1------------------------------------------------------------------------------ 2-- -- 3-- GNAT COMPILER COMPONENTS -- 4-- -- 5-- L I B . W R I T -- 6-- -- 7-- S p e c -- 8-- -- 9-- Copyright (C) 1992-2013, Free Software Foundation, Inc. -- 10-- -- 11-- GNAT is free software; you can redistribute it and/or modify it under -- 12-- terms of the GNU General Public License as published by the Free Soft- -- 13-- ware Foundation; either version 3, or (at your option) any later ver- -- 14-- sion. GNAT is distributed in the hope that it will be useful, but WITH- -- 15-- OUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY -- 16-- or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License -- 17-- for more details. You should have received a copy of the GNU General -- 18-- Public License distributed with GNAT; see file COPYING3. If not, go to -- 19-- http://www.gnu.org/licenses for a complete copy of the license. -- 20-- -- 21-- GNAT was originally developed by the GNAT team at New York University. -- 22-- Extensive contributions were provided by Ada Core Technologies Inc. -- 23-- -- 24------------------------------------------------------------------------------ 25 26-- This package contains the routines for writing the library information 27 28package Lib.Writ is 29 30 ----------------------------------- 31 -- Format of Library Information -- 32 ----------------------------------- 33 34 -- This section describes the format of the library information that is 35 -- associated with object files. The exact method of this association is 36 -- potentially implementation dependent and is described and implemented in 37 -- package ali. From the point of view of the description here, all we need 38 -- to know is that the information is represented as a string of characters 39 -- that is somehow associated with an object file, and can be retrieved. If 40 -- no library information exists for a given object file, then we take this 41 -- as equivalent to the non-existence of the object file, as if source file 42 -- has not been previously compiled. 43 44 -- The library information is written as a series of lines of the form: 45 46 -- Key_Character parameter parameter ... 47 48 -- The following sections describe the format of these lines in detail 49 50 -------------------------------------- 51 -- Making Changes to the ALI Format -- 52 -------------------------------------- 53 54 -- A number of tools use ali.adb to parse ali files. This means that 55 -- changes to this format can cause old versions of these tools to be 56 -- incompatible with new versions of the compiler. Any changes to ali file 57 -- formats must be carefully evaluated to understand any such possible 58 -- conflicts, and in particular, it is very undesirable to create conflicts 59 -- between older versions of GPS and newer versions of the compiler. 60 61 -- If the following guidelines are respected, downward compatibility 62 -- problems (old tools reading new ali files) should be minimized: 63 64 -- The basic key character format must be kept 65 66 -- The V line must be the first line, this is checked by ali.adb even in 67 -- Ignore_Errors mode, and is used to verify that the file at hand is 68 -- indeed likely intended to be an ali file. 69 70 -- The P line must be present, though may be modified in contents 71 -- according to remaining guidelines. Again, ali.adb assumes the P 72 -- line is present even in Ignore_Errors mode. 73 74 -- New modifiers can generally be added (in particular adding new two 75 -- letter modifiers to the P or U lines is always safe) 76 77 -- Adding entirely new lines (with a new key letter) to the ali file is 78 -- always safe, at any point (other than before the V line), since such 79 -- lines will be ignored. 80 81 -- Following the guidelines in this section should ensure that this problem 82 -- is minimized and that old tools will be able to deal successfully with 83 -- new ali formats. Note that this does not apply to the compiler itself, 84 -- which always requires consistency between the ali files and the binder. 85 -- That is because one of the main functions of the binder is to ensure 86 -- consistency of the partition, and this can be compromised if the ali 87 -- files are inconsistent. 88 89 ------------------ 90 -- Header Lines -- 91 ------------------ 92 93 -- The initial header lines in the file give information about the 94 -- compilation environment, and identify other special information such as 95 -- main program parameters. 96 97 -- ---------------- 98 -- -- V Version -- 99 -- ---------------- 100 101 -- V "xxxxxxxxxxxxxxxx" 102 -- 103 -- This line indicates the library output version, as defined in 104 -- Gnatvsn. It ensures that separate object modules of a program are 105 -- consistent. It has to be changed if anything changes which would 106 -- affect successful binding of separately compiled modules. Examples 107 -- of such changes are modifications in the format of the library info 108 -- described in this package, or modifications to calling sequences, or 109 -- to the way that data is represented. 110 111 -- Note: the V line absolutely must be the first line, and no change 112 -- to the ALI format should change this, since even in Ignore_Errors 113 -- mode, Scan_ALI insists on finding a V line. 114 115 -- --------------------- 116 -- -- M Main Program -- 117 -- --------------------- 118 119 -- M type [priority] [T=time-slice] [AB] [C=cpu] W=? 120 121 -- This line appears only if the main unit for this file is suitable 122 -- for use as a main program. The parameters are: 123 124 -- type 125 126 -- P for a parameterless procedure 127 -- F for a function returning a value of integral type 128 -- (used for writing a main program returning an exit status) 129 130 -- priority 131 132 -- Present only if there was a valid pragma Priority in the 133 -- corresponding unit to set the main task priority. It is an 134 -- unsigned decimal integer. 135 136 -- T=time-slice 137 138 -- Present only if there was a valid pragma Time_Slice in the 139 -- corresponding unit. It is an unsigned decimal integer in the 140 -- range 0 .. 10**9 giving the time slice value in units of 141 -- milliseconds. The actual significance of this parameter is 142 -- target dependent. 143 144 -- AB 145 146 -- Present if there is an allocator in the body of the procedure 147 -- after the BEGIN. This will be a violation of the restriction 148 -- No_Allocators_After_Elaboration if it is present, and this 149 -- unit is used as a main program (only the binder can find the 150 -- violation, since only the binder knows the main program). 151 152 -- C=cpu 153 154 -- Present only if there was a valid pragma CPU in the 155 -- corresponding unit to set the main task affinity. It is an 156 -- unsigned decimal integer. 157 158 -- W=? 159 160 -- This parameter indicates the wide character encoding method used 161 -- when compiling the main program file. The ? character is the 162 -- single character used in the -gnatW? switch. This is used to 163 -- provide the default wide-character encoding for Wide_Text_IO 164 -- files. 165 166 -- ----------------- 167 -- -- A Argument -- 168 -- ----------------- 169 170 -- A argument 171 172 -- One of these lines appears for each of the arguments present in the 173 -- call to the gnat1 program. This can be used if it is necessary to 174 -- reconstruct this call (e.g. for fix and continue). 175 176 -- ------------------- 177 -- -- P Parameters -- 178 -- ------------------- 179 180 -- P <<parameters>> 181 182 -- Indicates various information that applies to the compilation of the 183 -- corresponding source file. Parameters is a sequence of zero or more 184 -- two letter codes that indicate configuration pragmas and other 185 -- parameters that apply: 186 187 -- The arguments are as follows: 188 189 -- CE Compilation errors. If this is present it means that the ali 190 -- file resulted from a compilation with the -gnatQ switch set, 191 -- and illegalities were detected. The ali file contents may 192 -- not be completely reliable, but the format will be correct 193 -- and complete. Note that NO is always present if CE is 194 -- present. 195 196 -- DB Detect_Blocking pragma is in effect for all units in this 197 -- file. 198 199 -- Ex A valid Partition_Elaboration_Policy pragma applies to all 200 -- the units in this file, where x is the first character 201 -- (upper case) of the policy name (e.g. 'C' for Concurrent). 202 203 -- FD Configuration pragmas apply to all the units in this file 204 -- specifying a possibly non-standard floating point format 205 -- (VAX float with Long_Float using D_Float). 206 207 -- FG Configuration pragmas apply to all the units in this file 208 -- specifying a possibly non-standard floating point format 209 -- (VAX float with Long_Float using G_Float). 210 211 -- FI Configuration pragmas apply to all the units in this file 212 -- specifying a possibly non-standard floating point format 213 -- (IEEE Float). 214 215 -- Lx A valid Locking_Policy pragma applies to all the units in 216 -- this file, where x is the first character (upper case) of 217 -- the policy name (e.g. 'C' for Ceiling_Locking). 218 219 -- NO No object. This flag indicates that the units in this file 220 -- were not compiled to produce an object. This can occur as a 221 -- result of the use of -gnatc, or if no object can be produced 222 -- (e.g. when a package spec is compiled instead of the body, 223 -- or a subunit on its own). 224 225 -- NR No_Run_Time. Indicates that a pragma No_Run_Time applies 226 -- to all units in the file. 227 228 -- NS Normalize_Scalars pragma in effect for all units in 229 -- this file. 230 231 -- Qx A valid Queueing_Policy pragma applies to all the units 232 -- in this file, where x is the first character (upper case) 233 -- of the policy name (e.g. 'P' for Priority_Queueing). 234 235 -- SL Indicates that the unit is an Interface to a Standalone 236 -- Library. Note that this indication is never given by the 237 -- compiler, but is added by the Project Manager in gnatmake 238 -- when an Interface ALI file is copied to the library 239 -- directory. 240 241 -- SS This unit references System.Secondary_Stack (that is, 242 -- the unit makes use of the secondary stack facilities). 243 244 -- Tx A valid Task_Dispatching_Policy pragma applies to all 245 -- the units in this file, where x is the first character 246 -- (upper case) of the corresponding policy name (e.g. 'F' 247 -- for FIFO_Within_Priorities). 248 249 -- UA Unreserve_All_Interrupts pragma was processed in one or 250 -- more units in this file 251 252 -- ZX Units in this file use zero-cost exceptions and have 253 -- generated exception tables. If ZX is not present, the 254 -- longjmp/setjmp exception scheme is in use. 255 256 -- Note that language defined units never output policy (Lx, Tx, Qx) 257 -- parameters. Language defined units must correctly handle all 258 -- possible cases. These values are checked for consistency by the 259 -- binder and then copied to the generated binder output file. 260 261 -- Note: The P line must be present. Even in Ignore_Errors mode, Scan_ALI 262 -- insists on finding a P line. So if changes are made to the ALI format, 263 -- they should not include removing the P line. 264 265 -- --------------------- 266 -- -- R Restrictions -- 267 -- --------------------- 268 269 -- There are two forms for R lines, positional and named. The positional 270 -- notation is now considered obsolescent, it is not generated by the most 271 -- recent versions of the compiler except under control of the debug switch 272 -- -gnatdR, but is still recognized by the binder. 273 274 -- The recognition by the binder is to ease the transition, and better deal 275 -- with some cases of inconsistent builds using incompatible versions of 276 -- the compiler and binder. The named notation is the current preferred 277 -- approach. 278 279 -- Note that R lines are generated using the information in unit Rident, 280 -- and intepreted by the binder using the information in System.Rident. 281 -- Normally these two units should be effectively identical. However in 282 -- some cases of inconsistent builds, they may be different. This may lead 283 -- to binder diagnostics, which can be suppressed using the -C switch for 284 -- the binder, which results in ignoring unrecognized restrictions in the 285 -- ali files. 286 287 -- --------------------------------------- 288 -- -- R Restrictions (Positional Form) -- 289 -- --------------------------------------- 290 291 -- The first R line records the status of restrictions generated by pragma 292 -- Restrictions encountered, as well as information on what the compiler 293 -- has been able to determine with respect to restrictions violations. 294 -- The format is: 295 296 -- R <<restriction-characters>> <<restriction-param-id-entries>> 297 298 -- The first parameter is a string of characters that records 299 -- information regarding restrictions that do not take parameter not 300 -- take parameter values. It is a string of characters, one character 301 -- for each value (in order) in All_Boolean_Restrictions. There are 302 -- three possible settings for each restriction: 303 304 -- r Restricted. Unit was compiled under control of a pragma 305 -- Restrictions for the corresponding restriction. In this case 306 -- the unit certainly does not violate the Restriction, since 307 -- this would have been detected by the compiler. 308 309 -- n Not used. The unit was not compiled under control of a pragma 310 -- Restrictions for the corresponding restriction, and does not 311 -- make any use of the referenced feature. 312 313 -- v Violated. The unit was not compiled under control of a pragma 314 -- Restrictions for the corresponding restriction, and it does 315 -- indeed use the referenced feature. 316 317 -- This information is used in the binder to check consistency, i.e. to 318 -- detect cases where one unit has "r" and another unit has "v", which 319 -- is not permitted, since these restrictions are partition-wide. 320 321 -- The second parameter, which immediately follows the first (with no 322 -- separating space) gives restriction information for identifiers for 323 -- which a parameter is given. 324 325 -- The parameter is a string of entries, one for each value in 326 -- Restrict.All_Parameter_Restrictions. Each entry has two components 327 -- in sequence, the first indicating whether or not there is a 328 -- restriction, and the second indicating whether or not the compiler 329 -- detected violations. In the boolean case it is not necessary to 330 -- separate these, since if a restriction is set, and violated, that is 331 -- an error. But in the parameter case, this is not true. For example, 332 -- we can have a unit with a pragma Restrictions (Max_Tasks => 4), 333 -- where the compiler can detect that there are exactly three tasks 334 -- declared. Both of these pieces of information must be passed to the 335 -- binder. The parameter of 4 is important in case the total number of 336 -- tasks in the partition is greater than 4. The parameter of 3 is 337 -- important in case some other unit has a restrictions pragma with 338 -- Max_Tasks=>2. 339 340 -- The component for the presence of restriction has one of two 341 -- possible forms: 342 343 -- n No pragma for this restriction is present in the set of units 344 -- for this ali file. 345 346 -- rN At least one pragma for this restriction is present in the 347 -- set of units for this ali file. The value N is the minimum 348 -- parameter value encountered in any such pragma. N is in the 349 -- range of Integer (a value larger than N'Last causes the 350 -- pragma to be ignored). 351 352 -- The component for the violation detection has one of three 353 -- possible forms: 354 355 -- n No violations were detected by the compiler 356 357 -- vN A violation was detected. N is either the maximum or total 358 -- count of violations (depending on the checking type) in all 359 -- the units represented by the ali file). Note that this 360 -- setting is only allowed for restrictions that are in 361 -- Checked_[Max|Sum]_Parameter_Restrictions. The value here is 362 -- known to be exact by the compiler and is in the range of 363 -- Natural. 364 365 -- vN+ A violation was detected. The compiler cannot determine 366 -- the exact count of violations, but it is at least N. 367 368 -- There are no spaces within the parameter string, so the entry 369 -- described above in the header of this section for Max_Tasks would 370 -- appear as the string r4v3. 371 372 -- Note: The restrictions line is required to be present. Even in 373 -- Ignore_Errors mode, Scan_ALI expects to find an R line and will 374 -- signal a fatal error if it is missing. This means that future 375 -- changes to the ALI file format must retain the R line. 376 377 -- ---------------------------------- 378 -- -- R Restrictions (Named Form) -- 379 -- ---------------------------------- 380 381 -- The first R line for named form announces that named notation will be 382 -- used, and also assures that there is at least one R line present, which 383 -- makes parsing of ali files simpler. A blank line preceds the RN line. 384 385 -- RN 386 387 -- In named notation, the restrictions are given as a series of lines, one 388 -- per retrictions that is specified or violated (no information is present 389 -- for restrictions that are not specified or violated). In the following 390 -- name is the name of the restriction in all upper case. 391 392 -- For boolean restrictions, we have only two possibilities. A restrictions 393 -- pragma is present, or a violation is detected: 394 395 -- RR name 396 397 -- A restriction pragma is present for the named boolean restriction. 398 -- No violations were detected by the compiler (or the unit in question 399 -- would have been found to be illegal). 400 401 -- RV name 402 403 -- No restriction pragma is present for the named boolean restriction. 404 -- However, the compiler did detect one or more violations of this 405 -- restriction, which may require a binder consistency check. Note that 406 -- one case of a violation is the use of a Restriction_Set attribute for 407 -- the restriction that yielded False. 408 409 -- For the case of restrictions that take a parameter, we need both the 410 -- information from pragma if present, and the actual information about 411 -- what possible violations occur. For example, we can have a unit with 412 -- a pragma Restrictions (Max_Tasks => 4), where the compiler can detect 413 -- that there are exactly three tasks declared. Both of these pieces 414 -- of information must be passed to the binder. The parameter of 4 is 415 -- important in case the total number of tasks in the partition is greater 416 -- than 4. The parameter of 3 is important in case some other unit has a 417 -- restrictions pragma with Max_Tasks=>2. 418 419 -- RR name=N 420 421 -- A restriction pragma is present for the named restriction which is 422 -- one of the restrictions taking a parameter. The value N (a decimal 423 -- integer) is the value given in the restriction pragma. 424 425 -- RV name=N 426 427 -- A restriction pragma may or may not be present for the restriction 428 -- given by name (one of the restrictions taking a parameter). But in 429 -- either case, the compiler detected possible violations. N (a decimal 430 -- integer) is the maximum or total count of violations (depending 431 -- on the checking type) in all the units represented by the ali file). 432 -- The value here is known to be exact by the compiler and is in the 433 -- range of Natural. Note that if an RR line is present for the same 434 -- restriction, then the value in the RV line cannot exceed the value 435 -- in the RR line (since otherwise the compiler would have detected a 436 -- violation of the restriction). 437 438 -- RV name=N+ 439 440 -- Similar to the above, but the compiler cannot determine the exact 441 -- count of violations, but it is at least N. 442 443 -- ------------------------------------------------- 444 -- -- R Restrictions (No_Dependence Information) -- 445 -- ------------------------------------------------- 446 447 -- Subsequent R lines are present only if pragma Restriction No_Dependence 448 -- is used. There is one such line for each such pragma appearing in the 449 -- extended main unit. The format is: 450 451 -- R unit_name 452 453 -- Here the unit name is in all lower case. The components of the unit 454 -- name are separated by periods. The names themselves are in encoded 455 -- form, as documented in Namet. 456 457 -- ------------------------- 458 -- -- I Interrupt States -- 459 -- ------------------------- 460 461 -- I interrupt-number interrupt-state line-number 462 463 -- This line records information from an Interrupt_State pragma. There 464 -- is one line for each separate pragma, and if no such pragmas are 465 -- used, then no I lines are present. 466 467 -- The interrupt-number is an unsigned positive integer giving the 468 -- value of the interrupt as defined in Ada.Interrupts.Names. 469 470 -- The interrupt-state is one of r/s/u for Runtime/System/User 471 472 -- The line number is an unsigned decimal integer giving the line 473 -- number of the corresponding Interrupt_State pragma. This is used 474 -- in consistency messages. 475 476 -- -------------------------------------- 477 -- -- S Priority Specific Dispatching -- 478 -- -------------------------------------- 479 480 -- S policy_identifier first_priority last_priority line-number 481 482 -- This line records information from a Priority_Specific_Dispatching 483 -- pragma. There is one line for each separate pragma, and if no such 484 -- pragmas are used, then no S lines are present. 485 486 -- The policy_identifier is the first character (upper case) of the 487 -- corresponding policy name (e.g. 'F' for FIFO_Within_Priorities). 488 489 -- The first_priority and last_priority fields define the range of 490 -- priorities to which the specified dispatching policy apply. 491 492 -- The line number is an unsigned decimal integer giving the line 493 -- number of the corresponding Priority_Specific_Dispatching pragma. 494 -- This is used in consistency messages. 495 496 ---------------------------- 497 -- Compilation Unit Lines -- 498 ---------------------------- 499 500 -- Following these header lines, a set of information lines appears for 501 -- each compilation unit that appears in the corresponding object file. In 502 -- particular, when a package body or subprogram body is compiled, there 503 -- will be two sets of information, one for the spec and one for the body, 504 -- with the entry for the body appearing first. This is the only case in 505 -- which a single ALI file contains more than one unit (in particular note 506 -- that subunits do *not* count as compilation units for this purpose, and 507 -- generate no library information, since they are inlined). 508 509 -- -------------------- 510 -- -- U Unit Header -- 511 -- -------------------- 512 513 -- The lines for each compilation unit have the following form 514 515 -- U unit-name source-name version <<attributes>> 516 517 -- This line identifies the unit to which this section of the library 518 -- information file applies. The first three parameters are the unit 519 -- name in internal format, as described in package Uname, and the name 520 -- of the source file containing the unit. 521 522 -- Version is the version given as eight hexadecimal characters with 523 -- upper case letters. This value is the exclusive or of the source 524 -- checksums of the unit and all its semantically dependent units. 525 526 -- The <<attributes>> are a series of two letter codes indicating 527 -- information about the unit: 528 529 -- BD Unit does not have pragma Elaborate_Body, but the elaboration 530 -- circuit has determined that it would be a good idea if this 531 -- pragma were present, since the body of the package contains 532 -- elaboration code that modifies one or more variables in the 533 -- visible part of the package. The binder will try, but does 534 -- not promise, to keep the elaboration of the body close to 535 -- the elaboration of the spec. 536 537 -- DE Dynamic Elaboration. This unit was compiled with the dynamic 538 -- elaboration model, as set by either the -gnatE switch or 539 -- pragma Elaboration_Checks (Dynamic). 540 -- 541 -- EB Unit has pragma Elaborate_Body, or is a generic instance that 542 -- has a body. Set for instances because RM 12.3(20) requires 543 -- that the body be immediately elaborated after the spec (we 544 -- would normally do that anyway, because elaborate spec and 545 -- body together whenever possible, and for an instance it is 546 -- always possible; however setting EB ensures that this is done 547 -- even when using the -p gnatbind switch). 548 549 -- EE Elaboration entity is present which must be set true when 550 -- the unit is elaborated. The name of the elaboration entity is 551 -- formed from the unit name in the usual way. If EE is present, 552 -- then this boolean must be set True as part of the elaboration 553 -- processing routine generated by the binder. Note that EE can 554 -- be set even if NE is set. This happens when the boolean is 555 -- needed solely for checking for the case of access before 556 -- elaboration. 557 558 -- GE Unit is a generic declaration, or corresponding body 559 -- 560 -- IL Unit source uses a style with identifiers in all lower-case 561 -- IU (IL) or all upper case (IU). If the standard mixed-case usage 562 -- is detected, or the compiler cannot determine the style, then 563 -- no I parameter will appear. 564 565 -- IS Initialize_Scalars pragma applies to this unit, or else there 566 -- is at least one use of the Invalid_Value attribute. 567 568 -- KM Unit source uses a style with keywords in mixed case (KM) 569 -- KU or all upper case (KU). If the standard lower-case usage is 570 -- is detected, or the compiler cannot determine the style, then 571 -- no K parameter will appear. 572 573 -- NE Unit has no elaboration routine. All subprogram bodies and 574 -- specs are in this category. Package bodies and specs may or 575 -- may not have NE set, depending on whether or not elaboration 576 -- code is required. Set if N_Compilation_Unit node has flag 577 -- Has_No_Elaboration_Code set. 578 579 -- OL The units in this file are compiled with a local pragma 580 -- Optimize_Alignment, so no consistency requirement applies 581 -- to these units. All internal units have this status since 582 -- they have an automatic default of Optimize_Alignment (Off). 583 -- 584 -- OO Optimize_Alignment (Off) is the default setting for all 585 -- units in this file. All files in the partition that specify 586 -- a default must specify the same default. 587 588 -- OS Optimize_Alignment (Space) is the default setting for all 589 -- units in this file. All files in the partition that specify 590 -- a default must specify the same default. 591 592 -- OT Optimize_Alignment (Time) is the default setting for all 593 -- units in this file. All files in the partition that specify 594 -- a default must specify the same default. 595 596 -- PF The unit has a library-level (package) finalizer 597 598 -- PK Unit is package, rather than a subprogram 599 600 -- PU Unit has pragma Pure 601 602 -- PR Unit has pragma Preelaborate 603 604 -- RA Unit declares a Remote Access to Class-Wide (RACW) type 605 606 -- RC Unit has pragma Remote_Call_Interface 607 608 -- RT Unit has pragma Remote_Types 609 610 -- SP Unit has pragma Shared_Passive. 611 612 -- SU Unit is a subprogram, rather than a package 613 614 -- The attributes may appear in any order, separated by spaces. 615 616 -- ----------------------------- 617 -- -- W, Y and Z Withed Units -- 618 -- ----------------------------- 619 620 -- Following each U line, is a series of lines of the form 621 622 -- W unit-name [source-name lib-name] [E] [EA] [ED] [AD] 623 -- or 624 -- Y unit-name [source-name lib-name] [E] [EA] [ED] [AD] 625 -- or 626 -- Z unit-name [source-name lib-name] [E] [EA] [ED] [AD] 627 628 -- One W line is present for each unit that is mentioned in an explicit 629 -- non-limited with clause by the current unit. One Y line is present 630 -- for each unit that is mentioned in an explicit limited with clause 631 -- by the current unit. One Z line is present for each unit that is 632 -- only implicitly withed by the current unit. The first parameter is 633 -- the unit name in internal format. The second parameter is the file 634 -- name of the file that must be compiled to compile this unit. It is 635 -- usually the file for the body, except for packages which have no 636 -- body. For units that need a body, if the source file for the body 637 -- cannot be found, the file name of the spec is used instead. The 638 -- third parameter is the file name of the library information file 639 -- that contains the results of compiling this unit. The optional 640 -- modifiers are used as follows: 641 642 -- E pragma Elaborate applies to this unit 643 644 -- EA pragma Elaborate_All applies to this unit 645 646 -- ED Elaborate_Desirable set for this unit, which means that there 647 -- is no Elaborate, but the analysis suggests that Program_Error 648 -- may be raised if the Elaborate conditions cannot be satisfied. 649 -- The binder will attempt to treat ED as E if it can. 650 651 -- AD Elaborate_All_Desirable set for this unit, which means that 652 -- there is no Elaborate_All, but the analysis suggests that 653 -- Program_Error may be raised if the Elaborate_All conditions 654 -- cannot be satisfied. The binder will attempt to treat AD as 655 -- EA if it can. 656 657 -- The parameter source-name and lib-name are omitted for the case of a 658 -- generic unit compiled with earlier versions of GNAT which did not 659 -- generate object or ali files for generics. For compatibility in the 660 -- bootstrap path we continue to omit these entries for predefined 661 -- generic units, even though we do now generate object and ali files. 662 663 -- However, in SPARK mode, we always generate source-name and lib-name 664 -- parameters. Bootstrap issues do not apply there, and we need this 665 -- information to properly compute frame conditions of subprograms. 666 667 -- The parameter source-name and lib-name are also omitted for the W 668 -- lines that result from use of a Restriction_Set attribute which gets 669 -- a result of False from a No_Dependence check, in the case where the 670 -- unit is not in the semantic closure. In such a case, the bare W 671 -- line is generated, but no D (dependency) line. This will make the 672 -- binder do the consistency check, but not include the unit in the 673 -- partition closure (unless it is properly With'ed somewhere). 674 675 -- ----------------------- 676 -- -- L Linker_Options -- 677 -- ----------------------- 678 679 -- Following the W lines (if any, or the U line if not), are an optional 680 -- series of lines that indicates the usage of the pragma Linker_Options in 681 -- the associated unit. For each appearance of a pragma Linker_Options (or 682 -- Link_With) in the unit, a line is present with the form: 683 684 -- L "string" 685 686 -- where string is the string from the unit line enclosed in quotes. 687 -- Within the quotes the following can occur: 688 689 -- c graphic characters in range 20-7E other than " or { 690 -- "" indicating a single " character 691 -- {hh} indicating a character whose code is hex hh (0-9,A-F) 692 -- {00} [ASCII.NUL] is used as a separator character 693 -- to separate multiple arguments of a single 694 -- Linker_Options pragma. 695 696 -- For further details, see Stringt.Write_String_Table_Entry. Note that 697 -- wide characters in the form {hhhh} cannot be produced, since pragma 698 -- Linker_Option accepts only String, not Wide_String. 699 700 -- The L lines are required to appear in the same order as the 701 -- corresponding Linker_Options (or Link_With) pragmas appear in the 702 -- source file, so that this order is preserved by the binder in 703 -- constructing the set of linker arguments. 704 705 -- Note: Linker_Options lines never appear in the ALI file generated for 706 -- a predefined generic unit, and there is cicuitry in Sem_Prag to enforce 707 -- this restriction, which is needed because of not generating source name 708 -- and lib name parameters on the with lines for such files, as explained 709 -- above in the section on with lines. 710 711 -- -------------- 712 -- -- N Notes -- 713 -- -------------- 714 715 -- The final section of unit-specific lines contains notes which record 716 -- annotations inserted in source code for processing by external tools 717 -- using pragmas. For each occurrence of any of these pragmas, a line is 718 -- generated with the following syntax: 719 720 -- N x<sloc> [<arg_id>:]<arg> ... 721 722 -- x is one of: 723 -- A pragma Annotate 724 -- C pragma Comment 725 -- I pragma Ident 726 -- T pragma Title 727 -- S pragma Subtitle 728 729 -- <sloc> is the source location of the pragma in line:col format 730 731 -- Successive entries record the pragma_argument_associations. 732 733 -- If a pragma argument identifier is present, the entry is prefixed 734 -- with the pragma argument identifier <arg_id> followed by a colon. 735 736 -- <arg> represents the pragma argument, and has the following 737 -- conventions: 738 739 -- - identifiers are output verbatim 740 -- - static string expressions are output as literals encoded as 741 -- for L lines 742 -- - static integer expressions are output as decimal literals 743 -- - any other expression is replaced by the placeholder "<expr>" 744 745 --------------------- 746 -- Reference Lines -- 747 --------------------- 748 749 -- The reference lines contain information about references from any of the 750 -- units in the compilation (including body version and version attributes, 751 -- linker options pragmas and source dependencies). 752 753 -- ------------------------------------ 754 -- -- E External Version References -- 755 -- ------------------------------------ 756 757 -- One of these lines is present for each use of 'Body_Version or 'Version 758 -- in any of the units of the compilation. These are used by the linker to 759 -- determine which version symbols must be output. The format is simply: 760 761 -- E name 762 763 -- where name is the external name, i.e. the unit name with either a S or a 764 -- B for spec or body version referenced (Body_Version always references 765 -- the body, Version references the Spec, except in the case of a reference 766 -- to a subprogram with no separate spec). Upper half and wide character 767 -- codes are encoded using the same method as in Namet (Uhh for upper half, 768 -- Whhhh for wide character, where hh are hex digits). 769 770 -- --------------------- 771 -- -- D Dependencies -- 772 -- --------------------- 773 774 -- The dependency lines indicate the source files on which the compiled 775 -- units depend. This is used by the binder for consistency checking. 776 -- These lines are also referenced by the cross-reference information. 777 778 -- D source-name time-stamp checksum [subunit-name] line:file-name 779 780 -- source-name also includes preprocessing data file and preprocessing 781 -- definition file. These preprocessing files may be given as full 782 -- path names instead of simple file names. If a full path name 783 -- includes a directory with spaces, the path name is quoted (quote 784 -- characters (") added at start and end, and any internal quotes are 785 -- doubled). 786 787 -- The time-stamp field contains the time stamp of the corresponding 788 -- source file. See types.ads for details on time stamp representation. 789 790 -- The checksum is an 8-hex digit representation of the source file 791 -- checksum, with letters given in lower case. 792 793 -- The subunit name is present only if the dependency line is for a 794 -- subunit. It contains the fully qualified name of the subunit in all 795 -- lower case letters. 796 797 -- The line:file-name entry is present only if a Source_Reference 798 -- pragma appeared in the source file identified by source-name. In 799 -- this case, it gives the information from this pragma. Note that this 800 -- allows cross-reference information to be related back to the 801 -- original file. Note: the reason the line number comes first is that 802 -- a leading digit immediately identifies this as a Source_Reference 803 -- entry, rather than a subunit-name. 804 805 -- A line number of zero for line: in this entry indicates that there 806 -- is more than one source reference pragma. In this case, the line 807 -- numbers in the cross-reference are correct, and refer to the 808 -- original line number, but there is no information that allows a 809 -- reader of the ALI file to determine the exact mapping of physical 810 -- line numbers back to the original source. 811 812 -- Files with a zero checksum and a non-zero time stamp are in general 813 -- files on which the compilation depends but which are not Ada files 814 -- with further dependencies. This includes preprocessor data files 815 -- and preprocessor definition files. 816 817 -- Note: blank lines are ignored when the library information is read, 818 -- and separate sections of the file are separated by blank lines to 819 -- ease readability. Blanks between fields are also ignored. 820 821 -- For entries corresponding to files that were not present (and thus 822 -- resulted in error messages), or for files that are not part of the 823 -- dependency set, both the time stamp and checksum are set to all zero 824 -- characters. These dummy entries are ignored by the binder in 825 -- dependency checking, but must be present for proper interpretation 826 -- of the cross-reference data. 827 828 -------------------------- 829 -- Cross-Reference Data -- 830 -------------------------- 831 832 -- The cross-reference data follows the dependency lines. See the spec of 833 -- Lib.Xref in file lib-xref.ads for details on the format of this data. 834 835 --------------------------------- 836 -- Source Coverage Obligations -- 837 --------------------------------- 838 839 -- The Source Coverage Obligation (SCO) information follows the cross- 840 -- reference data. See the spec of Par_SCO in file par_sco.ads for full 841 -- details of the format. 842 843 --------------------------------------- 844 -- SPARK Cross-Reference Information -- 845 --------------------------------------- 846 847 -- The SPARK cross-reference information follows the SCO information. See 848 -- the spec of SPARK_Xrefs in file spark_xrefs.ads for full details of the 849 -- format. 850 851 ------------------------------- 852 -- ALI File Generation for C -- 853 ------------------------------- 854 855 -- The C compiler can also generate ALI files for use by the IDE's in 856 -- providing navigation services in C. These ALI files are a subset of 857 -- the specification above, lacking all Ada-specific output. Primarily 858 -- the IDE uses the cross-reference sections of such files. 859 860 ---------------------- 861 -- Global Variables -- 862 ---------------------- 863 864 -- The table defined here stores one entry for each Interrupt_State pragma 865 -- encountered either in the main source or in an ancillary with'ed source. 866 -- Since interrupt state values have to be consistent across all units in a 867 -- partition, we detect inconsistencies at compile time when we can. 868 869 type Interrupt_State_Entry is record 870 Interrupt_Number : Pos; 871 -- Interrupt number value 872 873 Interrupt_State : Character; 874 -- Set to r/s/u for Runtime/System/User 875 876 Pragma_Loc : Source_Ptr; 877 -- Location of pragma setting this value in place 878 end record; 879 880 package Interrupt_States is new Table.Table ( 881 Table_Component_Type => Interrupt_State_Entry, 882 Table_Index_Type => Nat, 883 Table_Low_Bound => 1, 884 Table_Initial => 30, 885 Table_Increment => 200, 886 Table_Name => "Name_Interrupt_States"); 887 888 -- The table structure defined here stores one entry for each 889 -- Priority_Specific_Dispatching pragma encountered either in the main 890 -- source or in an ancillary with'ed source. Since have to be consistent 891 -- across all units in a partition, we may as well detect inconsistencies 892 -- at compile time when we can. 893 894 type Specific_Dispatching_Entry is record 895 Dispatching_Policy : Character; 896 -- First character (upper case) of the corresponding policy name 897 898 First_Priority : Nat; 899 -- Lower bound of the priority range to which the specified dispatching 900 -- policy applies. 901 902 Last_Priority : Nat; 903 -- Upper bound of the priority range to which the specified dispatching 904 -- policy applies. 905 906 Pragma_Loc : Source_Ptr; 907 -- Location of pragma setting this value in place 908 end record; 909 910 package Specific_Dispatching is new Table.Table ( 911 Table_Component_Type => Specific_Dispatching_Entry, 912 Table_Index_Type => Nat, 913 Table_Low_Bound => 1, 914 Table_Initial => 10, 915 Table_Increment => 100, 916 Table_Name => "Name_Priority_Specific_Dispatching"); 917 918 ----------------- 919 -- Subprograms -- 920 ----------------- 921 922 procedure Ensure_System_Dependency; 923 -- This procedure ensures that a dependency is created on system.ads. Even 924 -- if there is no semantic dependency, Targparm has read the file to 925 -- acquire target parameters, so we need a source dependency. 926 927 procedure Write_ALI (Object : Boolean); 928 -- This procedure writes the library information for the current main unit 929 -- The Object parameter is true if an object file is created, and false 930 -- otherwise. 931 -- 932 -- Note: in the case where we are not generating code (-gnatc mode), this 933 -- routine only writes an ALI file if it cannot find an existing up to 934 -- date ALI file. If it *can* find an existing up to date ALI file, then 935 -- it reads this file and sets the Lib.Compilation_Arguments table from 936 -- the A lines in this file. 937 938 procedure Add_Preprocessing_Dependency (S : Source_File_Index); 939 -- Indicate that there is a dependency to be added on a preprocessing data 940 -- file or on a preprocessing definition file. 941 942end Lib.Writ; 943