
12086
@node No_Local_Timing_Events,No_Long_Long_Integers,No_Local_Protected_Objects,Partition-Wide Restrictions
12087
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-local-timing-events}@anchor{19b}
12088
@subsection No_Local_Timing_Events
12091
@geindex No_Local_Timing_Events
12093
[RM D.7] All objects of type Ada.Timing_Events.Timing_Event are
12094
declared at the library level.
12096
@node No_Long_Long_Integers,No_Multiple_Elaboration,No_Local_Timing_Events,Partition-Wide Restrictions
12097
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-long-long-integers}@anchor{19c}
12098
@subsection No_Long_Long_Integers
12101
@geindex No_Long_Long_Integers
12103
[GNAT] This partition-wide restriction forbids any explicit reference to
12104
type Standard.Long_Long_Integer, and also forbids declaring range types whose
12105
implicit base type is Long_Long_Integer, and modular types whose size exceeds
12108
@node No_Multiple_Elaboration,No_Nested_Finalization,No_Long_Long_Integers,Partition-Wide Restrictions
12109
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-multiple-elaboration}@anchor{19d}
12110
@subsection No_Multiple_Elaboration
12113
@geindex No_Multiple_Elaboration
12115
[GNAT] Normally each package contains a 16-bit counter used to check for access
12116
before elaboration, and to control multiple elaboration attempts.
12117
This counter is eliminated for units compiled with the static model
12118
of elaboration if restriction @cite{No_Elaboration_Code}
12119
is active but because of
12120
the need to check for multiple elaboration in the general case, these
12121
counters cannot be eliminated if elaboration code may be present. The
12122
restriction @cite{No_Multiple_Elaboration}
12123
allows suppression of these counters
12124
in static elaboration units even if they do have elaboration code. If this
12125
restriction is used, then the situations in which multiple elaboration is
12126
possible, including non-Ada main programs, and Stand Alone libraries, are not
12127
permitted, and will be diagnosed by the binder.
12129
@node No_Nested_Finalization,No_Protected_Type_Allocators,No_Multiple_Elaboration,Partition-Wide Restrictions
12130
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-nested-finalization}@anchor{19e}
12131
@subsection No_Nested_Finalization
12134
@geindex No_Nested_Finalization
12136
[RM D.7] All objects requiring finalization are declared at the library level.
12138
@node No_Protected_Type_Allocators,No_Protected_Types,No_Nested_Finalization,Partition-Wide Restrictions
12139
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-protected-type-allocators}@anchor{19f}
12140
@subsection No_Protected_Type_Allocators
12143
@geindex No_Protected_Type_Allocators
12145
[RM D.7] This restriction ensures at compile time that there are no allocator
12146
expressions that attempt to allocate protected objects.
12148
@node No_Protected_Types,No_Recursion,No_Protected_Type_Allocators,Partition-Wide Restrictions
12149
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-protected-types}@anchor{1a0}
12150
@subsection No_Protected_Types
12153
@geindex No_Protected_Types
12155
[RM H.4] This restriction ensures at compile time that there are no
12156
declarations of protected types or protected objects.
12158
@node No_Recursion,No_Reentrancy,No_Protected_Types,Partition-Wide Restrictions
12159
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-recursion}@anchor{1a1}
12160
@subsection No_Recursion
12163
@geindex No_Recursion
12165
[RM H.4] A program execution is erroneous if a subprogram is invoked as
12166
part of its execution.
12168
@node No_Reentrancy,No_Relative_Delay,No_Recursion,Partition-Wide Restrictions
12169
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-reentrancy}@anchor{1a2}
12170
@subsection No_Reentrancy
12173
@geindex No_Reentrancy
12175
[RM H.4] A program execution is erroneous if a subprogram is executed by
12176
two tasks at the same time.
12178
@node No_Relative_Delay,No_Requeue_Statements,No_Reentrancy,Partition-Wide Restrictions
12179
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-relative-delay}@anchor{1a3}
12180
@subsection No_Relative_Delay
12183
@geindex No_Relative_Delay
12185
[RM D.7] This restriction ensures at compile time that there are no delay
12186
relative statements and prevents expressions such as @cite{delay 1.23;} from
12187
appearing in source code.
12189
@node No_Requeue_Statements,No_Secondary_Stack,No_Relative_Delay,Partition-Wide Restrictions
12190
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-requeue-statements}@anchor{1a4}
12191
@subsection No_Requeue_Statements
12194
@geindex No_Requeue_Statements
12196
[RM D.7] This restriction ensures at compile time that no requeue statements
12197
are permitted and prevents keyword @cite{requeue} from being used in source
12200
@geindex No_Requeue
12202
The restriction @cite{No_Requeue} is recognized as a
12203
synonym for @cite{No_Requeue_Statements}. This is retained for historical
12204
compatibility purposes (and a warning will be generated for its use if
12205
warnings on oNobsolescent features are activated).
12207
@node No_Secondary_Stack,No_Select_Statements,No_Requeue_Statements,Partition-Wide Restrictions
12208
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-secondary-stack}@anchor{1a5}
12209
@subsection No_Secondary_Stack
12212
@geindex No_Secondary_Stack
12214
[GNAT] This restriction ensures at compile time that the generated code
12215
does not contain any reference to the secondary stack. The secondary
12216
stack is used to implement functions returning unconstrained objects
12217
(arrays or records) on some targets.
12219
@node No_Select_Statements,No_Specific_Termination_Handlers,No_Secondary_Stack,Partition-Wide Restrictions
12220
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-select-statements}@anchor{1a6}
12221
@subsection No_Select_Statements
12224
@geindex No_Select_Statements
12226
[RM D.7] This restriction ensures at compile time no select statements of any
12227
kind are permitted, that is the keyword @cite{select} may not appear.
12229
@node No_Specific_Termination_Handlers,No_Specification_of_Aspect,No_Select_Statements,Partition-Wide Restrictions
12230
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-specific-termination-handlers}@anchor{1a7}
12231
@subsection No_Specific_Termination_Handlers
12234
@geindex No_Specific_Termination_Handlers
12236
[RM D.7] There are no calls to Ada.Task_Termination.Set_Specific_Handler
12237
or to Ada.Task_Termination.Specific_Handler.
12239
@node No_Specification_of_Aspect,No_Standard_Allocators_After_Elaboration,No_Specific_Termination_Handlers,Partition-Wide Restrictions
12240
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-specification-of-aspect}@anchor{1a8}
12241
@subsection No_Specification_of_Aspect
12244
@geindex No_Specification_of_Aspect
12246
[RM 13.12.1] This restriction checks at compile time that no aspect
12247
specification, attribute definition clause, or pragma is given for a
12250
@node No_Standard_Allocators_After_Elaboration,No_Standard_Storage_Pools,No_Specification_of_Aspect,Partition-Wide Restrictions
12251
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-standard-allocators-after-elaboration}@anchor{1a9}
12252
@subsection No_Standard_Allocators_After_Elaboration
12255
@geindex No_Standard_Allocators_After_Elaboration
12257
[RM D.7] Specifies that an allocator using a standard storage pool
12258
should never be evaluated at run time after the elaboration of the
12259
library items of the partition has completed. Otherwise, Storage_Error
12262
@node No_Standard_Storage_Pools,No_Stream_Optimizations,No_Standard_Allocators_After_Elaboration,Partition-Wide Restrictions
12263
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-standard-storage-pools}@anchor{1aa}
12264
@subsection No_Standard_Storage_Pools
12267
@geindex No_Standard_Storage_Pools
12269
[GNAT] This restriction ensures at compile time that no access types
12270
use the standard default storage pool. Any access type declared must
12271
have an explicit Storage_Pool attribute defined specifying a
12272
user-defined storage pool.
12274
@node No_Stream_Optimizations,No_Streams,No_Standard_Storage_Pools,Partition-Wide Restrictions
12275
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-stream-optimizations}@anchor{1ab}
12276
@subsection No_Stream_Optimizations
12279
@geindex No_Stream_Optimizations
12281
[GNAT] This restriction affects the performance of stream operations on types
12282
@cite{String}, @cite{Wide_String} and @cite{Wide_Wide_String}. By default, the
12283
compiler uses block reads and writes when manipulating @cite{String} objects
12284
due to their supperior performance. When this restriction is in effect, the
12285
compiler performs all IO operations on a per-character basis.
12287
@node No_Streams,No_Task_Allocators,No_Stream_Optimizations,Partition-Wide Restrictions
12288
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-streams}@anchor{1ac}
12289
@subsection No_Streams
12292
@geindex No_Streams
12294
[GNAT] This restriction ensures at compile/bind time that there are no
12295
stream objects created and no use of stream attributes.
12296
This restriction does not forbid dependences on the package
12297
@cite{Ada.Streams}. So it is permissible to with
12298
@cite{Ada.Streams} (or another package that does so itself)
12299
as long as no actual stream objects are created and no
12300
stream attributes are used.
12302
Note that the use of restriction allows optimization of tagged types,
12303
since they do not need to worry about dispatching stream operations.
12304
To take maximum advantage of this space-saving optimization, any
12305
unit declaring a tagged type should be compiled with the restriction,
12306
though this is not required.
12308
@node No_Task_Allocators,No_Task_At_Interrupt_Priority,No_Streams,Partition-Wide Restrictions
12309
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-task-allocators}@anchor{1ad}
12310
@subsection No_Task_Allocators
12313
@geindex No_Task_Allocators
12315
[RM D.7] There are no allocators for task types
12316
or types containing task subcomponents.
12318
@node No_Task_At_Interrupt_Priority,No_Task_Attributes_Package,No_Task_Allocators,Partition-Wide Restrictions
12319
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-task-at-interrupt-priority}@anchor{1ae}
12320
@subsection No_Task_At_Interrupt_Priority
12323
@geindex No_Task_At_Interrupt_Priority
12325
[GNAT] This restriction ensures at compile time that there is no
12326
Interrupt_Priority aspect or pragma for a task or a task type. As
12327
a consequence, the tasks are always created with a priority below
12328
that an interrupt priority.
12330
@node No_Task_Attributes_Package,No_Task_Hierarchy,No_Task_At_Interrupt_Priority,Partition-Wide Restrictions
12331
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-task-attributes-package}@anchor{1af}
12332
@subsection No_Task_Attributes_Package
12335
@geindex No_Task_Attributes_Package
12337
[GNAT] This restriction ensures at compile time that there are no implicit or
12338
explicit dependencies on the package @cite{Ada.Task_Attributes}.
12340
@geindex No_Task_Attributes
12342
The restriction @cite{No_Task_Attributes} is recognized as a synonym
12343
for @cite{No_Task_Attributes_Package}. This is retained for historical
12344
compatibility purposes (and a warning will be generated for its use if
12345
warnings on obsolescent features are activated).
12347
@node No_Task_Hierarchy,No_Task_Termination,No_Task_Attributes_Package,Partition-Wide Restrictions
12348
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-task-hierarchy}@anchor{1b0}
12349
@subsection No_Task_Hierarchy
12352
@geindex No_Task_Hierarchy
12354
[RM D.7] All (non-environment) tasks depend
12355
directly on the environment task of the partition.
12357
@node No_Task_Termination,No_Tasking,No_Task_Hierarchy,Partition-Wide Restrictions
12358
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-task-termination}@anchor{1b1}
12359
@subsection No_Task_Termination
12362
@geindex No_Task_Termination
12364
[RM D.7] Tasks that terminate are erroneous.
12366
@node No_Tasking,No_Terminate_Alternatives,No_Task_Termination,Partition-Wide Restrictions
12367
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-tasking}@anchor{1b2}
12368
@subsection No_Tasking
12371
@geindex No_Tasking
12373
[GNAT] This restriction prevents the declaration of tasks or task types
12374
throughout the partition. It is similar in effect to the use of
12375
@cite{Max_Tasks => 0} except that violations are caught at compile time
12376
and cause an error message to be output either by the compiler or
12379
@node No_Terminate_Alternatives,No_Unchecked_Access,No_Tasking,Partition-Wide Restrictions
12380
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-terminate-alternatives}@anchor{1b3}
12381
@subsection No_Terminate_Alternatives
12384
@geindex No_Terminate_Alternatives
12386
[RM D.7] There are no selective accepts with terminate alternatives.
12388
@node No_Unchecked_Access,No_Unchecked_Conversion,No_Terminate_Alternatives,Partition-Wide Restrictions
12389
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-unchecked-access}@anchor{1b4}
12390
@subsection No_Unchecked_Access
12393
@geindex No_Unchecked_Access
12395
[RM H.4] This restriction ensures at compile time that there are no
12396
occurrences of the Unchecked_Access attribute.
12398
@node No_Unchecked_Conversion,No_Unchecked_Deallocation,No_Unchecked_Access,Partition-Wide Restrictions
12399
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-unchecked-conversion}@anchor{1b5}
12400
@subsection No_Unchecked_Conversion
12403
@geindex No_Unchecked_Conversion
12405
[RM J.13] This restriction ensures at compile time that there are no semantic
12406
dependences on the predefined generic function Unchecked_Conversion.
12408
@node No_Unchecked_Deallocation,No_Use_Of_Entity,No_Unchecked_Conversion,Partition-Wide Restrictions
12409
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-unchecked-deallocation}@anchor{1b6}
12410
@subsection No_Unchecked_Deallocation
12413
@geindex No_Unchecked_Deallocation
12415
[RM J.13] This restriction ensures at compile time that there are no semantic
12416
dependences on the predefined generic procedure Unchecked_Deallocation.
12418
@node No_Use_Of_Entity,Pure_Barriers,No_Unchecked_Deallocation,Partition-Wide Restrictions
12419
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-use-of-entity}@anchor{1b7}
12420
@subsection No_Use_Of_Entity
12423
@geindex No_Use_Of_Entity
12425
[GNAT] This restriction ensures at compile time that there are no references
12426
to the entity given in the form
12429
No_Use_Of_Entity => Name
12432
where @code{Name} is the fully qualified entity, for example
12435
No_Use_Of_Entity => Ada.Text_IO.Put_Line
12438
@node Pure_Barriers,Simple_Barriers,No_Use_Of_Entity,Partition-Wide Restrictions
12439
@anchor{gnat_rm/standard_and_implementation_defined_restrictions pure-barriers}@anchor{1b8}
12440
@subsection Pure_Barriers
12443
@geindex Pure_Barriers
12445
[GNAT] This restriction ensures at compile time that protected entry
12446
barriers are restricted to:
12452
simple variables defined in the private part of the
12453
protected type/object,
12456
constant declarations,
12462
enumeration literals,
12471
character literals,
12474
implicitly defined comparison operators,
12477
uses of the Standard."not" operator,
12480
short-circuit operator
12483
This restriction is a relaxation of the Simple_Barriers restriction,
12484
but still ensures absence of side effects, exceptions, and recursion
12485
during the evaluation of the barriers.
12487
@node Simple_Barriers,Static_Priorities,Pure_Barriers,Partition-Wide Restrictions
12488
@anchor{gnat_rm/standard_and_implementation_defined_restrictions simple-barriers}@anchor{1b9}
12489
@subsection Simple_Barriers
12492
@geindex Simple_Barriers
12494
[RM D.7] This restriction ensures at compile time that barriers in entry
12495
declarations for protected types are restricted to either static boolean
12496
expressions or references to simple boolean variables defined in the private
12497
part of the protected type. No other form of entry barriers is permitted.
12499
@geindex Boolean_Entry_Barriers
12501
The restriction @cite{Boolean_Entry_Barriers} is recognized as a
12502
synonym for @cite{Simple_Barriers}. This is retained for historical
12503
compatibility purposes (and a warning will be generated for its use if
12504
warnings on obsolescent features are activated).
12506
@node Static_Priorities,Static_Storage_Size,Simple_Barriers,Partition-Wide Restrictions
12507
@anchor{gnat_rm/standard_and_implementation_defined_restrictions static-priorities}@anchor{1ba}
12508
@subsection Static_Priorities
12511
@geindex Static_Priorities
12513
[GNAT] This restriction ensures at compile time that all priority expressions
12514
are static, and that there are no dependences on the package
12515
@cite{Ada.Dynamic_Priorities}.
12517
@node Static_Storage_Size,,Static_Priorities,Partition-Wide Restrictions
12518
@anchor{gnat_rm/standard_and_implementation_defined_restrictions static-storage-size}@anchor{1bb}
12519
@subsection Static_Storage_Size
12522
@geindex Static_Storage_Size
12524
[GNAT] This restriction ensures at compile time that any expression appearing
12525
in a Storage_Size pragma or attribute definition clause is static.
12527
@node Program Unit Level Restrictions,,Partition-Wide Restrictions,Standard and Implementation Defined Restrictions
12528
@anchor{gnat_rm/standard_and_implementation_defined_restrictions program-unit-level-restrictions}@anchor{1bc}@anchor{gnat_rm/standard_and_implementation_defined_restrictions id3}@anchor{1bd}
12529
@section Program Unit Level Restrictions
12532
The second set of restriction identifiers
12533
does not require partition-wide consistency.
12534
The restriction may be enforced for a single
12535
compilation unit without any effect on any of the
12536
other compilation units in the partition.
12539
* No_Elaboration_Code::
12540
* No_Dynamic_Sized_Objects::
12542
* No_Implementation_Aspect_Specifications::
12543
* No_Implementation_Attributes::
12544
* No_Implementation_Identifiers::
12545
* No_Implementation_Pragmas::
12546
* No_Implementation_Restrictions::
12547
* No_Implementation_Units::
12548
* No_Implicit_Aliasing::
12549
* No_Obsolescent_Features::
12550
* No_Wide_Characters::
12555
@node No_Elaboration_Code,No_Dynamic_Sized_Objects,,Program Unit Level Restrictions
12556
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-elaboration-code}@anchor{1be}
12557
@subsection No_Elaboration_Code
12560
@geindex No_Elaboration_Code
12562
[GNAT] This restriction ensures at compile time that no elaboration code is
12563
generated. Note that this is not the same condition as is enforced
12564
by pragma @cite{Preelaborate}. There are cases in which pragma
12565
@cite{Preelaborate} still permits code to be generated (e.g., code
12566
to initialize a large array to all zeroes), and there are cases of units
12567
which do not meet the requirements for pragma @cite{Preelaborate},
12568
but for which no elaboration code is generated. Generally, it is
12569
the case that preelaborable units will meet the restrictions, with
12570
the exception of large aggregates initialized with an others_clause,
12571
and exception declarations (which generate calls to a run-time
12572
registry procedure). This restriction is enforced on
12573
a unit by unit basis, it need not be obeyed consistently
12574
throughout a partition.
12576
In the case of aggregates with others, if the aggregate has a dynamic
12577
size, there is no way to eliminate the elaboration code (such dynamic
12578
bounds would be incompatible with @cite{Preelaborate} in any case). If
12579
the bounds are static, then use of this restriction actually modifies
12580
the code choice of the compiler to avoid generating a loop, and instead
12581
generate the aggregate statically if possible, no matter how many times
12582
the data for the others clause must be repeatedly generated.
12584
It is not possible to precisely document
12585
the constructs which are compatible with this restriction, since,
12586
unlike most other restrictions, this is not a restriction on the
12587
source code, but a restriction on the generated object code. For
12588
example, if the source contains a declaration:
12591
Val : constant Integer := X;
12594
where X is not a static constant, it may be possible, depending
12595
on complex optimization circuitry, for the compiler to figure
12596
out the value of X at compile time, in which case this initialization
12597
can be done by the loader, and requires no initialization code. It
12598
is not possible to document the precise conditions under which the
12599
optimizer can figure this out.
12601
Note that this the implementation of this restriction requires full
12602
code generation. If it is used in conjunction with "semantics only"
12603
checking, then some cases of violations may be missed.
12605
@node No_Dynamic_Sized_Objects,No_Entry_Queue,No_Elaboration_Code,Program Unit Level Restrictions
12606
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-dynamic-sized-objects}@anchor{1bf}
12607
@subsection No_Dynamic_Sized_Objects
12610
@geindex No_Dynamic_Sized_Objects
12612
[GNAT] This restriction disallows certain constructs that might lead to the
12613
creation of dynamic-sized composite objects (or array or discriminated type).
12614
An array subtype indication is illegal if the bounds are not static
12615
or references to discriminants of an enclosing type.
12616
A discriminated subtype indication is illegal if the type has
12617
discriminant-dependent array components or a variant part, and the
12618
discriminants are not static. In addition, array and record aggregates are
12619
illegal in corresponding cases. Note that this restriction does not forbid
12620
access discriminants. It is often a good idea to combine this restriction
12621
with No_Secondary_Stack.
12623
@node No_Entry_Queue,No_Implementation_Aspect_Specifications,No_Dynamic_Sized_Objects,Program Unit Level Restrictions
12624
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-entry-queue}@anchor{1c0}
12625
@subsection No_Entry_Queue
12628
@geindex No_Entry_Queue
12630
[GNAT] This restriction is a declaration that any protected entry compiled in
12631
the scope of the restriction has at most one task waiting on the entry
12632
at any one time, and so no queue is required. This restriction is not
12633
checked at compile time. A program execution is erroneous if an attempt
12634
is made to queue a second task on such an entry.
12636
@node No_Implementation_Aspect_Specifications,No_Implementation_Attributes,No_Entry_Queue,Program Unit Level Restrictions
12637
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implementation-aspect-specifications}@anchor{1c1}
12638
@subsection No_Implementation_Aspect_Specifications
12641
@geindex No_Implementation_Aspect_Specifications
12643
[RM 13.12.1] This restriction checks at compile time that no
12644
GNAT-defined aspects are present. With this restriction, the only
12645
aspects that can be used are those defined in the Ada Reference Manual.
12647
@node No_Implementation_Attributes,No_Implementation_Identifiers,No_Implementation_Aspect_Specifications,Program Unit Level Restrictions
12648
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implementation-attributes}@anchor{1c2}
12649
@subsection No_Implementation_Attributes
12652
@geindex No_Implementation_Attributes
12654
[RM 13.12.1] This restriction checks at compile time that no
12655
GNAT-defined attributes are present. With this restriction, the only
12656
attributes that can be used are those defined in the Ada Reference
12659
@node No_Implementation_Identifiers,No_Implementation_Pragmas,No_Implementation_Attributes,Program Unit Level Restrictions
12660
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implementation-identifiers}@anchor{1c3}
12661
@subsection No_Implementation_Identifiers
12664
@geindex No_Implementation_Identifiers
12666
[RM 13.12.1] This restriction checks at compile time that no
12667
implementation-defined identifiers (marked with pragma Implementation_Defined)
12668
occur within language-defined packages.
12670
@node No_Implementation_Pragmas,No_Implementation_Restrictions,No_Implementation_Identifiers,Program Unit Level Restrictions
12671
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implementation-pragmas}@anchor{1c4}
12672
@subsection No_Implementation_Pragmas
12675
@geindex No_Implementation_Pragmas
12677
[RM 13.12.1] This restriction checks at compile time that no
12678
GNAT-defined pragmas are present. With this restriction, the only
12679
pragmas that can be used are those defined in the Ada Reference Manual.
12681
@node No_Implementation_Restrictions,No_Implementation_Units,No_Implementation_Pragmas,Program Unit Level Restrictions
12682
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implementation-restrictions}@anchor{1c5}
12683
@subsection No_Implementation_Restrictions
12686
@geindex No_Implementation_Restrictions
12688
[GNAT] This restriction checks at compile time that no GNAT-defined restriction
12689
identifiers (other than @cite{No_Implementation_Restrictions} itself)
12690
are present. With this restriction, the only other restriction identifiers
12691
that can be used are those defined in the Ada Reference Manual.
12693
@node No_Implementation_Units,No_Implicit_Aliasing,No_Implementation_Restrictions,Program Unit Level Restrictions
12694
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implementation-units}@anchor{1c6}
12695
@subsection No_Implementation_Units
12698
@geindex No_Implementation_Units
12700
[RM 13.12.1] This restriction checks at compile time that there is no
12701
mention in the context clause of any implementation-defined descendants
12702
of packages Ada, Interfaces, or System.
12704
@node No_Implicit_Aliasing,No_Obsolescent_Features,No_Implementation_Units,Program Unit Level Restrictions
12705
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-implicit-aliasing}@anchor{1c7}
12706
@subsection No_Implicit_Aliasing
12709
@geindex No_Implicit_Aliasing
12711
[GNAT] This restriction, which is not required to be partition-wide consistent,
12712
requires an explicit aliased keyword for an object to which 'Access,
12713
'Unchecked_Access, or 'Address is applied, and forbids entirely the use of
12714
the 'Unrestricted_Access attribute for objects. Note: the reason that
12715
Unrestricted_Access is forbidden is that it would require the prefix
12716
to be aliased, and in such cases, it can always be replaced by
12717
the standard attribute Unchecked_Access which is preferable.
12719
@node No_Obsolescent_Features,No_Wide_Characters,No_Implicit_Aliasing,Program Unit Level Restrictions
12720
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-obsolescent-features}@anchor{1c8}
12721
@subsection No_Obsolescent_Features
12724
@geindex No_Obsolescent_Features
12726
[RM 13.12.1] This restriction checks at compile time that no obsolescent
12727
features are used, as defined in Annex J of the Ada Reference Manual.
12729
@node No_Wide_Characters,SPARK_05,No_Obsolescent_Features,Program Unit Level Restrictions
12730
@anchor{gnat_rm/standard_and_implementation_defined_restrictions no-wide-characters}@anchor{1c9}
12731
@subsection No_Wide_Characters
12734
@geindex No_Wide_Characters
12736
[GNAT] This restriction ensures at compile time that no uses of the types
12737
@cite{Wide_Character} or @cite{Wide_String} or corresponding wide
12739
appear, and that no wide or wide wide string or character literals
12740
appear in the program (that is literals representing characters not in
12741
type @cite{Character}).
12743
@node SPARK_05,,No_Wide_Characters,Program Unit Level Restrictions
12744
@anchor{gnat_rm/standard_and_implementation_defined_restrictions spark-05}@anchor{1ca}
12745
@subsection SPARK_05
12750
[GNAT] This restriction checks at compile time that some constructs
12751
forbidden in SPARK 2005 are not present. Error messages related to
12752
SPARK restriction have the form:
12755
violation of restriction "SPARK_05" at <source-location>
12761
The restriction @cite{SPARK} is recognized as a
12762
synonym for @cite{SPARK_05}. This is retained for historical
12763
compatibility purposes (and an unconditional warning will be generated
12764
for its use, advising replacement by @cite{SPARK}).
12766
This is not a replacement for the semantic checks performed by the
12767
SPARK Examiner tool, as the compiler currently only deals with code,
12768
not SPARK 2005 annotations, and does not guarantee catching all
12769
cases of constructs forbidden by SPARK 2005.
12771
Thus it may well be the case that code which passes the compiler with
12772
the SPARK restriction is rejected by the SPARK Examiner, e.g. due to
12773
the different visibility rules of the Examiner based on SPARK 2005
12774
@cite{inherit} annotations.
12776
This restriction can be useful in providing an initial filter for code
12777
developed using SPARK 2005, or in examining legacy code to see how far
12778
it is from meeting SPARK restrictions.
12780
The list below summarizes the checks that are performed when this
12781
restriction is in force:
12787
No block statements
12790
No case statements with only an others clause
12793
Exit statements in loops must respect the SPARK 2005 language restrictions
12799
Return can only appear as last statement in function
12802
Function must have return statement
12805
Loop parameter specification must include subtype mark
12808
Prefix of expanded name cannot be a loop statement
12811
Abstract subprogram not allowed
12814
User-defined operators not allowed
12817
Access type parameters not allowed
12820
Default expressions for parameters not allowed
12823
Default expressions for record fields not allowed
12826
No tasking constructs allowed
12829
Label needed at end of subprograms and packages
12832
No mixing of positional and named parameter association
12835
No access types as result type
12838
No unconstrained arrays as result types
12844
Initial and later declarations must be in correct order (declaration can't come after body)
12847
No attributes on private types if full declaration not visible
12850
No package declaration within package specification
12853
No controlled types
12856
No discriminant types
12862
Selector name cannot be operator symbol (i.e. operator symbol cannot be prefixed)
12865
Access attribute not allowed
12868
Allocator not allowed
12871
Result of catenation must be String
12874
Operands of catenation must be string literal, static char or another catenation
12877
No conditional expressions
12880
No explicit dereference
12883
Quantified expression not allowed
12886
Slicing not allowed
12889
No exception renaming
12892
No generic renaming
12901
Aggregates must be qualified
12904
Nonstatic choice in array aggregates not allowed
12907
The only view conversions which are allowed as in-out parameters are conversions of a tagged type to an ancestor type
12910
No mixing of positional and named association in aggregate, no multi choice
12913
AND, OR and XOR for arrays only allowed when operands have same static bounds
12916
Fixed point operands to * or / must be qualified or converted
12919
Comparison operators not allowed for Booleans or arrays (except strings)
12922
Equality not allowed for arrays with non-matching static bounds (except strings)
12925
Conversion / qualification not allowed for arrays with non-matching static bounds
12928
Subprogram declaration only allowed in package spec (unless followed by import)
12931
Access types not allowed
12934
Incomplete type declaration not allowed
12937
Object and subtype declarations must respect SPARK restrictions
12940
Digits or delta constraint not allowed
12943
Decimal fixed point type not allowed
12946
Aliasing of objects not allowed
12949
Modular type modulus must be power of 2
12952
Base not allowed on subtype mark
12955
Unary operators not allowed on modular types (except not)
12958
Untagged record cannot be null
12961
No class-wide operations
12964
Initialization expressions must respect SPARK restrictions
12967
Nonstatic ranges not allowed except in iteration schemes
12970
String subtypes must have lower bound of 1
12973
Subtype of Boolean cannot have constraint
12976
At most one tagged type or extension per package
12979
Interface is not allowed
12982
Character literal cannot be prefixed (selector name cannot be character literal)
12985
Record aggregate cannot contain 'others'
12988
Component association in record aggregate must contain a single choice
12991
Ancestor part cannot be a type mark
12994
Attributes 'Image, 'Width and 'Value not allowed
12997
Functions may not update globals
13000
Subprograms may not contain direct calls to themselves (prevents recursion within unit)
13003
Call to subprogram not allowed in same unit before body has been seen (prevents recursion within unit)
13006
The following restrictions are enforced, but note that they are actually more
13007
strict that the latest SPARK 2005 language definition:
13013
No derived types other than tagged type extensions
13016
Subtype of unconstrained array must have constraint
13019
This list summarises the main SPARK 2005 language rules that are not
13020
currently checked by the SPARK_05 restriction:
13026
SPARK annotations are treated as comments so are not checked at all
13029
Based real literals not allowed
13032
Objects cannot be initialized at declaration by calls to user-defined functions
13035
Objects cannot be initialized at declaration by assignments from variables
13038
Objects cannot be initialized at declaration by assignments from indexed/selected components
13041
Ranges shall not be null
13044
A fixed point delta expression must be a simple expression
13047
Restrictions on where renaming declarations may be placed
13050
Externals of mode 'out' cannot be referenced
13053
Externals of mode 'in' cannot be updated
13056
Loop with no iteration scheme or exits only allowed as last statement in main program or task
13059
Subprogram cannot have parent unit name
13062
SPARK 2005 inherited subprogram must be prefixed with overriding
13065
External variables (or functions that reference them) may not be passed as actual parameters
13068
Globals must be explicitly mentioned in contract
13071
Deferred constants cannot be completed by pragma Import
13074
Package initialization cannot read/write variables from other packages
13077
Prefix not allowed for entities that are directly visible
13080
Identifier declaration can't override inherited package name
13083
Cannot use Standard or other predefined packages as identifiers
13086
After renaming, cannot use the original name
13089
Subprograms can only be renamed to remove package prefix
13092
Pragma import must be immediately after entity it names
13095
No mutual recursion between multiple units (this can be checked with gnatcheck)
13098
Note that if a unit is compiled in Ada 95 mode with the SPARK restriction,
13099
violations will be reported for constructs forbidden in SPARK 95,
13100
instead of SPARK 2005.
13102
@node Implementation Advice,Implementation Defined Characteristics,Standard and Implementation Defined Restrictions,Top
13103
@anchor{gnat_rm/implementation_advice doc}@anchor{1cb}@anchor{gnat_rm/implementation_advice implementation-advice}@anchor{a}@anchor{gnat_rm/implementation_advice id1}@anchor{1cc}
13104
@chapter Implementation Advice
13107
The main text of the Ada Reference Manual describes the required
13108
behavior of all Ada compilers, and the GNAT compiler conforms to
13109
these requirements.
13111
In addition, there are sections throughout the Ada Reference Manual headed
13112
by the phrase 'Implementation advice'. These sections are not normative,
13113
i.e., they do not specify requirements that all compilers must
13114
follow. Rather they provide advice on generally desirable behavior.
13115
They are not requirements, because they describe behavior that cannot
13116
be provided on all systems, or may be undesirable on some systems.
13118
As far as practical, GNAT follows the implementation advice in
13119
the Ada Reference Manual. Each such RM section corresponds to a section
13120
in this chapter whose title specifies the
13121
RM section number and paragraph number and the subject of
13122
the advice. The contents of each section consists of the RM text within
13124
followed by the GNAT interpretation of the advice. Most often, this simply says
13125
'followed', which means that GNAT follows the advice. However, in a
13126
number of cases, GNAT deliberately deviates from this advice, in which
13127
case the text describes what GNAT does and why.
13129
@geindex Error detection
13132
* RM 1.1.3(20); Error Detection: RM 1 1 3 20 Error Detection.
13133
* RM 1.1.3(31); Child Units: RM 1 1 3 31 Child Units.
13134
* RM 1.1.5(12); Bounded Errors: RM 1 1 5 12 Bounded Errors.
13135
* RM 2.8(16); Pragmas: RM 2 8 16 Pragmas.
13136
* RM 2.8(17-19); Pragmas: RM 2 8 17-19 Pragmas.
13137
* RM 3.5.2(5); Alternative Character Sets: RM 3 5 2 5 Alternative Character Sets.
13138
* RM 3.5.4(28); Integer Types: RM 3 5 4 28 Integer Types.
13139
* RM 3.5.4(29); Integer Types: RM 3 5 4 29 Integer Types.
13140
* RM 3.5.5(8); Enumeration Values: RM 3 5 5 8 Enumeration Values.
13141
* RM 3.5.7(17); Float Types: RM 3 5 7 17 Float Types.
13142
* RM 3.6.2(11); Multidimensional Arrays: RM 3 6 2 11 Multidimensional Arrays.
13143
* RM 9.6(30-31); Duration'Small: RM 9 6 30-31 Duration'Small.
13144
* RM 10.2.1(12); Consistent Representation: RM 10 2 1 12 Consistent Representation.
13145
* RM 11.4.1(19); Exception Information: RM 11 4 1 19 Exception Information.
13146
* RM 11.5(28); Suppression of Checks: RM 11 5 28 Suppression of Checks.
13147
* RM 13.1 (21-24); Representation Clauses: RM 13 1 21-24 Representation Clauses.
13148
* RM 13.2(6-8); Packed Types: RM 13 2 6-8 Packed Types.
13149
* RM 13.3(14-19); Address Clauses: RM 13 3 14-19 Address Clauses.
13150
* RM 13.3(29-35); Alignment Clauses: RM 13 3 29-35 Alignment Clauses.
13151
* RM 13.3(42-43); Size Clauses: RM 13 3 42-43 Size Clauses.
13152
* RM 13.3(50-56); Size Clauses: RM 13 3 50-56 Size Clauses.
13153
* RM 13.3(71-73); Component Size Clauses: RM 13 3 71-73 Component Size Clauses.
13154
* RM 13.4(9-10); Enumeration Representation Clauses: RM 13 4 9-10 Enumeration Representation Clauses.
13155
* RM 13.5.1(17-22); Record Representation Clauses: RM 13 5 1 17-22 Record Representation Clauses.
13156
* RM 13.5.2(5); Storage Place Attributes: RM 13 5 2 5 Storage Place Attributes.
13157
* RM 13.5.3(7-8); Bit Ordering: RM 13 5 3 7-8 Bit Ordering.
13158
* RM 13.7(37); Address as Private: RM 13 7 37 Address as Private.
13159
* RM 13.7.1(16); Address Operations: RM 13 7 1 16 Address Operations.
13160
* RM 13.9(14-17); Unchecked Conversion: RM 13 9 14-17 Unchecked Conversion.
13161
* RM 13.11(23-25); Implicit Heap Usage: RM 13 11 23-25 Implicit Heap Usage.
13162
* RM 13.11.2(17); Unchecked Deallocation: RM 13 11 2 17 Unchecked Deallocation.
13163
* RM 13.13.2(17); Stream Oriented Attributes: RM 13 13 2 17 Stream Oriented Attributes.
13164
* RM A.1(52); Names of Predefined Numeric Types: RM A 1 52 Names of Predefined Numeric Types.
13165
* RM A.3.2(49); Ada.Characters.Handling: RM A 3 2 49 Ada Characters Handling.
13166
* RM A.4.4(106); Bounded-Length String Handling: RM A 4 4 106 Bounded-Length String Handling.
13167
* RM A.5.2(46-47); Random Number Generation: RM A 5 2 46-47 Random Number Generation.
13168
* RM A.10.7(23); Get_Immediate: RM A 10 7 23 Get_Immediate.
13169
* RM B.1(39-41); Pragma Export: RM B 1 39-41 Pragma Export.
13170
* RM B.2(12-13); Package Interfaces: RM B 2 12-13 Package Interfaces.
13171
* RM B.3(63-71); Interfacing with C: RM B 3 63-71 Interfacing with C.
13172
* RM B.4(95-98); Interfacing with COBOL: RM B 4 95-98 Interfacing with COBOL.
13173
* RM B.5(22-26); Interfacing with Fortran: RM B 5 22-26 Interfacing with Fortran.
13174
* RM C.1(3-5); Access to Machine Operations: RM C 1 3-5 Access to Machine Operations.
13175
* RM C.1(10-16); Access to Machine Operations: RM C 1 10-16 Access to Machine Operations.
13176
* RM C.3(28); Interrupt Support: RM C 3 28 Interrupt Support.
13177
* RM C.3.1(20-21); Protected Procedure Handlers: RM C 3 1 20-21 Protected Procedure Handlers.
13178
* RM C.3.2(25); Package Interrupts: RM C 3 2 25 Package Interrupts.
13179
* RM C.4(14); Pre-elaboration Requirements: RM C 4 14 Pre-elaboration Requirements.
13180
* RM C.5(8); Pragma Discard_Names: RM C 5 8 Pragma Discard_Names.
13181
* RM C.7.2(30); The Package Task_Attributes: RM C 7 2 30 The Package Task_Attributes.
13182
* RM D.3(17); Locking Policies: RM D 3 17 Locking Policies.
13183
* RM D.4(16); Entry Queuing Policies: RM D 4 16 Entry Queuing Policies.
13184
* RM D.6(9-10); Preemptive Abort: RM D 6 9-10 Preemptive Abort.
13185
* RM D.7(21); Tasking Restrictions: RM D 7 21 Tasking Restrictions.
13186
* RM D.8(47-49); Monotonic Time: RM D 8 47-49 Monotonic Time.
13187
* RM E.5(28-29); Partition Communication Subsystem: RM E 5 28-29 Partition Communication Subsystem.
13188
* RM F(7); COBOL Support: RM F 7 COBOL Support.
13189
* RM F.1(2); Decimal Radix Support: RM F 1 2 Decimal Radix Support.
13190
* RM G; Numerics: RM G Numerics.
13191
* RM G.1.1(56-58); Complex Types: RM G 1 1 56-58 Complex Types.
13192
* RM G.1.2(49); Complex Elementary Functions: RM G 1 2 49 Complex Elementary Functions.
13193
* RM G.2.4(19); Accuracy Requirements: RM G 2 4 19 Accuracy Requirements.
13194
* RM G.2.6(15); Complex Arithmetic Accuracy: RM G 2 6 15 Complex Arithmetic Accuracy.
13195
* RM H.6(15/2); Pragma Partition_Elaboration_Policy: RM H 6 15/2 Pragma Partition_Elaboration_Policy.
13199
@node RM 1 1 3 20 Error Detection,RM 1 1 3 31 Child Units,,Implementation Advice
13200
@anchor{gnat_rm/implementation_advice rm-1-1-3-20-error-detection}@anchor{1cd}
13201
@section RM 1.1.3(20): Error Detection
13206
"If an implementation detects the use of an unsupported Specialized Needs
13207
Annex feature at run time, it should raise @cite{Program_Error} if
13211
Not relevant. All specialized needs annex features are either supported,
13212
or diagnosed at compile time.
13214
@geindex Child Units
13216
@node RM 1 1 3 31 Child Units,RM 1 1 5 12 Bounded Errors,RM 1 1 3 20 Error Detection,Implementation Advice
13217
@anchor{gnat_rm/implementation_advice rm-1-1-3-31-child-units}@anchor{1ce}
13218
@section RM 1.1.3(31): Child Units
13223
"If an implementation wishes to provide implementation-defined
13224
extensions to the functionality of a language-defined library unit, it
13225
should normally do so by adding children to the library unit."
13230
@geindex Bounded errors
13232
@node RM 1 1 5 12 Bounded Errors,RM 2 8 16 Pragmas,RM 1 1 3 31 Child Units,Implementation Advice
13233
@anchor{gnat_rm/implementation_advice rm-1-1-5-12-bounded-errors}@anchor{1cf}
13234
@section RM 1.1.5(12): Bounded Errors
13239
"If an implementation detects a bounded error or erroneous
13240
execution, it should raise @cite{Program_Error}."
13243
Followed in all cases in which the implementation detects a bounded
13244
error or erroneous execution. Not all such situations are detected at
13249
@node RM 2 8 16 Pragmas,RM 2 8 17-19 Pragmas,RM 1 1 5 12 Bounded Errors,Implementation Advice
13250
@anchor{gnat_rm/implementation_advice id2}@anchor{1d0}@anchor{gnat_rm/implementation_advice rm-2-8-16-pragmas}@anchor{1d1}
13251
@section RM 2.8(16): Pragmas
13256
"Normally, implementation-defined pragmas should have no semantic effect
13257
for error-free programs; that is, if the implementation-defined pragmas
13258
are removed from a working program, the program should still be legal,
13259
and should still have the same semantics."
13262
The following implementation defined pragmas are exceptions to this
13266
@multitable {xxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxx}
13309
@emph{CPP_Constructor}
13325
@emph{Interface_Name}
13333
@emph{Machine_Attribute}
13341
@emph{Unimplemented_Unit}
13349
@emph{Unchecked_Union}
13358
In each of the above cases, it is essential to the purpose of the pragma
13359
that this advice not be followed. For details see
13360
@ref{7,,Implementation Defined Pragmas}.
13362
@node RM 2 8 17-19 Pragmas,RM 3 5 2 5 Alternative Character Sets,RM 2 8 16 Pragmas,Implementation Advice
13363
@anchor{gnat_rm/implementation_advice rm-2-8-17-19-pragmas}@anchor{1d2}
13364
@section RM 2.8(17-19): Pragmas
13369
"Normally, an implementation should not define pragmas that can
13370
make an illegal program legal, except as follows:
13376
A pragma used to complete a declaration, such as a pragma @cite{Import};
13379
A pragma used to configure the environment by adding, removing, or
13380
replacing @cite{library_items}."
13384
See @ref{1d1,,RM 2.8(16); Pragmas}.
13386
@geindex Character Sets
13388
@geindex Alternative Character Sets
13390
@node RM 3 5 2 5 Alternative Character Sets,RM 3 5 4 28 Integer Types,RM 2 8 17-19 Pragmas,Implementation Advice
13391
@anchor{gnat_rm/implementation_advice rm-3-5-2-5-alternative-character-sets}@anchor{1d3}
13392
@section RM 3.5.2(5): Alternative Character Sets
13397
"If an implementation supports a mode with alternative interpretations
13398
for @cite{Character} and @cite{Wide_Character}, the set of graphic
13399
characters of @cite{Character} should nevertheless remain a proper
13400
subset of the set of graphic characters of @cite{Wide_Character}. Any
13401
character set 'localizations' should be reflected in the results of
13402
the subprograms defined in the language-defined package
13403
@cite{Characters.Handling} (see A.3) available in such a mode. In a mode with
13404
an alternative interpretation of @cite{Character}, the implementation should
13405
also support a corresponding change in what is a legal
13406
@cite{identifier_letter}."
13409
Not all wide character modes follow this advice, in particular the JIS
13410
and IEC modes reflect standard usage in Japan, and in these encoding,
13411
the upper half of the Latin-1 set is not part of the wide-character
13412
subset, since the most significant bit is used for wide character
13413
encoding. However, this only applies to the external forms. Internally
13414
there is no such restriction.
13416
@geindex Integer types
13418
@node RM 3 5 4 28 Integer Types,RM 3 5 4 29 Integer Types,RM 3 5 2 5 Alternative Character Sets,Implementation Advice
13419
@anchor{gnat_rm/implementation_advice rm-3-5-4-28-integer-types}@anchor{1d4}
13420
@section RM 3.5.4(28): Integer Types
13425
"An implementation should support @cite{Long_Integer} in addition to
13426
@cite{Integer} if the target machine supports 32-bit (or longer)
13427
arithmetic. No other named integer subtypes are recommended for package
13428
@cite{Standard}. Instead, appropriate named integer subtypes should be
13429
provided in the library package @cite{Interfaces} (see B.2)."
13432
@cite{Long_Integer} is supported. Other standard integer types are supported
13433
so this advice is not fully followed. These types
13434
are supported for convenient interface to C, and so that all hardware
13435
types of the machine are easily available.
13437
@node RM 3 5 4 29 Integer Types,RM 3 5 5 8 Enumeration Values,RM 3 5 4 28 Integer Types,Implementation Advice
13438
@anchor{gnat_rm/implementation_advice rm-3-5-4-29-integer-types}@anchor{1d5}
13439
@section RM 3.5.4(29): Integer Types
13444
"An implementation for a two's complement machine should support
13445
modular types with a binary modulus up to @code{System.Max_Int*2+2}. An
13446
implementation should support a non-binary modules up to @cite{Integer'Last}."
13451
@geindex Enumeration values
13453
@node RM 3 5 5 8 Enumeration Values,RM 3 5 7 17 Float Types,RM 3 5 4 29 Integer Types,Implementation Advice
13454
@anchor{gnat_rm/implementation_advice rm-3-5-5-8-enumeration-values}@anchor{1d6}
13455
@section RM 3.5.5(8): Enumeration Values
13460
"For the evaluation of a call on @code{S'Pos} for an enumeration
13461
subtype, if the value of the operand does not correspond to the internal
13462
code for any enumeration literal of its type (perhaps due to an
13463
un-initialized variable), then the implementation should raise
13464
@cite{Program_Error}. This is particularly important for enumeration
13465
types with noncontiguous internal codes specified by an
13466
enumeration_representation_clause."
13471
@geindex Float types
13473
@node RM 3 5 7 17 Float Types,RM 3 6 2 11 Multidimensional Arrays,RM 3 5 5 8 Enumeration Values,Implementation Advice
13474
@anchor{gnat_rm/implementation_advice rm-3-5-7-17-float-types}@anchor{1d7}
13475
@section RM 3.5.7(17): Float Types
13480
"An implementation should support @cite{Long_Float} in addition to
13481
@cite{Float} if the target machine supports 11 or more digits of
13482
precision. No other named floating point subtypes are recommended for
13483
package @cite{Standard}. Instead, appropriate named floating point subtypes
13484
should be provided in the library package @cite{Interfaces} (see B.2)."
13487
@cite{Short_Float} and @cite{Long_Long_Float} are also provided. The
13488
former provides improved compatibility with other implementations
13489
supporting this type. The latter corresponds to the highest precision
13490
floating-point type supported by the hardware. On most machines, this
13491
will be the same as @cite{Long_Float}, but on some machines, it will
13492
correspond to the IEEE extended form. The notable case is all ia32
13493
(x86) implementations, where @cite{Long_Long_Float} corresponds to
13494
the 80-bit extended precision format supported in hardware on this
13495
processor. Note that the 128-bit format on SPARC is not supported,
13496
since this is a software rather than a hardware format.
13498
@geindex Multidimensional arrays
13501
@geindex multidimensional
13503
@node RM 3 6 2 11 Multidimensional Arrays,RM 9 6 30-31 Duration'Small,RM 3 5 7 17 Float Types,Implementation Advice
13504
@anchor{gnat_rm/implementation_advice rm-3-6-2-11-multidimensional-arrays}@anchor{1d8}
13505
@section RM 3.6.2(11): Multidimensional Arrays
13510
"An implementation should normally represent multidimensional arrays in
13511
row-major order, consistent with the notation used for multidimensional
13512
array aggregates (see 4.3.3). However, if a pragma @cite{Convention}
13513
(@cite{Fortran}, ...) applies to a multidimensional array type, then
13514
column-major order should be used instead (see B.5, @cite{Interfacing with Fortran})."
13519
@geindex Duration'Small
13521
@node RM 9 6 30-31 Duration'Small,RM 10 2 1 12 Consistent Representation,RM 3 6 2 11 Multidimensional Arrays,Implementation Advice
13522
@anchor{gnat_rm/implementation_advice rm-9-6-30-31-duration-small}@anchor{1d9}
13523
@section RM 9.6(30-31): Duration'Small
13528
"Whenever possible in an implementation, the value of @cite{Duration'Small}
13529
should be no greater than 100 microseconds."
13532
Followed. (@cite{Duration'Small} = 10**(-9)).
13536
"The time base for @cite{delay_relative_statements} should be monotonic;
13537
it need not be the same time base as used for @cite{Calendar.Clock}."
13542
@node RM 10 2 1 12 Consistent Representation,RM 11 4 1 19 Exception Information,RM 9 6 30-31 Duration'Small,Implementation Advice
13543
@anchor{gnat_rm/implementation_advice rm-10-2-1-12-consistent-representation}@anchor{1da}
13544
@section RM 10.2.1(12): Consistent Representation
13549
"In an implementation, a type declared in a pre-elaborated package should
13550
have the same representation in every elaboration of a given version of
13551
the package, whether the elaborations occur in distinct executions of
13552
the same program, or in executions of distinct programs or partitions
13553
that include the given version."
13556
Followed, except in the case of tagged types. Tagged types involve
13557
implicit pointers to a local copy of a dispatch table, and these pointers
13558
have representations which thus depend on a particular elaboration of the
13559
package. It is not easy to see how it would be possible to follow this
13560
advice without severely impacting efficiency of execution.
13562
@geindex Exception information
13564
@node RM 11 4 1 19 Exception Information,RM 11 5 28 Suppression of Checks,RM 10 2 1 12 Consistent Representation,Implementation Advice
13565
@anchor{gnat_rm/implementation_advice rm-11-4-1-19-exception-information}@anchor{1db}
13566
@section RM 11.4.1(19): Exception Information
13571
"@cite{Exception_Message} by default and @cite{Exception_Information}
13572
should produce information useful for
13573
debugging. @cite{Exception_Message} should be short, about one
13574
line. @cite{Exception_Information} can be long. @cite{Exception_Message}
13575
should not include the
13576
@cite{Exception_Name}. @cite{Exception_Information} should include both
13577
the @cite{Exception_Name} and the @cite{Exception_Message}."
13580
Followed. For each exception that doesn't have a specified
13581
@cite{Exception_Message}, the compiler generates one containing the location
13582
of the raise statement. This location has the form 'file_name:line', where
13583
file_name is the short file name (without path information) and line is the line
13584
number in the file. Note that in the case of the Zero Cost Exception
13585
mechanism, these messages become redundant with the Exception_Information that
13586
contains a full backtrace of the calling sequence, so they are disabled.
13587
To disable explicitly the generation of the source location message, use the
13588
Pragma @cite{Discard_Names}.
13590
@geindex Suppression of checks
13593
@geindex suppression of
13595
@node RM 11 5 28 Suppression of Checks,RM 13 1 21-24 Representation Clauses,RM 11 4 1 19 Exception Information,Implementation Advice
13596
@anchor{gnat_rm/implementation_advice rm-11-5-28-suppression-of-checks}@anchor{1dc}
13597
@section RM 11.5(28): Suppression of Checks
13602
"The implementation should minimize the code executed for checks that
13603
have been suppressed."
13608
@geindex Representation clauses
13610
@node RM 13 1 21-24 Representation Clauses,RM 13 2 6-8 Packed Types,RM 11 5 28 Suppression of Checks,Implementation Advice
13611
@anchor{gnat_rm/implementation_advice rm-13-1-21-24-representation-clauses}@anchor{1dd}
13612
@section RM 13.1 (21-24): Representation Clauses
13617
"The recommended level of support for all representation items is
13618
qualified as follows:
13620
An implementation need not support representation items containing
13621
nonstatic expressions, except that an implementation should support a
13622
representation item for a given entity if each nonstatic expression in
13623
the representation item is a name that statically denotes a constant
13624
declared before the entity."
13627
Followed. In fact, GNAT goes beyond the recommended level of support
13628
by allowing nonstatic expressions in some representation clauses even
13629
without the need to declare constants initialized with the values of
13636
for Y'Address use X'Address;>>
13639
"An implementation need not support a specification for the `Size`
13640
for a given composite subtype, nor the size or storage place for an
13641
object (including a component) of a given composite subtype, unless the
13642
constraints on the subtype and its composite subcomponents (if any) are
13643
all static constraints."
13646
Followed. Size Clauses are not permitted on nonstatic components, as
13651
"An aliased component, or a component whose type is by-reference, should
13652
always be allocated at an addressable location."
13657
@geindex Packed types
13659
@node RM 13 2 6-8 Packed Types,RM 13 3 14-19 Address Clauses,RM 13 1 21-24 Representation Clauses,Implementation Advice
13660
@anchor{gnat_rm/implementation_advice rm-13-2-6-8-packed-types}@anchor{1de}
13661
@section RM 13.2(6-8): Packed Types
13666
"If a type is packed, then the implementation should try to minimize
13667
storage allocated to objects of the type, possibly at the expense of
13668
speed of accessing components, subject to reasonable complexity in
13669
addressing calculations.
13671
The recommended level of support pragma @cite{Pack} is:
13673
For a packed record type, the components should be packed as tightly as
13674
possible subject to the Sizes of the component subtypes, and subject to
13675
any @cite{record_representation_clause} that applies to the type; the
13676
implementation may, but need not, reorder components or cross aligned
13677
word boundaries to improve the packing. A component whose @cite{Size} is
13678
greater than the word size may be allocated an integral number of words."
13681
Followed. Tight packing of arrays is supported for all component sizes
13682
up to 64-bits. If the array component size is 1 (that is to say, if
13683
the component is a boolean type or an enumeration type with two values)
13684
then values of the type are implicitly initialized to zero. This
13685
happens both for objects of the packed type, and for objects that have a
13686
subcomponent of the packed type.
13690
"An implementation should support Address clauses for imported
13696
@geindex Address clauses
13698
@node RM 13 3 14-19 Address Clauses,RM 13 3 29-35 Alignment Clauses,RM 13 2 6-8 Packed Types,Implementation Advice
13699
@anchor{gnat_rm/implementation_advice rm-13-3-14-19-address-clauses}@anchor{1df}
13700
@section RM 13.3(14-19): Address Clauses
13705
"For an array @cite{X}, @code{X'Address} should point at the first
13706
component of the array, and not at the array bounds."
13713
"The recommended level of support for the @cite{Address} attribute is:
13715
@code{X'Address} should produce a useful result if @cite{X} is an
13716
object that is aliased or of a by-reference type, or is an entity whose
13717
@cite{Address} has been specified."
13720
Followed. A valid address will be produced even if none of those
13721
conditions have been met. If necessary, the object is forced into
13722
memory to ensure the address is valid.
13726
"An implementation should support @cite{Address} clauses for imported
13734
"Objects (including subcomponents) that are aliased or of a by-reference
13735
type should be allocated on storage element boundaries."
13742
"If the @cite{Address} of an object is specified, or it is imported or exported,
13743
then the implementation should not perform optimizations based on
13744
assumptions of no aliases."
13749
@geindex Alignment clauses
13751
@node RM 13 3 29-35 Alignment Clauses,RM 13 3 42-43 Size Clauses,RM 13 3 14-19 Address Clauses,Implementation Advice
13752
@anchor{gnat_rm/implementation_advice rm-13-3-29-35-alignment-clauses}@anchor{1e0}
13753
@section RM 13.3(29-35): Alignment Clauses
13758
"The recommended level of support for the @cite{Alignment} attribute for
13761
An implementation should support specified Alignments that are factors
13762
and multiples of the number of storage elements per word, subject to the
13770
"An implementation need not support specified Alignments for
13771
combinations of Sizes and Alignments that cannot be easily
13772
loaded and stored by available machine instructions."
13779
"An implementation need not support specified Alignments that are
13780
greater than the maximum @cite{Alignment} the implementation ever returns by
13788
"The recommended level of support for the @cite{Alignment} attribute for
13791
Same as above, for subtypes, but in addition:"
13798
"For stand-alone library-level objects of statically constrained
13799
subtypes, the implementation should support all alignments
13800
supported by the target linker. For example, page alignment is likely to
13801
be supported for such objects, but not for subtypes."
13806
@geindex Size clauses
13808
@node RM 13 3 42-43 Size Clauses,RM 13 3 50-56 Size Clauses,RM 13 3 29-35 Alignment Clauses,Implementation Advice
13809
@anchor{gnat_rm/implementation_advice rm-13-3-42-43-size-clauses}@anchor{1e1}
13810
@section RM 13.3(42-43): Size Clauses
13815
"The recommended level of support for the @cite{Size} attribute of
13818
A @cite{Size} clause should be supported for an object if the specified
13819
@cite{Size} is at least as large as its subtype's @cite{Size}, and
13820
corresponds to a size in storage elements that is a multiple of the
13821
object's @cite{Alignment} (if the @cite{Alignment} is nonzero)."
13826
@node RM 13 3 50-56 Size Clauses,RM 13 3 71-73 Component Size Clauses,RM 13 3 42-43 Size Clauses,Implementation Advice
13827
@anchor{gnat_rm/implementation_advice rm-13-3-50-56-size-clauses}@anchor{1e2}
13828
@section RM 13.3(50-56): Size Clauses
13833
"If the @cite{Size} of a subtype is specified, and allows for efficient
13834
independent addressability (see 9.10) on the target architecture, then
13835
the @cite{Size} of the following objects of the subtype should equal the
13836
@cite{Size} of the subtype:
13838
Aliased objects (including components)."
13845
"@cite{Size} clause on a composite subtype should not affect the
13846
internal layout of components."
13849
Followed. But note that this can be overridden by use of the implementation
13850
pragma Implicit_Packing in the case of packed arrays.
13854
"The recommended level of support for the @cite{Size} attribute of subtypes is:
13856
The @cite{Size} (if not specified) of a static discrete or fixed point
13857
subtype should be the number of bits needed to represent each value
13858
belonging to the subtype using an unbiased representation, leaving space
13859
for a sign bit only if the subtype contains negative values. If such a
13860
subtype is a first subtype, then an implementation should support a
13861
specified @cite{Size} for it that reflects this representation."
13868
"For a subtype implemented with levels of indirection, the @cite{Size}
13869
should include the size of the pointers, but not the size of what they
13875
@geindex Component_Size clauses
13877
@node RM 13 3 71-73 Component Size Clauses,RM 13 4 9-10 Enumeration Representation Clauses,RM 13 3 50-56 Size Clauses,Implementation Advice
13878
@anchor{gnat_rm/implementation_advice rm-13-3-71-73-component-size-clauses}@anchor{1e3}
13879
@section RM 13.3(71-73): Component Size Clauses
13884
"The recommended level of support for the @cite{Component_Size}
13887
An implementation need not support specified @cite{Component_Sizes} that are
13888
less than the @cite{Size} of the component subtype."
13895
"An implementation should support specified Component_Sizes that
13896
are factors and multiples of the word size. For such
13897
Component_Sizes, the array should contain no gaps between
13898
components. For other Component_Sizes (if supported), the array
13899
should contain no gaps between components when packing is also
13900
specified; the implementation should forbid this combination in cases
13901
where it cannot support a no-gaps representation."
13906
@geindex Enumeration representation clauses
13908
@geindex Representation clauses
13909
@geindex enumeration
13911
@node RM 13 4 9-10 Enumeration Representation Clauses,RM 13 5 1 17-22 Record Representation Clauses,RM 13 3 71-73 Component Size Clauses,Implementation Advice
13912
@anchor{gnat_rm/implementation_advice rm-13-4-9-10-enumeration-representation-clauses}@anchor{1e4}
13913
@section RM 13.4(9-10): Enumeration Representation Clauses
13918
"The recommended level of support for enumeration representation clauses
13921
An implementation need not support enumeration representation clauses
13922
for boolean types, but should at minimum support the internal codes in
13923
the range @cite{System.Min_Int .. System.Max_Int}."
13928
@geindex Record representation clauses
13930
@geindex Representation clauses
13933
@node RM 13 5 1 17-22 Record Representation Clauses,RM 13 5 2 5 Storage Place Attributes,RM 13 4 9-10 Enumeration Representation Clauses,Implementation Advice
13934
@anchor{gnat_rm/implementation_advice rm-13-5-1-17-22-record-representation-clauses}@anchor{1e5}
13935
@section RM 13.5.1(17-22): Record Representation Clauses
13940
"The recommended level of support for
13941
@cite{record_representation_clauses} is:
13943
An implementation should support storage places that can be extracted
13944
with a load, mask, shift sequence of machine code, and set with a load,
13945
shift, mask, store sequence, given the available machine instructions
13946
and run-time model."
13953
"A storage place should be supported if its size is equal to the
13954
@cite{Size} of the component subtype, and it starts and ends on a
13955
boundary that obeys the @cite{Alignment} of the component subtype."
13962
"If the default bit ordering applies to the declaration of a given type,
13963
then for a component whose subtype's @cite{Size} is less than the word
13964
size, any storage place that does not cross an aligned word boundary
13965
should be supported."
13972
"An implementation may reserve a storage place for the tag field of a
13973
tagged type, and disallow other components from overlapping that place."
13976
Followed. The storage place for the tag field is the beginning of the tagged
13977
record, and its size is Address'Size. GNAT will reject an explicit component
13978
clause for the tag field.
13982
"An implementation need not support a @cite{component_clause} for a
13983
component of an extension part if the storage place is not after the
13984
storage places of all components of the parent type, whether or not
13985
those storage places had been specified."
13988
Followed. The above advice on record representation clauses is followed,
13989
and all mentioned features are implemented.
13991
@geindex Storage place attributes
13993
@node RM 13 5 2 5 Storage Place Attributes,RM 13 5 3 7-8 Bit Ordering,RM 13 5 1 17-22 Record Representation Clauses,Implementation Advice
13994
@anchor{gnat_rm/implementation_advice rm-13-5-2-5-storage-place-attributes}@anchor{1e6}
13995
@section RM 13.5.2(5): Storage Place Attributes
14000
"If a component is represented using some form of pointer (such as an
14001
offset) to the actual data of the component, and this data is contiguous
14002
with the rest of the object, then the storage place attributes should
14003
reflect the place of the actual data, not the pointer. If a component is
14004
allocated discontinuously from the rest of the object, then a warning
14005
should be generated upon reference to one of its storage place
14009
Followed. There are no such components in GNAT.
14011
@geindex Bit ordering
14013
@node RM 13 5 3 7-8 Bit Ordering,RM 13 7 37 Address as Private,RM 13 5 2 5 Storage Place Attributes,Implementation Advice
14014
@anchor{gnat_rm/implementation_advice rm-13-5-3-7-8-bit-ordering}@anchor{1e7}
14015
@section RM 13.5.3(7-8): Bit Ordering
14020
"The recommended level of support for the non-default bit ordering is:
14022
If @cite{Word_Size} = @cite{Storage_Unit}, then the implementation
14023
should support the non-default bit ordering in addition to the default
14027
Followed. Word size does not equal storage size in this implementation.
14028
Thus non-default bit ordering is not supported.
14031
@geindex as private type
14033
@node RM 13 7 37 Address as Private,RM 13 7 1 16 Address Operations,RM 13 5 3 7-8 Bit Ordering,Implementation Advice
14034
@anchor{gnat_rm/implementation_advice rm-13-7-37-address-as-private}@anchor{1e8}
14035
@section RM 13.7(37): Address as Private
14040
"@cite{Address} should be of a private type."
14045
@geindex Operations
14046
@geindex on `Address`
14049
@geindex operations of
14051
@node RM 13 7 1 16 Address Operations,RM 13 9 14-17 Unchecked Conversion,RM 13 7 37 Address as Private,Implementation Advice
14052
@anchor{gnat_rm/implementation_advice rm-13-7-1-16-address-operations}@anchor{1e9}
14053
@section RM 13.7.1(16): Address Operations
14058
"Operations in @cite{System} and its children should reflect the target
14059
environment semantics as closely as is reasonable. For example, on most
14060
machines, it makes sense for address arithmetic to 'wrap around'.
14061
Operations that do not make sense should raise @cite{Program_Error}."
14064
Followed. Address arithmetic is modular arithmetic that wraps around. No
14065
operation raises @cite{Program_Error}, since all operations make sense.
14067
@geindex Unchecked conversion
14069
@node RM 13 9 14-17 Unchecked Conversion,RM 13 11 23-25 Implicit Heap Usage,RM 13 7 1 16 Address Operations,Implementation Advice
14070
@anchor{gnat_rm/implementation_advice rm-13-9-14-17-unchecked-conversion}@anchor{1ea}
14071
@section RM 13.9(14-17): Unchecked Conversion
14076
"The @cite{Size} of an array object should not include its bounds; hence,
14077
the bounds should not be part of the converted data."
14084
"The implementation should not generate unnecessary run-time checks to
14085
ensure that the representation of @cite{S} is a representation of the
14086
target type. It should take advantage of the permission to return by
14087
reference when possible. Restrictions on unchecked conversions should be
14088
avoided unless required by the target environment."
14091
Followed. There are no restrictions on unchecked conversion. A warning is
14092
generated if the source and target types do not have the same size since
14093
the semantics in this case may be target dependent.
14097
"The recommended level of support for unchecked conversions is:
14099
Unchecked conversions should be supported and should be reversible in
14100
the cases where this clause defines the result. To enable meaningful use
14101
of unchecked conversion, a contiguous representation should be used for
14102
elementary subtypes, for statically constrained array subtypes whose
14103
component subtype is one of the subtypes described in this paragraph,
14104
and for record subtypes without discriminants whose component subtypes
14105
are described in this paragraph."
14110
@geindex Heap usage
14113
@node RM 13 11 23-25 Implicit Heap Usage,RM 13 11 2 17 Unchecked Deallocation,RM 13 9 14-17 Unchecked Conversion,Implementation Advice
14114
@anchor{gnat_rm/implementation_advice rm-13-11-23-25-implicit-heap-usage}@anchor{1eb}
14115
@section RM 13.11(23-25): Implicit Heap Usage
14120
"An implementation should document any cases in which it dynamically
14121
allocates heap storage for a purpose other than the evaluation of an
14125
Followed, the only other points at which heap storage is dynamically
14126
allocated are as follows:
14132
At initial elaboration time, to allocate dynamically sized global
14136
To allocate space for a task when a task is created.
14139
To extend the secondary stack dynamically when needed. The secondary
14140
stack is used for returning variable length results.
14146
"A default (implementation-provided) storage pool for an
14147
access-to-constant type should not have overhead to support deallocation of
14148
individual objects."
14155
"A storage pool for an anonymous access type should be created at the
14156
point of an allocator for the type, and be reclaimed when the designated
14157
object becomes inaccessible."
14162
@geindex Unchecked deallocation
14164
@node RM 13 11 2 17 Unchecked Deallocation,RM 13 13 2 17 Stream Oriented Attributes,RM 13 11 23-25 Implicit Heap Usage,Implementation Advice
14165
@anchor{gnat_rm/implementation_advice rm-13-11-2-17-unchecked-deallocation}@anchor{1ec}
14166
@section RM 13.11.2(17): Unchecked Deallocation
14171
"For a standard storage pool, @cite{Free} should actually reclaim the
14177
@geindex Stream oriented attributes
14179
@node RM 13 13 2 17 Stream Oriented Attributes,RM A 1 52 Names of Predefined Numeric Types,RM 13 11 2 17 Unchecked Deallocation,Implementation Advice
14180
@anchor{gnat_rm/implementation_advice rm-13-13-2-17-stream-oriented-attributes}@anchor{1ed}
14181
@section RM 13.13.2(17): Stream Oriented Attributes
14186
"If a stream element is the same size as a storage element, then the
14187
normal in-memory representation should be used by @cite{Read} and
14188
@cite{Write} for scalar objects. Otherwise, @cite{Read} and @cite{Write}
14189
should use the smallest number of stream elements needed to represent
14190
all values in the base range of the scalar type."
14193
Followed. By default, GNAT uses the interpretation suggested by AI-195,
14194
which specifies using the size of the first subtype.
14195
However, such an implementation is based on direct binary
14196
representations and is therefore target- and endianness-dependent.
14197
To address this issue, GNAT also supplies an alternate implementation
14198
of the stream attributes @cite{Read} and @cite{Write},
14199
which uses the target-independent XDR standard representation
14202
@geindex XDR representation
14204
@geindex Read attribute
14206
@geindex Write attribute
14208
@geindex Stream oriented attributes
14210
The XDR implementation is provided as an alternative body of the
14211
@cite{System.Stream_Attributes} package, in the file
14212
@code{s-stratt-xdr.adb} in the GNAT library.
14213
There is no @code{s-stratt-xdr.ads} file.
14214
In order to install the XDR implementation, do the following:
14220
Replace the default implementation of the
14221
@cite{System.Stream_Attributes} package with the XDR implementation.
14222
For example on a Unix platform issue the commands:
14225
$ mv s-stratt.adb s-stratt-default.adb
14226
$ mv s-stratt-xdr.adb s-stratt.adb
14230
Rebuild the GNAT run-time library as documented in
14231
the @cite{GNAT and Libraries} section of the @cite{GNAT User's Guide}.
14234
@node RM A 1 52 Names of Predefined Numeric Types,RM A 3 2 49 Ada Characters Handling,RM 13 13 2 17 Stream Oriented Attributes,Implementation Advice
14235
@anchor{gnat_rm/implementation_advice rm-a-1-52-names-of-predefined-numeric-types}@anchor{1ee}
14236
@section RM A.1(52): Names of Predefined Numeric Types
14241
"If an implementation provides additional named predefined integer types,
14242
then the names should end with @code{Integer} as in
14243
@code{Long_Integer}. If an implementation provides additional named
14244
predefined floating point types, then the names should end with
14245
@code{Float} as in @code{Long_Float}."
14250
@geindex Ada.Characters.Handling
14252
@node RM A 3 2 49 Ada Characters Handling,RM A 4 4 106 Bounded-Length String Handling,RM A 1 52 Names of Predefined Numeric Types,Implementation Advice
14253
@anchor{gnat_rm/implementation_advice rm-a-3-2-49-ada-characters-handling}@anchor{1ef}
14254
@section RM A.3.2(49): @cite{Ada.Characters.Handling}
14259
"If an implementation provides a localized definition of @cite{Character}
14260
or @cite{Wide_Character}, then the effects of the subprograms in
14261
@cite{Characters.Handling} should reflect the localizations.
14265
Followed. GNAT provides no such localized definitions.
14267
@geindex Bounded-length strings
14269
@node RM A 4 4 106 Bounded-Length String Handling,RM A 5 2 46-47 Random Number Generation,RM A 3 2 49 Ada Characters Handling,Implementation Advice
14270
@anchor{gnat_rm/implementation_advice rm-a-4-4-106-bounded-length-string-handling}@anchor{1f0}
14271
@section RM A.4.4(106): Bounded-Length String Handling
14276
"Bounded string objects should not be implemented by implicit pointers
14277
and dynamic allocation."
14280
Followed. No implicit pointers or dynamic allocation are used.
14282
@geindex Random number generation
14284
@node RM A 5 2 46-47 Random Number Generation,RM A 10 7 23 Get_Immediate,RM A 4 4 106 Bounded-Length String Handling,Implementation Advice
14285
@anchor{gnat_rm/implementation_advice rm-a-5-2-46-47-random-number-generation}@anchor{1f1}
14286
@section RM A.5.2(46-47): Random Number Generation
14291
"Any storage associated with an object of type @cite{Generator} should be
14292
reclaimed on exit from the scope of the object."
14299
"If the generator period is sufficiently long in relation to the number
14300
of distinct initiator values, then each possible value of
14301
@cite{Initiator} passed to @cite{Reset} should initiate a sequence of
14302
random numbers that does not, in a practical sense, overlap the sequence
14303
initiated by any other value. If this is not possible, then the mapping
14304
between initiator values and generator states should be a rapidly
14305
varying function of the initiator value."
14308
Followed. The generator period is sufficiently long for the first
14309
condition here to hold true.
14311
@geindex Get_Immediate
14313
@node RM A 10 7 23 Get_Immediate,RM B 1 39-41 Pragma Export,RM A 5 2 46-47 Random Number Generation,Implementation Advice
14314
@anchor{gnat_rm/implementation_advice rm-a-10-7-23-get-immediate}@anchor{1f2}
14315
@section RM A.10.7(23): @cite{Get_Immediate}
14320
"The @cite{Get_Immediate} procedures should be implemented with
14321
unbuffered input. For a device such as a keyboard, input should be
14322
available if a key has already been typed, whereas for a disk
14323
file, input should always be available except at end of file. For a file
14324
associated with a keyboard-like device, any line-editing features of the
14325
underlying operating system should be disabled during the execution of
14326
@cite{Get_Immediate}."
14329
Followed on all targets except VxWorks. For VxWorks, there is no way to
14330
provide this functionality that does not result in the input buffer being
14331
flushed before the @cite{Get_Immediate} call. A special unit
14332
@cite{Interfaces.Vxworks.IO} is provided that contains routines to enable
14333
this functionality.
14337
@node RM B 1 39-41 Pragma Export,RM B 2 12-13 Package Interfaces,RM A 10 7 23 Get_Immediate,Implementation Advice
14338
@anchor{gnat_rm/implementation_advice rm-b-1-39-41-pragma-export}@anchor{1f3}
14339
@section RM B.1(39-41): Pragma @cite{Export}
14344
"If an implementation supports pragma @cite{Export} to a given language,
14345
then it should also allow the main subprogram to be written in that
14346
language. It should support some mechanism for invoking the elaboration
14347
of the Ada library units included in the system, and for invoking the
14348
finalization of the environment task. On typical systems, the
14349
recommended mechanism is to provide two subprograms whose link names are
14350
@cite{adainit} and @cite{adafinal}. @cite{adainit} should contain the
14351
elaboration code for library units. @cite{adafinal} should contain the
14352
finalization code. These subprograms should have no effect the second
14353
and subsequent time they are called."
14360
"Automatic elaboration of pre-elaborated packages should be
14361
provided when pragma @cite{Export} is supported."
14364
Followed when the main program is in Ada. If the main program is in a
14365
foreign language, then
14366
@cite{adainit} must be called to elaborate pre-elaborated
14371
"For each supported convention @cite{L} other than @cite{Intrinsic}, an
14372
implementation should support @cite{Import} and @cite{Export} pragmas
14373
for objects of @cite{L}-compatible types and for subprograms, and pragma
14374
@cite{Convention} for @cite{L}-eligible types and for subprograms,
14375
presuming the other language has corresponding features. Pragma
14376
@cite{Convention} need not be supported for scalar types."
14381
@geindex Package Interfaces
14383
@geindex Interfaces
14385
@node RM B 2 12-13 Package Interfaces,RM B 3 63-71 Interfacing with C,RM B 1 39-41 Pragma Export,Implementation Advice
14386
@anchor{gnat_rm/implementation_advice rm-b-2-12-13-package-interfaces}@anchor{1f4}
14387
@section RM B.2(12-13): Package @cite{Interfaces}
14392
"For each implementation-defined convention identifier, there should be a
14393
child package of package Interfaces with the corresponding name. This
14394
package should contain any declarations that would be useful for
14395
interfacing to the language (implementation) represented by the
14396
convention. Any declarations useful for interfacing to any language on
14397
the given hardware architecture should be provided directly in
14398
@cite{Interfaces}."
14405
"An implementation supporting an interface to C, COBOL, or Fortran should
14406
provide the corresponding package or packages described in the following
14410
Followed. GNAT provides all the packages described in this section.
14413
@geindex interfacing with
14415
@node RM B 3 63-71 Interfacing with C,RM B 4 95-98 Interfacing with COBOL,RM B 2 12-13 Package Interfaces,Implementation Advice
14416
@anchor{gnat_rm/implementation_advice rm-b-3-63-71-interfacing-with-c}@anchor{1f5}
14417
@section RM B.3(63-71): Interfacing with C
14422
"An implementation should support the following interface correspondences
14423
between Ada and C."
14430
"An Ada procedure corresponds to a void-returning C function."
14437
"An Ada function corresponds to a non-void C function."
14444
"An Ada @cite{in} scalar parameter is passed as a scalar argument to a C
14452
"An Ada @cite{in} parameter of an access-to-object type with designated
14453
type @cite{T} is passed as a @code{t*} argument to a C function,
14454
where @code{t} is the C type corresponding to the Ada type @cite{T}."
14461
"An Ada access @cite{T} parameter, or an Ada @cite{out} or @cite{in out}
14462
parameter of an elementary type @cite{T}, is passed as a @code{t*}
14463
argument to a C function, where @code{t} is the C type corresponding to
14464
the Ada type @cite{T}. In the case of an elementary @cite{out} or
14465
@cite{in out} parameter, a pointer to a temporary copy is used to
14466
preserve by-copy semantics."
14473
"An Ada parameter of a record type @cite{T}, of any mode, is passed as a
14474
@code{t*} argument to a C function, where @code{t} is the C
14475
structure corresponding to the Ada type @cite{T}."
14478
Followed. This convention may be overridden by the use of the C_Pass_By_Copy
14479
pragma, or Convention, or by explicitly specifying the mechanism for a given
14480
call using an extended import or export pragma.
14484
"An Ada parameter of an array type with component type @cite{T}, of any
14485
mode, is passed as a @code{t*} argument to a C function, where
14486
@code{t} is the C type corresponding to the Ada type @cite{T}."
14493
"An Ada parameter of an access-to-subprogram type is passed as a pointer
14494
to a C function whose prototype corresponds to the designated
14495
subprogram's specification."
14501
@geindex interfacing with
14503
@node RM B 4 95-98 Interfacing with COBOL,RM B 5 22-26 Interfacing with Fortran,RM B 3 63-71 Interfacing with C,Implementation Advice
14504
@anchor{gnat_rm/implementation_advice rm-b-4-95-98-interfacing-with-cobol}@anchor{1f6}
14505
@section RM B.4(95-98): Interfacing with COBOL
14510
"An Ada implementation should support the following interface
14511
correspondences between Ada and COBOL."
14518
"An Ada access @cite{T} parameter is passed as a @code{BY REFERENCE} data item of
14519
the COBOL type corresponding to @cite{T}."
14526
"An Ada in scalar parameter is passed as a @code{BY CONTENT} data item of
14527
the corresponding COBOL type."
14534
"Any other Ada parameter is passed as a @code{BY REFERENCE} data item of the
14535
COBOL type corresponding to the Ada parameter type; for scalars, a local
14536
copy is used if necessary to ensure by-copy semantics."
14542
@geindex interfacing with
14544
@node RM B 5 22-26 Interfacing with Fortran,RM C 1 3-5 Access to Machine Operations,RM B 4 95-98 Interfacing with COBOL,Implementation Advice
14545
@anchor{gnat_rm/implementation_advice rm-b-5-22-26-interfacing-with-fortran}@anchor{1f7}
14546
@section RM B.5(22-26): Interfacing with Fortran
14551
"An Ada implementation should support the following interface
14552
correspondences between Ada and Fortran:"
14559
"An Ada procedure corresponds to a Fortran subroutine."
14566
"An Ada function corresponds to a Fortran function."
14573
"An Ada parameter of an elementary, array, or record type @cite{T} is
14574
passed as a @cite{T} argument to a Fortran procedure, where @cite{T} is
14575
the Fortran type corresponding to the Ada type @cite{T}, and where the
14576
INTENT attribute of the corresponding dummy argument matches the Ada
14577
formal parameter mode; the Fortran implementation's parameter passing
14578
conventions are used. For elementary types, a local copy is used if
14579
necessary to ensure by-copy semantics."
14586
"An Ada parameter of an access-to-subprogram type is passed as a
14587
reference to a Fortran procedure whose interface corresponds to the
14588
designated subprogram's specification."
14593
@geindex Machine operations
14595
@node RM C 1 3-5 Access to Machine Operations,RM C 1 10-16 Access to Machine Operations,RM B 5 22-26 Interfacing with Fortran,Implementation Advice
14596
@anchor{gnat_rm/implementation_advice rm-c-1-3-5-access-to-machine-operations}@anchor{1f8}
14597
@section RM C.1(3-5): Access to Machine Operations
14602
"The machine code or intrinsic support should allow access to all
14603
operations normally available to assembly language programmers for the
14604
target environment, including privileged instructions, if any."
14611
"The interfacing pragmas (see Annex B) should support interface to
14612
assembler; the default assembler should be associated with the
14613
convention identifier @cite{Assembler}."
14620
"If an entity is exported to assembly language, then the implementation
14621
should allocate it at an addressable location, and should ensure that it
14622
is retained by the linking process, even if not otherwise referenced
14623
from the Ada code. The implementation should assume that any call to a
14624
machine code or assembler subprogram is allowed to read or update every
14625
object that is specified as exported."
14630
@node RM C 1 10-16 Access to Machine Operations,RM C 3 28 Interrupt Support,RM C 1 3-5 Access to Machine Operations,Implementation Advice
14631
@anchor{gnat_rm/implementation_advice rm-c-1-10-16-access-to-machine-operations}@anchor{1f9}
14632
@section RM C.1(10-16): Access to Machine Operations
14637
"The implementation should ensure that little or no overhead is
14638
associated with calling intrinsic and machine-code subprograms."
14641
Followed for both intrinsics and machine-code subprograms.
14645
"It is recommended that intrinsic subprograms be provided for convenient
14646
access to any machine operations that provide special capabilities or
14647
efficiency and that are not otherwise available through the language
14651
Followed. A full set of machine operation intrinsic subprograms is provided.
14655
"Atomic read-modify-write operations---e.g., test and set, compare and
14656
swap, decrement and test, enqueue/dequeue."
14659
Followed on any target supporting such operations.
14663
"Standard numeric functions---e.g.:, sin, log."
14666
Followed on any target supporting such operations.
14670
"String manipulation operations---e.g.:, translate and test."
14673
Followed on any target supporting such operations.
14677
"Vector operations---e.g.:, compare vector against thresholds."
14680
Followed on any target supporting such operations.
14684
"Direct operations on I/O ports."
14687
Followed on any target supporting such operations.
14689
@geindex Interrupt support
14691
@node RM C 3 28 Interrupt Support,RM C 3 1 20-21 Protected Procedure Handlers,RM C 1 10-16 Access to Machine Operations,Implementation Advice
14692
@anchor{gnat_rm/implementation_advice rm-c-3-28-interrupt-support}@anchor{1fa}
14693
@section RM C.3(28): Interrupt Support
14698
"If the @cite{Ceiling_Locking} policy is not in effect, the
14699
implementation should provide means for the application to specify which
14700
interrupts are to be blocked during protected actions, if the underlying
14701
system allows for a finer-grain control of interrupt blocking."
14704
Followed. The underlying system does not allow for finer-grain control
14705
of interrupt blocking.
14707
@geindex Protected procedure handlers
14709
@node RM C 3 1 20-21 Protected Procedure Handlers,RM C 3 2 25 Package Interrupts,RM C 3 28 Interrupt Support,Implementation Advice
14710
@anchor{gnat_rm/implementation_advice rm-c-3-1-20-21-protected-procedure-handlers}@anchor{1fb}
14711
@section RM C.3.1(20-21): Protected Procedure Handlers
14716
"Whenever possible, the implementation should allow interrupt handlers to
14717
be called directly by the hardware."
14720
Followed on any target where the underlying operating system permits
14725
"Whenever practical, violations of any
14726
implementation-defined restrictions should be detected before run time."
14729
Followed. Compile time warnings are given when possible.
14731
@geindex Package `Interrupts`
14733
@geindex Interrupts
14735
@node RM C 3 2 25 Package Interrupts,RM C 4 14 Pre-elaboration Requirements,RM C 3 1 20-21 Protected Procedure Handlers,Implementation Advice
14736
@anchor{gnat_rm/implementation_advice rm-c-3-2-25-package-interrupts}@anchor{1fc}
14737
@section RM C.3.2(25): Package @cite{Interrupts}
14742
"If implementation-defined forms of interrupt handler procedures are
14743
supported, such as protected procedures with parameters, then for each
14744
such form of a handler, a type analogous to @cite{Parameterless_Handler}
14745
should be specified in a child package of @cite{Interrupts}, with the
14746
same operations as in the predefined package Interrupts."
14751
@geindex Pre-elaboration requirements
14753
@node RM C 4 14 Pre-elaboration Requirements,RM C 5 8 Pragma Discard_Names,RM C 3 2 25 Package Interrupts,Implementation Advice
14754
@anchor{gnat_rm/implementation_advice rm-c-4-14-pre-elaboration-requirements}@anchor{1fd}
14755
@section RM C.4(14): Pre-elaboration Requirements
14760
"It is recommended that pre-elaborated packages be implemented in such a
14761
way that there should be little or no code executed at run time for the
14762
elaboration of entities not already covered by the Implementation
14766
Followed. Executable code is generated in some cases, e.g., loops
14767
to initialize large arrays.
14769
@node RM C 5 8 Pragma Discard_Names,RM C 7 2 30 The Package Task_Attributes,RM C 4 14 Pre-elaboration Requirements,Implementation Advice
14770
@anchor{gnat_rm/implementation_advice rm-c-5-8-pragma-discard-names}@anchor{1fe}
14771
@section RM C.5(8): Pragma @cite{Discard_Names}
14776
"If the pragma applies to an entity, then the implementation should
14777
reduce the amount of storage used for storing names associated with that
14783
@geindex Package Task_Attributes
14785
@geindex Task_Attributes
14787
@node RM C 7 2 30 The Package Task_Attributes,RM D 3 17 Locking Policies,RM C 5 8 Pragma Discard_Names,Implementation Advice
14788
@anchor{gnat_rm/implementation_advice rm-c-7-2-30-the-package-task-attributes}@anchor{1ff}
14789
@section RM C.7.2(30): The Package Task_Attributes
14794
"Some implementations are targeted to domains in which memory use at run
14795
time must be completely deterministic. For such implementations, it is
14796
recommended that the storage for task attributes will be pre-allocated
14797
statically and not from the heap. This can be accomplished by either
14798
placing restrictions on the number and the size of the task's
14799
attributes, or by using the pre-allocated storage for the first @cite{N}
14800
attribute objects, and the heap for the others. In the latter case,
14801
@cite{N} should be documented."
14804
Not followed. This implementation is not targeted to such a domain.
14806
@geindex Locking Policies
14808
@node RM D 3 17 Locking Policies,RM D 4 16 Entry Queuing Policies,RM C 7 2 30 The Package Task_Attributes,Implementation Advice
14809
@anchor{gnat_rm/implementation_advice rm-d-3-17-locking-policies}@anchor{200}
14810
@section RM D.3(17): Locking Policies
14815
"The implementation should use names that end with @code{_Locking} for
14816
locking policies defined by the implementation."
14819
Followed. Two implementation-defined locking policies are defined,
14820
whose names (@cite{Inheritance_Locking} and
14821
@cite{Concurrent_Readers_Locking}) follow this suggestion.
14823
@geindex Entry queuing policies
14825
@node RM D 4 16 Entry Queuing Policies,RM D 6 9-10 Preemptive Abort,RM D 3 17 Locking Policies,Implementation Advice
14826
@anchor{gnat_rm/implementation_advice rm-d-4-16-entry-queuing-policies}@anchor{201}
14827
@section RM D.4(16): Entry Queuing Policies
14832
"Names that end with @code{_Queuing} should be used
14833
for all implementation-defined queuing policies."
14836
Followed. No such implementation-defined queuing policies exist.
14838
@geindex Preemptive abort
14840
@node RM D 6 9-10 Preemptive Abort,RM D 7 21 Tasking Restrictions,RM D 4 16 Entry Queuing Policies,Implementation Advice
14841
@anchor{gnat_rm/implementation_advice rm-d-6-9-10-preemptive-abort}@anchor{202}
14842
@section RM D.6(9-10): Preemptive Abort
14847
"Even though the @cite{abort_statement} is included in the list of
14848
potentially blocking operations (see 9.5.1), it is recommended that this
14849
statement be implemented in a way that never requires the task executing
14850
the @cite{abort_statement} to block."
14857
"On a multi-processor, the delay associated with aborting a task on
14858
another processor should be bounded; the implementation should use
14859
periodic polling, if necessary, to achieve this."
14864
@geindex Tasking restrictions
14866
@node RM D 7 21 Tasking Restrictions,RM D 8 47-49 Monotonic Time,RM D 6 9-10 Preemptive Abort,Implementation Advice
14867
@anchor{gnat_rm/implementation_advice rm-d-7-21-tasking-restrictions}@anchor{203}
14868
@section RM D.7(21): Tasking Restrictions
14873
"When feasible, the implementation should take advantage of the specified
14874
restrictions to produce a more efficient implementation."
14877
GNAT currently takes advantage of these restrictions by providing an optimized
14878
run time when the Ravenscar profile and the GNAT restricted run time set
14879
of restrictions are specified. See pragma @cite{Profile (Ravenscar)} and
14880
pragma @cite{Profile (Restricted)} for more details.
14885
@node RM D 8 47-49 Monotonic Time,RM E 5 28-29 Partition Communication Subsystem,RM D 7 21 Tasking Restrictions,Implementation Advice
14886
@anchor{gnat_rm/implementation_advice rm-d-8-47-49-monotonic-time}@anchor{204}
14887
@section RM D.8(47-49): Monotonic Time
14892
"When appropriate, implementations should provide configuration
14893
mechanisms to change the value of @cite{Tick}."
14896
Such configuration mechanisms are not appropriate to this implementation
14897
and are thus not supported.
14901
"It is recommended that @cite{Calendar.Clock} and @cite{Real_Time.Clock}
14902
be implemented as transformations of the same time base."
14909
"It is recommended that the best time base which exists in
14910
the underlying system be available to the application through
14911
@cite{Clock}. @cite{Best} may mean highest accuracy or largest range."
14916
@geindex Partition communication subsystem
14920
@node RM E 5 28-29 Partition Communication Subsystem,RM F 7 COBOL Support,RM D 8 47-49 Monotonic Time,Implementation Advice
14921
@anchor{gnat_rm/implementation_advice rm-e-5-28-29-partition-communication-subsystem}@anchor{205}
14922
@section RM E.5(28-29): Partition Communication Subsystem
14927
"Whenever possible, the PCS on the called partition should allow for
14928
multiple tasks to call the RPC-receiver with different messages and
14929
should allow them to block until the corresponding subprogram body
14933
Followed by GLADE, a separately supplied PCS that can be used with
14938
"The @cite{Write} operation on a stream of type @cite{Params_Stream_Type}
14939
should raise @cite{Storage_Error} if it runs out of space trying to
14940
write the @cite{Item} into the stream."
14943
Followed by GLADE, a separately supplied PCS that can be used with
14946
@geindex COBOL support
14948
@node RM F 7 COBOL Support,RM F 1 2 Decimal Radix Support,RM E 5 28-29 Partition Communication Subsystem,Implementation Advice
14949
@anchor{gnat_rm/implementation_advice rm-f-7-cobol-support}@anchor{206}
14950
@section RM F(7): COBOL Support
14955
"If COBOL (respectively, C) is widely supported in the target
14956
environment, implementations supporting the Information Systems Annex
14957
should provide the child package @cite{Interfaces.COBOL} (respectively,
14958
@cite{Interfaces.C}) specified in Annex B and should support a
14959
@cite{convention_identifier} of COBOL (respectively, C) in the interfacing
14960
pragmas (see Annex B), thus allowing Ada programs to interface with
14961
programs written in that language."
14966
@geindex Decimal radix support
14968
@node RM F 1 2 Decimal Radix Support,RM G Numerics,RM F 7 COBOL Support,Implementation Advice
14969
@anchor{gnat_rm/implementation_advice rm-f-1-2-decimal-radix-support}@anchor{207}
14970
@section RM F.1(2): Decimal Radix Support
14975
"Packed decimal should be used as the internal representation for objects
14976
of subtype @cite{S} when @cite{S}'Machine_Radix = 10."
14979
Not followed. GNAT ignores @cite{S}'Machine_Radix and always uses binary
14984
@node RM G Numerics,RM G 1 1 56-58 Complex Types,RM F 1 2 Decimal Radix Support,Implementation Advice
14985
@anchor{gnat_rm/implementation_advice rm-g-numerics}@anchor{208}
14986
@section RM G: Numerics
14991
"If Fortran (respectively, C) is widely supported in the target
14992
environment, implementations supporting the Numerics Annex
14993
should provide the child package @cite{Interfaces.Fortran} (respectively,
14994
@cite{Interfaces.C}) specified in Annex B and should support a
14995
@cite{convention_identifier} of Fortran (respectively, C) in the interfacing
14996
pragmas (see Annex B), thus allowing Ada programs to interface with
14997
programs written in that language."
15002
@geindex Complex types
15004
@node RM G 1 1 56-58 Complex Types,RM G 1 2 49 Complex Elementary Functions,RM G Numerics,Implementation Advice
15005
@anchor{gnat_rm/implementation_advice rm-g-1-1-56-58-complex-types}@anchor{209}
15006
@section RM G.1.1(56-58): Complex Types
15011
"Because the usual mathematical meaning of multiplication of a complex
15012
operand and a real operand is that of the scaling of both components of
15013
the former by the latter, an implementation should not perform this
15014
operation by first promoting the real operand to complex type and then
15015
performing a full complex multiplication. In systems that, in the
15016
future, support an Ada binding to IEC 559:1989, the latter technique
15017
will not generate the required result when one of the components of the
15018
complex operand is infinite. (Explicit multiplication of the infinite
15019
component by the zero component obtained during promotion yields a NaN
15020
that propagates into the final result.) Analogous advice applies in the
15021
case of multiplication of a complex operand and a pure-imaginary
15022
operand, and in the case of division of a complex operand by a real or
15023
pure-imaginary operand."
15030
"Similarly, because the usual mathematical meaning of addition of a
15031
complex operand and a real operand is that the imaginary operand remains
15032
unchanged, an implementation should not perform this operation by first
15033
promoting the real operand to complex type and then performing a full
15034
complex addition. In implementations in which the @cite{Signed_Zeros}
15035
attribute of the component type is @cite{True} (and which therefore
15036
conform to IEC 559:1989 in regard to the handling of the sign of zero in
15037
predefined arithmetic operations), the latter technique will not
15038
generate the required result when the imaginary component of the complex
15039
operand is a negatively signed zero. (Explicit addition of the negative
15040
zero to the zero obtained during promotion yields a positive zero.)
15041
Analogous advice applies in the case of addition of a complex operand
15042
and a pure-imaginary operand, and in the case of subtraction of a
15043
complex operand and a real or pure-imaginary operand."
15050
"Implementations in which @cite{Real'Signed_Zeros} is @cite{True} should
15051
attempt to provide a rational treatment of the signs of zero results and
15052
result components. As one example, the result of the @cite{Argument}
15053
function should have the sign of the imaginary component of the
15054
parameter @cite{X} when the point represented by that parameter lies on
15055
the positive real axis; as another, the sign of the imaginary component
15056
of the @cite{Compose_From_Polar} function should be the same as
15057
(respectively, the opposite of) that of the @cite{Argument} parameter when that
15058
parameter has a value of zero and the @cite{Modulus} parameter has a
15059
nonnegative (respectively, negative) value."
15064
@geindex Complex elementary functions
15066
@node RM G 1 2 49 Complex Elementary Functions,RM G 2 4 19 Accuracy Requirements,RM G 1 1 56-58 Complex Types,Implementation Advice
15067
@anchor{gnat_rm/implementation_advice rm-g-1-2-49-complex-elementary-functions}@anchor{20a}
15068
@section RM G.1.2(49): Complex Elementary Functions
15073
"Implementations in which @cite{Complex_Types.Real'Signed_Zeros} is
15074
@cite{True} should attempt to provide a rational treatment of the signs
15075
of zero results and result components. For example, many of the complex
15076
elementary functions have components that are odd functions of one of
15077
the parameter components; in these cases, the result component should
15078
have the sign of the parameter component at the origin. Other complex
15079
elementary functions have zero components whose sign is opposite that of
15080
a parameter component at the origin, or is always positive or always
15086
@geindex Accuracy requirements
15088
@node RM G 2 4 19 Accuracy Requirements,RM G 2 6 15 Complex Arithmetic Accuracy,RM G 1 2 49 Complex Elementary Functions,Implementation Advice
15089
@anchor{gnat_rm/implementation_advice rm-g-2-4-19-accuracy-requirements}@anchor{20b}
15090
@section RM G.2.4(19): Accuracy Requirements
15095
"The versions of the forward trigonometric functions without a
15096
@cite{Cycle} parameter should not be implemented by calling the
15097
corresponding version with a @cite{Cycle} parameter of
15098
@cite{2.0*Numerics.Pi}, since this will not provide the required
15099
accuracy in some portions of the domain. For the same reason, the
15100
version of @cite{Log} without a @cite{Base} parameter should not be
15101
implemented by calling the corresponding version with a @cite{Base}
15102
parameter of @cite{Numerics.e}."
15107
@geindex Complex arithmetic accuracy
15110
@geindex complex arithmetic
15112
@node RM G 2 6 15 Complex Arithmetic Accuracy,RM H 6 15/2 Pragma Partition_Elaboration_Policy,RM G 2 4 19 Accuracy Requirements,Implementation Advice
15113
@anchor{gnat_rm/implementation_advice rm-g-2-6-15-complex-arithmetic-accuracy}@anchor{20c}
15114
@section RM G.2.6(15): Complex Arithmetic Accuracy
15119
"The version of the @cite{Compose_From_Polar} function without a
15120
@cite{Cycle} parameter should not be implemented by calling the
15121
corresponding version with a @cite{Cycle} parameter of
15122
@cite{2.0*Numerics.Pi}, since this will not provide the required
15123
accuracy in some portions of the domain."
15128
@geindex Sequential elaboration policy
15130
@node RM H 6 15/2 Pragma Partition_Elaboration_Policy,,RM G 2 6 15 Complex Arithmetic Accuracy,Implementation Advice
15131
@anchor{gnat_rm/implementation_advice rm-h-6-15-2-pragma-partition-elaboration-policy}@anchor{20d}
15132
@section RM H.6(15/2): Pragma Partition_Elaboration_Policy
15137
"If the partition elaboration policy is @cite{Sequential} and the
15138
Environment task becomes permanently blocked during elaboration then the
15139
partition is deadlocked and it is recommended that the partition be
15140
immediately terminated."
15145
@node Implementation Defined Characteristics,Intrinsic Subprograms,Implementation Advice,Top
15146
@anchor{gnat_rm/implementation_defined_characteristics implementation-defined-characteristics}@anchor{b}@anchor{gnat_rm/implementation_defined_characteristics doc}@anchor{20e}@anchor{gnat_rm/implementation_defined_characteristics id1}@anchor{20f}
15147
@chapter Implementation Defined Characteristics
15150
In addition to the implementation dependent pragmas and attributes, and the
15151
implementation advice, there are a number of other Ada features that are
15152
potentially implementation dependent and are designated as
15153
implementation-defined. These are mentioned throughout the Ada Reference
15154
Manual, and are summarized in Annex M.
15156
A requirement for conforming Ada compilers is that they provide
15157
documentation describing how the implementation deals with each of these
15158
issues. In this chapter you will find each point in Annex M listed,
15159
followed by a description of how GNAT
15160
handles the implementation dependence.
15162
You can use this chapter as a guide to minimizing implementation
15163
dependent features in your programs if portability to other compilers
15164
and other operating systems is an important consideration. The numbers
15165
in each entry below correspond to the paragraph numbers in the Ada
15172
"Whether or not each recommendation given in Implementation
15173
Advice is followed. See 1.1.2(37)."
15176
See @ref{a,,Implementation Advice}.
15182
"Capacity limitations of the implementation. See 1.1.3(3)."
15185
The complexity of programs that can be processed is limited only by the
15186
total amount of available virtual memory, and disk space for the
15187
generated object files.
15193
"Variations from the standard that are impractical to avoid
15194
given the implementation's execution environment. See 1.1.3(6)."
15197
There are no variations from the standard.
15203
"Which code_statements cause external
15204
interactions. See 1.1.3(10)."
15207
Any @cite{code_statement} can potentially cause external interactions.
15213
"The coded representation for the text of an Ada
15214
program. See 2.1(4)."
15217
See separate section on source representation.
15223
"The control functions allowed in comments. See 2.1(14)."
15226
See separate section on source representation.
15232
"The representation for an end of line. See 2.2(2)."
15235
See separate section on source representation.
15241
"Maximum supported line length and lexical element
15242
length. See 2.2(15)."
15245
The maximum line length is 255 characters and the maximum length of
15246
a lexical element is also 255 characters. This is the default setting
15247
if not overridden by the use of compiler switch @emph{-gnaty} (which
15248
sets the maximum to 79) or @emph{-gnatyMnn} which allows the maximum
15249
line length to be specified to be any value up to 32767. The maximum
15250
length of a lexical element is the same as the maximum line length.
15256
"Implementation defined pragmas. See 2.8(14)."
15259
See @ref{7,,Implementation Defined Pragmas}.
15265
"Effect of pragma @cite{Optimize}. See 2.8(27)."
15268
Pragma @cite{Optimize}, if given with a @cite{Time} or @cite{Space}
15269
parameter, checks that the optimization flag is set, and aborts if it is
15276
"The sequence of characters of the value returned by
15277
@code{S'Image} when some of the graphic characters of
15278
@code{S'Wide_Image} are not defined in @cite{Character}. See
15282
The sequence of characters is as defined by the wide character encoding
15283
method used for the source. See section on source representation for
15290
"The predefined integer types declared in
15291
@cite{Standard}. See 3.5.4(25)."
15295
@multitable {xxxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx}
15306
@emph{Short_Short_Integer}
15314
@emph{Short_Integer}
15318
(Short) 16 bit signed
15330
@emph{Long_Integer}
15334
64 bit signed (on most 64 bit targets,
15335
depending on the C definition of long).
15336
32 bit signed (all other targets)
15340
@emph{Long_Long_Integer}
15353
"Any nonstandard integer types and the operators defined
15354
for them. See 3.5.4(26)."
15357
There are no nonstandard integer types.
15363
"Any nonstandard real types and the operators defined for
15364
them. See 3.5.6(8)."
15367
There are no nonstandard real types.
15373
"What combinations of requested decimal precision and range
15374
are supported for floating point types. See 3.5.7(7)."
15377
The precision and range is as defined by the IEEE standard.
15383
"The predefined floating point types declared in
15384
@cite{Standard}. See 3.5.7(16)."
15388
@multitable {xxxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx}
15411
(Short) 32 bit IEEE short
15423
@emph{Long_Long_Float}
15427
64 bit IEEE long (80 bit IEEE long on x86 processors)
15436
"The small of an ordinary fixed point type. See 3.5.9(8)."
15439
@cite{Fine_Delta} is 2**(-63)
15445
"What combinations of small, range, and digits are
15446
supported for fixed point types. See 3.5.9(10)."
15449
Any combinations are permitted that do not result in a small less than
15450
@cite{Fine_Delta} and do not result in a mantissa larger than 63 bits.
15451
If the mantissa is larger than 53 bits on machines where Long_Long_Float
15452
is 64 bits (true of all architectures except ia32), then the output from
15453
Text_IO is accurate to only 53 bits, rather than the full mantissa. This
15454
is because floating-point conversions are used to convert fixed point.
15460
"The result of @cite{Tags.Expanded_Name} for types declared
15461
within an unnamed @cite{block_statement}. See 3.9(10)."
15464
Block numbers of the form @cite{B`nnn`}, where @cite{nnn} is a
15465
decimal integer are allocated.
15471
"Implementation-defined attributes. See 4.1.4(12)."
15474
See @ref{8,,Implementation Defined Attributes}.
15480
"Any implementation-defined time types. See 9.6(6)."
15483
There are no implementation-defined time types.
15489
"The time base associated with relative delays."
15492
See 9.6(20). The time base used is that provided by the C library
15493
function @cite{gettimeofday}.
15499
"The time base of the type @cite{Calendar.Time}. See
15503
The time base used is that provided by the C library function
15504
@cite{gettimeofday}.
15510
"The time zone used for package @cite{Calendar}
15511
operations. See 9.6(24)."
15514
The time zone used by package @cite{Calendar} is the current system time zone
15515
setting for local time, as accessed by the C library function
15522
"Any limit on @cite{delay_until_statements} of
15523
@cite{select_statements}. See 9.6(29)."
15526
There are no such limits.
15532
"Whether or not two non-overlapping parts of a composite
15533
object are independently addressable, in the case where packing, record
15534
layout, or @cite{Component_Size} is specified for the object. See
15538
Separate components are independently addressable if they do not share
15539
overlapping storage units.
15545
"The representation for a compilation. See 10.1(2)."
15548
A compilation is represented by a sequence of files presented to the
15549
compiler in a single invocation of the @emph{gcc} command.
15555
"Any restrictions on compilations that contain multiple
15556
compilation_units. See 10.1(4)."
15559
No single file can contain more than one compilation unit, but any
15560
sequence of files can be presented to the compiler as a single
15567
"The mechanisms for creating an environment and for adding
15568
and replacing compilation units. See 10.1.4(3)."
15571
See separate section on compilation model.
15577
"The manner of explicitly assigning library units to a
15578
partition. See 10.2(2)."
15581
If a unit contains an Ada main program, then the Ada units for the partition
15582
are determined by recursive application of the rules in the Ada Reference
15583
Manual section 10.2(2-6). In other words, the Ada units will be those that
15584
are needed by the main program, and then this definition of need is applied
15585
recursively to those units, and the partition contains the transitive
15586
closure determined by this relationship. In short, all the necessary units
15587
are included, with no need to explicitly specify the list. If additional
15588
units are required, e.g., by foreign language units, then all units must be
15589
mentioned in the context clause of one of the needed Ada units.
15591
If the partition contains no main program, or if the main program is in
15592
a language other than Ada, then GNAT
15593
provides the binder options @emph{-z} and @emph{-n} respectively, and in
15594
this case a list of units can be explicitly supplied to the binder for
15595
inclusion in the partition (all units needed by these units will also
15596
be included automatically). For full details on the use of these
15597
options, refer to the @cite{GNAT Make Program gnatmake} in the
15598
@cite{GNAT User's Guide}.
15604
"The implementation-defined means, if any, of specifying
15605
which compilation units are needed by a given compilation unit. See
15609
The units needed by a given compilation unit are as defined in
15610
the Ada Reference Manual section 10.2(2-6). There are no
15611
implementation-defined pragmas or other implementation-defined
15612
means for specifying needed units.
15618
"The manner of designating the main subprogram of a
15619
partition. See 10.2(7)."
15622
The main program is designated by providing the name of the
15623
corresponding @code{ALI} file as the input parameter to the binder.
15629
"The order of elaboration of @cite{library_items}. See
15633
The first constraint on ordering is that it meets the requirements of
15634
Chapter 10 of the Ada Reference Manual. This still leaves some
15635
implementation dependent choices, which are resolved by first
15636
elaborating bodies as early as possible (i.e., in preference to specs
15637
where there is a choice), and second by evaluating the immediate with
15638
clauses of a unit to determine the probably best choice, and
15639
third by elaborating in alphabetical order of unit names
15640
where a choice still remains.
15646
"Parameter passing and function return for the main
15647
subprogram. See 10.2(21)."
15650
The main program has no parameters. It may be a procedure, or a function
15651
returning an integer type. In the latter case, the returned integer
15652
value is the return code of the program (overriding any value that
15653
may have been set by a call to @cite{Ada.Command_Line.Set_Exit_Status}).
15659
"The mechanisms for building and running partitions. See
15663
GNAT itself supports programs with only a single partition. The GNATDIST
15664
tool provided with the GLADE package (which also includes an implementation
15665
of the PCS) provides a completely flexible method for building and running
15666
programs consisting of multiple partitions. See the separate GLADE manual
15673
"The details of program execution, including program
15674
termination. See 10.2(25)."
15677
See separate section on compilation model.
15683
"The semantics of any non-active partitions supported by the
15684
implementation. See 10.2(28)."
15687
Passive partitions are supported on targets where shared memory is
15688
provided by the operating system. See the GLADE reference manual for
15695
"The information returned by @cite{Exception_Message}. See
15699
Exception message returns the null string unless a specific message has
15700
been passed by the program.
15706
"The result of @cite{Exceptions.Exception_Name} for types
15707
declared within an unnamed @cite{block_statement}. See 11.4.1(12)."
15710
Blocks have implementation defined names of the form @cite{B`nnn`}
15711
where @cite{nnn} is an integer.
15717
"The information returned by
15718
@cite{Exception_Information}. See 11.4.1(13)."
15721
@cite{Exception_Information} returns a string in the following format:
15724
*Exception_Name:* nnnnn
15727
*Load address:* 0xhhhh
15728
*Call stack traceback locations:*
15729
0xhhhh 0xhhhh 0xhhhh ... 0xhhh
15740
@cite{nnnn} is the fully qualified name of the exception in all upper
15741
case letters. This line is always present.
15744
@cite{mmmm} is the message (this line present only if message is non-null)
15747
@cite{ppp} is the Process Id value as a decimal integer (this line is
15748
present only if the Process Id is nonzero). Currently we are
15749
not making use of this field.
15752
The Load address line, the Call stack traceback locations line and the
15753
following values are present only if at least one traceback location was
15754
recorded. The Load address indicates the address at which the main executable
15755
was loaded; this line may not be present if operating system hasn't relocated
15756
the main executable. The values are given in C style format, with lower case
15757
letters for a-f, and only as many digits present as are necessary.
15758
The line terminator sequence at the end of each line, including
15759
the last line is a single @cite{LF} character (@cite{16#0A#}).
15767
"Implementation-defined check names. See 11.5(27)."
15770
The implementation defined check names include Alignment_Check,
15771
Atomic_Synchronization, Duplicated_Tag_Check, Container_Checks,
15772
Tampering_Check, Predicate_Check, and Validity_Check. In addition, a user
15773
program can add implementation-defined check names by means of the pragma
15774
Check_Name. See the description of pragma @cite{Suppress} for full details.
15780
"The interpretation of each aspect of representation. See
15784
See separate section on data representations.
15790
"Any restrictions placed upon representation items. See
15794
See separate section on data representations.
15800
"The meaning of @cite{Size} for indefinite subtypes. See
15804
Size for an indefinite subtype is the maximum possible size, except that
15805
for the case of a subprogram parameter, the size of the parameter object
15806
is the actual size.
15812
"The default external representation for a type tag. See
15816
The default external representation for a type tag is the fully expanded
15817
name of the type in upper case letters.
15823
"What determines whether a compilation unit is the same in
15824
two different partitions. See 13.3(76)."
15827
A compilation unit is the same in two different partitions if and only
15828
if it derives from the same source file.
15834
"Implementation-defined components. See 13.5.1(15)."
15837
The only implementation defined component is the tag for a tagged type,
15838
which contains a pointer to the dispatching table.
15844
"If @cite{Word_Size} = @cite{Storage_Unit}, the default bit
15845
ordering. See 13.5.3(5)."
15848
@cite{Word_Size} (32) is not the same as @cite{Storage_Unit} (8) for this
15849
implementation, so no non-default bit ordering is supported. The default
15850
bit ordering corresponds to the natural endianness of the target architecture.
15856
"The contents of the visible part of package @cite{System}
15857
and its language-defined children. See 13.7(2)."
15860
See the definition of these packages in files @code{system.ads} and
15861
@code{s-stoele.ads}. Note that two declarations are added to package
15865
Max_Priority : constant Positive := Priority'Last;
15866
Max_Interrupt_Priority : constant Positive := Interrupt_Priority'Last;
15873
"The contents of the visible part of package
15874
@cite{System.Machine_Code}, and the meaning of
15875
@cite{code_statements}. See 13.8(7)."
15878
See the definition and documentation in file @code{s-maccod.ads}.
15884
"The effect of unchecked conversion. See 13.9(11)."
15887
Unchecked conversion between types of the same size
15888
results in an uninterpreted transmission of the bits from one type
15889
to the other. If the types are of unequal sizes, then in the case of
15890
discrete types, a shorter source is first zero or sign extended as
15891
necessary, and a shorter target is simply truncated on the left.
15892
For all non-discrete types, the source is first copied if necessary
15893
to ensure that the alignment requirements of the target are met, then
15894
a pointer is constructed to the source value, and the result is obtained
15895
by dereferencing this pointer after converting it to be a pointer to the
15896
target type. Unchecked conversions where the target subtype is an
15897
unconstrained array are not permitted. If the target alignment is
15898
greater than the source alignment, then a copy of the result is
15899
made with appropriate alignment
15905
"The semantics of operations on invalid representations.
15906
See 13.9.2(10-11)."
15909
For assignments and other operations where the use of invalid values cannot
15910
result in erroneous behavior, the compiler ignores the possibility of invalid
15911
values. An exception is raised at the point where an invalid value would
15912
result in erroneous behavior. For example executing:
15915
procedure invalidvals is
15917
Y : Natural range 1 .. 10;
15918
for Y'Address use X'Address;
15919
Z : Natural range 1 .. 10;
15920
A : array (Natural range 1 .. 10) of Integer;
15922
Z := Y; -- no exception
15923
A (Z) := 3; -- exception raised;
15927
As indicated, an exception is raised on the array assignment, but not
15928
on the simple assignment of the invalid negative value from Y to Z.
15934
"The manner of choosing a storage pool for an access type
15935
when @cite{Storage_Pool} is not specified for the type. See 13.11(17)."
15938
There are 3 different standard pools used by the compiler when
15939
@cite{Storage_Pool} is not specified depending whether the type is local
15940
to a subprogram or defined at the library level and whether
15941
@cite{Storage_Size`is specified or not. See documentation in the runtime library units `System.Pool_Global}, @cite{System.Pool_Size} and
15942
@cite{System.Pool_Local} in files @code{s-poosiz.ads},
15943
@code{s-pooglo.ads} and @code{s-pooloc.ads} for full details on the
15944
default pools used.
15950
"Whether or not the implementation provides user-accessible
15951
names for the standard pool type(s). See 13.11(17)."
15954
See documentation in the sources of the run time mentioned in the previous
15955
paragraph. All these pools are accessible by means of @cite{with}'ing
15962
"The meaning of @cite{Storage_Size}. See 13.11(18)."
15965
@cite{Storage_Size} is measured in storage units, and refers to the
15966
total space available for an access type collection, or to the primary
15967
stack space for a task.
15973
"Implementation-defined aspects of storage pools. See
15977
See documentation in the sources of the run time mentioned in the
15978
paragraph about standard storage pools above
15979
for details on GNAT-defined aspects of storage pools.
15985
"The set of restrictions allowed in a pragma
15986
@cite{Restrictions}. See 13.12(7)."
15989
See @ref{9,,Standard and Implementation Defined Restrictions}.
15995
"The consequences of violating limitations on
15996
@cite{Restrictions} pragmas. See 13.12(9)."
15999
Restrictions that can be checked at compile time result in illegalities
16000
if violated. Currently there are no other consequences of violating
16007
"The representation used by the @cite{Read} and
16008
@cite{Write} attributes of elementary types in terms of stream
16009
elements. See 13.13.2(9)."
16012
The representation is the in-memory representation of the base type of
16013
the type, using the number of bits corresponding to the
16014
@code{type'Size} value, and the natural ordering of the machine.
16020
"The names and characteristics of the numeric subtypes
16021
declared in the visible part of package @cite{Standard}. See A.1(3)."
16024
See items describing the integer and floating-point types supported.
16030
"The string returned by @cite{Character_Set_Version}.
16034
@cite{Ada.Wide_Characters.Handling.Character_Set_Version} returns
16035
the string "Unicode 4.0", referring to version 4.0 of the
16036
Unicode specification.
16042
"The accuracy actually achieved by the elementary
16043
functions. See A.5.1(1)."
16046
The elementary functions correspond to the functions available in the C
16047
library. Only fast math mode is implemented.
16053
"The sign of a zero result from some of the operators or
16054
functions in @cite{Numerics.Generic_Elementary_Functions}, when
16055
@cite{Float_Type'Signed_Zeros} is @cite{True}. See A.5.1(46)."
16058
The sign of zeroes follows the requirements of the IEEE 754 standard on
16066
@cite{Numerics.Float_Random.Max_Image_Width}. See A.5.2(27)."
16069
Maximum image width is 6864, see library file @code{s-rannum.ads}.
16076
@cite{Numerics.Discrete_Random.Max_Image_Width}. See A.5.2(27)."
16079
Maximum image width is 6864, see library file @code{s-rannum.ads}.
16085
"The algorithms for random number generation. See
16089
The algorithm is the Mersenne Twister, as documented in the source file
16090
@code{s-rannum.adb}. This version of the algorithm has a period of
16097
"The string representation of a random number generator's
16098
state. See A.5.2(38)."
16101
The value returned by the Image function is the concatenation of
16102
the fixed-width decimal representations of the 624 32-bit integers
16103
of the state vector.
16109
"The minimum time interval between calls to the
16110
time-dependent Reset procedure that are guaranteed to initiate different
16111
random number sequences. See A.5.2(45)."
16114
The minimum period between reset calls to guarantee distinct series of
16115
random numbers is one microsecond.
16121
"The values of the @cite{Model_Mantissa},
16122
@cite{Model_Emin}, @cite{Model_Epsilon}, @cite{Model},
16123
@cite{Safe_First}, and @cite{Safe_Last} attributes, if the Numerics
16124
Annex is not supported. See A.5.3(72)."
16127
Run the compiler with @emph{-gnatS} to produce a listing of package
16128
@cite{Standard}, has the values of all numeric attributes.
16134
"Any implementation-defined characteristics of the
16135
input-output packages. See A.7(14)."
16138
There are no special implementation defined characteristics for these
16145
"The value of @cite{Buffer_Size} in @cite{Storage_IO}. See
16149
All type representations are contiguous, and the @cite{Buffer_Size} is
16150
the value of @code{type'Size} rounded up to the next storage unit
16157
"External files for standard input, standard output, and
16158
standard error See A.10(5)."
16161
These files are mapped onto the files provided by the C streams
16162
libraries. See source file @code{i-cstrea.ads} for further details.
16168
"The accuracy of the value produced by @cite{Put}. See
16172
If more digits are requested in the output than are represented by the
16173
precision of the value, zeroes are output in the corresponding least
16174
significant digit positions.
16180
"The meaning of @cite{Argument_Count}, @cite{Argument}, and
16181
@cite{Command_Name}. See A.15(1)."
16184
These are mapped onto the @cite{argv} and @cite{argc} parameters of the
16185
main program in the natural manner.
16191
"The interpretation of the @cite{Form} parameter in procedure
16192
@cite{Create_Directory}. See A.16(56)."
16195
The @cite{Form} parameter is not used.
16201
"The interpretation of the @cite{Form} parameter in procedure
16202
@cite{Create_Path}. See A.16(60)."
16205
The @cite{Form} parameter is not used.
16211
"The interpretation of the @cite{Form} parameter in procedure
16212
@cite{Copy_File}. See A.16(68)."
16215
The @cite{Form} parameter is case-insensitive.
16216
Two fields are recognized in the @cite{Form} parameter:
16223
<value> starts immediately after the character '=' and ends with the
16224
character immediately preceding the next comma (',') or with the last
16225
character of the parameter.
16227
The only possible values for preserve= are:
16230
@multitable {xxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx}
16241
@emph{no_attributes}
16245
Do not try to preserve any file attributes. This is the
16246
default if no preserve= is found in Form.
16250
@emph{all_attributes}
16254
Try to preserve all file attributes (timestamps, access rights).
16262
Preserve the timestamp of the copied file, but not the other
16268
The only possible values for mode= are:
16271
@multitable {xxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx}
16286
Only do the copy if the destination file does not already exist.
16287
If it already exists, Copy_File fails.
16295
Copy the file in all cases. Overwrite an already existing destination file.
16303
Append the original file to the destination file. If the destination file
16304
does not exist, the destination file is a copy of the source file.
16305
When mode=append, the field preserve=, if it exists, is not taken into account.
16310
If the Form parameter includes one or both of the fields and the value or
16311
values are incorrect, Copy_file fails with Use_Error.
16313
Examples of correct Forms:
16316
Form => "preserve=no_attributes,mode=overwrite" (the default)
16317
Form => "mode=append"
16318
Form => "mode=copy, preserve=all_attributes"
16321
Examples of incorrect Forms:
16324
Form => "preserve=junk"
16325
Form => "mode=internal, preserve=timestamps"
16332
"The interpretation of the @cite{Pattern} parameter, when not the null string,
16333
in the @cite{Start_Search} and @cite{Search} procedures.
16334
See A.16(104) and A.16(112)."
16337
When the @cite{Pattern} parameter is not the null string, it is interpreted
16338
according to the syntax of regular expressions as defined in the
16339
@cite{GNAT.Regexp} package.
16341
See @ref{210,,GNAT.Regexp (g-regexp.ads)}.
16347
"Implementation-defined convention names. See B.1(11)."
16350
The following convention names are supported
16353
@multitable {xxxxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx}
16372
@emph{Ada_Pass_By_Copy}
16376
Allowed for any types except by-reference types such as limited
16377
records. Compatible with convention Ada, but causes any parameters
16378
with this convention to be passed by copy.
16382
@emph{Ada_Pass_By_Reference}
16386
Allowed for any types except by-copy types such as scalars.
16387
Compatible with convention Ada, but causes any parameters
16388
with this convention to be passed by reference.
16404
Synonym for Assembler
16412
Synonym for Assembler
16424
@emph{C_Pass_By_Copy}
16428
Allowed only for record types, like C, but also notes that record
16429
is to be passed by copy rather than reference.
16441
@emph{C_Plus_Plus (or CPP)}
16453
Treated the same as C
16461
Treated the same as C
16477
For support of pragma @cite{Import} with convention Intrinsic, see
16478
separate section on Intrinsic Subprograms.
16486
Stdcall (used for Windows implementations only). This convention correspond
16487
to the WINAPI (previously called Pascal convention) C/C++ convention under
16488
Windows. A routine with this convention cleans the stack before
16489
exit. This pragma cannot be applied to a dispatching call.
16497
Synonym for Stdcall
16505
Synonym for Stdcall
16513
Stubbed is a special convention used to indicate that the body of the
16514
subprogram will be entirely ignored. Any call to the subprogram
16515
is converted into a raise of the @cite{Program_Error} exception. If a
16516
pragma @cite{Import} specifies convention @cite{stubbed} then no body need
16517
be present at all. This convention is useful during development for the
16518
inclusion of subprograms whose body has not yet been written.
16519
In addition, all otherwise unrecognized convention names are also
16520
treated as being synonymous with convention C. In all implementations
16521
except for VMS, use of such other names results in a warning. In VMS
16522
implementations, these names are accepted silently.
16531
"The meaning of link names. See B.1(36)."
16534
Link names are the actual names used by the linker.
16540
"The manner of choosing link names when neither the link
16541
name nor the address of an imported or exported entity is specified. See
16545
The default linker name is that which would be assigned by the relevant
16546
external language, interpreting the Ada name as being in all lower case
16553
"The effect of pragma @cite{Linker_Options}. See B.1(37)."
16556
The string passed to @cite{Linker_Options} is presented uninterpreted as
16557
an argument to the link command, unless it contains ASCII.NUL characters.
16558
NUL characters if they appear act as argument separators, so for example
16561
pragma Linker_Options ("-labc" & ASCII.NUL & "-ldef");
16564
causes two separate arguments @cite{-labc} and @cite{-ldef} to be passed to the
16565
linker. The order of linker options is preserved for a given unit. The final
16566
list of options passed to the linker is in reverse order of the elaboration
16567
order. For example, linker options for a body always appear before the options
16568
from the corresponding package spec.
16574
"The contents of the visible part of package
16575
@cite{Interfaces} and its language-defined descendants. See B.2(1)."
16578
See files with prefix @code{i-} in the distributed library.
16584
"Implementation-defined children of package
16585
@cite{Interfaces}. The contents of the visible part of package
16586
@cite{Interfaces}. See B.2(11)."
16589
See files with prefix @code{i-} in the distributed library.
16595
"The types @cite{Floating}, @cite{Long_Floating},
16596
@cite{Binary}, @cite{Long_Binary}, @cite{Decimal_ Element}, and
16597
@cite{COBOL_Character}; and the initialization of the variables
16598
@cite{Ada_To_COBOL} and @cite{COBOL_To_Ada}, in
16599
@cite{Interfaces.COBOL}. See B.4(50)."
16603
@multitable {xxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx}
16622
@emph{Long_Floating}
16626
(Floating) Long_Float
16646
@emph{Decimal_Element}
16654
@emph{COBOL_Character}
16663
For initialization, see the file @code{i-cobol.ads} in the distributed library.
16669
"Support for access to machine instructions. See C.1(1)."
16672
See documentation in file @code{s-maccod.ads} in the distributed library.
16678
"Implementation-defined aspects of access to machine
16679
operations. See C.1(9)."
16682
See documentation in file @code{s-maccod.ads} in the distributed library.
16688
"Implementation-defined aspects of interrupts. See C.3(2)."
16691
Interrupts are mapped to signals or conditions as appropriate. See
16693
@cite{Ada.Interrupt_Names} in source file @code{a-intnam.ads} for details
16694
on the interrupts supported on a particular target.
16700
"Implementation-defined aspects of pre-elaboration. See
16704
GNAT does not permit a partition to be restarted without reloading,
16705
except under control of the debugger.
16711
"The semantics of pragma @cite{Discard_Names}. See C.5(7)."
16714
Pragma @cite{Discard_Names} causes names of enumeration literals to
16715
be suppressed. In the presence of this pragma, the Image attribute
16716
provides the image of the Pos of the literal, and Value accepts
16723
"The result of the @cite{Task_Identification.Image}
16724
attribute. See C.7.1(7)."
16727
The result of this attribute is a string that identifies
16728
the object or component that denotes a given task. If a variable @cite{Var}
16729
has a task type, the image for this task will have the form @cite{Var_`XXXXXXXX`},
16731
is the hexadecimal representation of the virtual address of the corresponding
16732
task control block. If the variable is an array of tasks, the image of each
16733
task will have the form of an indexed component indicating the position of a
16734
given task in the array, e.g., @cite{Group(5)_`XXXXXXX`}. If the task is a
16735
component of a record, the image of the task will have the form of a selected
16736
component. These rules are fully recursive, so that the image of a task that
16737
is a subcomponent of a composite object corresponds to the expression that
16738
designates this task.
16740
If a task is created by an allocator, its image depends on the context. If the
16741
allocator is part of an object declaration, the rules described above are used
16742
to construct its image, and this image is not affected by subsequent
16743
assignments. If the allocator appears within an expression, the image
16744
includes only the name of the task type.
16746
If the configuration pragma Discard_Names is present, or if the restriction
16747
No_Implicit_Heap_Allocation is in effect, the image reduces to
16748
the numeric suffix, that is to say the hexadecimal representation of the
16749
virtual address of the control block of the task.
16755
"The value of @cite{Current_Task} when in a protected entry
16756
or interrupt handler. See C.7.1(17)."
16759
Protected entries or interrupt handlers can be executed by any
16760
convenient thread, so the value of @cite{Current_Task} is undefined.
16766
"The effect of calling @cite{Current_Task} from an entry
16767
body or interrupt handler. See C.7.1(19)."
16770
The effect of calling @cite{Current_Task} from an entry body or
16771
interrupt handler is to return the identification of the task currently
16772
executing the code.
16778
"Implementation-defined aspects of
16779
@cite{Task_Attributes}. See C.7.2(19)."
16782
There are no implementation-defined aspects of @cite{Task_Attributes}.
16788
"Values of all @cite{Metrics}. See D(2)."
16791
The metrics information for GNAT depends on the performance of the
16792
underlying operating system. The sources of the run-time for tasking
16793
implementation, together with the output from @emph{-gnatG} can be
16794
used to determine the exact sequence of operating systems calls made
16795
to implement various tasking constructs. Together with appropriate
16796
information on the performance of the underlying operating system,
16797
on the exact target in use, this information can be used to determine
16798
the required metrics.
16804
"The declarations of @cite{Any_Priority} and
16805
@cite{Priority}. See D.1(11)."
16808
See declarations in file @code{system.ads}.
16814
"Implementation-defined execution resources. See D.1(15)."
16817
There are no implementation-defined execution resources.
16823
"Whether, on a multiprocessor, a task that is waiting for
16824
access to a protected object keeps its processor busy. See D.2.1(3)."
16827
On a multi-processor, a task that is waiting for access to a protected
16828
object does not keep its processor busy.
16834
"The affect of implementation defined execution resources
16835
on task dispatching. See D.2.1(9)."
16838
Tasks map to threads in the threads package used by GNAT. Where possible
16839
and appropriate, these threads correspond to native threads of the
16840
underlying operating system.
16846
"Implementation-defined @cite{policy_identifiers} allowed
16847
in a pragma @cite{Task_Dispatching_Policy}. See D.2.2(3)."
16850
There are no implementation-defined policy-identifiers allowed in this
16857
"Implementation-defined aspects of priority inversion. See
16861
Execution of a task cannot be preempted by the implementation processing
16862
of delay expirations for lower priority tasks.
16868
"Implementation-defined task dispatching. See D.2.2(18)."
16871
The policy is the same as that of the underlying threads implementation.
16877
"Implementation-defined @cite{policy_identifiers} allowed
16878
in a pragma @cite{Locking_Policy}. See D.3(4)."
16881
The two implementation defined policies permitted in GNAT are
16882
@cite{Inheritance_Locking} and @cite{Conccurent_Readers_Locking}. On
16883
targets that support the @cite{Inheritance_Locking} policy, locking is
16884
implemented by inheritance, i.e., the task owning the lock operates
16885
at a priority equal to the highest priority of any task currently
16886
requesting the lock. On targets that support the
16887
@cite{Conccurent_Readers_Locking} policy, locking is implemented with a
16888
read/write lock allowing multiple propected object functions to enter
16895
"Default ceiling priorities. See D.3(10)."
16898
The ceiling priority of protected objects of the type
16899
@cite{System.Interrupt_Priority'Last} as described in the Ada
16900
Reference Manual D.3(10),
16906
"The ceiling of any protected object used internally by
16907
the implementation. See D.3(16)."
16910
The ceiling priority of internal protected objects is
16911
@cite{System.Priority'Last}.
16917
"Implementation-defined queuing policies. See D.4(1)."
16920
There are no implementation-defined queuing policies.
16926
"On a multiprocessor, any conditions that cause the
16927
completion of an aborted construct to be delayed later than what is
16928
specified for a single processor. See D.6(3)."
16931
The semantics for abort on a multi-processor is the same as on a single
16932
processor, there are no further delays.
16938
"Any operations that implicitly require heap storage
16939
allocation. See D.7(8)."
16942
The only operation that implicitly requires heap storage allocation is
16949
"What happens when a task terminates in the presence of
16950
pragma @cite{No_Task_Termination}. See D.7(15)."
16953
Execution is erroneous in that case.
16959
"Implementation-defined aspects of pragma
16960
@cite{Restrictions}. See D.7(20)."
16963
There are no such implementation-defined aspects.
16969
"Implementation-defined aspects of package
16970
@cite{Real_Time}. See D.8(17)."
16973
There are no implementation defined aspects of package @cite{Real_Time}.
16979
"Implementation-defined aspects of
16980
@cite{delay_statements}. See D.9(8)."
16983
Any difference greater than one microsecond will cause the task to be
16984
delayed (see D.9(7)).
16990
"The upper bound on the duration of interrupt blocking
16991
caused by the implementation. See D.12(5)."
16994
The upper bound is determined by the underlying operating system. In
16995
no cases is it more than 10 milliseconds.
17001
"The means for creating and executing distributed
17002
programs. See E(5)."
17005
The GLADE package provides a utility GNATDIST for creating and executing
17006
distributed programs. See the GLADE reference manual for further details.
17012
"Any events that can result in a partition becoming
17013
inaccessible. See E.1(7)."
17016
See the GLADE reference manual for full details on such events.
17022
"The scheduling policies, treatment of priorities, and
17023
management of shared resources between partitions in certain cases. See
17027
See the GLADE reference manual for full details on these aspects of
17028
multi-partition execution.
17034
"Events that cause the version of a compilation unit to
17035
change. See E.3(5)."
17038
Editing the source file of a compilation unit, or the source files of
17039
any units on which it is dependent in a significant way cause the version
17040
to change. No other actions cause the version number to change. All changes
17041
are significant except those which affect only layout, capitalization or
17048
"Whether the execution of the remote subprogram is
17049
immediately aborted as a result of cancellation. See E.4(13)."
17052
See the GLADE reference manual for details on the effect of abort in
17053
a distributed application.
17059
"Implementation-defined aspects of the PCS. See E.5(25)."
17062
See the GLADE reference manual for a full description of all implementation
17063
defined aspects of the PCS.
17069
"Implementation-defined interfaces in the PCS. See
17073
See the GLADE reference manual for a full description of all
17074
implementation defined interfaces.
17080
"The values of named numbers in the package
17081
@cite{Decimal}. See F.2(7)."
17085
@multitable {xxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxx}
17128
@emph{Max_Decimal_Digits}
17141
"The value of @cite{Max_Picture_Length} in the package
17142
@cite{Text_IO.Editing}. See F.3.3(16)."
17151
"The value of @cite{Max_Picture_Length} in the package
17152
@cite{Wide_Text_IO.Editing}. See F.3.4(5)."
17161
"The accuracy actually achieved by the complex elementary
17162
functions and by other complex arithmetic operations. See G.1(1)."
17165
Standard library functions are used for the complex arithmetic
17166
operations. Only fast math mode is currently supported.
17172
"The sign of a zero result (or a component thereof) from
17173
any operator or function in @cite{Numerics.Generic_Complex_Types}, when
17174
@cite{Real'Signed_Zeros} is True. See G.1.1(53)."
17177
The signs of zero values are as recommended by the relevant
17178
implementation advice.
17184
"The sign of a zero result (or a component thereof) from
17185
any operator or function in
17186
@cite{Numerics.Generic_Complex_Elementary_Functions}, when
17187
@cite{Real'Signed_Zeros} is @cite{True}. See G.1.2(45)."
17190
The signs of zero values are as recommended by the relevant
17191
implementation advice.
17197
"Whether the strict mode or the relaxed mode is the
17198
default. See G.2(2)."
17201
The strict mode is the default. There is no separate relaxed mode. GNAT
17202
provides a highly efficient implementation of strict mode.
17208
"The result interval in certain cases of fixed-to-float
17209
conversion. See G.2.1(10)."
17212
For cases where the result interval is implementation dependent, the
17213
accuracy is that provided by performing all operations in 64-bit IEEE
17214
floating-point format.
17220
"The result of a floating point arithmetic operation in
17221
overflow situations, when the @cite{Machine_Overflows} attribute of the
17222
result type is @cite{False}. See G.2.1(13)."
17225
Infinite and NaN values are produced as dictated by the IEEE
17226
floating-point standard.
17227
Note that on machines that are not fully compliant with the IEEE
17228
floating-point standard, such as Alpha, the @emph{-mieee} compiler flag
17229
must be used for achieving IEEE conforming behavior (although at the cost
17230
of a significant performance penalty), so infinite and NaN values are
17231
properly generated.
17237
"The result interval for division (or exponentiation by a
17238
negative exponent), when the floating point hardware implements division
17239
as multiplication by a reciprocal. See G.2.1(16)."
17242
Not relevant, division is IEEE exact.
17248
"The definition of close result set, which determines the
17249
accuracy of certain fixed point multiplications and divisions. See
17253
Operations in the close result set are performed using IEEE long format
17254
floating-point arithmetic. The input operands are converted to
17255
floating-point, the operation is done in floating-point, and the result
17256
is converted to the target type.
17262
"Conditions on a @cite{universal_real} operand of a fixed
17263
point multiplication or division for which the result shall be in the
17264
perfect result set. See G.2.3(22)."
17267
The result is only defined to be in the perfect result set if the result
17268
can be computed by a single scaling operation involving a scale factor
17269
representable in 64-bits.
17275
"The result of a fixed point arithmetic operation in
17276
overflow situations, when the @cite{Machine_Overflows} attribute of the
17277
result type is @cite{False}. See G.2.3(27)."
17280
Not relevant, @cite{Machine_Overflows} is @cite{True} for fixed-point
17287
"The result of an elementary function reference in
17288
overflow situations, when the @cite{Machine_Overflows} attribute of the
17289
result type is @cite{False}. See G.2.4(4)."
17292
IEEE infinite and Nan values are produced as appropriate.
17298
"The value of the angle threshold, within which certain
17299
elementary functions, complex arithmetic operations, and complex
17300
elementary functions yield results conforming to a maximum relative
17301
error bound. See G.2.4(10)."
17304
Information on this subject is not yet available.
17310
"The accuracy of certain elementary functions for
17311
parameters beyond the angle threshold. See G.2.4(10)."
17314
Information on this subject is not yet available.
17320
"The result of a complex arithmetic operation or complex
17321
elementary function reference in overflow situations, when the
17322
@cite{Machine_Overflows} attribute of the corresponding real type is
17323
@cite{False}. See G.2.6(5)."
17326
IEEE infinite and Nan values are produced as appropriate.
17332
"The accuracy of certain complex arithmetic operations and
17333
certain complex elementary functions for parameters (or components
17334
thereof) beyond the angle threshold. See G.2.6(8)."
17337
Information on those subjects is not yet available.
17343
"Information regarding bounded errors and erroneous
17344
execution. See H.2(1)."
17347
Information on this subject is not yet available.
17353
"Implementation-defined aspects of pragma
17354
@cite{Inspection_Point}. See H.3.2(8)."
17357
Pragma @cite{Inspection_Point} ensures that the variable is live and can
17358
be examined by the debugger at the inspection point.
17364
"Implementation-defined aspects of pragma
17365
@cite{Restrictions}. See H.4(25)."
17368
There are no implementation-defined aspects of pragma @cite{Restrictions}. The
17369
use of pragma @cite{Restrictions [No_Exceptions]} has no effect on the
17370
generated code. Checks must suppressed by use of pragma @cite{Suppress}.
17376
"Any restrictions on pragma @cite{Restrictions}. See
17380
There are no restrictions on pragma @cite{Restrictions}.
17382
@node Intrinsic Subprograms,Representation Clauses and Pragmas,Implementation Defined Characteristics,Top
17383
@anchor{gnat_rm/intrinsic_subprograms doc}@anchor{211}@anchor{gnat_rm/intrinsic_subprograms intrinsic-subprograms}@anchor{c}@anchor{gnat_rm/intrinsic_subprograms id1}@anchor{212}
17384
@chapter Intrinsic Subprograms
17387
@geindex Intrinsic Subprograms
17389
GNAT allows a user application program to write the declaration:
17392
pragma Import (Intrinsic, name);
17395
providing that the name corresponds to one of the implemented intrinsic
17396
subprograms in GNAT, and that the parameter profile of the referenced
17397
subprogram meets the requirements. This chapter describes the set of
17398
implemented intrinsic subprograms, and the requirements on parameter profiles.
17399
Note that no body is supplied; as with other uses of pragma Import, the
17400
body is supplied elsewhere (in this case by the compiler itself). Note
17401
that any use of this feature is potentially non-portable, since the
17402
Ada standard does not require Ada compilers to implement this feature.
17405
* Intrinsic Operators::
17406
* Compilation_Date::
17407
* Compilation_Time::
17408
* Enclosing_Entity::
17409
* Exception_Information::
17410
* Exception_Message::
17414
* Shifts and Rotates::
17415
* Source_Location::
17419
@node Intrinsic Operators,Compilation_Date,,Intrinsic Subprograms
17420
@anchor{gnat_rm/intrinsic_subprograms id2}@anchor{213}@anchor{gnat_rm/intrinsic_subprograms intrinsic-operators}@anchor{214}
17421
@section Intrinsic Operators
17424
@geindex Intrinsic operator
17426
All the predefined numeric operators in package Standard
17427
in @cite{pragma Import (Intrinsic@comma{}..)}
17428
declarations. In the binary operator case, the operands must have the same
17429
size. The operand or operands must also be appropriate for
17430
the operator. For example, for addition, the operands must
17431
both be floating-point or both be fixed-point, and the
17432
right operand for @cite{"**"} must have a root type of
17433
@cite{Standard.Integer'Base}.
17434
You can use an intrinsic operator declaration as in the following example:
17437
type Int1 is new Integer;
17438
type Int2 is new Integer;
17440
function "+" (X1 : Int1; X2 : Int2) return Int1;
17441
function "+" (X1 : Int1; X2 : Int2) return Int2;
17442
pragma Import (Intrinsic, "+");
17445
This declaration would permit 'mixed mode' arithmetic on items
17446
of the differing types @cite{Int1} and @cite{Int2}.
17447
It is also possible to specify such operators for private types, if the
17448
full views are appropriate arithmetic types.
17450
@node Compilation_Date,Compilation_Time,Intrinsic Operators,Intrinsic Subprograms
17451
@anchor{gnat_rm/intrinsic_subprograms compilation-date}@anchor{215}@anchor{gnat_rm/intrinsic_subprograms id3}@anchor{216}
17452
@section Compilation_Date
17455
@geindex Compilation_Date
17457
This intrinsic subprogram is used in the implementation of the
17458
library package @cite{GNAT.Source_Info}. The only useful use of the
17459
intrinsic import in this case is the one in this unit, so an
17460
application program should simply call the function
17461
@cite{GNAT.Source_Info.Compilation_Date} to obtain the date of
17462
the current compilation (in local time format MMM DD YYYY).
17464
@node Compilation_Time,Enclosing_Entity,Compilation_Date,Intrinsic Subprograms
17465
@anchor{gnat_rm/intrinsic_subprograms compilation-time}@anchor{217}@anchor{gnat_rm/intrinsic_subprograms id4}@anchor{218}
17466
@section Compilation_Time
17469
@geindex Compilation_Time
17471
This intrinsic subprogram is used in the implementation of the
17472
library package @cite{GNAT.Source_Info}. The only useful use of the
17473
intrinsic import in this case is the one in this unit, so an
17474
application program should simply call the function
17475
@cite{GNAT.Source_Info.Compilation_Time} to obtain the time of
17476
the current compilation (in local time format HH:MM:SS).
17478
@node Enclosing_Entity,Exception_Information,Compilation_Time,Intrinsic Subprograms
17479
@anchor{gnat_rm/intrinsic_subprograms id5}@anchor{219}@anchor{gnat_rm/intrinsic_subprograms enclosing-entity}@anchor{21a}
17480
@section Enclosing_Entity
17483
@geindex Enclosing_Entity
17485
This intrinsic subprogram is used in the implementation of the
17486
library package @cite{GNAT.Source_Info}. The only useful use of the
17487
intrinsic import in this case is the one in this unit, so an
17488
application program should simply call the function
17489
@cite{GNAT.Source_Info.Enclosing_Entity} to obtain the name of
17490
the current subprogram, package, task, entry, or protected subprogram.
17492
@node Exception_Information,Exception_Message,Enclosing_Entity,Intrinsic Subprograms
17493
@anchor{gnat_rm/intrinsic_subprograms id6}@anchor{21b}@anchor{gnat_rm/intrinsic_subprograms exception-information}@anchor{21c}
17494
@section Exception_Information
17497
@geindex Exception_Information'
17499
This intrinsic subprogram is used in the implementation of the
17500
library package @cite{GNAT.Current_Exception}. The only useful
17501
use of the intrinsic import in this case is the one in this unit,
17502
so an application program should simply call the function
17503
@cite{GNAT.Current_Exception.Exception_Information} to obtain
17504
the exception information associated with the current exception.
17506
@node Exception_Message,Exception_Name,Exception_Information,Intrinsic Subprograms
17507
@anchor{gnat_rm/intrinsic_subprograms exception-message}@anchor{21d}@anchor{gnat_rm/intrinsic_subprograms id7}@anchor{21e}
17508
@section Exception_Message
17511
@geindex Exception_Message
17513
This intrinsic subprogram is used in the implementation of the
17514
library package @cite{GNAT.Current_Exception}. The only useful
17515
use of the intrinsic import in this case is the one in this unit,
17516
so an application program should simply call the function
17517
@cite{GNAT.Current_Exception.Exception_Message} to obtain
17518
the message associated with the current exception.
17520
@node Exception_Name,File,Exception_Message,Intrinsic Subprograms
17521
@anchor{gnat_rm/intrinsic_subprograms exception-name}@anchor{21f}@anchor{gnat_rm/intrinsic_subprograms id8}@anchor{220}
17522
@section Exception_Name
17525
@geindex Exception_Name
17527
This intrinsic subprogram is used in the implementation of the
17528
library package @cite{GNAT.Current_Exception}. The only useful
17529
use of the intrinsic import in this case is the one in this unit,
17530
so an application program should simply call the function
17531
@cite{GNAT.Current_Exception.Exception_Name} to obtain
17532
the name of the current exception.
17534
@node File,Line,Exception_Name,Intrinsic Subprograms
17535
@anchor{gnat_rm/intrinsic_subprograms file}@anchor{221}@anchor{gnat_rm/intrinsic_subprograms id9}@anchor{222}
17541
This intrinsic subprogram is used in the implementation of the
17542
library package @cite{GNAT.Source_Info}. The only useful use of the
17543
intrinsic import in this case is the one in this unit, so an
17544
application program should simply call the function
17545
@cite{GNAT.Source_Info.File} to obtain the name of the current
17548
@node Line,Shifts and Rotates,File,Intrinsic Subprograms
17549
@anchor{gnat_rm/intrinsic_subprograms id10}@anchor{223}@anchor{gnat_rm/intrinsic_subprograms line}@anchor{224}
17555
This intrinsic subprogram is used in the implementation of the
17556
library package @cite{GNAT.Source_Info}. The only useful use of the
17557
intrinsic import in this case is the one in this unit, so an
17558
application program should simply call the function
17559
@cite{GNAT.Source_Info.Line} to obtain the number of the current
17562
@node Shifts and Rotates,Source_Location,Line,Intrinsic Subprograms
17563
@anchor{gnat_rm/intrinsic_subprograms id11}@anchor{225}@anchor{gnat_rm/intrinsic_subprograms shifts-and-rotates}@anchor{226}
17564
@section Shifts and Rotates
17567
@geindex Shift_Left
17569
@geindex Shift_Right
17571
@geindex Shift_Right_Arithmetic
17573
@geindex Rotate_Left
17575
@geindex Rotate_Right
17577
In standard Ada, the shift and rotate functions are available only
17578
for the predefined modular types in package @cite{Interfaces}. However, in
17579
GNAT it is possible to define these functions for any integer
17580
type (signed or modular), as in this example:
17583
function Shift_Left
17585
Amount : Natural) return T;
17588
The function name must be one of
17589
Shift_Left, Shift_Right, Shift_Right_Arithmetic, Rotate_Left, or
17590
Rotate_Right. T must be an integer type. T'Size must be
17591
8, 16, 32 or 64 bits; if T is modular, the modulus
17592
must be 2**8, 2**16, 2**32 or 2**64.
17593
The result type must be the same as the type of @cite{Value}.
17594
The shift amount must be Natural.
17595
The formal parameter names can be anything.
17597
A more convenient way of providing these shift operators is to use
17598
the Provide_Shift_Operators pragma, which provides the function declarations
17599
and corresponding pragma Import's for all five shift functions.
17601
@node Source_Location,,Shifts and Rotates,Intrinsic Subprograms
17602
@anchor{gnat_rm/intrinsic_subprograms source-location}@anchor{227}@anchor{gnat_rm/intrinsic_subprograms id12}@anchor{228}
17603
@section Source_Location
17606
@geindex Source_Location
17608
This intrinsic subprogram is used in the implementation of the
17609
library routine @cite{GNAT.Source_Info}. The only useful use of the
17610
intrinsic import in this case is the one in this unit, so an
17611
application program should simply call the function
17612
@cite{GNAT.Source_Info.Source_Location} to obtain the current
17613
source file location.
17615
@node Representation Clauses and Pragmas,Standard Library Routines,Intrinsic Subprograms,Top
17616
@anchor{gnat_rm/representation_clauses_and_pragmas representation-clauses-and-pragmas}@anchor{d}@anchor{gnat_rm/representation_clauses_and_pragmas doc}@anchor{229}@anchor{gnat_rm/representation_clauses_and_pragmas id1}@anchor{22a}
17617
@chapter Representation Clauses and Pragmas
17620
@geindex Representation Clauses
17622
@geindex Representation Clause
17624
@geindex Representation Pragma
17627
@geindex representation
17629
This section describes the representation clauses accepted by GNAT, and
17630
their effect on the representation of corresponding data objects.
17632
GNAT fully implements Annex C (Systems Programming). This means that all
17633
the implementation advice sections in chapter 13 are fully implemented.
17634
However, these sections only require a minimal level of support for
17635
representation clauses. GNAT provides much more extensive capabilities,
17636
and this section describes the additional capabilities provided.
17639
* Alignment Clauses::
17641
* Storage_Size Clauses::
17642
* Size of Variant Record Objects::
17643
* Biased Representation::
17644
* Value_Size and Object_Size Clauses::
17645
* Component_Size Clauses::
17646
* Bit_Order Clauses::
17647
* Effect of Bit_Order on Byte Ordering::
17648
* Pragma Pack for Arrays::
17649
* Pragma Pack for Records::
17650
* Record Representation Clauses::
17651
* Handling of Records with Holes::
17652
* Enumeration Clauses::
17653
* Address Clauses::
17654
* Use of Address Clauses for Memory-Mapped I/O::
17655
* Effect of Convention on Representation::
17656
* Conventions and Anonymous Access Types::
17657
* Determining the Representations chosen by GNAT::
17661
@node Alignment Clauses,Size Clauses,,Representation Clauses and Pragmas
17662
@anchor{gnat_rm/representation_clauses_and_pragmas id2}@anchor{22b}@anchor{gnat_rm/representation_clauses_and_pragmas alignment-clauses}@anchor{22c}
17663
@section Alignment Clauses
17666
@geindex Alignment Clause
17668
GNAT requires that all alignment clauses specify a power of 2, and all
17669
default alignments are always a power of 2. The default alignment
17670
values are as follows:
17676
@emph{Primitive Types}.
17678
For primitive types, the alignment is the minimum of the actual size of
17679
objects of the type divided by @cite{Storage_Unit},
17680
and the maximum alignment supported by the target.
17681
(This maximum alignment is given by the GNAT-specific attribute
17682
@cite{Standard'Maximum_Alignment}; see @ref{149,,Attribute Maximum_Alignment}.)
17684
@geindex Maximum_Alignment attribute
17686
For example, for type @cite{Long_Float}, the object size is 8 bytes, and the
17687
default alignment will be 8 on any target that supports alignments
17688
this large, but on some targets, the maximum alignment may be smaller
17689
than 8, in which case objects of type @cite{Long_Float} will be maximally
17695
For arrays, the alignment is equal to the alignment of the component type
17696
for the normal case where no packing or component size is given. If the
17697
array is packed, and the packing is effective (see separate section on
17698
packed arrays), then the alignment will be one for long packed arrays,
17699
or arrays whose length is not known at compile time. For short packed
17700
arrays, which are handled internally as modular types, the alignment
17701
will be as described for primitive types, e.g., a packed array of length
17702
31 bits will have an object size of four bytes, and an alignment of 4.
17707
For the normal non-packed case, the alignment of a record is equal to
17708
the maximum alignment of any of its components. For tagged records, this
17709
includes the implicit access type used for the tag. If a pragma @cite{Pack}
17710
is used and all components are packable (see separate section on pragma
17711
@cite{Pack}), then the resulting alignment is 1, unless the layout of the
17712
record makes it profitable to increase it.
17714
A special case is when:
17720
the size of the record is given explicitly, or a
17721
full record representation clause is given, and
17724
the size of the record is 2, 4, or 8 bytes.
17727
In this case, an alignment is chosen to match the
17728
size of the record. For example, if we have:
17731
type Small is record
17734
for Small'Size use 16;
17737
then the default alignment of the record type @cite{Small} is 2, not 1. This
17738
leads to more efficient code when the record is treated as a unit, and also
17739
allows the type to specified as @cite{Atomic} on architectures requiring
17743
An alignment clause may specify a larger alignment than the default value
17744
up to some maximum value dependent on the target (obtainable by using the
17745
attribute reference @cite{Standard'Maximum_Alignment}). It may also specify
17746
a smaller alignment than the default value for enumeration, integer and
17747
fixed point types, as well as for record types, for example
17754
for V'alignment use 1;
17760
The default alignment for the type @cite{V} is 4, as a result of the
17761
Integer field in the record, but it is permissible, as shown, to
17762
override the default alignment of the record with a smaller value.
17767
Note that according to the Ada standard, an alignment clause applies only
17768
to the first named subtype. If additional subtypes are declared, then the
17769
compiler is allowed to choose any alignment it likes, and there is no way
17770
to control this choice. Consider:
17773
type R is range 1 .. 10_000;
17774
for R'Alignment use 1;
17775
subtype RS is R range 1 .. 1000;
17778
The alignment clause specifies an alignment of 1 for the first named subtype
17779
@cite{R} but this does not necessarily apply to @cite{RS}. When writing
17780
portable Ada code, you should avoid writing code that explicitly or
17781
implicitly relies on the alignment of such subtypes.
17783
For the GNAT compiler, if an explicit alignment clause is given, this
17784
value is also used for any subsequent subtypes. So for GNAT, in the
17785
above example, you can count on the alignment of @cite{RS} being 1. But this
17786
assumption is non-portable, and other compilers may choose different
17787
alignments for the subtype @cite{RS}.
17789
@node Size Clauses,Storage_Size Clauses,Alignment Clauses,Representation Clauses and Pragmas
17790
@anchor{gnat_rm/representation_clauses_and_pragmas id3}@anchor{22d}@anchor{gnat_rm/representation_clauses_and_pragmas size-clauses}@anchor{22e}
17791
@section Size Clauses
17794
@geindex Size Clause
17796
The default size for a type @cite{T} is obtainable through the
17797
language-defined attribute @cite{T'Size} and also through the
17798
equivalent GNAT-defined attribute @cite{T'Value_Size}.
17799
For objects of type @cite{T}, GNAT will generally increase the type size
17800
so that the object size (obtainable through the GNAT-defined attribute
17801
@cite{T'Object_Size})
17802
is a multiple of @cite{T'Alignment * Storage_Unit}.
17807
type Smallint is range 1 .. 6;
17815
In this example, @cite{Smallint'Size} = @cite{Smallint'Value_Size} = 3,
17816
as specified by the RM rules,
17817
but objects of this type will have a size of 8
17818
(@cite{Smallint'Object_Size} = 8),
17819
since objects by default occupy an integral number
17820
of storage units. On some targets, notably older
17821
versions of the Digital Alpha, the size of stand
17822
alone objects of this type may be 32, reflecting
17823
the inability of the hardware to do byte load/stores.
17825
Similarly, the size of type @cite{Rec} is 40 bits
17826
(@cite{Rec'Size} = @cite{Rec'Value_Size} = 40), but
17827
the alignment is 4, so objects of this type will have
17828
their size increased to 64 bits so that it is a multiple
17829
of the alignment (in bits). This decision is
17830
in accordance with the specific Implementation Advice in RM 13.3(43):
17834
"A @cite{Size} clause should be supported for an object if the specified
17835
@cite{Size} is at least as large as its subtype's @cite{Size}, and corresponds
17836
to a size in storage elements that is a multiple of the object's
17837
@cite{Alignment} (if the @cite{Alignment} is nonzero)."
17840
An explicit size clause may be used to override the default size by
17841
increasing it. For example, if we have:
17844
type My_Boolean is new Boolean;
17845
for My_Boolean'Size use 32;
17848
then values of this type will always be 32 bits long. In the case of
17849
discrete types, the size can be increased up to 64 bits, with the effect
17850
that the entire specified field is used to hold the value, sign- or
17851
zero-extended as appropriate. If more than 64 bits is specified, then
17852
padding space is allocated after the value, and a warning is issued that
17853
there are unused bits.
17855
Similarly the size of records and arrays may be increased, and the effect
17856
is to add padding bits after the value. This also causes a warning message
17859
The largest Size value permitted in GNAT is 2**31-1. Since this is a
17860
Size in bits, this corresponds to an object of size 256 megabytes (minus
17861
one). This limitation is true on all targets. The reason for this
17862
limitation is that it improves the quality of the code in many cases
17863
if it is known that a Size value can be accommodated in an object of
17866
@node Storage_Size Clauses,Size of Variant Record Objects,Size Clauses,Representation Clauses and Pragmas
17867
@anchor{gnat_rm/representation_clauses_and_pragmas storage-size-clauses}@anchor{22f}@anchor{gnat_rm/representation_clauses_and_pragmas id4}@anchor{230}
17868
@section Storage_Size Clauses
17871
@geindex Storage_Size Clause
17873
For tasks, the @cite{Storage_Size} clause specifies the amount of space
17874
to be allocated for the task stack. This cannot be extended, and if the
17875
stack is exhausted, then @cite{Storage_Error} will be raised (if stack
17876
checking is enabled). Use a @cite{Storage_Size} attribute definition clause,
17877
or a @cite{Storage_Size} pragma in the task definition to set the
17878
appropriate required size. A useful technique is to include in every
17879
task definition a pragma of the form:
17882
pragma Storage_Size (Default_Stack_Size);
17885
Then @cite{Default_Stack_Size} can be defined in a global package, and
17886
modified as required. Any tasks requiring stack sizes different from the
17887
default can have an appropriate alternative reference in the pragma.
17889
You can also use the @emph{-d} binder switch to modify the default stack
17892
For access types, the @cite{Storage_Size} clause specifies the maximum
17893
space available for allocation of objects of the type. If this space is
17894
exceeded then @cite{Storage_Error} will be raised by an allocation attempt.
17895
In the case where the access type is declared local to a subprogram, the
17896
use of a @cite{Storage_Size} clause triggers automatic use of a special
17897
predefined storage pool (@cite{System.Pool_Size}) that ensures that all
17898
space for the pool is automatically reclaimed on exit from the scope in
17899
which the type is declared.
17901
A special case recognized by the compiler is the specification of a
17902
@cite{Storage_Size} of zero for an access type. This means that no
17903
items can be allocated from the pool, and this is recognized at compile
17904
time, and all the overhead normally associated with maintaining a fixed
17905
size storage pool is eliminated. Consider the following example:
17909
type R is array (Natural) of Character;
17910
type P is access all R;
17911
for P'Storage_Size use 0;
17912
-- Above access type intended only for interfacing purposes
17916
procedure g (m : P);
17917
pragma Import (C, g);
17927
As indicated in this example, these dummy storage pools are often useful in
17928
connection with interfacing where no object will ever be allocated. If you
17929
compile the above example, you get the warning:
17932
p.adb:16:09: warning: allocation from empty storage pool
17933
p.adb:16:09: warning: Storage_Error will be raised at run time
17936
Of course in practice, there will not be any explicit allocators in the
17937
case of such an access declaration.
17939
@node Size of Variant Record Objects,Biased Representation,Storage_Size Clauses,Representation Clauses and Pragmas
17940
@anchor{gnat_rm/representation_clauses_and_pragmas id5}@anchor{231}@anchor{gnat_rm/representation_clauses_and_pragmas size-of-variant-record-objects}@anchor{232}
17941
@section Size of Variant Record Objects
17945
@geindex variant record objects
17947
@geindex Variant record objects
17950
In the case of variant record objects, there is a question whether Size gives
17951
information about a particular variant, or the maximum size required
17952
for any variant. Consider the following program
17955
with Text_IO; use Text_IO;
17957
type R1 (A : Boolean := False) is record
17959
when True => X : Character;
17960
when False => null;
17968
Put_Line (Integer'Image (V1'Size));
17969
Put_Line (Integer'Image (V2'Size));
17973
Here we are dealing with a variant record, where the True variant
17974
requires 16 bits, and the False variant requires 8 bits.
17975
In the above example, both V1 and V2 contain the False variant,
17976
which is only 8 bits long. However, the result of running the
17984
The reason for the difference here is that the discriminant value of
17985
V1 is fixed, and will always be False. It is not possible to assign
17986
a True variant value to V1, therefore 8 bits is sufficient. On the
17987
other hand, in the case of V2, the initial discriminant value is
17988
False (from the default), but it is possible to assign a True
17989
variant value to V2, therefore 16 bits must be allocated for V2
17990
in the general case, even fewer bits may be needed at any particular
17991
point during the program execution.
17993
As can be seen from the output of this program, the @cite{'Size}
17994
attribute applied to such an object in GNAT gives the actual allocated
17995
size of the variable, which is the largest size of any of the variants.
17996
The Ada Reference Manual is not completely clear on what choice should
17997
be made here, but the GNAT behavior seems most consistent with the
17998
language in the RM.
18000
In some cases, it may be desirable to obtain the size of the current
18001
variant, rather than the size of the largest variant. This can be
18002
achieved in GNAT by making use of the fact that in the case of a
18003
subprogram parameter, GNAT does indeed return the size of the current
18004
variant (because a subprogram has no way of knowing how much space
18005
is actually allocated for the actual).
18007
Consider the following modified version of the above program:
18010
with Text_IO; use Text_IO;
18012
type R1 (A : Boolean := False) is record
18014
when True => X : Character;
18015
when False => null;
18021
function Size (V : R1) return Integer is
18027
Put_Line (Integer'Image (V2'Size));
18028
Put_Line (Integer'Image (Size (V2)));
18030
Put_Line (Integer'Image (V2'Size));
18031
Put_Line (Integer'Image (Size (V2)));
18035
The output from this program is
18044
Here we see that while the @cite{'Size} attribute always returns
18045
the maximum size, regardless of the current variant value, the
18046
@cite{Size} function does indeed return the size of the current
18049
@node Biased Representation,Value_Size and Object_Size Clauses,Size of Variant Record Objects,Representation Clauses and Pragmas
18050
@anchor{gnat_rm/representation_clauses_and_pragmas id6}@anchor{233}@anchor{gnat_rm/representation_clauses_and_pragmas biased-representation}@anchor{234}
18051
@section Biased Representation
18054
@geindex Size for biased representation
18056
@geindex Biased representation
18058
In the case of scalars with a range starting at other than zero, it is
18059
possible in some cases to specify a size smaller than the default minimum
18060
value, and in such cases, GNAT uses an unsigned biased representation,
18061
in which zero is used to represent the lower bound, and successive values
18062
represent successive values of the type.
18064
For example, suppose we have the declaration:
18067
type Small is range -7 .. -4;
18068
for Small'Size use 2;
18071
Although the default size of type @cite{Small} is 4, the @cite{Size}
18072
clause is accepted by GNAT and results in the following representation
18076
-7 is represented as 2#00#
18077
-6 is represented as 2#01#
18078
-5 is represented as 2#10#
18079
-4 is represented as 2#11#
18082
Biased representation is only used if the specified @cite{Size} clause
18083
cannot be accepted in any other manner. These reduced sizes that force
18084
biased representation can be used for all discrete types except for
18085
enumeration types for which a representation clause is given.
18087
@node Value_Size and Object_Size Clauses,Component_Size Clauses,Biased Representation,Representation Clauses and Pragmas
18088
@anchor{gnat_rm/representation_clauses_and_pragmas id7}@anchor{235}@anchor{gnat_rm/representation_clauses_and_pragmas value-size-and-object-size-clauses}@anchor{236}
18089
@section Value_Size and Object_Size Clauses
18092
@geindex Value_Size
18094
@geindex Object_Size
18097
@geindex of objects
18099
In Ada 95 and Ada 2005, @cite{T'Size} for a type @cite{T} is the minimum
18100
number of bits required to hold values of type @cite{T}.
18101
Although this interpretation was allowed in Ada 83, it was not required,
18102
and this requirement in practice can cause some significant difficulties.
18103
For example, in most Ada 83 compilers, @cite{Natural'Size} was 32.
18104
However, in Ada 95 and Ada 2005,
18105
@cite{Natural'Size} is
18106
typically 31. This means that code may change in behavior when moving
18107
from Ada 83 to Ada 95 or Ada 2005. For example, consider:
18110
type Rec is record;
18116
at 0 range 0 .. Natural'Size - 1;
18117
at 0 range Natural'Size .. 2 * Natural'Size - 1;
18121
In the above code, since the typical size of @cite{Natural} objects
18122
is 32 bits and @cite{Natural'Size} is 31, the above code can cause
18123
unexpected inefficient packing in Ada 95 and Ada 2005, and in general
18124
there are cases where the fact that the object size can exceed the
18125
size of the type causes surprises.
18127
To help get around this problem GNAT provides two implementation
18128
defined attributes, @cite{Value_Size} and @cite{Object_Size}. When
18129
applied to a type, these attributes yield the size of the type
18130
(corresponding to the RM defined size attribute), and the size of
18131
objects of the type respectively.
18133
The @cite{Object_Size} is used for determining the default size of
18134
objects and components. This size value can be referred to using the
18135
@cite{Object_Size} attribute. The phrase 'is used' here means that it is
18136
the basis of the determination of the size. The backend is free to
18137
pad this up if necessary for efficiency, e.g., an 8-bit stand-alone
18138
character might be stored in 32 bits on a machine with no efficient
18139
byte access instructions such as the Alpha.
18141
The default rules for the value of @cite{Object_Size} for
18142
discrete types are as follows:
18148
The @cite{Object_Size} for base subtypes reflect the natural hardware
18149
size in bits (run the compiler with @emph{-gnatS} to find those values
18150
for numeric types). Enumeration types and fixed-point base subtypes have
18151
8, 16, 32 or 64 bits for this size, depending on the range of values
18155
The @cite{Object_Size} of a subtype is the same as the
18156
@cite{Object_Size} of
18157
the type from which it is obtained.
18160
The @cite{Object_Size} of a derived base type is copied from the parent
18161
base type, and the @cite{Object_Size} of a derived first subtype is copied
18162
from the parent first subtype.
18165
The @cite{Value_Size} attribute
18166
is the (minimum) number of bits required to store a value
18168
This value is used to determine how tightly to pack
18169
records or arrays with components of this type, and also affects
18170
the semantics of unchecked conversion (unchecked conversions where
18171
the @cite{Value_Size} values differ generate a warning, and are potentially
18174
The default rules for the value of @cite{Value_Size} are as follows:
18180
The @cite{Value_Size} for a base subtype is the minimum number of bits
18181
required to store all values of the type (including the sign bit
18182
only if negative values are possible).
18185
If a subtype statically matches the first subtype of a given type, then it has
18186
by default the same @cite{Value_Size} as the first subtype. This is a
18187
consequence of RM 13.1(14): "if two subtypes statically match,
18188
then their subtype-specific aspects are the same".)
18191
All other subtypes have a @cite{Value_Size} corresponding to the minimum
18192
number of bits required to store all values of the subtype. For
18193
dynamic bounds, it is assumed that the value can range down or up
18194
to the corresponding bound of the ancestor
18197
The RM defined attribute @cite{Size} corresponds to the
18198
@cite{Value_Size} attribute.
18200
The @cite{Size} attribute may be defined for a first-named subtype. This sets
18201
the @cite{Value_Size} of
18202
the first-named subtype to the given value, and the
18203
@cite{Object_Size} of this first-named subtype to the given value padded up
18204
to an appropriate boundary. It is a consequence of the default rules
18205
above that this @cite{Object_Size} will apply to all further subtypes. On the
18206
other hand, @cite{Value_Size} is affected only for the first subtype, any
18207
dynamic subtypes obtained from it directly, and any statically matching
18208
subtypes. The @cite{Value_Size} of any other static subtypes is not affected.
18210
@cite{Value_Size} and
18211
@cite{Object_Size} may be explicitly set for any subtype using
18212
an attribute definition clause. Note that the use of these attributes
18213
can cause the RM 13.1(14) rule to be violated. If two access types
18214
reference aliased objects whose subtypes have differing @cite{Object_Size}
18215
values as a result of explicit attribute definition clauses, then it
18216
is illegal to convert from one access subtype to the other. For a more
18217
complete description of this additional legality rule, see the
18218
description of the @cite{Object_Size} attribute.
18220
To get a feel for the difference, consider the following examples (note
18221
that in each case the base is @cite{Short_Short_Integer} with a size of 8):
18224
@multitable {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxx}
18227
Type or subtype declaration
18239
@code{type x1 is range 0 .. 5;}
18251
@code{type x2 is range 0 .. 5;}
18252
@code{for x2'size use 12;}
18264
@code{subtype x3 is x2 range 0 .. 3;}
18276
@code{subtype x4 is x2'base range 0 .. 10;}
18288
@code{dynamic : x2'Base range -64 .. +63;}
18296
@code{subtype x5 is x2 range 0 .. dynamic;}
18308
@code{subtype x6 is x2'base range 0 .. dynamic;}
18321
Note: the entries marked '*' are not actually specified by the Ada
18322
Reference Manual, which has nothing to say about size in the dynamic
18323
case. What GNAT does is to allocate sufficient bits to accomodate any
18324
possible dynamic values for the bounds at run-time.
18326
So far, so good, but GNAT has to obey the RM rules, so the question is
18327
under what conditions must the RM @cite{Size} be used.
18328
The following is a list
18329
of the occasions on which the RM @cite{Size} must be used:
18335
Component size for packed arrays or records
18338
Value of the attribute @cite{Size} for a type
18341
Warning about sizes not matching for unchecked conversion
18344
For record types, the @cite{Object_Size} is always a multiple of the
18345
alignment of the type (this is true for all types). In some cases the
18346
@cite{Value_Size} can be smaller. Consider:
18355
On a typical 32-bit architecture, the X component will be four bytes, and
18356
require four-byte alignment, and the Y component will be one byte. In this
18357
case @cite{R'Value_Size} will be 40 (bits) since this is the minimum size
18358
required to store a value of this type, and for example, it is permissible
18359
to have a component of type R in an outer array whose component size is
18360
specified to be 48 bits. However, @cite{R'Object_Size} will be 64 (bits),
18361
since it must be rounded up so that this value is a multiple of the
18362
alignment (4 bytes = 32 bits).
18364
For all other types, the @cite{Object_Size}
18365
and @cite{Value_Size} are the same (and equivalent to the RM attribute @cite{Size}).
18366
Only @cite{Size} may be specified for such types.
18368
Note that @cite{Value_Size} can be used to force biased representation
18369
for a particular subtype. Consider this example:
18372
type R is (A, B, C, D, E, F);
18373
subtype RAB is R range A .. B;
18374
subtype REF is R range E .. F;
18377
By default, @cite{RAB}
18378
has a size of 1 (sufficient to accommodate the representation
18379
of @cite{A} and @cite{B}, 0 and 1), and @cite{REF}
18380
has a size of 3 (sufficient to accommodate the representation
18381
of @cite{E} and @cite{F}, 4 and 5). But if we add the
18382
following @cite{Value_Size} attribute definition clause:
18385
for REF'Value_Size use 1;
18388
then biased representation is forced for @cite{REF},
18389
and 0 will represent @cite{E} and 1 will represent @cite{F}.
18390
A warning is issued when a @cite{Value_Size} attribute
18391
definition clause forces biased representation. This
18392
warning can be turned off using @cite{-gnatw.B}.
18394
@node Component_Size Clauses,Bit_Order Clauses,Value_Size and Object_Size Clauses,Representation Clauses and Pragmas
18395
@anchor{gnat_rm/representation_clauses_and_pragmas id8}@anchor{237}@anchor{gnat_rm/representation_clauses_and_pragmas component-size-clauses}@anchor{238}
18396
@section Component_Size Clauses
18399
@geindex Component_Size Clause
18401
Normally, the value specified in a component size clause must be consistent
18402
with the subtype of the array component with regard to size and alignment.
18403
In other words, the value specified must be at least equal to the size
18404
of this subtype, and must be a multiple of the alignment value.
18406
In addition, component size clauses are allowed which cause the array
18407
to be packed, by specifying a smaller value. A first case is for
18408
component size values in the range 1 through 63. The value specified
18409
must not be smaller than the Size of the subtype. GNAT will accurately
18410
honor all packing requests in this range. For example, if we have:
18413
type r is array (1 .. 8) of Natural;
18414
for r'Component_Size use 31;
18417
then the resulting array has a length of 31 bytes (248 bits = 8 * 31).
18418
Of course access to the components of such an array is considerably
18419
less efficient than if the natural component size of 32 is used.
18420
A second case is when the subtype of the component is a record type
18421
padded because of its default alignment. For example, if we have:
18430
type a is array (1 .. 8) of r;
18431
for a'Component_Size use 72;
18434
then the resulting array has a length of 72 bytes, instead of 96 bytes
18435
if the alignment of the record (4) was obeyed.
18437
Note that there is no point in giving both a component size clause
18438
and a pragma Pack for the same array type. if such duplicate
18439
clauses are given, the pragma Pack will be ignored.
18441
@node Bit_Order Clauses,Effect of Bit_Order on Byte Ordering,Component_Size Clauses,Representation Clauses and Pragmas
18442
@anchor{gnat_rm/representation_clauses_and_pragmas bit-order-clauses}@anchor{239}@anchor{gnat_rm/representation_clauses_and_pragmas id9}@anchor{23a}
18443
@section Bit_Order Clauses
18446
@geindex Bit_Order Clause
18448
@geindex bit ordering
18453
For record subtypes, GNAT permits the specification of the @cite{Bit_Order}
18454
attribute. The specification may either correspond to the default bit
18455
order for the target, in which case the specification has no effect and
18456
places no additional restrictions, or it may be for the non-standard
18457
setting (that is the opposite of the default).
18459
In the case where the non-standard value is specified, the effect is
18460
to renumber bits within each byte, but the ordering of bytes is not
18461
affected. There are certain
18462
restrictions placed on component clauses as follows:
18468
Components fitting within a single storage unit.
18470
These are unrestricted, and the effect is merely to renumber bits. For
18471
example if we are on a little-endian machine with @cite{Low_Order_First}
18472
being the default, then the following two declarations have exactly
18478
B : Integer range 1 .. 120;
18482
A at 0 range 0 .. 0;
18483
B at 0 range 1 .. 7;
18488
B : Integer range 1 .. 120;
18491
for R2'Bit_Order use High_Order_First;
18494
A at 0 range 7 .. 7;
18495
B at 0 range 0 .. 6;
18499
The useful application here is to write the second declaration with the
18500
@cite{Bit_Order} attribute definition clause, and know that it will be treated
18501
the same, regardless of whether the target is little-endian or big-endian.
18504
Components occupying an integral number of bytes.
18506
These are components that exactly fit in two or more bytes. Such component
18507
declarations are allowed, but have no effect, since it is important to realize
18508
that the @cite{Bit_Order} specification does not affect the ordering of bytes.
18509
In particular, the following attempt at getting an endian-independent integer
18517
for R2'Bit_Order use High_Order_First;
18520
A at 0 range 0 .. 31;
18524
This declaration will result in a little-endian integer on a
18525
little-endian machine, and a big-endian integer on a big-endian machine.
18526
If byte flipping is required for interoperability between big- and
18527
little-endian machines, this must be explicitly programmed. This capability
18528
is not provided by @cite{Bit_Order}.
18531
Components that are positioned across byte boundaries
18533
but do not occupy an integral number of bytes. Given that bytes are not
18534
reordered, such fields would occupy a non-contiguous sequence of bits
18535
in memory, requiring non-trivial code to reassemble. They are for this
18536
reason not permitted, and any component clause specifying such a layout
18537
will be flagged as illegal by GNAT.
18540
Since the misconception that Bit_Order automatically deals with all
18541
endian-related incompatibilities is a common one, the specification of
18542
a component field that is an integral number of bytes will always
18543
generate a warning. This warning may be suppressed using @cite{pragma Warnings (Off)}
18544
if desired. The following section contains additional
18545
details regarding the issue of byte ordering.
18547
@node Effect of Bit_Order on Byte Ordering,Pragma Pack for Arrays,Bit_Order Clauses,Representation Clauses and Pragmas
18548
@anchor{gnat_rm/representation_clauses_and_pragmas id10}@anchor{23b}@anchor{gnat_rm/representation_clauses_and_pragmas effect-of-bit-order-on-byte-ordering}@anchor{23c}
18549
@section Effect of Bit_Order on Byte Ordering
18552
@geindex byte ordering
18557
In this section we will review the effect of the @cite{Bit_Order} attribute
18558
definition clause on byte ordering. Briefly, it has no effect at all, but
18559
a detailed example will be helpful. Before giving this
18560
example, let us review the precise
18561
definition of the effect of defining @cite{Bit_Order}. The effect of a
18562
non-standard bit order is described in section 13.5.3 of the Ada
18567
"2 A bit ordering is a method of interpreting the meaning of
18568
the storage place attributes."
18571
To understand the precise definition of storage place attributes in
18572
this context, we visit section 13.5.1 of the manual:
18576
"13 A record_representation_clause (without the mod_clause)
18577
specifies the layout. The storage place attributes (see 13.5.2)
18578
are taken from the values of the position, first_bit, and last_bit
18579
expressions after normalizing those values so that first_bit is
18580
less than Storage_Unit."
18583
The critical point here is that storage places are taken from
18584
the values after normalization, not before. So the @cite{Bit_Order}
18585
interpretation applies to normalized values. The interpretation
18586
is described in the later part of the 13.5.3 paragraph:
18590
"2 A bit ordering is a method of interpreting the meaning of
18591
the storage place attributes. High_Order_First (known in the
18592
vernacular as 'big endian') means that the first bit of a
18593
storage element (bit 0) is the most significant bit (interpreting
18594
the sequence of bits that represent a component as an unsigned
18595
integer value). Low_Order_First (known in the vernacular as
18596
'little endian') means the opposite: the first bit is the
18597
least significant."
18600
Note that the numbering is with respect to the bits of a storage
18601
unit. In other words, the specification affects only the numbering
18602
of bits within a single storage unit.
18604
We can make the effect clearer by giving an example.
18606
Suppose that we have an external device which presents two bytes, the first
18607
byte presented, which is the first (low addressed byte) of the two byte
18608
record is called Master, and the second byte is called Slave.
18610
The left most (most significant bit is called Control for each byte, and
18611
the remaining 7 bits are called V1, V2, ... V7, where V7 is the rightmost
18612
(least significant) bit.
18614
On a big-endian machine, we can write the following representation clause
18617
type Data is record
18618
Master_Control : Bit;
18626
Slave_Control : Bit;
18636
for Data use record
18637
Master_Control at 0 range 0 .. 0;
18638
Master_V1 at 0 range 1 .. 1;
18639
Master_V2 at 0 range 2 .. 2;
18640
Master_V3 at 0 range 3 .. 3;
18641
Master_V4 at 0 range 4 .. 4;
18642
Master_V5 at 0 range 5 .. 5;
18643
Master_V6 at 0 range 6 .. 6;
18644
Master_V7 at 0 range 7 .. 7;
18645
Slave_Control at 1 range 0 .. 0;
18646
Slave_V1 at 1 range 1 .. 1;
18647
Slave_V2 at 1 range 2 .. 2;
18648
Slave_V3 at 1 range 3 .. 3;
18649
Slave_V4 at 1 range 4 .. 4;
18650
Slave_V5 at 1 range 5 .. 5;
18651
Slave_V6 at 1 range 6 .. 6;
18652
Slave_V7 at 1 range 7 .. 7;
18656
Now if we move this to a little endian machine, then the bit ordering within
18657
the byte is backwards, so we have to rewrite the record rep clause as:
18660
for Data use record
18661
Master_Control at 0 range 7 .. 7;
18662
Master_V1 at 0 range 6 .. 6;
18663
Master_V2 at 0 range 5 .. 5;
18664
Master_V3 at 0 range 4 .. 4;
18665
Master_V4 at 0 range 3 .. 3;
18666
Master_V5 at 0 range 2 .. 2;
18667
Master_V6 at 0 range 1 .. 1;
18668
Master_V7 at 0 range 0 .. 0;
18669
Slave_Control at 1 range 7 .. 7;
18670
Slave_V1 at 1 range 6 .. 6;
18671
Slave_V2 at 1 range 5 .. 5;
18672
Slave_V3 at 1 range 4 .. 4;
18673
Slave_V4 at 1 range 3 .. 3;
18674
Slave_V5 at 1 range 2 .. 2;
18675
Slave_V6 at 1 range 1 .. 1;
18676
Slave_V7 at 1 range 0 .. 0;
18680
It is a nuisance to have to rewrite the clause, especially if
18681
the code has to be maintained on both machines. However,
18682
this is a case that we can handle with the
18683
@cite{Bit_Order} attribute if it is implemented.
18684
Note that the implementation is not required on byte addressed
18685
machines, but it is indeed implemented in GNAT.
18686
This means that we can simply use the
18687
first record clause, together with the declaration
18690
for Data'Bit_Order use High_Order_First;
18693
and the effect is what is desired, namely the layout is exactly the same,
18694
independent of whether the code is compiled on a big-endian or little-endian
18697
The important point to understand is that byte ordering is not affected.
18698
A @cite{Bit_Order} attribute definition never affects which byte a field
18699
ends up in, only where it ends up in that byte.
18700
To make this clear, let us rewrite the record rep clause of the previous
18704
for Data'Bit_Order use High_Order_First;
18705
for Data use record
18706
Master_Control at 0 range 0 .. 0;
18707
Master_V1 at 0 range 1 .. 1;
18708
Master_V2 at 0 range 2 .. 2;
18709
Master_V3 at 0 range 3 .. 3;
18710
Master_V4 at 0 range 4 .. 4;
18711
Master_V5 at 0 range 5 .. 5;
18712
Master_V6 at 0 range 6 .. 6;
18713
Master_V7 at 0 range 7 .. 7;
18714
Slave_Control at 0 range 8 .. 8;
18715
Slave_V1 at 0 range 9 .. 9;
18716
Slave_V2 at 0 range 10 .. 10;
18717
Slave_V3 at 0 range 11 .. 11;
18718
Slave_V4 at 0 range 12 .. 12;
18719
Slave_V5 at 0 range 13 .. 13;
18720
Slave_V6 at 0 range 14 .. 14;
18721
Slave_V7 at 0 range 15 .. 15;
18725
This is exactly equivalent to saying (a repeat of the first example):
18728
for Data'Bit_Order use High_Order_First;
18729
for Data use record
18730
Master_Control at 0 range 0 .. 0;
18731
Master_V1 at 0 range 1 .. 1;
18732
Master_V2 at 0 range 2 .. 2;
18733
Master_V3 at 0 range 3 .. 3;
18734
Master_V4 at 0 range 4 .. 4;
18735
Master_V5 at 0 range 5 .. 5;
18736
Master_V6 at 0 range 6 .. 6;
18737
Master_V7 at 0 range 7 .. 7;
18738
Slave_Control at 1 range 0 .. 0;
18739
Slave_V1 at 1 range 1 .. 1;
18740
Slave_V2 at 1 range 2 .. 2;
18741
Slave_V3 at 1 range 3 .. 3;
18742
Slave_V4 at 1 range 4 .. 4;
18743
Slave_V5 at 1 range 5 .. 5;
18744
Slave_V6 at 1 range 6 .. 6;
18745
Slave_V7 at 1 range 7 .. 7;
18749
Why are they equivalent? Well take a specific field, the @cite{Slave_V2}
18750
field. The storage place attributes are obtained by normalizing the
18751
values given so that the @cite{First_Bit} value is less than 8. After
18752
normalizing the values (0,10,10) we get (1,2,2) which is exactly what
18753
we specified in the other case.
18755
Now one might expect that the @cite{Bit_Order} attribute might affect
18756
bit numbering within the entire record component (two bytes in this
18757
case, thus affecting which byte fields end up in), but that is not
18758
the way this feature is defined, it only affects numbering of bits,
18759
not which byte they end up in.
18761
Consequently it never makes sense to specify a starting bit number
18762
greater than 7 (for a byte addressable field) if an attribute
18763
definition for @cite{Bit_Order} has been given, and indeed it
18764
may be actively confusing to specify such a value, so the compiler
18765
generates a warning for such usage.
18767
If you do need to control byte ordering then appropriate conditional
18768
values must be used. If in our example, the slave byte came first on
18769
some machines we might write:
18772
Master_Byte_First constant Boolean := ...;
18774
Master_Byte : constant Natural :=
18775
1 - Boolean'Pos (Master_Byte_First);
18776
Slave_Byte : constant Natural :=
18777
Boolean'Pos (Master_Byte_First);
18779
for Data'Bit_Order use High_Order_First;
18780
for Data use record
18781
Master_Control at Master_Byte range 0 .. 0;
18782
Master_V1 at Master_Byte range 1 .. 1;
18783
Master_V2 at Master_Byte range 2 .. 2;
18784
Master_V3 at Master_Byte range 3 .. 3;
18785
Master_V4 at Master_Byte range 4 .. 4;
18786
Master_V5 at Master_Byte range 5 .. 5;
18787
Master_V6 at Master_Byte range 6 .. 6;
18788
Master_V7 at Master_Byte range 7 .. 7;
18789
Slave_Control at Slave_Byte range 0 .. 0;
18790
Slave_V1 at Slave_Byte range 1 .. 1;
18791
Slave_V2 at Slave_Byte range 2 .. 2;
18792
Slave_V3 at Slave_Byte range 3 .. 3;
18793
Slave_V4 at Slave_Byte range 4 .. 4;
18794
Slave_V5 at Slave_Byte range 5 .. 5;
18795
Slave_V6 at Slave_Byte range 6 .. 6;
18796
Slave_V7 at Slave_Byte range 7 .. 7;
18800
Now to switch between machines, all that is necessary is
18801
to set the boolean constant @cite{Master_Byte_First} in
18802
an appropriate manner.
18804
@node Pragma Pack for Arrays,Pragma Pack for Records,Effect of Bit_Order on Byte Ordering,Representation Clauses and Pragmas
18805
@anchor{gnat_rm/representation_clauses_and_pragmas pragma-pack-for-arrays}@anchor{23d}@anchor{gnat_rm/representation_clauses_and_pragmas id11}@anchor{23e}
18806
@section Pragma Pack for Arrays
18809
@geindex Pragma Pack (for arrays)
18811
Pragma @cite{Pack} applied to an array has no effect unless the component type
18812
is packable. For a component type to be packable, it must be one of the
18822
Any type whose size is specified with a size clause
18825
Any packed array type with a static size
18828
Any record type padded because of its default alignment
18831
For all these cases, if the component subtype size is in the range
18832
1 through 63, then the effect of the pragma @cite{Pack} is exactly as though a
18833
component size were specified giving the component subtype size.
18834
For example if we have:
18837
type r is range 0 .. 17;
18839
type ar is array (1 .. 8) of r;
18843
Then the component size of @cite{ar} will be set to 5 (i.e., to @cite{r'size},
18844
and the size of the array @cite{ar} will be exactly 40 bits.
18846
Note that in some cases this rather fierce approach to packing can produce
18847
unexpected effects. For example, in Ada 95 and Ada 2005,
18848
subtype @cite{Natural} typically has a size of 31, meaning that if you
18849
pack an array of @cite{Natural}, you get 31-bit
18850
close packing, which saves a few bits, but results in far less efficient
18851
access. Since many other Ada compilers will ignore such a packing request,
18852
GNAT will generate a warning on some uses of pragma @cite{Pack} that it guesses
18853
might not be what is intended. You can easily remove this warning by
18854
using an explicit @cite{Component_Size} setting instead, which never generates
18855
a warning, since the intention of the programmer is clear in this case.
18857
GNAT treats packed arrays in one of two ways. If the size of the array is
18858
known at compile time and is less than 64 bits, then internally the array
18859
is represented as a single modular type, of exactly the appropriate number
18860
of bits. If the length is greater than 63 bits, or is not known at compile
18861
time, then the packed array is represented as an array of bytes, and the
18862
length is always a multiple of 8 bits.
18864
Note that to represent a packed array as a modular type, the alignment must
18865
be suitable for the modular type involved. For example, on typical machines
18866
a 32-bit packed array will be represented by a 32-bit modular integer with
18867
an alignment of four bytes. If you explicitly override the default alignment
18868
with an alignment clause that is too small, the modular representation
18869
cannot be used. For example, consider the following set of declarations:
18872
type R is range 1 .. 3;
18873
type S is array (1 .. 31) of R;
18874
for S'Component_Size use 2;
18876
for S'Alignment use 1;
18879
If the alignment clause were not present, then a 62-bit modular
18880
representation would be chosen (typically with an alignment of 4 or 8
18881
bytes depending on the target). But the default alignment is overridden
18882
with the explicit alignment clause. This means that the modular
18883
representation cannot be used, and instead the array of bytes
18884
representation must be used, meaning that the length must be a multiple
18885
of 8. Thus the above set of declarations will result in a diagnostic
18886
rejecting the size clause and noting that the minimum size allowed is 64.
18888
@geindex Pragma Pack (for type Natural)
18890
@geindex Pragma Pack warning
18892
One special case that is worth noting occurs when the base type of the
18893
component size is 8/16/32 and the subtype is one bit less. Notably this
18894
occurs with subtype @cite{Natural}. Consider:
18897
type Arr is array (1 .. 32) of Natural;
18901
In all commonly used Ada 83 compilers, this pragma Pack would be ignored,
18902
since typically @cite{Natural'Size} is 32 in Ada 83, and in any case most
18903
Ada 83 compilers did not attempt 31 bit packing.
18905
In Ada 95 and Ada 2005, @cite{Natural'Size} is required to be 31. Furthermore,
18906
GNAT really does pack 31-bit subtype to 31 bits. This may result in a
18907
substantial unintended performance penalty when porting legacy Ada 83 code.
18908
To help prevent this, GNAT generates a warning in such cases. If you really
18909
want 31 bit packing in a case like this, you can set the component size
18913
type Arr is array (1 .. 32) of Natural;
18914
for Arr'Component_Size use 31;
18917
Here 31-bit packing is achieved as required, and no warning is generated,
18918
since in this case the programmer intention is clear.
18920
@node Pragma Pack for Records,Record Representation Clauses,Pragma Pack for Arrays,Representation Clauses and Pragmas
18921
@anchor{gnat_rm/representation_clauses_and_pragmas pragma-pack-for-records}@anchor{23f}@anchor{gnat_rm/representation_clauses_and_pragmas id12}@anchor{240}
18922
@section Pragma Pack for Records
18925
@geindex Pragma Pack (for records)
18927
Pragma @cite{Pack} applied to a record will pack the components to reduce
18928
wasted space from alignment gaps and by reducing the amount of space
18929
taken by components. We distinguish between @emph{packable} components and
18930
@emph{non-packable} components.
18931
Components of the following types are considered packable:
18937
Components of a primitive type are packable unless they are aliased
18938
or of an atomic type.
18941
Small packed arrays, whose size does not exceed 64 bits, and where the
18942
size is statically known at compile time, are represented internally
18943
as modular integers, and so they are also packable.
18946
All packable components occupy the exact number of bits corresponding to
18947
their @cite{Size} value, and are packed with no padding bits, i.e., they
18948
can start on an arbitrary bit boundary.
18950
All other types are non-packable, they occupy an integral number of
18952
are placed at a boundary corresponding to their alignment requirements.
18954
For example, consider the record
18957
type Rb1 is array (1 .. 13) of Boolean;
18960
type Rb2 is array (1 .. 65) of Boolean;
18963
type AF is new Float with Atomic;
18976
The representation for the record X2 is as follows:
18979
for X2'Size use 224;
18981
L1 at 0 range 0 .. 0;
18982
L2 at 0 range 1 .. 64;
18983
L3 at 12 range 0 .. 31;
18984
L4 at 16 range 0 .. 0;
18985
L5 at 16 range 1 .. 13;
18986
L6 at 18 range 0 .. 71;
18990
Studying this example, we see that the packable fields @cite{L1}
18992
of length equal to their sizes, and placed at specific bit boundaries (and
18993
not byte boundaries) to
18994
eliminate padding. But @cite{L3} is of a non-packable float type (because
18995
it is aliased), so it is on the next appropriate alignment boundary.
18997
The next two fields are fully packable, so @cite{L4} and @cite{L5} are
18998
minimally packed with no gaps. However, type @cite{Rb2} is a packed
18999
array that is longer than 64 bits, so it is itself non-packable. Thus
19000
the @cite{L6} field is aligned to the next byte boundary, and takes an
19001
integral number of bytes, i.e., 72 bits.
19003
@node Record Representation Clauses,Handling of Records with Holes,Pragma Pack for Records,Representation Clauses and Pragmas
19004
@anchor{gnat_rm/representation_clauses_and_pragmas id13}@anchor{241}@anchor{gnat_rm/representation_clauses_and_pragmas record-representation-clauses}@anchor{242}
19005
@section Record Representation Clauses
19008
@geindex Record Representation Clause
19010
Record representation clauses may be given for all record types, including
19011
types obtained by record extension. Component clauses are allowed for any
19012
static component. The restrictions on component clauses depend on the type
19015
@geindex Component Clause
19017
For all components of an elementary type, the only restriction on component
19018
clauses is that the size must be at least the 'Size value of the type
19019
(actually the Value_Size). There are no restrictions due to alignment,
19020
and such components may freely cross storage boundaries.
19022
Packed arrays with a size up to and including 64 bits are represented
19023
internally using a modular type with the appropriate number of bits, and
19024
thus the same lack of restriction applies. For example, if you declare:
19027
type R is array (1 .. 49) of Boolean;
19032
then a component clause for a component of type R may start on any
19033
specified bit boundary, and may specify a value of 49 bits or greater.
19035
For packed bit arrays that are longer than 64 bits, there are two
19036
cases. If the component size is a power of 2 (1,2,4,8,16,32 bits),
19037
including the important case of single bits or boolean values, then
19038
there are no limitations on placement of such components, and they
19039
may start and end at arbitrary bit boundaries.
19041
If the component size is not a power of 2 (e.g., 3 or 5), then
19042
an array of this type longer than 64 bits must always be placed on
19043
on a storage unit (byte) boundary and occupy an integral number
19044
of storage units (bytes). Any component clause that does not
19045
meet this requirement will be rejected.
19047
Any aliased component, or component of an aliased type, must
19048
have its normal alignment and size. A component clause that
19049
does not meet this requirement will be rejected.
19051
The tag field of a tagged type always occupies an address sized field at
19052
the start of the record. No component clause may attempt to overlay this
19053
tag. When a tagged type appears as a component, the tag field must have
19056
In the case of a record extension T1, of a type T, no component clause applied
19057
to the type T1 can specify a storage location that would overlap the first
19058
T'Size bytes of the record.
19060
For all other component types, including non-bit-packed arrays,
19061
the component can be placed at an arbitrary bit boundary,
19062
so for example, the following is permitted:
19065
type R is array (1 .. 10) of Boolean;
19074
G at 0 range 0 .. 0;
19075
H at 0 range 1 .. 1;
19076
L at 0 range 2 .. 81;
19077
R at 0 range 82 .. 161;
19081
Note: the above rules apply to recent releases of GNAT 5.
19082
In GNAT 3, there are more severe restrictions on larger components.
19083
For non-primitive types, including packed arrays with a size greater than
19084
64 bits, component clauses must respect the alignment requirement of the
19085
type, in particular, always starting on a byte boundary, and the length
19086
must be a multiple of the storage unit.
19088
@node Handling of Records with Holes,Enumeration Clauses,Record Representation Clauses,Representation Clauses and Pragmas
19089
@anchor{gnat_rm/representation_clauses_and_pragmas handling-of-records-with-holes}@anchor{243}@anchor{gnat_rm/representation_clauses_and_pragmas id14}@anchor{244}
19090
@section Handling of Records with Holes
19093
@geindex Handling of Records with Holes
19095
As a result of alignment considerations, records may contain "holes"
19097
which do not correspond to the data bits of any of the components.
19098
Record representation clauses can also result in holes in records.
19100
GNAT does not attempt to clear these holes, so in record objects,
19101
they should be considered to hold undefined rubbish. The generated
19102
equality routine just tests components so does not access these
19103
undefined bits, and assignment and copy operations may or may not
19104
preserve the contents of these holes (for assignments, the holes
19105
in the target will in practice contain either the bits that are
19106
present in the holes in the source, or the bits that were present
19107
in the target before the assignment).
19109
If it is necessary to ensure that holes in records have all zero
19110
bits, then record objects for which this initialization is desired
19111
should be explicitly set to all zero values using Unchecked_Conversion
19112
or address overlays. For example
19115
type HRec is record
19121
On typical machines, integers need to be aligned on a four-byte
19122
boundary, resulting in three bytes of undefined rubbish following
19123
the 8-bit field for C. To ensure that the hole in a variable of
19124
type HRec is set to all zero bits,
19125
you could for example do:
19128
type Base is record
19129
Dummy1, Dummy2 : Integer := 0;
19134
for RealVar'Address use BaseVar'Address;
19137
Now the 8-bytes of the value of RealVar start out containing all zero
19138
bits. A safer approach is to just define dummy fields, avoiding the
19142
type HRec is record
19144
Dummy1 : Short_Short_Integer := 0;
19145
Dummy2 : Short_Short_Integer := 0;
19146
Dummy3 : Short_Short_Integer := 0;
19151
And to make absolutely sure that the intent of this is followed, you
19152
can use representation clauses:
19155
for Hrec use record
19156
C at 0 range 0 .. 7;
19157
Dummy1 at 1 range 0 .. 7;
19158
Dummy2 at 2 range 0 .. 7;
19159
Dummy3 at 3 range 0 .. 7;
19160
I at 4 range 0 .. 31;
19162
for Hrec'Size use 64;
19165
@node Enumeration Clauses,Address Clauses,Handling of Records with Holes,Representation Clauses and Pragmas
19166
@anchor{gnat_rm/representation_clauses_and_pragmas enumeration-clauses}@anchor{245}@anchor{gnat_rm/representation_clauses_and_pragmas id15}@anchor{246}
19167
@section Enumeration Clauses
19170
The only restriction on enumeration clauses is that the range of values
19171
must be representable. For the signed case, if one or more of the
19172
representation values are negative, all values must be in the range:
19175
System.Min_Int .. System.Max_Int
19178
For the unsigned case, where all values are nonnegative, the values must
19182
0 .. System.Max_Binary_Modulus;
19185
A @emph{confirming} representation clause is one in which the values range
19186
from 0 in sequence, i.e., a clause that confirms the default representation
19187
for an enumeration type.
19188
Such a confirming representation
19189
is permitted by these rules, and is specially recognized by the compiler so
19190
that no extra overhead results from the use of such a clause.
19192
If an array has an index type which is an enumeration type to which an
19193
enumeration clause has been applied, then the array is stored in a compact
19194
manner. Consider the declarations:
19197
type r is (A, B, C);
19198
for r use (A => 1, B => 5, C => 10);
19199
type t is array (r) of Character;
19202
The array type t corresponds to a vector with exactly three elements and
19203
has a default size equal to @cite{3*Character'Size}. This ensures efficient
19204
use of space, but means that accesses to elements of the array will incur
19205
the overhead of converting representation values to the corresponding
19206
positional values, (i.e., the value delivered by the @cite{Pos} attribute).
19208
@node Address Clauses,Use of Address Clauses for Memory-Mapped I/O,Enumeration Clauses,Representation Clauses and Pragmas
19209
@anchor{gnat_rm/representation_clauses_and_pragmas id16}@anchor{247}@anchor{gnat_rm/representation_clauses_and_pragmas address-clauses}@anchor{248}
19210
@section Address Clauses
19213
@geindex Address Clause
19215
The reference manual allows a general restriction on representation clauses,
19216
as found in RM 13.1(22):
19220
"An implementation need not support representation
19221
items containing nonstatic expressions, except that
19222
an implementation should support a representation item
19223
for a given entity if each nonstatic expression in the
19224
representation item is a name that statically denotes
19225
a constant declared before the entity."
19228
In practice this is applicable only to address clauses, since this is the
19229
only case in which a nonstatic expression is permitted by the syntax. As
19230
the AARM notes in sections 13.1 (22.a-22.h):
19234
22.a Reason: This is to avoid the following sort of thing:
19236
22.b X : Integer := F(...);
19237
Y : Address := G(...);
19238
for X'Address use Y;
19240
22.c In the above, we have to evaluate the
19241
initialization expression for X before we
19242
know where to put the result. This seems
19243
like an unreasonable implementation burden.
19245
22.d The above code should instead be written
19248
22.e Y : constant Address := G(...);
19249
X : Integer := F(...);
19250
for X'Address use Y;
19252
22.f This allows the expression 'Y' to be safely
19253
evaluated before X is created.
19255
22.g The constant could be a formal parameter of mode in.
19257
22.h An implementation can support other nonstatic
19258
expressions if it wants to. Expressions of type
19259
Address are hardly ever static, but their value
19260
might be known at compile time anyway in many
19264
GNAT does indeed permit many additional cases of nonstatic expressions. In
19265
particular, if the type involved is elementary there are no restrictions
19266
(since in this case, holding a temporary copy of the initialization value,
19267
if one is present, is inexpensive). In addition, if there is no implicit or
19268
explicit initialization, then there are no restrictions. GNAT will reject
19269
only the case where all three of these conditions hold:
19275
The type of the item is non-elementary (e.g., a record or array).
19278
There is explicit or implicit initialization required for the object.
19279
Note that access values are always implicitly initialized.
19282
The address value is nonstatic. Here GNAT is more permissive than the
19283
RM, and allows the address value to be the address of a previously declared
19284
stand-alone variable, as long as it does not itself have an address clause.
19287
Anchor : Some_Initialized_Type;
19288
Overlay : Some_Initialized_Type;
19289
for Overlay'Address use Anchor'Address;
19292
However, the prefix of the address clause cannot be an array component, or
19293
a component of a discriminated record.
19296
As noted above in section 22.h, address values are typically nonstatic. In
19297
particular the To_Address function, even if applied to a literal value, is
19298
a nonstatic function call. To avoid this minor annoyance, GNAT provides
19299
the implementation defined attribute 'To_Address. The following two
19300
expressions have identical values:
19304
@geindex To_Address
19307
To_Address (16#1234_0000#)
19308
System'To_Address (16#1234_0000#);
19311
except that the second form is considered to be a static expression, and
19312
thus when used as an address clause value is always permitted.
19314
Additionally, GNAT treats as static an address clause that is an
19315
unchecked_conversion of a static integer value. This simplifies the porting
19316
of legacy code, and provides a portable equivalent to the GNAT attribute
19319
Another issue with address clauses is the interaction with alignment
19320
requirements. When an address clause is given for an object, the address
19321
value must be consistent with the alignment of the object (which is usually
19322
the same as the alignment of the type of the object). If an address clause
19323
is given that specifies an inappropriately aligned address value, then the
19324
program execution is erroneous.
19326
Since this source of erroneous behavior can have unfortunate effects on
19327
machines with strict alignment requirements, GNAT
19328
checks (at compile time if possible, generating a warning, or at execution
19329
time with a run-time check) that the alignment is appropriate. If the
19330
run-time check fails, then @cite{Program_Error} is raised. This run-time
19331
check is suppressed if range checks are suppressed, or if the special GNAT
19332
check Alignment_Check is suppressed, or if
19333
@cite{pragma Restrictions (No_Elaboration_Code)} is in effect. It is also
19334
suppressed by default on non-strict alignment machines (such as the x86).
19336
Finally, GNAT does not permit overlaying of objects of controlled types or
19337
composite types containing a controlled component. In most cases, the compiler
19338
can detect an attempt at such overlays and will generate a warning at compile
19339
time and a Program_Error exception at run time.
19343
An address clause cannot be given for an exported object. More
19344
understandably the real restriction is that objects with an address
19345
clause cannot be exported. This is because such variables are not
19346
defined by the Ada program, so there is no external object to export.
19350
It is permissible to give an address clause and a pragma Import for the
19351
same object. In this case, the variable is not really defined by the
19352
Ada program, so there is no external symbol to be linked. The link name
19353
and the external name are ignored in this case. The reason that we allow this
19354
combination is that it provides a useful idiom to avoid unwanted
19355
initializations on objects with address clauses.
19357
When an address clause is given for an object that has implicit or
19358
explicit initialization, then by default initialization takes place. This
19359
means that the effect of the object declaration is to overwrite the
19360
memory at the specified address. This is almost always not what the
19361
programmer wants, so GNAT will output a warning:
19371
for Ext'Address use System'To_Address (16#1234_1234#);
19373
>>> warning: implicit initialization of "Ext" may
19374
modify overlaid storage
19375
>>> warning: use pragma Import for "Ext" to suppress
19376
initialization (RM B(24))
19381
As indicated by the warning message, the solution is to use a (dummy) pragma
19382
Import to suppress this initialization. The pragma tell the compiler that the
19383
object is declared and initialized elsewhere. The following package compiles
19384
without warnings (and the initialization is suppressed):
19394
for Ext'Address use System'To_Address (16#1234_1234#);
19395
pragma Import (Ada, Ext);
19399
A final issue with address clauses involves their use for overlaying
19400
variables, as in the following example:
19402
@geindex Overlaying of objects
19407
for B'Address use A'Address;
19410
or alternatively, using the form recommended by the RM:
19414
Addr : constant Address := A'Address;
19416
for B'Address use Addr;
19419
In both of these cases, @cite{A} and @cite{B} become aliased to one another
19420
via the address clause. This use of address clauses to overlay
19421
variables, achieving an effect similar to unchecked conversion
19422
was erroneous in Ada 83, but in Ada 95 and Ada 2005
19423
the effect is implementation defined. Furthermore, the
19424
Ada RM specifically recommends that in a situation
19425
like this, @cite{B} should be subject to the following
19426
implementation advice (RM 13.3(19)):
19430
"19 If the Address of an object is specified, or it is imported
19431
or exported, then the implementation should not perform
19432
optimizations based on assumptions of no aliases."
19435
GNAT follows this recommendation, and goes further by also applying
19436
this recommendation to the overlaid variable (@cite{A} in the above example)
19437
in this case. This means that the overlay works "as expected", in that
19438
a modification to one of the variables will affect the value of the other.
19440
More generally, GNAT interprets this recommendation conservatively for
19441
address clauses: in the cases other than overlays, it considers that the
19442
object is effectively subject to pragma @cite{Volatile} and implements the
19443
associated semantics.
19445
Note that when address clause overlays are used in this way, there is an
19446
issue of unintentional initialization, as shown by this example:
19449
package Overwrite_Record is
19451
A : Character := 'C';
19452
B : Character := 'A';
19454
X : Short_Integer := 3;
19456
for Y'Address use X'Address;
19458
>>> warning: default initialization of "Y" may
19459
modify "X", use pragma Import for "Y" to
19460
suppress initialization (RM B.1(24))
19462
end Overwrite_Record;
19465
Here the default initialization of @cite{Y} will clobber the value
19466
of @cite{X}, which justifies the warning. The warning notes that
19467
this effect can be eliminated by adding a @cite{pragma Import}
19468
which suppresses the initialization:
19471
package Overwrite_Record is
19473
A : Character := 'C';
19474
B : Character := 'A';
19476
X : Short_Integer := 3;
19478
for Y'Address use X'Address;
19479
pragma Import (Ada, Y);
19480
end Overwrite_Record;
19483
Note that the use of @cite{pragma Initialize_Scalars} may cause variables to
19484
be initialized when they would not otherwise have been in the absence
19485
of the use of this pragma. This may cause an overlay to have this
19486
unintended clobbering effect. The compiler avoids this for scalar
19487
types, but not for composite objects (where in general the effect
19488
of @cite{Initialize_Scalars} is part of the initialization routine
19489
for the composite object:
19492
pragma Initialize_Scalars;
19493
with Ada.Text_IO; use Ada.Text_IO;
19494
procedure Overwrite_Array is
19495
type Arr is array (1 .. 5) of Integer;
19496
X : Arr := (others => 1);
19498
for A'Address use X'Address;
19500
>>> warning: default initialization of "A" may
19501
modify "X", use pragma Import for "A" to
19502
suppress initialization (RM B.1(24))
19505
if X /= Arr'(others => 1) then
19506
Put_Line ("X was clobbered");
19508
Put_Line ("X was not clobbered");
19510
end Overwrite_Array;
19513
The above program generates the warning as shown, and at execution
19514
time, prints @cite{X was clobbered}. If the @cite{pragma Import} is
19515
added as suggested:
19518
pragma Initialize_Scalars;
19519
with Ada.Text_IO; use Ada.Text_IO;
19520
procedure Overwrite_Array is
19521
type Arr is array (1 .. 5) of Integer;
19522
X : Arr := (others => 1);
19524
for A'Address use X'Address;
19525
pragma Import (Ada, A);
19527
if X /= Arr'(others => 1) then
19528
Put_Line ("X was clobbered");
19530
Put_Line ("X was not clobbered");
19532
end Overwrite_Array;
19535
then the program compiles without the warning and when run will generate
19536
the output @cite{X was not clobbered}.
19538
@node Use of Address Clauses for Memory-Mapped I/O,Effect of Convention on Representation,Address Clauses,Representation Clauses and Pragmas
19539
@anchor{gnat_rm/representation_clauses_and_pragmas id17}@anchor{249}@anchor{gnat_rm/representation_clauses_and_pragmas use-of-address-clauses-for-memory-mapped-i-o}@anchor{24a}
19540
@section Use of Address Clauses for Memory-Mapped I/O
19543
@geindex Memory-mapped I/O
19545
A common pattern is to use an address clause to map an atomic variable to
19546
a location in memory that corresponds to a memory-mapped I/O operation or
19547
operations, for example:
19550
type Mem_Word is record
19553
pragma Atomic (Mem_Word);
19554
for Mem_Word_Size use 32;
19557
for Mem'Address use some-address;
19564
For a full access (reference or modification) of the variable (Mem) in this
19565
case, as in the above examples, GNAT guarantees that the entire atomic word
19566
will be accessed, in accordance with the RM C.6(15) clause.
19568
A problem arises with a component access such as:
19574
Note that the component A is not declared as atomic. This means that it is
19575
not clear what this assignment means. It could correspond to full word read
19576
and write as given in the first example, or on architectures that supported
19577
such an operation it might be a single byte store instruction. The RM does
19578
not have anything to say in this situation, and GNAT does not make any
19579
guarantee. The code generated may vary from target to target. GNAT will issue
19580
a warning in such a case:
19585
>>> warning: access to non-atomic component of atomic array,
19586
may cause unexpected accesses to atomic object
19589
It is best to be explicit in this situation, by either declaring the
19590
components to be atomic if you want the byte store, or explicitly writing
19591
the full word access sequence if that is what the hardware requires.
19592
Alternatively, if the full word access sequence is required, GNAT also
19593
provides the pragma @cite{Volatile_Full_Access} which can be used in lieu of
19594
pragma @cite{Atomic} and will give the additional guarantee.
19596
@node Effect of Convention on Representation,Conventions and Anonymous Access Types,Use of Address Clauses for Memory-Mapped I/O,Representation Clauses and Pragmas
19597
@anchor{gnat_rm/representation_clauses_and_pragmas id18}@anchor{24b}@anchor{gnat_rm/representation_clauses_and_pragmas effect-of-convention-on-representation}@anchor{24c}
19598
@section Effect of Convention on Representation
19601
@geindex Convention
19602
@geindex effect on representation
19604
Normally the specification of a foreign language convention for a type or
19605
an object has no effect on the chosen representation. In particular, the
19606
representation chosen for data in GNAT generally meets the standard system
19607
conventions, and for example records are laid out in a manner that is
19608
consistent with C. This means that specifying convention C (for example)
19611
There are four exceptions to this general rule:
19617
@emph{Convention Fortran and array subtypes}.
19619
If pragma Convention Fortran is specified for an array subtype, then in
19620
accordance with the implementation advice in section 3.6.2(11) of the
19621
Ada Reference Manual, the array will be stored in a Fortran-compatible
19622
column-major manner, instead of the normal default row-major order.
19625
@emph{Convention C and enumeration types}
19627
GNAT normally stores enumeration types in 8, 16, or 32 bits as required
19628
to accommodate all values of the type. For example, for the enumeration
19632
type Color is (Red, Green, Blue);
19635
8 bits is sufficient to store all values of the type, so by default, objects
19636
of type @cite{Color} will be represented using 8 bits. However, normal C
19637
convention is to use 32 bits for all enum values in C, since enum values
19638
are essentially of type int. If pragma @cite{Convention C} is specified for an
19639
Ada enumeration type, then the size is modified as necessary (usually to
19640
32 bits) to be consistent with the C convention for enum values.
19642
Note that this treatment applies only to types. If Convention C is given for
19643
an enumeration object, where the enumeration type is not Convention C, then
19644
Object_Size bits are allocated. For example, for a normal enumeration type,
19645
with less than 256 elements, only 8 bits will be allocated for the object.
19646
Since this may be a surprise in terms of what C expects, GNAT will issue a
19647
warning in this situation. The warning can be suppressed by giving an explicit
19648
size clause specifying the desired size.
19651
@emph{Convention C/Fortran and Boolean types}
19653
In C, the usual convention for boolean values, that is values used for
19654
conditions, is that zero represents false, and nonzero values represent
19655
true. In Ada, the normal convention is that two specific values, typically
19656
0/1, are used to represent false/true respectively.
19658
Fortran has a similar convention for @cite{LOGICAL} values (any nonzero
19659
value represents true).
19661
To accommodate the Fortran and C conventions, if a pragma Convention specifies
19662
C or Fortran convention for a derived Boolean, as in the following example:
19665
type C_Switch is new Boolean;
19666
pragma Convention (C, C_Switch);
19669
then the GNAT generated code will treat any nonzero value as true. For truth
19670
values generated by GNAT, the conventional value 1 will be used for True, but
19671
when one of these values is read, any nonzero value is treated as True.
19674
@node Conventions and Anonymous Access Types,Determining the Representations chosen by GNAT,Effect of Convention on Representation,Representation Clauses and Pragmas
19675
@anchor{gnat_rm/representation_clauses_and_pragmas conventions-and-anonymous-access-types}@anchor{24d}@anchor{gnat_rm/representation_clauses_and_pragmas id19}@anchor{24e}
19676
@section Conventions and Anonymous Access Types
19679
@geindex Anonymous access types
19681
@geindex Convention for anonymous access types
19683
The RM is not entirely clear on convention handling in a number of cases,
19684
and in particular, it is not clear on the convention to be given to
19685
anonymous access types in general, and in particular what is to be
19686
done for the case of anonymous access-to-subprogram.
19688
In GNAT, we decide that if an explicit Convention is applied
19689
to an object or component, and its type is such an anonymous type,
19690
then the convention will apply to this anonymous type as well. This
19691
seems to make sense since it is anomolous in any case to have a
19692
different convention for an object and its type, and there is clearly
19693
no way to explicitly specify a convention for an anonymous type, since
19694
it doesn't have a name to specify!
19696
Furthermore, we decide that if a convention is applied to a record type,
19697
then this convention is inherited by any of its components that are of an
19698
anonymous access type which do not have an explicitly specified convention.
19700
The following program shows these conventions in action:
19703
package ConvComp is
19704
type Foo is range 1 .. 10;
19706
A : access function (X : Foo) return Integer;
19709
pragma Convention (C, T1);
19712
A : access function (X : Foo) return Integer;
19713
pragma Convention (C, A);
19716
pragma Convention (COBOL, T2);
19719
A : access function (X : Foo) return Integer;
19720
pragma Convention (COBOL, A);
19723
pragma Convention (C, T3);
19726
A : access function (X : Foo) return Integer;
19729
pragma Convention (COBOL, T4);
19731
function F (X : Foo) return Integer;
19732
pragma Convention (C, F);
19734
function F (X : Foo) return Integer is (13);
19736
TV1 : T1 := (F'Access, 12); -- OK
19737
TV2 : T2 := (F'Access, 13); -- OK
19739
TV3 : T3 := (F'Access, 13); -- ERROR
19741
>>> subprogram "F" has wrong convention
19742
>>> does not match access to subprogram declared at line 17
19743
38. TV4 : T4 := (F'Access, 13); -- ERROR
19745
>>> subprogram "F" has wrong convention
19746
>>> does not match access to subprogram declared at line 24
19750
@node Determining the Representations chosen by GNAT,,Conventions and Anonymous Access Types,Representation Clauses and Pragmas
19751
@anchor{gnat_rm/representation_clauses_and_pragmas id20}@anchor{24f}@anchor{gnat_rm/representation_clauses_and_pragmas determining-the-representations-chosen-by-gnat}@anchor{250}
19752
@section Determining the Representations chosen by GNAT
19755
@geindex Representation
19756
@geindex determination of
19758
@geindex -gnatR (gcc)
19760
Although the descriptions in this section are intended to be complete, it is
19761
often easier to simply experiment to see what GNAT accepts and what the
19762
effect is on the layout of types and objects.
19764
As required by the Ada RM, if a representation clause is not accepted, then
19765
it must be rejected as illegal by the compiler. However, when a
19766
representation clause or pragma is accepted, there can still be questions
19767
of what the compiler actually does. For example, if a partial record
19768
representation clause specifies the location of some components and not
19769
others, then where are the non-specified components placed? Or if pragma
19770
@cite{Pack} is used on a record, then exactly where are the resulting
19771
fields placed? The section on pragma @cite{Pack} in this chapter can be
19772
used to answer the second question, but it is often easier to just see
19773
what the compiler does.
19775
For this purpose, GNAT provides the option @emph{-gnatR}. If you compile
19776
with this option, then the compiler will output information on the actual
19777
representations chosen, in a format similar to source representation
19778
clauses. For example, if we compile the package:
19782
type r (x : boolean) is tagged record
19784
when True => S : String (1 .. 100);
19785
when False => null;
19789
type r2 is new r (false) with record
19794
y2 at 16 range 0 .. 31;
19801
type x1 is array (1 .. 10) of x;
19802
for x1'component_size use 11;
19804
type ia is access integer;
19806
type Rb1 is array (1 .. 13) of Boolean;
19809
type Rb2 is array (1 .. 65) of Boolean;
19824
using the switch @emph{-gnatR} we obtain the following output:
19827
Representation information for unit q
19828
-------------------------------------
19831
for r'Alignment use 4;
19833
x at 4 range 0 .. 7;
19834
_tag at 0 range 0 .. 31;
19835
s at 5 range 0 .. 799;
19838
for r2'Size use 160;
19839
for r2'Alignment use 4;
19841
x at 4 range 0 .. 7;
19842
_tag at 0 range 0 .. 31;
19843
_parent at 0 range 0 .. 63;
19844
y2 at 16 range 0 .. 31;
19848
for x'Alignment use 1;
19850
y at 0 range 0 .. 7;
19853
for x1'Size use 112;
19854
for x1'Alignment use 1;
19855
for x1'Component_Size use 11;
19857
for rb1'Size use 13;
19858
for rb1'Alignment use 2;
19859
for rb1'Component_Size use 1;
19861
for rb2'Size use 72;
19862
for rb2'Alignment use 1;
19863
for rb2'Component_Size use 1;
19865
for x2'Size use 224;
19866
for x2'Alignment use 4;
19868
l1 at 0 range 0 .. 0;
19869
l2 at 0 range 1 .. 64;
19870
l3 at 12 range 0 .. 31;
19871
l4 at 16 range 0 .. 0;
19872
l5 at 16 range 1 .. 13;
19873
l6 at 18 range 0 .. 71;
19877
The Size values are actually the Object_Size, i.e., the default size that
19878
will be allocated for objects of the type.
19879
The @code{??} size for type r indicates that we have a variant record, and the
19880
actual size of objects will depend on the discriminant value.
19882
The Alignment values show the actual alignment chosen by the compiler
19883
for each record or array type.
19885
The record representation clause for type r shows where all fields
19886
are placed, including the compiler generated tag field (whose location
19887
cannot be controlled by the programmer).
19889
The record representation clause for the type extension r2 shows all the
19890
fields present, including the parent field, which is a copy of the fields
19891
of the parent type of r2, i.e., r1.
19893
The component size and size clauses for types rb1 and rb2 show
19894
the exact effect of pragma @cite{Pack} on these arrays, and the record
19895
representation clause for type x2 shows how pragma @cite{Pack} affects
19898
In some cases, it may be useful to cut and paste the representation clauses
19899
generated by the compiler into the original source to fix and guarantee
19900
the actual representation to be used.
19902
@node Standard Library Routines,The Implementation of Standard I/O,Representation Clauses and Pragmas,Top
19903
@anchor{gnat_rm/standard_library_routines standard-library-routines}@anchor{e}@anchor{gnat_rm/standard_library_routines doc}@anchor{251}@anchor{gnat_rm/standard_library_routines id1}@anchor{252}
19904
@chapter Standard Library Routines
19907
The Ada Reference Manual contains in Annex A a full description of an
19908
extensive set of standard library routines that can be used in any Ada
19909
program, and which must be provided by all Ada compilers. They are
19910
analogous to the standard C library used by C programs.
19912
GNAT implements all of the facilities described in annex A, and for most
19913
purposes the description in the Ada Reference Manual, or appropriate Ada
19914
text book, will be sufficient for making use of these facilities.
19916
In the case of the input-output facilities,
19917
@ref{f,,The Implementation of Standard I/O},
19918
gives details on exactly how GNAT interfaces to the
19919
file system. For the remaining packages, the Ada Reference Manual
19920
should be sufficient. The following is a list of the packages included,
19921
together with a brief description of the functionality that is provided.
19923
For completeness, references are included to other predefined library
19924
routines defined in other sections of the Ada Reference Manual (these are
19925
cross-indexed from Annex A). For further details see the relevant
19926
package declarations in the run-time library. In particular, a few units
19927
are not implemented, as marked by the presence of pragma Unimplemented_Unit,
19928
and in this case the package declaration contains comments explaining why
19929
the unit is not implemented.
19934
@item @code{Ada} @emph{(A.2)}
19936
This is a parent package for all the standard library packages. It is
19937
usually included implicitly in your program, and itself contains no
19938
useful data or routines.
19940
@item @code{Ada.Assertions} @emph{(11.4.2)}
19942
@cite{Assertions} provides the @cite{Assert} subprograms, and also
19943
the declaration of the @cite{Assertion_Error} exception.
19945
@item @code{Ada.Asynchronous_Task_Control} @emph{(D.11)}
19947
@cite{Asynchronous_Task_Control} provides low level facilities for task
19948
synchronization. It is typically not implemented. See package spec for details.
19950
@item @code{Ada.Calendar} @emph{(9.6)}
19952
@cite{Calendar} provides time of day access, and routines for
19953
manipulating times and durations.
19955
@item @code{Ada.Calendar.Arithmetic} @emph{(9.6.1)}
19957
This package provides additional arithmetic
19958
operations for @cite{Calendar}.
19960
@item @code{Ada.Calendar.Formatting} @emph{(9.6.1)}
19962
This package provides formatting operations for @cite{Calendar}.
19964
@item @code{Ada.Calendar.Time_Zones} @emph{(9.6.1)}
19966
This package provides additional @cite{Calendar} facilities
19967
for handling time zones.
19969
@item @code{Ada.Characters} @emph{(A.3.1)}
19971
This is a dummy parent package that contains no useful entities
19973
@item @code{Ada.Characters.Conversions} @emph{(A.3.2)}
19975
This package provides character conversion functions.
19977
@item @code{Ada.Characters.Handling} @emph{(A.3.2)}
19979
This package provides some basic character handling capabilities,
19980
including classification functions for classes of characters (e.g., test
19981
for letters, or digits).
19983
@item @code{Ada.Characters.Latin_1} @emph{(A.3.3)}
19985
This package includes a complete set of definitions of the characters
19986
that appear in type CHARACTER. It is useful for writing programs that
19987
will run in international environments. For example, if you want an
19988
upper case E with an acute accent in a string, it is often better to use
19989
the definition of @cite{UC_E_Acute} in this package. Then your program
19990
will print in an understandable manner even if your environment does not
19991
support these extended characters.
19993
@item @code{Ada.Command_Line} @emph{(A.15)}
19995
This package provides access to the command line parameters and the name
19996
of the current program (analogous to the use of @cite{argc} and @cite{argv}
19997
in C), and also allows the exit status for the program to be set in a
19998
system-independent manner.
20000
@item @code{Ada.Complex_Text_IO} @emph{(G.1.3)}
20002
This package provides text input and output of complex numbers.
20004
@item @code{Ada.Containers} @emph{(A.18.1)}
20006
A top level package providing a few basic definitions used by all the
20007
following specific child packages that provide specific kinds of
20011
@code{Ada.Containers.Bounded_Priority_Queues} @emph{(A.18.31)}
20013
@code{Ada.Containers.Bounded_Synchronized_Queues} @emph{(A.18.29)}
20015
@code{Ada.Containers.Doubly_Linked_Lists} @emph{(A.18.3)}
20017
@code{Ada.Containers.Generic_Array_Sort} @emph{(A.18.26)}
20019
@code{Ada.Containers.Generic_Constrained_Array_Sort} @emph{(A.18.26)}
20021
@code{Ada.Containers.Generic_Sort} @emph{(A.18.26)}
20023
@code{Ada.Containers.Hashed_Maps} @emph{(A.18.5)}
20025
@code{Ada.Containers.Hashed_Sets} @emph{(A.18.8)}
20027
@code{Ada.Containers.Indefinite_Doubly_Linked_Lists} @emph{(A.18.12)}
20029
@code{Ada.Containers.Indefinite_Hashed_Maps} @emph{(A.18.13)}
20031
@code{Ada.Containers.Indefinite_Hashed_Sets} @emph{(A.18.15)}
20033
@code{Ada.Containers.Indefinite_Holders} @emph{(A.18.18)}
20035
@code{Ada.Containers.Indefinite_Multiway_Trees} @emph{(A.18.17)}
20037
@code{Ada.Containers.Indefinite_Ordered_Maps} @emph{(A.18.14)}
20039
@code{Ada.Containers.Indefinite_Ordered_Sets} @emph{(A.18.16)}
20041
@code{Ada.Containers.Indefinite_Vectors} @emph{(A.18.11)}
20043
@code{Ada.Containers.Multiway_Trees} @emph{(A.18.10)}
20045
@code{Ada.Containers.Ordered_Maps} @emph{(A.18.6)}
20047
@code{Ada.Containers.Ordered_Sets} @emph{(A.18.9)}
20049
@code{Ada.Containers.Synchronized_Queue_Interfaces} @emph{(A.18.27)}
20051
@code{Ada.Containers.Unbounded_Priority_Queues} @emph{(A.18.30)}
20053
@code{Ada.Containers.Unbounded_Synchronized_Queues} @emph{(A.18.28)}
20055
@code{Ada.Containers.Vectors} @emph{(A.18.2)}
20060
@item @code{Ada.Directories} @emph{(A.16)}
20062
This package provides operations on directories.
20064
@item @code{Ada.Directories.Hierarchical_File_Names} @emph{(A.16.1)}
20066
This package provides additional directory operations handling
20067
hiearchical file names.
20069
@item @code{Ada.Directories.Information} @emph{(A.16)}
20071
This is an implementation defined package for additional directory
20072
operations, which is not implemented in GNAT.
20074
@item @code{Ada.Decimal} @emph{(F.2)}
20076
This package provides constants describing the range of decimal numbers
20077
implemented, and also a decimal divide routine (analogous to the COBOL
20078
verb DIVIDE ... GIVING ... REMAINDER ...)
20080
@item @code{Ada.Direct_IO} @emph{(A.8.4)}
20082
This package provides input-output using a model of a set of records of
20083
fixed-length, containing an arbitrary definite Ada type, indexed by an
20084
integer record number.
20086
@item @code{Ada.Dispatching} @emph{(D.2.1)}
20088
A parent package containing definitions for task dispatching operations.
20090
@item @code{Ada.Dispatching.EDF} @emph{(D.2.6)}
20092
Not implemented in GNAT.
20094
@item @code{Ada.Dispatching.Non_Preemptive} @emph{(D.2.4)}
20096
Not implemented in GNAT.
20098
@item @code{Ada.Dispatching.Round_Robin} @emph{(D.2.5)}
20100
Not implemented in GNAT.
20102
@item @code{Ada.Dynamic_Priorities} @emph{(D.5)}
20104
This package allows the priorities of a task to be adjusted dynamically
20105
as the task is running.
20107
@item @code{Ada.Environment_Variables} @emph{(A.17)}
20109
This package provides facilities for accessing environment variables.
20111
@item @code{Ada.Exceptions} @emph{(11.4.1)}
20113
This package provides additional information on exceptions, and also
20114
contains facilities for treating exceptions as data objects, and raising
20115
exceptions with associated messages.
20117
@item @code{Ada.Execution_Time} @emph{(D.14)}
20119
Not implemented in GNAT.
20121
@item @code{Ada.Execution_Time.Group_Budgets} @emph{(D.14.2)}
20123
Not implemented in GNAT.
20125
@item @code{Ada.Execution_Time.Timers} @emph{(D.14.1)'}
20127
Not implemented in GNAT.
20129
@item @code{Ada.Finalization} @emph{(7.6)}
20131
This package contains the declarations and subprograms to support the
20132
use of controlled types, providing for automatic initialization and
20133
finalization (analogous to the constructors and destructors of C++).
20135
@item @code{Ada.Float_Text_IO} @emph{(A.10.9)}
20137
A library level instantiation of Text_IO.Float_IO for type Float.
20139
@item @code{Ada.Float_Wide_Text_IO} @emph{(A.10.9)}
20141
A library level instantiation of Wide_Text_IO.Float_IO for type Float.
20143
@item @code{Ada.Float_Wide_Wide_Text_IO} @emph{(A.10.9)}
20145
A library level instantiation of Wide_Wide_Text_IO.Float_IO for type Float.
20147
@item @code{Ada.Integer_Text_IO} @emph{(A.10.9)}
20149
A library level instantiation of Text_IO.Integer_IO for type Integer.
20151
@item @code{Ada.Integer_Wide_Text_IO} @emph{(A.10.9)}
20153
A library level instantiation of Wide_Text_IO.Integer_IO for type Integer.
20155
@item @code{Ada.Integer_Wide_Wide_Text_IO} @emph{(A.10.9)}
20157
A library level instantiation of Wide_Wide_Text_IO.Integer_IO for type Integer.
20159
@item @code{Ada.Interrupts} @emph{(C.3.2)}
20161
This package provides facilities for interfacing to interrupts, which
20162
includes the set of signals or conditions that can be raised and
20163
recognized as interrupts.
20165
@item @code{Ada.Interrupts.Names} @emph{(C.3.2)}
20167
This package provides the set of interrupt names (actually signal
20168
or condition names) that can be handled by GNAT.
20170
@item @code{Ada.IO_Exceptions} @emph{(A.13)}
20172
This package defines the set of exceptions that can be raised by use of
20173
the standard IO packages.
20175
@item @code{Ada.Iterator_Interfaces} @emph{(5.5.1)}
20177
This package provides a generic interface to generalized iterators.
20179
@item @code{Ada.Locales} @emph{(A.19)}
20181
This package provides declarations providing information (Language
20182
and Country) about the current locale.
20184
@item @code{Ada.Numerics}
20186
This package contains some standard constants and exceptions used
20187
throughout the numerics packages. Note that the constants pi and e are
20188
defined here, and it is better to use these definitions than rolling
20191
@item @code{Ada.Numerics.Complex_Arrays} @emph{(G.3.2)}
20193
Provides operations on arrays of complex numbers.
20195
@item @code{Ada.Numerics.Complex_Elementary_Functions}
20197
Provides the implementation of standard elementary functions (such as
20198
log and trigonometric functions) operating on complex numbers using the
20199
standard @cite{Float} and the @cite{Complex} and @cite{Imaginary} types
20200
created by the package @cite{Numerics.Complex_Types}.
20202
@item @code{Ada.Numerics.Complex_Types}
20204
This is a predefined instantiation of
20205
@cite{Numerics.Generic_Complex_Types} using @cite{Standard.Float} to
20206
build the type @cite{Complex} and @cite{Imaginary}.
20208
@item @code{Ada.Numerics.Discrete_Random}
20210
This generic package provides a random number generator suitable for generating
20211
uniformly distributed values of a specified discrete subtype.
20213
@item @code{Ada.Numerics.Float_Random}
20215
This package provides a random number generator suitable for generating
20216
uniformly distributed floating point values in the unit interval.
20218
@item @code{Ada.Numerics.Generic_Complex_Elementary_Functions}
20220
This is a generic version of the package that provides the
20221
implementation of standard elementary functions (such as log and
20222
trigonometric functions) for an arbitrary complex type.
20224
The following predefined instantiations of this package are provided:
20232
@cite{Ada.Numerics.Short_Complex_Elementary_Functions}
20237
@cite{Ada.Numerics.Complex_Elementary_Functions}
20242
@cite{Ada.Numerics.Long_Complex_Elementary_Functions}
20245
@item @code{Ada.Numerics.Generic_Complex_Types}
20247
This is a generic package that allows the creation of complex types,
20248
with associated complex arithmetic operations.
20250
The following predefined instantiations of this package exist
20258
@cite{Ada.Numerics.Short_Complex_Complex_Types}
20263
@cite{Ada.Numerics.Complex_Complex_Types}
20268
@cite{Ada.Numerics.Long_Complex_Complex_Types}
20271
@item @code{Ada.Numerics.Generic_Elementary_Functions}
20273
This is a generic package that provides the implementation of standard
20274
elementary functions (such as log an trigonometric functions) for an
20275
arbitrary float type.
20277
The following predefined instantiations of this package exist
20285
@cite{Ada.Numerics.Short_Elementary_Functions}
20290
@cite{Ada.Numerics.Elementary_Functions}
20295
@cite{Ada.Numerics.Long_Elementary_Functions}
20298
@item @code{Ada.Numerics.Generic_Real_Arrays} @emph{(G.3.1)}
20300
Generic operations on arrays of reals
20302
@item @code{Ada.Numerics.Real_Arrays} @emph{(G.3.1)}
20304
Preinstantiation of Ada.Numerics.Generic_Real_Arrays (Float).
20306
@item @code{Ada.Real_Time} @emph{(D.8)}
20308
This package provides facilities similar to those of @cite{Calendar}, but
20309
operating with a finer clock suitable for real time control. Note that
20310
annex D requires that there be no backward clock jumps, and GNAT generally
20311
guarantees this behavior, but of course if the external clock on which
20312
the GNAT runtime depends is deliberately reset by some external event,
20313
then such a backward jump may occur.
20315
@item @code{Ada.Real_Time.Timing_Events} @emph{(D.15)}
20317
Not implemented in GNAT.
20319
@item @code{Ada.Sequential_IO} @emph{(A.8.1)}
20321
This package provides input-output facilities for sequential files,
20322
which can contain a sequence of values of a single type, which can be
20323
any Ada type, including indefinite (unconstrained) types.
20325
@item @code{Ada.Storage_IO} @emph{(A.9)}
20327
This package provides a facility for mapping arbitrary Ada types to and
20328
from a storage buffer. It is primarily intended for the creation of new
20331
@item @code{Ada.Streams} @emph{(13.13.1)}
20333
This is a generic package that provides the basic support for the
20334
concept of streams as used by the stream attributes (@cite{Input},
20335
@cite{Output}, @cite{Read} and @cite{Write}).
20337
@item @code{Ada.Streams.Stream_IO} @emph{(A.12.1)}
20339
This package is a specialization of the type @cite{Streams} defined in
20340
package @cite{Streams} together with a set of operations providing
20341
Stream_IO capability. The Stream_IO model permits both random and
20342
sequential access to a file which can contain an arbitrary set of values
20343
of one or more Ada types.
20345
@item @code{Ada.Strings} @emph{(A.4.1)}
20347
This package provides some basic constants used by the string handling
20350
@item @code{Ada.Strings.Bounded} @emph{(A.4.4)}
20352
This package provides facilities for handling variable length
20353
strings. The bounded model requires a maximum length. It is thus
20354
somewhat more limited than the unbounded model, but avoids the use of
20355
dynamic allocation or finalization.
20357
@item @code{Ada.Strings.Bounded.Equal_Case_Insensitive} @emph{(A.4.10)}
20359
Provides case-insensitive comparisons of bounded strings
20361
@item @code{Ada.Strings.Bounded.Hash} @emph{(A.4.9)}
20363
This package provides a generic hash function for bounded strings
20365
@item @code{Ada.Strings.Bounded.Hash_Case_Insensitive} @emph{(A.4.9)}
20367
This package provides a generic hash function for bounded strings that
20368
converts the string to be hashed to lower case.
20370
@item @code{Ada.Strings.Bounded.Less_Case_Insensitive} @emph{(A.4.10)}
20372
This package provides a comparison function for bounded strings that works
20373
in a case insensitive manner by converting to lower case before the comparison.
20375
@item @code{Ada.Strings.Fixed} @emph{(A.4.3)}
20377
This package provides facilities for handling fixed length strings.
20379
@item @code{Ada.Strings.Fixed.Equal_Case_Insensitive} @emph{(A.4.10)}
20381
This package provides an equality function for fixed strings that compares
20382
the strings after converting both to lower case.
20384
@item @code{Ada.Strings.Fixed.Hash_Case_Insensitive} @emph{(A.4.9)}
20386
This package provides a case insensitive hash function for fixed strings that
20387
converts the string to lower case before computing the hash.
20389
@item @code{Ada.Strings.Fixed.Less_Case_Insensitive} @emph{(A.4.10)}
20391
This package provides a comparison function for fixed strings that works
20392
in a case insensitive manner by converting to lower case before the comparison.
20394
@item @code{Ada.Strings.Hash} @emph{(A.4.9)}
20396
This package provides a hash function for strings.
20398
@item @code{Ada.Strings.Hash_Case_Insensitive} @emph{(A.4.9)}
20400
This package provides a hash function for strings that is case insensitive.
20401
The string is converted to lower case before computing the hash.
20403
@item @code{Ada.Strings.Less_Case_Insensitive} @emph{(A.4.10)}
20405
This package provides a comparison function for\strings that works
20406
in a case insensitive manner by converting to lower case before the comparison.
20408
@item @code{Ada.Strings.Maps} @emph{(A.4.2)}
20410
This package provides facilities for handling character mappings and
20411
arbitrarily defined subsets of characters. For instance it is useful in
20412
defining specialized translation tables.
20414
@item @code{Ada.Strings.Maps.Constants} @emph{(A.4.6)}
20416
This package provides a standard set of predefined mappings and
20417
predefined character sets. For example, the standard upper to lower case
20418
conversion table is found in this package. Note that upper to lower case
20419
conversion is non-trivial if you want to take the entire set of
20420
characters, including extended characters like E with an acute accent,
20421
into account. You should use the mappings in this package (rather than
20422
adding 32 yourself) to do case mappings.
20424
@item @code{Ada.Strings.Unbounded} @emph{(A.4.5)}
20426
This package provides facilities for handling variable length
20427
strings. The unbounded model allows arbitrary length strings, but
20428
requires the use of dynamic allocation and finalization.
20430
@item @code{Ada.Strings.Unbounded.Equal_Case_Insensitive} @emph{(A.4.10)}
20432
Provides case-insensitive comparisons of unbounded strings
20434
@item @code{Ada.Strings.Unbounded.Hash} @emph{(A.4.9)}
20436
This package provides a generic hash function for unbounded strings
20438
@item @code{Ada.Strings.Unbounded.Hash_Case_Insensitive} @emph{(A.4.9)}
20440
This package provides a generic hash function for unbounded strings that
20441
converts the string to be hashed to lower case.
20443
@item @code{Ada.Strings.Unbounded.Less_Case_Insensitive} @emph{(A.4.10)}
20445
This package provides a comparison function for unbounded strings that works
20446
in a case insensitive manner by converting to lower case before the comparison.
20448
@item @code{Ada.Strings.UTF_Encoding} @emph{(A.4.11)}
20450
This package provides basic definitions for dealing with UTF-encoded strings.
20452
@item @code{Ada.Strings.UTF_Encoding.Conversions} @emph{(A.4.11)}
20454
This package provides conversion functions for UTF-encoded strings.
20457
@code{Ada.Strings.UTF_Encoding.Strings} @emph{(A.4.11)}
20459
@code{Ada.Strings.UTF_Encoding.Wide_Strings} @emph{(A.4.11)}
20464
@item @code{Ada.Strings.UTF_Encoding.Wide_Wide_Strings} @emph{(A.4.11)}
20466
These packages provide facilities for handling UTF encodings for
20467
Strings, Wide_Strings and Wide_Wide_Strings.
20470
@code{Ada.Strings.Wide_Bounded} @emph{(A.4.7)}
20472
@code{Ada.Strings.Wide_Fixed} @emph{(A.4.7)}
20474
@code{Ada.Strings.Wide_Maps} @emph{(A.4.7)}
20479
@item @code{Ada.Strings.Wide_Unbounded} @emph{(A.4.7)}
20481
These packages provide analogous capabilities to the corresponding
20482
packages without @code{Wide_} in the name, but operate with the types
20483
@cite{Wide_String} and @cite{Wide_Character} instead of @cite{String}
20484
and @cite{Character}. Versions of all the child packages are available.
20487
@code{Ada.Strings.Wide_Wide_Bounded} @emph{(A.4.7)}
20489
@code{Ada.Strings.Wide_Wide_Fixed} @emph{(A.4.7)}
20491
@code{Ada.Strings.Wide_Wide_Maps} @emph{(A.4.7)}
20496
@item @code{Ada.Strings.Wide_Wide_Unbounded} @emph{(A.4.7)}
20498
These packages provide analogous capabilities to the corresponding
20499
packages without @code{Wide_} in the name, but operate with the types
20500
@cite{Wide_Wide_String} and @cite{Wide_Wide_Character} instead
20501
of @cite{String} and @cite{Character}.
20503
@item @code{Ada.Synchronous_Barriers} @emph{(D.10.1)}
20505
This package provides facilities for synchronizing tasks at a low level
20508
@item @code{Ada.Synchronous_Task_Control} @emph{(D.10)}
20510
This package provides some standard facilities for controlling task
20511
communication in a synchronous manner.
20513
@item @code{Ada.Synchronous_Task_Control.EDF} @emph{(D.10)}
20515
Not implemented in GNAT.
20517
@item @code{Ada.Tags}
20519
This package contains definitions for manipulation of the tags of tagged
20522
@item @code{Ada.Tags.Generic_Dispatching_Constructor} @emph{(3.9)}
20524
This package provides a way of constructing tagged class-wide values given
20525
only the tag value.
20527
@item @code{Ada.Task_Attributes} @emph{(C.7.2)}
20529
This package provides the capability of associating arbitrary
20530
task-specific data with separate tasks.
20532
@item @code{Ada.Task_Identifification} @emph{(C.7.1)}
20534
This package provides capabilities for task identification.
20536
@item @code{Ada.Task_Termination} @emph{(C.7.3)}
20538
This package provides control over task termination.
20540
@item @code{Ada.Text_IO}
20542
This package provides basic text input-output capabilities for
20543
character, string and numeric data. The subpackages of this
20544
package are listed next. Note that although these are defined
20545
as subpackages in the RM, they are actually transparently
20546
implemented as child packages in GNAT, meaning that they
20547
are only loaded if needed.
20549
@item @code{Ada.Text_IO.Decimal_IO}
20551
Provides input-output facilities for decimal fixed-point types
20553
@item @code{Ada.Text_IO.Enumeration_IO}
20555
Provides input-output facilities for enumeration types.
20557
@item @code{Ada.Text_IO.Fixed_IO}
20559
Provides input-output facilities for ordinary fixed-point types.
20561
@item @code{Ada.Text_IO.Float_IO}
20563
Provides input-output facilities for float types. The following
20564
predefined instantiations of this generic package are available:
20572
@cite{Short_Float_Text_IO}
20577
@cite{Float_Text_IO}
20582
@cite{Long_Float_Text_IO}
20585
@item @code{Ada.Text_IO.Integer_IO}
20587
Provides input-output facilities for integer types. The following
20588
predefined instantiations of this generic package are available:
20594
@code{Short_Short_Integer}
20596
@cite{Ada.Short_Short_Integer_Text_IO}
20599
@code{Short_Integer}
20601
@cite{Ada.Short_Integer_Text_IO}
20606
@cite{Ada.Integer_Text_IO}
20609
@code{Long_Integer}
20611
@cite{Ada.Long_Integer_Text_IO}
20614
@code{Long_Long_Integer}
20616
@cite{Ada.Long_Long_Integer_Text_IO}
20619
@item @code{Ada.Text_IO.Modular_IO}
20621
Provides input-output facilities for modular (unsigned) types.
20623
@item @code{Ada.Text_IO.Bounded_IO (A.10.11)}
20625
Provides input-output facilities for bounded strings.
20627
@item @code{Ada.Text_IO.Complex_IO (G.1.3)}
20629
This package provides basic text input-output capabilities for complex
20632
@item @code{Ada.Text_IO.Editing (F.3.3)}
20634
This package contains routines for edited output, analogous to the use
20635
of pictures in COBOL. The picture formats used by this package are a
20636
close copy of the facility in COBOL.
20638
@item @code{Ada.Text_IO.Text_Streams (A.12.2)}
20640
This package provides a facility that allows Text_IO files to be treated
20641
as streams, so that the stream attributes can be used for writing
20642
arbitrary data, including binary data, to Text_IO files.
20644
@item @code{Ada.Text_IO.Unbounded_IO (A.10.12)}
20646
This package provides input-output facilities for unbounded strings.
20648
@item @code{Ada.Unchecked_Conversion (13.9)}
20650
This generic package allows arbitrary conversion from one type to
20651
another of the same size, providing for breaking the type safety in
20652
special circumstances.
20654
If the types have the same Size (more accurately the same Value_Size),
20655
then the effect is simply to transfer the bits from the source to the
20656
target type without any modification. This usage is well defined, and
20657
for simple types whose representation is typically the same across
20658
all implementations, gives a portable method of performing such
20661
If the types do not have the same size, then the result is implementation
20662
defined, and thus may be non-portable. The following describes how GNAT
20663
handles such unchecked conversion cases.
20665
If the types are of different sizes, and are both discrete types, then
20666
the effect is of a normal type conversion without any constraint checking.
20667
In particular if the result type has a larger size, the result will be
20668
zero or sign extended. If the result type has a smaller size, the result
20669
will be truncated by ignoring high order bits.
20671
If the types are of different sizes, and are not both discrete types,
20672
then the conversion works as though pointers were created to the source
20673
and target, and the pointer value is converted. The effect is that bits
20674
are copied from successive low order storage units and bits of the source
20675
up to the length of the target type.
20677
A warning is issued if the lengths differ, since the effect in this
20678
case is implementation dependent, and the above behavior may not match
20679
that of some other compiler.
20681
A pointer to one type may be converted to a pointer to another type using
20682
unchecked conversion. The only case in which the effect is undefined is
20683
when one or both pointers are pointers to unconstrained array types. In
20684
this case, the bounds information may get incorrectly transferred, and in
20685
particular, GNAT uses double size pointers for such types, and it is
20686
meaningless to convert between such pointer types. GNAT will issue a
20687
warning if the alignment of the target designated type is more strict
20688
than the alignment of the source designated type (since the result may
20689
be unaligned in this case).
20691
A pointer other than a pointer to an unconstrained array type may be
20692
converted to and from System.Address. Such usage is common in Ada 83
20693
programs, but note that Ada.Address_To_Access_Conversions is the
20694
preferred method of performing such conversions in Ada 95 and Ada 2005.
20696
unchecked conversion nor Ada.Address_To_Access_Conversions should be
20697
used in conjunction with pointers to unconstrained objects, since
20698
the bounds information cannot be handled correctly in this case.
20700
@item @code{Ada.Unchecked_Deallocation} @emph{(13.11.2)}
20702
This generic package allows explicit freeing of storage previously
20703
allocated by use of an allocator.
20705
@item @code{Ada.Wide_Text_IO} @emph{(A.11)}
20707
This package is similar to @cite{Ada.Text_IO}, except that the external
20708
file supports wide character representations, and the internal types are
20709
@cite{Wide_Character} and @cite{Wide_String} instead of @cite{Character}
20710
and @cite{String}. The corresponding set of nested packages and child
20711
packages are defined.
20713
@item @code{Ada.Wide_Wide_Text_IO} @emph{(A.11)}
20715
This package is similar to @cite{Ada.Text_IO}, except that the external
20716
file supports wide character representations, and the internal types are
20717
@cite{Wide_Character} and @cite{Wide_String} instead of @cite{Character}
20718
and @cite{String}. The corresponding set of nested packages and child
20719
packages are defined.
20722
For packages in Interfaces and System, all the RM defined packages are
20723
available in GNAT, see the Ada 2012 RM for full details.
20725
@node The Implementation of Standard I/O,The GNAT Library,Standard Library Routines,Top
20726
@anchor{gnat_rm/the_implementation_of_standard_i_o the-implementation-of-standard-i-o}@anchor{f}@anchor{gnat_rm/the_implementation_of_standard_i_o doc}@anchor{253}@anchor{gnat_rm/the_implementation_of_standard_i_o id1}@anchor{254}
20727
@chapter The Implementation of Standard I/O
20730
GNAT implements all the required input-output facilities described in
20731
A.6 through A.14. These sections of the Ada Reference Manual describe the
20732
required behavior of these packages from the Ada point of view, and if
20733
you are writing a portable Ada program that does not need to know the
20734
exact manner in which Ada maps to the outside world when it comes to
20735
reading or writing external files, then you do not need to read this
20736
chapter. As long as your files are all regular files (not pipes or
20737
devices), and as long as you write and read the files only from Ada, the
20738
description in the Ada Reference Manual is sufficient.
20740
However, if you want to do input-output to pipes or other devices, such
20741
as the keyboard or screen, or if the files you are dealing with are
20742
either generated by some other language, or to be read by some other
20743
language, then you need to know more about the details of how the GNAT
20744
implementation of these input-output facilities behaves.
20746
In this chapter we give a detailed description of exactly how GNAT
20747
interfaces to the file system. As always, the sources of the system are
20748
available to you for answering questions at an even more detailed level,
20749
but for most purposes the information in this chapter will suffice.
20751
Another reason that you may need to know more about how input-output is
20752
implemented arises when you have a program written in mixed languages
20753
where, for example, files are shared between the C and Ada sections of
20754
the same program. GNAT provides some additional facilities, in the form
20755
of additional child library packages, that facilitate this sharing, and
20756
these additional facilities are also described in this chapter.
20759
* Standard I/O Packages::
20765
* Wide_Wide_Text_IO::
20767
* Text Translation::
20769
* Filenames encoding::
20770
* File content encoding::
20772
* Operations on C Streams::
20773
* Interfacing to C Streams::
20777
@node Standard I/O Packages,FORM Strings,,The Implementation of Standard I/O
20778
@anchor{gnat_rm/the_implementation_of_standard_i_o standard-i-o-packages}@anchor{255}@anchor{gnat_rm/the_implementation_of_standard_i_o id2}@anchor{256}
20779
@section Standard I/O Packages
20782
The Standard I/O packages described in Annex A for
20791
Ada.Text_IO.Complex_IO
20794
Ada.Text_IO.Text_Streams
20800
Ada.Wide_Text_IO.Complex_IO
20803
Ada.Wide_Text_IO.Text_Streams
20806
Ada.Wide_Wide_Text_IO
20809
Ada.Wide_Wide_Text_IO.Complex_IO
20812
Ada.Wide_Wide_Text_IO.Text_Streams
20824
are implemented using the C
20825
library streams facility; where
20831
All files are opened using @cite{fopen}.
20834
All input/output operations use @cite{fread}/@cite{fwrite}.
20837
There is no internal buffering of any kind at the Ada library level. The only
20838
buffering is that provided at the system level in the implementation of the
20839
library routines that support streams. This facilitates shared use of these
20840
streams by mixed language programs. Note though that system level buffering is
20841
explicitly enabled at elaboration of the standard I/O packages and that can
20842
have an impact on mixed language programs, in particular those using I/O before
20843
calling the Ada elaboration routine (e.g., adainit). It is recommended to call
20844
the Ada elaboration routine before performing any I/O or when impractical,
20845
flush the common I/O streams and in particular Standard_Output before
20846
elaborating the Ada code.
20848
@node FORM Strings,Direct_IO,Standard I/O Packages,The Implementation of Standard I/O
20849
@anchor{gnat_rm/the_implementation_of_standard_i_o form-strings}@anchor{257}@anchor{gnat_rm/the_implementation_of_standard_i_o id3}@anchor{258}
20850
@section FORM Strings
20853
The format of a FORM string in GNAT is:
20856
"keyword=value,keyword=value,...,keyword=value"
20859
where letters may be in upper or lower case, and there are no spaces
20860
between values. The order of the entries is not important. Currently
20861
the following keywords defined.
20864
TEXT_TRANSLATION=[YES|NO|TEXT|BINARY|U8TEXT|WTEXT|U16TEXT]
20866
WCEM=[n|h|u|s|e|8|b]
20867
ENCODING=[UTF8|8BITS]
20870
The use of these parameters is described later in this section. If an
20871
unrecognized keyword appears in a form string, it is silently ignored
20872
and not considered invalid.
20874
@node Direct_IO,Sequential_IO,FORM Strings,The Implementation of Standard I/O
20875
@anchor{gnat_rm/the_implementation_of_standard_i_o direct-io}@anchor{259}@anchor{gnat_rm/the_implementation_of_standard_i_o id4}@anchor{25a}
20879
Direct_IO can only be instantiated for definite types. This is a
20880
restriction of the Ada language, which means that the records are fixed
20881
length (the length being determined by @code{type'Size}, rounded
20882
up to the next storage unit boundary if necessary).
20884
The records of a Direct_IO file are simply written to the file in index
20885
sequence, with the first record starting at offset zero, and subsequent
20886
records following. There is no control information of any kind. For
20887
example, if 32-bit integers are being written, each record takes
20888
4-bytes, so the record at index @cite{K} starts at offset
20891
There is no limit on the size of Direct_IO files, they are expanded as
20892
necessary to accommodate whatever records are written to the file.
20894
@node Sequential_IO,Text_IO,Direct_IO,The Implementation of Standard I/O
20895
@anchor{gnat_rm/the_implementation_of_standard_i_o sequential-io}@anchor{25b}@anchor{gnat_rm/the_implementation_of_standard_i_o id5}@anchor{25c}
20896
@section Sequential_IO
20899
Sequential_IO may be instantiated with either a definite (constrained)
20900
or indefinite (unconstrained) type.
20902
For the definite type case, the elements written to the file are simply
20903
the memory images of the data values with no control information of any
20904
kind. The resulting file should be read using the same type, no validity
20905
checking is performed on input.
20907
For the indefinite type case, the elements written consist of two
20908
parts. First is the size of the data item, written as the memory image
20909
of a @cite{Interfaces.C.size_t} value, followed by the memory image of
20910
the data value. The resulting file can only be read using the same
20911
(unconstrained) type. Normal assignment checks are performed on these
20912
read operations, and if these checks fail, @cite{Data_Error} is
20913
raised. In particular, in the array case, the lengths must match, and in
20914
the variant record case, if the variable for a particular read operation
20915
is constrained, the discriminants must match.
20917
Note that it is not possible to use Sequential_IO to write variable
20918
length array items, and then read the data back into different length
20919
arrays. For example, the following will raise @cite{Data_Error}:
20922
package IO is new Sequential_IO (String);
20927
IO.Write (F, "hello!")
20928
IO.Reset (F, Mode=>In_File);
20933
On some Ada implementations, this will print @cite{hell}, but the program is
20934
clearly incorrect, since there is only one element in the file, and that
20935
element is the string @cite{hello!}.
20937
In Ada 95 and Ada 2005, this kind of behavior can be legitimately achieved
20938
using Stream_IO, and this is the preferred mechanism. In particular, the
20939
above program fragment rewritten to use Stream_IO will work correctly.
20941
@node Text_IO,Wide_Text_IO,Sequential_IO,The Implementation of Standard I/O
20942
@anchor{gnat_rm/the_implementation_of_standard_i_o id6}@anchor{25d}@anchor{gnat_rm/the_implementation_of_standard_i_o text-io}@anchor{25e}
20946
Text_IO files consist of a stream of characters containing the following
20947
special control characters:
20950
LF (line feed, 16#0A#) Line Mark
20951
FF (form feed, 16#0C#) Page Mark
20954
A canonical Text_IO file is defined as one in which the following
20955
conditions are met:
20961
The character @cite{LF} is used only as a line mark, i.e., to mark the end
20965
The character @cite{FF} is used only as a page mark, i.e., to mark the
20966
end of a page and consequently can appear only immediately following a
20967
@cite{LF} (line mark) character.
20970
The file ends with either @cite{LF} (line mark) or @cite{LF}-@cite{FF}
20971
(line mark, page mark). In the former case, the page mark is implicitly
20972
assumed to be present.
20975
A file written using Text_IO will be in canonical form provided that no
20976
explicit @cite{LF} or @cite{FF} characters are written using @cite{Put}
20977
or @cite{Put_Line}. There will be no @cite{FF} character at the end of
20978
the file unless an explicit @cite{New_Page} operation was performed
20979
before closing the file.
20981
A canonical Text_IO file that is a regular file (i.e., not a device or a
20982
pipe) can be read using any of the routines in Text_IO. The
20983
semantics in this case will be exactly as defined in the Ada Reference
20984
Manual, and all the routines in Text_IO are fully implemented.
20986
A text file that does not meet the requirements for a canonical Text_IO
20987
file has one of the following:
20993
The file contains @cite{FF} characters not immediately following a
20994
@cite{LF} character.
20997
The file contains @cite{LF} or @cite{FF} characters written by
20998
@cite{Put} or @cite{Put_Line}, which are not logically considered to be
20999
line marks or page marks.
21002
The file ends in a character other than @cite{LF} or @cite{FF},
21003
i.e., there is no explicit line mark or page mark at the end of the file.
21006
Text_IO can be used to read such non-standard text files but subprograms
21007
to do with line or page numbers do not have defined meanings. In
21008
particular, a @cite{FF} character that does not follow a @cite{LF}
21009
character may or may not be treated as a page mark from the point of
21010
view of page and line numbering. Every @cite{LF} character is considered
21011
to end a line, and there is an implied @cite{LF} character at the end of
21015
* Stream Pointer Positioning::
21016
* Reading and Writing Non-Regular Files::
21018
* Treating Text_IO Files as Streams::
21019
* Text_IO Extensions::
21020
* Text_IO Facilities for Unbounded Strings::
21024
@node Stream Pointer Positioning,Reading and Writing Non-Regular Files,,Text_IO
21025
@anchor{gnat_rm/the_implementation_of_standard_i_o id7}@anchor{25f}@anchor{gnat_rm/the_implementation_of_standard_i_o stream-pointer-positioning}@anchor{260}
21026
@subsection Stream Pointer Positioning
21029
@cite{Ada.Text_IO} has a definition of current position for a file that
21030
is being read. No internal buffering occurs in Text_IO, and usually the
21031
physical position in the stream used to implement the file corresponds
21032
to this logical position defined by Text_IO. There are two exceptions:
21038
After a call to @cite{End_Of_Page} that returns @cite{True}, the stream
21039
is positioned past the @cite{LF} (line mark) that precedes the page
21040
mark. Text_IO maintains an internal flag so that subsequent read
21041
operations properly handle the logical position which is unchanged by
21042
the @cite{End_Of_Page} call.
21045
After a call to @cite{End_Of_File} that returns @cite{True}, if the
21046
Text_IO file was positioned before the line mark at the end of file
21047
before the call, then the logical position is unchanged, but the stream
21048
is physically positioned right at the end of file (past the line mark,
21049
and past a possible page mark following the line mark. Again Text_IO
21050
maintains internal flags so that subsequent read operations properly
21051
handle the logical position.
21054
These discrepancies have no effect on the observable behavior of
21055
Text_IO, but if a single Ada stream is shared between a C program and
21056
Ada program, or shared (using @code{shared=yes} in the form string)
21057
between two Ada files, then the difference may be observable in some
21060
@node Reading and Writing Non-Regular Files,Get_Immediate,Stream Pointer Positioning,Text_IO
21061
@anchor{gnat_rm/the_implementation_of_standard_i_o reading-and-writing-non-regular-files}@anchor{261}@anchor{gnat_rm/the_implementation_of_standard_i_o id8}@anchor{262}
21062
@subsection Reading and Writing Non-Regular Files
21065
A non-regular file is a device (such as a keyboard), or a pipe. Text_IO
21066
can be used for reading and writing. Writing is not affected and the
21067
sequence of characters output is identical to the normal file case, but
21068
for reading, the behavior of Text_IO is modified to avoid undesirable
21069
look-ahead as follows:
21071
An input file that is not a regular file is considered to have no page
21072
marks. Any @cite{Ascii.FF} characters (the character normally used for a
21073
page mark) appearing in the file are considered to be data
21074
characters. In particular:
21080
@cite{Get_Line} and @cite{Skip_Line} do not test for a page mark
21081
following a line mark. If a page mark appears, it will be treated as a
21085
This avoids the need to wait for an extra character to be typed or
21086
entered from the pipe to complete one of these operations.
21089
@cite{End_Of_Page} always returns @cite{False}
21092
@cite{End_Of_File} will return @cite{False} if there is a page mark at
21093
the end of the file.
21096
Output to non-regular files is the same as for regular files. Page marks
21097
may be written to non-regular files using @cite{New_Page}, but as noted
21098
above they will not be treated as page marks on input if the output is
21099
piped to another Ada program.
21101
Another important discrepancy when reading non-regular files is that the end
21102
of file indication is not 'sticky'. If an end of file is entered, e.g., by
21103
pressing the @code{EOT} key,
21105
is signaled once (i.e., the test @cite{End_Of_File}
21106
will yield @cite{True}, or a read will
21107
raise @cite{End_Error}), but then reading can resume
21108
to read data past that end of
21109
file indication, until another end of file indication is entered.
21111
@node Get_Immediate,Treating Text_IO Files as Streams,Reading and Writing Non-Regular Files,Text_IO
21112
@anchor{gnat_rm/the_implementation_of_standard_i_o get-immediate}@anchor{263}@anchor{gnat_rm/the_implementation_of_standard_i_o id9}@anchor{264}
21113
@subsection Get_Immediate
21116
@geindex Get_Immediate
21118
Get_Immediate returns the next character (including control characters)
21119
from the input file. In particular, Get_Immediate will return LF or FF
21120
characters used as line marks or page marks. Such operations leave the
21121
file positioned past the control character, and it is thus not treated
21122
as having its normal function. This means that page, line and column
21123
counts after this kind of Get_Immediate call are set as though the mark
21124
did not occur. In the case where a Get_Immediate leaves the file
21125
positioned between the line mark and page mark (which is not normally
21126
possible), it is undefined whether the FF character will be treated as a
21129
@node Treating Text_IO Files as Streams,Text_IO Extensions,Get_Immediate,Text_IO
21130
@anchor{gnat_rm/the_implementation_of_standard_i_o id10}@anchor{265}@anchor{gnat_rm/the_implementation_of_standard_i_o treating-text-io-files-as-streams}@anchor{266}
21131
@subsection Treating Text_IO Files as Streams
21134
@geindex Stream files
21136
The package @cite{Text_IO.Streams} allows a Text_IO file to be treated
21137
as a stream. Data written to a Text_IO file in this stream mode is
21138
binary data. If this binary data contains bytes 16#0A# (@cite{LF}) or
21139
16#0C# (@cite{FF}), the resulting file may have non-standard
21140
format. Similarly if read operations are used to read from a Text_IO
21141
file treated as a stream, then @cite{LF} and @cite{FF} characters may be
21142
skipped and the effect is similar to that described above for
21143
@cite{Get_Immediate}.
21145
@node Text_IO Extensions,Text_IO Facilities for Unbounded Strings,Treating Text_IO Files as Streams,Text_IO
21146
@anchor{gnat_rm/the_implementation_of_standard_i_o id11}@anchor{267}@anchor{gnat_rm/the_implementation_of_standard_i_o text-io-extensions}@anchor{268}
21147
@subsection Text_IO Extensions
21150
@geindex Text_IO extensions
21152
A package GNAT.IO_Aux in the GNAT library provides some useful extensions
21153
to the standard @cite{Text_IO} package:
21159
function File_Exists (Name : String) return Boolean;
21160
Determines if a file of the given name exists.
21163
function Get_Line return String;
21164
Reads a string from the standard input file. The value returned is exactly
21165
the length of the line that was read.
21168
function Get_Line (File : Ada.Text_IO.File_Type) return String;
21169
Similar, except that the parameter File specifies the file from which
21170
the string is to be read.
21173
@node Text_IO Facilities for Unbounded Strings,,Text_IO Extensions,Text_IO
21174
@anchor{gnat_rm/the_implementation_of_standard_i_o text-io-facilities-for-unbounded-strings}@anchor{269}@anchor{gnat_rm/the_implementation_of_standard_i_o id12}@anchor{26a}
21175
@subsection Text_IO Facilities for Unbounded Strings
21178
@geindex Text_IO for unbounded strings
21180
@geindex Unbounded_String
21181
@geindex Text_IO operations
21183
The package @cite{Ada.Strings.Unbounded.Text_IO}
21184
in library files @cite{a-suteio.ads/adb} contains some GNAT-specific
21185
subprograms useful for Text_IO operations on unbounded strings:
21191
function Get_Line (File : File_Type) return Unbounded_String;
21192
Reads a line from the specified file
21193
and returns the result as an unbounded string.
21196
procedure Put (File : File_Type; U : Unbounded_String);
21197
Writes the value of the given unbounded string to the specified file
21198
Similar to the effect of
21199
@cite{Put (To_String (U))} except that an extra copy is avoided.
21202
procedure Put_Line (File : File_Type; U : Unbounded_String);
21203
Writes the value of the given unbounded string to the specified file,
21204
followed by a @cite{New_Line}.
21205
Similar to the effect of @cite{Put_Line (To_String (U))} except
21206
that an extra copy is avoided.
21209
In the above procedures, @cite{File} is of type @cite{Ada.Text_IO.File_Type}
21210
and is optional. If the parameter is omitted, then the standard input or
21211
output file is referenced as appropriate.
21213
The package @cite{Ada.Strings.Wide_Unbounded.Wide_Text_IO} in library
21214
files @code{a-swuwti.ads} and @code{a-swuwti.adb} provides similar extended
21215
@cite{Wide_Text_IO} functionality for unbounded wide strings.
21217
The package @cite{Ada.Strings.Wide_Wide_Unbounded.Wide_Wide_Text_IO} in library
21218
files @code{a-szuzti.ads} and @code{a-szuzti.adb} provides similar extended
21219
@cite{Wide_Wide_Text_IO} functionality for unbounded wide wide strings.
21221
@node Wide_Text_IO,Wide_Wide_Text_IO,Text_IO,The Implementation of Standard I/O
21222
@anchor{gnat_rm/the_implementation_of_standard_i_o wide-text-io}@anchor{26b}@anchor{gnat_rm/the_implementation_of_standard_i_o id13}@anchor{26c}
21223
@section Wide_Text_IO
21226
@cite{Wide_Text_IO} is similar in most respects to Text_IO, except that
21227
both input and output files may contain special sequences that represent
21228
wide character values. The encoding scheme for a given file may be
21229
specified using a FORM parameter:
21235
as part of the FORM string (WCEM = wide character encoding method),
21236
where @cite{x} is one of the following characters
21239
@multitable {xxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxx}
21262
Upper half encoding
21299
The encoding methods match those that
21300
can be used in a source
21301
program, but there is no requirement that the encoding method used for
21302
the source program be the same as the encoding method used for files,
21303
and different files may use different encoding methods.
21305
The default encoding method for the standard files, and for opened files
21306
for which no WCEM parameter is given in the FORM string matches the
21307
wide character encoding specified for the main program (the default
21308
being brackets encoding if no coding method was specified with -gnatW).
21313
@item @emph{Hex Coding}
21315
In this encoding, a wide character is represented by a five character
21326
where @cite{a}, @cite{b}, @cite{c}, @cite{d} are the four hexadecimal
21327
characters (using upper case letters) of the wide character code. For
21328
example, ESC A345 is used to represent the wide character with code
21329
16#A345#. This scheme is compatible with use of the full
21330
@cite{Wide_Character} set.
21336
@item @emph{Upper Half Coding}
21338
The wide character with encoding 16#abcd#, where the upper bit is on
21339
(i.e., a is in the range 8-F) is represented as two bytes 16#ab# and
21340
16#cd#. The second byte may never be a format control character, but is
21341
not required to be in the upper half. This method can be also used for
21342
shift-JIS or EUC where the internal coding matches the external coding.
21344
@item @emph{Shift JIS Coding}
21346
A wide character is represented by a two character sequence 16#ab# and
21347
16#cd#, with the restrictions described for upper half encoding as
21348
described above. The internal character code is the corresponding JIS
21349
character according to the standard algorithm for Shift-JIS
21350
conversion. Only characters defined in the JIS code set table can be
21351
used with this encoding method.
21353
@item @emph{EUC Coding}
21355
A wide character is represented by a two character sequence 16#ab# and
21356
16#cd#, with both characters being in the upper half. The internal
21357
character code is the corresponding JIS character according to the EUC
21358
encoding algorithm. Only characters defined in the JIS code set table
21359
can be used with this encoding method.
21361
@item @emph{UTF-8 Coding}
21363
A wide character is represented using
21364
UCS Transformation Format 8 (UTF-8) as defined in Annex R of ISO
21365
10646-1/Am.2. Depending on the character value, the representation
21366
is a one, two, or three byte sequence:
21370
16#0000#-16#007f#: 2#0xxxxxxx#
21371
16#0080#-16#07ff#: 2#110xxxxx# 2#10xxxxxx#
21372
16#0800#-16#ffff#: 2#1110xxxx# 2#10xxxxxx# 2#10xxxxxx#
21378
where the @cite{xxx} bits correspond to the left-padded bits of the
21379
16-bit character value. Note that all lower half ASCII characters
21380
are represented as ASCII bytes and all upper half characters and
21381
other wide characters are represented as sequences of upper-half
21382
(The full UTF-8 scheme allows for encoding 31-bit characters as
21383
6-byte sequences, but in this implementation, all UTF-8 sequences
21384
of four or more bytes length will raise a Constraint_Error, as
21385
will all invalid UTF-8 sequences.)
21391
@item @emph{Brackets Coding}
21393
In this encoding, a wide character is represented by the following eight
21394
character sequence:
21404
where @cite{a}, @cite{b}, @cite{c}, @cite{d} are the four hexadecimal
21405
characters (using uppercase letters) of the wide character code. For
21406
example, @cite{["A345"]} is used to represent the wide character with code
21408
This scheme is compatible with use of the full Wide_Character set.
21409
On input, brackets coding can also be used for upper half characters,
21410
e.g., @cite{["C1"]} for lower case a. However, on output, brackets notation
21411
is only used for wide characters with a code greater than @cite{16#FF#}.
21413
Note that brackets coding is not normally used in the context of
21414
Wide_Text_IO or Wide_Wide_Text_IO, since it is really just designed as
21415
a portable way of encoding source files. In the context of Wide_Text_IO
21416
or Wide_Wide_Text_IO, it can only be used if the file does not contain
21417
any instance of the left bracket character other than to encode wide
21418
character values using the brackets encoding method. In practice it is
21419
expected that some standard wide character encoding method such
21420
as UTF-8 will be used for text input output.
21422
If brackets notation is used, then any occurrence of a left bracket
21423
in the input file which is not the start of a valid wide character
21424
sequence will cause Constraint_Error to be raised. It is possible to
21425
encode a left bracket as ["5B"] and Wide_Text_IO and Wide_Wide_Text_IO
21426
input will interpret this as a left bracket.
21428
However, when a left bracket is output, it will be output as a left bracket
21429
and not as ["5B"]. We make this decision because for normal use of
21430
Wide_Text_IO for outputting messages, it is unpleasant to clobber left
21431
brackets. For example, if we write:
21434
Put_Line ("Start of output [first run]");
21437
we really do not want to have the left bracket in this message clobbered so
21438
that the output reads:
21442
Start of output ["5B"]first run]
21448
In practice brackets encoding is reasonably useful for normal Put_Line use
21449
since we won't get confused between left brackets and wide character
21450
sequences in the output. But for input, or when files are written out
21451
and read back in, it really makes better sense to use one of the standard
21452
encoding methods such as UTF-8.
21455
For the coding schemes other than UTF-8, Hex, or Brackets encoding,
21456
not all wide character
21457
values can be represented. An attempt to output a character that cannot
21458
be represented using the encoding scheme for the file causes
21459
Constraint_Error to be raised. An invalid wide character sequence on
21460
input also causes Constraint_Error to be raised.
21463
* Stream Pointer Positioning: Stream Pointer Positioning<2>.
21464
* Reading and Writing Non-Regular Files: Reading and Writing Non-Regular Files<2>.
21468
@node Stream Pointer Positioning<2>,Reading and Writing Non-Regular Files<2>,,Wide_Text_IO
21469
@anchor{gnat_rm/the_implementation_of_standard_i_o stream-pointer-positioning-1}@anchor{26d}@anchor{gnat_rm/the_implementation_of_standard_i_o id14}@anchor{26e}
21470
@subsection Stream Pointer Positioning
21473
@cite{Ada.Wide_Text_IO} is similar to @cite{Ada.Text_IO} in its handling
21474
of stream pointer positioning (@ref{25e,,Text_IO}). There is one additional
21477
If @cite{Ada.Wide_Text_IO.Look_Ahead} reads a character outside the
21478
normal lower ASCII set (i.e., a character in the range:
21481
Wide_Character'Val (16#0080#) .. Wide_Character'Val (16#FFFF#)
21484
then although the logical position of the file pointer is unchanged by
21485
the @cite{Look_Ahead} call, the stream is physically positioned past the
21486
wide character sequence. Again this is to avoid the need for buffering
21487
or backup, and all @cite{Wide_Text_IO} routines check the internal
21488
indication that this situation has occurred so that this is not visible
21489
to a normal program using @cite{Wide_Text_IO}. However, this discrepancy
21490
can be observed if the wide text file shares a stream with another file.
21492
@node Reading and Writing Non-Regular Files<2>,,Stream Pointer Positioning<2>,Wide_Text_IO
21493
@anchor{gnat_rm/the_implementation_of_standard_i_o reading-and-writing-non-regular-files-1}@anchor{26f}@anchor{gnat_rm/the_implementation_of_standard_i_o id15}@anchor{270}
21494
@subsection Reading and Writing Non-Regular Files
21497
As in the case of Text_IO, when a non-regular file is read, it is
21498
assumed that the file contains no page marks (any form characters are
21499
treated as data characters), and @cite{End_Of_Page} always returns
21500
@cite{False}. Similarly, the end of file indication is not sticky, so
21501
it is possible to read beyond an end of file.
21503
@node Wide_Wide_Text_IO,Stream_IO,Wide_Text_IO,The Implementation of Standard I/O
21504
@anchor{gnat_rm/the_implementation_of_standard_i_o id16}@anchor{271}@anchor{gnat_rm/the_implementation_of_standard_i_o wide-wide-text-io}@anchor{272}
21505
@section Wide_Wide_Text_IO
21508
@cite{Wide_Wide_Text_IO} is similar in most respects to Text_IO, except that
21509
both input and output files may contain special sequences that represent
21510
wide wide character values. The encoding scheme for a given file may be
21511
specified using a FORM parameter:
21517
as part of the FORM string (WCEM = wide character encoding method),
21518
where @cite{x} is one of the following characters
21521
@multitable {xxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxx}
21544
Upper half encoding
21581
The encoding methods match those that
21582
can be used in a source
21583
program, but there is no requirement that the encoding method used for
21584
the source program be the same as the encoding method used for files,
21585
and different files may use different encoding methods.
21587
The default encoding method for the standard files, and for opened files
21588
for which no WCEM parameter is given in the FORM string matches the
21589
wide character encoding specified for the main program (the default
21590
being brackets encoding if no coding method was specified with -gnatW).
21595
@item @emph{UTF-8 Coding}
21597
A wide character is represented using
21598
UCS Transformation Format 8 (UTF-8) as defined in Annex R of ISO
21599
10646-1/Am.2. Depending on the character value, the representation
21600
is a one, two, three, or four byte sequence:
21604
16#000000#-16#00007f#: 2#0xxxxxxx#
21605
16#000080#-16#0007ff#: 2#110xxxxx# 2#10xxxxxx#
21606
16#000800#-16#00ffff#: 2#1110xxxx# 2#10xxxxxx# 2#10xxxxxx#
21607
16#010000#-16#10ffff#: 2#11110xxx# 2#10xxxxxx# 2#10xxxxxx# 2#10xxxxxx#
21613
where the @cite{xxx} bits correspond to the left-padded bits of the
21614
21-bit character value. Note that all lower half ASCII characters
21615
are represented as ASCII bytes and all upper half characters and
21616
other wide characters are represented as sequences of upper-half
21623
@item @emph{Brackets Coding}
21625
In this encoding, a wide wide character is represented by the following eight
21626
character sequence if is in wide character range
21636
and by the following ten character sequence if not
21640
[ " a b c d e f " ]
21646
where @cite{a}, @cite{b}, @cite{c}, @cite{d}, @cite{e}, and @cite{f}
21647
are the four or six hexadecimal
21648
characters (using uppercase letters) of the wide wide character code. For
21649
example, @cite{["01A345"]} is used to represent the wide wide character
21650
with code @cite{16#01A345#}.
21652
This scheme is compatible with use of the full Wide_Wide_Character set.
21653
On input, brackets coding can also be used for upper half characters,
21654
e.g., @cite{["C1"]} for lower case a. However, on output, brackets notation
21655
is only used for wide characters with a code greater than @cite{16#FF#}.
21658
If is also possible to use the other Wide_Character encoding methods,
21659
such as Shift-JIS, but the other schemes cannot support the full range
21660
of wide wide characters.
21661
An attempt to output a character that cannot
21662
be represented using the encoding scheme for the file causes
21663
Constraint_Error to be raised. An invalid wide character sequence on
21664
input also causes Constraint_Error to be raised.
21667
* Stream Pointer Positioning: Stream Pointer Positioning<3>.
21668
* Reading and Writing Non-Regular Files: Reading and Writing Non-Regular Files<3>.
21672
@node Stream Pointer Positioning<3>,Reading and Writing Non-Regular Files<3>,,Wide_Wide_Text_IO
21673
@anchor{gnat_rm/the_implementation_of_standard_i_o stream-pointer-positioning-2}@anchor{273}@anchor{gnat_rm/the_implementation_of_standard_i_o id17}@anchor{274}
21674
@subsection Stream Pointer Positioning
21677
@cite{Ada.Wide_Wide_Text_IO} is similar to @cite{Ada.Text_IO} in its handling
21678
of stream pointer positioning (@ref{25e,,Text_IO}). There is one additional
21681
If @cite{Ada.Wide_Wide_Text_IO.Look_Ahead} reads a character outside the
21682
normal lower ASCII set (i.e., a character in the range:
21685
Wide_Wide_Character'Val (16#0080#) .. Wide_Wide_Character'Val (16#10FFFF#)
21688
then although the logical position of the file pointer is unchanged by
21689
the @cite{Look_Ahead} call, the stream is physically positioned past the
21690
wide character sequence. Again this is to avoid the need for buffering
21691
or backup, and all @cite{Wide_Wide_Text_IO} routines check the internal
21692
indication that this situation has occurred so that this is not visible
21693
to a normal program using @cite{Wide_Wide_Text_IO}. However, this discrepancy
21694
can be observed if the wide text file shares a stream with another file.
21696
@node Reading and Writing Non-Regular Files<3>,,Stream Pointer Positioning<3>,Wide_Wide_Text_IO
21697
@anchor{gnat_rm/the_implementation_of_standard_i_o id18}@anchor{275}@anchor{gnat_rm/the_implementation_of_standard_i_o reading-and-writing-non-regular-files-2}@anchor{276}
21698
@subsection Reading and Writing Non-Regular Files
21701
As in the case of Text_IO, when a non-regular file is read, it is
21702
assumed that the file contains no page marks (any form characters are
21703
treated as data characters), and @cite{End_Of_Page} always returns
21704
@cite{False}. Similarly, the end of file indication is not sticky, so
21705
it is possible to read beyond an end of file.
21707
@node Stream_IO,Text Translation,Wide_Wide_Text_IO,The Implementation of Standard I/O
21708
@anchor{gnat_rm/the_implementation_of_standard_i_o id19}@anchor{277}@anchor{gnat_rm/the_implementation_of_standard_i_o stream-io}@anchor{278}
21712
A stream file is a sequence of bytes, where individual elements are
21713
written to the file as described in the Ada Reference Manual. The type
21714
@cite{Stream_Element} is simply a byte. There are two ways to read or
21715
write a stream file.
21721
The operations @cite{Read} and @cite{Write} directly read or write a
21722
sequence of stream elements with no control information.
21725
The stream attributes applied to a stream file transfer data in the
21726
manner described for stream attributes.
21729
@node Text Translation,Shared Files,Stream_IO,The Implementation of Standard I/O
21730
@anchor{gnat_rm/the_implementation_of_standard_i_o id20}@anchor{279}@anchor{gnat_rm/the_implementation_of_standard_i_o text-translation}@anchor{27a}
21731
@section Text Translation
21734
@code{Text_Translation=xxx} may be used as the Form parameter
21735
passed to Text_IO.Create and Text_IO.Open. @code{Text_Translation=xxx}
21736
has no effect on Unix systems. Possible values are:
21742
@code{Yes} or @code{Text} is the default, which means to
21743
translate LF to/from CR/LF on Windows systems.
21745
@code{No} disables this translation; i.e. it
21746
uses binary mode. For output files, @code{Text_Translation=No}
21747
may be used to create Unix-style files on
21751
@code{wtext} translation enabled in Unicode mode.
21752
(corresponds to _O_WTEXT).
21755
@code{u8text} translation enabled in Unicode UTF-8 mode.
21756
(corresponds to O_U8TEXT).
21759
@code{u16text} translation enabled in Unicode UTF-16
21760
mode. (corresponds to_O_U16TEXT).
21763
@node Shared Files,Filenames encoding,Text Translation,The Implementation of Standard I/O
21764
@anchor{gnat_rm/the_implementation_of_standard_i_o id21}@anchor{27b}@anchor{gnat_rm/the_implementation_of_standard_i_o shared-files}@anchor{27c}
21765
@section Shared Files
21768
Section A.14 of the Ada Reference Manual allows implementations to
21769
provide a wide variety of behavior if an attempt is made to access the
21770
same external file with two or more internal files.
21772
To provide a full range of functionality, while at the same time
21773
minimizing the problems of portability caused by this implementation
21774
dependence, GNAT handles file sharing as follows:
21780
In the absence of a @code{shared=xxx} form parameter, an attempt
21781
to open two or more files with the same full name is considered an error
21782
and is not supported. The exception @cite{Use_Error} will be
21783
raised. Note that a file that is not explicitly closed by the program
21784
remains open until the program terminates.
21787
If the form parameter @code{shared=no} appears in the form string, the
21788
file can be opened or created with its own separate stream identifier,
21789
regardless of whether other files sharing the same external file are
21790
opened. The exact effect depends on how the C stream routines handle
21791
multiple accesses to the same external files using separate streams.
21794
If the form parameter @code{shared=yes} appears in the form string for
21795
each of two or more files opened using the same full name, the same
21796
stream is shared between these files, and the semantics are as described
21797
in Ada Reference Manual, Section A.14.
21800
When a program that opens multiple files with the same name is ported
21801
from another Ada compiler to GNAT, the effect will be that
21802
@cite{Use_Error} is raised.
21804
The documentation of the original compiler and the documentation of the
21805
program should then be examined to determine if file sharing was
21806
expected, and @code{shared=xxx} parameters added to @cite{Open}
21807
and @cite{Create} calls as required.
21809
When a program is ported from GNAT to some other Ada compiler, no
21810
special attention is required unless the @code{shared=xxx} form
21811
parameter is used in the program. In this case, you must examine the
21812
documentation of the new compiler to see if it supports the required
21813
file sharing semantics, and form strings modified appropriately. Of
21814
course it may be the case that the program cannot be ported if the
21815
target compiler does not support the required functionality. The best
21816
approach in writing portable code is to avoid file sharing (and hence
21817
the use of the @code{shared=xxx} parameter in the form string)
21820
One common use of file sharing in Ada 83 is the use of instantiations of
21821
Sequential_IO on the same file with different types, to achieve
21822
heterogeneous input-output. Although this approach will work in GNAT if
21823
@code{shared=yes} is specified, it is preferable in Ada to use Stream_IO
21824
for this purpose (using the stream attributes)
21826
@node Filenames encoding,File content encoding,Shared Files,The Implementation of Standard I/O
21827
@anchor{gnat_rm/the_implementation_of_standard_i_o filenames-encoding}@anchor{27d}@anchor{gnat_rm/the_implementation_of_standard_i_o id22}@anchor{27e}
21828
@section Filenames encoding
21831
An encoding form parameter can be used to specify the filename
21832
encoding @code{encoding=xxx}.
21838
If the form parameter @code{encoding=utf8} appears in the form string, the
21839
filename must be encoded in UTF-8.
21842
If the form parameter @code{encoding=8bits} appears in the form
21843
string, the filename must be a standard 8bits string.
21846
In the absence of a @code{encoding=xxx} form parameter, the
21847
encoding is controlled by the @code{GNAT_CODE_PAGE} environment
21848
variable. And if not set @code{utf8} is assumed.
21853
@item @emph{CP_ACP}
21855
The current system Windows ANSI code page.
21857
@item @emph{CP_UTF8}
21862
This encoding form parameter is only supported on the Windows
21863
platform. On the other Operating Systems the run-time is supporting
21866
@node File content encoding,Open Modes,Filenames encoding,The Implementation of Standard I/O
21867
@anchor{gnat_rm/the_implementation_of_standard_i_o file-content-encoding}@anchor{27f}@anchor{gnat_rm/the_implementation_of_standard_i_o id23}@anchor{280}
21868
@section File content encoding
21871
For text files it is possible to specify the encoding to use. This is
21872
controlled by the by the @code{GNAT_CCS_ENCODING} environment
21873
variable. And if not set @code{TEXT} is assumed.
21875
The possible values are those supported on Windows:
21882
Translated text mode
21886
Translated unicode encoding
21888
@item @emph{U16TEXT}
21890
Unicode 16-bit encoding
21892
@item @emph{U8TEXT}
21894
Unicode 8-bit encoding
21897
This encoding is only supported on the Windows platform.
21899
@node Open Modes,Operations on C Streams,File content encoding,The Implementation of Standard I/O
21900
@anchor{gnat_rm/the_implementation_of_standard_i_o open-modes}@anchor{281}@anchor{gnat_rm/the_implementation_of_standard_i_o id24}@anchor{282}
21901
@section Open Modes
21904
@cite{Open} and @cite{Create} calls result in a call to @cite{fopen}
21905
using the mode shown in the following table:
21908
@multitable {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxx}
21911
@cite{Open} and @cite{Create} Call Modes
21953
Out_File (Direct_IO)
21965
Out_File (all other cases)
21990
If text file translation is required, then either @code{b} or @code{t}
21991
is added to the mode, depending on the setting of Text. Text file
21992
translation refers to the mapping of CR/LF sequences in an external file
21993
to LF characters internally. This mapping only occurs in DOS and
21994
DOS-like systems, and is not relevant to other systems.
21996
A special case occurs with Stream_IO. As shown in the above table, the
21997
file is initially opened in @code{r} or @code{w} mode for the
21998
@cite{In_File} and @cite{Out_File} cases. If a @cite{Set_Mode} operation
21999
subsequently requires switching from reading to writing or vice-versa,
22000
then the file is reopened in @code{r+} mode to permit the required operation.
22002
@node Operations on C Streams,Interfacing to C Streams,Open Modes,The Implementation of Standard I/O
22003
@anchor{gnat_rm/the_implementation_of_standard_i_o operations-on-c-streams}@anchor{283}@anchor{gnat_rm/the_implementation_of_standard_i_o id25}@anchor{284}
22004
@section Operations on C Streams
22007
The package @cite{Interfaces.C_Streams} provides an Ada program with direct
22008
access to the C library functions for operations on C streams:
22011
package Interfaces.C_Streams is
22012
-- Note: the reason we do not use the types that are in
22013
-- Interfaces.C is that we want to avoid dragging in the
22014
-- code in this unit if possible.
22015
subtype chars is System.Address;
22016
-- Pointer to null-terminated array of characters
22017
subtype FILEs is System.Address;
22018
-- Corresponds to the C type FILE*
22019
subtype voids is System.Address;
22020
-- Corresponds to the C type void*
22021
subtype int is Integer;
22022
subtype long is Long_Integer;
22023
-- Note: the above types are subtypes deliberately, and it
22024
-- is part of this spec that the above correspondences are
22025
-- guaranteed. This means that it is legitimate to, for
22026
-- example, use Integer instead of int. We provide these
22027
-- synonyms for clarity, but in some cases it may be
22028
-- convenient to use the underlying types (for example to
22029
-- avoid an unnecessary dependency of a spec on the spec
22031
type size_t is mod 2 ** Standard'Address_Size;
22032
NULL_Stream : constant FILEs;
22033
-- Value returned (NULL in C) to indicate an
22034
-- fdopen/fopen/tmpfile error
22035
----------------------------------
22036
-- Constants Defined in stdio.h --
22037
----------------------------------
22038
EOF : constant int;
22039
-- Used by a number of routines to indicate error or
22041
IOFBF : constant int;
22042
IOLBF : constant int;
22043
IONBF : constant int;
22044
-- Used to indicate buffering mode for setvbuf call
22045
SEEK_CUR : constant int;
22046
SEEK_END : constant int;
22047
SEEK_SET : constant int;
22048
-- Used to indicate origin for fseek call
22049
function stdin return FILEs;
22050
function stdout return FILEs;
22051
function stderr return FILEs;
22052
-- Streams associated with standard files
22053
--------------------------
22054
-- Standard C functions --
22055
--------------------------
22056
-- The functions selected below are ones that are
22057
-- available in UNIX (but not necessarily in ANSI C).
22058
-- These are very thin interfaces
22059
-- which copy exactly the C headers. For more
22060
-- documentation on these functions, see the Microsoft C
22061
-- "Run-Time Library Reference" (Microsoft Press, 1990,
22062
-- ISBN 1-55615-225-6), which includes useful information
22063
-- on system compatibility.
22064
procedure clearerr (stream : FILEs);
22065
function fclose (stream : FILEs) return int;
22066
function fdopen (handle : int; mode : chars) return FILEs;
22067
function feof (stream : FILEs) return int;
22068
function ferror (stream : FILEs) return int;
22069
function fflush (stream : FILEs) return int;
22070
function fgetc (stream : FILEs) return int;
22071
function fgets (strng : chars; n : int; stream : FILEs)
22073
function fileno (stream : FILEs) return int;
22074
function fopen (filename : chars; Mode : chars)
22076
-- Note: to maintain target independence, use
22077
-- text_translation_required, a boolean variable defined in
22078
-- a-sysdep.c to deal with the target dependent text
22079
-- translation requirement. If this variable is set,
22080
-- then b/t should be appended to the standard mode
22081
-- argument to set the text translation mode off or on
22083
function fputc (C : int; stream : FILEs) return int;
22084
function fputs (Strng : chars; Stream : FILEs) return int;
22101
function ftell (stream : FILEs) return long;
22108
function isatty (handle : int) return int;
22109
procedure mktemp (template : chars);
22110
-- The return value (which is just a pointer to template)
22112
procedure rewind (stream : FILEs);
22113
function rmtmp return int;
22121
function tmpfile return FILEs;
22122
function ungetc (c : int; stream : FILEs) return int;
22123
function unlink (filename : chars) return int;
22124
---------------------
22125
-- Extra functions --
22126
---------------------
22127
-- These functions supply slightly thicker bindings than
22128
-- those above. They are derived from functions in the
22129
-- C Run-Time Library, but may do a bit more work than
22130
-- just directly calling one of the Library functions.
22131
function is_regular_file (handle : int) return int;
22132
-- Tests if given handle is for a regular file (result 1)
22133
-- or for a non-regular file (pipe or device, result 0).
22134
---------------------------------
22135
-- Control of Text/Binary Mode --
22136
---------------------------------
22137
-- If text_translation_required is true, then the following
22138
-- functions may be used to dynamically switch a file from
22139
-- binary to text mode or vice versa. These functions have
22140
-- no effect if text_translation_required is false (i.e., in
22141
-- normal UNIX mode). Use fileno to get a stream handle.
22142
procedure set_binary_mode (handle : int);
22143
procedure set_text_mode (handle : int);
22144
----------------------------
22145
-- Full Path Name support --
22146
----------------------------
22147
procedure full_name (nam : chars; buffer : chars);
22148
-- Given a NUL terminated string representing a file
22149
-- name, returns in buffer a NUL terminated string
22150
-- representing the full path name for the file name.
22151
-- On systems where it is relevant the drive is also
22152
-- part of the full path name. It is the responsibility
22153
-- of the caller to pass an actual parameter for buffer
22154
-- that is big enough for any full path name. Use
22155
-- max_path_len given below as the size of buffer.
22156
max_path_len : integer;
22157
-- Maximum length of an allowable full path name on the
22158
-- system, including a terminating NUL character.
22159
end Interfaces.C_Streams;
22162
@node Interfacing to C Streams,,Operations on C Streams,The Implementation of Standard I/O
22163
@anchor{gnat_rm/the_implementation_of_standard_i_o interfacing-to-c-streams}@anchor{285}@anchor{gnat_rm/the_implementation_of_standard_i_o id26}@anchor{286}
22164
@section Interfacing to C Streams
22167
The packages in this section permit interfacing Ada files to C Stream
22171
with Interfaces.C_Streams;
22172
package Ada.Sequential_IO.C_Streams is
22173
function C_Stream (F : File_Type)
22174
return Interfaces.C_Streams.FILEs;
22176
(File : in out File_Type;
22177
Mode : in File_Mode;
22178
C_Stream : in Interfaces.C_Streams.FILEs;
22179
Form : in String := "");
22180
end Ada.Sequential_IO.C_Streams;
22182
with Interfaces.C_Streams;
22183
package Ada.Direct_IO.C_Streams is
22184
function C_Stream (F : File_Type)
22185
return Interfaces.C_Streams.FILEs;
22187
(File : in out File_Type;
22188
Mode : in File_Mode;
22189
C_Stream : in Interfaces.C_Streams.FILEs;
22190
Form : in String := "");
22191
end Ada.Direct_IO.C_Streams;
22193
with Interfaces.C_Streams;
22194
package Ada.Text_IO.C_Streams is
22195
function C_Stream (F : File_Type)
22196
return Interfaces.C_Streams.FILEs;
22198
(File : in out File_Type;
22199
Mode : in File_Mode;
22200
C_Stream : in Interfaces.C_Streams.FILEs;
22201
Form : in String := "");
22202
end Ada.Text_IO.C_Streams;
22204
with Interfaces.C_Streams;
22205
package Ada.Wide_Text_IO.C_Streams is
22206
function C_Stream (F : File_Type)
22207
return Interfaces.C_Streams.FILEs;
22209
(File : in out File_Type;
22210
Mode : in File_Mode;
22211
C_Stream : in Interfaces.C_Streams.FILEs;
22212
Form : in String := "");
22213
end Ada.Wide_Text_IO.C_Streams;
22215
with Interfaces.C_Streams;
22216
package Ada.Wide_Wide_Text_IO.C_Streams is
22217
function C_Stream (F : File_Type)
22218
return Interfaces.C_Streams.FILEs;
22220
(File : in out File_Type;
22221
Mode : in File_Mode;
22222
C_Stream : in Interfaces.C_Streams.FILEs;
22223
Form : in String := "");
22224
end Ada.Wide_Wide_Text_IO.C_Streams;
22226
with Interfaces.C_Streams;
22227
package Ada.Stream_IO.C_Streams is
22228
function C_Stream (F : File_Type)
22229
return Interfaces.C_Streams.FILEs;
22231
(File : in out File_Type;
22232
Mode : in File_Mode;
22233
C_Stream : in Interfaces.C_Streams.FILEs;
22234
Form : in String := "");
22235
end Ada.Stream_IO.C_Streams;
22238
In each of these six packages, the @cite{C_Stream} function obtains the
22239
@cite{FILE} pointer from a currently opened Ada file. It is then
22240
possible to use the @cite{Interfaces.C_Streams} package to operate on
22241
this stream, or the stream can be passed to a C program which can
22242
operate on it directly. Of course the program is responsible for
22243
ensuring that only appropriate sequences of operations are executed.
22245
One particular use of relevance to an Ada program is that the
22246
@cite{setvbuf} function can be used to control the buffering of the
22247
stream used by an Ada file. In the absence of such a call the standard
22248
default buffering is used.
22250
The @cite{Open} procedures in these packages open a file giving an
22251
existing C Stream instead of a file name. Typically this stream is
22252
imported from a C program, allowing an Ada file to operate on an
22255
@node The GNAT Library,Interfacing to Other Languages,The Implementation of Standard I/O,Top
22256
@anchor{gnat_rm/the_gnat_library the-gnat-library}@anchor{10}@anchor{gnat_rm/the_gnat_library doc}@anchor{287}@anchor{gnat_rm/the_gnat_library id1}@anchor{288}
22257
@chapter The GNAT Library
22260
The GNAT library contains a number of general and special purpose packages.
22261
It represents functionality that the GNAT developers have found useful, and
22262
which is made available to GNAT users. The packages described here are fully
22263
supported, and upwards compatibility will be maintained in future releases,
22264
so you can use these facilities with the confidence that the same functionality
22265
will be available in future releases.
22267
The chapter here simply gives a brief summary of the facilities available.
22268
The full documentation is found in the spec file for the package. The full
22269
sources of these library packages, including both spec and body, are provided
22270
with all GNAT releases. For example, to find out the full specifications of
22271
the SPITBOL pattern matching capability, including a full tutorial and
22272
extensive examples, look in the @code{g-spipat.ads} file in the library.
22274
For each entry here, the package name (as it would appear in a @cite{with}
22275
clause) is given, followed by the name of the corresponding spec file in
22276
parentheses. The packages are children in four hierarchies, @cite{Ada},
22277
@cite{Interfaces}, @cite{System}, and @cite{GNAT}, the latter being a
22278
GNAT-specific hierarchy.
22280
Note that an application program should only use packages in one of these
22281
four hierarchies if the package is defined in the Ada Reference Manual,
22282
or is listed in this section of the GNAT Programmers Reference Manual.
22283
All other units should be considered internal implementation units and
22284
should not be directly @cite{with}'ed by application code. The use of
22285
a @cite{with} statement that references one of these internal implementation
22286
units makes an application potentially dependent on changes in versions
22287
of GNAT, and will generate a warning message.
22290
* Ada.Characters.Latin_9 (a-chlat9.ads): Ada Characters Latin_9 a-chlat9 ads.
22291
* Ada.Characters.Wide_Latin_1 (a-cwila1.ads): Ada Characters Wide_Latin_1 a-cwila1 ads.
22292
* Ada.Characters.Wide_Latin_9 (a-cwila1.ads): Ada Characters Wide_Latin_9 a-cwila1 ads.
22293
* Ada.Characters.Wide_Wide_Latin_1 (a-chzla1.ads): Ada Characters Wide_Wide_Latin_1 a-chzla1 ads.
22294
* Ada.Characters.Wide_Wide_Latin_9 (a-chzla9.ads): Ada Characters Wide_Wide_Latin_9 a-chzla9 ads.
22295
* Ada.Containers.Formal_Doubly_Linked_Lists (a-cfdlli.ads): Ada Containers Formal_Doubly_Linked_Lists a-cfdlli ads.
22296
* Ada.Containers.Formal_Hashed_Maps (a-cfhama.ads): Ada Containers Formal_Hashed_Maps a-cfhama ads.
22297
* Ada.Containers.Formal_Hashed_Sets (a-cfhase.ads): Ada Containers Formal_Hashed_Sets a-cfhase ads.
22298
* Ada.Containers.Formal_Ordered_Maps (a-cforma.ads): Ada Containers Formal_Ordered_Maps a-cforma ads.
22299
* Ada.Containers.Formal_Ordered_Sets (a-cforse.ads): Ada Containers Formal_Ordered_Sets a-cforse ads.
22300
* Ada.Containers.Formal_Vectors (a-cofove.ads): Ada Containers Formal_Vectors a-cofove ads.
22301
* Ada.Containers.Formal_Indefinite_Vectors (a-cfinve.ads): Ada Containers Formal_Indefinite_Vectors a-cfinve ads.
22302
* Ada.Containers.Bounded_Holders (a-coboho.ads): Ada Containers Bounded_Holders a-coboho ads.
22303
* Ada.Command_Line.Environment (a-colien.ads): Ada Command_Line Environment a-colien ads.
22304
* Ada.Command_Line.Remove (a-colire.ads): Ada Command_Line Remove a-colire ads.
22305
* Ada.Command_Line.Response_File (a-clrefi.ads): Ada Command_Line Response_File a-clrefi ads.
22306
* Ada.Direct_IO.C_Streams (a-diocst.ads): Ada Direct_IO C_Streams a-diocst ads.
22307
* Ada.Exceptions.Is_Null_Occurrence (a-einuoc.ads): Ada Exceptions Is_Null_Occurrence a-einuoc ads.
22308
* Ada.Exceptions.Last_Chance_Handler (a-elchha.ads): Ada Exceptions Last_Chance_Handler a-elchha ads.
22309
* Ada.Exceptions.Traceback (a-exctra.ads): Ada Exceptions Traceback a-exctra ads.
22310
* Ada.Sequential_IO.C_Streams (a-siocst.ads): Ada Sequential_IO C_Streams a-siocst ads.
22311
* Ada.Streams.Stream_IO.C_Streams (a-ssicst.ads): Ada Streams Stream_IO C_Streams a-ssicst ads.
22312
* Ada.Strings.Unbounded.Text_IO (a-suteio.ads): Ada Strings Unbounded Text_IO a-suteio ads.
22313
* Ada.Strings.Wide_Unbounded.Wide_Text_IO (a-swuwti.ads): Ada Strings Wide_Unbounded Wide_Text_IO a-swuwti ads.
22314
* Ada.Strings.Wide_Wide_Unbounded.Wide_Wide_Text_IO (a-szuzti.ads): Ada Strings Wide_Wide_Unbounded Wide_Wide_Text_IO a-szuzti ads.
22315
* Ada.Text_IO.C_Streams (a-tiocst.ads): Ada Text_IO C_Streams a-tiocst ads.
22316
* Ada.Text_IO.Reset_Standard_Files (a-tirsfi.ads): Ada Text_IO Reset_Standard_Files a-tirsfi ads.
22317
* Ada.Wide_Characters.Unicode (a-wichun.ads): Ada Wide_Characters Unicode a-wichun ads.
22318
* Ada.Wide_Text_IO.C_Streams (a-wtcstr.ads): Ada Wide_Text_IO C_Streams a-wtcstr ads.
22319
* Ada.Wide_Text_IO.Reset_Standard_Files (a-wrstfi.ads): Ada Wide_Text_IO Reset_Standard_Files a-wrstfi ads.
22320
* Ada.Wide_Wide_Characters.Unicode (a-zchuni.ads): Ada Wide_Wide_Characters Unicode a-zchuni ads.
22321
* Ada.Wide_Wide_Text_IO.C_Streams (a-ztcstr.ads): Ada Wide_Wide_Text_IO C_Streams a-ztcstr ads.
22322
* Ada.Wide_Wide_Text_IO.Reset_Standard_Files (a-zrstfi.ads): Ada Wide_Wide_Text_IO Reset_Standard_Files a-zrstfi ads.
22323
* GNAT.Altivec (g-altive.ads): GNAT Altivec g-altive ads.
22324
* GNAT.Altivec.Conversions (g-altcon.ads): GNAT Altivec Conversions g-altcon ads.
22325
* GNAT.Altivec.Vector_Operations (g-alveop.ads): GNAT Altivec Vector_Operations g-alveop ads.
22326
* GNAT.Altivec.Vector_Types (g-alvety.ads): GNAT Altivec Vector_Types g-alvety ads.
22327
* GNAT.Altivec.Vector_Views (g-alvevi.ads): GNAT Altivec Vector_Views g-alvevi ads.
22328
* GNAT.Array_Split (g-arrspl.ads): GNAT Array_Split g-arrspl ads.
22329
* GNAT.AWK (g-awk.ads): GNAT AWK g-awk ads.
22330
* GNAT.Bind_Environment (g-binenv.ads): GNAT Bind_Environment g-binenv ads.
22331
* GNAT.Bounded_Buffers (g-boubuf.ads): GNAT Bounded_Buffers g-boubuf ads.
22332
* GNAT.Bounded_Mailboxes (g-boumai.ads): GNAT Bounded_Mailboxes g-boumai ads.
22333
* GNAT.Bubble_Sort (g-bubsor.ads): GNAT Bubble_Sort g-bubsor ads.
22334
* GNAT.Bubble_Sort_A (g-busora.ads): GNAT Bubble_Sort_A g-busora ads.
22335
* GNAT.Bubble_Sort_G (g-busorg.ads): GNAT Bubble_Sort_G g-busorg ads.
22336
* GNAT.Byte_Order_Mark (g-byorma.ads): GNAT Byte_Order_Mark g-byorma ads.
22337
* GNAT.Byte_Swapping (g-bytswa.ads): GNAT Byte_Swapping g-bytswa ads.
22338
* GNAT.Calendar (g-calend.ads): GNAT Calendar g-calend ads.
22339
* GNAT.Calendar.Time_IO (g-catiio.ads): GNAT Calendar Time_IO g-catiio ads.
22340
* GNAT.CRC32 (g-crc32.ads): GNAT CRC32 g-crc32 ads.
22341
* GNAT.Case_Util (g-casuti.ads): GNAT Case_Util g-casuti ads.
22342
* GNAT.CGI (g-cgi.ads): GNAT CGI g-cgi ads.
22343
* GNAT.CGI.Cookie (g-cgicoo.ads): GNAT CGI Cookie g-cgicoo ads.
22344
* GNAT.CGI.Debug (g-cgideb.ads): GNAT CGI Debug g-cgideb ads.
22345
* GNAT.Command_Line (g-comlin.ads): GNAT Command_Line g-comlin ads.
22346
* GNAT.Compiler_Version (g-comver.ads): GNAT Compiler_Version g-comver ads.
22347
* GNAT.Ctrl_C (g-ctrl_c.ads): GNAT Ctrl_C g-ctrl_c ads.
22348
* GNAT.Current_Exception (g-curexc.ads): GNAT Current_Exception g-curexc ads.
22349
* GNAT.Debug_Pools (g-debpoo.ads): GNAT Debug_Pools g-debpoo ads.
22350
* GNAT.Debug_Utilities (g-debuti.ads): GNAT Debug_Utilities g-debuti ads.
22351
* GNAT.Decode_String (g-decstr.ads): GNAT Decode_String g-decstr ads.
22352
* GNAT.Decode_UTF8_String (g-deutst.ads): GNAT Decode_UTF8_String g-deutst ads.
22353
* GNAT.Directory_Operations (g-dirope.ads): GNAT Directory_Operations g-dirope ads.
22354
* GNAT.Directory_Operations.Iteration (g-diopit.ads): GNAT Directory_Operations Iteration g-diopit ads.
22355
* GNAT.Dynamic_HTables (g-dynhta.ads): GNAT Dynamic_HTables g-dynhta ads.
22356
* GNAT.Dynamic_Tables (g-dyntab.ads): GNAT Dynamic_Tables g-dyntab ads.
22357
* GNAT.Encode_String (g-encstr.ads): GNAT Encode_String g-encstr ads.
22358
* GNAT.Encode_UTF8_String (g-enutst.ads): GNAT Encode_UTF8_String g-enutst ads.
22359
* GNAT.Exception_Actions (g-excact.ads): GNAT Exception_Actions g-excact ads.
22360
* GNAT.Exception_Traces (g-exctra.ads): GNAT Exception_Traces g-exctra ads.
22361
* GNAT.Exceptions (g-expect.ads): GNAT Exceptions g-expect ads.
22362
* GNAT.Expect (g-expect.ads): GNAT Expect g-expect ads.
22363
* GNAT.Expect.TTY (g-exptty.ads): GNAT Expect TTY g-exptty ads.
22364
* GNAT.Float_Control (g-flocon.ads): GNAT Float_Control g-flocon ads.
22365
* GNAT.Formatted_String (g-forstr.ads): GNAT Formatted_String g-forstr ads.
22366
* GNAT.Heap_Sort (g-heasor.ads): GNAT Heap_Sort g-heasor ads.
22367
* GNAT.Heap_Sort_A (g-hesora.ads): GNAT Heap_Sort_A g-hesora ads.
22368
* GNAT.Heap_Sort_G (g-hesorg.ads): GNAT Heap_Sort_G g-hesorg ads.
22369
* GNAT.HTable (g-htable.ads): GNAT HTable g-htable ads.
22370
* GNAT.IO (g-io.ads): GNAT IO g-io ads.
22371
* GNAT.IO_Aux (g-io_aux.ads): GNAT IO_Aux g-io_aux ads.
22372
* GNAT.Lock_Files (g-locfil.ads): GNAT Lock_Files g-locfil ads.
22373
* GNAT.MBBS_Discrete_Random (g-mbdira.ads): GNAT MBBS_Discrete_Random g-mbdira ads.
22374
* GNAT.MBBS_Float_Random (g-mbflra.ads): GNAT MBBS_Float_Random g-mbflra ads.
22375
* GNAT.MD5 (g-md5.ads): GNAT MD5 g-md5 ads.
22376
* GNAT.Memory_Dump (g-memdum.ads): GNAT Memory_Dump g-memdum ads.
22377
* GNAT.Most_Recent_Exception (g-moreex.ads): GNAT Most_Recent_Exception g-moreex ads.
22378
* GNAT.OS_Lib (g-os_lib.ads): GNAT OS_Lib g-os_lib ads.
22379
* GNAT.Perfect_Hash_Generators (g-pehage.ads): GNAT Perfect_Hash_Generators g-pehage ads.
22380
* GNAT.Random_Numbers (g-rannum.ads): GNAT Random_Numbers g-rannum ads.
22381
* GNAT.Regexp (g-regexp.ads): GNAT Regexp g-regexp ads.
22382
* GNAT.Registry (g-regist.ads): GNAT Registry g-regist ads.
22383
* GNAT.Regpat (g-regpat.ads): GNAT Regpat g-regpat ads.
22384
* GNAT.Rewrite_Data (g-rewdat.ads): GNAT Rewrite_Data g-rewdat ads.
22385
* GNAT.Secondary_Stack_Info (g-sestin.ads): GNAT Secondary_Stack_Info g-sestin ads.
22386
* GNAT.Semaphores (g-semaph.ads): GNAT Semaphores g-semaph ads.
22387
* GNAT.Serial_Communications (g-sercom.ads): GNAT Serial_Communications g-sercom ads.
22388
* GNAT.SHA1 (g-sha1.ads): GNAT SHA1 g-sha1 ads.
22389
* GNAT.SHA224 (g-sha224.ads): GNAT SHA224 g-sha224 ads.
22390
* GNAT.SHA256 (g-sha256.ads): GNAT SHA256 g-sha256 ads.
22391
* GNAT.SHA384 (g-sha384.ads): GNAT SHA384 g-sha384 ads.
22392
* GNAT.SHA512 (g-sha512.ads): GNAT SHA512 g-sha512 ads.
22393
* GNAT.Signals (g-signal.ads): GNAT Signals g-signal ads.
22394
* GNAT.Sockets (g-socket.ads): GNAT Sockets g-socket ads.
22395
* GNAT.Source_Info (g-souinf.ads): GNAT Source_Info g-souinf ads.
22396
* GNAT.Spelling_Checker (g-speche.ads): GNAT Spelling_Checker g-speche ads.
22397
* GNAT.Spelling_Checker_Generic (g-spchge.ads): GNAT Spelling_Checker_Generic g-spchge ads.
22398
* GNAT.Spitbol.Patterns (g-spipat.ads): GNAT Spitbol Patterns g-spipat ads.
22399
* GNAT.Spitbol (g-spitbo.ads): GNAT Spitbol g-spitbo ads.
22400
* GNAT.Spitbol.Table_Boolean (g-sptabo.ads): GNAT Spitbol Table_Boolean g-sptabo ads.
22401
* GNAT.Spitbol.Table_Integer (g-sptain.ads): GNAT Spitbol Table_Integer g-sptain ads.
22402
* GNAT.Spitbol.Table_VString (g-sptavs.ads): GNAT Spitbol Table_VString g-sptavs ads.
22403
* GNAT.SSE (g-sse.ads): GNAT SSE g-sse ads.
22404
* GNAT.SSE.Vector_Types (g-ssvety.ads): GNAT SSE Vector_Types g-ssvety ads.
22405
* GNAT.Strings (g-string.ads): GNAT Strings g-string ads.
22406
* GNAT.String_Split (g-strspl.ads): GNAT String_Split g-strspl ads.
22407
* GNAT.Table (g-table.ads): GNAT Table g-table ads.
22408
* GNAT.Task_Lock (g-tasloc.ads): GNAT Task_Lock g-tasloc ads.
22409
* GNAT.Time_Stamp (g-timsta.ads): GNAT Time_Stamp g-timsta ads.
22410
* GNAT.Threads (g-thread.ads): GNAT Threads g-thread ads.
22411
* GNAT.Traceback (g-traceb.ads): GNAT Traceback g-traceb ads.
22412
* GNAT.Traceback.Symbolic (g-trasym.ads): GNAT Traceback Symbolic g-trasym ads.
22413
* GNAT.UTF_32 (g-table.ads): GNAT UTF_32 g-table ads.
22414
* GNAT.Wide_Spelling_Checker (g-u3spch.ads): GNAT Wide_Spelling_Checker g-u3spch ads.
22415
* GNAT.Wide_Spelling_Checker (g-wispch.ads): GNAT Wide_Spelling_Checker g-wispch ads.
22416
* GNAT.Wide_String_Split (g-wistsp.ads): GNAT Wide_String_Split g-wistsp ads.
22417
* GNAT.Wide_Wide_Spelling_Checker (g-zspche.ads): GNAT Wide_Wide_Spelling_Checker g-zspche ads.
22418
* GNAT.Wide_Wide_String_Split (g-zistsp.ads): GNAT Wide_Wide_String_Split g-zistsp ads.
22419
* Interfaces.C.Extensions (i-cexten.ads): Interfaces C Extensions i-cexten ads.
22420
* Interfaces.C.Streams (i-cstrea.ads): Interfaces C Streams i-cstrea ads.
22421
* Interfaces.Packed_Decimal (i-pacdec.ads): Interfaces Packed_Decimal i-pacdec ads.
22422
* Interfaces.VxWorks (i-vxwork.ads): Interfaces VxWorks i-vxwork ads.
22423
* Interfaces.VxWorks.IO (i-vxwoio.ads): Interfaces VxWorks IO i-vxwoio ads.
22424
* System.Address_Image (s-addima.ads): System Address_Image s-addima ads.
22425
* System.Assertions (s-assert.ads): System Assertions s-assert ads.
22426
* System.Atomic_Counters (s-atocou.ads): System Atomic_Counters s-atocou ads.
22427
* System.Memory (s-memory.ads): System Memory s-memory ads.
22428
* System.Multiprocessors (s-multip.ads): System Multiprocessors s-multip ads.
22429
* System.Multiprocessors.Dispatching_Domains (s-mudido.ads): System Multiprocessors Dispatching_Domains s-mudido ads.
22430
* System.Partition_Interface (s-parint.ads): System Partition_Interface s-parint ads.
22431
* System.Pool_Global (s-pooglo.ads): System Pool_Global s-pooglo ads.
22432
* System.Pool_Local (s-pooloc.ads): System Pool_Local s-pooloc ads.
22433
* System.Restrictions (s-restri.ads): System Restrictions s-restri ads.
22434
* System.Rident (s-rident.ads): System Rident s-rident ads.
22435
* System.Strings.Stream_Ops (s-ststop.ads): System Strings Stream_Ops s-ststop ads.
22436
* System.Unsigned_Types (s-unstyp.ads): System Unsigned_Types s-unstyp ads.
22437
* System.Wch_Cnv (s-wchcnv.ads): System Wch_Cnv s-wchcnv ads.
22438
* System.Wch_Con (s-wchcon.ads): System Wch_Con s-wchcon ads.
22442
@node Ada Characters Latin_9 a-chlat9 ads,Ada Characters Wide_Latin_1 a-cwila1 ads,,The GNAT Library
22443
@anchor{gnat_rm/the_gnat_library id2}@anchor{289}@anchor{gnat_rm/the_gnat_library ada-characters-latin-9-a-chlat9-ads}@anchor{28a}
22444
@section @cite{Ada.Characters.Latin_9} (@code{a-chlat9.ads})
22447
@geindex Ada.Characters.Latin_9 (a-chlat9.ads)
22449
@geindex Latin_9 constants for Character
22451
This child of @cite{Ada.Characters}
22452
provides a set of definitions corresponding to those in the
22453
RM-defined package @cite{Ada.Characters.Latin_1} but with the
22454
few modifications required for @cite{Latin-9}
22455
The provision of such a package
22456
is specifically authorized by the Ada Reference Manual
22459
@node Ada Characters Wide_Latin_1 a-cwila1 ads,Ada Characters Wide_Latin_9 a-cwila1 ads,Ada Characters Latin_9 a-chlat9 ads,The GNAT Library
22460
@anchor{gnat_rm/the_gnat_library ada-characters-wide-latin-1-a-cwila1-ads}@anchor{28b}@anchor{gnat_rm/the_gnat_library id3}@anchor{28c}
22461
@section @cite{Ada.Characters.Wide_Latin_1} (@code{a-cwila1.ads})
22464
@geindex Ada.Characters.Wide_Latin_1 (a-cwila1.ads)
22466
@geindex Latin_1 constants for Wide_Character
22468
This child of @cite{Ada.Characters}
22469
provides a set of definitions corresponding to those in the
22470
RM-defined package @cite{Ada.Characters.Latin_1} but with the
22471
types of the constants being @cite{Wide_Character}
22472
instead of @cite{Character}. The provision of such a package
22473
is specifically authorized by the Ada Reference Manual
22476
@node Ada Characters Wide_Latin_9 a-cwila1 ads,Ada Characters Wide_Wide_Latin_1 a-chzla1 ads,Ada Characters Wide_Latin_1 a-cwila1 ads,The GNAT Library
22477
@anchor{gnat_rm/the_gnat_library id4}@anchor{28d}@anchor{gnat_rm/the_gnat_library ada-characters-wide-latin-9-a-cwila1-ads}@anchor{28e}
22478
@section @cite{Ada.Characters.Wide_Latin_9} (@code{a-cwila1.ads})
22481
@geindex Ada.Characters.Wide_Latin_9 (a-cwila1.ads)
22483
@geindex Latin_9 constants for Wide_Character
22485
This child of @cite{Ada.Characters}
22486
provides a set of definitions corresponding to those in the
22487
GNAT defined package @cite{Ada.Characters.Latin_9} but with the
22488
types of the constants being @cite{Wide_Character}
22489
instead of @cite{Character}. The provision of such a package
22490
is specifically authorized by the Ada Reference Manual
22493
@node Ada Characters Wide_Wide_Latin_1 a-chzla1 ads,Ada Characters Wide_Wide_Latin_9 a-chzla9 ads,Ada Characters Wide_Latin_9 a-cwila1 ads,The GNAT Library
22494
@anchor{gnat_rm/the_gnat_library ada-characters-wide-wide-latin-1-a-chzla1-ads}@anchor{28f}@anchor{gnat_rm/the_gnat_library id5}@anchor{290}
22495
@section @cite{Ada.Characters.Wide_Wide_Latin_1} (@code{a-chzla1.ads})
22498
@geindex Ada.Characters.Wide_Wide_Latin_1 (a-chzla1.ads)
22500
@geindex Latin_1 constants for Wide_Wide_Character
22502
This child of @cite{Ada.Characters}
22503
provides a set of definitions corresponding to those in the
22504
RM-defined package @cite{Ada.Characters.Latin_1} but with the
22505
types of the constants being @cite{Wide_Wide_Character}
22506
instead of @cite{Character}. The provision of such a package
22507
is specifically authorized by the Ada Reference Manual
22510
@node Ada Characters Wide_Wide_Latin_9 a-chzla9 ads,Ada Containers Formal_Doubly_Linked_Lists a-cfdlli ads,Ada Characters Wide_Wide_Latin_1 a-chzla1 ads,The GNAT Library
22511
@anchor{gnat_rm/the_gnat_library ada-characters-wide-wide-latin-9-a-chzla9-ads}@anchor{291}@anchor{gnat_rm/the_gnat_library id6}@anchor{292}
22512
@section @cite{Ada.Characters.Wide_Wide_Latin_9} (@code{a-chzla9.ads})
22515
@geindex Ada.Characters.Wide_Wide_Latin_9 (a-chzla9.ads)
22517
@geindex Latin_9 constants for Wide_Wide_Character
22519
This child of @cite{Ada.Characters}
22520
provides a set of definitions corresponding to those in the
22521
GNAT defined package @cite{Ada.Characters.Latin_9} but with the
22522
types of the constants being @cite{Wide_Wide_Character}
22523
instead of @cite{Character}. The provision of such a package
22524
is specifically authorized by the Ada Reference Manual
22527
@node Ada Containers Formal_Doubly_Linked_Lists a-cfdlli ads,Ada Containers Formal_Hashed_Maps a-cfhama ads,Ada Characters Wide_Wide_Latin_9 a-chzla9 ads,The GNAT Library
22528
@anchor{gnat_rm/the_gnat_library id7}@anchor{293}@anchor{gnat_rm/the_gnat_library ada-containers-formal-doubly-linked-lists-a-cfdlli-ads}@anchor{294}
22529
@section @cite{Ada.Containers.Formal_Doubly_Linked_Lists} (@code{a-cfdlli.ads})
22532
@geindex Ada.Containers.Formal_Doubly_Linked_Lists (a-cfdlli.ads)
22534
@geindex Formal container for doubly linked lists
22536
This child of @cite{Ada.Containers} defines a modified version of the
22537
Ada 2005 container for doubly linked lists, meant to facilitate formal
22538
verification of code using such containers. The specification of this
22539
unit is compatible with SPARK 2014.
22541
Note that although this container was designed with formal verification
22542
in mind, it may well be generally useful in that it is a simplified more
22543
efficient version than the one defined in the standard. In particular it
22544
does not have the complex overhead required to detect cursor tampering.
22546
@node Ada Containers Formal_Hashed_Maps a-cfhama ads,Ada Containers Formal_Hashed_Sets a-cfhase ads,Ada Containers Formal_Doubly_Linked_Lists a-cfdlli ads,The GNAT Library
22547
@anchor{gnat_rm/the_gnat_library id8}@anchor{295}@anchor{gnat_rm/the_gnat_library ada-containers-formal-hashed-maps-a-cfhama-ads}@anchor{296}
22548
@section @cite{Ada.Containers.Formal_Hashed_Maps} (@code{a-cfhama.ads})
22551
@geindex Ada.Containers.Formal_Hashed_Maps (a-cfhama.ads)
22553
@geindex Formal container for hashed maps
22555
This child of @cite{Ada.Containers} defines a modified version of the
22556
Ada 2005 container for hashed maps, meant to facilitate formal
22557
verification of code using such containers. The specification of this
22558
unit is compatible with SPARK 2014.
22560
Note that although this container was designed with formal verification
22561
in mind, it may well be generally useful in that it is a simplified more
22562
efficient version than the one defined in the standard. In particular it
22563
does not have the complex overhead required to detect cursor tampering.
22565
@node Ada Containers Formal_Hashed_Sets a-cfhase ads,Ada Containers Formal_Ordered_Maps a-cforma ads,Ada Containers Formal_Hashed_Maps a-cfhama ads,The GNAT Library
22566
@anchor{gnat_rm/the_gnat_library id9}@anchor{297}@anchor{gnat_rm/the_gnat_library ada-containers-formal-hashed-sets-a-cfhase-ads}@anchor{298}
22567
@section @cite{Ada.Containers.Formal_Hashed_Sets} (@code{a-cfhase.ads})
22570
@geindex Ada.Containers.Formal_Hashed_Sets (a-cfhase.ads)
22572
@geindex Formal container for hashed sets
22574
This child of @cite{Ada.Containers} defines a modified version of the
22575
Ada 2005 container for hashed sets, meant to facilitate formal
22576
verification of code using such containers. The specification of this
22577
unit is compatible with SPARK 2014.
22579
Note that although this container was designed with formal verification
22580
in mind, it may well be generally useful in that it is a simplified more
22581
efficient version than the one defined in the standard. In particular it
22582
does not have the complex overhead required to detect cursor tampering.
22584
@node Ada Containers Formal_Ordered_Maps a-cforma ads,Ada Containers Formal_Ordered_Sets a-cforse ads,Ada Containers Formal_Hashed_Sets a-cfhase ads,The GNAT Library
22585
@anchor{gnat_rm/the_gnat_library id10}@anchor{299}@anchor{gnat_rm/the_gnat_library ada-containers-formal-ordered-maps-a-cforma-ads}@anchor{29a}
22586
@section @cite{Ada.Containers.Formal_Ordered_Maps} (@code{a-cforma.ads})
22589
@geindex Ada.Containers.Formal_Ordered_Maps (a-cforma.ads)
22591
@geindex Formal container for ordered maps
22593
This child of @cite{Ada.Containers} defines a modified version of the
22594
Ada 2005 container for ordered maps, meant to facilitate formal
22595
verification of code using such containers. The specification of this
22596
unit is compatible with SPARK 2014.
22598
Note that although this container was designed with formal verification
22599
in mind, it may well be generally useful in that it is a simplified more
22600
efficient version than the one defined in the standard. In particular it
22601
does not have the complex overhead required to detect cursor tampering.
22603
@node Ada Containers Formal_Ordered_Sets a-cforse ads,Ada Containers Formal_Vectors a-cofove ads,Ada Containers Formal_Ordered_Maps a-cforma ads,The GNAT Library
22604
@anchor{gnat_rm/the_gnat_library ada-containers-formal-ordered-sets-a-cforse-ads}@anchor{29b}@anchor{gnat_rm/the_gnat_library id11}@anchor{29c}
22605
@section @cite{Ada.Containers.Formal_Ordered_Sets} (@code{a-cforse.ads})
22608
@geindex Ada.Containers.Formal_Ordered_Sets (a-cforse.ads)
22610
@geindex Formal container for ordered sets
22612
This child of @cite{Ada.Containers} defines a modified version of the
22613
Ada 2005 container for ordered sets, meant to facilitate formal
22614
verification of code using such containers. The specification of this
22615
unit is compatible with SPARK 2014.
22617
Note that although this container was designed with formal verification
22618
in mind, it may well be generally useful in that it is a simplified more
22619
efficient version than the one defined in the standard. In particular it
22620
does not have the complex overhead required to detect cursor tampering.
22622
@node Ada Containers Formal_Vectors a-cofove ads,Ada Containers Formal_Indefinite_Vectors a-cfinve ads,Ada Containers Formal_Ordered_Sets a-cforse ads,The GNAT Library
22623
@anchor{gnat_rm/the_gnat_library id12}@anchor{29d}@anchor{gnat_rm/the_gnat_library ada-containers-formal-vectors-a-cofove-ads}@anchor{29e}
22624
@section @cite{Ada.Containers.Formal_Vectors} (@code{a-cofove.ads})
22627
@geindex Ada.Containers.Formal_Vectors (a-cofove.ads)
22629
@geindex Formal container for vectors
22631
This child of @cite{Ada.Containers} defines a modified version of the
22632
Ada 2005 container for vectors, meant to facilitate formal
22633
verification of code using such containers. The specification of this
22634
unit is compatible with SPARK 2014.
22636
Note that although this container was designed with formal verification
22637
in mind, it may well be generally useful in that it is a simplified more
22638
efficient version than the one defined in the standard. In particular it
22639
does not have the complex overhead required to detect cursor tampering.
22641
@node Ada Containers Formal_Indefinite_Vectors a-cfinve ads,Ada Containers Bounded_Holders a-coboho ads,Ada Containers Formal_Vectors a-cofove ads,The GNAT Library
22642
@anchor{gnat_rm/the_gnat_library id13}@anchor{29f}@anchor{gnat_rm/the_gnat_library ada-containers-formal-indefinite-vectors-a-cfinve-ads}@anchor{2a0}
22643
@section @cite{Ada.Containers.Formal_Indefinite_Vectors} (@code{a-cfinve.ads})
22646
@geindex Ada.Containers.Formal_Indefinite_Vectors (a-cfinve.ads)
22648
@geindex Formal container for vectors
22650
This child of @cite{Ada.Containers} defines a modified version of the
22651
Ada 2005 container for vectors of indefinite elements, meant to
22652
facilitate formal verification of code using such containers. The
22653
specification of this unit is compatible with SPARK 2014.
22655
Note that although this container was designed with formal verification
22656
in mind, it may well be generally useful in that it is a simplified more
22657
efficient version than the one defined in the standard. In particular it
22658
does not have the complex overhead required to detect cursor tampering.
22660
@node Ada Containers Bounded_Holders a-coboho ads,Ada Command_Line Environment a-colien ads,Ada Containers Formal_Indefinite_Vectors a-cfinve ads,The GNAT Library
22661
@anchor{gnat_rm/the_gnat_library id14}@anchor{2a1}@anchor{gnat_rm/the_gnat_library ada-containers-bounded-holders-a-coboho-ads}@anchor{2a2}
22662
@section @cite{Ada.Containers.Bounded_Holders} (@code{a-coboho.ads})
22665
@geindex Ada.Containers.Bounded_Holders (a-coboho.ads)
22667
@geindex Formal container for vectors
22669
This child of @cite{Ada.Containers} defines a modified version of
22670
Indefinite_Holders that avoids heap allocation.
22672
@node Ada Command_Line Environment a-colien ads,Ada Command_Line Remove a-colire ads,Ada Containers Bounded_Holders a-coboho ads,The GNAT Library
22673
@anchor{gnat_rm/the_gnat_library ada-command-line-environment-a-colien-ads}@anchor{2a3}@anchor{gnat_rm/the_gnat_library id15}@anchor{2a4}
22674
@section @cite{Ada.Command_Line.Environment} (@code{a-colien.ads})
22677
@geindex Ada.Command_Line.Environment (a-colien.ads)
22679
@geindex Environment entries
22681
This child of @cite{Ada.Command_Line}
22682
provides a mechanism for obtaining environment values on systems
22683
where this concept makes sense.
22685
@node Ada Command_Line Remove a-colire ads,Ada Command_Line Response_File a-clrefi ads,Ada Command_Line Environment a-colien ads,The GNAT Library
22686
@anchor{gnat_rm/the_gnat_library id16}@anchor{2a5}@anchor{gnat_rm/the_gnat_library ada-command-line-remove-a-colire-ads}@anchor{2a6}
22687
@section @cite{Ada.Command_Line.Remove} (@code{a-colire.ads})
22690
@geindex Ada.Command_Line.Remove (a-colire.ads)
22692
@geindex Removing command line arguments
22694
@geindex Command line
22695
@geindex argument removal
22697
This child of @cite{Ada.Command_Line}
22698
provides a mechanism for logically removing
22699
arguments from the argument list. Once removed, an argument is not visible
22700
to further calls on the subprograms in @cite{Ada.Command_Line} will not
22701
see the removed argument.
22703
@node Ada Command_Line Response_File a-clrefi ads,Ada Direct_IO C_Streams a-diocst ads,Ada Command_Line Remove a-colire ads,The GNAT Library
22704
@anchor{gnat_rm/the_gnat_library ada-command-line-response-file-a-clrefi-ads}@anchor{2a7}@anchor{gnat_rm/the_gnat_library id17}@anchor{2a8}
22705
@section @cite{Ada.Command_Line.Response_File} (@code{a-clrefi.ads})
22708
@geindex Ada.Command_Line.Response_File (a-clrefi.ads)
22710
@geindex Response file for command line
22712
@geindex Command line
22713
@geindex response file
22715
@geindex Command line
22716
@geindex handling long command lines
22718
This child of @cite{Ada.Command_Line} provides a mechanism facilities for
22719
getting command line arguments from a text file, called a "response file".
22720
Using a response file allow passing a set of arguments to an executable longer
22721
than the maximum allowed by the system on the command line.
22723
@node Ada Direct_IO C_Streams a-diocst ads,Ada Exceptions Is_Null_Occurrence a-einuoc ads,Ada Command_Line Response_File a-clrefi ads,The GNAT Library
22724
@anchor{gnat_rm/the_gnat_library id18}@anchor{2a9}@anchor{gnat_rm/the_gnat_library ada-direct-io-c-streams-a-diocst-ads}@anchor{2aa}
22725
@section @cite{Ada.Direct_IO.C_Streams} (@code{a-diocst.ads})
22728
@geindex Ada.Direct_IO.C_Streams (a-diocst.ads)
22731
@geindex Interfacing with Direct_IO
22733
This package provides subprograms that allow interfacing between
22734
C streams and @cite{Direct_IO}. The stream identifier can be
22735
extracted from a file opened on the Ada side, and an Ada file
22736
can be constructed from a stream opened on the C side.
22738
@node Ada Exceptions Is_Null_Occurrence a-einuoc ads,Ada Exceptions Last_Chance_Handler a-elchha ads,Ada Direct_IO C_Streams a-diocst ads,The GNAT Library
22739
@anchor{gnat_rm/the_gnat_library id19}@anchor{2ab}@anchor{gnat_rm/the_gnat_library ada-exceptions-is-null-occurrence-a-einuoc-ads}@anchor{2ac}
22740
@section @cite{Ada.Exceptions.Is_Null_Occurrence} (@code{a-einuoc.ads})
22743
@geindex Ada.Exceptions.Is_Null_Occurrence (a-einuoc.ads)
22745
@geindex Null_Occurrence
22746
@geindex testing for
22748
This child subprogram provides a way of testing for the null
22749
exception occurrence (@cite{Null_Occurrence}) without raising
22752
@node Ada Exceptions Last_Chance_Handler a-elchha ads,Ada Exceptions Traceback a-exctra ads,Ada Exceptions Is_Null_Occurrence a-einuoc ads,The GNAT Library
22753
@anchor{gnat_rm/the_gnat_library id20}@anchor{2ad}@anchor{gnat_rm/the_gnat_library ada-exceptions-last-chance-handler-a-elchha-ads}@anchor{2ae}
22754
@section @cite{Ada.Exceptions.Last_Chance_Handler} (@code{a-elchha.ads})
22757
@geindex Ada.Exceptions.Last_Chance_Handler (a-elchha.ads)
22759
@geindex Null_Occurrence
22760
@geindex testing for
22762
This child subprogram is used for handling otherwise unhandled
22763
exceptions (hence the name last chance), and perform clean ups before
22764
terminating the program. Note that this subprogram never returns.
22766
@node Ada Exceptions Traceback a-exctra ads,Ada Sequential_IO C_Streams a-siocst ads,Ada Exceptions Last_Chance_Handler a-elchha ads,The GNAT Library
22767
@anchor{gnat_rm/the_gnat_library ada-exceptions-traceback-a-exctra-ads}@anchor{2af}@anchor{gnat_rm/the_gnat_library id21}@anchor{2b0}
22768
@section @cite{Ada.Exceptions.Traceback} (@code{a-exctra.ads})
22771
@geindex Ada.Exceptions.Traceback (a-exctra.ads)
22773
@geindex Traceback for Exception Occurrence
22775
This child package provides the subprogram (@cite{Tracebacks}) to
22776
give a traceback array of addresses based on an exception
22779
@node Ada Sequential_IO C_Streams a-siocst ads,Ada Streams Stream_IO C_Streams a-ssicst ads,Ada Exceptions Traceback a-exctra ads,The GNAT Library
22780
@anchor{gnat_rm/the_gnat_library ada-sequential-io-c-streams-a-siocst-ads}@anchor{2b1}@anchor{gnat_rm/the_gnat_library id22}@anchor{2b2}
22781
@section @cite{Ada.Sequential_IO.C_Streams} (@code{a-siocst.ads})
22784
@geindex Ada.Sequential_IO.C_Streams (a-siocst.ads)
22787
@geindex Interfacing with Sequential_IO
22789
This package provides subprograms that allow interfacing between
22790
C streams and @cite{Sequential_IO}. The stream identifier can be
22791
extracted from a file opened on the Ada side, and an Ada file
22792
can be constructed from a stream opened on the C side.
22794
@node Ada Streams Stream_IO C_Streams a-ssicst ads,Ada Strings Unbounded Text_IO a-suteio ads,Ada Sequential_IO C_Streams a-siocst ads,The GNAT Library
22795
@anchor{gnat_rm/the_gnat_library id23}@anchor{2b3}@anchor{gnat_rm/the_gnat_library ada-streams-stream-io-c-streams-a-ssicst-ads}@anchor{2b4}
22796
@section @cite{Ada.Streams.Stream_IO.C_Streams} (@code{a-ssicst.ads})
22799
@geindex Ada.Streams.Stream_IO.C_Streams (a-ssicst.ads)
22802
@geindex Interfacing with Stream_IO
22804
This package provides subprograms that allow interfacing between
22805
C streams and @cite{Stream_IO}. The stream identifier can be
22806
extracted from a file opened on the Ada side, and an Ada file
22807
can be constructed from a stream opened on the C side.
22809
@node Ada Strings Unbounded Text_IO a-suteio ads,Ada Strings Wide_Unbounded Wide_Text_IO a-swuwti ads,Ada Streams Stream_IO C_Streams a-ssicst ads,The GNAT Library
22810
@anchor{gnat_rm/the_gnat_library ada-strings-unbounded-text-io-a-suteio-ads}@anchor{2b5}@anchor{gnat_rm/the_gnat_library id24}@anchor{2b6}
22811
@section @cite{Ada.Strings.Unbounded.Text_IO} (@code{a-suteio.ads})
22814
@geindex Ada.Strings.Unbounded.Text_IO (a-suteio.ads)
22816
@geindex Unbounded_String
22817
@geindex IO support
22820
@geindex extensions for unbounded strings
22822
This package provides subprograms for Text_IO for unbounded
22823
strings, avoiding the necessity for an intermediate operation
22824
with ordinary strings.
22826
@node Ada Strings Wide_Unbounded Wide_Text_IO a-swuwti ads,Ada Strings Wide_Wide_Unbounded Wide_Wide_Text_IO a-szuzti ads,Ada Strings Unbounded Text_IO a-suteio ads,The GNAT Library
22827
@anchor{gnat_rm/the_gnat_library id25}@anchor{2b7}@anchor{gnat_rm/the_gnat_library ada-strings-wide-unbounded-wide-text-io-a-swuwti-ads}@anchor{2b8}
22828
@section @cite{Ada.Strings.Wide_Unbounded.Wide_Text_IO} (@code{a-swuwti.ads})
22831
@geindex Ada.Strings.Wide_Unbounded.Wide_Text_IO (a-swuwti.ads)
22833
@geindex Unbounded_Wide_String
22834
@geindex IO support
22837
@geindex extensions for unbounded wide strings
22839
This package provides subprograms for Text_IO for unbounded
22840
wide strings, avoiding the necessity for an intermediate operation
22841
with ordinary wide strings.
22843
@node Ada Strings Wide_Wide_Unbounded Wide_Wide_Text_IO a-szuzti ads,Ada Text_IO C_Streams a-tiocst ads,Ada Strings Wide_Unbounded Wide_Text_IO a-swuwti ads,The GNAT Library
22844
@anchor{gnat_rm/the_gnat_library ada-strings-wide-wide-unbounded-wide-wide-text-io-a-szuzti-ads}@anchor{2b9}@anchor{gnat_rm/the_gnat_library id26}@anchor{2ba}
22845
@section @cite{Ada.Strings.Wide_Wide_Unbounded.Wide_Wide_Text_IO} (@code{a-szuzti.ads})
22848
@geindex Ada.Strings.Wide_Wide_Unbounded.Wide_Wide_Text_IO (a-szuzti.ads)
22850
@geindex Unbounded_Wide_Wide_String
22851
@geindex IO support
22854
@geindex extensions for unbounded wide wide strings
22856
This package provides subprograms for Text_IO for unbounded
22857
wide wide strings, avoiding the necessity for an intermediate operation
22858
with ordinary wide wide strings.
22860
@node Ada Text_IO C_Streams a-tiocst ads,Ada Text_IO Reset_Standard_Files a-tirsfi ads,Ada Strings Wide_Wide_Unbounded Wide_Wide_Text_IO a-szuzti ads,The GNAT Library
22861
@anchor{gnat_rm/the_gnat_library ada-text-io-c-streams-a-tiocst-ads}@anchor{2bb}@anchor{gnat_rm/the_gnat_library id27}@anchor{2bc}
22862
@section @cite{Ada.Text_IO.C_Streams} (@code{a-tiocst.ads})
22865
@geindex Ada.Text_IO.C_Streams (a-tiocst.ads)
22868
@geindex Interfacing with `Text_IO`
22870
This package provides subprograms that allow interfacing between
22871
C streams and @cite{Text_IO}. The stream identifier can be
22872
extracted from a file opened on the Ada side, and an Ada file
22873
can be constructed from a stream opened on the C side.
22875
@node Ada Text_IO Reset_Standard_Files a-tirsfi ads,Ada Wide_Characters Unicode a-wichun ads,Ada Text_IO C_Streams a-tiocst ads,The GNAT Library
22876
@anchor{gnat_rm/the_gnat_library id28}@anchor{2bd}@anchor{gnat_rm/the_gnat_library ada-text-io-reset-standard-files-a-tirsfi-ads}@anchor{2be}
22877
@section @cite{Ada.Text_IO.Reset_Standard_Files} (@code{a-tirsfi.ads})
22880
@geindex Ada.Text_IO.Reset_Standard_Files (a-tirsfi.ads)
22882
@geindex Text_IO resetting standard files
22884
This procedure is used to reset the status of the standard files used
22885
by Ada.Text_IO. This is useful in a situation (such as a restart in an
22886
embedded application) where the status of the files may change during
22887
execution (for example a standard input file may be redefined to be
22890
@node Ada Wide_Characters Unicode a-wichun ads,Ada Wide_Text_IO C_Streams a-wtcstr ads,Ada Text_IO Reset_Standard_Files a-tirsfi ads,The GNAT Library
22891
@anchor{gnat_rm/the_gnat_library id29}@anchor{2bf}@anchor{gnat_rm/the_gnat_library ada-wide-characters-unicode-a-wichun-ads}@anchor{2c0}
22892
@section @cite{Ada.Wide_Characters.Unicode} (@code{a-wichun.ads})
22895
@geindex Ada.Wide_Characters.Unicode (a-wichun.ads)
22897
@geindex Unicode categorization
22898
@geindex Wide_Character
22900
This package provides subprograms that allow categorization of
22901
Wide_Character values according to Unicode categories.
22903
@node Ada Wide_Text_IO C_Streams a-wtcstr ads,Ada Wide_Text_IO Reset_Standard_Files a-wrstfi ads,Ada Wide_Characters Unicode a-wichun ads,The GNAT Library
22904
@anchor{gnat_rm/the_gnat_library ada-wide-text-io-c-streams-a-wtcstr-ads}@anchor{2c1}@anchor{gnat_rm/the_gnat_library id30}@anchor{2c2}
22905
@section @cite{Ada.Wide_Text_IO.C_Streams} (@code{a-wtcstr.ads})
22908
@geindex Ada.Wide_Text_IO.C_Streams (a-wtcstr.ads)
22911
@geindex Interfacing with `Wide_Text_IO`
22913
This package provides subprograms that allow interfacing between
22914
C streams and @cite{Wide_Text_IO}. The stream identifier can be
22915
extracted from a file opened on the Ada side, and an Ada file
22916
can be constructed from a stream opened on the C side.
22918
@node Ada Wide_Text_IO Reset_Standard_Files a-wrstfi ads,Ada Wide_Wide_Characters Unicode a-zchuni ads,Ada Wide_Text_IO C_Streams a-wtcstr ads,The GNAT Library
22919
@anchor{gnat_rm/the_gnat_library ada-wide-text-io-reset-standard-files-a-wrstfi-ads}@anchor{2c3}@anchor{gnat_rm/the_gnat_library id31}@anchor{2c4}
22920
@section @cite{Ada.Wide_Text_IO.Reset_Standard_Files} (@code{a-wrstfi.ads})
22923
@geindex Ada.Wide_Text_IO.Reset_Standard_Files (a-wrstfi.ads)
22925
@geindex Wide_Text_IO resetting standard files
22927
This procedure is used to reset the status of the standard files used
22928
by Ada.Wide_Text_IO. This is useful in a situation (such as a restart in an
22929
embedded application) where the status of the files may change during
22930
execution (for example a standard input file may be redefined to be
22933
@node Ada Wide_Wide_Characters Unicode a-zchuni ads,Ada Wide_Wide_Text_IO C_Streams a-ztcstr ads,Ada Wide_Text_IO Reset_Standard_Files a-wrstfi ads,The GNAT Library
22934
@anchor{gnat_rm/the_gnat_library id32}@anchor{2c5}@anchor{gnat_rm/the_gnat_library ada-wide-wide-characters-unicode-a-zchuni-ads}@anchor{2c6}
22935
@section @cite{Ada.Wide_Wide_Characters.Unicode} (@code{a-zchuni.ads})
22938
@geindex Ada.Wide_Wide_Characters.Unicode (a-zchuni.ads)
22940
@geindex Unicode categorization
22941
@geindex Wide_Wide_Character
22943
This package provides subprograms that allow categorization of
22944
Wide_Wide_Character values according to Unicode categories.
22946
@node Ada Wide_Wide_Text_IO C_Streams a-ztcstr ads,Ada Wide_Wide_Text_IO Reset_Standard_Files a-zrstfi ads,Ada Wide_Wide_Characters Unicode a-zchuni ads,The GNAT Library
22947
@anchor{gnat_rm/the_gnat_library id33}@anchor{2c7}@anchor{gnat_rm/the_gnat_library ada-wide-wide-text-io-c-streams-a-ztcstr-ads}@anchor{2c8}
22948
@section @cite{Ada.Wide_Wide_Text_IO.C_Streams} (@code{a-ztcstr.ads})
22951
@geindex Ada.Wide_Wide_Text_IO.C_Streams (a-ztcstr.ads)
22954
@geindex Interfacing with `Wide_Wide_Text_IO`
22956
This package provides subprograms that allow interfacing between
22957
C streams and @cite{Wide_Wide_Text_IO}. The stream identifier can be
22958
extracted from a file opened on the Ada side, and an Ada file
22959
can be constructed from a stream opened on the C side.
22961
@node Ada Wide_Wide_Text_IO Reset_Standard_Files a-zrstfi ads,GNAT Altivec g-altive ads,Ada Wide_Wide_Text_IO C_Streams a-ztcstr ads,The GNAT Library
22962
@anchor{gnat_rm/the_gnat_library id34}@anchor{2c9}@anchor{gnat_rm/the_gnat_library ada-wide-wide-text-io-reset-standard-files-a-zrstfi-ads}@anchor{2ca}
22963
@section @cite{Ada.Wide_Wide_Text_IO.Reset_Standard_Files} (@code{a-zrstfi.ads})
22966
@geindex Ada.Wide_Wide_Text_IO.Reset_Standard_Files (a-zrstfi.ads)
22968
@geindex Wide_Wide_Text_IO resetting standard files
22970
This procedure is used to reset the status of the standard files used
22971
by Ada.Wide_Wide_Text_IO. This is useful in a situation (such as a
22972
restart in an embedded application) where the status of the files may
22973
change during execution (for example a standard input file may be
22974
redefined to be interactive).
22976
@node GNAT Altivec g-altive ads,GNAT Altivec Conversions g-altcon ads,Ada Wide_Wide_Text_IO Reset_Standard_Files a-zrstfi ads,The GNAT Library
22977
@anchor{gnat_rm/the_gnat_library gnat-altivec-g-altive-ads}@anchor{2cb}@anchor{gnat_rm/the_gnat_library id35}@anchor{2cc}
22978
@section @cite{GNAT.Altivec} (@code{g-altive.ads})
22981
@geindex GNAT.Altivec (g-altive.ads)
22985
This is the root package of the GNAT AltiVec binding. It provides
22986
definitions of constants and types common to all the versions of the
22989
@node GNAT Altivec Conversions g-altcon ads,GNAT Altivec Vector_Operations g-alveop ads,GNAT Altivec g-altive ads,The GNAT Library
22990
@anchor{gnat_rm/the_gnat_library id36}@anchor{2cd}@anchor{gnat_rm/the_gnat_library gnat-altivec-conversions-g-altcon-ads}@anchor{2ce}
22991
@section @cite{GNAT.Altivec.Conversions} (@code{g-altcon.ads})
22994
@geindex GNAT.Altivec.Conversions (g-altcon.ads)
22998
This package provides the Vector/View conversion routines.
23000
@node GNAT Altivec Vector_Operations g-alveop ads,GNAT Altivec Vector_Types g-alvety ads,GNAT Altivec Conversions g-altcon ads,The GNAT Library
23001
@anchor{gnat_rm/the_gnat_library id37}@anchor{2cf}@anchor{gnat_rm/the_gnat_library gnat-altivec-vector-operations-g-alveop-ads}@anchor{2d0}
23002
@section @cite{GNAT.Altivec.Vector_Operations} (@code{g-alveop.ads})
23005
@geindex GNAT.Altivec.Vector_Operations (g-alveop.ads)
23009
This package exposes the Ada interface to the AltiVec operations on
23010
vector objects. A soft emulation is included by default in the GNAT
23011
library. The hard binding is provided as a separate package. This unit
23012
is common to both bindings.
23014
@node GNAT Altivec Vector_Types g-alvety ads,GNAT Altivec Vector_Views g-alvevi ads,GNAT Altivec Vector_Operations g-alveop ads,The GNAT Library
23015
@anchor{gnat_rm/the_gnat_library gnat-altivec-vector-types-g-alvety-ads}@anchor{2d1}@anchor{gnat_rm/the_gnat_library id38}@anchor{2d2}
23016
@section @cite{GNAT.Altivec.Vector_Types} (@code{g-alvety.ads})
23019
@geindex GNAT.Altivec.Vector_Types (g-alvety.ads)
23023
This package exposes the various vector types part of the Ada binding
23024
to AltiVec facilities.
23026
@node GNAT Altivec Vector_Views g-alvevi ads,GNAT Array_Split g-arrspl ads,GNAT Altivec Vector_Types g-alvety ads,The GNAT Library
23027
@anchor{gnat_rm/the_gnat_library gnat-altivec-vector-views-g-alvevi-ads}@anchor{2d3}@anchor{gnat_rm/the_gnat_library id39}@anchor{2d4}
23028
@section @cite{GNAT.Altivec.Vector_Views} (@code{g-alvevi.ads})
23031
@geindex GNAT.Altivec.Vector_Views (g-alvevi.ads)
23035
This package provides public 'View' data types from/to which private
23036
vector representations can be converted via
23037
GNAT.Altivec.Conversions. This allows convenient access to individual
23038
vector elements and provides a simple way to initialize vector
23041
@node GNAT Array_Split g-arrspl ads,GNAT AWK g-awk ads,GNAT Altivec Vector_Views g-alvevi ads,The GNAT Library
23042
@anchor{gnat_rm/the_gnat_library gnat-array-split-g-arrspl-ads}@anchor{2d5}@anchor{gnat_rm/the_gnat_library id40}@anchor{2d6}
23043
@section @cite{GNAT.Array_Split} (@code{g-arrspl.ads})
23046
@geindex GNAT.Array_Split (g-arrspl.ads)
23048
@geindex Array splitter
23050
Useful array-manipulation routines: given a set of separators, split
23051
an array wherever the separators appear, and provide direct access
23052
to the resulting slices.
23054
@node GNAT AWK g-awk ads,GNAT Bind_Environment g-binenv ads,GNAT Array_Split g-arrspl ads,The GNAT Library
23055
@anchor{gnat_rm/the_gnat_library id41}@anchor{2d7}@anchor{gnat_rm/the_gnat_library gnat-awk-g-awk-ads}@anchor{2d8}
23056
@section @cite{GNAT.AWK} (@code{g-awk.ads})
23059
@geindex GNAT.AWK (g-awk.ads)
23065
Provides AWK-like parsing functions, with an easy interface for parsing one
23066
or more files containing formatted data. The file is viewed as a database
23067
where each record is a line and a field is a data element in this line.
23069
@node GNAT Bind_Environment g-binenv ads,GNAT Bounded_Buffers g-boubuf ads,GNAT AWK g-awk ads,The GNAT Library
23070
@anchor{gnat_rm/the_gnat_library gnat-bind-environment-g-binenv-ads}@anchor{2d9}@anchor{gnat_rm/the_gnat_library id42}@anchor{2da}
23071
@section @cite{GNAT.Bind_Environment} (@code{g-binenv.ads})
23074
@geindex GNAT.Bind_Environment (g-binenv.ads)
23076
@geindex Bind environment
23078
Provides access to key=value associations captured at bind time.
23079
These associations can be specified using the @cite{-V} binder command
23082
@node GNAT Bounded_Buffers g-boubuf ads,GNAT Bounded_Mailboxes g-boumai ads,GNAT Bind_Environment g-binenv ads,The GNAT Library
23083
@anchor{gnat_rm/the_gnat_library gnat-bounded-buffers-g-boubuf-ads}@anchor{2db}@anchor{gnat_rm/the_gnat_library id43}@anchor{2dc}
23084
@section @cite{GNAT.Bounded_Buffers} (@code{g-boubuf.ads})
23087
@geindex GNAT.Bounded_Buffers (g-boubuf.ads)
23091
@geindex Bounded Buffers
23093
Provides a concurrent generic bounded buffer abstraction. Instances are
23094
useful directly or as parts of the implementations of other abstractions,
23097
@node GNAT Bounded_Mailboxes g-boumai ads,GNAT Bubble_Sort g-bubsor ads,GNAT Bounded_Buffers g-boubuf ads,The GNAT Library
23098
@anchor{gnat_rm/the_gnat_library id44}@anchor{2dd}@anchor{gnat_rm/the_gnat_library gnat-bounded-mailboxes-g-boumai-ads}@anchor{2de}
23099
@section @cite{GNAT.Bounded_Mailboxes} (@code{g-boumai.ads})
23102
@geindex GNAT.Bounded_Mailboxes (g-boumai.ads)
23108
Provides a thread-safe asynchronous intertask mailbox communication facility.
23110
@node GNAT Bubble_Sort g-bubsor ads,GNAT Bubble_Sort_A g-busora ads,GNAT Bounded_Mailboxes g-boumai ads,The GNAT Library
23111
@anchor{gnat_rm/the_gnat_library gnat-bubble-sort-g-bubsor-ads}@anchor{2df}@anchor{gnat_rm/the_gnat_library id45}@anchor{2e0}
23112
@section @cite{GNAT.Bubble_Sort} (@code{g-bubsor.ads})
23115
@geindex GNAT.Bubble_Sort (g-bubsor.ads)
23119
@geindex Bubble sort
23121
Provides a general implementation of bubble sort usable for sorting arbitrary
23122
data items. Exchange and comparison procedures are provided by passing
23123
access-to-procedure values.
23125
@node GNAT Bubble_Sort_A g-busora ads,GNAT Bubble_Sort_G g-busorg ads,GNAT Bubble_Sort g-bubsor ads,The GNAT Library
23126
@anchor{gnat_rm/the_gnat_library id46}@anchor{2e1}@anchor{gnat_rm/the_gnat_library gnat-bubble-sort-a-g-busora-ads}@anchor{2e2}
23127
@section @cite{GNAT.Bubble_Sort_A} (@code{g-busora.ads})
23130
@geindex GNAT.Bubble_Sort_A (g-busora.ads)
23134
@geindex Bubble sort
23136
Provides a general implementation of bubble sort usable for sorting arbitrary
23137
data items. Move and comparison procedures are provided by passing
23138
access-to-procedure values. This is an older version, retained for
23139
compatibility. Usually @cite{GNAT.Bubble_Sort} will be preferable.
23141
@node GNAT Bubble_Sort_G g-busorg ads,GNAT Byte_Order_Mark g-byorma ads,GNAT Bubble_Sort_A g-busora ads,The GNAT Library
23142
@anchor{gnat_rm/the_gnat_library id47}@anchor{2e3}@anchor{gnat_rm/the_gnat_library gnat-bubble-sort-g-g-busorg-ads}@anchor{2e4}
23143
@section @cite{GNAT.Bubble_Sort_G} (@code{g-busorg.ads})
23146
@geindex GNAT.Bubble_Sort_G (g-busorg.ads)
23150
@geindex Bubble sort
23152
Similar to @cite{Bubble_Sort_A} except that the move and sorting procedures
23153
are provided as generic parameters, this improves efficiency, especially
23154
if the procedures can be inlined, at the expense of duplicating code for
23155
multiple instantiations.
23157
@node GNAT Byte_Order_Mark g-byorma ads,GNAT Byte_Swapping g-bytswa ads,GNAT Bubble_Sort_G g-busorg ads,The GNAT Library
23158
@anchor{gnat_rm/the_gnat_library gnat-byte-order-mark-g-byorma-ads}@anchor{2e5}@anchor{gnat_rm/the_gnat_library id48}@anchor{2e6}
23159
@section @cite{GNAT.Byte_Order_Mark} (@code{g-byorma.ads})
23162
@geindex GNAT.Byte_Order_Mark (g-byorma.ads)
23164
@geindex UTF-8 representation
23166
@geindex Wide characte representations
23168
Provides a routine which given a string, reads the start of the string to
23169
see whether it is one of the standard byte order marks (BOM's) which signal
23170
the encoding of the string. The routine includes detection of special XML
23171
sequences for various UCS input formats.
23173
@node GNAT Byte_Swapping g-bytswa ads,GNAT Calendar g-calend ads,GNAT Byte_Order_Mark g-byorma ads,The GNAT Library
23174
@anchor{gnat_rm/the_gnat_library gnat-byte-swapping-g-bytswa-ads}@anchor{2e7}@anchor{gnat_rm/the_gnat_library id49}@anchor{2e8}
23175
@section @cite{GNAT.Byte_Swapping} (@code{g-bytswa.ads})
23178
@geindex GNAT.Byte_Swapping (g-bytswa.ads)
23180
@geindex Byte swapping
23182
@geindex Endianness
23184
General routines for swapping the bytes in 2-, 4-, and 8-byte quantities.
23185
Machine-specific implementations are available in some cases.
23187
@node GNAT Calendar g-calend ads,GNAT Calendar Time_IO g-catiio ads,GNAT Byte_Swapping g-bytswa ads,The GNAT Library
23188
@anchor{gnat_rm/the_gnat_library id50}@anchor{2e9}@anchor{gnat_rm/the_gnat_library gnat-calendar-g-calend-ads}@anchor{2ea}
23189
@section @cite{GNAT.Calendar} (@code{g-calend.ads})
23192
@geindex GNAT.Calendar (g-calend.ads)
23196
Extends the facilities provided by @cite{Ada.Calendar} to include handling
23197
of days of the week, an extended @cite{Split} and @cite{Time_Of} capability.
23198
Also provides conversion of @cite{Ada.Calendar.Time} values to and from the
23199
C @cite{timeval} format.
23201
@node GNAT Calendar Time_IO g-catiio ads,GNAT CRC32 g-crc32 ads,GNAT Calendar g-calend ads,The GNAT Library
23202
@anchor{gnat_rm/the_gnat_library gnat-calendar-time-io-g-catiio-ads}@anchor{2eb}@anchor{gnat_rm/the_gnat_library id51}@anchor{2ec}
23203
@section @cite{GNAT.Calendar.Time_IO} (@code{g-catiio.ads})
23210
@geindex GNAT.Calendar.Time_IO (g-catiio.ads)
23212
@node GNAT CRC32 g-crc32 ads,GNAT Case_Util g-casuti ads,GNAT Calendar Time_IO g-catiio ads,The GNAT Library
23213
@anchor{gnat_rm/the_gnat_library id52}@anchor{2ed}@anchor{gnat_rm/the_gnat_library gnat-crc32-g-crc32-ads}@anchor{2ee}
23214
@section @cite{GNAT.CRC32} (@code{g-crc32.ads})
23217
@geindex GNAT.CRC32 (g-crc32.ads)
23221
@geindex Cyclic Redundancy Check
23223
This package implements the CRC-32 algorithm. For a full description
23224
of this algorithm see
23225
@emph{Computation of Cyclic Redundancy Checks via Table Look-Up},
23226
@cite{Communications of the ACM}, Vol. 31 No. 8, pp. 1008-1013,
23227
Aug. 1988. Sarwate, D.V.
23229
@node GNAT Case_Util g-casuti ads,GNAT CGI g-cgi ads,GNAT CRC32 g-crc32 ads,The GNAT Library
23230
@anchor{gnat_rm/the_gnat_library gnat-case-util-g-casuti-ads}@anchor{2ef}@anchor{gnat_rm/the_gnat_library id53}@anchor{2f0}
23231
@section @cite{GNAT.Case_Util} (@code{g-casuti.ads})
23234
@geindex GNAT.Case_Util (g-casuti.ads)
23236
@geindex Casing utilities
23238
@geindex Character handling (`GNAT.Case_Util`)
23240
A set of simple routines for handling upper and lower casing of strings
23241
without the overhead of the full casing tables
23242
in @cite{Ada.Characters.Handling}.
23244
@node GNAT CGI g-cgi ads,GNAT CGI Cookie g-cgicoo ads,GNAT Case_Util g-casuti ads,The GNAT Library
23245
@anchor{gnat_rm/the_gnat_library id54}@anchor{2f1}@anchor{gnat_rm/the_gnat_library gnat-cgi-g-cgi-ads}@anchor{2f2}
23246
@section @cite{GNAT.CGI} (@code{g-cgi.ads})
23249
@geindex GNAT.CGI (g-cgi.ads)
23251
@geindex CGI (Common Gateway Interface)
23253
This is a package for interfacing a GNAT program with a Web server via the
23254
Common Gateway Interface (CGI). Basically this package parses the CGI
23255
parameters, which are a set of key/value pairs sent by the Web server. It
23256
builds a table whose index is the key and provides some services to deal
23259
@node GNAT CGI Cookie g-cgicoo ads,GNAT CGI Debug g-cgideb ads,GNAT CGI g-cgi ads,The GNAT Library
23260
@anchor{gnat_rm/the_gnat_library gnat-cgi-cookie-g-cgicoo-ads}@anchor{2f3}@anchor{gnat_rm/the_gnat_library id55}@anchor{2f4}
23261
@section @cite{GNAT.CGI.Cookie} (@code{g-cgicoo.ads})
23264
@geindex GNAT.CGI.Cookie (g-cgicoo.ads)
23266
@geindex CGI (Common Gateway Interface) cookie support
23268
@geindex Cookie support in CGI
23270
This is a package to interface a GNAT program with a Web server via the
23271
Common Gateway Interface (CGI). It exports services to deal with Web
23272
cookies (piece of information kept in the Web client software).
23274
@node GNAT CGI Debug g-cgideb ads,GNAT Command_Line g-comlin ads,GNAT CGI Cookie g-cgicoo ads,The GNAT Library
23275
@anchor{gnat_rm/the_gnat_library gnat-cgi-debug-g-cgideb-ads}@anchor{2f5}@anchor{gnat_rm/the_gnat_library id56}@anchor{2f6}
23276
@section @cite{GNAT.CGI.Debug} (@code{g-cgideb.ads})
23279
@geindex GNAT.CGI.Debug (g-cgideb.ads)
23281
@geindex CGI (Common Gateway Interface) debugging
23283
This is a package to help debugging CGI (Common Gateway Interface)
23284
programs written in Ada.
23286
@node GNAT Command_Line g-comlin ads,GNAT Compiler_Version g-comver ads,GNAT CGI Debug g-cgideb ads,The GNAT Library
23287
@anchor{gnat_rm/the_gnat_library id57}@anchor{2f7}@anchor{gnat_rm/the_gnat_library gnat-command-line-g-comlin-ads}@anchor{2f8}
23288
@section @cite{GNAT.Command_Line} (@code{g-comlin.ads})
23291
@geindex GNAT.Command_Line (g-comlin.ads)
23293
@geindex Command line
23295
Provides a high level interface to @cite{Ada.Command_Line} facilities,
23296
including the ability to scan for named switches with optional parameters
23297
and expand file names using wild card notations.
23299
@node GNAT Compiler_Version g-comver ads,GNAT Ctrl_C g-ctrl_c ads,GNAT Command_Line g-comlin ads,The GNAT Library
23300
@anchor{gnat_rm/the_gnat_library gnat-compiler-version-g-comver-ads}@anchor{2f9}@anchor{gnat_rm/the_gnat_library id58}@anchor{2fa}
23301
@section @cite{GNAT.Compiler_Version} (@code{g-comver.ads})
23304
@geindex GNAT.Compiler_Version (g-comver.ads)
23306
@geindex Compiler Version
23309
@geindex of compiler
23311
Provides a routine for obtaining the version of the compiler used to
23312
compile the program. More accurately this is the version of the binder
23313
used to bind the program (this will normally be the same as the version
23314
of the compiler if a consistent tool set is used to compile all units
23317
@node GNAT Ctrl_C g-ctrl_c ads,GNAT Current_Exception g-curexc ads,GNAT Compiler_Version g-comver ads,The GNAT Library
23318
@anchor{gnat_rm/the_gnat_library gnat-ctrl-c-g-ctrl-c-ads}@anchor{2fb}@anchor{gnat_rm/the_gnat_library id59}@anchor{2fc}
23319
@section @cite{GNAT.Ctrl_C} (@code{g-ctrl_c.ads})
23322
@geindex GNAT.Ctrl_C (g-ctrl_c.ads)
23326
Provides a simple interface to handle Ctrl-C keyboard events.
23328
@node GNAT Current_Exception g-curexc ads,GNAT Debug_Pools g-debpoo ads,GNAT Ctrl_C g-ctrl_c ads,The GNAT Library
23329
@anchor{gnat_rm/the_gnat_library id60}@anchor{2fd}@anchor{gnat_rm/the_gnat_library gnat-current-exception-g-curexc-ads}@anchor{2fe}
23330
@section @cite{GNAT.Current_Exception} (@code{g-curexc.ads})
23333
@geindex GNAT.Current_Exception (g-curexc.ads)
23335
@geindex Current exception
23337
@geindex Exception retrieval
23339
Provides access to information on the current exception that has been raised
23340
without the need for using the Ada 95 / Ada 2005 exception choice parameter
23341
specification syntax.
23342
This is particularly useful in simulating typical facilities for
23343
obtaining information about exceptions provided by Ada 83 compilers.
23345
@node GNAT Debug_Pools g-debpoo ads,GNAT Debug_Utilities g-debuti ads,GNAT Current_Exception g-curexc ads,The GNAT Library
23346
@anchor{gnat_rm/the_gnat_library gnat-debug-pools-g-debpoo-ads}@anchor{2ff}@anchor{gnat_rm/the_gnat_library id61}@anchor{300}
23347
@section @cite{GNAT.Debug_Pools} (@code{g-debpoo.ads})
23350
@geindex GNAT.Debug_Pools (g-debpoo.ads)
23354
@geindex Debug pools
23356
@geindex Memory corruption debugging
23358
Provide a debugging storage pools that helps tracking memory corruption
23360
See @cite{The GNAT Debug_Pool Facility} section in the @cite{GNAT User's Guide}.
23362
@node GNAT Debug_Utilities g-debuti ads,GNAT Decode_String g-decstr ads,GNAT Debug_Pools g-debpoo ads,The GNAT Library
23363
@anchor{gnat_rm/the_gnat_library gnat-debug-utilities-g-debuti-ads}@anchor{301}@anchor{gnat_rm/the_gnat_library id62}@anchor{302}
23364
@section @cite{GNAT.Debug_Utilities} (@code{g-debuti.ads})
23367
@geindex GNAT.Debug_Utilities (g-debuti.ads)
23371
Provides a few useful utilities for debugging purposes, including conversion
23372
to and from string images of address values. Supports both C and Ada formats
23373
for hexadecimal literals.
23375
@node GNAT Decode_String g-decstr ads,GNAT Decode_UTF8_String g-deutst ads,GNAT Debug_Utilities g-debuti ads,The GNAT Library
23376
@anchor{gnat_rm/the_gnat_library gnat-decode-string-g-decstr-ads}@anchor{303}@anchor{gnat_rm/the_gnat_library id63}@anchor{304}
23377
@section @cite{GNAT.Decode_String} (@code{g-decstr.ads})
23380
@geindex GNAT.Decode_String (g-decstr.ads)
23382
@geindex Decoding strings
23384
@geindex String decoding
23386
@geindex Wide character encoding
23392
A generic package providing routines for decoding wide character and wide wide
23393
character strings encoded as sequences of 8-bit characters using a specified
23394
encoding method. Includes validation routines, and also routines for stepping
23395
to next or previous encoded character in an encoded string.
23396
Useful in conjunction with Unicode character coding. Note there is a
23397
preinstantiation for UTF-8. See next entry.
23399
@node GNAT Decode_UTF8_String g-deutst ads,GNAT Directory_Operations g-dirope ads,GNAT Decode_String g-decstr ads,The GNAT Library
23400
@anchor{gnat_rm/the_gnat_library gnat-decode-utf8-string-g-deutst-ads}@anchor{305}@anchor{gnat_rm/the_gnat_library id64}@anchor{306}
23401
@section @cite{GNAT.Decode_UTF8_String} (@code{g-deutst.ads})
23404
@geindex GNAT.Decode_UTF8_String (g-deutst.ads)
23406
@geindex Decoding strings
23408
@geindex Decoding UTF-8 strings
23410
@geindex UTF-8 string decoding
23412
@geindex Wide character decoding
23418
A preinstantiation of GNAT.Decode_Strings for UTF-8 encoding.
23420
@node GNAT Directory_Operations g-dirope ads,GNAT Directory_Operations Iteration g-diopit ads,GNAT Decode_UTF8_String g-deutst ads,The GNAT Library
23421
@anchor{gnat_rm/the_gnat_library id65}@anchor{307}@anchor{gnat_rm/the_gnat_library gnat-directory-operations-g-dirope-ads}@anchor{308}
23422
@section @cite{GNAT.Directory_Operations} (@code{g-dirope.ads})
23425
@geindex GNAT.Directory_Operations (g-dirope.ads)
23427
@geindex Directory operations
23429
Provides a set of routines for manipulating directories, including changing
23430
the current directory, making new directories, and scanning the files in a
23433
@node GNAT Directory_Operations Iteration g-diopit ads,GNAT Dynamic_HTables g-dynhta ads,GNAT Directory_Operations g-dirope ads,The GNAT Library
23434
@anchor{gnat_rm/the_gnat_library id66}@anchor{309}@anchor{gnat_rm/the_gnat_library gnat-directory-operations-iteration-g-diopit-ads}@anchor{30a}
23435
@section @cite{GNAT.Directory_Operations.Iteration} (@code{g-diopit.ads})
23438
@geindex GNAT.Directory_Operations.Iteration (g-diopit.ads)
23440
@geindex Directory operations iteration
23442
A child unit of GNAT.Directory_Operations providing additional operations
23443
for iterating through directories.
23445
@node GNAT Dynamic_HTables g-dynhta ads,GNAT Dynamic_Tables g-dyntab ads,GNAT Directory_Operations Iteration g-diopit ads,The GNAT Library
23446
@anchor{gnat_rm/the_gnat_library id67}@anchor{30b}@anchor{gnat_rm/the_gnat_library gnat-dynamic-htables-g-dynhta-ads}@anchor{30c}
23447
@section @cite{GNAT.Dynamic_HTables} (@code{g-dynhta.ads})
23450
@geindex GNAT.Dynamic_HTables (g-dynhta.ads)
23452
@geindex Hash tables
23454
A generic implementation of hash tables that can be used to hash arbitrary
23455
data. Provided in two forms, a simple form with built in hash functions,
23456
and a more complex form in which the hash function is supplied.
23458
This package provides a facility similar to that of @cite{GNAT.HTable},
23459
except that this package declares a type that can be used to define
23460
dynamic instances of the hash table, while an instantiation of
23461
@cite{GNAT.HTable} creates a single instance of the hash table.
23463
@node GNAT Dynamic_Tables g-dyntab ads,GNAT Encode_String g-encstr ads,GNAT Dynamic_HTables g-dynhta ads,The GNAT Library
23464
@anchor{gnat_rm/the_gnat_library gnat-dynamic-tables-g-dyntab-ads}@anchor{30d}@anchor{gnat_rm/the_gnat_library id68}@anchor{30e}
23465
@section @cite{GNAT.Dynamic_Tables} (@code{g-dyntab.ads})
23468
@geindex GNAT.Dynamic_Tables (g-dyntab.ads)
23470
@geindex Table implementation
23473
@geindex extendable
23475
A generic package providing a single dimension array abstraction where the
23476
length of the array can be dynamically modified.
23478
This package provides a facility similar to that of @cite{GNAT.Table},
23479
except that this package declares a type that can be used to define
23480
dynamic instances of the table, while an instantiation of
23481
@cite{GNAT.Table} creates a single instance of the table type.
23483
@node GNAT Encode_String g-encstr ads,GNAT Encode_UTF8_String g-enutst ads,GNAT Dynamic_Tables g-dyntab ads,The GNAT Library
23484
@anchor{gnat_rm/the_gnat_library id69}@anchor{30f}@anchor{gnat_rm/the_gnat_library gnat-encode-string-g-encstr-ads}@anchor{310}
23485
@section @cite{GNAT.Encode_String} (@code{g-encstr.ads})
23488
@geindex GNAT.Encode_String (g-encstr.ads)
23490
@geindex Encoding strings
23492
@geindex String encoding
23494
@geindex Wide character encoding
23500
A generic package providing routines for encoding wide character and wide
23501
wide character strings as sequences of 8-bit characters using a specified
23502
encoding method. Useful in conjunction with Unicode character coding.
23503
Note there is a preinstantiation for UTF-8. See next entry.
23505
@node GNAT Encode_UTF8_String g-enutst ads,GNAT Exception_Actions g-excact ads,GNAT Encode_String g-encstr ads,The GNAT Library
23506
@anchor{gnat_rm/the_gnat_library gnat-encode-utf8-string-g-enutst-ads}@anchor{311}@anchor{gnat_rm/the_gnat_library id70}@anchor{312}
23507
@section @cite{GNAT.Encode_UTF8_String} (@code{g-enutst.ads})
23510
@geindex GNAT.Encode_UTF8_String (g-enutst.ads)
23512
@geindex Encoding strings
23514
@geindex Encoding UTF-8 strings
23516
@geindex UTF-8 string encoding
23518
@geindex Wide character encoding
23524
A preinstantiation of GNAT.Encode_Strings for UTF-8 encoding.
23526
@node GNAT Exception_Actions g-excact ads,GNAT Exception_Traces g-exctra ads,GNAT Encode_UTF8_String g-enutst ads,The GNAT Library
23527
@anchor{gnat_rm/the_gnat_library id71}@anchor{313}@anchor{gnat_rm/the_gnat_library gnat-exception-actions-g-excact-ads}@anchor{314}
23528
@section @cite{GNAT.Exception_Actions} (@code{g-excact.ads})
23531
@geindex GNAT.Exception_Actions (g-excact.ads)
23533
@geindex Exception actions
23535
Provides callbacks when an exception is raised. Callbacks can be registered
23536
for specific exceptions, or when any exception is raised. This
23537
can be used for instance to force a core dump to ease debugging.
23539
@node GNAT Exception_Traces g-exctra ads,GNAT Exceptions g-expect ads,GNAT Exception_Actions g-excact ads,The GNAT Library
23540
@anchor{gnat_rm/the_gnat_library id72}@anchor{315}@anchor{gnat_rm/the_gnat_library gnat-exception-traces-g-exctra-ads}@anchor{316}
23541
@section @cite{GNAT.Exception_Traces} (@code{g-exctra.ads})
23544
@geindex GNAT.Exception_Traces (g-exctra.ads)
23546
@geindex Exception traces
23550
Provides an interface allowing to control automatic output upon exception
23553
@node GNAT Exceptions g-expect ads,GNAT Expect g-expect ads,GNAT Exception_Traces g-exctra ads,The GNAT Library
23554
@anchor{gnat_rm/the_gnat_library id73}@anchor{317}@anchor{gnat_rm/the_gnat_library gnat-exceptions-g-expect-ads}@anchor{318}
23555
@section @cite{GNAT.Exceptions} (@code{g-expect.ads})
23558
@geindex GNAT.Exceptions (g-expect.ads)
23560
@geindex Exceptions
23563
@geindex Pure packages
23564
@geindex exceptions
23566
Normally it is not possible to raise an exception with
23567
a message from a subprogram in a pure package, since the
23568
necessary types and subprograms are in @cite{Ada.Exceptions}
23569
which is not a pure unit. @cite{GNAT.Exceptions} provides a
23570
facility for getting around this limitation for a few
23571
predefined exceptions, and for example allow raising
23572
@cite{Constraint_Error} with a message from a pure subprogram.
23574
@node GNAT Expect g-expect ads,GNAT Expect TTY g-exptty ads,GNAT Exceptions g-expect ads,The GNAT Library
23575
@anchor{gnat_rm/the_gnat_library gnat-expect-g-expect-ads}@anchor{319}@anchor{gnat_rm/the_gnat_library id74}@anchor{31a}
23576
@section @cite{GNAT.Expect} (@code{g-expect.ads})
23579
@geindex GNAT.Expect (g-expect.ads)
23581
Provides a set of subprograms similar to what is available
23582
with the standard Tcl Expect tool.
23583
It allows you to easily spawn and communicate with an external process.
23584
You can send commands or inputs to the process, and compare the output
23585
with some expected regular expression. Currently @cite{GNAT.Expect}
23586
is implemented on all native GNAT ports.
23587
It is not implemented for cross ports, and in particular is not
23588
implemented for VxWorks or LynxOS.
23590
@node GNAT Expect TTY g-exptty ads,GNAT Float_Control g-flocon ads,GNAT Expect g-expect ads,The GNAT Library
23591
@anchor{gnat_rm/the_gnat_library gnat-expect-tty-g-exptty-ads}@anchor{31b}@anchor{gnat_rm/the_gnat_library id75}@anchor{31c}
23592
@section @cite{GNAT.Expect.TTY} (@code{g-exptty.ads})
23595
@geindex GNAT.Expect.TTY (g-exptty.ads)
23597
As GNAT.Expect but using pseudo-terminal.
23598
Currently @cite{GNAT.Expect.TTY} is implemented on all native GNAT
23599
ports. It is not implemented for cross ports, and
23600
in particular is not implemented for VxWorks or LynxOS.
23602
@node GNAT Float_Control g-flocon ads,GNAT Formatted_String g-forstr ads,GNAT Expect TTY g-exptty ads,The GNAT Library
23603
@anchor{gnat_rm/the_gnat_library id76}@anchor{31d}@anchor{gnat_rm/the_gnat_library gnat-float-control-g-flocon-ads}@anchor{31e}
23604
@section @cite{GNAT.Float_Control} (@code{g-flocon.ads})
23607
@geindex GNAT.Float_Control (g-flocon.ads)
23609
@geindex Floating-Point Processor
23611
Provides an interface for resetting the floating-point processor into the
23612
mode required for correct semantic operation in Ada. Some third party
23613
library calls may cause this mode to be modified, and the Reset procedure
23614
in this package can be used to reestablish the required mode.
23616
@node GNAT Formatted_String g-forstr ads,GNAT Heap_Sort g-heasor ads,GNAT Float_Control g-flocon ads,The GNAT Library
23617
@anchor{gnat_rm/the_gnat_library gnat-formatted-string-g-forstr-ads}@anchor{31f}@anchor{gnat_rm/the_gnat_library id77}@anchor{320}
23618
@section @cite{GNAT.Formatted_String} (@code{g-forstr.ads})
23621
@geindex GNAT.Formatted_String (g-forstr.ads)
23623
@geindex Formatted String
23625
Provides support for C/C++ printf() formatted strings. The format is
23626
copied from the printf() routine and should therefore gives identical
23627
output. Some generic routines are provided to be able to use types
23628
derived from Integer, Float or enumerations as values for the
23631
@node GNAT Heap_Sort g-heasor ads,GNAT Heap_Sort_A g-hesora ads,GNAT Formatted_String g-forstr ads,The GNAT Library
23632
@anchor{gnat_rm/the_gnat_library gnat-heap-sort-g-heasor-ads}@anchor{321}@anchor{gnat_rm/the_gnat_library id78}@anchor{322}
23633
@section @cite{GNAT.Heap_Sort} (@code{g-heasor.ads})
23636
@geindex GNAT.Heap_Sort (g-heasor.ads)
23640
Provides a general implementation of heap sort usable for sorting arbitrary
23641
data items. Exchange and comparison procedures are provided by passing
23642
access-to-procedure values. The algorithm used is a modified heap sort
23643
that performs approximately N*log(N) comparisons in the worst case.
23645
@node GNAT Heap_Sort_A g-hesora ads,GNAT Heap_Sort_G g-hesorg ads,GNAT Heap_Sort g-heasor ads,The GNAT Library
23646
@anchor{gnat_rm/the_gnat_library id79}@anchor{323}@anchor{gnat_rm/the_gnat_library gnat-heap-sort-a-g-hesora-ads}@anchor{324}
23647
@section @cite{GNAT.Heap_Sort_A} (@code{g-hesora.ads})
23650
@geindex GNAT.Heap_Sort_A (g-hesora.ads)
23654
Provides a general implementation of heap sort usable for sorting arbitrary
23655
data items. Move and comparison procedures are provided by passing
23656
access-to-procedure values. The algorithm used is a modified heap sort
23657
that performs approximately N*log(N) comparisons in the worst case.
23658
This differs from @cite{GNAT.Heap_Sort} in having a less convenient
23659
interface, but may be slightly more efficient.
23661
@node GNAT Heap_Sort_G g-hesorg ads,GNAT HTable g-htable ads,GNAT Heap_Sort_A g-hesora ads,The GNAT Library
23662
@anchor{gnat_rm/the_gnat_library id80}@anchor{325}@anchor{gnat_rm/the_gnat_library gnat-heap-sort-g-g-hesorg-ads}@anchor{326}
23663
@section @cite{GNAT.Heap_Sort_G} (@code{g-hesorg.ads})
23666
@geindex GNAT.Heap_Sort_G (g-hesorg.ads)
23670
Similar to @cite{Heap_Sort_A} except that the move and sorting procedures
23671
are provided as generic parameters, this improves efficiency, especially
23672
if the procedures can be inlined, at the expense of duplicating code for
23673
multiple instantiations.
23675
@node GNAT HTable g-htable ads,GNAT IO g-io ads,GNAT Heap_Sort_G g-hesorg ads,The GNAT Library
23676
@anchor{gnat_rm/the_gnat_library id81}@anchor{327}@anchor{gnat_rm/the_gnat_library gnat-htable-g-htable-ads}@anchor{328}
23677
@section @cite{GNAT.HTable} (@code{g-htable.ads})
23680
@geindex GNAT.HTable (g-htable.ads)
23682
@geindex Hash tables
23684
A generic implementation of hash tables that can be used to hash arbitrary
23685
data. Provides two approaches, one a simple static approach, and the other
23686
allowing arbitrary dynamic hash tables.
23688
@node GNAT IO g-io ads,GNAT IO_Aux g-io_aux ads,GNAT HTable g-htable ads,The GNAT Library
23689
@anchor{gnat_rm/the_gnat_library id82}@anchor{329}@anchor{gnat_rm/the_gnat_library gnat-io-g-io-ads}@anchor{32a}
23690
@section @cite{GNAT.IO} (@code{g-io.ads})
23693
@geindex GNAT.IO (g-io.ads)
23695
@geindex Simple I/O
23697
@geindex Input/Output facilities
23699
A simple preelaborable input-output package that provides a subset of
23700
simple Text_IO functions for reading characters and strings from
23701
Standard_Input, and writing characters, strings and integers to either
23702
Standard_Output or Standard_Error.
23704
@node GNAT IO_Aux g-io_aux ads,GNAT Lock_Files g-locfil ads,GNAT IO g-io ads,The GNAT Library
23705
@anchor{gnat_rm/the_gnat_library id83}@anchor{32b}@anchor{gnat_rm/the_gnat_library gnat-io-aux-g-io-aux-ads}@anchor{32c}
23706
@section @cite{GNAT.IO_Aux} (@code{g-io_aux.ads})
23709
@geindex GNAT.IO_Aux (g-io_aux.ads)
23713
@geindex Input/Output facilities
23715
Provides some auxiliary functions for use with Text_IO, including a test
23716
for whether a file exists, and functions for reading a line of text.
23718
@node GNAT Lock_Files g-locfil ads,GNAT MBBS_Discrete_Random g-mbdira ads,GNAT IO_Aux g-io_aux ads,The GNAT Library
23719
@anchor{gnat_rm/the_gnat_library gnat-lock-files-g-locfil-ads}@anchor{32d}@anchor{gnat_rm/the_gnat_library id84}@anchor{32e}
23720
@section @cite{GNAT.Lock_Files} (@code{g-locfil.ads})
23723
@geindex GNAT.Lock_Files (g-locfil.ads)
23725
@geindex File locking
23727
@geindex Locking using files
23729
Provides a general interface for using files as locks. Can be used for
23730
providing program level synchronization.
23732
@node GNAT MBBS_Discrete_Random g-mbdira ads,GNAT MBBS_Float_Random g-mbflra ads,GNAT Lock_Files g-locfil ads,The GNAT Library
23733
@anchor{gnat_rm/the_gnat_library id85}@anchor{32f}@anchor{gnat_rm/the_gnat_library gnat-mbbs-discrete-random-g-mbdira-ads}@anchor{330}
23734
@section @cite{GNAT.MBBS_Discrete_Random} (@code{g-mbdira.ads})
23737
@geindex GNAT.MBBS_Discrete_Random (g-mbdira.ads)
23739
@geindex Random number generation
23741
The original implementation of @cite{Ada.Numerics.Discrete_Random}. Uses
23742
a modified version of the Blum-Blum-Shub generator.
23744
@node GNAT MBBS_Float_Random g-mbflra ads,GNAT MD5 g-md5 ads,GNAT MBBS_Discrete_Random g-mbdira ads,The GNAT Library
23745
@anchor{gnat_rm/the_gnat_library id86}@anchor{331}@anchor{gnat_rm/the_gnat_library gnat-mbbs-float-random-g-mbflra-ads}@anchor{332}
23746
@section @cite{GNAT.MBBS_Float_Random} (@code{g-mbflra.ads})
23749
@geindex GNAT.MBBS_Float_Random (g-mbflra.ads)
23751
@geindex Random number generation
23753
The original implementation of @cite{Ada.Numerics.Float_Random}. Uses
23754
a modified version of the Blum-Blum-Shub generator.
23756
@node GNAT MD5 g-md5 ads,GNAT Memory_Dump g-memdum ads,GNAT MBBS_Float_Random g-mbflra ads,The GNAT Library
23757
@anchor{gnat_rm/the_gnat_library id87}@anchor{333}@anchor{gnat_rm/the_gnat_library gnat-md5-g-md5-ads}@anchor{334}
23758
@section @cite{GNAT.MD5} (@code{g-md5.ads})
23761
@geindex GNAT.MD5 (g-md5.ads)
23763
@geindex Message Digest MD5
23765
Implements the MD5 Message-Digest Algorithm as described in RFC 1321, and
23766
the HMAC-MD5 message authentication function as described in RFC 2104 and
23769
@node GNAT Memory_Dump g-memdum ads,GNAT Most_Recent_Exception g-moreex ads,GNAT MD5 g-md5 ads,The GNAT Library
23770
@anchor{gnat_rm/the_gnat_library id88}@anchor{335}@anchor{gnat_rm/the_gnat_library gnat-memory-dump-g-memdum-ads}@anchor{336}
23771
@section @cite{GNAT.Memory_Dump} (@code{g-memdum.ads})
23774
@geindex GNAT.Memory_Dump (g-memdum.ads)
23776
@geindex Dump Memory
23778
Provides a convenient routine for dumping raw memory to either the
23779
standard output or standard error files. Uses GNAT.IO for actual
23782
@node GNAT Most_Recent_Exception g-moreex ads,GNAT OS_Lib g-os_lib ads,GNAT Memory_Dump g-memdum ads,The GNAT Library
23783
@anchor{gnat_rm/the_gnat_library id89}@anchor{337}@anchor{gnat_rm/the_gnat_library gnat-most-recent-exception-g-moreex-ads}@anchor{338}
23784
@section @cite{GNAT.Most_Recent_Exception} (@code{g-moreex.ads})
23787
@geindex GNAT.Most_Recent_Exception (g-moreex.ads)
23790
@geindex obtaining most recent
23792
Provides access to the most recently raised exception. Can be used for
23793
various logging purposes, including duplicating functionality of some
23794
Ada 83 implementation dependent extensions.
23796
@node GNAT OS_Lib g-os_lib ads,GNAT Perfect_Hash_Generators g-pehage ads,GNAT Most_Recent_Exception g-moreex ads,The GNAT Library
23797
@anchor{gnat_rm/the_gnat_library id90}@anchor{339}@anchor{gnat_rm/the_gnat_library gnat-os-lib-g-os-lib-ads}@anchor{33a}
23798
@section @cite{GNAT.OS_Lib} (@code{g-os_lib.ads})
23801
@geindex GNAT.OS_Lib (g-os_lib.ads)
23803
@geindex Operating System interface
23805
@geindex Spawn capability
23807
Provides a range of target independent operating system interface functions,
23808
including time/date management, file operations, subprocess management,
23809
including a portable spawn procedure, and access to environment variables
23810
and error return codes.
23812
@node GNAT Perfect_Hash_Generators g-pehage ads,GNAT Random_Numbers g-rannum ads,GNAT OS_Lib g-os_lib ads,The GNAT Library
23813
@anchor{gnat_rm/the_gnat_library gnat-perfect-hash-generators-g-pehage-ads}@anchor{33b}@anchor{gnat_rm/the_gnat_library id91}@anchor{33c}
23814
@section @cite{GNAT.Perfect_Hash_Generators} (@code{g-pehage.ads})
23817
@geindex GNAT.Perfect_Hash_Generators (g-pehage.ads)
23819
@geindex Hash functions
23821
Provides a generator of static minimal perfect hash functions. No
23822
collisions occur and each item can be retrieved from the table in one
23823
probe (perfect property). The hash table size corresponds to the exact
23824
size of the key set and no larger (minimal property). The key set has to
23825
be know in advance (static property). The hash functions are also order
23826
preserving. If w2 is inserted after w1 in the generator, their
23827
hashcode are in the same order. These hashing functions are very
23828
convenient for use with realtime applications.
23830
@node GNAT Random_Numbers g-rannum ads,GNAT Regexp g-regexp ads,GNAT Perfect_Hash_Generators g-pehage ads,The GNAT Library
23831
@anchor{gnat_rm/the_gnat_library gnat-random-numbers-g-rannum-ads}@anchor{33d}@anchor{gnat_rm/the_gnat_library id92}@anchor{33e}
23832
@section @cite{GNAT.Random_Numbers} (@code{g-rannum.ads})
23835
@geindex GNAT.Random_Numbers (g-rannum.ads)
23837
@geindex Random number generation
23839
Provides random number capabilities which extend those available in the
23840
standard Ada library and are more convenient to use.
23842
@node GNAT Regexp g-regexp ads,GNAT Registry g-regist ads,GNAT Random_Numbers g-rannum ads,The GNAT Library
23843
@anchor{gnat_rm/the_gnat_library gnat-regexp-g-regexp-ads}@anchor{210}@anchor{gnat_rm/the_gnat_library id93}@anchor{33f}
23844
@section @cite{GNAT.Regexp} (@code{g-regexp.ads})
23847
@geindex GNAT.Regexp (g-regexp.ads)
23849
@geindex Regular expressions
23851
@geindex Pattern matching
23853
A simple implementation of regular expressions, using a subset of regular
23854
expression syntax copied from familiar Unix style utilities. This is the
23855
simplest of the three pattern matching packages provided, and is particularly
23856
suitable for 'file globbing' applications.
23858
@node GNAT Registry g-regist ads,GNAT Regpat g-regpat ads,GNAT Regexp g-regexp ads,The GNAT Library
23859
@anchor{gnat_rm/the_gnat_library id94}@anchor{340}@anchor{gnat_rm/the_gnat_library gnat-registry-g-regist-ads}@anchor{341}
23860
@section @cite{GNAT.Registry} (@code{g-regist.ads})
23863
@geindex GNAT.Registry (g-regist.ads)
23865
@geindex Windows Registry
23867
This is a high level binding to the Windows registry. It is possible to
23868
do simple things like reading a key value, creating a new key. For full
23869
registry API, but at a lower level of abstraction, refer to the Win32.Winreg
23870
package provided with the Win32Ada binding
23872
@node GNAT Regpat g-regpat ads,GNAT Rewrite_Data g-rewdat ads,GNAT Registry g-regist ads,The GNAT Library
23873
@anchor{gnat_rm/the_gnat_library gnat-regpat-g-regpat-ads}@anchor{342}@anchor{gnat_rm/the_gnat_library id95}@anchor{343}
23874
@section @cite{GNAT.Regpat} (@code{g-regpat.ads})
23877
@geindex GNAT.Regpat (g-regpat.ads)
23879
@geindex Regular expressions
23881
@geindex Pattern matching
23883
A complete implementation of Unix-style regular expression matching, copied
23884
from the original V7 style regular expression library written in C by
23885
Henry Spencer (and binary compatible with this C library).
23887
@node GNAT Rewrite_Data g-rewdat ads,GNAT Secondary_Stack_Info g-sestin ads,GNAT Regpat g-regpat ads,The GNAT Library
23888
@anchor{gnat_rm/the_gnat_library id96}@anchor{344}@anchor{gnat_rm/the_gnat_library gnat-rewrite-data-g-rewdat-ads}@anchor{345}
23889
@section @cite{GNAT.Rewrite_Data} (@code{g-rewdat.ads})
23892
@geindex GNAT.Rewrite_Data (g-rewdat.ads)
23894
@geindex Rewrite data
23896
A unit to rewrite on-the-fly string occurrences in a stream of
23897
data. The implementation has a very minimal memory footprint as the
23898
full content to be processed is not loaded into memory all at once. This makes
23899
this interface usable for large files or socket streams.
23901
@node GNAT Secondary_Stack_Info g-sestin ads,GNAT Semaphores g-semaph ads,GNAT Rewrite_Data g-rewdat ads,The GNAT Library
23902
@anchor{gnat_rm/the_gnat_library gnat-secondary-stack-info-g-sestin-ads}@anchor{346}@anchor{gnat_rm/the_gnat_library id97}@anchor{347}
23903
@section @cite{GNAT.Secondary_Stack_Info} (@code{g-sestin.ads})
23906
@geindex GNAT.Secondary_Stack_Info (g-sestin.ads)
23908
@geindex Secondary Stack Info
23910
Provide the capability to query the high water mark of the current task's
23913
@node GNAT Semaphores g-semaph ads,GNAT Serial_Communications g-sercom ads,GNAT Secondary_Stack_Info g-sestin ads,The GNAT Library
23914
@anchor{gnat_rm/the_gnat_library id98}@anchor{348}@anchor{gnat_rm/the_gnat_library gnat-semaphores-g-semaph-ads}@anchor{349}
23915
@section @cite{GNAT.Semaphores} (@code{g-semaph.ads})
23918
@geindex GNAT.Semaphores (g-semaph.ads)
23920
@geindex Semaphores
23922
Provides classic counting and binary semaphores using protected types.
23924
@node GNAT Serial_Communications g-sercom ads,GNAT SHA1 g-sha1 ads,GNAT Semaphores g-semaph ads,The GNAT Library
23925
@anchor{gnat_rm/the_gnat_library gnat-serial-communications-g-sercom-ads}@anchor{34a}@anchor{gnat_rm/the_gnat_library id99}@anchor{34b}
23926
@section @cite{GNAT.Serial_Communications} (@code{g-sercom.ads})
23929
@geindex GNAT.Serial_Communications (g-sercom.ads)
23931
@geindex Serial_Communications
23933
Provides a simple interface to send and receive data over a serial
23934
port. This is only supported on GNU/Linux and Windows.
23936
@node GNAT SHA1 g-sha1 ads,GNAT SHA224 g-sha224 ads,GNAT Serial_Communications g-sercom ads,The GNAT Library
23937
@anchor{gnat_rm/the_gnat_library gnat-sha1-g-sha1-ads}@anchor{34c}@anchor{gnat_rm/the_gnat_library id100}@anchor{34d}
23938
@section @cite{GNAT.SHA1} (@code{g-sha1.ads})
23941
@geindex GNAT.SHA1 (g-sha1.ads)
23943
@geindex Secure Hash Algorithm SHA-1
23945
Implements the SHA-1 Secure Hash Algorithm as described in FIPS PUB 180-3
23946
and RFC 3174, and the HMAC-SHA1 message authentication function as described
23947
in RFC 2104 and FIPS PUB 198.
23949
@node GNAT SHA224 g-sha224 ads,GNAT SHA256 g-sha256 ads,GNAT SHA1 g-sha1 ads,The GNAT Library
23950
@anchor{gnat_rm/the_gnat_library id101}@anchor{34e}@anchor{gnat_rm/the_gnat_library gnat-sha224-g-sha224-ads}@anchor{34f}
23951
@section @cite{GNAT.SHA224} (@code{g-sha224.ads})
23954
@geindex GNAT.SHA224 (g-sha224.ads)
23956
@geindex Secure Hash Algorithm SHA-224
23958
Implements the SHA-224 Secure Hash Algorithm as described in FIPS PUB 180-3,
23959
and the HMAC-SHA224 message authentication function as described
23960
in RFC 2104 and FIPS PUB 198.
23962
@node GNAT SHA256 g-sha256 ads,GNAT SHA384 g-sha384 ads,GNAT SHA224 g-sha224 ads,The GNAT Library
23963
@anchor{gnat_rm/the_gnat_library id102}@anchor{350}@anchor{gnat_rm/the_gnat_library gnat-sha256-g-sha256-ads}@anchor{351}
23964
@section @cite{GNAT.SHA256} (@code{g-sha256.ads})
23967
@geindex GNAT.SHA256 (g-sha256.ads)
23969
@geindex Secure Hash Algorithm SHA-256
23971
Implements the SHA-256 Secure Hash Algorithm as described in FIPS PUB 180-3,
23972
and the HMAC-SHA256 message authentication function as described
23973
in RFC 2104 and FIPS PUB 198.
23975
@node GNAT SHA384 g-sha384 ads,GNAT SHA512 g-sha512 ads,GNAT SHA256 g-sha256 ads,The GNAT Library
23976
@anchor{gnat_rm/the_gnat_library id103}@anchor{352}@anchor{gnat_rm/the_gnat_library gnat-sha384-g-sha384-ads}@anchor{353}
23977
@section @cite{GNAT.SHA384} (@code{g-sha384.ads})
23980
@geindex GNAT.SHA384 (g-sha384.ads)
23982
@geindex Secure Hash Algorithm SHA-384
23984
Implements the SHA-384 Secure Hash Algorithm as described in FIPS PUB 180-3,
23985
and the HMAC-SHA384 message authentication function as described
23986
in RFC 2104 and FIPS PUB 198.
23988
@node GNAT SHA512 g-sha512 ads,GNAT Signals g-signal ads,GNAT SHA384 g-sha384 ads,The GNAT Library
23989
@anchor{gnat_rm/the_gnat_library gnat-sha512-g-sha512-ads}@anchor{354}@anchor{gnat_rm/the_gnat_library id104}@anchor{355}
23990
@section @cite{GNAT.SHA512} (@code{g-sha512.ads})
23993
@geindex GNAT.SHA512 (g-sha512.ads)
23995
@geindex Secure Hash Algorithm SHA-512
23997
Implements the SHA-512 Secure Hash Algorithm as described in FIPS PUB 180-3,
23998
and the HMAC-SHA512 message authentication function as described
23999
in RFC 2104 and FIPS PUB 198.
24001
@node GNAT Signals g-signal ads,GNAT Sockets g-socket ads,GNAT SHA512 g-sha512 ads,The GNAT Library
24002
@anchor{gnat_rm/the_gnat_library gnat-signals-g-signal-ads}@anchor{356}@anchor{gnat_rm/the_gnat_library id105}@anchor{357}
24003
@section @cite{GNAT.Signals} (@code{g-signal.ads})
24006
@geindex GNAT.Signals (g-signal.ads)
24010
Provides the ability to manipulate the blocked status of signals on supported
24013
@node GNAT Sockets g-socket ads,GNAT Source_Info g-souinf ads,GNAT Signals g-signal ads,The GNAT Library
24014
@anchor{gnat_rm/the_gnat_library gnat-sockets-g-socket-ads}@anchor{358}@anchor{gnat_rm/the_gnat_library id106}@anchor{359}
24015
@section @cite{GNAT.Sockets} (@code{g-socket.ads})
24018
@geindex GNAT.Sockets (g-socket.ads)
24022
A high level and portable interface to develop sockets based applications.
24023
This package is based on the sockets thin binding found in
24024
@cite{GNAT.Sockets.Thin}. Currently @cite{GNAT.Sockets} is implemented
24025
on all native GNAT ports and on VxWorks cross prots. It is not implemented for
24026
the LynxOS cross port.
24028
@node GNAT Source_Info g-souinf ads,GNAT Spelling_Checker g-speche ads,GNAT Sockets g-socket ads,The GNAT Library
24029
@anchor{gnat_rm/the_gnat_library gnat-source-info-g-souinf-ads}@anchor{35a}@anchor{gnat_rm/the_gnat_library id107}@anchor{35b}
24030
@section @cite{GNAT.Source_Info} (@code{g-souinf.ads})
24033
@geindex GNAT.Source_Info (g-souinf.ads)
24035
@geindex Source Information
24037
Provides subprograms that give access to source code information known at
24038
compile time, such as the current file name and line number. Also provides
24039
subprograms yielding the date and time of the current compilation (like the
24040
C macros @cite{__DATE__} and @cite{__TIME__})
24042
@node GNAT Spelling_Checker g-speche ads,GNAT Spelling_Checker_Generic g-spchge ads,GNAT Source_Info g-souinf ads,The GNAT Library
24043
@anchor{gnat_rm/the_gnat_library gnat-spelling-checker-g-speche-ads}@anchor{35c}@anchor{gnat_rm/the_gnat_library id108}@anchor{35d}
24044
@section @cite{GNAT.Spelling_Checker} (@code{g-speche.ads})
24047
@geindex GNAT.Spelling_Checker (g-speche.ads)
24049
@geindex Spell checking
24051
Provides a function for determining whether one string is a plausible
24052
near misspelling of another string.
24054
@node GNAT Spelling_Checker_Generic g-spchge ads,GNAT Spitbol Patterns g-spipat ads,GNAT Spelling_Checker g-speche ads,The GNAT Library
24055
@anchor{gnat_rm/the_gnat_library id109}@anchor{35e}@anchor{gnat_rm/the_gnat_library gnat-spelling-checker-generic-g-spchge-ads}@anchor{35f}
24056
@section @cite{GNAT.Spelling_Checker_Generic} (@code{g-spchge.ads})
24059
@geindex GNAT.Spelling_Checker_Generic (g-spchge.ads)
24061
@geindex Spell checking
24063
Provides a generic function that can be instantiated with a string type for
24064
determining whether one string is a plausible near misspelling of another
24067
@node GNAT Spitbol Patterns g-spipat ads,GNAT Spitbol g-spitbo ads,GNAT Spelling_Checker_Generic g-spchge ads,The GNAT Library
24068
@anchor{gnat_rm/the_gnat_library id110}@anchor{360}@anchor{gnat_rm/the_gnat_library gnat-spitbol-patterns-g-spipat-ads}@anchor{361}
24069
@section @cite{GNAT.Spitbol.Patterns} (@code{g-spipat.ads})
24072
@geindex GNAT.Spitbol.Patterns (g-spipat.ads)
24074
@geindex SPITBOL pattern matching
24076
@geindex Pattern matching
24078
A complete implementation of SNOBOL4 style pattern matching. This is the
24079
most elaborate of the pattern matching packages provided. It fully duplicates
24080
the SNOBOL4 dynamic pattern construction and matching capabilities, using the
24081
efficient algorithm developed by Robert Dewar for the SPITBOL system.
24083
@node GNAT Spitbol g-spitbo ads,GNAT Spitbol Table_Boolean g-sptabo ads,GNAT Spitbol Patterns g-spipat ads,The GNAT Library
24084
@anchor{gnat_rm/the_gnat_library id111}@anchor{362}@anchor{gnat_rm/the_gnat_library gnat-spitbol-g-spitbo-ads}@anchor{363}
24085
@section @cite{GNAT.Spitbol} (@code{g-spitbo.ads})
24088
@geindex GNAT.Spitbol (g-spitbo.ads)
24090
@geindex SPITBOL interface
24092
The top level package of the collection of SPITBOL-style functionality, this
24093
package provides basic SNOBOL4 string manipulation functions, such as
24094
Pad, Reverse, Trim, Substr capability, as well as a generic table function
24095
useful for constructing arbitrary mappings from strings in the style of
24096
the SNOBOL4 TABLE function.
24098
@node GNAT Spitbol Table_Boolean g-sptabo ads,GNAT Spitbol Table_Integer g-sptain ads,GNAT Spitbol g-spitbo ads,The GNAT Library
24099
@anchor{gnat_rm/the_gnat_library id112}@anchor{364}@anchor{gnat_rm/the_gnat_library gnat-spitbol-table-boolean-g-sptabo-ads}@anchor{365}
24100
@section @cite{GNAT.Spitbol.Table_Boolean} (@code{g-sptabo.ads})
24103
@geindex GNAT.Spitbol.Table_Boolean (g-sptabo.ads)
24105
@geindex Sets of strings
24107
@geindex SPITBOL Tables
24109
A library level of instantiation of @cite{GNAT.Spitbol.Patterns.Table}
24110
for type @cite{Standard.Boolean}, giving an implementation of sets of
24113
@node GNAT Spitbol Table_Integer g-sptain ads,GNAT Spitbol Table_VString g-sptavs ads,GNAT Spitbol Table_Boolean g-sptabo ads,The GNAT Library
24114
@anchor{gnat_rm/the_gnat_library id113}@anchor{366}@anchor{gnat_rm/the_gnat_library gnat-spitbol-table-integer-g-sptain-ads}@anchor{367}
24115
@section @cite{GNAT.Spitbol.Table_Integer} (@code{g-sptain.ads})
24118
@geindex GNAT.Spitbol.Table_Integer (g-sptain.ads)
24120
@geindex Integer maps
24124
@geindex SPITBOL Tables
24126
A library level of instantiation of @cite{GNAT.Spitbol.Patterns.Table}
24127
for type @cite{Standard.Integer}, giving an implementation of maps
24128
from string to integer values.
24130
@node GNAT Spitbol Table_VString g-sptavs ads,GNAT SSE g-sse ads,GNAT Spitbol Table_Integer g-sptain ads,The GNAT Library
24131
@anchor{gnat_rm/the_gnat_library id114}@anchor{368}@anchor{gnat_rm/the_gnat_library gnat-spitbol-table-vstring-g-sptavs-ads}@anchor{369}
24132
@section @cite{GNAT.Spitbol.Table_VString} (@code{g-sptavs.ads})
24135
@geindex GNAT.Spitbol.Table_VString (g-sptavs.ads)
24137
@geindex String maps
24141
@geindex SPITBOL Tables
24143
A library level of instantiation of @cite{GNAT.Spitbol.Patterns.Table} for
24144
a variable length string type, giving an implementation of general
24145
maps from strings to strings.
24147
@node GNAT SSE g-sse ads,GNAT SSE Vector_Types g-ssvety ads,GNAT Spitbol Table_VString g-sptavs ads,The GNAT Library
24148
@anchor{gnat_rm/the_gnat_library id115}@anchor{36a}@anchor{gnat_rm/the_gnat_library gnat-sse-g-sse-ads}@anchor{36b}
24149
@section @cite{GNAT.SSE} (@code{g-sse.ads})
24152
@geindex GNAT.SSE (g-sse.ads)
24154
Root of a set of units aimed at offering Ada bindings to a subset of
24155
the Intel(r) Streaming SIMD Extensions with GNAT on the x86 family of
24156
targets. It exposes vector component types together with a general
24157
introduction to the binding contents and use.
24159
@node GNAT SSE Vector_Types g-ssvety ads,GNAT Strings g-string ads,GNAT SSE g-sse ads,The GNAT Library
24160
@anchor{gnat_rm/the_gnat_library gnat-sse-vector-types-g-ssvety-ads}@anchor{36c}@anchor{gnat_rm/the_gnat_library id116}@anchor{36d}
24161
@section @cite{GNAT.SSE.Vector_Types} (@code{g-ssvety.ads})
24164
@geindex GNAT.SSE.Vector_Types (g-ssvety.ads)
24166
SSE vector types for use with SSE related intrinsics.
24168
@node GNAT Strings g-string ads,GNAT String_Split g-strspl ads,GNAT SSE Vector_Types g-ssvety ads,The GNAT Library
24169
@anchor{gnat_rm/the_gnat_library gnat-strings-g-string-ads}@anchor{36e}@anchor{gnat_rm/the_gnat_library id117}@anchor{36f}
24170
@section @cite{GNAT.Strings} (@code{g-string.ads})
24173
@geindex GNAT.Strings (g-string.ads)
24175
Common String access types and related subprograms. Basically it
24176
defines a string access and an array of string access types.
24178
@node GNAT String_Split g-strspl ads,GNAT Table g-table ads,GNAT Strings g-string ads,The GNAT Library
24179
@anchor{gnat_rm/the_gnat_library gnat-string-split-g-strspl-ads}@anchor{370}@anchor{gnat_rm/the_gnat_library id118}@anchor{371}
24180
@section @cite{GNAT.String_Split} (@code{g-strspl.ads})
24183
@geindex GNAT.String_Split (g-strspl.ads)
24185
@geindex String splitter
24187
Useful string manipulation routines: given a set of separators, split
24188
a string wherever the separators appear, and provide direct access
24189
to the resulting slices. This package is instantiated from
24190
@cite{GNAT.Array_Split}.
24192
@node GNAT Table g-table ads,GNAT Task_Lock g-tasloc ads,GNAT String_Split g-strspl ads,The GNAT Library
24193
@anchor{gnat_rm/the_gnat_library gnat-table-g-table-ads}@anchor{372}@anchor{gnat_rm/the_gnat_library id119}@anchor{373}
24194
@section @cite{GNAT.Table} (@code{g-table.ads})
24197
@geindex GNAT.Table (g-table.ads)
24199
@geindex Table implementation
24202
@geindex extendable
24204
A generic package providing a single dimension array abstraction where the
24205
length of the array can be dynamically modified.
24207
This package provides a facility similar to that of @cite{GNAT.Dynamic_Tables},
24208
except that this package declares a single instance of the table type,
24209
while an instantiation of @cite{GNAT.Dynamic_Tables} creates a type that can be
24210
used to define dynamic instances of the table.
24212
@node GNAT Task_Lock g-tasloc ads,GNAT Time_Stamp g-timsta ads,GNAT Table g-table ads,The GNAT Library
24213
@anchor{gnat_rm/the_gnat_library gnat-task-lock-g-tasloc-ads}@anchor{374}@anchor{gnat_rm/the_gnat_library id120}@anchor{375}
24214
@section @cite{GNAT.Task_Lock} (@code{g-tasloc.ads})
24217
@geindex GNAT.Task_Lock (g-tasloc.ads)
24219
@geindex Task synchronization
24221
@geindex Task locking
24225
A very simple facility for locking and unlocking sections of code using a
24226
single global task lock. Appropriate for use in situations where contention
24227
between tasks is very rarely expected.
24229
@node GNAT Time_Stamp g-timsta ads,GNAT Threads g-thread ads,GNAT Task_Lock g-tasloc ads,The GNAT Library
24230
@anchor{gnat_rm/the_gnat_library gnat-time-stamp-g-timsta-ads}@anchor{376}@anchor{gnat_rm/the_gnat_library id121}@anchor{377}
24231
@section @cite{GNAT.Time_Stamp} (@code{g-timsta.ads})
24234
@geindex GNAT.Time_Stamp (g-timsta.ads)
24236
@geindex Time stamp
24238
@geindex Current time
24240
Provides a simple function that returns a string YYYY-MM-DD HH:MM:SS.SS that
24241
represents the current date and time in ISO 8601 format. This is a very simple
24242
routine with minimal code and there are no dependencies on any other unit.
24244
@node GNAT Threads g-thread ads,GNAT Traceback g-traceb ads,GNAT Time_Stamp g-timsta ads,The GNAT Library
24245
@anchor{gnat_rm/the_gnat_library gnat-threads-g-thread-ads}@anchor{378}@anchor{gnat_rm/the_gnat_library id122}@anchor{379}
24246
@section @cite{GNAT.Threads} (@code{g-thread.ads})
24249
@geindex GNAT.Threads (g-thread.ads)
24251
@geindex Foreign threads
24256
Provides facilities for dealing with foreign threads which need to be known
24257
by the GNAT run-time system. Consult the documentation of this package for
24258
further details if your program has threads that are created by a non-Ada
24259
environment which then accesses Ada code.
24261
@node GNAT Traceback g-traceb ads,GNAT Traceback Symbolic g-trasym ads,GNAT Threads g-thread ads,The GNAT Library
24262
@anchor{gnat_rm/the_gnat_library id123}@anchor{37a}@anchor{gnat_rm/the_gnat_library gnat-traceback-g-traceb-ads}@anchor{37b}
24263
@section @cite{GNAT.Traceback} (@code{g-traceb.ads})
24266
@geindex GNAT.Traceback (g-traceb.ads)
24268
@geindex Trace back facilities
24270
Provides a facility for obtaining non-symbolic traceback information, useful
24271
in various debugging situations.
24273
@node GNAT Traceback Symbolic g-trasym ads,GNAT UTF_32 g-table ads,GNAT Traceback g-traceb ads,The GNAT Library
24274
@anchor{gnat_rm/the_gnat_library gnat-traceback-symbolic-g-trasym-ads}@anchor{37c}@anchor{gnat_rm/the_gnat_library id124}@anchor{37d}
24275
@section @cite{GNAT.Traceback.Symbolic} (@code{g-trasym.ads})
24278
@geindex GNAT.Traceback.Symbolic (g-trasym.ads)
24280
@geindex Trace back facilities
24282
@node GNAT UTF_32 g-table ads,GNAT Wide_Spelling_Checker g-u3spch ads,GNAT Traceback Symbolic g-trasym ads,The GNAT Library
24283
@anchor{gnat_rm/the_gnat_library id125}@anchor{37e}@anchor{gnat_rm/the_gnat_library gnat-utf-32-g-table-ads}@anchor{37f}
24284
@section @cite{GNAT.UTF_32} (@code{g-table.ads})
24287
@geindex GNAT.UTF_32 (g-table.ads)
24289
@geindex Wide character codes
24291
This is a package intended to be used in conjunction with the
24292
@cite{Wide_Character} type in Ada 95 and the
24293
@cite{Wide_Wide_Character} type in Ada 2005 (available
24294
in @cite{GNAT} in Ada 2005 mode). This package contains
24295
Unicode categorization routines, as well as lexical
24296
categorization routines corresponding to the Ada 2005
24297
lexical rules for identifiers and strings, and also a
24298
lower case to upper case fold routine corresponding to
24299
the Ada 2005 rules for identifier equivalence.
24301
@node GNAT Wide_Spelling_Checker g-u3spch ads,GNAT Wide_Spelling_Checker g-wispch ads,GNAT UTF_32 g-table ads,The GNAT Library
24302
@anchor{gnat_rm/the_gnat_library gnat-wide-spelling-checker-g-u3spch-ads}@anchor{380}@anchor{gnat_rm/the_gnat_library id126}@anchor{381}
24303
@section @cite{GNAT.Wide_Spelling_Checker} (@code{g-u3spch.ads})
24306
@geindex GNAT.Wide_Spelling_Checker (g-u3spch.ads)
24308
@geindex Spell checking
24310
Provides a function for determining whether one wide wide string is a plausible
24311
near misspelling of another wide wide string, where the strings are represented
24312
using the UTF_32_String type defined in System.Wch_Cnv.
24314
@node GNAT Wide_Spelling_Checker g-wispch ads,GNAT Wide_String_Split g-wistsp ads,GNAT Wide_Spelling_Checker g-u3spch ads,The GNAT Library
24315
@anchor{gnat_rm/the_gnat_library gnat-wide-spelling-checker-g-wispch-ads}@anchor{382}@anchor{gnat_rm/the_gnat_library id127}@anchor{383}
24316
@section @cite{GNAT.Wide_Spelling_Checker} (@code{g-wispch.ads})
24319
@geindex GNAT.Wide_Spelling_Checker (g-wispch.ads)
24321
@geindex Spell checking
24323
Provides a function for determining whether one wide string is a plausible
24324
near misspelling of another wide string.
24326
@node GNAT Wide_String_Split g-wistsp ads,GNAT Wide_Wide_Spelling_Checker g-zspche ads,GNAT Wide_Spelling_Checker g-wispch ads,The GNAT Library
24327
@anchor{gnat_rm/the_gnat_library gnat-wide-string-split-g-wistsp-ads}@anchor{384}@anchor{gnat_rm/the_gnat_library id128}@anchor{385}
24328
@section @cite{GNAT.Wide_String_Split} (@code{g-wistsp.ads})
24331
@geindex GNAT.Wide_String_Split (g-wistsp.ads)
24333
@geindex Wide_String splitter
24335
Useful wide string manipulation routines: given a set of separators, split
24336
a wide string wherever the separators appear, and provide direct access
24337
to the resulting slices. This package is instantiated from
24338
@cite{GNAT.Array_Split}.
24340
@node GNAT Wide_Wide_Spelling_Checker g-zspche ads,GNAT Wide_Wide_String_Split g-zistsp ads,GNAT Wide_String_Split g-wistsp ads,The GNAT Library
24341
@anchor{gnat_rm/the_gnat_library gnat-wide-wide-spelling-checker-g-zspche-ads}@anchor{386}@anchor{gnat_rm/the_gnat_library id129}@anchor{387}
24342
@section @cite{GNAT.Wide_Wide_Spelling_Checker} (@code{g-zspche.ads})
24345
@geindex GNAT.Wide_Wide_Spelling_Checker (g-zspche.ads)
24347
@geindex Spell checking
24349
Provides a function for determining whether one wide wide string is a plausible
24350
near misspelling of another wide wide string.
24352
@node GNAT Wide_Wide_String_Split g-zistsp ads,Interfaces C Extensions i-cexten ads,GNAT Wide_Wide_Spelling_Checker g-zspche ads,The GNAT Library
24353
@anchor{gnat_rm/the_gnat_library gnat-wide-wide-string-split-g-zistsp-ads}@anchor{388}@anchor{gnat_rm/the_gnat_library id130}@anchor{389}
24354
@section @cite{GNAT.Wide_Wide_String_Split} (@code{g-zistsp.ads})
24357
@geindex GNAT.Wide_Wide_String_Split (g-zistsp.ads)
24359
@geindex Wide_Wide_String splitter
24361
Useful wide wide string manipulation routines: given a set of separators, split
24362
a wide wide string wherever the separators appear, and provide direct access
24363
to the resulting slices. This package is instantiated from
24364
@cite{GNAT.Array_Split}.
24366
@node Interfaces C Extensions i-cexten ads,Interfaces C Streams i-cstrea ads,GNAT Wide_Wide_String_Split g-zistsp ads,The GNAT Library
24367
@anchor{gnat_rm/the_gnat_library id131}@anchor{38a}@anchor{gnat_rm/the_gnat_library interfaces-c-extensions-i-cexten-ads}@anchor{38b}
24368
@section @cite{Interfaces.C.Extensions} (@code{i-cexten.ads})
24371
@geindex Interfaces.C.Extensions (i-cexten.ads)
24373
This package contains additional C-related definitions, intended
24374
for use with either manually or automatically generated bindings
24377
@node Interfaces C Streams i-cstrea ads,Interfaces Packed_Decimal i-pacdec ads,Interfaces C Extensions i-cexten ads,The GNAT Library
24378
@anchor{gnat_rm/the_gnat_library id132}@anchor{38c}@anchor{gnat_rm/the_gnat_library interfaces-c-streams-i-cstrea-ads}@anchor{38d}
24379
@section @cite{Interfaces.C.Streams} (@code{i-cstrea.ads})
24382
@geindex Interfaces.C.Streams (i-cstrea.ads)
24385
@geindex interfacing
24387
This package is a binding for the most commonly used operations
24390
@node Interfaces Packed_Decimal i-pacdec ads,Interfaces VxWorks i-vxwork ads,Interfaces C Streams i-cstrea ads,The GNAT Library
24391
@anchor{gnat_rm/the_gnat_library interfaces-packed-decimal-i-pacdec-ads}@anchor{38e}@anchor{gnat_rm/the_gnat_library id133}@anchor{38f}
24392
@section @cite{Interfaces.Packed_Decimal} (@code{i-pacdec.ads})
24395
@geindex Interfaces.Packed_Decimal (i-pacdec.ads)
24397
@geindex IBM Packed Format
24399
@geindex Packed Decimal
24401
This package provides a set of routines for conversions to and
24402
from a packed decimal format compatible with that used on IBM
24405
@node Interfaces VxWorks i-vxwork ads,Interfaces VxWorks IO i-vxwoio ads,Interfaces Packed_Decimal i-pacdec ads,The GNAT Library
24406
@anchor{gnat_rm/the_gnat_library id134}@anchor{390}@anchor{gnat_rm/the_gnat_library interfaces-vxworks-i-vxwork-ads}@anchor{391}
24407
@section @cite{Interfaces.VxWorks} (@code{i-vxwork.ads})
24410
@geindex Interfaces.VxWorks (i-vxwork.ads)
24412
@geindex Interfacing to VxWorks
24415
@geindex interfacing
24417
This package provides a limited binding to the VxWorks API.
24418
In particular, it interfaces with the
24419
VxWorks hardware interrupt facilities.
24421
@node Interfaces VxWorks IO i-vxwoio ads,System Address_Image s-addima ads,Interfaces VxWorks i-vxwork ads,The GNAT Library
24422
@anchor{gnat_rm/the_gnat_library interfaces-vxworks-io-i-vxwoio-ads}@anchor{392}@anchor{gnat_rm/the_gnat_library id135}@anchor{393}
24423
@section @cite{Interfaces.VxWorks.IO} (@code{i-vxwoio.ads})
24426
@geindex Interfaces.VxWorks.IO (i-vxwoio.ads)
24428
@geindex Interfacing to VxWorks' I/O
24431
@geindex I/O interfacing
24434
@geindex Get_Immediate
24436
@geindex Get_Immediate
24439
This package provides a binding to the ioctl (IO/Control)
24440
function of VxWorks, defining a set of option values and
24441
function codes. A particular use of this package is
24442
to enable the use of Get_Immediate under VxWorks.
24444
@node System Address_Image s-addima ads,System Assertions s-assert ads,Interfaces VxWorks IO i-vxwoio ads,The GNAT Library
24445
@anchor{gnat_rm/the_gnat_library system-address-image-s-addima-ads}@anchor{394}@anchor{gnat_rm/the_gnat_library id136}@anchor{395}
24446
@section @cite{System.Address_Image} (@code{s-addima.ads})
24449
@geindex System.Address_Image (s-addima.ads)
24451
@geindex Address image
24454
@geindex of an address
24456
This function provides a useful debugging
24457
function that gives an (implementation dependent)
24458
string which identifies an address.
24460
@node System Assertions s-assert ads,System Atomic_Counters s-atocou ads,System Address_Image s-addima ads,The GNAT Library
24461
@anchor{gnat_rm/the_gnat_library system-assertions-s-assert-ads}@anchor{396}@anchor{gnat_rm/the_gnat_library id137}@anchor{397}
24462
@section @cite{System.Assertions} (@code{s-assert.ads})
24465
@geindex System.Assertions (s-assert.ads)
24467
@geindex Assertions
24469
@geindex Assert_Failure
24472
This package provides the declaration of the exception raised
24473
by an run-time assertion failure, as well as the routine that
24474
is used internally to raise this assertion.
24476
@node System Atomic_Counters s-atocou ads,System Memory s-memory ads,System Assertions s-assert ads,The GNAT Library
24477
@anchor{gnat_rm/the_gnat_library id138}@anchor{398}@anchor{gnat_rm/the_gnat_library system-atomic-counters-s-atocou-ads}@anchor{399}
24478
@section @cite{System.Atomic_Counters} (@code{s-atocou.ads})
24481
@geindex System.Atomic_Counters (s-atocou.ads)
24483
This package provides the declaration of an atomic counter type,
24484
together with efficient routines (using hardware
24485
synchronization primitives) for incrementing, decrementing,
24486
and testing of these counters. This package is implemented
24487
on most targets, including all Alpha, ia64, PowerPC, SPARC V9,
24488
x86, and x86_64 platforms.
24490
@node System Memory s-memory ads,System Multiprocessors s-multip ads,System Atomic_Counters s-atocou ads,The GNAT Library
24491
@anchor{gnat_rm/the_gnat_library system-memory-s-memory-ads}@anchor{39a}@anchor{gnat_rm/the_gnat_library id139}@anchor{39b}
24492
@section @cite{System.Memory} (@code{s-memory.ads})
24495
@geindex System.Memory (s-memory.ads)
24497
@geindex Memory allocation
24499
This package provides the interface to the low level routines used
24500
by the generated code for allocation and freeing storage for the
24501
default storage pool (analogous to the C routines malloc and free.
24502
It also provides a reallocation interface analogous to the C routine
24503
realloc. The body of this unit may be modified to provide alternative
24504
allocation mechanisms for the default pool, and in addition, direct
24505
calls to this unit may be made for low level allocation uses (for
24506
example see the body of @cite{GNAT.Tables}).
24508
@node System Multiprocessors s-multip ads,System Multiprocessors Dispatching_Domains s-mudido ads,System Memory s-memory ads,The GNAT Library
24509
@anchor{gnat_rm/the_gnat_library id140}@anchor{39c}@anchor{gnat_rm/the_gnat_library system-multiprocessors-s-multip-ads}@anchor{39d}
24510
@section @cite{System.Multiprocessors} (@code{s-multip.ads})
24513
@geindex System.Multiprocessors (s-multip.ads)
24515
@geindex Multiprocessor interface
24517
This is an Ada 2012 unit defined in the Ada 2012 Reference Manual, but
24518
in GNAT we also make it available in Ada 95 and Ada 2005 (where it is
24519
technically an implementation-defined addition).
24521
@node System Multiprocessors Dispatching_Domains s-mudido ads,System Partition_Interface s-parint ads,System Multiprocessors s-multip ads,The GNAT Library
24522
@anchor{gnat_rm/the_gnat_library system-multiprocessors-dispatching-domains-s-mudido-ads}@anchor{39e}@anchor{gnat_rm/the_gnat_library id141}@anchor{39f}
24523
@section @cite{System.Multiprocessors.Dispatching_Domains} (@code{s-mudido.ads})
24526
@geindex System.Multiprocessors.Dispatching_Domains (s-mudido.ads)
24528
@geindex Multiprocessor interface
24530
This is an Ada 2012 unit defined in the Ada 2012 Reference Manual, but
24531
in GNAT we also make it available in Ada 95 and Ada 2005 (where it is
24532
technically an implementation-defined addition).
24534
@node System Partition_Interface s-parint ads,System Pool_Global s-pooglo ads,System Multiprocessors Dispatching_Domains s-mudido ads,The GNAT Library
24535
@anchor{gnat_rm/the_gnat_library id142}@anchor{3a0}@anchor{gnat_rm/the_gnat_library system-partition-interface-s-parint-ads}@anchor{3a1}
24536
@section @cite{System.Partition_Interface} (@code{s-parint.ads})
24539
@geindex System.Partition_Interface (s-parint.ads)
24541
@geindex Partition interfacing functions
24543
This package provides facilities for partition interfacing. It
24544
is used primarily in a distribution context when using Annex E
24547
@node System Pool_Global s-pooglo ads,System Pool_Local s-pooloc ads,System Partition_Interface s-parint ads,The GNAT Library
24548
@anchor{gnat_rm/the_gnat_library id143}@anchor{3a2}@anchor{gnat_rm/the_gnat_library system-pool-global-s-pooglo-ads}@anchor{3a3}
24549
@section @cite{System.Pool_Global} (@code{s-pooglo.ads})
24552
@geindex System.Pool_Global (s-pooglo.ads)
24554
@geindex Storage pool
24557
@geindex Global storage pool
24559
This package provides a storage pool that is equivalent to the default
24560
storage pool used for access types for which no pool is specifically
24561
declared. It uses malloc/free to allocate/free and does not attempt to
24562
do any automatic reclamation.
24564
@node System Pool_Local s-pooloc ads,System Restrictions s-restri ads,System Pool_Global s-pooglo ads,The GNAT Library
24565
@anchor{gnat_rm/the_gnat_library id144}@anchor{3a4}@anchor{gnat_rm/the_gnat_library system-pool-local-s-pooloc-ads}@anchor{3a5}
24566
@section @cite{System.Pool_Local} (@code{s-pooloc.ads})
24569
@geindex System.Pool_Local (s-pooloc.ads)
24571
@geindex Storage pool
24574
@geindex Local storage pool
24576
This package provides a storage pool that is intended for use with locally
24577
defined access types. It uses malloc/free for allocate/free, and maintains
24578
a list of allocated blocks, so that all storage allocated for the pool can
24579
be freed automatically when the pool is finalized.
24581
@node System Restrictions s-restri ads,System Rident s-rident ads,System Pool_Local s-pooloc ads,The GNAT Library
24582
@anchor{gnat_rm/the_gnat_library id145}@anchor{3a6}@anchor{gnat_rm/the_gnat_library system-restrictions-s-restri-ads}@anchor{3a7}
24583
@section @cite{System.Restrictions} (@code{s-restri.ads})
24586
@geindex System.Restrictions (s-restri.ads)
24588
@geindex Run-time restrictions access
24590
This package provides facilities for accessing at run time
24591
the status of restrictions specified at compile time for
24592
the partition. Information is available both with regard
24593
to actual restrictions specified, and with regard to
24594
compiler determined information on which restrictions
24595
are violated by one or more packages in the partition.
24597
@node System Rident s-rident ads,System Strings Stream_Ops s-ststop ads,System Restrictions s-restri ads,The GNAT Library
24598
@anchor{gnat_rm/the_gnat_library system-rident-s-rident-ads}@anchor{3a8}@anchor{gnat_rm/the_gnat_library id146}@anchor{3a9}
24599
@section @cite{System.Rident} (@code{s-rident.ads})
24602
@geindex System.Rident (s-rident.ads)
24604
@geindex Restrictions definitions
24606
This package provides definitions of the restrictions
24607
identifiers supported by GNAT, and also the format of
24608
the restrictions provided in package System.Restrictions.
24609
It is not normally necessary to @cite{with} this generic package
24610
since the necessary instantiation is included in
24611
package System.Restrictions.
24613
@node System Strings Stream_Ops s-ststop ads,System Unsigned_Types s-unstyp ads,System Rident s-rident ads,The GNAT Library
24614
@anchor{gnat_rm/the_gnat_library id147}@anchor{3aa}@anchor{gnat_rm/the_gnat_library system-strings-stream-ops-s-ststop-ads}@anchor{3ab}
24615
@section @cite{System.Strings.Stream_Ops} (@code{s-ststop.ads})
24618
@geindex System.Strings.Stream_Ops (s-ststop.ads)
24620
@geindex Stream operations
24622
@geindex String stream operations
24624
This package provides a set of stream subprograms for standard string types.
24625
It is intended primarily to support implicit use of such subprograms when
24626
stream attributes are applied to string types, but the subprograms in this
24627
package can be used directly by application programs.
24629
@node System Unsigned_Types s-unstyp ads,System Wch_Cnv s-wchcnv ads,System Strings Stream_Ops s-ststop ads,The GNAT Library
24630
@anchor{gnat_rm/the_gnat_library system-unsigned-types-s-unstyp-ads}@anchor{3ac}@anchor{gnat_rm/the_gnat_library id148}@anchor{3ad}
24631
@section @cite{System.Unsigned_Types} (@code{s-unstyp.ads})
24634
@geindex System.Unsigned_Types (s-unstyp.ads)
24636
This package contains definitions of standard unsigned types that
24637
correspond in size to the standard signed types declared in Standard,
24638
and (unlike the types in Interfaces) have corresponding names. It
24639
also contains some related definitions for other specialized types
24640
used by the compiler in connection with packed array types.
24642
@node System Wch_Cnv s-wchcnv ads,System Wch_Con s-wchcon ads,System Unsigned_Types s-unstyp ads,The GNAT Library
24643
@anchor{gnat_rm/the_gnat_library system-wch-cnv-s-wchcnv-ads}@anchor{3ae}@anchor{gnat_rm/the_gnat_library id149}@anchor{3af}
24644
@section @cite{System.Wch_Cnv} (@code{s-wchcnv.ads})
24647
@geindex System.Wch_Cnv (s-wchcnv.ads)
24649
@geindex Wide Character
24650
@geindex Representation
24652
@geindex Wide String
24653
@geindex Conversion
24655
@geindex Representation of wide characters
24657
This package provides routines for converting between
24658
wide and wide wide characters and a representation as a value of type
24659
@cite{Standard.String}, using a specified wide character
24660
encoding method. It uses definitions in
24661
package @cite{System.Wch_Con}.
24663
@node System Wch_Con s-wchcon ads,,System Wch_Cnv s-wchcnv ads,The GNAT Library
24664
@anchor{gnat_rm/the_gnat_library system-wch-con-s-wchcon-ads}@anchor{3b0}@anchor{gnat_rm/the_gnat_library id150}@anchor{3b1}
24665
@section @cite{System.Wch_Con} (@code{s-wchcon.ads})
24668
@geindex System.Wch_Con (s-wchcon.ads)
24670
This package provides definitions and descriptions of
24671
the various methods used for encoding wide characters
24672
in ordinary strings. These definitions are used by
24673
the package @cite{System.Wch_Cnv}.
24675
@node Interfacing to Other Languages,Specialized Needs Annexes,The GNAT Library,Top
24676
@anchor{gnat_rm/interfacing_to_other_languages interfacing-to-other-languages}@anchor{11}@anchor{gnat_rm/interfacing_to_other_languages doc}@anchor{3b2}@anchor{gnat_rm/interfacing_to_other_languages id1}@anchor{3b3}
24677
@chapter Interfacing to Other Languages
24680
The facilities in Annex B of the Ada Reference Manual are fully
24681
implemented in GNAT, and in addition, a full interface to C++ is
24685
* Interfacing to C::
24686
* Interfacing to C++::
24687
* Interfacing to COBOL::
24688
* Interfacing to Fortran::
24689
* Interfacing to non-GNAT Ada code::
24693
@node Interfacing to C,Interfacing to C++,,Interfacing to Other Languages
24694
@anchor{gnat_rm/interfacing_to_other_languages interfacing-to-c}@anchor{3b4}@anchor{gnat_rm/interfacing_to_other_languages id2}@anchor{3b5}
24695
@section Interfacing to C
24698
Interfacing to C with GNAT can use one of two approaches:
24704
The types in the package @cite{Interfaces.C} may be used.
24707
Standard Ada types may be used directly. This may be less portable to
24708
other compilers, but will work on all GNAT compilers, which guarantee
24709
correspondence between the C and Ada types.
24712
Pragma @cite{Convention C} may be applied to Ada types, but mostly has no
24713
effect, since this is the default. The following table shows the
24714
correspondence between Ada scalar types and the corresponding C types.
24717
@multitable {xxxxxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx}
24736
@code{Short_Integer}
24744
@code{Short_Short_Integer}
24752
@code{Long_Integer}
24760
@code{Long_Long_Integer}
24792
@code{Long_Long_Float}
24796
This is the longest floating-point type supported by the hardware.
24801
Additionally, there are the following general correspondences between Ada
24808
Ada enumeration types map to C enumeration types directly if pragma
24809
@cite{Convention C} is specified, which causes them to have int
24810
length. Without pragma @cite{Convention C}, Ada enumeration types map to
24811
8, 16, or 32 bits (i.e., C types @cite{signed char}, @cite{short},
24812
@cite{int}, respectively) depending on the number of values passed.
24813
This is the only case in which pragma @cite{Convention C} affects the
24814
representation of an Ada type.
24817
Ada access types map to C pointers, except for the case of pointers to
24818
unconstrained types in Ada, which have no direct C equivalent.
24821
Ada arrays map directly to C arrays.
24824
Ada records map directly to C structures.
24827
Packed Ada records map to C structures where all members are bit fields
24828
of the length corresponding to the @code{type'Size} value in Ada.
24831
@node Interfacing to C++,Interfacing to COBOL,Interfacing to C,Interfacing to Other Languages
24832
@anchor{gnat_rm/interfacing_to_other_languages id4}@anchor{3b6}@anchor{gnat_rm/interfacing_to_other_languages id3}@anchor{3f}
24833
@section Interfacing to C++
24836
The interface to C++ makes use of the following pragmas, which are
24837
primarily intended to be constructed automatically using a binding generator
24838
tool, although it is possible to construct them by hand.
24840
Using these pragmas it is possible to achieve complete
24841
inter-operability between Ada tagged types and C++ class definitions.
24842
See @ref{7,,Implementation Defined Pragmas}, for more details.
24847
@item @emph{pragma CPP_Class ([Entity =>] `LOCAL_NAME`)}
24849
The argument denotes an entity in the current declarative region that is
24850
declared as a tagged or untagged record type. It indicates that the type
24851
corresponds to an externally declared C++ class type, and is to be laid
24852
out the same way that C++ would lay out the type.
24854
Note: Pragma @cite{CPP_Class} is currently obsolete. It is supported
24855
for backward compatibility but its functionality is available
24856
using pragma @cite{Import} with @cite{Convention} = @cite{CPP}.
24858
@item @emph{pragma CPP_Constructor ([Entity =>] `LOCAL_NAME`)}
24860
This pragma identifies an imported function (imported in the usual way
24861
with pragma @cite{Import}) as corresponding to a C++ constructor.
24864
A few restrictions are placed on the use of the @cite{Access} attribute
24865
in conjunction with subprograms subject to convention @cite{CPP}: the
24866
attribute may be used neither on primitive operations of a tagged
24867
record type with convention @cite{CPP}, imported or not, nor on
24868
subprograms imported with pragma @cite{CPP_Constructor}.
24870
In addition, C++ exceptions are propagated and can be handled in an
24871
@cite{others} choice of an exception handler. The corresponding Ada
24872
occurrence has no message, and the simple name of the exception identity
24873
contains @code{Foreign_Exception}. Finalization and awaiting dependent
24874
tasks works properly when such foreign exceptions are propagated.
24876
It is also possible to import a C++ exception using the following syntax:
24879
LOCAL_NAME : exception;
24880
pragma Import (Cpp,
24881
[Entity =>] LOCAL_NAME,
24882
[External_Name =>] static_string_EXPRESSION);
24885
The @cite{External_Name} is the name of the C++ RTTI symbol. You can then
24886
cover a specific C++ exception in an exception handler.
24888
@node Interfacing to COBOL,Interfacing to Fortran,Interfacing to C++,Interfacing to Other Languages
24889
@anchor{gnat_rm/interfacing_to_other_languages id5}@anchor{3b7}@anchor{gnat_rm/interfacing_to_other_languages interfacing-to-cobol}@anchor{3b8}
24890
@section Interfacing to COBOL
24893
Interfacing to COBOL is achieved as described in section B.4 of
24894
the Ada Reference Manual.
24896
@node Interfacing to Fortran,Interfacing to non-GNAT Ada code,Interfacing to COBOL,Interfacing to Other Languages
24897
@anchor{gnat_rm/interfacing_to_other_languages id6}@anchor{3b9}@anchor{gnat_rm/interfacing_to_other_languages interfacing-to-fortran}@anchor{3ba}
24898
@section Interfacing to Fortran
24901
Interfacing to Fortran is achieved as described in section B.5 of the
24902
Ada Reference Manual. The pragma @cite{Convention Fortran}, applied to a
24903
multi-dimensional array causes the array to be stored in column-major
24904
order as required for convenient interface to Fortran.
24906
@node Interfacing to non-GNAT Ada code,,Interfacing to Fortran,Interfacing to Other Languages
24907
@anchor{gnat_rm/interfacing_to_other_languages interfacing-to-non-gnat-ada-code}@anchor{3bb}@anchor{gnat_rm/interfacing_to_other_languages id7}@anchor{3bc}
24908
@section Interfacing to non-GNAT Ada code
24911
It is possible to specify the convention @cite{Ada} in a pragma
24912
@cite{Import} or pragma @cite{Export}. However this refers to
24913
the calling conventions used by GNAT, which may or may not be
24914
similar enough to those used by some other Ada 83 / Ada 95 / Ada 2005
24915
compiler to allow interoperation.
24917
If arguments types are kept simple, and if the foreign compiler generally
24918
follows system calling conventions, then it may be possible to integrate
24919
files compiled by other Ada compilers, provided that the elaboration
24920
issues are adequately addressed (for example by eliminating the
24921
need for any load time elaboration).
24923
In particular, GNAT running on VMS is designed to
24924
be highly compatible with the DEC Ada 83 compiler, so this is one
24925
case in which it is possible to import foreign units of this type,
24926
provided that the data items passed are restricted to simple scalar
24927
values or simple record types without variants, or simple array
24928
types with fixed bounds.
24930
@node Specialized Needs Annexes,Implementation of Specific Ada Features,Interfacing to Other Languages,Top
24931
@anchor{gnat_rm/specialized_needs_annexes specialized-needs-annexes}@anchor{12}@anchor{gnat_rm/specialized_needs_annexes doc}@anchor{3bd}@anchor{gnat_rm/specialized_needs_annexes id1}@anchor{3be}
24932
@chapter Specialized Needs Annexes
24935
Ada 95, Ada 2005, and Ada 2012 define a number of Specialized Needs Annexes, which are not
24936
required in all implementations. However, as described in this chapter,
24937
GNAT implements all of these annexes:
24942
@item @emph{Systems Programming (Annex C)}
24944
The Systems Programming Annex is fully implemented.
24946
@item @emph{Real-Time Systems (Annex D)}
24948
The Real-Time Systems Annex is fully implemented.
24950
@item @emph{Distributed Systems (Annex E)}
24952
Stub generation is fully implemented in the GNAT compiler. In addition,
24953
a complete compatible PCS is available as part of the GLADE system,
24954
a separate product. When the two
24955
products are used in conjunction, this annex is fully implemented.
24957
@item @emph{Information Systems (Annex F)}
24959
The Information Systems annex is fully implemented.
24961
@item @emph{Numerics (Annex G)}
24963
The Numerics Annex is fully implemented.
24965
@item @emph{Safety and Security / High-Integrity Systems (Annex H)}
24967
The Safety and Security Annex (termed the High-Integrity Systems Annex
24968
in Ada 2005) is fully implemented.
24971
@node Implementation of Specific Ada Features,Implementation of Ada 2012 Features,Specialized Needs Annexes,Top
24972
@anchor{gnat_rm/implementation_of_specific_ada_features implementation-of-specific-ada-features}@anchor{13}@anchor{gnat_rm/implementation_of_specific_ada_features doc}@anchor{3bf}@anchor{gnat_rm/implementation_of_specific_ada_features id1}@anchor{3c0}
24973
@chapter Implementation of Specific Ada Features
24976
This chapter describes the GNAT implementation of several Ada language
24980
* Machine Code Insertions::
24981
* GNAT Implementation of Tasking::
24982
* GNAT Implementation of Shared Passive Packages::
24983
* Code Generation for Array Aggregates::
24984
* The Size of Discriminated Records with Default Discriminants::
24985
* Strict Conformance to the Ada Reference Manual::
24989
@node Machine Code Insertions,GNAT Implementation of Tasking,,Implementation of Specific Ada Features
24990
@anchor{gnat_rm/implementation_of_specific_ada_features machine-code-insertions}@anchor{125}@anchor{gnat_rm/implementation_of_specific_ada_features id2}@anchor{3c1}
24991
@section Machine Code Insertions
24994
@geindex Machine Code insertions
24996
Package @cite{Machine_Code} provides machine code support as described
24997
in the Ada Reference Manual in two separate forms:
25003
Machine code statements, consisting of qualified expressions that
25004
fit the requirements of RM section 13.8.
25007
An intrinsic callable procedure, providing an alternative mechanism of
25008
including machine instructions in a subprogram.
25011
The two features are similar, and both are closely related to the mechanism
25012
provided by the asm instruction in the GNU C compiler. Full understanding
25013
and use of the facilities in this package requires understanding the asm
25014
instruction, see the section on Extended Asm in
25015
@cite{Using_the_GNU_Compiler_Collection_(GCC)}.
25017
Calls to the function @cite{Asm} and the procedure @cite{Asm} have identical
25018
semantic restrictions and effects as described below. Both are provided so
25019
that the procedure call can be used as a statement, and the function call
25020
can be used to form a code_statement.
25022
Consider this C @cite{asm} instruction:
25025
asm ("fsinx %1 %0" : "=f" (result) : "f" (angle));
25028
The equivalent can be written for GNAT as:
25031
Asm ("fsinx %1 %0",
25032
My_Float'Asm_Output ("=f", result),
25033
My_Float'Asm_Input ("f", angle));
25036
The first argument to @cite{Asm} is the assembler template, and is
25037
identical to what is used in GNU C. This string must be a static
25038
expression. The second argument is the output operand list. It is
25039
either a single @cite{Asm_Output} attribute reference, or a list of such
25040
references enclosed in parentheses (technically an array aggregate of
25043
The @cite{Asm_Output} attribute denotes a function that takes two
25044
parameters. The first is a string, the second is the name of a variable
25045
of the type designated by the attribute prefix. The first (string)
25046
argument is required to be a static expression and designates the
25047
constraint (see the section on Constraints in
25048
@cite{Using_the_GNU_Compiler_Collection_(GCC)})
25049
for the parameter; e.g., what kind of register is required. The second
25050
argument is the variable to be written or updated with the
25051
result. The possible values for constraint are the same as those used in
25052
the RTL, and are dependent on the configuration file used to build the
25053
GCC back end. If there are no output operands, then this argument may
25054
either be omitted, or explicitly given as @cite{No_Output_Operands}.
25055
No support is provided for GNU C's symbolic names for output parameters.
25057
The second argument of @code{my_float'Asm_Output} functions as
25058
though it were an @cite{out} parameter, which is a little curious, but
25059
all names have the form of expressions, so there is no syntactic
25060
irregularity, even though normally functions would not be permitted
25061
@cite{out} parameters. The third argument is the list of input
25062
operands. It is either a single @cite{Asm_Input} attribute reference, or
25063
a list of such references enclosed in parentheses (technically an array
25064
aggregate of such references).
25066
The @cite{Asm_Input} attribute denotes a function that takes two
25067
parameters. The first is a string, the second is an expression of the
25068
type designated by the prefix. The first (string) argument is required
25069
to be a static expression, and is the constraint for the parameter,
25070
(e.g., what kind of register is required). The second argument is the
25071
value to be used as the input argument. The possible values for the
25072
constraint are the same as those used in the RTL, and are dependent on
25073
the configuration file used to built the GCC back end.
25074
No support is provided for GNU C's symbolic names for input parameters.
25076
If there are no input operands, this argument may either be omitted, or
25077
explicitly given as @cite{No_Input_Operands}. The fourth argument, not
25078
present in the above example, is a list of register names, called the
25079
@emph{clobber} argument. This argument, if given, must be a static string
25080
expression, and is a space or comma separated list of names of registers
25081
that must be considered destroyed as a result of the @cite{Asm} call. If
25082
this argument is the null string (the default value), then the code
25083
generator assumes that no additional registers are destroyed.
25084
In addition to registers, the special clobbers @cite{memory} and
25085
@cite{cc} as described in the GNU C docs are both supported.
25087
The fifth argument, not present in the above example, called the
25088
@emph{volatile} argument, is by default @cite{False}. It can be set to
25089
the literal value @cite{True} to indicate to the code generator that all
25090
optimizations with respect to the instruction specified should be
25091
suppressed, and in particular an instruction that has outputs
25092
will still be generated, even if none of the outputs are
25093
used. See @cite{Using_the_GNU_Compiler_Collection_(GCC)}
25094
for the full description.
25095
Generally it is strongly advisable to use Volatile for any ASM statement
25096
that is missing either input or output operands or to avoid unwanted
25097
optimizations. A warning is generated if this advice is not followed.
25099
No support is provided for GNU C's @cite{asm goto} feature.
25101
The @cite{Asm} subprograms may be used in two ways. First the procedure
25102
forms can be used anywhere a procedure call would be valid, and
25103
correspond to what the RM calls 'intrinsic' routines. Such calls can
25104
be used to intersperse machine instructions with other Ada statements.
25105
Second, the function forms, which return a dummy value of the limited
25106
private type @cite{Asm_Insn}, can be used in code statements, and indeed
25107
this is the only context where such calls are allowed. Code statements
25108
appear as aggregates of the form:
25111
Asm_Insn'(Asm (...));
25112
Asm_Insn'(Asm_Volatile (...));
25115
In accordance with RM rules, such code statements are allowed only
25116
within subprograms whose entire body consists of such statements. It is
25117
not permissible to intermix such statements with other Ada statements.
25119
Typically the form using intrinsic procedure calls is more convenient
25120
and more flexible. The code statement form is provided to meet the RM
25121
suggestion that such a facility should be made available. The following
25122
is the exact syntax of the call to @cite{Asm}. As usual, if named notation
25123
is used, the arguments may be given in arbitrary order, following the
25124
normal rules for use of positional and named arguments:
25128
[Template =>] static_string_EXPRESSION
25129
[,[Outputs =>] OUTPUT_OPERAND_LIST ]
25130
[,[Inputs =>] INPUT_OPERAND_LIST ]
25131
[,[Clobber =>] static_string_EXPRESSION ]
25132
[,[Volatile =>] static_boolean_EXPRESSION] )
25134
OUTPUT_OPERAND_LIST ::=
25135
[PREFIX.]No_Output_Operands
25136
| OUTPUT_OPERAND_ATTRIBUTE
25137
| (OUTPUT_OPERAND_ATTRIBUTE @{,OUTPUT_OPERAND_ATTRIBUTE@})
25139
OUTPUT_OPERAND_ATTRIBUTE ::=
25140
SUBTYPE_MARK'Asm_Output (static_string_EXPRESSION, NAME)
25142
INPUT_OPERAND_LIST ::=
25143
[PREFIX.]No_Input_Operands
25144
| INPUT_OPERAND_ATTRIBUTE
25145
| (INPUT_OPERAND_ATTRIBUTE @{,INPUT_OPERAND_ATTRIBUTE@})
25147
INPUT_OPERAND_ATTRIBUTE ::=
25148
SUBTYPE_MARK'Asm_Input (static_string_EXPRESSION, EXPRESSION)
25151
The identifiers @cite{No_Input_Operands} and @cite{No_Output_Operands}
25152
are declared in the package @cite{Machine_Code} and must be referenced
25153
according to normal visibility rules. In particular if there is no
25154
@cite{use} clause for this package, then appropriate package name
25155
qualification is required.
25157
@node GNAT Implementation of Tasking,GNAT Implementation of Shared Passive Packages,Machine Code Insertions,Implementation of Specific Ada Features
25158
@anchor{gnat_rm/implementation_of_specific_ada_features id3}@anchor{3c2}@anchor{gnat_rm/implementation_of_specific_ada_features gnat-implementation-of-tasking}@anchor{3c3}
25159
@section GNAT Implementation of Tasking
25162
This chapter outlines the basic GNAT approach to tasking (in particular,
25163
a multi-layered library for portability) and discusses issues related
25164
to compliance with the Real-Time Systems Annex.
25167
* Mapping Ada Tasks onto the Underlying Kernel Threads::
25168
* Ensuring Compliance with the Real-Time Annex::
25172
@node Mapping Ada Tasks onto the Underlying Kernel Threads,Ensuring Compliance with the Real-Time Annex,,GNAT Implementation of Tasking
25173
@anchor{gnat_rm/implementation_of_specific_ada_features mapping-ada-tasks-onto-the-underlying-kernel-threads}@anchor{3c4}@anchor{gnat_rm/implementation_of_specific_ada_features id4}@anchor{3c5}
25174
@subsection Mapping Ada Tasks onto the Underlying Kernel Threads
25177
GNAT's run-time support comprises two layers:
25183
GNARL (GNAT Run-time Layer)
25186
GNULL (GNAT Low-level Library)
25189
In GNAT, Ada's tasking services rely on a platform and OS independent
25190
layer known as GNARL. This code is responsible for implementing the
25191
correct semantics of Ada's task creation, rendezvous, protected
25194
GNARL decomposes Ada's tasking semantics into simpler lower level
25195
operations such as create a thread, set the priority of a thread,
25196
yield, create a lock, lock/unlock, etc. The spec for these low-level
25197
operations constitutes GNULLI, the GNULL Interface. This interface is
25198
directly inspired from the POSIX real-time API.
25200
If the underlying executive or OS implements the POSIX standard
25201
faithfully, the GNULL Interface maps as is to the services offered by
25202
the underlying kernel. Otherwise, some target dependent glue code maps
25203
the services offered by the underlying kernel to the semantics expected
25206
Whatever the underlying OS (VxWorks, UNIX, Windows, etc.) the
25207
key point is that each Ada task is mapped on a thread in the underlying
25208
kernel. For example, in the case of VxWorks, one Ada task = one VxWorks task.
25210
In addition Ada task priorities map onto the underlying thread priorities.
25211
Mapping Ada tasks onto the underlying kernel threads has several advantages:
25217
The underlying scheduler is used to schedule the Ada tasks. This
25218
makes Ada tasks as efficient as kernel threads from a scheduling
25222
Interaction with code written in C containing threads is eased
25223
since at the lowest level Ada tasks and C threads map onto the same
25224
underlying kernel concept.
25227
When an Ada task is blocked during I/O the remaining Ada tasks are
25231
On multiprocessor systems Ada tasks can execute in parallel.
25234
Some threads libraries offer a mechanism to fork a new process, with the
25235
child process duplicating the threads from the parent.
25237
support this functionality when the parent contains more than one task.
25239
@geindex Forking a new process
25241
@node Ensuring Compliance with the Real-Time Annex,,Mapping Ada Tasks onto the Underlying Kernel Threads,GNAT Implementation of Tasking
25242
@anchor{gnat_rm/implementation_of_specific_ada_features id5}@anchor{3c6}@anchor{gnat_rm/implementation_of_specific_ada_features ensuring-compliance-with-the-real-time-annex}@anchor{3c7}
25243
@subsection Ensuring Compliance with the Real-Time Annex
25246
@geindex Real-Time Systems Annex compliance
25248
Although mapping Ada tasks onto
25249
the underlying threads has significant advantages, it does create some
25250
complications when it comes to respecting the scheduling semantics
25251
specified in the real-time annex (Annex D).
25253
For instance the Annex D requirement for the @cite{FIFO_Within_Priorities}
25254
scheduling policy states:
25258
@emph{When the active priority of a ready task that is not running
25259
changes, or the setting of its base priority takes effect, the
25260
task is removed from the ready queue for its old active priority
25261
and is added at the tail of the ready queue for its new active
25262
priority, except in the case where the active priority is lowered
25263
due to the loss of inherited priority, in which case the task is
25264
added at the head of the ready queue for its new active priority.}
25267
While most kernels do put tasks at the end of the priority queue when
25268
a task changes its priority, (which respects the main
25269
FIFO_Within_Priorities requirement), almost none keep a thread at the
25270
beginning of its priority queue when its priority drops from the loss
25271
of inherited priority.
25273
As a result most vendors have provided incomplete Annex D implementations.
25275
The GNAT run-time, has a nice cooperative solution to this problem
25276
which ensures that accurate FIFO_Within_Priorities semantics are
25279
The principle is as follows. When an Ada task T is about to start
25280
running, it checks whether some other Ada task R with the same
25281
priority as T has been suspended due to the loss of priority
25282
inheritance. If this is the case, T yields and is placed at the end of
25283
its priority queue. When R arrives at the front of the queue it
25286
Note that this simple scheme preserves the relative order of the tasks
25287
that were ready to execute in the priority queue where R has been
25290
@node GNAT Implementation of Shared Passive Packages,Code Generation for Array Aggregates,GNAT Implementation of Tasking,Implementation of Specific Ada Features
25291
@anchor{gnat_rm/implementation_of_specific_ada_features id6}@anchor{3c8}@anchor{gnat_rm/implementation_of_specific_ada_features gnat-implementation-of-shared-passive-packages}@anchor{3c9}
25292
@section GNAT Implementation of Shared Passive Packages
25295
@geindex Shared passive packages
25297
GNAT fully implements the
25298
@geindex pragma Shared_Passive
25300
@cite{Shared_Passive} for
25301
the purpose of designating shared passive packages.
25302
This allows the use of passive partitions in the
25303
context described in the Ada Reference Manual; i.e., for communication
25304
between separate partitions of a distributed application using the
25305
features in Annex E.
25309
@geindex Distribution Systems Annex
25311
However, the implementation approach used by GNAT provides for more
25312
extensive usage as follows:
25317
@item @emph{Communication between separate programs}
25319
This allows separate programs to access the data in passive
25320
partitions, using protected objects for synchronization where
25321
needed. The only requirement is that the two programs have a
25322
common shared file system. It is even possible for programs
25323
running on different machines with different architectures
25324
(e.g., different endianness) to communicate via the data in
25325
a passive partition.
25327
@item @emph{Persistence between program runs}
25329
The data in a passive package can persist from one run of a
25330
program to another, so that a later program sees the final
25331
values stored by a previous run of the same program.
25334
The implementation approach used is to store the data in files. A
25335
separate stream file is created for each object in the package, and
25336
an access to an object causes the corresponding file to be read or
25339
@geindex SHARED_MEMORY_DIRECTORY environment variable
25341
The environment variable @cite{SHARED_MEMORY_DIRECTORY} should be
25342
set to the directory to be used for these files.
25343
The files in this directory
25344
have names that correspond to their fully qualified names. For
25345
example, if we have the package
25349
pragma Shared_Passive (X);
25355
and the environment variable is set to @cite{/stemp/}, then the files created
25356
will have the names:
25363
These files are created when a value is initially written to the object, and
25364
the files are retained until manually deleted. This provides the persistence
25365
semantics. If no file exists, it means that no partition has assigned a value
25366
to the variable; in this case the initial value declared in the package
25367
will be used. This model ensures that there are no issues in synchronizing
25368
the elaboration process, since elaboration of passive packages elaborates the
25369
initial values, but does not create the files.
25371
The files are written using normal @cite{Stream_IO} access.
25372
If you want to be able
25373
to communicate between programs or partitions running on different
25374
architectures, then you should use the XDR versions of the stream attribute
25375
routines, since these are architecture independent.
25377
If active synchronization is required for access to the variables in the
25378
shared passive package, then as described in the Ada Reference Manual, the
25379
package may contain protected objects used for this purpose. In this case
25380
a lock file (whose name is @code{___lock} (three underscores)
25381
is created in the shared memory directory.
25383
@geindex ___lock file (for shared passive packages)
25385
This is used to provide the required locking
25386
semantics for proper protected object synchronization.
25388
GNAT supports shared passive packages on all platforms
25389
except for OpenVMS.
25391
@node Code Generation for Array Aggregates,The Size of Discriminated Records with Default Discriminants,GNAT Implementation of Shared Passive Packages,Implementation of Specific Ada Features
25392
@anchor{gnat_rm/implementation_of_specific_ada_features code-generation-for-array-aggregates}@anchor{3ca}@anchor{gnat_rm/implementation_of_specific_ada_features id7}@anchor{3cb}
25393
@section Code Generation for Array Aggregates
25396
Aggregates have a rich syntax and allow the user to specify the values of
25397
complex data structures by means of a single construct. As a result, the
25398
code generated for aggregates can be quite complex and involve loops, case
25399
statements and multiple assignments. In the simplest cases, however, the
25400
compiler will recognize aggregates whose components and constraints are
25401
fully static, and in those cases the compiler will generate little or no
25402
executable code. The following is an outline of the code that GNAT generates
25403
for various aggregate constructs. For further details, you will find it
25404
useful to examine the output produced by the -gnatG flag to see the expanded
25405
source that is input to the code generator. You may also want to examine
25406
the assembly code generated at various levels of optimization.
25408
The code generated for aggregates depends on the context, the component values,
25409
and the type. In the context of an object declaration the code generated is
25410
generally simpler than in the case of an assignment. As a general rule, static
25411
component values and static subtypes also lead to simpler code.
25414
* Static constant aggregates with static bounds::
25415
* Constant aggregates with unconstrained nominal types::
25416
* Aggregates with static bounds::
25417
* Aggregates with nonstatic bounds::
25418
* Aggregates in assignment statements::
25422
@node Static constant aggregates with static bounds,Constant aggregates with unconstrained nominal types,,Code Generation for Array Aggregates
25423
@anchor{gnat_rm/implementation_of_specific_ada_features static-constant-aggregates-with-static-bounds}@anchor{3cc}@anchor{gnat_rm/implementation_of_specific_ada_features id8}@anchor{3cd}
25424
@subsection Static constant aggregates with static bounds
25427
For the declarations:
25430
type One_Dim is array (1..10) of integer;
25431
ar0 : constant One_Dim := (1, 2, 3, 4, 5, 6, 7, 8, 9, 0);
25434
GNAT generates no executable code: the constant ar0 is placed in static memory.
25435
The same is true for constant aggregates with named associations:
25438
Cr1 : constant One_Dim := (4 => 16, 2 => 4, 3 => 9, 1 => 1, 5 .. 10 => 0);
25439
Cr3 : constant One_Dim := (others => 7777);
25442
The same is true for multidimensional constant arrays such as:
25445
type two_dim is array (1..3, 1..3) of integer;
25446
Unit : constant two_dim := ( (1,0,0), (0,1,0), (0,0,1));
25449
The same is true for arrays of one-dimensional arrays: the following are
25453
type ar1b is array (1..3) of boolean;
25454
type ar_ar is array (1..3) of ar1b;
25455
None : constant ar1b := (others => false); -- fully static
25456
None2 : constant ar_ar := (1..3 => None); -- fully static
25459
However, for multidimensional aggregates with named associations, GNAT will
25460
generate assignments and loops, even if all associations are static. The
25461
following two declarations generate a loop for the first dimension, and
25462
individual component assignments for the second dimension:
25465
Zero1: constant two_dim := (1..3 => (1..3 => 0));
25466
Zero2: constant two_dim := (others => (others => 0));
25469
@node Constant aggregates with unconstrained nominal types,Aggregates with static bounds,Static constant aggregates with static bounds,Code Generation for Array Aggregates
25470
@anchor{gnat_rm/implementation_of_specific_ada_features constant-aggregates-with-unconstrained-nominal-types}@anchor{3ce}@anchor{gnat_rm/implementation_of_specific_ada_features id9}@anchor{3cf}
25471
@subsection Constant aggregates with unconstrained nominal types
25474
In such cases the aggregate itself establishes the subtype, so that
25475
associations with @cite{others} cannot be used. GNAT determines the
25476
bounds for the actual subtype of the aggregate, and allocates the
25477
aggregate statically as well. No code is generated for the following:
25480
type One_Unc is array (natural range <>) of integer;
25481
Cr_Unc : constant One_Unc := (12,24,36);
25484
@node Aggregates with static bounds,Aggregates with nonstatic bounds,Constant aggregates with unconstrained nominal types,Code Generation for Array Aggregates
25485
@anchor{gnat_rm/implementation_of_specific_ada_features id10}@anchor{3d0}@anchor{gnat_rm/implementation_of_specific_ada_features aggregates-with-static-bounds}@anchor{3d1}
25486
@subsection Aggregates with static bounds
25489
In all previous examples the aggregate was the initial (and immutable) value
25490
of a constant. If the aggregate initializes a variable, then code is generated
25491
for it as a combination of individual assignments and loops over the target
25492
object. The declarations
25495
Cr_Var1 : One_Dim := (2, 5, 7, 11, 0, 0, 0, 0, 0, 0);
25496
Cr_Var2 : One_Dim := (others > -1);
25499
generate the equivalent of
25507
for I in Cr_Var2'range loop
25512
@node Aggregates with nonstatic bounds,Aggregates in assignment statements,Aggregates with static bounds,Code Generation for Array Aggregates
25513
@anchor{gnat_rm/implementation_of_specific_ada_features id11}@anchor{3d2}@anchor{gnat_rm/implementation_of_specific_ada_features aggregates-with-nonstatic-bounds}@anchor{3d3}
25514
@subsection Aggregates with nonstatic bounds
25517
If the bounds of the aggregate are not statically compatible with the bounds
25518
of the nominal subtype of the target, then constraint checks have to be
25519
generated on the bounds. For a multidimensional array, constraint checks may
25520
have to be applied to sub-arrays individually, if they do not have statically
25521
compatible subtypes.
25523
@node Aggregates in assignment statements,,Aggregates with nonstatic bounds,Code Generation for Array Aggregates
25524
@anchor{gnat_rm/implementation_of_specific_ada_features id12}@anchor{3d4}@anchor{gnat_rm/implementation_of_specific_ada_features aggregates-in-assignment-statements}@anchor{3d5}
25525
@subsection Aggregates in assignment statements
25528
In general, aggregate assignment requires the construction of a temporary,
25529
and a copy from the temporary to the target of the assignment. This is because
25530
it is not always possible to convert the assignment into a series of individual
25531
component assignments. For example, consider the simple case:
25537
This cannot be converted into:
25544
So the aggregate has to be built first in a separate location, and then
25545
copied into the target. GNAT recognizes simple cases where this intermediate
25546
step is not required, and the assignments can be performed in place, directly
25547
into the target. The following sufficient criteria are applied:
25553
The bounds of the aggregate are static, and the associations are static.
25556
The components of the aggregate are static constants, names of
25557
simple variables that are not renamings, or expressions not involving
25558
indexed components whose operands obey these rules.
25561
If any of these conditions are violated, the aggregate will be built in
25562
a temporary (created either by the front-end or the code generator) and then
25563
that temporary will be copied onto the target.
25565
@node The Size of Discriminated Records with Default Discriminants,Strict Conformance to the Ada Reference Manual,Code Generation for Array Aggregates,Implementation of Specific Ada Features
25566
@anchor{gnat_rm/implementation_of_specific_ada_features id13}@anchor{3d6}@anchor{gnat_rm/implementation_of_specific_ada_features the-size-of-discriminated-records-with-default-discriminants}@anchor{3d7}
25567
@section The Size of Discriminated Records with Default Discriminants
25570
If a discriminated type @cite{T} has discriminants with default values, it is
25571
possible to declare an object of this type without providing an explicit
25575
type Size is range 1..100;
25577
type Rec (D : Size := 15) is record
25578
Name : String (1..D);
25584
Such an object is said to be @emph{unconstrained}.
25585
The discriminant of the object
25586
can be modified by a full assignment to the object, as long as it preserves the
25587
relation between the value of the discriminant, and the value of the components
25591
Word := (3, "yes");
25593
Word := (5, "maybe");
25595
Word := (5, "no"); -- raises Constraint_Error
25598
In order to support this behavior efficiently, an unconstrained object is
25599
given the maximum size that any value of the type requires. In the case
25600
above, @cite{Word} has storage for the discriminant and for
25601
a @cite{String} of length 100.
25602
It is important to note that unconstrained objects do not require dynamic
25603
allocation. It would be an improper implementation to place on the heap those
25604
components whose size depends on discriminants. (This improper implementation
25605
was used by some Ada83 compilers, where the @cite{Name} component above
25607
been stored as a pointer to a dynamic string). Following the principle that
25608
dynamic storage management should never be introduced implicitly,
25609
an Ada compiler should reserve the full size for an unconstrained declared
25610
object, and place it on the stack.
25612
This maximum size approach
25613
has been a source of surprise to some users, who expect the default
25614
values of the discriminants to determine the size reserved for an
25615
unconstrained object: "If the default is 15, why should the object occupy
25617
The answer, of course, is that the discriminant may be later modified,
25618
and its full range of values must be taken into account. This is why the
25622
type Rec (D : Positive := 15) is record
25623
Name : String (1..D);
25629
is flagged by the compiler with a warning:
25630
an attempt to create @cite{Too_Large} will raise @cite{Storage_Error},
25631
because the required size includes @cite{Positive'Last}
25632
bytes. As the first example indicates, the proper approach is to declare an
25633
index type of 'reasonable' range so that unconstrained objects are not too
25636
One final wrinkle: if the object is declared to be @cite{aliased}, or if it is
25637
created in the heap by means of an allocator, then it is @emph{not}
25639
it is constrained by the default values of the discriminants, and those values
25640
cannot be modified by full assignment. This is because in the presence of
25641
aliasing all views of the object (which may be manipulated by different tasks,
25642
say) must be consistent, so it is imperative that the object, once created,
25645
@node Strict Conformance to the Ada Reference Manual,,The Size of Discriminated Records with Default Discriminants,Implementation of Specific Ada Features
25646
@anchor{gnat_rm/implementation_of_specific_ada_features strict-conformance-to-the-ada-reference-manual}@anchor{3d8}@anchor{gnat_rm/implementation_of_specific_ada_features id14}@anchor{3d9}
25647
@section Strict Conformance to the Ada Reference Manual
25650
The dynamic semantics defined by the Ada Reference Manual impose a set of
25651
run-time checks to be generated. By default, the GNAT compiler will insert many
25652
run-time checks into the compiled code, including most of those required by the
25653
Ada Reference Manual. However, there are two checks that are not enabled in
25654
the default mode for efficiency reasons: checks for access before elaboration
25655
on subprogram calls, and stack overflow checking (most operating systems do not
25656
perform this check by default).
25658
Strict conformance to the Ada Reference Manual can be achieved by adding two
25659
compiler options for dynamic checks for access-before-elaboration on subprogram
25660
calls and generic instantiations (@emph{-gnatE}), and stack overflow checking
25661
(@emph{-fstack-check}).
25663
Note that the result of a floating point arithmetic operation in overflow and
25664
invalid situations, when the @cite{Machine_Overflows} attribute of the result
25665
type is @cite{False}, is to generate IEEE NaN and infinite values. This is the
25666
case for machines compliant with the IEEE floating-point standard, but on
25667
machines that are not fully compliant with this standard, such as Alpha, the
25668
@emph{-mieee} compiler flag must be used for achieving IEEE confirming
25669
behavior (although at the cost of a significant performance penalty), so
25670
infinite and NaN values are properly generated.
25672
@node Implementation of Ada 2012 Features,Obsolescent Features,Implementation of Specific Ada Features,Top
25673
@anchor{gnat_rm/implementation_of_ada_2012_features doc}@anchor{3da}@anchor{gnat_rm/implementation_of_ada_2012_features implementation-of-ada-2012-features}@anchor{14}@anchor{gnat_rm/implementation_of_ada_2012_features id1}@anchor{3db}
25674
@chapter Implementation of Ada 2012 Features
25677
@geindex Ada 2012 implementation status
25679
@geindex -gnat12 option (gcc)
25681
@geindex pragma Ada_2012
25683
@geindex configuration pragma Ada_2012
25685
@geindex Ada_2012 configuration pragma
25687
This chapter contains a complete list of Ada 2012 features that have been
25689
Generally, these features are only
25690
available if the @emph{-gnat12} (Ada 2012 features enabled) option is set,
25691
which is the default behavior,
25692
or if the configuration pragma @cite{Ada_2012} is used.
25694
However, new pragmas, attributes, and restrictions are
25695
unconditionally available, since the Ada 95 standard allows the addition of
25696
new pragmas, attributes, and restrictions (there are exceptions, which are
25697
documented in the individual descriptions), and also certain packages
25698
were made available in earlier versions of Ada.
25700
An ISO date (YYYY-MM-DD) appears in parentheses on the description line.
25701
This date shows the implementation date of the feature. Any wavefront
25702
subsequent to this date will contain the indicated feature, as will any
25703
subsequent releases. A date of 0000-00-00 means that GNAT has always
25704
implemented the feature, or implemented it as soon as it appeared as a
25705
binding interpretation.
25707
Each feature corresponds to an Ada Issue ('AI') approved by the Ada
25708
standardization group (ISO/IEC JTC1/SC22/WG9) for inclusion in Ada 2012.
25709
The features are ordered based on the relevant sections of the Ada
25710
Reference Manual ("RM"). When a given AI relates to multiple points
25711
in the RM, the earliest is used.
25713
A complete description of the AIs may be found in
25714
@indicateurl{http://www.ada-auth.org/ai05-summary.html}.
25716
@geindex AI-0176 (Ada 2012 feature)
25722
@emph{AI-0176 Quantified expressions (2010-09-29)}
25724
Both universally and existentially quantified expressions are implemented.
25725
They use the new syntax for iterators proposed in AI05-139-2, as well as
25726
the standard Ada loop syntax.
25728
RM References: 1.01.04 (12) 2.09 (2/2) 4.04 (7) 4.05.09 (0)
25731
@geindex AI-0079 (Ada 2012 feature)
25737
@emph{AI-0079 Allow other_format characters in source (2010-07-10)}
25739
Wide characters in the unicode category @emph{other_format} are now allowed in
25740
source programs between tokens, but not within a token such as an identifier.
25742
RM References: 2.01 (4/2) 2.02 (7)
25745
@geindex AI-0091 (Ada 2012 feature)
25751
@emph{AI-0091 Do not allow other_format in identifiers (0000-00-00)}
25753
Wide characters in the unicode category @emph{other_format} are not permitted
25754
within an identifier, since this can be a security problem. The error
25755
message for this case has been improved to be more specific, but GNAT has
25756
never allowed such characters to appear in identifiers.
25758
RM References: 2.03 (3.1/2) 2.03 (4/2) 2.03 (5/2) 2.03 (5.1/2) 2.03 (5.2/2) 2.03 (5.3/2) 2.09 (2/2)
25761
@geindex AI-0100 (Ada 2012 feature)
25767
@emph{AI-0100 Placement of pragmas (2010-07-01)}
25769
This AI is an earlier version of AI-163. It simplifies the rules
25770
for legal placement of pragmas. In the case of lists that allow pragmas, if
25771
the list may have no elements, then the list may consist solely of pragmas.
25773
RM References: 2.08 (7)
25776
@geindex AI-0163 (Ada 2012 feature)
25782
@emph{AI-0163 Pragmas in place of null (2010-07-01)}
25784
A statement sequence may be composed entirely of pragmas. It is no longer
25785
necessary to add a dummy @cite{null} statement to make the sequence legal.
25787
RM References: 2.08 (7) 2.08 (16)
25790
@geindex AI-0080 (Ada 2012 feature)
25796
@emph{AI-0080 'View of' not needed if clear from context (0000-00-00)}
25798
This is an editorial change only, described as non-testable in the AI.
25800
RM References: 3.01 (7)
25803
@geindex AI-0183 (Ada 2012 feature)
25809
@emph{AI-0183 Aspect specifications (2010-08-16)}
25811
Aspect specifications have been fully implemented except for pre and post-
25812
conditions, and type invariants, which have their own separate AI's. All
25813
forms of declarations listed in the AI are supported. The following is a
25814
list of the aspects supported (with GNAT implementation aspects marked)
25818
@multitable {xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx} {xxxxxxxxxxxxx}
25863
@cite{Atomic_Components}
25875
@cite{Component_Size}
25881
@cite{Contract_Cases}
25889
@cite{Discard_Names}
25895
@cite{External_Tag}
25901
@cite{Favor_Top_Level}
25915
@cite{Inline_Always}
25931
@cite{Machine_Radix}
25957
@cite{Persistent_BSS}
25983
@cite{Preelaborable_Initialization}
25989
@cite{Pure_Function}
25997
@cite{Remote_Access_Type}
26019
@cite{Storage_Pool}
26025
@cite{Storage_Size}
26043
@cite{Suppress_Debug_Info}
26059
@cite{Thread_Local_Storage}
26067
@cite{Type_Invariant}
26073
@cite{Unchecked_Union}
26079
@cite{Universal_Aliasing}
26095
@cite{Unreferenced}
26103
@cite{Unreferenced_Objects}
26131
@cite{Volatile_Components}
26148
Note that for aspects with an expression, e.g. @cite{Size}, the expression is
26149
treated like a default expression (visibility is analyzed at the point of
26150
occurrence of the aspect, but evaluation of the expression occurs at the
26151
freeze point of the entity involved).
26153
RM References: 3.02.01 (3) 3.02.02 (2) 3.03.01 (2/2) 3.08 (6)
26154
3.09.03 (1.1/2) 6.01 (2/2) 6.07 (2/2) 9.05.02 (2/2) 7.01 (3) 7.03
26155
(2) 7.03 (3) 9.01 (2/2) 9.01 (3/2) 9.04 (2/2) 9.04 (3/2)
26156
9.05.02 (2/2) 11.01 (2) 12.01 (3) 12.03 (2/2) 12.04 (2/2) 12.05 (2)
26157
12.06 (2.1/2) 12.06 (2.2/2) 12.07 (2) 13.01 (0.1/2) 13.03 (5/1)
26161
@geindex AI-0128 (Ada 2012 feature)
26167
@emph{AI-0128 Inequality is a primitive operation (0000-00-00)}
26169
If an equality operator ("=") is declared for a type, then the implicitly
26170
declared inequality operator ("/=") is a primitive operation of the type.
26171
This is the only reasonable interpretation, and is the one always implemented
26172
by GNAT, but the RM was not entirely clear in making this point.
26174
RM References: 3.02.03 (6) 6.06 (6)
26177
@geindex AI-0003 (Ada 2012 feature)
26183
@emph{AI-0003 Qualified expressions as names (2010-07-11)}
26185
In Ada 2012, a qualified expression is considered to be syntactically a name,
26186
meaning that constructs such as @cite{A'(F(X)).B} are now legal. This is
26187
useful in disambiguating some cases of overloading.
26189
RM References: 3.03 (11) 3.03 (21) 4.01 (2) 4.04 (7) 4.07 (3)
26193
@geindex AI-0120 (Ada 2012 feature)
26199
@emph{AI-0120 Constant instance of protected object (0000-00-00)}
26201
This is an RM editorial change only. The section that lists objects that are
26202
constant failed to include the current instance of a protected object
26203
within a protected function. This has always been treated as a constant
26206
RM References: 3.03 (21)
26209
@geindex AI-0008 (Ada 2012 feature)
26215
@emph{AI-0008 General access to constrained objects (0000-00-00)}
26217
The wording in the RM implied that if you have a general access to a
26218
constrained object, it could be used to modify the discriminants. This was
26219
obviously not intended. @cite{Constraint_Error} should be raised, and GNAT
26220
has always done so in this situation.
26222
RM References: 3.03 (23) 3.10.02 (26/2) 4.01 (9) 6.04.01 (17) 8.05.01 (5/2)
26225
@geindex AI-0093 (Ada 2012 feature)
26231
@emph{AI-0093 Additional rules use immutably limited (0000-00-00)}
26233
This is an editorial change only, to make more widespread use of the Ada 2012
26234
'immutably limited'.
26236
RM References: 3.03 (23.4/3)
26239
@geindex AI-0096 (Ada 2012 feature)
26245
@emph{AI-0096 Deriving from formal private types (2010-07-20)}
26247
In general it is illegal for a type derived from a formal limited type to be
26248
nonlimited. This AI makes an exception to this rule: derivation is legal
26249
if it appears in the private part of the generic, and the formal type is not
26250
tagged. If the type is tagged, the legality check must be applied to the
26251
private part of the package.
26253
RM References: 3.04 (5.1/2) 6.02 (7)
26256
@geindex AI-0181 (Ada 2012 feature)
26262
@emph{AI-0181 Soft hyphen is a non-graphic character (2010-07-23)}
26264
From Ada 2005 on, soft hyphen is considered a non-graphic character, which
26265
means that it has a special name (@cite{SOFT_HYPHEN}) in conjunction with the
26266
@cite{Image} and @cite{Value} attributes for the character types. Strictly
26267
speaking this is an inconsistency with Ada 95, but in practice the use of
26268
these attributes is so obscure that it will not cause problems.
26270
RM References: 3.05.02 (2/2) A.01 (35/2) A.03.03 (21)
26273
@geindex AI-0182 (Ada 2012 feature)
26279
@emph{AI-0182 Additional forms for `Character'Value} (0000-00-00)`
26281
This AI allows @cite{Character'Value} to accept the string @cite{'?'} where
26282
@cite{?} is any character including non-graphic control characters. GNAT has
26283
always accepted such strings. It also allows strings such as
26284
@cite{HEX_00000041} to be accepted, but GNAT does not take advantage of this
26285
permission and raises @cite{Constraint_Error}, as is certainly still
26288
RM References: 3.05 (56/2)
26291
@geindex AI-0214 (Ada 2012 feature)
26297
@emph{AI-0214 Defaulted discriminants for limited tagged (2010-10-01)}
26299
Ada 2012 relaxes the restriction that forbids discriminants of tagged types
26300
to have default expressions by allowing them when the type is limited. It
26301
is often useful to define a default value for a discriminant even though
26302
it can't be changed by assignment.
26304
RM References: 3.07 (9.1/2) 3.07.02 (3)
26307
@geindex AI-0102 (Ada 2012 feature)
26313
@emph{AI-0102 Some implicit conversions are illegal (0000-00-00)}
26315
It is illegal to assign an anonymous access constant to an anonymous access
26316
variable. The RM did not have a clear rule to prevent this, but GNAT has
26317
always generated an error for this usage.
26319
RM References: 3.07 (16) 3.07.01 (9) 6.04.01 (6) 8.06 (27/2)
26322
@geindex AI-0158 (Ada 2012 feature)
26328
@emph{AI-0158 Generalizing membership tests (2010-09-16)}
26330
This AI extends the syntax of membership tests to simplify complex conditions
26331
that can be expressed as membership in a subset of values of any type. It
26332
introduces syntax for a list of expressions that may be used in loop contexts
26335
RM References: 3.08.01 (5) 4.04 (3) 4.05.02 (3) 4.05.02 (5) 4.05.02 (27)
26338
@geindex AI-0173 (Ada 2012 feature)
26344
@emph{AI-0173 Testing if tags represent abstract types (2010-07-03)}
26346
The function @cite{Ada.Tags.Type_Is_Abstract} returns @cite{True} if invoked
26347
with the tag of an abstract type, and @cite{False} otherwise.
26349
RM References: 3.09 (7.4/2) 3.09 (12.4/2)
26352
@geindex AI-0076 (Ada 2012 feature)
26358
@emph{AI-0076 function with controlling result (0000-00-00)}
26360
This is an editorial change only. The RM defines calls with controlling
26361
results, but uses the term 'function with controlling result' without an
26362
explicit definition.
26364
RM References: 3.09.02 (2/2)
26367
@geindex AI-0126 (Ada 2012 feature)
26373
@emph{AI-0126 Dispatching with no declared operation (0000-00-00)}
26375
This AI clarifies dispatching rules, and simply confirms that dispatching
26376
executes the operation of the parent type when there is no explicitly or
26377
implicitly declared operation for the descendant type. This has always been
26378
the case in all versions of GNAT.
26380
RM References: 3.09.02 (20/2) 3.09.02 (20.1/2) 3.09.02 (20.2/2)
26383
@geindex AI-0097 (Ada 2012 feature)
26389
@emph{AI-0097 Treatment of abstract null extension (2010-07-19)}
26391
The RM as written implied that in some cases it was possible to create an
26392
object of an abstract type, by having an abstract extension inherit a non-
26393
abstract constructor from its parent type. This mistake has been corrected
26394
in GNAT and in the RM, and this construct is now illegal.
26396
RM References: 3.09.03 (4/2)
26399
@geindex AI-0203 (Ada 2012 feature)
26405
@emph{AI-0203 Extended return cannot be abstract (0000-00-00)}
26407
A return_subtype_indication cannot denote an abstract subtype. GNAT has never
26408
permitted such usage.
26410
RM References: 3.09.03 (8/3)
26413
@geindex AI-0198 (Ada 2012 feature)
26419
@emph{AI-0198 Inheriting abstract operators (0000-00-00)}
26421
This AI resolves a conflict between two rules involving inherited abstract
26422
operations and predefined operators. If a derived numeric type inherits
26423
an abstract operator, it overrides the predefined one. This interpretation
26424
was always the one implemented in GNAT.
26426
RM References: 3.09.03 (4/3)
26429
@geindex AI-0073 (Ada 2012 feature)
26435
@emph{AI-0073 Functions returning abstract types (2010-07-10)}
26437
This AI covers a number of issues regarding returning abstract types. In
26438
particular generic functions cannot have abstract result types or access
26439
result types designated an abstract type. There are some other cases which
26440
are detailed in the AI. Note that this binding interpretation has not been
26441
retrofitted to operate before Ada 2012 mode, since it caused a significant
26442
number of regressions.
26444
RM References: 3.09.03 (8) 3.09.03 (10) 6.05 (8/2)
26447
@geindex AI-0070 (Ada 2012 feature)
26453
@emph{AI-0070 Elaboration of interface types (0000-00-00)}
26455
This is an editorial change only, there are no testable consequences short of
26456
checking for the absence of generated code for an interface declaration.
26458
RM References: 3.09.04 (18/2)
26461
@geindex AI-0208 (Ada 2012 feature)
26467
@emph{AI-0208 Characteristics of incomplete views (0000-00-00)}
26469
The wording in the Ada 2005 RM concerning characteristics of incomplete views
26470
was incorrect and implied that some programs intended to be legal were now
26471
illegal. GNAT had never considered such programs illegal, so it has always
26472
implemented the intent of this AI.
26474
RM References: 3.10.01 (2.4/2) 3.10.01 (2.6/2)
26477
@geindex AI-0162 (Ada 2012 feature)
26483
@emph{AI-0162 Incomplete type completed by partial view (2010-09-15)}
26485
Incomplete types are made more useful by allowing them to be completed by
26486
private types and private extensions.
26488
RM References: 3.10.01 (2.5/2) 3.10.01 (2.6/2) 3.10.01 (3) 3.10.01 (4/2)
26491
@geindex AI-0098 (Ada 2012 feature)
26497
@emph{AI-0098 Anonymous subprogram access restrictions (0000-00-00)}
26499
An unintentional omission in the RM implied some inconsistent restrictions on
26500
the use of anonymous access to subprogram values. These restrictions were not
26501
intentional, and have never been enforced by GNAT.
26503
RM References: 3.10.01 (6) 3.10.01 (9.2/2)
26506
@geindex AI-0199 (Ada 2012 feature)
26512
@emph{AI-0199 Aggregate with anonymous access components (2010-07-14)}
26514
A choice list in a record aggregate can include several components of
26515
(distinct) anonymous access types as long as they have matching designated
26518
RM References: 4.03.01 (16)
26521
@geindex AI-0220 (Ada 2012 feature)
26527
@emph{AI-0220 Needed components for aggregates (0000-00-00)}
26529
This AI addresses a wording problem in the RM that appears to permit some
26530
complex cases of aggregates with nonstatic discriminants. GNAT has always
26531
implemented the intended semantics.
26533
RM References: 4.03.01 (17)
26536
@geindex AI-0147 (Ada 2012 feature)
26542
@emph{AI-0147 Conditional expressions (2009-03-29)}
26544
Conditional expressions are permitted. The form of such an expression is:
26547
(if expr then expr @{elsif expr then expr@} [else expr])
26550
The parentheses can be omitted in contexts where parentheses are present
26551
anyway, such as subprogram arguments and pragma arguments. If the @strong{else}
26552
clause is omitted, @strong{else} @emph{True} is assumed;
26553
thus @code{(if A then B)} is a way to conveniently represent
26554
@emph{(A implies B)} in standard logic.
26556
RM References: 4.03.03 (15) 4.04 (1) 4.04 (7) 4.05.07 (0) 4.07 (2)
26557
4.07 (3) 4.09 (12) 4.09 (33) 5.03 (3) 5.03 (4) 7.05 (2.1/2)
26560
@geindex AI-0037 (Ada 2012 feature)
26566
@emph{AI-0037 Out-of-range box associations in aggregate (0000-00-00)}
26568
This AI confirms that an association of the form @cite{Indx => <>} in an
26569
array aggregate must raise @cite{Constraint_Error} if @cite{Indx}
26570
is out of range. The RM specified a range check on other associations, but
26571
not when the value of the association was defaulted. GNAT has always inserted
26572
a constraint check on the index value.
26574
RM References: 4.03.03 (29)
26577
@geindex AI-0123 (Ada 2012 feature)
26583
@emph{AI-0123 Composability of equality (2010-04-13)}
26585
Equality of untagged record composes, so that the predefined equality for a
26586
composite type that includes a component of some untagged record type
26587
@cite{R} uses the equality operation of @cite{R} (which may be user-defined
26588
or predefined). This makes the behavior of untagged records identical to that
26589
of tagged types in this respect.
26591
This change is an incompatibility with previous versions of Ada, but it
26592
corrects a non-uniformity that was often a source of confusion. Analysis of
26593
a large number of industrial programs indicates that in those rare cases
26594
where a composite type had an untagged record component with a user-defined
26595
equality, either there was no use of the composite equality, or else the code
26596
expected the same composability as for tagged types, and thus had a bug that
26597
would be fixed by this change.
26599
RM References: 4.05.02 (9.7/2) 4.05.02 (14) 4.05.02 (15) 4.05.02 (24)
26603
@geindex AI-0088 (Ada 2012 feature)
26609
@emph{AI-0088 The value of exponentiation (0000-00-00)}
26611
This AI clarifies the equivalence rule given for the dynamic semantics of
26612
exponentiation: the value of the operation can be obtained by repeated
26613
multiplication, but the operation can be implemented otherwise (for example
26614
using the familiar divide-by-two-and-square algorithm, even if this is less
26615
accurate), and does not imply repeated reads of a volatile base.
26617
RM References: 4.05.06 (11)
26620
@geindex AI-0188 (Ada 2012 feature)
26626
@emph{AI-0188 Case expressions (2010-01-09)}
26628
Case expressions are permitted. This allows use of constructs such as:
26631
X := (case Y is when 1 => 2, when 2 => 3, when others => 31)
26634
RM References: 4.05.07 (0) 4.05.08 (0) 4.09 (12) 4.09 (33)
26637
@geindex AI-0104 (Ada 2012 feature)
26643
@emph{AI-0104 Null exclusion and uninitialized allocator (2010-07-15)}
26645
The assignment @code{Ptr := new not null Some_Ptr;} will raise
26646
@code{Constraint_Error} because the default value of the allocated object is
26647
@strong{null}. This useless construct is illegal in Ada 2012.
26649
RM References: 4.08 (2)
26652
@geindex AI-0157 (Ada 2012 feature)
26658
@emph{AI-0157 Allocation/Deallocation from empty pool (2010-07-11)}
26660
Allocation and Deallocation from an empty storage pool (i.e. allocation or
26661
deallocation of a pointer for which a static storage size clause of zero
26662
has been given) is now illegal and is detected as such. GNAT
26663
previously gave a warning but not an error.
26665
RM References: 4.08 (5.3/2) 13.11.02 (4) 13.11.02 (17)
26668
@geindex AI-0179 (Ada 2012 feature)
26674
@emph{AI-0179 Statement not required after label (2010-04-10)}
26676
It is not necessary to have a statement following a label, so a label
26677
can appear at the end of a statement sequence without the need for putting a
26678
null statement afterwards, but it is not allowable to have only labels and
26679
no real statements in a statement sequence.
26681
RM References: 5.01 (2)
26684
@geindex AI-0139-2 (Ada 2012 feature)
26690
@emph{AI-0139-2 Syntactic sugar for iterators (2010-09-29)}
26692
The new syntax for iterating over arrays and containers is now implemented.
26693
Iteration over containers is for now limited to read-only iterators. Only
26694
default iterators are supported, with the syntax: @cite{for Elem of C}.
26696
RM References: 5.05
26699
@geindex AI-0134 (Ada 2012 feature)
26705
@emph{AI-0134 Profiles must match for full conformance (0000-00-00)}
26707
For full conformance, the profiles of anonymous-access-to-subprogram
26708
parameters must match. GNAT has always enforced this rule.
26710
RM References: 6.03.01 (18)
26713
@geindex AI-0207 (Ada 2012 feature)
26719
@emph{AI-0207 Mode conformance and access constant (0000-00-00)}
26721
This AI confirms that access_to_constant indication must match for mode
26722
conformance. This was implemented in GNAT when the qualifier was originally
26723
introduced in Ada 2005.
26725
RM References: 6.03.01 (16/2)
26728
@geindex AI-0046 (Ada 2012 feature)
26734
@emph{AI-0046 Null exclusion match for full conformance (2010-07-17)}
26736
For full conformance, in the case of access parameters, the null exclusion
26737
must match (either both or neither must have @code{not null}).
26739
RM References: 6.03.02 (18)
26742
@geindex AI-0118 (Ada 2012 feature)
26748
@emph{AI-0118 The association of parameter associations (0000-00-00)}
26750
This AI clarifies the rules for named associations in subprogram calls and
26751
generic instantiations. The rules have been in place since Ada 83.
26753
RM References: 6.04.01 (2) 12.03 (9)
26756
@geindex AI-0196 (Ada 2012 feature)
26762
@emph{AI-0196 Null exclusion tests for out parameters (0000-00-00)}
26764
Null exclusion checks are not made for @cite{**out**} parameters when
26765
evaluating the actual parameters. GNAT has never generated these checks.
26767
RM References: 6.04.01 (13)
26770
@geindex AI-0015 (Ada 2012 feature)
26776
@emph{AI-0015 Constant return objects (0000-00-00)}
26778
The return object declared in an @emph{extended_return_statement} may be
26779
declared constant. This was always intended, and GNAT has always allowed it.
26781
RM References: 6.05 (2.1/2) 3.03 (10/2) 3.03 (21) 6.05 (5/2)
26785
@geindex AI-0032 (Ada 2012 feature)
26791
@emph{AI-0032 Extended return for class-wide functions (0000-00-00)}
26793
If a function returns a class-wide type, the object of an extended return
26794
statement can be declared with a specific type that is covered by the class-
26795
wide type. This has been implemented in GNAT since the introduction of
26796
extended returns. Note AI-0103 complements this AI by imposing matching
26797
rules for constrained return types.
26799
RM References: 6.05 (5.2/2) 6.05 (5.3/2) 6.05 (5.6/2) 6.05 (5.8/2)
26803
@geindex AI-0103 (Ada 2012 feature)
26809
@emph{AI-0103 Static matching for extended return (2010-07-23)}
26811
If the return subtype of a function is an elementary type or a constrained
26812
type, the subtype indication in an extended return statement must match
26813
statically this return subtype.
26815
RM References: 6.05 (5.2/2)
26818
@geindex AI-0058 (Ada 2012 feature)
26824
@emph{AI-0058 Abnormal completion of an extended return (0000-00-00)}
26826
The RM had some incorrect wording implying wrong treatment of abnormal
26827
completion in an extended return. GNAT has always implemented the intended
26828
correct semantics as described by this AI.
26830
RM References: 6.05 (22/2)
26833
@geindex AI-0050 (Ada 2012 feature)
26839
@emph{AI-0050 Raising Constraint_Error early for function call (0000-00-00)}
26841
The implementation permissions for raising @cite{Constraint_Error} early on a function call
26842
when it was clear an exception would be raised were over-permissive and allowed
26843
mishandling of discriminants in some cases. GNAT did
26844
not take advantage of these incorrect permissions in any case.
26846
RM References: 6.05 (24/2)
26849
@geindex AI-0125 (Ada 2012 feature)
26855
@emph{AI-0125 Nonoverridable operations of an ancestor (2010-09-28)}
26857
In Ada 2012, the declaration of a primitive operation of a type extension
26858
or private extension can also override an inherited primitive that is not
26859
visible at the point of this declaration.
26861
RM References: 7.03.01 (6) 8.03 (23) 8.03.01 (5/2) 8.03.01 (6/2)
26864
@geindex AI-0062 (Ada 2012 feature)
26870
@emph{AI-0062 Null exclusions and deferred constants (0000-00-00)}
26872
A full constant may have a null exclusion even if its associated deferred
26873
constant does not. GNAT has always allowed this.
26875
RM References: 7.04 (6/2) 7.04 (7.1/2)
26878
@geindex AI-0178 (Ada 2012 feature)
26884
@emph{AI-0178 Incomplete views are limited (0000-00-00)}
26886
This AI clarifies the role of incomplete views and plugs an omission in the
26887
RM. GNAT always correctly restricted the use of incomplete views and types.
26889
RM References: 7.05 (3/2) 7.05 (6/2)
26892
@geindex AI-0087 (Ada 2012 feature)
26898
@emph{AI-0087 Actual for formal nonlimited derived type (2010-07-15)}
26900
The actual for a formal nonlimited derived type cannot be limited. In
26901
particular, a formal derived type that extends a limited interface but which
26902
is not explicitly limited cannot be instantiated with a limited type.
26904
RM References: 7.05 (5/2) 12.05.01 (5.1/2)
26907
@geindex AI-0099 (Ada 2012 feature)
26913
@emph{AI-0099 Tag determines whether finalization needed (0000-00-00)}
26915
This AI clarifies that 'needs finalization' is part of dynamic semantics,
26916
and therefore depends on the run-time characteristics of an object (i.e. its
26917
tag) and not on its nominal type. As the AI indicates: "we do not expect
26918
this to affect any implementation'@w{'}.
26920
RM References: 7.06.01 (6) 7.06.01 (7) 7.06.01 (8) 7.06.01 (9/2)
26923
@geindex AI-0064 (Ada 2012 feature)
26929
@emph{AI-0064 Redundant finalization rule (0000-00-00)}
26931
This is an editorial change only. The intended behavior is already checked
26932
by an existing ACATS test, which GNAT has always executed correctly.
26934
RM References: 7.06.01 (17.1/1)
26937
@geindex AI-0026 (Ada 2012 feature)
26943
@emph{AI-0026 Missing rules for Unchecked_Union (2010-07-07)}
26945
Record representation clauses concerning Unchecked_Union types cannot mention
26946
the discriminant of the type. The type of a component declared in the variant
26947
part of an Unchecked_Union cannot be controlled, have controlled components,
26948
nor have protected or task parts. If an Unchecked_Union type is declared
26949
within the body of a generic unit or its descendants, then the type of a
26950
component declared in the variant part cannot be a formal private type or a
26951
formal private extension declared within the same generic unit.
26953
RM References: 7.06 (9.4/2) B.03.03 (9/2) B.03.03 (10/2)
26956
@geindex AI-0205 (Ada 2012 feature)
26962
@emph{AI-0205 Extended return declares visible name (0000-00-00)}
26964
This AI corrects a simple omission in the RM. Return objects have always
26965
been visible within an extended return statement.
26967
RM References: 8.03 (17)
26970
@geindex AI-0042 (Ada 2012 feature)
26976
@emph{AI-0042 Overriding versus implemented-by (0000-00-00)}
26978
This AI fixes a wording gap in the RM. An operation of a synchronized
26979
interface can be implemented by a protected or task entry, but the abstract
26980
operation is not being overridden in the usual sense, and it must be stated
26981
separately that this implementation is legal. This has always been the case
26984
RM References: 9.01 (9.2/2) 9.04 (11.1/2)
26987
@geindex AI-0030 (Ada 2012 feature)
26993
@emph{AI-0030 Requeue on synchronized interfaces (2010-07-19)}
26995
Requeue is permitted to a protected, synchronized or task interface primitive
26996
providing it is known that the overriding operation is an entry. Otherwise
26997
the requeue statement has the same effect as a procedure call. Use of pragma
26998
@cite{Implemented} provides a way to impose a static requirement on the
26999
overriding operation by adhering to one of the implementation kinds: entry,
27000
protected procedure or any of the above.
27002
RM References: 9.05 (9) 9.05.04 (2) 9.05.04 (3) 9.05.04 (5)
27003
9.05.04 (6) 9.05.04 (7) 9.05.04 (12)
27006
@geindex AI-0201 (Ada 2012 feature)
27012
@emph{AI-0201 Independence of atomic object components (2010-07-22)}
27014
If an Atomic object has a pragma @cite{Pack} or a @cite{Component_Size}
27015
attribute, then individual components may not be addressable by independent
27016
tasks. However, if the representation clause has no effect (is confirming),
27017
then independence is not compromised. Furthermore, in GNAT, specification of
27018
other appropriately addressable component sizes (e.g. 16 for 8-bit
27019
characters) also preserves independence. GNAT now gives very clear warnings
27020
both for the declaration of such a type, and for any assignment to its components.
27022
RM References: 9.10 (1/3) C.06 (22/2) C.06 (23/2)
27025
@geindex AI-0009 (Ada 2012 feature)
27031
@emph{AI-0009 Pragma Independent[_Components] (2010-07-23)}
27033
This AI introduces the new pragmas @cite{Independent} and
27034
@cite{Independent_Components},
27035
which control guaranteeing independence of access to objects and components.
27036
The AI also requires independence not unaffected by confirming rep clauses.
27038
RM References: 9.10 (1) 13.01 (15/1) 13.02 (9) 13.03 (13) C.06 (2)
27039
C.06 (4) C.06 (6) C.06 (9) C.06 (13) C.06 (14)
27042
@geindex AI-0072 (Ada 2012 feature)
27048
@emph{AI-0072 Task signalling using 'Terminated (0000-00-00)}
27050
This AI clarifies that task signalling for reading @cite{'Terminated} only
27051
occurs if the result is True. GNAT semantics has always been consistent with
27052
this notion of task signalling.
27054
RM References: 9.10 (6.1/1)
27057
@geindex AI-0108 (Ada 2012 feature)
27063
@emph{AI-0108 Limited incomplete view and discriminants (0000-00-00)}
27065
This AI confirms that an incomplete type from a limited view does not have
27066
discriminants. This has always been the case in GNAT.
27068
RM References: 10.01.01 (12.3/2)
27071
@geindex AI-0129 (Ada 2012 feature)
27077
@emph{AI-0129 Limited views and incomplete types (0000-00-00)}
27079
This AI clarifies the description of limited views: a limited view of a
27080
package includes only one view of a type that has an incomplete declaration
27081
and a full declaration (there is no possible ambiguity in a client package).
27082
This AI also fixes an omission: a nested package in the private part has no
27083
limited view. GNAT always implemented this correctly.
27085
RM References: 10.01.01 (12.2/2) 10.01.01 (12.3/2)
27088
@geindex AI-0077 (Ada 2012 feature)
27094
@emph{AI-0077 Limited withs and scope of declarations (0000-00-00)}
27096
This AI clarifies that a declaration does not include a context clause,
27097
and confirms that it is illegal to have a context in which both a limited
27098
and a nonlimited view of a package are accessible. Such double visibility
27099
was always rejected by GNAT.
27101
RM References: 10.01.02 (12/2) 10.01.02 (21/2) 10.01.02 (22/2)
27104
@geindex AI-0122 (Ada 2012 feature)
27110
@emph{AI-0122 Private with and children of generics (0000-00-00)}
27112
This AI clarifies the visibility of private children of generic units within
27113
instantiations of a parent. GNAT has always handled this correctly.
27115
RM References: 10.01.02 (12/2)
27118
@geindex AI-0040 (Ada 2012 feature)
27124
@emph{AI-0040 Limited with clauses on descendant (0000-00-00)}
27126
This AI confirms that a limited with clause in a child unit cannot name
27127
an ancestor of the unit. This has always been checked in GNAT.
27129
RM References: 10.01.02 (20/2)
27132
@geindex AI-0132 (Ada 2012 feature)
27138
@emph{AI-0132 Placement of library unit pragmas (0000-00-00)}
27140
This AI fills a gap in the description of library unit pragmas. The pragma
27141
clearly must apply to a library unit, even if it does not carry the name
27142
of the enclosing unit. GNAT has always enforced the required check.
27144
RM References: 10.01.05 (7)
27147
@geindex AI-0034 (Ada 2012 feature)
27153
@emph{AI-0034 Categorization of limited views (0000-00-00)}
27155
The RM makes certain limited with clauses illegal because of categorization
27156
considerations, when the corresponding normal with would be legal. This is
27157
not intended, and GNAT has always implemented the recommended behavior.
27159
RM References: 10.02.01 (11/1) 10.02.01 (17/2)
27162
@geindex AI-0035 (Ada 2012 feature)
27168
@emph{AI-0035 Inconsistencies with Pure units (0000-00-00)}
27170
This AI remedies some inconsistencies in the legality rules for Pure units.
27171
Derived access types are legal in a pure unit (on the assumption that the
27172
rule for a zero storage pool size has been enforced on the ancestor type).
27173
The rules are enforced in generic instances and in subunits. GNAT has always
27174
implemented the recommended behavior.
27176
RM References: 10.02.01 (15.1/2) 10.02.01 (15.4/2) 10.02.01 (15.5/2) 10.02.01 (17/2)
27179
@geindex AI-0219 (Ada 2012 feature)
27185
@emph{AI-0219 Pure permissions and limited parameters (2010-05-25)}
27187
This AI refines the rules for the cases with limited parameters which do not
27188
allow the implementations to omit 'redundant'. GNAT now properly conforms
27189
to the requirements of this binding interpretation.
27191
RM References: 10.02.01 (18/2)
27194
@geindex AI-0043 (Ada 2012 feature)
27200
@emph{AI-0043 Rules about raising exceptions (0000-00-00)}
27202
This AI covers various omissions in the RM regarding the raising of
27203
exceptions. GNAT has always implemented the intended semantics.
27205
RM References: 11.04.01 (10.1/2) 11 (2)
27208
@geindex AI-0200 (Ada 2012 feature)
27214
@emph{AI-0200 Mismatches in formal package declarations (0000-00-00)}
27216
This AI plugs a gap in the RM which appeared to allow some obviously intended
27217
illegal instantiations. GNAT has never allowed these instantiations.
27219
RM References: 12.07 (16)
27222
@geindex AI-0112 (Ada 2012 feature)
27228
@emph{AI-0112 Detection of duplicate pragmas (2010-07-24)}
27230
This AI concerns giving names to various representation aspects, but the
27231
practical effect is simply to make the use of duplicate
27232
@cite{Atomic[_Components]},
27233
@cite{Volatile[_Components]}, and
27234
@cite{Independent[_Components]} pragmas illegal, and GNAT
27235
now performs this required check.
27237
RM References: 13.01 (8)
27240
@geindex AI-0106 (Ada 2012 feature)
27246
@emph{AI-0106 No representation pragmas on generic formals (0000-00-00)}
27248
The RM appeared to allow representation pragmas on generic formal parameters,
27249
but this was not intended, and GNAT has never permitted this usage.
27251
RM References: 13.01 (9.1/1)
27254
@geindex AI-0012 (Ada 2012 feature)
27260
@emph{AI-0012 Pack/Component_Size for aliased/atomic (2010-07-15)}
27262
It is now illegal to give an inappropriate component size or a pragma
27263
@cite{Pack} that attempts to change the component size in the case of atomic
27264
or aliased components. Previously GNAT ignored such an attempt with a
27267
RM References: 13.02 (6.1/2) 13.02 (7) C.06 (10) C.06 (11) C.06 (21)
27270
@geindex AI-0039 (Ada 2012 feature)
27276
@emph{AI-0039 Stream attributes cannot be dynamic (0000-00-00)}
27278
The RM permitted the use of dynamic expressions (such as @code{ptr.all})`
27279
for stream attributes, but these were never useful and are now illegal. GNAT
27280
has always regarded such expressions as illegal.
27282
RM References: 13.03 (4) 13.03 (6) 13.13.02 (38/2)
27285
@geindex AI-0095 (Ada 2012 feature)
27291
@emph{AI-0095 Address of intrinsic subprograms (0000-00-00)}
27293
The prefix of @cite{'Address} cannot statically denote a subprogram with
27294
convention @cite{Intrinsic}. The use of the @cite{Address} attribute raises
27295
@cite{Program_Error} if the prefix denotes a subprogram with convention
27298
RM References: 13.03 (11/1)
27301
@geindex AI-0116 (Ada 2012 feature)
27307
@emph{AI-0116 Alignment of class-wide objects (0000-00-00)}
27309
This AI requires that the alignment of a class-wide object be no greater
27310
than the alignment of any type in the class. GNAT has always followed this
27313
RM References: 13.03 (29) 13.11 (16)
27316
@geindex AI-0146 (Ada 2012 feature)
27322
@emph{AI-0146 Type invariants (2009-09-21)}
27324
Type invariants may be specified for private types using the aspect notation.
27325
Aspect @cite{Type_Invariant} may be specified for any private type,
27326
@cite{Type_Invariant'Class} can
27327
only be specified for tagged types, and is inherited by any descendent of the
27328
tagged types. The invariant is a boolean expression that is tested for being
27329
true in the following situations: conversions to the private type, object
27330
declarations for the private type that are default initialized, and
27331
[@strong{in}] @strong{out}
27332
parameters and returned result on return from any primitive operation for
27333
the type that is visible to a client.
27334
GNAT defines the synonyms @cite{Invariant} for @cite{Type_Invariant} and
27335
@cite{Invariant'Class} for @cite{Type_Invariant'Class}.
27337
RM References: 13.03.03 (00)
27340
@geindex AI-0078 (Ada 2012 feature)
27346
@emph{AI-0078 Relax Unchecked_Conversion alignment rules (0000-00-00)}
27348
In Ada 2012, compilers are required to support unchecked conversion where the
27349
target alignment is a multiple of the source alignment. GNAT always supported
27350
this case (and indeed all cases of differing alignments, doing copies where
27351
required if the alignment was reduced).
27353
RM References: 13.09 (7)
27356
@geindex AI-0195 (Ada 2012 feature)
27362
@emph{AI-0195 Invalid value handling is implementation defined (2010-07-03)}
27364
The handling of invalid values is now designated to be implementation
27365
defined. This is a documentation change only, requiring Annex M in the GNAT
27366
Reference Manual to document this handling.
27367
In GNAT, checks for invalid values are made
27368
only when necessary to avoid erroneous behavior. Operations like assignments
27369
which cannot cause erroneous behavior ignore the possibility of invalid
27370
values and do not do a check. The date given above applies only to the
27371
documentation change, this behavior has always been implemented by GNAT.
27373
RM References: 13.09.01 (10)
27376
@geindex AI-0193 (Ada 2012 feature)
27382
@emph{AI-0193 Alignment of allocators (2010-09-16)}
27384
This AI introduces a new attribute @cite{Max_Alignment_For_Allocation},
27385
analogous to @cite{Max_Size_In_Storage_Elements}, but for alignment instead
27388
RM References: 13.11 (16) 13.11 (21) 13.11.01 (0) 13.11.01 (1)
27389
13.11.01 (2) 13.11.01 (3)
27392
@geindex AI-0177 (Ada 2012 feature)
27398
@emph{AI-0177 Parameterized expressions (2010-07-10)}
27400
The new Ada 2012 notion of parameterized expressions is implemented. The form
27404
function-specification is (expression)
27407
This is exactly equivalent to the
27408
corresponding function body that returns the expression, but it can appear
27409
in a package spec. Note that the expression must be parenthesized.
27411
RM References: 13.11.01 (3/2)
27414
@geindex AI-0033 (Ada 2012 feature)
27420
@emph{AI-0033 Attach/Interrupt_Handler in generic (2010-07-24)}
27422
Neither of these two pragmas may appear within a generic template, because
27423
the generic might be instantiated at other than the library level.
27425
RM References: 13.11.02 (16) C.03.01 (7/2) C.03.01 (8/2)
27428
@geindex AI-0161 (Ada 2012 feature)
27434
@emph{AI-0161 Restriction No_Default_Stream_Attributes (2010-09-11)}
27436
A new restriction @cite{No_Default_Stream_Attributes} prevents the use of any
27437
of the default stream attributes for elementary types. If this restriction is
27438
in force, then it is necessary to provide explicit subprograms for any
27439
stream attributes used.
27441
RM References: 13.12.01 (4/2) 13.13.02 (40/2) 13.13.02 (52/2)
27444
@geindex AI-0194 (Ada 2012 feature)
27450
@emph{AI-0194 Value of Stream_Size attribute (0000-00-00)}
27452
The @cite{Stream_Size} attribute returns the default number of bits in the
27453
stream representation of the given type.
27454
This value is not affected by the presence
27455
of stream subprogram attributes for the type. GNAT has always implemented
27456
this interpretation.
27458
RM References: 13.13.02 (1.2/2)
27461
@geindex AI-0109 (Ada 2012 feature)
27467
@emph{AI-0109 Redundant check in S'Class'Input (0000-00-00)}
27469
This AI is an editorial change only. It removes the need for a tag check
27470
that can never fail.
27472
RM References: 13.13.02 (34/2)
27475
@geindex AI-0007 (Ada 2012 feature)
27481
@emph{AI-0007 Stream read and private scalar types (0000-00-00)}
27483
The RM as written appeared to limit the possibilities of declaring read
27484
attribute procedures for private scalar types. This limitation was not
27485
intended, and has never been enforced by GNAT.
27487
RM References: 13.13.02 (50/2) 13.13.02 (51/2)
27490
@geindex AI-0065 (Ada 2012 feature)
27496
@emph{AI-0065 Remote access types and external streaming (0000-00-00)}
27498
This AI clarifies the fact that all remote access types support external
27499
streaming. This fixes an obvious oversight in the definition of the
27500
language, and GNAT always implemented the intended correct rules.
27502
RM References: 13.13.02 (52/2)
27505
@geindex AI-0019 (Ada 2012 feature)
27511
@emph{AI-0019 Freezing of primitives for tagged types (0000-00-00)}
27513
The RM suggests that primitive subprograms of a specific tagged type are
27514
frozen when the tagged type is frozen. This would be an incompatible change
27515
and is not intended. GNAT has never attempted this kind of freezing and its
27516
behavior is consistent with the recommendation of this AI.
27518
RM References: 13.14 (2) 13.14 (3/1) 13.14 (8.1/1) 13.14 (10) 13.14 (14) 13.14 (15.1/2)
27521
@geindex AI-0017 (Ada 2012 feature)
27527
@emph{AI-0017 Freezing and incomplete types (0000-00-00)}
27529
So-called 'Taft-amendment types' (i.e., types that are completed in package
27530
bodies) are not frozen by the occurrence of bodies in the
27531
enclosing declarative part. GNAT always implemented this properly.
27533
RM References: 13.14 (3/1)
27536
@geindex AI-0060 (Ada 2012 feature)
27542
@emph{AI-0060 Extended definition of remote access types (0000-00-00)}
27544
This AI extends the definition of remote access types to include access
27545
to limited, synchronized, protected or task class-wide interface types.
27546
GNAT already implemented this extension.
27548
RM References: A (4) E.02.02 (9/1) E.02.02 (9.2/1) E.02.02 (14/2) E.02.02 (18)
27551
@geindex AI-0114 (Ada 2012 feature)
27557
@emph{AI-0114 Classification of letters (0000-00-00)}
27559
The code points 170 (@cite{FEMININE ORDINAL INDICATOR}),
27560
181 (@cite{MICRO SIGN}), and
27561
186 (@cite{MASCULINE ORDINAL INDICATOR}) are technically considered
27562
lower case letters by Unicode.
27563
However, they are not allowed in identifiers, and they
27564
return @cite{False} to @cite{Ada.Characters.Handling.Is_Letter/Is_Lower}.
27565
This behavior is consistent with that defined in Ada 95.
27567
RM References: A.03.02 (59) A.04.06 (7)
27570
@geindex AI-0185 (Ada 2012 feature)
27576
@emph{AI-0185 Ada.Wide_[Wide_]Characters.Handling (2010-07-06)}
27578
Two new packages @cite{Ada.Wide_[Wide_]Characters.Handling} provide
27579
classification functions for @cite{Wide_Character} and
27580
@cite{Wide_Wide_Character}, as well as providing
27581
case folding routines for @cite{Wide_[Wide_]Character} and
27582
@cite{Wide_[Wide_]String}.
27584
RM References: A.03.05 (0) A.03.06 (0)
27587
@geindex AI-0031 (Ada 2012 feature)
27593
@emph{AI-0031 Add From parameter to Find_Token (2010-07-25)}
27595
A new version of @cite{Find_Token} is added to all relevant string packages,
27596
with an extra parameter @cite{From}. Instead of starting at the first
27597
character of the string, the search for a matching Token starts at the
27598
character indexed by the value of @cite{From}.
27599
These procedures are available in all versions of Ada
27600
but if used in versions earlier than Ada 2012 they will generate a warning
27601
that an Ada 2012 subprogram is being used.
27603
RM References: A.04.03 (16) A.04.03 (67) A.04.03 (68/1) A.04.04 (51)
27607
@geindex AI-0056 (Ada 2012 feature)
27613
@emph{AI-0056 Index on null string returns zero (0000-00-00)}
27615
The wording in the Ada 2005 RM implied an incompatible handling of the
27616
@cite{Index} functions, resulting in raising an exception instead of
27617
returning zero in some situations.
27618
This was not intended and has been corrected.
27619
GNAT always returned zero, and is thus consistent with this AI.
27621
RM References: A.04.03 (56.2/2) A.04.03 (58.5/2)
27624
@geindex AI-0137 (Ada 2012 feature)
27630
@emph{AI-0137 String encoding package (2010-03-25)}
27632
The packages @cite{Ada.Strings.UTF_Encoding}, together with its child
27633
packages, @cite{Conversions}, @cite{Strings}, @cite{Wide_Strings},
27634
and @cite{Wide_Wide_Strings} have been
27635
implemented. These packages (whose documentation can be found in the spec
27636
files @code{a-stuten.ads}, @code{a-suenco.ads}, @code{a-suenst.ads},
27637
@code{a-suewst.ads}, @code{a-suezst.ads}) allow encoding and decoding of
27638
@cite{String}, @cite{Wide_String}, and @cite{Wide_Wide_String}
27639
values using UTF coding schemes (including UTF-8, UTF-16LE, UTF-16BE, and
27640
UTF-16), as well as conversions between the different UTF encodings. With
27641
the exception of @cite{Wide_Wide_Strings}, these packages are available in
27642
Ada 95 and Ada 2005 mode as well as Ada 2012 mode.
27643
The @cite{Wide_Wide_Strings package}
27644
is available in Ada 2005 mode as well as Ada 2012 mode (but not in Ada 95
27645
mode since it uses @cite{Wide_Wide_Character}).
27647
RM References: A.04.11
27650
@geindex AI-0038 (Ada 2012 feature)
27656
@emph{AI-0038 Minor errors in Text_IO (0000-00-00)}
27658
These are minor errors in the description on three points. The intent on
27659
all these points has always been clear, and GNAT has always implemented the
27660
correct intended semantics.
27662
RM References: A.10.05 (37) A.10.07 (8/1) A.10.07 (10) A.10.07 (12) A.10.08 (10) A.10.08 (24)
27665
@geindex AI-0044 (Ada 2012 feature)
27671
@emph{AI-0044 Restrictions on container instantiations (0000-00-00)}
27673
This AI places restrictions on allowed instantiations of generic containers.
27674
These restrictions are not checked by the compiler, so there is nothing to
27675
change in the implementation. This affects only the RM documentation.
27677
RM References: A.18 (4/2) A.18.02 (231/2) A.18.03 (145/2) A.18.06 (56/2) A.18.08 (66/2) A.18.09 (79/2) A.18.26 (5/2) A.18.26 (9/2)
27680
@geindex AI-0127 (Ada 2012 feature)
27686
@emph{AI-0127 Adding Locale Capabilities (2010-09-29)}
27688
This package provides an interface for identifying the current locale.
27690
RM References: A.19 A.19.01 A.19.02 A.19.03 A.19.05 A.19.06
27691
A.19.07 A.19.08 A.19.09 A.19.10 A.19.11 A.19.12 A.19.13
27694
@geindex AI-0002 (Ada 2012 feature)
27700
@emph{AI-0002 Export C with unconstrained arrays (0000-00-00)}
27702
The compiler is not required to support exporting an Ada subprogram with
27703
convention C if there are parameters or a return type of an unconstrained
27704
array type (such as @cite{String}). GNAT allows such declarations but
27705
generates warnings. It is possible, but complicated, to write the
27706
corresponding C code and certainly such code would be specific to GNAT and
27709
RM References: B.01 (17) B.03 (62) B.03 (71.1/2)
27712
@geindex AI05-0216 (Ada 2012 feature)
27718
@emph{AI-0216 No_Task_Hierarchy forbids local tasks (0000-00-00)}
27720
It is clearly the intention that @cite{No_Task_Hierarchy} is intended to
27721
forbid tasks declared locally within subprograms, or functions returning task
27722
objects, and that is the implementation that GNAT has always provided.
27723
However the language in the RM was not sufficiently clear on this point.
27724
Thus this is a documentation change in the RM only.
27726
RM References: D.07 (3/3)
27729
@geindex AI-0211 (Ada 2012 feature)
27735
@emph{AI-0211 No_Relative_Delays forbids Set_Handler use (2010-07-09)}
27737
The restriction @cite{No_Relative_Delays} forbids any calls to the subprogram
27738
@cite{Ada.Real_Time.Timing_Events.Set_Handler}.
27740
RM References: D.07 (5) D.07 (10/2) D.07 (10.4/2) D.07 (10.7/2)
27743
@geindex AI-0190 (Ada 2012 feature)
27749
@emph{AI-0190 pragma Default_Storage_Pool (2010-09-15)}
27751
This AI introduces a new pragma @cite{Default_Storage_Pool}, which can be
27752
used to control storage pools globally.
27753
In particular, you can force every access
27754
type that is used for allocation (@strong{new}) to have an explicit storage pool,
27755
or you can declare a pool globally to be used for all access types that lack
27758
RM References: D.07 (8)
27761
@geindex AI-0189 (Ada 2012 feature)
27767
@emph{AI-0189 No_Allocators_After_Elaboration (2010-01-23)}
27769
This AI introduces a new restriction @cite{No_Allocators_After_Elaboration},
27770
which says that no dynamic allocation will occur once elaboration is
27772
In general this requires a run-time check, which is not required, and which
27773
GNAT does not attempt. But the static cases of allocators in a task body or
27774
in the body of the main program are detected and flagged at compile or bind
27777
RM References: D.07 (19.1/2) H.04 (23.3/2)
27780
@geindex AI-0171 (Ada 2012 feature)
27786
@emph{AI-0171 Pragma CPU and Ravenscar Profile (2010-09-24)}
27788
A new package @cite{System.Multiprocessors} is added, together with the
27789
definition of pragma @cite{CPU} for controlling task affinity. A new no
27790
dependence restriction, on @cite{System.Multiprocessors.Dispatching_Domains},
27791
is added to the Ravenscar profile.
27793
RM References: D.13.01 (4/2) D.16
27796
@geindex AI-0210 (Ada 2012 feature)
27802
@emph{AI-0210 Correct Timing_Events metric (0000-00-00)}
27804
This is a documentation only issue regarding wording of metric requirements,
27805
that does not affect the implementation of the compiler.
27807
RM References: D.15 (24/2)
27810
@geindex AI-0206 (Ada 2012 feature)
27816
@emph{AI-0206 Remote types packages and preelaborate (2010-07-24)}
27818
Remote types packages are now allowed to depend on preelaborated packages.
27819
This was formerly considered illegal.
27821
RM References: E.02.02 (6)
27824
@geindex AI-0152 (Ada 2012 feature)
27830
@emph{AI-0152 Restriction No_Anonymous_Allocators (2010-09-08)}
27832
Restriction @cite{No_Anonymous_Allocators} prevents the use of allocators
27833
where the type of the returned value is an anonymous access type.
27835
RM References: H.04 (8/1)
27838
@node Obsolescent Features,Compatibility and Porting Guide,Implementation of Ada 2012 Features,Top
27839
@anchor{gnat_rm/obsolescent_features id1}@anchor{3dc}@anchor{gnat_rm/obsolescent_features doc}@anchor{3dd}@anchor{gnat_rm/obsolescent_features obsolescent-features}@anchor{15}
27840
@chapter Obsolescent Features
27843
This chapter describes features that are provided by GNAT, but are
27844
considered obsolescent since there are preferred ways of achieving
27845
the same effect. These features are provided solely for historical
27846
compatibility purposes.
27849
* pragma No_Run_Time::
27850
* pragma Ravenscar::
27851
* pragma Restricted_Run_Time::
27852
* pragma Task_Info::
27853
* package System.Task_Info (s-tasinf.ads): package System Task_Info s-tasinf ads.
27857
@node pragma No_Run_Time,pragma Ravenscar,,Obsolescent Features
27858
@anchor{gnat_rm/obsolescent_features id2}@anchor{3de}@anchor{gnat_rm/obsolescent_features pragma-no-run-time}@anchor{3df}
27859
@section pragma No_Run_Time
27862
The pragma @cite{No_Run_Time} is used to achieve an affect similar
27863
to the use of the "Zero Foot Print" configurable run time, but without
27864
requiring a specially configured run time. The result of using this
27865
pragma, which must be used for all units in a partition, is to restrict
27866
the use of any language features requiring run-time support code. The
27867
preferred usage is to use an appropriately configured run-time that
27868
includes just those features that are to be made accessible.
27870
@node pragma Ravenscar,pragma Restricted_Run_Time,pragma No_Run_Time,Obsolescent Features
27871
@anchor{gnat_rm/obsolescent_features id3}@anchor{3e0}@anchor{gnat_rm/obsolescent_features pragma-ravenscar}@anchor{3e1}
27872
@section pragma Ravenscar
27875
The pragma @cite{Ravenscar} has exactly the same effect as pragma
27876
@cite{Profile (Ravenscar)}. The latter usage is preferred since it
27877
is part of the new Ada 2005 standard.
27879
@node pragma Restricted_Run_Time,pragma Task_Info,pragma Ravenscar,Obsolescent Features
27880
@anchor{gnat_rm/obsolescent_features pragma-restricted-run-time}@anchor{3e2}@anchor{gnat_rm/obsolescent_features id4}@anchor{3e3}
27881
@section pragma Restricted_Run_Time
27884
The pragma @cite{Restricted_Run_Time} has exactly the same effect as
27885
pragma @cite{Profile (Restricted)}. The latter usage is
27886
preferred since the Ada 2005 pragma @cite{Profile} is intended for
27887
this kind of implementation dependent addition.
27889
@node pragma Task_Info,package System Task_Info s-tasinf ads,pragma Restricted_Run_Time,Obsolescent Features
27890
@anchor{gnat_rm/obsolescent_features pragma-task-info}@anchor{3e4}@anchor{gnat_rm/obsolescent_features id5}@anchor{3e5}
27891
@section pragma Task_Info
27894
The functionality provided by pragma @cite{Task_Info} is now part of the
27895
Ada language. The @cite{CPU} aspect and the package
27896
@cite{System.Multiprocessors} offer a less system-dependent way to specify
27897
task affinity or to query the number of processsors.
27902
pragma Task_Info (EXPRESSION);
27905
This pragma appears within a task definition (like pragma
27906
@cite{Priority}) and applies to the task in which it appears. The
27907
argument must be of type @cite{System.Task_Info.Task_Info_Type}.
27908
The @cite{Task_Info} pragma provides system dependent control over
27909
aspects of tasking implementation, for example, the ability to map
27910
tasks to specific processors. For details on the facilities available
27911
for the version of GNAT that you are using, see the documentation
27912
in the spec of package System.Task_Info in the runtime
27915
@node package System Task_Info s-tasinf ads,,pragma Task_Info,Obsolescent Features
27916
@anchor{gnat_rm/obsolescent_features package-system-task-info}@anchor{3e6}@anchor{gnat_rm/obsolescent_features package-system-task-info-s-tasinf-ads}@anchor{3e7}
27917
@section package System.Task_Info (@code{s-tasinf.ads})
27920
This package provides target dependent functionality that is used
27921
to support the @cite{Task_Info} pragma. The predefined Ada package
27922
@cite{System.Multiprocessors} and the @cite{CPU} aspect now provide a
27923
standard replacement for GNAT's @cite{Task_Info} functionality.
27925
@node Compatibility and Porting Guide,GNU Free Documentation License,Obsolescent Features,Top
27926
@anchor{gnat_rm/compatibility_and_porting_guide compatibility-and-porting-guide}@anchor{16}@anchor{gnat_rm/compatibility_and_porting_guide doc}@anchor{3e8}@anchor{gnat_rm/compatibility_and_porting_guide id1}@anchor{3e9}
27927
@chapter Compatibility and Porting Guide
27930
This chapter presents some guidelines for developing portable Ada code,
27931
describes the compatibility issues that may arise between
27932
GNAT and other Ada compilation systems (including those for Ada 83),
27933
and shows how GNAT can expedite porting
27934
applications developed in other Ada environments.
27937
* Writing Portable Fixed-Point Declarations::
27938
* Compatibility with Ada 83::
27939
* Compatibility between Ada 95 and Ada 2005::
27940
* Implementation-dependent characteristics::
27941
* Compatibility with Other Ada Systems::
27942
* Representation Clauses::
27943
* Compatibility with HP Ada 83::
27947
@node Writing Portable Fixed-Point Declarations,Compatibility with Ada 83,,Compatibility and Porting Guide
27948
@anchor{gnat_rm/compatibility_and_porting_guide id2}@anchor{3ea}@anchor{gnat_rm/compatibility_and_porting_guide writing-portable-fixed-point-declarations}@anchor{3eb}
27949
@section Writing Portable Fixed-Point Declarations
27952
The Ada Reference Manual gives an implementation freedom to choose bounds
27953
that are narrower by @cite{Small} from the given bounds.
27954
For example, if we write
27957
type F1 is delta 1.0 range -128.0 .. +128.0;
27960
then the implementation is allowed to choose -128.0 .. +127.0 if it
27961
likes, but is not required to do so.
27963
This leads to possible portability problems, so let's have a closer
27964
look at this, and figure out how to avoid these problems.
27966
First, why does this freedom exist, and why would an implementation
27967
take advantage of it? To answer this, take a closer look at the type
27968
declaration for @cite{F1} above. If the compiler uses the given bounds,
27969
it would need 9 bits to hold the largest positive value (and typically
27970
that means 16 bits on all machines). But if the implementation chooses
27971
the +127.0 bound then it can fit values of the type in 8 bits.
27973
Why not make the user write +127.0 if that's what is wanted?
27974
The rationale is that if you are thinking of fixed point
27975
as a kind of 'poor man's floating-point', then you don't want
27976
to be thinking about the scaled integers that are used in its
27977
representation. Let's take another example:
27980
type F2 is delta 2.0**(-15) range -1.0 .. +1.0;
27983
Looking at this declaration, it seems casually as though
27984
it should fit in 16 bits, but again that extra positive value
27985
+1.0 has the scaled integer equivalent of 2**15 which is one too
27986
big for signed 16 bits. The implementation can treat this as:
27989
type F2 is delta 2.0**(-15) range -1.0 .. +1.0-(2.0**(-15));
27992
and the Ada language design team felt that this was too annoying
27993
to require. We don't need to debate this decision at this point,
27994
since it is well established (the rule about narrowing the ranges
27997
But the important point is that an implementation is not required
27998
to do this narrowing, so we have a potential portability problem.
27999
We could imagine three types of implementation:
28005
those that narrow the range automatically if they can figure
28006
out that the narrower range will allow storage in a smaller machine unit,
28009
those that will narrow only if forced to by a @cite{'Size} clause, and
28012
those that will never narrow.
28015
Now if we are language theoreticians, we can imagine a fourth
28016
approach: to narrow all the time, e.g. to treat
28019
type F3 is delta 1.0 range -10.0 .. +23.0;
28022
as though it had been written:
28025
type F3 is delta 1.0 range -9.0 .. +22.0;
28028
But although technically allowed, such a behavior would be hostile and silly,
28029
and no real compiler would do this. All real compilers will fall into one of
28030
the categories (a), (b) or (c) above.
28032
So, how do you get the compiler to do what you want? The answer is give the
28033
actual bounds you want, and then use a @cite{'Small} clause and a
28034
@cite{'Size} clause to absolutely pin down what the compiler does.
28035
E.g., for @cite{F2} above, we will write:
28038
My_Small : constant := 2.0**(-15);
28039
My_First : constant := -1.0;
28040
My_Last : constant := +1.0 - My_Small;
28042
type F2 is delta My_Small range My_First .. My_Last;
28048
for F2'Small use my_Small;
28049
for F2'Size use 16;
28052
In practice all compilers will do the same thing here and will give you
28053
what you want, so the above declarations are fully portable. If you really
28054
want to play language lawyer and guard against ludicrous behavior by the
28055
compiler you could add
28058
Test1 : constant := 1 / Boolean'Pos (F2'First = My_First);
28059
Test2 : constant := 1 / Boolean'Pos (F2'Last = My_Last);
28062
One or other or both are allowed to be illegal if the compiler is
28063
behaving in a silly manner, but at least the silly compiler will not
28064
get away with silently messing with your (very clear) intentions.
28066
If you follow this scheme you will be guaranteed that your fixed-point
28067
types will be portable.
28069
@node Compatibility with Ada 83,Compatibility between Ada 95 and Ada 2005,Writing Portable Fixed-Point Declarations,Compatibility and Porting Guide
28070
@anchor{gnat_rm/compatibility_and_porting_guide compatibility-with-ada-83}@anchor{3ec}@anchor{gnat_rm/compatibility_and_porting_guide id3}@anchor{3ed}
28071
@section Compatibility with Ada 83
28074
@geindex Compatibility (between Ada 83 and Ada 95 / Ada 2005 / Ada 2012)
28076
Ada 95 and the subsequent revisions Ada 2005 and Ada 2012
28077
are highly upwards compatible with Ada 83. In
28078
particular, the design intention was that the difficulties associated
28079
with moving from Ada 83 to later versions of the standard should be no greater
28080
than those that occur when moving from one Ada 83 system to another.
28082
However, there are a number of points at which there are minor
28083
incompatibilities. The @cite{Ada 95 Annotated Reference Manual} contains
28084
full details of these issues as they relate to Ada 95,
28085
and should be consulted for a complete treatment.
28087
following subsections treat the most likely issues to be encountered.
28090
* Legal Ada 83 programs that are illegal in Ada 95::
28091
* More deterministic semantics::
28092
* Changed semantics::
28093
* Other language compatibility issues::
28097
@node Legal Ada 83 programs that are illegal in Ada 95,More deterministic semantics,,Compatibility with Ada 83
28098
@anchor{gnat_rm/compatibility_and_porting_guide id4}@anchor{3ee}@anchor{gnat_rm/compatibility_and_porting_guide legal-ada-83-programs-that-are-illegal-in-ada-95}@anchor{3ef}
28099
@subsection Legal Ada 83 programs that are illegal in Ada 95
28102
Some legal Ada 83 programs are illegal (i.e., they will fail to compile) in
28103
Ada 95 and later versions of the standard:
28109
@emph{Character literals}
28111
Some uses of character literals are ambiguous. Since Ada 95 has introduced
28112
@cite{Wide_Character} as a new predefined character type, some uses of
28113
character literals that were legal in Ada 83 are illegal in Ada 95.
28117
for Char in 'A' .. 'Z' loop ... end loop;
28120
The problem is that 'A' and 'Z' could be from either
28121
@cite{Character} or @cite{Wide_Character}. The simplest correction
28122
is to make the type explicit; e.g.:
28125
for Char in Character range 'A' .. 'Z' loop ... end loop;
28129
@emph{New reserved words}
28131
The identifiers @cite{abstract}, @cite{aliased}, @cite{protected},
28132
@cite{requeue}, @cite{tagged}, and @cite{until} are reserved in Ada 95.
28133
Existing Ada 83 code using any of these identifiers must be edited to
28134
use some alternative name.
28137
@emph{Freezing rules}
28139
The rules in Ada 95 are slightly different with regard to the point at
28140
which entities are frozen, and representation pragmas and clauses are
28141
not permitted past the freeze point. This shows up most typically in
28142
the form of an error message complaining that a representation item
28143
appears too late, and the appropriate corrective action is to move
28144
the item nearer to the declaration of the entity to which it refers.
28146
A particular case is that representation pragmas
28147
cannot be applied to a subprogram body. If necessary, a separate subprogram
28148
declaration must be introduced to which the pragma can be applied.
28151
@emph{Optional bodies for library packages}
28153
In Ada 83, a package that did not require a package body was nevertheless
28154
allowed to have one. This lead to certain surprises in compiling large
28155
systems (situations in which the body could be unexpectedly ignored by the
28156
binder). In Ada 95, if a package does not require a body then it is not
28157
permitted to have a body. To fix this problem, simply remove a redundant
28158
body if it is empty, or, if it is non-empty, introduce a dummy declaration
28159
into the spec that makes the body required. One approach is to add a private
28160
part to the package declaration (if necessary), and define a parameterless
28161
procedure called @cite{Requires_Body}, which must then be given a dummy
28162
procedure body in the package body, which then becomes required.
28163
Another approach (assuming that this does not introduce elaboration
28164
circularities) is to add an @cite{Elaborate_Body} pragma to the package spec,
28165
since one effect of this pragma is to require the presence of a package body.
28168
@emph{Numeric_Error is the same exception as Constraint_Error}
28170
In Ada 95, the exception @cite{Numeric_Error} is a renaming of @cite{Constraint_Error}.
28171
This means that it is illegal to have separate exception handlers for
28172
the two exceptions. The fix is simply to remove the handler for the
28173
@cite{Numeric_Error} case (since even in Ada 83, a compiler was free to raise
28174
@cite{Constraint_Error} in place of @cite{Numeric_Error} in all cases).
28177
@emph{Indefinite subtypes in generics}
28179
In Ada 83, it was permissible to pass an indefinite type (e.g, @cite{String})
28180
as the actual for a generic formal private type, but then the instantiation
28181
would be illegal if there were any instances of declarations of variables
28182
of this type in the generic body. In Ada 95, to avoid this clear violation
28183
of the methodological principle known as the 'contract model',
28184
the generic declaration explicitly indicates whether
28185
or not such instantiations are permitted. If a generic formal parameter
28186
has explicit unknown discriminants, indicated by using @cite{(<>)} after the
28187
subtype name, then it can be instantiated with indefinite types, but no
28188
stand-alone variables can be declared of this type. Any attempt to declare
28189
such a variable will result in an illegality at the time the generic is
28190
declared. If the @cite{(<>)} notation is not used, then it is illegal
28191
to instantiate the generic with an indefinite type.
28192
This is the potential incompatibility issue when porting Ada 83 code to Ada 95.
28193
It will show up as a compile time error, and
28194
the fix is usually simply to add the @cite{(<>)} to the generic declaration.
28197
@node More deterministic semantics,Changed semantics,Legal Ada 83 programs that are illegal in Ada 95,Compatibility with Ada 83
28198
@anchor{gnat_rm/compatibility_and_porting_guide more-deterministic-semantics}@anchor{3f0}@anchor{gnat_rm/compatibility_and_porting_guide id5}@anchor{3f1}
28199
@subsection More deterministic semantics
28208
Conversions from real types to integer types round away from 0. In Ada 83
28209
the conversion Integer(2.5) could deliver either 2 or 3 as its value. This
28210
implementation freedom was intended to support unbiased rounding in
28211
statistical applications, but in practice it interfered with portability.
28212
In Ada 95 the conversion semantics are unambiguous, and rounding away from 0
28213
is required. Numeric code may be affected by this change in semantics.
28214
Note, though, that this issue is no worse than already existed in Ada 83
28215
when porting code from one vendor to another.
28220
The Real-Time Annex introduces a set of policies that define the behavior of
28221
features that were implementation dependent in Ada 83, such as the order in
28222
which open select branches are executed.
28225
@node Changed semantics,Other language compatibility issues,More deterministic semantics,Compatibility with Ada 83
28226
@anchor{gnat_rm/compatibility_and_porting_guide id6}@anchor{3f2}@anchor{gnat_rm/compatibility_and_porting_guide changed-semantics}@anchor{3f3}
28227
@subsection Changed semantics
28230
The worst kind of incompatibility is one where a program that is legal in
28231
Ada 83 is also legal in Ada 95 but can have an effect in Ada 95 that was not
28232
possible in Ada 83. Fortunately this is extremely rare, but the one
28233
situation that you should be alert to is the change in the predefined type
28234
@cite{Character} from 7-bit ASCII to 8-bit Latin-1.
28245
@emph{Range of type `Character`}
28247
The range of @cite{Standard.Character} is now the full 256 characters
28248
of Latin-1, whereas in most Ada 83 implementations it was restricted
28249
to 128 characters. Although some of the effects of
28250
this change will be manifest in compile-time rejection of legal
28251
Ada 83 programs it is possible for a working Ada 83 program to have
28252
a different effect in Ada 95, one that was not permitted in Ada 83.
28253
As an example, the expression
28254
@cite{Character'Pos(Character'Last)} returned @cite{127} in Ada 83 and now
28255
delivers @cite{255} as its value.
28256
In general, you should look at the logic of any
28257
character-processing Ada 83 program and see whether it needs to be adapted
28258
to work correctly with Latin-1. Note that the predefined Ada 95 API has a
28259
character handling package that may be relevant if code needs to be adapted
28260
to account for the additional Latin-1 elements.
28261
The desirable fix is to
28262
modify the program to accommodate the full character set, but in some cases
28263
it may be convenient to define a subtype or derived type of Character that
28264
covers only the restricted range.
28267
@node Other language compatibility issues,,Changed semantics,Compatibility with Ada 83
28268
@anchor{gnat_rm/compatibility_and_porting_guide other-language-compatibility-issues}@anchor{3f4}@anchor{gnat_rm/compatibility_and_porting_guide id7}@anchor{3f5}
28269
@subsection Other language compatibility issues
28276
@emph{-gnat83} switch
28278
All implementations of GNAT provide a switch that causes GNAT to operate
28279
in Ada 83 mode. In this mode, some but not all compatibility problems
28280
of the type described above are handled automatically. For example, the
28281
new reserved words introduced in Ada 95 and Ada 2005 are treated simply
28282
as identifiers as in Ada 83. However,
28283
in practice, it is usually advisable to make the necessary modifications
28284
to the program to remove the need for using this switch.
28285
See the @cite{Compiling Different Versions of Ada} section in
28286
the @cite{GNAT User's Guide}.
28289
Support for removed Ada 83 pragmas and attributes
28291
A number of pragmas and attributes from Ada 83 were removed from Ada 95,
28292
generally because they were replaced by other mechanisms. Ada 95 and Ada 2005
28293
compilers are allowed, but not required, to implement these missing
28294
elements. In contrast with some other compilers, GNAT implements all
28295
such pragmas and attributes, eliminating this compatibility concern. These
28296
include @cite{pragma Interface} and the floating point type attributes
28297
(@cite{Emax}, @cite{Mantissa}, etc.), among other items.
28300
@node Compatibility between Ada 95 and Ada 2005,Implementation-dependent characteristics,Compatibility with Ada 83,Compatibility and Porting Guide
28301
@anchor{gnat_rm/compatibility_and_porting_guide compatibility-between-ada-95-and-ada-2005}@anchor{3f6}@anchor{gnat_rm/compatibility_and_porting_guide id8}@anchor{3f7}
28302
@section Compatibility between Ada 95 and Ada 2005
28305
@geindex Compatibility between Ada 95 and Ada 2005
28307
Although Ada 2005 was designed to be upwards compatible with Ada 95, there are
28308
a number of incompatibilities. Several are enumerated below;
28309
for a complete description please see the
28310
@cite{Annotated Ada 2005 Reference Manual}, or section 9.1.1 in
28311
@cite{Rationale for Ada 2005}.
28317
@emph{New reserved words.}
28319
The words @cite{interface}, @cite{overriding} and @cite{synchronized} are
28320
reserved in Ada 2005.
28321
A pre-Ada 2005 program that uses any of these as an identifier will be
28325
@emph{New declarations in predefined packages.}
28327
A number of packages in the predefined environment contain new declarations:
28328
@cite{Ada.Exceptions}, @cite{Ada.Real_Time}, @cite{Ada.Strings},
28329
@cite{Ada.Strings.Fixed}, @cite{Ada.Strings.Bounded},
28330
@cite{Ada.Strings.Unbounded}, @cite{Ada.Strings.Wide_Fixed},
28331
@cite{Ada.Strings.Wide_Bounded}, @cite{Ada.Strings.Wide_Unbounded},
28332
@cite{Ada.Tags}, @cite{Ada.Text_IO}, and @cite{Interfaces.C}.
28333
If an Ada 95 program does a @cite{with} and @cite{use} of any of these
28334
packages, the new declarations may cause name clashes.
28337
@emph{Access parameters.}
28339
A nondispatching subprogram with an access parameter cannot be renamed
28340
as a dispatching operation. This was permitted in Ada 95.
28343
@emph{Access types, discriminants, and constraints.}
28345
Rule changes in this area have led to some incompatibilities; for example,
28346
constrained subtypes of some access types are not permitted in Ada 2005.
28349
@emph{Aggregates for limited types.}
28351
The allowance of aggregates for limited types in Ada 2005 raises the
28352
possibility of ambiguities in legal Ada 95 programs, since additional types
28353
now need to be considered in expression resolution.
28356
@emph{Fixed-point multiplication and division.}
28358
Certain expressions involving '*' or '/' for a fixed-point type, which
28359
were legal in Ada 95 and invoked the predefined versions of these operations,
28361
The ambiguity may be resolved either by applying a type conversion to the
28362
expression, or by explicitly invoking the operation from package
28366
@emph{Return-by-reference types.}
28368
The Ada 95 return-by-reference mechanism has been removed. Instead, the user
28369
can declare a function returning a value from an anonymous access type.
28372
@node Implementation-dependent characteristics,Compatibility with Other Ada Systems,Compatibility between Ada 95 and Ada 2005,Compatibility and Porting Guide
28373
@anchor{gnat_rm/compatibility_and_porting_guide implementation-dependent-characteristics}@anchor{3f8}@anchor{gnat_rm/compatibility_and_porting_guide id9}@anchor{3f9}
28374
@section Implementation-dependent characteristics
28377
Although the Ada language defines the semantics of each construct as
28378
precisely as practical, in some situations (for example for reasons of
28379
efficiency, or where the effect is heavily dependent on the host or target
28380
platform) the implementation is allowed some freedom. In porting Ada 83
28381
code to GNAT, you need to be aware of whether / how the existing code
28382
exercised such implementation dependencies. Such characteristics fall into
28383
several categories, and GNAT offers specific support in assisting the
28384
transition from certain Ada 83 compilers.
28387
* Implementation-defined pragmas::
28388
* Implementation-defined attributes::
28390
* Elaboration order::
28391
* Target-specific aspects::
28395
@node Implementation-defined pragmas,Implementation-defined attributes,,Implementation-dependent characteristics
28396
@anchor{gnat_rm/compatibility_and_porting_guide implementation-defined-pragmas}@anchor{3fa}@anchor{gnat_rm/compatibility_and_porting_guide id10}@anchor{3fb}
28397
@subsection Implementation-defined pragmas
28400
Ada compilers are allowed to supplement the language-defined pragmas, and
28401
these are a potential source of non-portability. All GNAT-defined pragmas
28402
are described in the @cite{Implementation Defined Pragmas} chapter of the
28403
@cite{GNAT Reference Manual}, and these include several that are specifically
28404
intended to correspond to other vendors' Ada 83 pragmas.
28405
For migrating from VADS, the pragma @cite{Use_VADS_Size} may be useful.
28406
For compatibility with HP Ada 83, GNAT supplies the pragmas
28407
@cite{Extend_System}, @cite{Ident}, @cite{Inline_Generic},
28408
@cite{Interface_Name}, @cite{Passive}, @cite{Suppress_All},
28409
and @cite{Volatile}.
28410
Other relevant pragmas include @cite{External} and @cite{Link_With}.
28411
Some vendor-specific
28412
Ada 83 pragmas (@cite{Share_Generic}, @cite{Subtitle}, and @cite{Title}) are
28414
avoiding compiler rejection of units that contain such pragmas; they are not
28415
relevant in a GNAT context and hence are not otherwise implemented.
28417
@node Implementation-defined attributes,Libraries,Implementation-defined pragmas,Implementation-dependent characteristics
28418
@anchor{gnat_rm/compatibility_and_porting_guide id11}@anchor{3fc}@anchor{gnat_rm/compatibility_and_porting_guide implementation-defined-attributes}@anchor{3fd}
28419
@subsection Implementation-defined attributes
28422
Analogous to pragmas, the set of attributes may be extended by an
28423
implementation. All GNAT-defined attributes are described in
28424
@cite{Implementation Defined Attributes} section of the
28425
@cite{GNAT Reference Manual}, and these include several that are specifically intended
28426
to correspond to other vendors' Ada 83 attributes. For migrating from VADS,
28427
the attribute @cite{VADS_Size} may be useful. For compatibility with HP
28428
Ada 83, GNAT supplies the attributes @cite{Bit}, @cite{Machine_Size} and
28431
@node Libraries,Elaboration order,Implementation-defined attributes,Implementation-dependent characteristics
28432
@anchor{gnat_rm/compatibility_and_porting_guide libraries}@anchor{3fe}@anchor{gnat_rm/compatibility_and_porting_guide id12}@anchor{3ff}
28433
@subsection Libraries
28436
Vendors may supply libraries to supplement the standard Ada API. If Ada 83
28437
code uses vendor-specific libraries then there are several ways to manage
28438
this in Ada 95 and later versions of the standard:
28444
If the source code for the libraries (specs and bodies) are
28445
available, then the libraries can be migrated in the same way as the
28449
If the source code for the specs but not the bodies are
28450
available, then you can reimplement the bodies.
28453
Some features introduced by Ada 95 obviate the need for library support. For
28454
example most Ada 83 vendors supplied a package for unsigned integers. The
28455
Ada 95 modular type feature is the preferred way to handle this need, so
28456
instead of migrating or reimplementing the unsigned integer package it may
28457
be preferable to retrofit the application using modular types.
28460
@node Elaboration order,Target-specific aspects,Libraries,Implementation-dependent characteristics
28461
@anchor{gnat_rm/compatibility_and_porting_guide elaboration-order}@anchor{400}@anchor{gnat_rm/compatibility_and_porting_guide id13}@anchor{401}
28462
@subsection Elaboration order
28465
The implementation can choose any elaboration order consistent with the unit
28466
dependency relationship. This freedom means that some orders can result in
28467
Program_Error being raised due to an 'Access Before Elaboration': an attempt
28468
to invoke a subprogram before its body has been elaborated, or to instantiate
28469
a generic before the generic body has been elaborated. By default GNAT
28470
attempts to choose a safe order (one that will not encounter access before
28471
elaboration problems) by implicitly inserting @cite{Elaborate} or
28472
@cite{Elaborate_All} pragmas where
28473
needed. However, this can lead to the creation of elaboration circularities
28474
and a resulting rejection of the program by gnatbind. This issue is
28475
thoroughly described in the @cite{Elaboration Order Handling in GNAT} appendix
28476
in the @cite{GNAT User's Guide}.
28477
In brief, there are several
28478
ways to deal with this situation:
28484
Modify the program to eliminate the circularities, e.g., by moving
28485
elaboration-time code into explicitly-invoked procedures
28488
Constrain the elaboration order by including explicit @cite{Elaborate_Body} or
28489
@cite{Elaborate} pragmas, and then inhibit the generation of implicit
28490
@cite{Elaborate_All}
28491
pragmas either globally (as an effect of the @emph{-gnatE} switch) or locally
28492
(by selectively suppressing elaboration checks via pragma
28493
@cite{Suppress(Elaboration_Check)} when it is safe to do so).
28496
@node Target-specific aspects,,Elaboration order,Implementation-dependent characteristics
28497
@anchor{gnat_rm/compatibility_and_porting_guide target-specific-aspects}@anchor{402}@anchor{gnat_rm/compatibility_and_porting_guide id14}@anchor{403}
28498
@subsection Target-specific aspects
28501
Low-level applications need to deal with machine addresses, data
28502
representations, interfacing with assembler code, and similar issues. If
28503
such an Ada 83 application is being ported to different target hardware (for
28504
example where the byte endianness has changed) then you will need to
28505
carefully examine the program logic; the porting effort will heavily depend
28506
on the robustness of the original design. Moreover, Ada 95 (and thus
28507
Ada 2005 and Ada 2012) are sometimes
28508
incompatible with typical Ada 83 compiler practices regarding implicit
28509
packing, the meaning of the Size attribute, and the size of access values.
28510
GNAT's approach to these issues is described in @ref{404,,Representation Clauses}.
28512
@node Compatibility with Other Ada Systems,Representation Clauses,Implementation-dependent characteristics,Compatibility and Porting Guide
28513
@anchor{gnat_rm/compatibility_and_porting_guide id15}@anchor{405}@anchor{gnat_rm/compatibility_and_porting_guide compatibility-with-other-ada-systems}@anchor{406}
28514
@section Compatibility with Other Ada Systems
28517
If programs avoid the use of implementation dependent and
28518
implementation defined features, as documented in the
28519
@cite{Ada Reference Manual}, there should be a high degree of portability between
28520
GNAT and other Ada systems. The following are specific items which
28521
have proved troublesome in moving Ada 95 programs from GNAT to other Ada 95
28522
compilers, but do not affect porting code to GNAT.
28523
(As of January 2007, GNAT is the only compiler available for Ada 2005;
28524
the following issues may or may not arise for Ada 2005 programs
28525
when other compilers appear.)
28531
@emph{Ada 83 Pragmas and Attributes}
28533
Ada 95 compilers are allowed, but not required, to implement the missing
28534
Ada 83 pragmas and attributes that are no longer defined in Ada 95.
28535
GNAT implements all such pragmas and attributes, eliminating this as
28536
a compatibility concern, but some other Ada 95 compilers reject these
28537
pragmas and attributes.
28540
@emph{Specialized Needs Annexes}
28542
GNAT implements the full set of special needs annexes. At the
28543
current time, it is the only Ada 95 compiler to do so. This means that
28544
programs making use of these features may not be portable to other Ada
28545
95 compilation systems.
28548
@emph{Representation Clauses}
28550
Some other Ada 95 compilers implement only the minimal set of
28551
representation clauses required by the Ada 95 reference manual. GNAT goes
28552
far beyond this minimal set, as described in the next section.
28555
@node Representation Clauses,Compatibility with HP Ada 83,Compatibility with Other Ada Systems,Compatibility and Porting Guide
28556
@anchor{gnat_rm/compatibility_and_porting_guide representation-clauses}@anchor{404}@anchor{gnat_rm/compatibility_and_porting_guide id16}@anchor{407}
28557
@section Representation Clauses
28560
The Ada 83 reference manual was quite vague in describing both the minimal
28561
required implementation of representation clauses, and also their precise
28562
effects. Ada 95 (and thus also Ada 2005) are much more explicit, but the
28563
minimal set of capabilities required is still quite limited.
28565
GNAT implements the full required set of capabilities in
28566
Ada 95 and Ada 2005, but also goes much further, and in particular
28567
an effort has been made to be compatible with existing Ada 83 usage to the
28568
greatest extent possible.
28570
A few cases exist in which Ada 83 compiler behavior is incompatible with
28571
the requirements in Ada 95 (and thus also Ada 2005). These are instances of
28572
intentional or accidental dependence on specific implementation dependent
28573
characteristics of these Ada 83 compilers. The following is a list of
28574
the cases most likely to arise in existing Ada 83 code.
28580
@emph{Implicit Packing}
28582
Some Ada 83 compilers allowed a Size specification to cause implicit
28583
packing of an array or record. This could cause expensive implicit
28584
conversions for change of representation in the presence of derived
28585
types, and the Ada design intends to avoid this possibility.
28586
Subsequent AI's were issued to make it clear that such implicit
28587
change of representation in response to a Size clause is inadvisable,
28588
and this recommendation is represented explicitly in the Ada 95 (and Ada 2005)
28589
Reference Manuals as implementation advice that is followed by GNAT.
28590
The problem will show up as an error
28591
message rejecting the size clause. The fix is simply to provide
28592
the explicit pragma @cite{Pack}, or for more fine tuned control, provide
28593
a Component_Size clause.
28596
@emph{Meaning of Size Attribute}
28598
The Size attribute in Ada 95 (and Ada 2005) for discrete types is defined as
28599
the minimal number of bits required to hold values of the type. For example,
28600
on a 32-bit machine, the size of @cite{Natural} will typically be 31 and not
28601
32 (since no sign bit is required). Some Ada 83 compilers gave 31, and
28602
some 32 in this situation. This problem will usually show up as a compile
28603
time error, but not always. It is a good idea to check all uses of the
28604
'Size attribute when porting Ada 83 code. The GNAT specific attribute
28605
Object_Size can provide a useful way of duplicating the behavior of
28606
some Ada 83 compiler systems.
28609
@emph{Size of Access Types}
28611
A common assumption in Ada 83 code is that an access type is in fact a pointer,
28612
and that therefore it will be the same size as a System.Address value. This
28613
assumption is true for GNAT in most cases with one exception. For the case of
28614
a pointer to an unconstrained array type (where the bounds may vary from one
28615
value of the access type to another), the default is to use a 'fat pointer',
28616
which is represented as two separate pointers, one to the bounds, and one to
28617
the array. This representation has a number of advantages, including improved
28618
efficiency. However, it may cause some difficulties in porting existing Ada 83
28619
code which makes the assumption that, for example, pointers fit in 32 bits on
28620
a machine with 32-bit addressing.
28622
To get around this problem, GNAT also permits the use of 'thin pointers' for
28623
access types in this case (where the designated type is an unconstrained array
28624
type). These thin pointers are indeed the same size as a System.Address value.
28625
To specify a thin pointer, use a size clause for the type, for example:
28628
type X is access all String;
28629
for X'Size use Standard'Address_Size;
28632
which will cause the type X to be represented using a single pointer.
28633
When using this representation, the bounds are right behind the array.
28634
This representation is slightly less efficient, and does not allow quite
28635
such flexibility in the use of foreign pointers or in using the
28636
Unrestricted_Access attribute to create pointers to non-aliased objects.
28637
But for any standard portable use of the access type it will work in
28638
a functionally correct manner and allow porting of existing code.
28639
Note that another way of forcing a thin pointer representation
28640
is to use a component size clause for the element size in an array,
28641
or a record representation clause for an access field in a record.
28643
See the documentation of Unrestricted_Access in the GNAT RM for a
28644
full discussion of possible problems using this attribute in conjunction
28645
with thin pointers.
28648
@node Compatibility with HP Ada 83,,Representation Clauses,Compatibility and Porting Guide
28649
@anchor{gnat_rm/compatibility_and_porting_guide compatibility-with-hp-ada-83}@anchor{408}@anchor{gnat_rm/compatibility_and_porting_guide id17}@anchor{409}
28650
@section Compatibility with HP Ada 83
28653
All the HP Ada 83 pragmas and attributes are recognized, although only a subset
28654
of them can sensibly be implemented. The description of pragmas in
28655
@ref{7,,Implementation Defined Pragmas} indicates whether or not they are
28656
applicable to GNAT.
28662
@emph{Default floating-point representation}
28664
In GNAT, the default floating-point format is IEEE, whereas in HP Ada 83,
28670
the package System in GNAT exactly corresponds to the definition in the
28671
Ada 95 reference manual, which means that it excludes many of the
28672
HP Ada 83 extensions. However, a separate package Aux_DEC is provided
28673
that contains the additional definitions, and a special pragma,
28674
Extend_System allows this package to be treated transparently as an
28675
extension of package System.
28678
@node GNU Free Documentation License,Index,Compatibility and Porting Guide,Top
28679
@anchor{share/gnu_free_documentation_license gnu-fdl}@anchor{1}@anchor{share/gnu_free_documentation_license doc}@anchor{40a}@anchor{share/gnu_free_documentation_license gnu-free-documentation-license}@anchor{40b}
28680
@chapter GNU Free Documentation License
28683
Version 1.3, 3 November 2008
28685
Copyright 2000, 2001, 2002, 2007, 2008 Free Software Foundation, Inc
28686
@indicateurl{http://fsf.org/}
28688
Everyone is permitted to copy and distribute verbatim copies of this
28689
license document, but changing it is not allowed.
28693
The purpose of this License is to make a manual, textbook, or other
28694
functional and useful document "free" in the sense of freedom: to
28695
assure everyone the effective freedom to copy and redistribute it,
28696
with or without modifying it, either commercially or noncommercially.
28697
Secondarily, this License preserves for the author and publisher a way
28698
to get credit for their work, while not being considered responsible
28699
for modifications made by others.
28701
This License is a kind of "copyleft", which means that derivative
28702
works of the document must themselves be free in the same sense. It
28703
complements the GNU General Public License, which is a copyleft
28704
license designed for free software.
28706
We have designed this License in order to use it for manuals for free
28707
software, because free software needs free documentation: a free
28708
program should come with manuals providing the same freedoms that the
28709
software does. But this License is not limited to software manuals;
28710
it can be used for any textual work, regardless of subject matter or
28711
whether it is published as a printed book. We recommend this License
28712
principally for works whose purpose is instruction or reference.
28714
@strong{1. APPLICABILITY AND DEFINITIONS}
28716
This License applies to any manual or other work, in any medium, that
28717
contains a notice placed by the copyright holder saying it can be
28718
distributed under the terms of this License. Such a notice grants a
28719
world-wide, royalty-free license, unlimited in duration, to use that
28720
work under the conditions stated herein. The @strong{Document}, below,
28721
refers to any such manual or work. Any member of the public is a
28722
licensee, and is addressed as "@strong{you}". You accept the license if you
28723
copy, modify or distribute the work in a way requiring permission
28724
under copyright law.
28726
A "@strong{Modified Version}" of the Document means any work containing the
28727
Document or a portion of it, either copied verbatim, or with
28728
modifications and/or translated into another language.
28730
A "@strong{Secondary Section}" is a named appendix or a front-matter section of
28731
the Document that deals exclusively with the relationship of the
28732
publishers or authors of the Document to the Document's overall subject
28733
(or to related matters) and contains nothing that could fall directly
28734
within that overall subject. (Thus, if the Document is in part a
28735
textbook of mathematics, a Secondary Section may not explain any
28736
mathematics.) The relationship could be a matter of historical
28737
connection with the subject or with related matters, or of legal,
28738
commercial, philosophical, ethical or political position regarding
28741
The "@strong{Invariant Sections}" are certain Secondary Sections whose titles
28742
are designated, as being those of Invariant Sections, in the notice
28743
that says that the Document is released under this License. If a
28744
section does not fit the above definition of Secondary then it is not
28745
allowed to be designated as Invariant. The Document may contain zero
28746
Invariant Sections. If the Document does not identify any Invariant
28747
Sections then there are none.
28749
The "@strong{Cover Texts}" are certain short passages of text that are listed,
28750
as Front-Cover Texts or Back-Cover Texts, in the notice that says that
28751
the Document is released under this License. A Front-Cover Text may
28752
be at most 5 words, and a Back-Cover Text may be at most 25 words.
28754
A "@strong{Transparent}" copy of the Document means a machine-readable copy,
28755
represented in a format whose specification is available to the
28756
general public, that is suitable for revising the document
28757
straightforwardly with generic text editors or (for images composed of
28758
pixels) generic paint programs or (for drawings) some widely available
28759
drawing editor, and that is suitable for input to text formatters or
28760
for automatic translation to a variety of formats suitable for input
28761
to text formatters. A copy made in an otherwise Transparent file
28762
format whose markup, or absence of markup, has been arranged to thwart
28763
or discourage subsequent modification by readers is not Transparent.
28764
An image format is not Transparent if used for any substantial amount
28765
of text. A copy that is not "Transparent" is called @strong{Opaque}.
28767
Examples of suitable formats for Transparent copies include plain
28768
ASCII without markup, Texinfo input format, LaTeX input format, SGML
28769
or XML using a publicly available DTD, and standard-conforming simple
28770
HTML, PostScript or PDF designed for human modification. Examples of
28771
transparent image formats include PNG, XCF and JPG. Opaque formats
28772
include proprietary formats that can be read and edited only by
28773
proprietary word processors, SGML or XML for which the DTD and/or
28774
processing tools are not generally available, and the
28775
machine-generated HTML, PostScript or PDF produced by some word
28776
processors for output purposes only.
28778
The "@strong{Title Page}" means, for a printed book, the title page itself,
28779
plus such following pages as are needed to hold, legibly, the material
28780
this License requires to appear in the title page. For works in
28781
formats which do not have any title page as such, "Title Page" means
28782
the text near the most prominent appearance of the work's title,
28783
preceding the beginning of the body of the text.
28785
The "@strong{publisher}" means any person or entity that distributes
28786
copies of the Document to the public.
28788
A section "@strong{Entitled XYZ}" means a named subunit of the Document whose
28789
title either is precisely XYZ or contains XYZ in parentheses following
28790
text that translates XYZ in another language. (Here XYZ stands for a
28791
specific section name mentioned below, such as "@strong{Acknowledgements}",
28792
"@strong{Dedications}", "@strong{Endorsements}", or "@strong{History}".)
28793
To "@strong{Preserve the Title}"
28794
of such a section when you modify the Document means that it remains a
28795
section "Entitled XYZ" according to this definition.
28797
The Document may include Warranty Disclaimers next to the notice which
28798
states that this License applies to the Document. These Warranty
28799
Disclaimers are considered to be included by reference in this
28800
License, but only as regards disclaiming warranties: any other
28801
implication that these Warranty Disclaimers may have is void and has
28802
no effect on the meaning of this License.
28804
@strong{2. VERBATIM COPYING}
28806
You may copy and distribute the Document in any medium, either
28807
commercially or noncommercially, provided that this License, the
28808
copyright notices, and the license notice saying this License applies
28809
to the Document are reproduced in all copies, and that you add no other
28810
conditions whatsoever to those of this License. You may not use
28811
technical measures to obstruct or control the reading or further
28812
copying of the copies you make or distribute. However, you may accept
28813
compensation in exchange for copies. If you distribute a large enough
28814
number of copies you must also follow the conditions in section 3.
28816
You may also lend copies, under the same conditions stated above, and
28817
you may publicly display copies.
28819
@strong{3. COPYING IN QUANTITY}
28821
If you publish printed copies (or copies in media that commonly have
28822
printed covers) of the Document, numbering more than 100, and the
28823
Document's license notice requires Cover Texts, you must enclose the
28824
copies in covers that carry, clearly and legibly, all these Cover
28825
Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on
28826
the back cover. Both covers must also clearly and legibly identify
28827
you as the publisher of these copies. The front cover must present
28828
the full title with all words of the title equally prominent and
28829
visible. You may add other material on the covers in addition.
28830
Copying with changes limited to the covers, as long as they preserve
28831
the title of the Document and satisfy these conditions, can be treated
28832
as verbatim copying in other respects.
28834
If the required texts for either cover are too voluminous to fit
28835
legibly, you should put the first ones listed (as many as fit
28836
reasonably) on the actual cover, and continue the rest onto adjacent
28839
If you publish or distribute Opaque copies of the Document numbering
28840
more than 100, you must either include a machine-readable Transparent
28841
copy along with each Opaque copy, or state in or with each Opaque copy
28842
a computer-network location from which the general network-using
28843
public has access to download using public-standard network protocols
28844
a complete Transparent copy of the Document, free of added material.
28845
If you use the latter option, you must take reasonably prudent steps,
28846
when you begin distribution of Opaque copies in quantity, to ensure
28847
that this Transparent copy will remain thus accessible at the stated
28848
location until at least one year after the last time you distribute an
28849
Opaque copy (directly or through your agents or retailers) of that
28850
edition to the public.
28852
It is requested, but not required, that you contact the authors of the
28853
Document well before redistributing any large number of copies, to give
28854
them a chance to provide you with an updated version of the Document.
28856
@strong{4. MODIFICATIONS}
28858
You may copy and distribute a Modified Version of the Document under
28859
the conditions of sections 2 and 3 above, provided that you release
28860
the Modified Version under precisely this License, with the Modified
28861
Version filling the role of the Document, thus licensing distribution
28862
and modification of the Modified Version to whoever possesses a copy
28863
of it. In addition, you must do these things in the Modified Version:
28869
Use in the Title Page (and on the covers, if any) a title distinct
28870
from that of the Document, and from those of previous versions
28871
(which should, if there were any, be listed in the History section
28872
of the Document). You may use the same title as a previous version
28873
if the original publisher of that version gives permission.
28876
List on the Title Page, as authors, one or more persons or entities
28877
responsible for authorship of the modifications in the Modified
28878
Version, together with at least five of the principal authors of the
28879
Document (all of its principal authors, if it has fewer than five),
28880
unless they release you from this requirement.
28883
State on the Title page the name of the publisher of the
28884
Modified Version, as the publisher.
28887
Preserve all the copyright notices of the Document.
28890
Add an appropriate copyright notice for your modifications
28891
adjacent to the other copyright notices.
28894
Include, immediately after the copyright notices, a license notice
28895
giving the public permission to use the Modified Version under the
28896
terms of this License, in the form shown in the Addendum below.
28899
Preserve in that license notice the full lists of Invariant Sections
28900
and required Cover Texts given in the Document's license notice.
28903
Include an unaltered copy of this License.
28906
Preserve the section Entitled "History", Preserve its Title, and add
28907
to it an item stating at least the title, year, new authors, and
28908
publisher of the Modified Version as given on the Title Page. If
28909
there is no section Entitled "History" in the Document, create one
28910
stating the title, year, authors, and publisher of the Document as
28911
given on its Title Page, then add an item describing the Modified
28912
Version as stated in the previous sentence.
28915
Preserve the network location, if any, given in the Document for
28916
public access to a Transparent copy of the Document, and likewise
28917
the network locations given in the Document for previous versions
28918
it was based on. These may be placed in the "History" section.
28919
You may omit a network location for a work that was published at
28920
least four years before the Document itself, or if the original
28921
publisher of the version it refers to gives permission.
28924
For any section Entitled "Acknowledgements" or "Dedications",
28925
Preserve the Title of the section, and preserve in the section all
28926
the substance and tone of each of the contributor acknowledgements
28927
and/or dedications given therein.
28930
Preserve all the Invariant Sections of the Document,
28931
unaltered in their text and in their titles. Section numbers
28932
or the equivalent are not considered part of the section titles.
28935
Delete any section Entitled "Endorsements". Such a section
28936
may not be included in the Modified Version.
28939
Do not retitle any existing section to be Entitled "Endorsements"
28940
or to conflict in title with any Invariant Section.
28943
Preserve any Warranty Disclaimers.
28946
If the Modified Version includes new front-matter sections or
28947
appendices that qualify as Secondary Sections and contain no material
28948
copied from the Document, you may at your option designate some or all
28949
of these sections as invariant. To do this, add their titles to the
28950
list of Invariant Sections in the Modified Version's license notice.
28951
These titles must be distinct from any other section titles.
28953
You may add a section Entitled "Endorsements", provided it contains
28954
nothing but endorsements of your Modified Version by various
28955
parties---for example, statements of peer review or that the text has
28956
been approved by an organization as the authoritative definition of a
28959
You may add a passage of up to five words as a Front-Cover Text, and a
28960
passage of up to 25 words as a Back-Cover Text, to the end of the list
28961
of Cover Texts in the Modified Version. Only one passage of
28962
Front-Cover Text and one of Back-Cover Text may be added by (or
28963
through arrangements made by) any one entity. If the Document already
28964
includes a cover text for the same cover, previously added by you or
28965
by arrangement made by the same entity you are acting on behalf of,
28966
you may not add another; but you may replace the old one, on explicit
28967
permission from the previous publisher that added the old one.
28969
The author(s) and publisher(s) of the Document do not by this License
28970
give permission to use their names for publicity for or to assert or
28971
imply endorsement of any Modified Version.
28973
@strong{5. COMBINING DOCUMENTS}
28975
You may combine the Document with other documents released under this
28976
License, under the terms defined in section 4 above for modified
28977
versions, provided that you include in the combination all of the
28978
Invariant Sections of all of the original documents, unmodified, and
28979
list them all as Invariant Sections of your combined work in its
28980
license notice, and that you preserve all their Warranty Disclaimers.
28982
The combined work need only contain one copy of this License, and
28983
multiple identical Invariant Sections may be replaced with a single
28984
copy. If there are multiple Invariant Sections with the same name but
28985
different contents, make the title of each such section unique by
28986
adding at the end of it, in parentheses, the name of the original
28987
author or publisher of that section if known, or else a unique number.
28988
Make the same adjustment to the section titles in the list of
28989
Invariant Sections in the license notice of the combined work.
28991
In the combination, you must combine any sections Entitled "History"
28992
in the various original documents, forming one section Entitled
28993
"History"; likewise combine any sections Entitled "Acknowledgements",
28994
and any sections Entitled "Dedications". You must delete all sections
28995
Entitled "Endorsements".
28997
@strong{6. COLLECTIONS OF DOCUMENTS}
28999
You may make a collection consisting of the Document and other documents
29000
released under this License, and replace the individual copies of this
29001
License in the various documents with a single copy that is included in
29002
the collection, provided that you follow the rules of this License for
29003
verbatim copying of each of the documents in all other respects.
29005
You may extract a single document from such a collection, and distribute
29006
it individually under this License, provided you insert a copy of this
29007
License into the extracted document, and follow this License in all
29008
other respects regarding verbatim copying of that document.
29010
@strong{7. AGGREGATION WITH INDEPENDENT WORKS}
29012
A compilation of the Document or its derivatives with other separate
29013
and independent documents or works, in or on a volume of a storage or
29014
distribution medium, is called an "aggregate" if the copyright
29015
resulting from the compilation is not used to limit the legal rights
29016
of the compilation's users beyond what the individual works permit.
29017
When the Document is included in an aggregate, this License does not
29018
apply to the other works in the aggregate which are not themselves
29019
derivative works of the Document.
29021
If the Cover Text requirement of section 3 is applicable to these
29022
copies of the Document, then if the Document is less than one half of
29023
the entire aggregate, the Document's Cover Texts may be placed on
29024
covers that bracket the Document within the aggregate, or the
29025
electronic equivalent of covers if the Document is in electronic form.
29026
Otherwise they must appear on printed covers that bracket the whole
29029
@strong{8. TRANSLATION}
29031
Translation is considered a kind of modification, so you may
29032
distribute translations of the Document under the terms of section 4.
29033
Replacing Invariant Sections with translations requires special
29034
permission from their copyright holders, but you may include
29035
translations of some or all Invariant Sections in addition to the
29036
original versions of these Invariant Sections. You may include a
29037
translation of this License, and all the license notices in the
29038
Document, and any Warranty Disclaimers, provided that you also include
29039
the original English version of this License and the original versions
29040
of those notices and disclaimers. In case of a disagreement between
29041
the translation and the original version of this License or a notice
29042
or disclaimer, the original version will prevail.
29044
If a section in the Document is Entitled "Acknowledgements",
29045
"Dedications", or "History", the requirement (section 4) to Preserve
29046
its Title (section 1) will typically require changing the actual
29049
@strong{9. TERMINATION}
29051
You may not copy, modify, sublicense, or distribute the Document
29052
except as expressly provided under this License. Any attempt
29053
otherwise to copy, modify, sublicense, or distribute it is void, and
29054
will automatically terminate your rights under this License.
29056
However, if you cease all violation of this License, then your license
29057
from a particular copyright holder is reinstated (a) provisionally,
29058
unless and until the copyright holder explicitly and finally
29059
terminates your license, and (b) permanently, if the copyright holder
29060
fails to notify you of the violation by some reasonable means prior to
29061
60 days after the cessation.
29063
Moreover, your license from a particular copyright holder is
29064
reinstated permanently if the copyright holder notifies you of the
29065
violation by some reasonable means, this is the first time you have
29066
received notice of violation of this License (for any work) from that
29067
copyright holder, and you cure the violation prior to 30 days after
29068
your receipt of the notice.
29070
Termination of your rights under this section does not terminate the
29071
licenses of parties who have received copies or rights from you under
29072
this License. If your rights have been terminated and not permanently
29073
reinstated, receipt of a copy of some or all of the same material does
29074
not give you any rights to use it.
29076
@strong{10. FUTURE REVISIONS OF THIS LICENSE}
29078
The Free Software Foundation may publish new, revised versions
29079
of the GNU Free Documentation License from time to time. Such new
29080
versions will be similar in spirit to the present version, but may
29081
differ in detail to address new problems or concerns. See
29082
@indicateurl{http://www.gnu.org/copyleft/}.
29084
Each version of the License is given a distinguishing version number.
29085
If the Document specifies that a particular numbered version of this
29086
License "or any later version" applies to it, you have the option of
29087
following the terms and conditions either of that specified version or
29088
of any later version that has been published (not as a draft) by the
29089
Free Software Foundation. If the Document does not specify a version
29090
number of this License, you may choose any version ever published (not
29091
as a draft) by the Free Software Foundation. If the Document
29092
specifies that a proxy can decide which future versions of this
29093
License can be used, that proxy's public statement of acceptance of a
29094
version permanently authorizes you to choose that version for the
29097
@strong{11. RELICENSING}
29099
"Massive Multiauthor Collaboration Site" (or "MMC Site") means any
29100
World Wide Web server that publishes copyrightable works and also
29101
provides prominent facilities for anybody to edit those works. A
29102
public wiki that anybody can edit is an example of such a server. A
29103
"Massive Multiauthor Collaboration" (or "MMC") contained in the
29104
site means any set of copyrightable works thus published on the MMC
29107
"CC-BY-SA" means the Creative Commons Attribution-Share Alike 3.0
29108
license published by Creative Commons Corporation, a not-for-profit
29109
corporation with a principal place of business in San Francisco,
29110
California, as well as future copyleft versions of that license
29111
published by that same organization.
29113
"Incorporate" means to publish or republish a Document, in whole or
29114
in part, as part of another Document.
29116
An MMC is "eligible for relicensing" if it is licensed under this
29117
License, and if all works that were first published under this License
29118
somewhere other than this MMC, and subsequently incorporated in whole
29119
or in part into the MMC, (1) had no cover texts or invariant sections,
29120
and (2) were thus incorporated prior to November 1, 2008.
29122
The operator of an MMC Site may republish an MMC contained in the site
29123
under CC-BY-SA on the same site at any time before August 1, 2009,
29124
provided the MMC is eligible for relicensing.
29126
@strong{ADDENDUM: How to use this License for your documents}
29128
To use this License in a document you have written, include a copy of
29129
the License in the document and put the following copyright and
29130
license notices just after the title page:
29134
Copyright © YEAR YOUR NAME.
29135
Permission is granted to copy, distribute and/or modify this document
29136
under the terms of the GNU Free Documentation License, Version 1.3
29137
or any later version published by the Free Software Foundation;
29138
with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts.
29139
A copy of the license is included in the section entitled "GNU
29140
Free Documentation License".
29143
If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts,
29144
replace the "with ... Texts." line with this:
29148
with the Invariant Sections being LIST THEIR TITLES, with the
29149
Front-Cover Texts being LIST, and with the Back-Cover Texts being LIST.
29152
If you have Invariant Sections without Cover Texts, or some other
29153
combination of the three, merge those two alternatives to suit the
29156
If your document contains nontrivial examples of program code, we
29157
recommend releasing these examples in parallel under your choice of
29158
free software license, such as the GNU General Public License,
29159
to permit their use in free software.
29161
@node Index,,GNU Free Documentation License,Top