1------------------------------------------------------------------------------ 2-- -- 3-- GNAT COMPILER COMPONENTS -- 4-- -- 5-- C H E C K S -- 6-- -- 7-- S p e c -- 8-- -- 9-- Copyright (C) 1992-2019, 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-- Package containing routines used to deal with runtime checks. These 27-- routines are used both by the semantics and by the expander. In some 28-- cases, checks are enabled simply by setting flags for gigi, and in 29-- other cases the code for the check is expanded. 30 31-- The approach used for range and length checks, in regards to suppressed 32-- checks, is to attempt to detect at compilation time that a constraint 33-- error will occur. If this is detected a warning or error is issued and the 34-- offending expression or statement replaced with a constraint error node. 35-- This always occurs whether checks are suppressed or not. Dynamic range 36-- checks are, of course, not inserted if checks are suppressed. 37 38with Errout; use Errout; 39with Namet; use Namet; 40with Table; 41with Types; use Types; 42with Uintp; use Uintp; 43with Urealp; use Urealp; 44 45package Checks is 46 47 procedure Initialize; 48 -- Called for each new main source program, to initialize internal 49 -- variables used in the package body of the Checks unit. 50 51 function Access_Checks_Suppressed (E : Entity_Id) return Boolean; 52 function Accessibility_Checks_Suppressed (E : Entity_Id) return Boolean; 53 function Alignment_Checks_Suppressed (E : Entity_Id) return Boolean; 54 function Allocation_Checks_Suppressed (E : Entity_Id) return Boolean; 55 function Atomic_Synchronization_Disabled (E : Entity_Id) return Boolean; 56 function Discriminant_Checks_Suppressed (E : Entity_Id) return Boolean; 57 function Division_Checks_Suppressed (E : Entity_Id) return Boolean; 58 function Duplicated_Tag_Checks_Suppressed (E : Entity_Id) return Boolean; 59 function Elaboration_Checks_Suppressed (E : Entity_Id) return Boolean; 60 function Index_Checks_Suppressed (E : Entity_Id) return Boolean; 61 function Length_Checks_Suppressed (E : Entity_Id) return Boolean; 62 function Overflow_Checks_Suppressed (E : Entity_Id) return Boolean; 63 function Predicate_Checks_Suppressed (E : Entity_Id) return Boolean; 64 function Range_Checks_Suppressed (E : Entity_Id) return Boolean; 65 function Storage_Checks_Suppressed (E : Entity_Id) return Boolean; 66 function Tag_Checks_Suppressed (E : Entity_Id) return Boolean; 67 function Validity_Checks_Suppressed (E : Entity_Id) return Boolean; 68 -- These functions check to see if the named check is suppressed, either 69 -- by an active scope suppress setting, or because the check has been 70 -- specifically suppressed for the given entity. If no entity is relevant 71 -- for the current check, then Empty is used as an argument. Note: the 72 -- reason we insist on specifying Empty is to force the caller to think 73 -- about whether there is any relevant entity that should be checked. 74 75 function Is_Check_Suppressed (E : Entity_Id; C : Check_Id) return Boolean; 76 -- This function is called if Checks_May_Be_Suppressed (E) is True to 77 -- determine whether check C is suppressed either on the entity E or 78 -- as the result of a scope suppress pragma. If Checks_May_Be_Suppressed 79 -- is False, then the status of the check can be determined simply by 80 -- examining Scope_Suppress, so this routine is not called in that case. 81 82 function Overflow_Check_Mode return Overflow_Mode_Type; 83 -- Returns current overflow checking mode, taking into account whether 84 -- we are inside an assertion expression and the assertion policy. 85 86 ----------------------------------------- 87 -- Control of Alignment Check Warnings -- 88 ----------------------------------------- 89 90 -- When we have address clauses, there is an issue of whether the address 91 -- specified is appropriate to the alignment. In the general case where the 92 -- address is dynamic, we generate a check and a possible warning (this 93 -- warning occurs for example if we have a restricted run time with the 94 -- restriction No_Exception_Propagation). We also issue this warning in 95 -- the case where the address is static, but we don't know the alignment 96 -- at the time we process the address clause. In such a case, we issue the 97 -- warning, but we may be able to find out later (after the back end has 98 -- annotated the actual alignment chosen) that the warning was not needed. 99 100 -- To deal with deleting these potentially annoying warnings, we save the 101 -- warning information in a table, and then delete the waranings in the 102 -- post compilation validation stage if we can tell that the check would 103 -- never fail (in general the back end will also optimize away the check 104 -- in such cases). 105 106 -- Table used to record information 107 108 type Alignment_Warnings_Record is record 109 E : Entity_Id; 110 -- Entity whose alignment possibly warrants a warning 111 112 A : Uint; 113 -- Compile time known value of address clause for which the alignment 114 -- is to be checked once we know the alignment. 115 116 W : Error_Msg_Id; 117 -- Id of warning message we might delete 118 end record; 119 120 package Alignment_Warnings is new Table.Table ( 121 Table_Component_Type => Alignment_Warnings_Record, 122 Table_Index_Type => Int, 123 Table_Low_Bound => 0, 124 Table_Initial => 10, 125 Table_Increment => 200, 126 Table_Name => "Alignment_Warnings"); 127 128 procedure Validate_Alignment_Check_Warnings; 129 -- This routine is called after back annotation of type data to delete any 130 -- alignment warnings that turn out to be false alarms, based on knowing 131 -- the actual alignment, and a compile-time known alignment value. 132 133 ------------------------------------------- 134 -- Procedures to Activate Checking Flags -- 135 ------------------------------------------- 136 137 procedure Activate_Division_Check (N : Node_Id); 138 pragma Inline (Activate_Division_Check); 139 -- Sets Do_Division_Check flag in node N, and handles possible local raise. 140 -- Always call this routine rather than calling Set_Do_Division_Check to 141 -- set an explicit value of True, to ensure handling the local raise case. 142 143 procedure Activate_Overflow_Check (N : Node_Id); 144 pragma Inline (Activate_Overflow_Check); 145 -- Sets Do_Overflow_Check flag in node N, and handles possible local raise. 146 -- Always call this routine rather than calling Set_Do_Overflow_Check to 147 -- set an explicit value of True, to ensure handling the local raise case. 148 -- Note that for discrete types, this call has no effect for MOD, REM, and 149 -- unary "+" for which overflow is never possible in any case. 150 -- 151 -- Note: for the discrete-type case, it is legitimate to call this routine 152 -- on an unanalyzed node where the Etype field is not set. However, for the 153 -- floating-point case, Etype must be set (to a floating-point type). 154 -- 155 -- For floating-point, we set the flag if we have automatic overflow checks 156 -- on the target, or if Check_Float_Overflow mode is set. For the floating- 157 -- point case, we ignore all the unary operators ("+", "-", and abs) since 158 -- none of these can result in overflow. If there are no overflow checks on 159 -- the target, and Check_Float_Overflow mode is not set, then the call has 160 -- no effect, since in such cases we want to generate NaN's and infinities. 161 162 procedure Activate_Range_Check (N : Node_Id); 163 pragma Inline (Activate_Range_Check); 164 -- Sets Do_Range_Check flag in node N, and handles possible local raise 165 -- Always call this routine rather than calling Set_Do_Range_Check to 166 -- set an explicit value of True, to ensure handling the local raise case. 167 168 -------------------------------- 169 -- Procedures to Apply Checks -- 170 -------------------------------- 171 172 -- General note on following checks. These checks are always active if 173 -- Expander_Active and not Inside_A_Generic. They are inactive and have 174 -- no effect Inside_A_Generic. In the case where not Expander_Active 175 -- and not Inside_A_Generic, most of them are inactive, but some of them 176 -- operate anyway since they may generate useful compile time warnings. 177 178 procedure Apply_Access_Check (N : Node_Id); 179 -- Determines whether an expression node requires a runtime access 180 -- check and if so inserts the appropriate run-time check. 181 182 procedure Apply_Accessibility_Check 183 (N : Node_Id; 184 Typ : Entity_Id; 185 Insert_Node : Node_Id); 186 -- Given a name N denoting an access parameter, emits a run-time 187 -- accessibility check (if necessary), checking that the level of 188 -- the object denoted by the access parameter is not deeper than the 189 -- level of the type Typ. Program_Error is raised if the check fails. 190 -- Insert_Node indicates the node where the check should be inserted. 191 192 procedure Apply_Address_Clause_Check (E : Entity_Id; N : Node_Id); 193 -- E is the entity for an object which has an address clause. If checks 194 -- are enabled, then this procedure generates a check that the specified 195 -- address has an alignment consistent with the alignment of the object, 196 -- raising PE if this is not the case. The resulting check (if one is 197 -- generated) is prepended to the Actions list of N_Freeze_Entity node N. 198 -- Note that the check references E'Alignment, so it cannot be emitted 199 -- before N (its freeze node), otherwise this would cause an illegal 200 -- access before elaboration error in GIGI. For the case of a clear overlay 201 -- situation, we also check that the size of the overlaying object is not 202 -- larger than the overlaid object. 203 204 procedure Apply_Arithmetic_Overflow_Check (N : Node_Id); 205 -- Handle overflow checking for an arithmetic operator. Also handles the 206 -- cases of ELIMINATED and MINIMIZED overflow checking mode. If the mode 207 -- is one of the latter two, then this routine can also be called with 208 -- an if or case expression node to make sure that we properly handle 209 -- overflow checking for dependent expressions. This routine handles 210 -- front end vs back end overflow checks (in the front end case it expands 211 -- the necessary check). Note that divide is handled separately using 212 -- Apply_Divide_Checks. Node N may or may not have Do_Overflow_Check. 213 -- In STRICT mode, there is nothing to do if this flag is off, but in 214 -- MINIMIZED/ELIMINATED mode we still have to deal with possible use 215 -- of doing operations in Long_Long_Integer or Bignum mode. 216 217 procedure Apply_Constraint_Check 218 (N : Node_Id; 219 Typ : Entity_Id; 220 No_Sliding : Boolean := False); 221 -- Top-level procedure, calls all the others depending on the class of 222 -- Typ. Checks that expression N satisfies the constraint of type Typ. 223 -- No_Sliding is only relevant for constrained array types, if set to 224 -- True, it checks that indexes are in range. 225 226 procedure Apply_Discriminant_Check 227 (N : Node_Id; 228 Typ : Entity_Id; 229 Lhs : Node_Id := Empty); 230 -- Given an expression N of a discriminated type, or of an access type 231 -- whose designated type is a discriminanted type, generates a check to 232 -- ensure that the expression can be converted to the subtype given as 233 -- the second parameter. Lhs is empty except in the case of assignments, 234 -- where the target object may be needed to determine the subtype to 235 -- check against (such as the cases of unconstrained formal parameters 236 -- and unconstrained aliased objects). For the case of unconstrained 237 -- formals, the check is performed only if the corresponding actual is 238 -- constrained, i.e., whether Lhs'Constrained is True. 239 240 procedure Apply_Divide_Checks (N : Node_Id); 241 -- The node kind is N_Op_Divide, N_Op_Mod, or N_Op_Rem if either of the 242 -- flags Do_Division_Check or Do_Overflow_Check is set, then this routine 243 -- ensures that the appropriate checks are made. Note that overflow can 244 -- occur in the signed case for the case of the largest negative number 245 -- divided by minus one. This procedure only applies to Integer types. 246 247 procedure Apply_Parameter_Aliasing_Checks 248 (Call : Node_Id; 249 Subp : Entity_Id); 250 -- Given a subprogram call Call, add a check to verify that none of the 251 -- actuals overlap. Subp denotes the subprogram being called. 252 253 procedure Apply_Parameter_Validity_Checks (Subp : Entity_Id); 254 -- Given a subprogram Subp, add both a pre and post condition pragmas that 255 -- verify the proper initialization of scalars in parameters and function 256 -- results. 257 258 procedure Apply_Predicate_Check 259 (N : Node_Id; 260 Typ : Entity_Id; 261 Fun : Entity_Id := Empty); 262 -- N is an expression to which a predicate check may need to be applied for 263 -- Typ, if Typ has a predicate function. When N is an actual in a call, Fun 264 -- is the function being called, which is used to generate a better warning 265 -- if the call leads to an infinite recursion. 266 267 procedure Apply_Type_Conversion_Checks (N : Node_Id); 268 -- N is an N_Type_Conversion node. A type conversion actually involves 269 -- two sorts of checks. The first check is the checks that ensures that 270 -- the operand in the type conversion fits onto the base type of the 271 -- subtype it is being converted to (see RM 4.6 (28)-(50)). The second 272 -- check is there to ensure that once the operand has been converted to 273 -- a value of the target type, this converted value meets the 274 -- constraints imposed by the target subtype (see RM 4.6 (51)). 275 276 procedure Apply_Universal_Integer_Attribute_Checks (N : Node_Id); 277 -- The argument N is an attribute reference node intended for processing 278 -- by gigi. The attribute is one that returns a universal integer, but 279 -- the attribute reference node is currently typed with the expected 280 -- result type. This routine deals with range and overflow checks needed 281 -- to make sure that the universal result is in range. 282 283 function Build_Discriminant_Checks 284 (N : Node_Id; 285 T_Typ : Entity_Id) 286 return Node_Id; 287 -- Subsidiary routine for Apply_Discriminant_Check. Builds the expression 288 -- that compares discriminants of the expression with discriminants of the 289 -- type. Also used directly for membership tests (see Exp_Ch4.Expand_N_In). 290 291 function Convert_From_Bignum (N : Node_Id) return Node_Id; 292 -- Returns result of converting node N from Bignum. The returned value is 293 -- not analyzed, the caller takes responsibility for this. Node N must be 294 -- a subexpression node of type Bignum. The result is Long_Long_Integer. 295 296 function Convert_To_Bignum (N : Node_Id) return Node_Id; 297 -- Returns result of converting node N to Bignum. The returned value is not 298 -- analyzed, the caller takes responsibility for this. Node N must be a 299 -- subexpression node of a signed integer type or Bignum type (if it is 300 -- already a Bignum, the returned value is Relocate_Node (N)). 301 302 procedure Determine_Range 303 (N : Node_Id; 304 OK : out Boolean; 305 Lo : out Uint; 306 Hi : out Uint; 307 Assume_Valid : Boolean := False); 308 -- N is a node for a subexpression. If N is of a discrete type with no 309 -- error indications, and no other peculiarities (e.g. missing Etype), 310 -- then OK is True on return, and Lo and Hi are set to a conservative 311 -- estimate of the possible range of values of N. Thus if OK is True on 312 -- return, the value of the subexpression N is known to lie in the range 313 -- Lo .. Hi (inclusive). For enumeration and character literals the values 314 -- returned are the Pos value in the relevant enumeration type. If the 315 -- expression is not of a discrete type, or some kind of error condition 316 -- is detected, then OK is False on exit, and Lo/Hi are set to No_Uint. 317 -- Thus the significance of OK being False on return is that no useful 318 -- information is available on the range of the expression. Assume_Valid 319 -- determines whether the processing is allowed to assume that values are 320 -- in range of their subtypes. If it is set to True, then this assumption 321 -- is valid, if False, then processing is done using base types to allow 322 -- invalid values. 323 324 procedure Determine_Range_R 325 (N : Node_Id; 326 OK : out Boolean; 327 Lo : out Ureal; 328 Hi : out Ureal; 329 Assume_Valid : Boolean := False); 330 -- Similar to Determine_Range, but for a node N of floating-point type. OK 331 -- is True on return only for IEEE floating-point types and only if we do 332 -- not have to worry about extended precision (i.e. on the x86, we must be 333 -- using -msse2 -mfpmath=sse). At the current time, this is used only in 334 -- GNATprove, though we could consider using it more generally in future. 335 -- For that to happen, the possibility of arguments of infinite or NaN 336 -- value should be taken into account, which is not the case currently. 337 338 procedure Install_Null_Excluding_Check (N : Node_Id); 339 -- Determines whether an access node requires a runtime access check and 340 -- if so inserts the appropriate run-time check. 341 342 procedure Install_Primitive_Elaboration_Check (Subp_Body : Node_Id); 343 -- Insert a check which ensures that subprogram body Subp_Body has been 344 -- properly elaborated. The check is installed only when Subp_Body is the 345 -- body of a nonabstract library-level primitive of a tagged type. Further 346 -- restrictions may apply, see the body for details. 347 348 function Make_Bignum_Block (Loc : Source_Ptr) return Node_Id; 349 -- This function is used by top level overflow checking routines to do a 350 -- mark/release operation on the secondary stack around bignum operations. 351 -- The block created looks like: 352 -- 353 -- declare 354 -- M : Mark_Id := SS_Mark; 355 -- begin 356 -- SS_Release (M); 357 -- end; 358 -- 359 -- The idea is that the caller will insert any needed extra declarations 360 -- after the declaration of M, and any needed statements (in particular 361 -- the bignum operations) before the call to SS_Release, and then do an 362 -- Insert_Action of the whole block (it is returned unanalyzed). The Loc 363 -- parameter is used to supply Sloc values for the constructed tree. 364 365 procedure Minimize_Eliminate_Overflows 366 (N : Node_Id; 367 Lo : out Uint; 368 Hi : out Uint; 369 Top_Level : Boolean); 370 -- This is the main routine for handling MINIMIZED and ELIMINATED overflow 371 -- processing. On entry N is a node whose result is a signed integer 372 -- subtype. The Do_Overflow_Check flag may or may not be set on N. If the 373 -- node is an arithmetic operation, then a range analysis is carried out, 374 -- and there are three possibilities: 375 -- 376 -- The node is left unchanged (apart from expansion of an exponentiation 377 -- operation). This happens if the routine can determine that the result 378 -- is definitely in range. The Do_Overflow_Check flag is turned off in 379 -- this case. 380 -- 381 -- The node is transformed into an arithmetic operation with a result 382 -- type of Long_Long_Integer. 383 -- 384 -- The node is transformed into a function call that calls an appropriate 385 -- function in the System.Bignums package to compute a Bignum result. 386 -- 387 -- In the first two cases, Lo and Hi are set to the bounds of the possible 388 -- range of results, computed as accurately as possible. In the third case 389 -- Lo and Hi are set to No_Uint (there are some cases where we could get an 390 -- advantage from keeping result ranges for Bignum values, but it could use 391 -- a lot of space and is very unlikely to be valuable). 392 -- 393 -- If the node is not an arithmetic operation, then it is unchanged but 394 -- Lo and Hi are still set (to the bounds of the result subtype if nothing 395 -- better can be determined). 396 -- 397 -- Note: this function is recursive, if called with an arithmetic operator, 398 -- recursive calls are made to process the operands using this procedure. 399 -- So we end up doing things top down. Nothing happens to an arithmetic 400 -- expression until this procedure is called on the top level node and 401 -- then the recursive calls process all the children. We have to do it 402 -- this way. If we try to do it bottom up in natural expansion order, then 403 -- there are two problems. First, where do we stash the bounds, and more 404 -- importantly, semantic processing will be messed up. Consider A+B+C where 405 -- A,B,C are all of type integer, if we processed A+B before doing semantic 406 -- analysis of the addition of this result to C, that addition could end up 407 -- with a Long_Long_Integer left operand and an Integer right operand, and 408 -- we would get a semantic error. 409 -- 410 -- The routine is called in three situations if we are operating in either 411 -- MINIMIZED or ELIMINATED modes. 412 -- 413 -- Overflow processing applied to the top node of an expression tree when 414 -- that node is an arithmetic operator. In this case the result is 415 -- converted to the appropriate result type (there is special processing 416 -- when the parent is a conversion, see body for details). 417 -- 418 -- Overflow processing applied to the operands of a comparison operation. 419 -- In this case, the comparison is done on the result Long_Long_Integer 420 -- or Bignum values, without raising any exceptions. 421 -- 422 -- Overflow processing applied to the left operand of a membership test. 423 -- In this case no exception is raised if a Long_Long_Integer or Bignum 424 -- result is outside the range of the type of that left operand (it is 425 -- just that the result of IN is false in that case). 426 -- 427 -- Note that if Bignum values appear, the caller must take care of doing 428 -- the appropriate mark/release operations on the secondary stack. 429 -- 430 -- Top_Level is used to avoid inefficient unnecessary transitions into the 431 -- Bignum domain. If Top_Level is True, it means that the caller will have 432 -- to convert any Bignum value back to Long_Long_Integer, possibly checking 433 -- that the value is in range. This is the normal case for a top level 434 -- operator in a subexpression. There is no point in going into Bignum mode 435 -- to avoid an overflow just so we can check for overflow the next moment. 436 -- For calls from comparisons and membership tests, and for all recursive 437 -- calls, we do want to transition into the Bignum domain if necessary. 438 -- Note that this setting is only relevant in ELIMINATED mode. 439 440 ------------------------------------------------------- 441 -- Control and Optimization of Range/Overflow Checks -- 442 ------------------------------------------------------- 443 444 -- Range checks are controlled by the Do_Range_Check flag. The front end 445 -- is responsible for setting this flag in relevant nodes. Originally 446 -- the back end generated all corresponding range checks. But later on 447 -- we decided to generate many range checks in the front end. We are now 448 -- in the transitional phase where some of these checks are still done 449 -- by the back end, but many are done by the front end. It is possible 450 -- that in the future we might move all the checks to the front end. The 451 -- main remaining back end checks are for subscript checking. 452 453 -- Overflow checks are similarly controlled by the Do_Overflow_Check flag. 454 -- The difference here is that if back end overflow checks are inactive 455 -- (Backend_Overflow_Checks_On_Target set False), then the actual overflow 456 -- checks are generated by the front end, but if back end overflow checks 457 -- are active (Backend_Overflow_Checks_On_Target set True), then the back 458 -- end does generate the checks. 459 460 -- The following two routines are used to set these flags, they allow 461 -- for the possibility of eliminating checks. Checks can be eliminated 462 -- if an identical check has already been performed. 463 464 procedure Enable_Overflow_Check (N : Node_Id); 465 -- First this routine determines if an overflow check is needed by doing 466 -- an appropriate range check. If a check is not needed, then the call 467 -- has no effect. If a check is needed then this routine sets the flag 468 -- Do_Overflow_Check in node N to True, unless it can be determined that 469 -- the check is not needed. The only condition under which this is the 470 -- case is if there was an identical check earlier on. 471 472 procedure Enable_Range_Check (N : Node_Id); 473 -- Set Do_Range_Check flag in node N True, unless it can be determined 474 -- that the check is not needed. The only condition under which this is 475 -- the case is if there was an identical check earlier on. This routine 476 -- is not responsible for doing range analysis to determine whether or 477 -- not such a check is needed -- the caller is expected to do this. The 478 -- one other case in which the request to set the flag is ignored is 479 -- when Kill_Range_Check is set in an N_Unchecked_Conversion node. 480 481 -- The following routines are used to keep track of processing sequences 482 -- of statements (e.g. the THEN statements of an IF statement). A check 483 -- that appears within such a sequence can eliminate an identical check 484 -- within this sequence of statements. However, after the end of the 485 -- sequence of statements, such a check is no longer of interest, since 486 -- it may not have been executed. 487 488 procedure Conditional_Statements_Begin; 489 -- This call marks the start of processing of a sequence of statements. 490 -- Every call to this procedure must be followed by a matching call to 491 -- Conditional_Statements_End. 492 493 procedure Conditional_Statements_End; 494 -- This call removes from consideration all saved checks since the 495 -- corresponding call to Conditional_Statements_Begin. These two 496 -- procedures operate in a stack like manner. 497 498 -- The mechanism for optimizing checks works by remembering checks 499 -- that have already been made, but certain conditions, for example 500 -- an assignment to a variable involved in a check, may mean that the 501 -- remembered check is no longer valid, in the sense that if the same 502 -- expression appears again, another check is required because the 503 -- value may have changed. 504 505 -- The following routines are used to note conditions which may render 506 -- some or all of the stored and remembered checks to be invalidated. 507 508 procedure Kill_Checks (V : Entity_Id); 509 -- This procedure records an assignment or other condition that causes 510 -- the value of the variable to be changed, invalidating any stored 511 -- checks that reference the value. Note that all such checks must 512 -- be discarded, even if they are not in the current statement range. 513 514 procedure Kill_All_Checks; 515 -- This procedure kills all remembered checks 516 517 ----------------------------- 518 -- Length and Range Checks -- 519 ----------------------------- 520 521 -- In the following procedures, there are three arguments which have 522 -- a common meaning as follows: 523 524 -- Expr The expression to be checked. If a check is required, 525 -- the appropriate flag will be placed on this node. Whether 526 -- this node is further examined depends on the setting of 527 -- the parameter Source_Typ, as described below. 528 529 -- ??? Apply_Length_Check and Apply_Range_Check do not have an Expr 530 -- formal 531 532 -- ??? Apply_Length_Check and Apply_Range_Check have a Ck_Node formal 533 -- which is undocumented, is it the same as Expr? 534 535 -- Target_Typ The target type on which the check is to be based. For 536 -- example, if we have a scalar range check, then the check 537 -- is that we are in range of this type. 538 539 -- Source_Typ Normally Empty, but can be set to a type, in which case 540 -- this type is used for the check, see below. 541 542 -- The checks operate in one of two modes: 543 544 -- If Source_Typ is Empty, then the node Expr is examined, at the very 545 -- least to get the source subtype. In addition for some of the checks, 546 -- the actual form of the node may be examined. For example, a node of 547 -- type Integer whose actual form is an Integer conversion from a type 548 -- with range 0 .. 3 can be determined to have a value in range 0 .. 3. 549 550 -- If Source_Typ is given, then nothing can be assumed about the Expr, 551 -- and indeed its contents are not examined. In this case the check is 552 -- based on the assumption that Expr can be an arbitrary value of the 553 -- given Source_Typ. 554 555 -- Currently, the only case in which a Source_Typ is explicitly supplied 556 -- is for the case of Out and In_Out parameters, where, for the conversion 557 -- on return (the Out direction), the types must be reversed. This is 558 -- handled by the caller. 559 560 procedure Apply_Length_Check 561 (Ck_Node : Node_Id; 562 Target_Typ : Entity_Id; 563 Source_Typ : Entity_Id := Empty); 564 -- This procedure builds a sequence of declarations to do a length check 565 -- that checks if the lengths of the two arrays Target_Typ and source type 566 -- are the same. The resulting actions are inserted at Node using a call 567 -- to Insert_Actions. 568 -- 569 -- For access types, the Directly_Designated_Type is retrieved and 570 -- processing continues as enumerated above, with a guard against null 571 -- values. 572 -- 573 -- Note: calls to Apply_Length_Check currently never supply an explicit 574 -- Source_Typ parameter, but Apply_Length_Check takes this parameter and 575 -- processes it as described above for consistency with the other routines 576 -- in this section. 577 578 procedure Apply_Range_Check 579 (Ck_Node : Node_Id; 580 Target_Typ : Entity_Id; 581 Source_Typ : Entity_Id := Empty); 582 -- For a Node of kind N_Range, constructs a range check action that tests 583 -- first that the range is not null and then that the range is contained in 584 -- the Target_Typ range. 585 -- 586 -- For scalar types, constructs a range check action that first tests that 587 -- the expression is contained in the Target_Typ range. The difference 588 -- between this and Apply_Scalar_Range_Check is that the latter generates 589 -- the actual checking code against the Etype of the expression. 590 -- 591 -- For constrained array types, construct series of range check actions 592 -- to check that each Expr range is properly contained in the range of 593 -- Target_Typ. 594 -- 595 -- For a type conversion to an unconstrained array type, constructs a range 596 -- check action to check that the bounds of the source type are within the 597 -- constraints imposed by the Target_Typ. 598 -- 599 -- For access types, the Directly_Designated_Type is retrieved and 600 -- processing continues as enumerated above, with a guard against null 601 -- values. 602 -- 603 -- The source type is used by type conversions to unconstrained array 604 -- types to retrieve the corresponding bounds. 605 606 procedure Apply_Static_Length_Check 607 (Expr : Node_Id; 608 Target_Typ : Entity_Id; 609 Source_Typ : Entity_Id := Empty); 610 -- Tries to determine statically whether the two array types source type 611 -- and Target_Typ have the same length. If it can be determined at compile 612 -- time that they do not, then an N_Raise_Constraint_Error node replaces 613 -- Expr, and a warning message is issued. 614 615 procedure Apply_Scalar_Range_Check 616 (Expr : Node_Id; 617 Target_Typ : Entity_Id; 618 Source_Typ : Entity_Id := Empty; 619 Fixed_Int : Boolean := False); 620 -- For scalar types, determines whether an expression node should be 621 -- flagged as needing a runtime range check. If the node requires such a 622 -- check, the Do_Range_Check flag is turned on. The Fixed_Int flag if set 623 -- causes any fixed-point values to be treated as though they were discrete 624 -- values (i.e. the underlying integer value is used). 625 626 type Check_Result is private; 627 -- Type used to return result of Get_Range_Checks call, for later use in 628 -- call to Insert_Range_Checks procedure. 629 630 function Get_Range_Checks 631 (Ck_Node : Node_Id; 632 Target_Typ : Entity_Id; 633 Source_Typ : Entity_Id := Empty; 634 Warn_Node : Node_Id := Empty) return Check_Result; 635 -- Like Apply_Range_Check, except it does not modify anything. Instead 636 -- it returns an encapsulated result of the check operations for later 637 -- use in a call to Insert_Range_Checks. If Warn_Node is non-empty, its 638 -- Sloc is used, in the static case, for the generated warning or error. 639 -- Additionally, it is used rather than Expr (or Low/High_Bound of Expr) 640 -- in constructing the check. 641 642 procedure Append_Range_Checks 643 (Checks : Check_Result; 644 Stmts : List_Id; 645 Suppress_Typ : Entity_Id; 646 Static_Sloc : Source_Ptr; 647 Flag_Node : Node_Id); 648 -- Called to append range checks as returned by a call to Get_Range_Checks. 649 -- Stmts is a list to which either the dynamic check is appended or the 650 -- raise Constraint_Error statement is appended (for static checks). 651 -- Static_Sloc is the Sloc at which the raise CE node points, Flag_Node is 652 -- used as the node at which to set the Has_Dynamic_Check flag. Checks_On 653 -- is a boolean value that says if range and index checking is on or not. 654 655 procedure Insert_Range_Checks 656 (Checks : Check_Result; 657 Node : Node_Id; 658 Suppress_Typ : Entity_Id; 659 Static_Sloc : Source_Ptr := No_Location; 660 Flag_Node : Node_Id := Empty; 661 Do_Before : Boolean := False); 662 -- Called to insert range checks as returned by a call to Get_Range_Checks. 663 -- Node is the node after which either the dynamic check is inserted or 664 -- the raise Constraint_Error statement is inserted (for static checks). 665 -- Suppress_Typ is the type to check to determine if checks are suppressed. 666 -- Static_Sloc, if passed, is the Sloc at which the raise CE node points, 667 -- otherwise Sloc (Node) is used. The Has_Dynamic_Check flag is normally 668 -- set at Node. If Flag_Node is present, then this is used instead as the 669 -- node at which to set the Has_Dynamic_Check flag. Normally the check is 670 -- inserted after, if Do_Before is True, the check is inserted before 671 -- Node. 672 673 ----------------------- 674 -- Expander Routines -- 675 ----------------------- 676 677 -- Some of the earlier processing for checks results in temporarily setting 678 -- the Do_Range_Check flag rather than actually generating checks. Now we 679 -- are moving the generation of such checks into the front end for reasons 680 -- of efficiency and simplicity (there were difficulties in handling this 681 -- in the back end when side effects were present in the expressions being 682 -- checked). 683 684 -- Probably we could eliminate the Do_Range_Check flag entirely and 685 -- generate the checks earlier, but this is a delicate area and it 686 -- seemed safer to implement the following routines, which are called 687 -- late on in the expansion process. They check the Do_Range_Check flag 688 -- and if it is set, generate the actual checks and reset the flag. 689 690 procedure Generate_Range_Check 691 (N : Node_Id; 692 Target_Type : Entity_Id; 693 Reason : RT_Exception_Code); 694 -- This procedure is called to actually generate and insert a range check. 695 -- A check is generated to ensure that the value of N lies within the range 696 -- of the target type. Note that the base type of N may be different from 697 -- the base type of the target type. This happens in the conversion case. 698 -- The Reason parameter is the exception code to be used for the exception 699 -- if raised. 700 -- 701 -- Note: if the expander is not active, or if we are in GNATprove mode, 702 -- then we do not generate explicit range code. Instead we just turn the 703 -- Do_Range_Check flag on, since in these cases that's what we want to see 704 -- in the tree (GNATprove in particular depends on this flag being set). If 705 -- we generate the actual range check, then we make sure the flag is off, 706 -- since the code we generate takes complete care of the check. 707 -- 708 -- Historical note: We used to just pass on the Do_Range_Check flag to the 709 -- back end to generate the check, but now in code-generation mode we never 710 -- have this flag set, since the front end takes care of the check. The 711 -- normal processing flow now is that the analyzer typically turns on the 712 -- Do_Range_Check flag, and if it is set, this routine is called, which 713 -- turns the flag off in code-generation mode. 714 715 procedure Generate_Index_Checks (N : Node_Id); 716 -- This procedure is called to generate index checks on the subscripts for 717 -- the indexed component node N. Each subscript expression is examined, and 718 -- if the Do_Range_Check flag is set, an appropriate index check is 719 -- generated and the flag is reset. 720 721 -- Similarly, we set the flag Do_Discriminant_Check in the semantic 722 -- analysis to indicate that a discriminant check is required for selected 723 -- component of a discriminated type. The following routine is called from 724 -- the expander to actually generate the call. 725 726 procedure Generate_Discriminant_Check (N : Node_Id); 727 -- N is a selected component for which a discriminant check is required to 728 -- make sure that the discriminants have appropriate values for the 729 -- selection. This is done by calling the appropriate discriminant checking 730 -- routine for the selector. 731 732 ----------------------- 733 -- Validity Checking -- 734 ----------------------- 735 736 -- In (RM 13.9.1(9-11)) we have the following rules on invalid values 737 738 -- If the representation of a scalar object does not represent value of 739 -- the object's subtype (perhaps because the object was not initialized), 740 -- the object is said to have an invalid representation. It is a bounded 741 -- error to evaluate the value of such an object. If the error is 742 -- detected, either Constraint_Error or Program_Error is raised. 743 -- Otherwise, execution continues using the invalid representation. The 744 -- rules of the language outside this subclause assume that all objects 745 -- have valid representations. The semantics of operations on invalid 746 -- representations are as follows: 747 -- 748 -- 10 If the representation of the object represents a value of the 749 -- object's type, the value of the type is used. 750 -- 751 -- 11 If the representation of the object does not represent a value 752 -- of the object's type, the semantics of operations on such 753 -- representations is implementation-defined, but does not by 754 -- itself lead to erroneous or unpredictable execution, or to 755 -- other objects becoming abnormal. 756 757 -- We quote the rules in full here since they are quite delicate. Most 758 -- of the time, we can just compute away with wrong values, and get a 759 -- possibly wrong result, which is well within the range of allowed 760 -- implementation defined behavior. The two tricky cases are subscripted 761 -- array assignments, where we don't want to do wild stores, and case 762 -- statements where we don't want to do wild jumps. 763 764 -- In GNAT, we control validity checking with a switch -gnatV that can take 765 -- three parameters, n/d/f for None/Default/Full. These modes have the 766 -- following meanings: 767 768 -- None (no validity checking) 769 770 -- In this mode, there is no specific checking for invalid values 771 -- and the code generator assumes that all stored values are always 772 -- within the bounds of the object subtype. The consequences are as 773 -- follows: 774 775 -- For case statements, an out of range invalid value will cause 776 -- Constraint_Error to be raised, or an arbitrary one of the case 777 -- alternatives will be executed. Wild jumps cannot result even 778 -- in this mode, since we always do a range check 779 780 -- For subscripted array assignments, wild stores will result in 781 -- the expected manner when addresses are calculated using values 782 -- of subscripts that are out of range. 783 784 -- It could perhaps be argued that this mode is still conformant with 785 -- the letter of the RM, since implementation defined is a rather 786 -- broad category, but certainly it is not in the spirit of the 787 -- RM requirement, since wild stores certainly seem to be a case of 788 -- erroneous behavior. 789 790 -- Default (default standard RM-compatible validity checking) 791 792 -- In this mode, which is the default, minimal validity checking is 793 -- performed to ensure no erroneous behavior as follows: 794 795 -- For case statements, an out of range invalid value will cause 796 -- Constraint_Error to be raised. 797 798 -- For subscripted array assignments, invalid out of range 799 -- subscript values will cause Constraint_Error to be raised. 800 801 -- Full (Full validity checking) 802 803 -- In this mode, the protections guaranteed by the standard mode are 804 -- in place, and the following additional checks are made: 805 806 -- For every assignment, the right side is checked for validity 807 808 -- For every call, IN and IN OUT parameters are checked for validity 809 810 -- For every subscripted array reference, both for stores and loads, 811 -- all subscripts are checked for validity. 812 813 -- These checks are not required by the RM, but will in practice 814 -- improve the detection of uninitialized variables, particularly 815 -- if used in conjunction with pragma Normalize_Scalars. 816 817 -- In the above description, we talk about performing validity checks, 818 -- but we don't actually generate a check in a case where the compiler 819 -- can be sure that the value is valid. Note that this assurance must 820 -- be achieved without assuming that any uninitialized value lies within 821 -- the range of its type. The following are cases in which values are 822 -- known to be valid. The flag Is_Known_Valid is used to keep track of 823 -- some of these cases. 824 825 -- If all possible stored values are valid, then any uninitialized 826 -- value must be valid. 827 828 -- Literals, including enumeration literals, are clearly always valid 829 830 -- Constants are always assumed valid, with a validity check being 831 -- performed on the initializing value where necessary to ensure that 832 -- this is the case. 833 834 -- For variables, the status is set to known valid if there is an 835 -- initializing expression. Again a check is made on the initializing 836 -- value if necessary to ensure that this assumption is valid. The 837 -- status can change as a result of local assignments to a variable. 838 -- If a known valid value is unconditionally assigned, then we mark 839 -- the left side as known valid. If a value is assigned that is not 840 -- known to be valid, then we mark the left side as invalid. This 841 -- kind of processing does NOT apply to non-local variables since we 842 -- are not following the flow graph (more properly the flow of actual 843 -- processing only corresponds to the flow graph for local assignments). 844 -- For non-local variables, we preserve the current setting, i.e. a 845 -- validity check is performed when assigning to a knonwn valid global. 846 847 -- Note: no validity checking is required if range checks are suppressed 848 -- regardless of the setting of the validity checking mode. 849 850 -- The following procedures are used in handling validity checking 851 852 procedure Apply_Subscript_Validity_Checks (Expr : Node_Id); 853 -- Expr is the node for an indexed component. If validity checking and 854 -- range checking are enabled, all subscripts for this indexed component 855 -- are checked for validity. 856 857 procedure Check_Valid_Lvalue_Subscripts (Expr : Node_Id); 858 -- Expr is a lvalue, i.e. an expression representing the target of an 859 -- assignment. This procedure checks for this expression involving an 860 -- assignment to an array value. We have to be sure that all the subscripts 861 -- in such a case are valid, since according to the rules in (RM 862 -- 13.9.1(9-11)) such assignments are not permitted to result in erroneous 863 -- behavior in the case of invalid subscript values. 864 865 procedure Ensure_Valid 866 (Expr : Node_Id; 867 Holes_OK : Boolean := False; 868 Related_Id : Entity_Id := Empty; 869 Is_Low_Bound : Boolean := False; 870 Is_High_Bound : Boolean := False); 871 -- Ensure that Expr represents a valid value of its type. If this type 872 -- is not a scalar type, then the call has no effect, since validity 873 -- is only an issue for scalar types. The effect of this call is to 874 -- check if the value is known valid, if so, nothing needs to be done. 875 -- If this is not known, then either Expr is set to be range checked, 876 -- or specific checking code is inserted so that an exception is raised 877 -- if the value is not valid. 878 -- 879 -- The optional argument Holes_OK indicates whether it is necessary to 880 -- worry about enumeration types with non-standard representations leading 881 -- to "holes" in the range of possible representations. If Holes_OK is 882 -- True, then such values are assumed valid (this is used when the caller 883 -- will make a separate check for this case anyway). If Holes_OK is False, 884 -- then this case is checked, and code is inserted to ensure that Expr is 885 -- valid, raising Constraint_Error if the value is not valid. 886 -- 887 -- Related_Id denotes the entity of the context where Expr appears. Flags 888 -- Is_Low_Bound and Is_High_Bound specify whether the expression to check 889 -- is the low or the high bound of a range. These three optional arguments 890 -- signal Remove_Side_Effects to create an external symbol of the form 891 -- Chars (Related_Id)_FIRST/_LAST. For suggested use of these parameters 892 -- see the warning in the body of Sem_Ch3.Process_Range_Expr_In_Decl. 893 894 function Expr_Known_Valid (Expr : Node_Id) return Boolean; 895 -- This function tests it the value of Expr is known to be valid in the 896 -- sense of RM 13.9.1(9-11). In the case of GNAT, it is only discrete types 897 -- which are a concern, since for non-discrete types we simply continue 898 -- computation with invalid values, which does not lead to erroneous 899 -- behavior. Thus Expr_Known_Valid always returns True if the type of Expr 900 -- is non-discrete. For discrete types the value returned is True only if 901 -- it can be determined that the value is Valid. Otherwise False is 902 -- returned. 903 904 procedure Insert_Valid_Check 905 (Expr : Node_Id; 906 Related_Id : Entity_Id := Empty; 907 Is_Low_Bound : Boolean := False; 908 Is_High_Bound : Boolean := False); 909 -- Inserts code that will check for the value of Expr being valid, in the 910 -- sense of the 'Valid attribute returning True. Constraint_Error will be 911 -- raised if the value is not valid. 912 -- 913 -- Related_Id denotes the entity of the context where Expr appears. Flags 914 -- Is_Low_Bound and Is_High_Bound specify whether the expression to check 915 -- is the low or the high bound of a range. These three optional arguments 916 -- signal Remove_Side_Effects to create an external symbol of the form 917 -- Chars (Related_Id)_FIRST/_LAST. For suggested use of these parameters 918 -- see the warning in the body of Sem_Ch3.Process_Range_Expr_In_Decl. 919 920 procedure Null_Exclusion_Static_Checks 921 (N : Node_Id; 922 Comp : Node_Id := Empty; 923 Array_Comp : Boolean := False); 924 -- Ada 2005 (AI-231): Test for and warn on null-excluding objects or 925 -- components that will raise an exception due to initialization by null. 926 -- 927 -- When a value for Comp is supplied (as in the case of an uninitialized 928 -- null-excluding component within a composite object), a reported warning 929 -- will indicate the offending component instead of the object itself. 930 -- Array_Comp being True indicates an array object with null-excluding 931 -- components, and any reported warning will indicate that. 932 933 procedure Remove_Checks (Expr : Node_Id); 934 -- Remove all checks from Expr except those that are only executed 935 -- conditionally (on the right side of And Then/Or Else. This call 936 -- removes only embedded checks (Do_Range_Check, Do_Overflow_Check). 937 938 procedure Validity_Check_Range 939 (N : Node_Id; 940 Related_Id : Entity_Id := Empty); 941 -- If N is an N_Range node, then Ensure_Valid is called on its bounds, if 942 -- validity checking of operands is enabled. Related_Id denotes the entity 943 -- of the context where N appears. 944 945 ----------------------------- 946 -- Handling of Check Names -- 947 ----------------------------- 948 949 -- The following table contains Name_Id's for recognized checks. The first 950 -- entries (corresponding to the values of the subtype Predefined_Check_Id) 951 -- contain the Name_Id values for the checks that are predefined, including 952 -- All_Checks (see Types). Remaining entries are those that are introduced 953 -- by pragma Check_Names. 954 955 package Check_Names is new Table.Table ( 956 Table_Component_Type => Name_Id, 957 Table_Index_Type => Check_Id, 958 Table_Low_Bound => 1, 959 Table_Initial => 30, 960 Table_Increment => 200, 961 Table_Name => "Name_Check_Names"); 962 963 function Get_Check_Id (N : Name_Id) return Check_Id; 964 -- Function to search above table for matching name. If found returns the 965 -- corresponding Check_Id value in the range 1 .. Check_Name.Last. If not 966 -- found returns No_Check_Id. 967 968private 969 970 type Check_Result is array (Positive range 1 .. 2) of Node_Id; 971 -- There are two cases for the result returned by Range_Check: 972 -- 973 -- For the static case the result is one or two nodes that should cause 974 -- a Constraint_Error. Typically these will include Expr itself or the 975 -- direct descendants of Expr, such as Low/High_Bound (Expr)). It is the 976 -- responsibility of the caller to rewrite and substitute the nodes with 977 -- N_Raise_Constraint_Error nodes. 978 -- 979 -- For the non-static case a single N_Raise_Constraint_Error node with a 980 -- non-empty Condition field is returned. 981 -- 982 -- Unused entries in Check_Result, if any, are simply set to Empty For 983 -- external clients, the required processing on this result is achieved 984 -- using the Insert_Range_Checks routine. 985 986 pragma Inline (Apply_Length_Check); 987 pragma Inline (Apply_Range_Check); 988 pragma Inline (Apply_Static_Length_Check); 989end Checks; 990