1 ------------------------------------------------------------------------------
3 -- GNAT COMPILER COMPONENTS --
9 -- Copyright (C) 1992-2007, Free Software Foundation, Inc. --
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 2, 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 COPYING. If not, write --
19 -- to the Free Software Foundation, 51 Franklin Street, Fifth Floor, --
20 -- Boston, MA 02110-1301, USA. --
22 -- GNAT was originally developed by the GNAT team at New York University. --
23 -- Extensive contributions were provided by Ada Core Technologies Inc. --
25 ------------------------------------------------------------------------------
27 -- Package containing routines used to deal with runtime checks. These
28 -- routines are used both by the semantics and by the expander. In some
29 -- cases, checks are enabled simply by setting flags for gigi, and in
30 -- other cases the code for the check is expanded.
32 -- The approach used for range and length checks, in regards to suppressed
33 -- checks, is to attempt to detect at compilation time that a constraint
34 -- error will occur. If this is detected a warning or error is issued and the
35 -- offending expression or statement replaced with a constraint error node.
36 -- This always occurs whether checks are suppressed or not. Dynamic range
37 -- checks are, of course, not inserted if checks are suppressed.
39 with Types; use Types;
40 with Uintp; use Uintp;
45 -- Called for each new main source program, to initialize internal
46 -- variables used in the package body of the Checks unit.
48 function Access_Checks_Suppressed (E : Entity_Id) return Boolean;
49 function Accessibility_Checks_Suppressed (E : Entity_Id) return Boolean;
50 function Alignment_Checks_Suppressed (E : Entity_Id) return Boolean;
51 function Discriminant_Checks_Suppressed (E : Entity_Id) return Boolean;
52 function Division_Checks_Suppressed (E : Entity_Id) return Boolean;
53 function Elaboration_Checks_Suppressed (E : Entity_Id) return Boolean;
54 function Index_Checks_Suppressed (E : Entity_Id) return Boolean;
55 function Length_Checks_Suppressed (E : Entity_Id) return Boolean;
56 function Overflow_Checks_Suppressed (E : Entity_Id) return Boolean;
57 function Range_Checks_Suppressed (E : Entity_Id) return Boolean;
58 function Storage_Checks_Suppressed (E : Entity_Id) return Boolean;
59 function Tag_Checks_Suppressed (E : Entity_Id) return Boolean;
60 function Validity_Checks_Suppressed (E : Entity_Id) return Boolean;
61 -- These functions check to see if the named check is suppressed, either
62 -- by an active scope suppress setting, or because the check has been
63 -- specifically suppressed for the given entity. If no entity is relevant
64 -- for the current check, then Empty is used as an argument. Note: the
65 -- reason we insist on specifying Empty is to force the caller to think
66 -- about whether there is any relevant entity that should be checked.
68 -------------------------------------------
69 -- Procedures to Activate Checking Flags --
70 -------------------------------------------
72 procedure Activate_Division_Check (N : Node_Id);
73 pragma Inline (Activate_Division_Check);
74 -- Sets Do_Division_Check flag in node N, and handles possible local raise.
75 -- Always call this routine rather than calling Set_Do_Division_Check to
76 -- set an explicit value of True, to ensure handling the local raise case.
78 procedure Activate_Overflow_Check (N : Node_Id);
79 pragma Inline (Activate_Overflow_Check);
80 -- Sets Do_Overflow_Check flag in node N, and handles possible local raise.
81 -- Always call this routine rather than calling Set_Do_Overflow_Check to
82 -- set an explicit value of True, to ensure handling the local raise case.
84 procedure Activate_Range_Check (N : Node_Id);
85 pragma Inline (Activate_Range_Check);
86 -- Sets Do_Range_Check flag in node N, and handles possible local raise
87 -- Always call this routine rather than calling Set_Do_Range_Check to
88 -- set an explicit value of True, to ensure handling the local raise case.
90 --------------------------------
91 -- Procedures to Apply Checks --
92 --------------------------------
94 -- General note on following checks. These checks are always active if
95 -- Expander_Active and not Inside_A_Generic. They are inactive and have
96 -- no effect Inside_A_Generic. In the case where not Expander_Active
97 -- and not Inside_A_Generic, most of them are inactive, but some of them
98 -- operate anyway since they may generate useful compile time warnings.
100 procedure Apply_Access_Check (N : Node_Id);
101 -- Determines whether an expression node requires a runtime access
102 -- check and if so inserts the appropriate run-time check.
104 procedure Apply_Accessibility_Check (N : Node_Id; Typ : Entity_Id);
105 -- Given a name N denoting an access parameter, emits a run-time
106 -- accessibility check (if necessary), checking that the level of
107 -- the object denoted by the access parameter is not deeper than the
108 -- level of the type Typ. Program_Error is raised if the check fails.
110 procedure Apply_Address_Clause_Check (E : Entity_Id; N : Node_Id);
111 -- E is the entity for an object which has an address clause. If checks
112 -- are enabled, then this procedure generates a check that the specified
113 -- address has an alignment consistent with the alignment of the object,
114 -- raising PE if this is not the case. The resulting check (if one is
115 -- generated) is inserted before node N. check is also made for the case of
116 -- a clear overlay situation that the size of the overlaying object is not
117 -- larger than the overlaid object.
119 procedure Apply_Arithmetic_Overflow_Check (N : Node_Id);
120 -- Given a binary arithmetic operator (+ - *) expand a software integer
121 -- overflow check using range checks on a larger checking type or a call
122 -- to an appropriate runtime routine. This is used for all three operators
123 -- for the signed integer case, and for +/- in the fixed-point case. The
124 -- check is expanded only if Software_Overflow_Checking is enabled and
125 -- Do_Overflow_Check is set on node N. Note that divide is handled
126 -- separately using Apply_Arithmetic_Divide_Overflow_Check.
128 procedure Apply_Constraint_Check
131 No_Sliding : Boolean := False);
132 -- Top-level procedure, calls all the others depending on the class of Typ.
133 -- Checks that expression N verifies the constraint of type Typ. No_Sliding
134 -- is only relevant for constrained array types, id set to true, it
135 -- checks that indexes are in range.
137 procedure Apply_Discriminant_Check
140 Lhs : Node_Id := Empty);
141 -- Given an expression N of a discriminated type, or of an access type
142 -- whose designated type is a discriminanted type, generates a check to
143 -- ensure that the expression can be converted to the subtype given as
144 -- the second parameter. Lhs is empty except in the case of assignments,
145 -- where the target object may be needed to determine the subtype to
146 -- check against (such as the cases of unconstrained formal parameters
147 -- and unconstrained aliased objects). For the case of unconstrained
148 -- formals, the check is peformed only if the corresponding actual is
149 -- constrained, i.e., whether Lhs'Constrained is True.
151 function Build_Discriminant_Checks
155 -- Subsidiary routine for Apply_Discriminant_Check. Builds the expression
156 -- that compares discriminants of the expression with discriminants of the
157 -- type. Also used directly for membership tests (see Exp_Ch4.Expand_N_In).
159 procedure Apply_Divide_Check (N : Node_Id);
160 -- The node kind is N_Op_Divide, N_Op_Mod, or N_Op_Rem. An appropriate
161 -- check is generated to ensure that the right operand is non-zero. In
162 -- the divide case, we also check that we do not have the annoying case
163 -- of the largest negative number divided by minus one.
165 procedure Apply_Type_Conversion_Checks (N : Node_Id);
166 -- N is an N_Type_Conversion node. A type conversion actually involves
167 -- two sorts of checks. The first check is the checks that ensures that
168 -- the operand in the type conversion fits onto the base type of the
169 -- subtype it is being converted to (see RM 4.6 (28)-(50)). The second
170 -- check is there to ensure that once the operand has been converted to
171 -- a value of the target type, this converted value meets the
172 -- constraints imposed by the target subtype (see RM 4.6 (51)).
174 procedure Apply_Universal_Integer_Attribute_Checks (N : Node_Id);
175 -- The argument N is an attribute reference node intended for processing
176 -- by gigi. The attribute is one that returns a universal integer, but
177 -- the attribute reference node is currently typed with the expected
178 -- result type. This routine deals with range and overflow checks needed
179 -- to make sure that the universal result is in range.
181 procedure Determine_Range
186 -- N is a node for a subexpression. If N is of a discrete type with no
187 -- error indications, and no other peculiarities (e.g. missing type
188 -- fields), then OK is True on return, and Lo and Hi are set to a
189 -- conservative estimate of the possible range of values of N. Thus if OK
190 -- is True on return, the value of the subexpression N is known to like in
191 -- the range Lo .. Hi (inclusive). If the expression is not of a discrete
192 -- type, or some kind of error condition is detected, then OK is False on
193 -- exit, and Lo/Hi are set to No_Uint. Thus the significance of OK being
194 -- False on return is that no useful information is available on the range
195 -- of the expression.
197 procedure Install_Null_Excluding_Check (N : Node_Id);
198 -- Determines whether an access node requires a runtime access check and
199 -- if so inserts the appropriate run-time check.
201 -------------------------------------------------------
202 -- Control and Optimization of Range/Overflow Checks --
203 -------------------------------------------------------
205 -- Range checks are controlled by the Do_Range_Check flag. The front end
206 -- is responsible for setting this flag in relevant nodes. Originally
207 -- the back end generated all corresponding range checks. But later on
208 -- we decided to generate all range checks in the front end. We are now
209 -- in the transitional phase where some of these checks are still done
210 -- by the back end, but many are done by the front end.
212 -- Overflow checks are similarly controlled by the Do_Overflow_Check flag.
213 -- The difference here is that if Backend_Overflow_Checks is is
214 -- (Backend_Overflow_Checks_On_Target set False), then the actual overflow
215 -- checks are generated by the front end, but if back end overflow checks
216 -- are active (Backend_Overflow_Checks_On_Target set True), then the back
217 -- end does generate the checks.
219 -- The following two routines are used to set these flags, they allow
220 -- for the possibility of eliminating checks. Checks can be eliminated
221 -- if an identical check has already been performed.
223 procedure Enable_Overflow_Check (N : Node_Id);
224 -- First this routine determines if an overflow check is needed by doing
225 -- an appropriate range check. If a check is not needed, then the call
226 -- has no effect. If a check is needed then this routine sets the flag
227 -- Set Do_Overflow_Check in node N to True, unless it can be determined
228 -- that the check is not needed. The only condition under which this is
229 -- the case is if there was an identical check earlier on.
231 procedure Enable_Range_Check (N : Node_Id);
232 -- Set Do_Range_Check flag in node N True, unless it can be determined
233 -- that the check is not needed. The only condition under which this is
234 -- the case is if there was an identical check earlier on. This routine
235 -- is not responsible for doing range analysis to determine whether or
236 -- not such a check is needed -- the caller is expected to do this. The
237 -- one other case in which the request to set the flag is ignored is
238 -- when Kill_Range_Check is set in an N_Unchecked_Conversion node.
240 -- The following routines are used to keep track of processing sequences
241 -- of statements (e.g. the THEN statements of an IF statement). A check
242 -- that appears within such a sequence can eliminate an identical check
243 -- within this sequence of statements. However, after the end of the
244 -- sequence of statements, such a check is no longer of interest, since
245 -- it may not have been executed.
247 procedure Conditional_Statements_Begin;
248 -- This call marks the start of processing of a sequence of statements.
249 -- Every call to this procedure must be followed by a matching call to
250 -- Conditional_Statements_End.
252 procedure Conditional_Statements_End;
253 -- This call removes from consideration all saved checks since the
254 -- corresponding call to Conditional_Statements_Begin. These two
255 -- procedures operate in a stack like manner.
257 -- The mechanism for optimizing checks works by remembering checks
258 -- that have already been made, but certain conditions, for example
259 -- an assignment to a variable involved in a check, may mean that the
260 -- remembered check is no longer valid, in the sense that if the same
261 -- expression appears again, another check is required because the
262 -- value may have changed.
264 -- The following routines are used to note conditions which may render
265 -- some or all of the stored and remembered checks to be invalidated.
267 procedure Kill_Checks (V : Entity_Id);
268 -- This procedure records an assignment or other condition that causes
269 -- the value of the variable to be changed, invalidating any stored
270 -- checks that reference the value. Note that all such checks must
271 -- be discarded, even if they are not in the current statement range.
273 procedure Kill_All_Checks;
274 -- This procedure kills all remembered checks
276 -----------------------------
277 -- Length and Range Checks --
278 -----------------------------
280 -- In the following procedures, there are three arguments which have
281 -- a common meaning as follows:
283 -- Expr The expression to be checked. If a check is required,
284 -- the appropriate flag will be placed on this node. Whether
285 -- this node is further examined depends on the setting of
286 -- the parameter Source_Typ, as described below.
288 -- ??? Apply_Length_Check and Apply_Range_Check do not have an Expr
291 -- ??? Apply_Length_Check and Apply_Range_Check have a Ck_Node formal
292 -- which is undocumented, is it the same as Expr?
294 -- Target_Typ The target type on which the check is to be based. For
295 -- example, if we have a scalar range check, then the check
296 -- is that we are in range of this type.
298 -- Source_Typ Normally Empty, but can be set to a type, in which case
299 -- this type is used for the check, see below.
301 -- The checks operate in one of two modes:
303 -- If Source_Typ is Empty, then the node Expr is examined, at the very
304 -- least to get the source subtype. In addition for some of the checks,
305 -- the actual form of the node may be examined. For example, a node of
306 -- type Integer whose actual form is an Integer conversion from a type
307 -- with range 0 .. 3 can be determined to have a value in range 0 .. 3.
309 -- If Source_Typ is given, then nothing can be assumed about the Expr,
310 -- and indeed its contents are not examined. In this case the check is
311 -- based on the assumption that Expr can be an arbitrary value of the
314 -- Currently, the only case in which a Source_Typ is explicitly supplied
315 -- is for the case of Out and In_Out parameters, where, for the conversion
316 -- on return (the Out direction), the types must be reversed. This is
317 -- handled by the caller.
319 procedure Apply_Length_Check
321 Target_Typ : Entity_Id;
322 Source_Typ : Entity_Id := Empty);
323 -- This procedure builds a sequence of declarations to do a length check
324 -- that checks if the lengths of the two arrays Target_Typ and source type
325 -- are the same. The resulting actions are inserted at Node using a call
326 -- to Insert_Actions.
328 -- For access types, the Directly_Designated_Type is retrieved and
329 -- processing continues as enumerated above, with a guard against null
332 -- Note: calls to Apply_Length_Check currently never supply an explicit
333 -- Source_Typ parameter, but Apply_Length_Check takes this parameter and
334 -- processes it as described above for consistency with the other routines
337 procedure Apply_Range_Check
339 Target_Typ : Entity_Id;
340 Source_Typ : Entity_Id := Empty);
341 -- For a Node of kind N_Range, constructs a range check action that tests
342 -- first that the range is not null and then that the range is contained in
343 -- the Target_Typ range.
345 -- For scalar types, constructs a range check action that first tests that
346 -- the expression is contained in the Target_Typ range. The difference
347 -- between this and Apply_Scalar_Range_Check is that the latter generates
348 -- the actual checking code in gigi against the Etype of the expression.
350 -- For constrained array types, construct series of range check actions
351 -- to check that each Expr range is properly contained in the range of
354 -- For a type conversion to an unconstrained array type, constructs a range
355 -- check action to check that the bounds of the source type are within the
356 -- constraints imposed by the Target_Typ.
358 -- For access types, the Directly_Designated_Type is retrieved and
359 -- processing continues as enumerated above, with a guard against null
362 -- The source type is used by type conversions to unconstrained array
363 -- types to retrieve the corresponding bounds.
365 procedure Apply_Static_Length_Check
367 Target_Typ : Entity_Id;
368 Source_Typ : Entity_Id := Empty);
369 -- Tries to determine statically whether the two array types source type
370 -- and Target_Typ have the same length. If it can be determined at compile
371 -- time that they do not, then an N_Raise_Constraint_Error node replaces
372 -- Expr, and a warning message is issued.
374 procedure Apply_Scalar_Range_Check
376 Target_Typ : Entity_Id;
377 Source_Typ : Entity_Id := Empty;
378 Fixed_Int : Boolean := False);
379 -- For scalar types, determines whether an expression node should be
380 -- flagged as needing a runtime range check. If the node requires such a
381 -- check, the Do_Range_Check flag is turned on. The Fixed_Int flag if set
382 -- causes any fixed-point values to be treated as though they were discrete
383 -- values (i.e. the underlying integer value is used).
385 type Check_Result is private;
386 -- Type used to return result of Range_Check call, for later use in
387 -- call to Insert_Range_Checks procedure.
389 procedure Append_Range_Checks
390 (Checks : Check_Result;
392 Suppress_Typ : Entity_Id;
393 Static_Sloc : Source_Ptr;
394 Flag_Node : Node_Id);
395 -- Called to append range checks as returned by a call to Range_Check.
396 -- Stmts is a list to which either the dynamic check is appended or the
397 -- raise Constraint_Error statement is appended (for static checks).
398 -- Static_Sloc is the Sloc at which the raise CE node points, Flag_Node is
399 -- used as the node at which to set the Has_Dynamic_Check flag. Checks_On
400 -- is a boolean value that says if range and index checking is on or not.
402 procedure Insert_Range_Checks
403 (Checks : Check_Result;
405 Suppress_Typ : Entity_Id;
406 Static_Sloc : Source_Ptr := No_Location;
407 Flag_Node : Node_Id := Empty;
408 Do_Before : Boolean := False);
409 -- Called to insert range checks as returned by a call to Range_Check.
410 -- Node is the node after which either the dynamic check is inserted or
411 -- the raise Constraint_Error statement is inserted (for static checks).
412 -- Suppress_Typ is the type to check to determine if checks are suppressed.
413 -- Static_Sloc, if passed, is the Sloc at which the raise CE node points,
414 -- otherwise Sloc (Node) is used. The Has_Dynamic_Check flag is normally
415 -- set at Node. If Flag_Node is present, then this is used instead as the
416 -- node at which to set the Has_Dynamic_Check flag. Normally the check is
417 -- inserted after, if Do_Before is True, the check is inserted before
422 Target_Typ : Entity_Id;
423 Source_Typ : Entity_Id := Empty;
424 Warn_Node : Node_Id := Empty)
426 -- Like Apply_Range_Check, except it does not modify anything. Instead
427 -- it returns an encapsulated result of the check operations for later
428 -- use in a call to Insert_Range_Checks. If Warn_Node is non-empty, its
429 -- Sloc is used, in the static case, for the generated warning or error.
430 -- Additionally, it is used rather than Expr (or Low/High_Bound of Expr)
431 -- in constructing the check.
433 -----------------------
434 -- Expander Routines --
435 -----------------------
437 -- Some of the earlier processing for checks results in temporarily setting
438 -- the Do_Range_Check flag rather than actually generating checks. Now we
439 -- are moving the generation of such checks into the front end for reasons
440 -- of efficiency and simplicity (there were difficutlies in handling this
441 -- in the back end when side effects were present in the expressions being
444 -- Probably we could eliminate the Do_Range_Check flag entirely and
445 -- generate the checks earlier, but this is a delicate area and it
446 -- seemed safer to implement the following routines, which are called
447 -- late on in the expansion process. They check the Do_Range_Check flag
448 -- and if it is set, generate the actual checks and reset the flag.
450 procedure Generate_Range_Check
452 Target_Type : Entity_Id;
453 Reason : RT_Exception_Code);
454 -- This procedure is called to actually generate and insert a range check.
455 -- A check is generated to ensure that the value of N lies within the range
456 -- of the target type. Note that the base type of N may be different from
457 -- the base type of the target type. This happens in the conversion case.
458 -- The Reason parameter is the exception code to be used for the exception
461 -- Note on the relation of this routine to the Do_Range_Check flag. Mostly
462 -- for historical reasons, we often set the Do_Range_Check flag and then
463 -- later we call Generate_Range_Check if this flag is set. Most probably we
464 -- could eliminate this intermediate setting of the flag (historically the
465 -- back end dealt with range checks, using this flag to indicate if a check
466 -- was required, then we moved checks into the front end).
468 procedure Generate_Index_Checks (N : Node_Id);
469 -- This procedure is called to generate index checks on the subscripts for
470 -- the indexed component node N. Each subscript expression is examined, and
471 -- if the Do_Range_Check flag is set, an appropriate index check is
472 -- generated and the flag is reset.
474 -- Similarly, we set the flag Do_Discriminant_Check in the semantic
475 -- analysis to indicate that a discriminant check is required for selected
476 -- component of a discriminated type. The following routine is called from
477 -- the expander to actually generate the call.
479 procedure Generate_Discriminant_Check (N : Node_Id);
480 -- N is a selected component for which a discriminant check is required to
481 -- make sure that the discriminants have appropriate values for the
482 -- selection. This is done by calling the appropriate discriminant checking
483 -- routine for the selector.
485 -----------------------
486 -- Validity Checking --
487 -----------------------
489 -- In (RM 13.9.1(9-11)) we have the following rules on invalid values
491 -- If the representation of a scalar object does not represent value of
492 -- the object's subtype (perhaps because the object was not initialized),
493 -- the object is said to have an invalid representation. It is a bounded
494 -- error to evaluate the value of such an object. If the error is
495 -- detected, either Constraint_Error or Program_Error is raised.
496 -- Otherwise, execution continues using the invalid representation. The
497 -- rules of the language outside this subclause assume that all objects
498 -- have valid representations. The semantics of operations on invalid
499 -- representations are as follows:
501 -- 10 If the representation of the object represents a value of the
502 -- object's type, the value of the type is used.
504 -- 11 If the representation of the object does not represent a value
505 -- of the object's type, the semantics of operations on such
506 -- representations is implementation-defined, but does not by
507 -- itself lead to erroneous or unpredictable execution, or to
508 -- other objects becoming abnormal.
510 -- We quote the rules in full here since they are quite delicate. Most
511 -- of the time, we can just compute away with wrong values, and get a
512 -- possibly wrong result, which is well within the range of allowed
513 -- implementation defined behavior. The two tricky cases are subscripted
514 -- array assignments, where we don't want to do wild stores, and case
515 -- statements where we don't want to do wild jumps.
517 -- In GNAT, we control validity checking with a switch -gnatV that can take
518 -- three parameters, n/d/f for None/Default/Full. These modes have the
519 -- following meanings:
521 -- None (no validity checking)
523 -- In this mode, there is no specific checking for invalid values
524 -- and the code generator assumes that all stored values are always
525 -- within the bounds of the object subtype. The consequences are as
528 -- For case statements, an out of range invalid value will cause
529 -- Constraint_Error to be raised, or an arbitrary one of the case
530 -- alternatives will be executed. Wild jumps cannot result even
531 -- in this mode, since we always do a range check
533 -- For subscripted array assignments, wild stores will result in
534 -- the expected manner when addresses are calculated using values
535 -- of subscripts that are out of range.
537 -- It could perhaps be argued that this mode is still conformant with
538 -- the letter of the RM, since implementation defined is a rather
539 -- broad category, but certainly it is not in the spirit of the
540 -- RM requirement, since wild stores certainly seem to be a case of
541 -- erroneous behavior.
543 -- Default (default standard RM-compatible validity checking)
545 -- In this mode, which is the default, minimal validity checking is
546 -- performed to ensure no erroneous behavior as follows:
548 -- For case statements, an out of range invalid value will cause
549 -- Constraint_Error to be raised.
551 -- For subscripted array assignments, invalid out of range
552 -- subscript values will cause Constraint_Error to be raised.
554 -- Full (Full validity checking)
556 -- In this mode, the protections guaranteed by the standard mode are
557 -- in place, and the following additional checks are made:
559 -- For every assignment, the right side is checked for validity
561 -- For every call, IN and IN OUT parameters are checked for validity
563 -- For every subscripted array reference, both for stores and loads,
564 -- all subscripts are checked for validity.
566 -- These checks are not required by the RM, but will in practice
567 -- improve the detection of uninitialized variables, particularly
568 -- if used in conjunction with pragma Normalize_Scalars.
570 -- In the above description, we talk about performing validity checks,
571 -- but we don't actually generate a check in a case where the compiler
572 -- can be sure that the value is valid. Note that this assurance must
573 -- be achieved without assuming that any uninitialized value lies within
574 -- the range of its type. The following are cases in which values are
575 -- known to be valid. The flag Is_Known_Valid is used to keep track of
576 -- some of these cases.
578 -- If all possible stored values are valid, then any uninitialized
579 -- value must be valid.
581 -- Literals, including enumeration literals, are clearly always valid
583 -- Constants are always assumed valid, with a validity check being
584 -- performed on the initializing value where necessary to ensure that
587 -- For variables, the status is set to known valid if there is an
588 -- initializing expression. Again a check is made on the initializing
589 -- value if necessary to ensure that this assumption is valid. The
590 -- status can change as a result of local assignments to a variable.
591 -- If a known valid value is unconditionally assigned, then we mark
592 -- the left side as known valid. If a value is assigned that is not
593 -- known to be valid, then we mark the left side as invalid. This
594 -- kind of processing does NOT apply to non-local variables since we
595 -- are not following the flow graph (more properly the flow of actual
596 -- processing only corresponds to the flow graph for local assignments).
597 -- For non-local variables, we preserve the current setting, i.e. a
598 -- validity check is performed when assigning to a knonwn valid global.
600 -- Note: no validity checking is required if range checks are suppressed
601 -- regardless of the setting of the validity checking mode.
603 -- The following procedures are used in handling validity checking
605 procedure Apply_Subscript_Validity_Checks (Expr : Node_Id);
606 -- Expr is the node for an indexed component. If validity checking and
607 -- range checking are enabled, all subscripts for this indexed component
608 -- are checked for validity.
610 procedure Check_Valid_Lvalue_Subscripts (Expr : Node_Id);
611 -- Expr is a lvalue, i.e. an expression representing the target of an
612 -- assignment. This procedure checks for this expression involving an
613 -- assignment to an array value. We have to be sure that all the subscripts
614 -- in such a case are valid, since according to the rules in (RM
615 -- 13.9.1(9-11)) such assignments are not permitted to result in erroneous
616 -- behavior in the case of invalid subscript values.
618 procedure Ensure_Valid (Expr : Node_Id; Holes_OK : Boolean := False);
619 -- Ensure that Expr represents a valid value of its type. If this type
620 -- is not a scalar type, then the call has no effect, since validity
621 -- is only an issue for scalar types. The effect of this call is to
622 -- check if the value is known valid, if so, nothing needs to be done.
623 -- If this is not known, then either Expr is set to be range checked,
624 -- or specific checking code is inserted so that an exception is raised
625 -- if the value is not valid.
627 -- The optional argument Holes_OK indicates whether it is necessary to
628 -- worry about enumeration types with non-standard representations leading
629 -- to "holes" in the range of possible representations. If Holes_OK is
630 -- True, then such values are assumed valid (this is used when the caller
631 -- will make a separate check for this case anyway). If Holes_OK is False,
632 -- then this case is checked, and code is inserted to ensure that Expr is
633 -- valid, raising Constraint_Error if the value is not valid.
635 function Expr_Known_Valid (Expr : Node_Id) return Boolean;
636 -- This function tests it the value of Expr is known to be valid in the
637 -- sense of RM 13.9.1(9-11). In the case of GNAT, it is only discrete types
638 -- which are a concern, since for non-discrete types we simply continue
639 -- computation with invalid values, which does not lead to erroneous
640 -- behavior. Thus Expr_Known_Valid always returns True if the type of Expr
641 -- is non-discrete. For discrete types the value returned is True only if
642 -- it can be determined that the value is Valid. Otherwise False is
645 procedure Insert_Valid_Check (Expr : Node_Id);
646 -- Inserts code that will check for the value of Expr being valid, in
647 -- the sense of the 'Valid attribute returning True. Constraint_Error
648 -- will be raised if the value is not valid.
650 procedure Null_Exclusion_Static_Checks (N : Node_Id);
651 -- Ada 2005 (AI-231): Check bad usages of the null-exclusion issue
653 procedure Remove_Checks (Expr : Node_Id);
654 -- Remove all checks from Expr except those that are only executed
655 -- conditionally (on the right side of And Then/Or Else. This call
656 -- removes only embedded checks (Do_Range_Check, Do_Overflow_Check).
658 procedure Validity_Check_Range (N : Node_Id);
659 -- If N is an N_Range node, then Ensure_Valid is called on its bounds,
660 -- if validity checking of operands is enabled.
664 type Check_Result is array (Positive range 1 .. 2) of Node_Id;
665 -- There are two cases for the result returned by Range_Check:
667 -- For the static case the result is one or two nodes that should cause
668 -- a Constraint_Error. Typically these will include Expr itself or the
669 -- direct descendents of Expr, such as Low/High_Bound (Expr)). It is the
670 -- responsibility of the caller to rewrite and substitute the nodes with
671 -- N_Raise_Constraint_Error nodes.
673 -- For the non-static case a single N_Raise_Constraint_Error node with a
674 -- non-empty Condition field is returned.
676 -- Unused entries in Check_Result, if any, are simply set to Empty For
677 -- external clients, the required processing on this result is achieved
678 -- using the Insert_Range_Checks routine.
680 pragma Inline (Apply_Length_Check);
681 pragma Inline (Apply_Range_Check);
682 pragma Inline (Apply_Static_Length_Check);
OSDN Git Service
