cparser
15 years agoAnother attempt to solve the halting problem: Detect variables, which are never read...
Christoph Mallon [Mon, 20 Oct 2008 10:46:10 +0000 (10:46 +0000)]
Another attempt to solve the halting problem: Detect variables, which are never read.  The test is cheap and could be improved by using dataflow, SSA and fixpoint magic, but of course it is lightyears ahead of GCC.  For example in x = x + 1; x is not marked as read, because it is only used to assign to itself.

[r23026]

15 years agoMake check_unreachable() more tail recursion friendly.
Christoph Mallon [Mon, 20 Oct 2008 08:40:04 +0000 (08:40 +0000)]
Make check_unreachable() more tail recursion friendly.

[r23021]

15 years agoResolve assertion, when using -fsyntax-only.
Christoph Mallon [Mon, 20 Oct 2008 08:35:58 +0000 (08:35 +0000)]
Resolve assertion, when using -fsyntax-only.

[r23020]

15 years agoRemove stale (since r21011) comment.
Christoph Mallon [Sun, 19 Oct 2008 17:30:24 +0000 (17:30 +0000)]
Remove stale (since r21011) comment.

[r23016]

15 years agoFix r23014.
Christoph Mallon [Sun, 19 Oct 2008 17:25:56 +0000 (17:25 +0000)]
Fix r23014.

[r23015]

15 years agoDo not generate an error message about incompatible types in ?: if the error type...
Christoph Mallon [Sun, 19 Oct 2008 16:49:14 +0000 (16:49 +0000)]
Do not generate an error message about incompatible types in ?: if the error type is involved.

[r23014]

15 years agoResolve some warnings.
Christoph Mallon [Sun, 19 Oct 2008 16:42:19 +0000 (16:42 +0000)]
Resolve some warnings.

[r23013]

15 years agoAdd forgotten conversion specifiers.
Christoph Mallon [Sun, 19 Oct 2008 16:27:40 +0000 (16:27 +0000)]
Add forgotten conversion specifiers.

[r23012]

15 years agoResolve some warnings.
Christoph Mallon [Sun, 19 Oct 2008 16:19:50 +0000 (16:19 +0000)]
Resolve some warnings.

[r23011]

15 years ago-fopt-foo means activate, not deactivate.
Christoph Mallon [Tue, 14 Oct 2008 15:16:52 +0000 (15:16 +0000)]
-fopt-foo means activate, not deactivate.

[r22879]

15 years agoAdd debug info to implicitly created != 0 tests in ifs.
Christoph Mallon [Mon, 13 Oct 2008 10:37:57 +0000 (10:37 +0000)]
Add debug info to implicitly created != 0 tests in ifs.

[r22835]

15 years agono need for bitfiddle here
Matthias Braun [Sat, 11 Oct 2008 18:53:41 +0000 (18:53 +0000)]
no need for bitfiddle here

[r22709]

15 years agoPrint AST and error/warning summary even in the presence of preprocessor errors.
Christoph Mallon [Tue, 7 Oct 2008 07:33:26 +0000 (07:33 +0000)]
Print AST and error/warning summary even in the presence of preprocessor errors.

[r22560]

15 years agoRemove unreachable statement.
Christoph Mallon [Tue, 7 Oct 2008 07:13:48 +0000 (07:13 +0000)]
Remove unreachable statement.

[r22559]

15 years agoShow the number of the conversion specification in the format check warnings.
Christoph Mallon [Tue, 7 Oct 2008 07:12:59 +0000 (07:12 +0000)]
Show the number of the conversion specification in the format check warnings.

[r22558]

15 years agoRemove pointless local variable.
Christoph Mallon [Tue, 7 Oct 2008 07:01:16 +0000 (07:01 +0000)]
Remove pointless local variable.

[r22557]

15 years agoRemove redundant assignment.
Christoph Mallon [Tue, 7 Oct 2008 06:39:22 +0000 (06:39 +0000)]
Remove redundant assignment.

[r22556]

15 years ago- improve error message even more ;-)
Michael Beck [Mon, 6 Oct 2008 22:40:10 +0000 (22:40 +0000)]
- improve error message even more ;-)

[r22555]

15 years ago- improve error message
Michael Beck [Mon, 6 Oct 2008 22:36:55 +0000 (22:36 +0000)]
- improve error message

[r22554]

15 years ago- allow '-' for s and S format
Michael Beck [Mon, 6 Oct 2008 21:50:23 +0000 (21:50 +0000)]
- allow '-' for s and S format
- improve error message

[r22553]

15 years agoonly query primitive types for their base type
Matthias Braun [Mon, 6 Oct 2008 18:54:52 +0000 (18:54 +0000)]
only query primitive types for their base type

[r22549]

15 years agofix missing skip_typeref buf
Matthias Braun [Mon, 6 Oct 2008 13:27:51 +0000 (13:27 +0000)]
fix missing skip_typeref buf

[r22533]

15 years agodisable fix for parsetest/fehler053.c for now (as it breaks more common use cases)
Matthias Braun [Mon, 6 Oct 2008 13:20:23 +0000 (13:20 +0000)]
disable fix for parsetest/fehler053.c for now (as it breaks more common use cases)

[r22530]

15 years agofix bitfield2.c
Matthias Braun [Fri, 3 Oct 2008 17:16:32 +0000 (17:16 +0000)]
fix bitfield2.c

[r22436]

15 years agotricky bitfield test from gcc (extended)
Matthias Braun [Fri, 3 Oct 2008 17:16:10 +0000 (17:16 +0000)]
tricky bitfield test from gcc (extended)

[r22435]

15 years agoImplement reference_addr() for DECLARATION_KIND_FUNCTION.
Christoph Mallon [Fri, 3 Oct 2008 16:36:40 +0000 (16:36 +0000)]
Implement reference_addr() for DECLARATION_KIND_FUNCTION.

[r22434]

15 years agoskip_typeref().
Christoph Mallon [Thu, 2 Oct 2008 21:51:20 +0000 (21:51 +0000)]
skip_typeref().

[r22424]

15 years agoFix assertion failure introduced in r22072.
Christoph Mallon [Thu, 2 Oct 2008 19:49:19 +0000 (19:49 +0000)]
Fix assertion failure introduced in r22072.

[r22422]

15 years agoskip_typeref().
Christoph Mallon [Thu, 2 Oct 2008 18:56:14 +0000 (18:56 +0000)]
skip_typeref().

[r22421]

15 years agoTreat expressions with error type as lvalues. There was a parse error before, which...
Christoph Mallon [Thu, 2 Oct 2008 18:39:58 +0000 (18:39 +0000)]
Treat expressions with error type as lvalues.  There was a parse error before, which maybe prevented properly recognizing it as lvalue.

[r22419]

15 years agoImprove error recovery for if-statements.
Christoph Mallon [Thu, 2 Oct 2008 18:29:30 +0000 (18:29 +0000)]
Improve error recovery for if-statements.

[r22418]

15 years agoParse __builtin_inf{,f,l} and correct __builtin_nand to __builtin_nanl.
Christoph Mallon [Thu, 2 Oct 2008 07:57:02 +0000 (07:57 +0000)]
Parse __builtin_inf{,f,l} and correct __builtin_nand to __builtin_nanl.

[r22406]

15 years ago- set the new combo flag
Michael Beck [Mon, 29 Sep 2008 23:39:17 +0000 (23:39 +0000)]
- set the new combo flag

[r22354]

15 years agouse firm array to avoid name/symbol clashes
Matthias Braun [Sun, 28 Sep 2008 18:31:20 +0000 (18:31 +0000)]
use firm array to avoid name/symbol clashes

[r22332]

16 years agoProcess declaration modifiers for variables in ast2firm, too.
Christoph Mallon [Fri, 26 Sep 2008 10:46:50 +0000 (10:46 +0000)]
Process declaration modifiers for variables in ast2firm, too.

[r22284]

16 years ago- add -finline-max-size= and -finline-threshold= parameter
Michael Beck [Fri, 26 Sep 2008 10:07:33 +0000 (10:07 +0000)]
- add -finline-max-size= and -finline-threshold= parameter

[r22279]

16 years agoincrease inline limit
Matthias Braun [Wed, 24 Sep 2008 14:50:51 +0000 (14:50 +0000)]
increase inline limit

[r22244]

16 years agoGenerate AST for some semantically invalid initializers.
Christoph Mallon [Tue, 23 Sep 2008 20:50:14 +0000 (20:50 +0000)]
Generate AST for some semantically invalid initializers.

[r22222]

16 years ago- add missing initialization of have_const_functions
Michael Beck [Tue, 23 Sep 2008 17:13:40 +0000 (17:13 +0000)]
- add missing initialization of have_const_functions

[r22214]

16 years agohandle DM_CONST, lower inline limit, force optimize_funccalls
Matthias Braun [Tue, 23 Sep 2008 14:57:08 +0000 (14:57 +0000)]
handle DM_CONST, lower inline limit, force optimize_funccalls

[r22210]

16 years agoTreat size_constant like a bool should be, i.e. do not explicitly compare with =...
Christoph Mallon [Tue, 23 Sep 2008 14:45:28 +0000 (14:45 +0000)]
Treat size_constant like a bool should be, i.e. do not explicitly compare with == 1/true.

[r22206]

16 years agodisable builtins by default
Matthias Braun [Tue, 23 Sep 2008 09:12:41 +0000 (09:12 +0000)]
disable builtins by default

[r22187]

16 years agopossibility to disable builtin functions
Matthias Braun [Mon, 22 Sep 2008 18:42:50 +0000 (18:42 +0000)]
possibility to disable builtin functions

[r22178]

16 years agomake -fsync default
Matthias Braun [Sun, 21 Sep 2008 22:48:19 +0000 (22:48 +0000)]
make -fsync default

[r22157]

16 years agoright_to_left() now returns 0 for left-to-right associativity instead of -1.
Christoph Mallon [Sat, 20 Sep 2008 12:17:52 +0000 (12:17 +0000)]
right_to_left() now returns 0 for left-to-right associativity instead of -1.

[r22138]

16 years agofix gcse
Matthias Braun [Fri, 19 Sep 2008 11:27:03 +0000 (11:27 +0000)]
fix gcse

[r22122]

16 years agoResolve off-by-one (and off-by-two) errors in the source position of statements.
Christoph Mallon [Fri, 19 Sep 2008 04:56:20 +0000 (04:56 +0000)]
Resolve off-by-one (and off-by-two) errors in the source position of statements.

[r22112]

16 years agoSet the source position in allocate_statement_zero().
Christoph Mallon [Fri, 19 Sep 2008 04:48:26 +0000 (04:48 +0000)]
Set the source position in allocate_statement_zero().

[r22111]

16 years agoMake stray ; outside of functions an error in strict mode.
Christoph Mallon [Thu, 18 Sep 2008 22:30:18 +0000 (22:30 +0000)]
Make stray ; outside of functions an error in strict mode.

[r22102]

16 years agoResolve compile error.
Christoph Mallon [Thu, 18 Sep 2008 22:29:34 +0000 (22:29 +0000)]
Resolve compile error.

[r22101]

16 years ago- BugFix: unitializer must be created using qualifiers from array-skipped typed
Michael Beck [Thu, 18 Sep 2008 19:17:54 +0000 (19:17 +0000)]
- BugFix: unitializer must be created using qualifiers from array-skipped typed

[r22092]

16 years ago- add get_type_qualifiers() allowing to get the type qualifiers WITHOUT
Michael Beck [Thu, 18 Sep 2008 19:17:11 +0000 (19:17 +0000)]
- add get_type_qualifiers() allowing to get the type qualifiers WITHOUT
  using skip_typeref() AND allowing to skip ARRAY types

[r22091]

16 years agoThe standard calls it "file scope", so rename global_scope to file_scope.
Christoph Mallon [Thu, 18 Sep 2008 13:03:31 +0000 (13:03 +0000)]
The standard calls it "file scope", so rename global_scope to file_scope.

[r22079]

16 years agoResolve unreachable statement warnings.
Christoph Mallon [Thu, 18 Sep 2008 11:52:46 +0000 (11:52 +0000)]
Resolve unreachable statement warnings.

[r22077]

16 years agoPut T_EOF into the anchor set.
Christoph Mallon [Thu, 18 Sep 2008 11:43:02 +0000 (11:43 +0000)]
Put T_EOF into the anchor set.

[r22075]

16 years agodepend on libfirm (if path is specified in config.mak)
Matthias Braun [Thu, 18 Sep 2008 11:15:36 +0000 (11:15 +0000)]
depend on libfirm (if path is specified in config.mak)

[r22074]

16 years agoSlightly simplify stack_pop_to().
Christoph Mallon [Thu, 18 Sep 2008 09:41:03 +0000 (09:41 +0000)]
Slightly simplify stack_pop_to().

[r22072]

16 years agoenvironment_top() and local_label_top() return size_t, not int.
Christoph Mallon [Thu, 18 Sep 2008 09:02:58 +0000 (09:02 +0000)]
environment_top() and local_label_top() return size_t, not int.

[r22070]

16 years agoSlightly simplify stack_push().
Christoph Mallon [Thu, 18 Sep 2008 08:53:45 +0000 (08:53 +0000)]
Slightly simplify stack_push().

[r22069]

16 years agofixed parameter redeclaration error
Michael Beck [Thu, 18 Sep 2008 00:32:53 +0000 (00:32 +0000)]
fixed parameter redeclaration error

[r22068]

16 years agoenhanced
Michael Beck [Thu, 18 Sep 2008 00:32:13 +0000 (00:32 +0000)]
enhanced

[r22067]

16 years ago- functions cannot be lvalues
Michael Beck [Thu, 18 Sep 2008 00:22:55 +0000 (00:22 +0000)]
- functions cannot be lvalues

[r22066]

16 years ago- add more attributes to support inner functions
Michael Beck [Thu, 18 Sep 2008 00:15:34 +0000 (00:15 +0000)]
- add more attributes to support inner functions
- preliminary support for code generation for inner functions (closure must NOT be used yet)

[r22065]

16 years ago- added outer function jump to example
Michael Beck [Thu, 18 Sep 2008 00:13:05 +0000 (00:13 +0000)]
- added outer function jump to example

[r22064]

16 years agoshould warn about struct inside parameter scope definition
Michael Beck [Wed, 17 Sep 2008 23:20:48 +0000 (23:20 +0000)]
should warn about struct inside parameter scope definition

[r22063]

16 years ago- replaced set_scope by scope_push()/scope_pop()
Michael Beck [Wed, 17 Sep 2008 21:52:36 +0000 (21:52 +0000)]
- replaced set_scope by scope_push()/scope_pop()
- detect access to variables of outer functions and set the address_taken flag
- detect redeclaration of parameters and issue an error
- set last_declaration in the parameter scope
- mark the scope of parameters

[r22060]

16 years agoAdd { as anchor while parsing k&r declaration list.
Christoph Mallon [Wed, 17 Sep 2008 21:49:45 +0000 (21:49 +0000)]
Add { as anchor while parsing k&r declaration list.

[r22059]

16 years agoImplement -Waddress.
Christoph Mallon [Wed, 17 Sep 2008 21:40:23 +0000 (21:40 +0000)]
Implement -Waddress.

[r22057]

16 years agoRestore value of in_gcc_extension using old_gcc_extension.
Christoph Mallon [Wed, 17 Sep 2008 20:04:31 +0000 (20:04 +0000)]
Restore value of in_gcc_extension using old_gcc_extension.

[r22054]

16 years agoAdd { as anchor while parsing external declaration.
Christoph Mallon [Wed, 17 Sep 2008 19:58:13 +0000 (19:58 +0000)]
Add { as anchor while parsing external declaration.

[r22053]

16 years agofixed comments
Michael Beck [Wed, 17 Sep 2008 19:28:50 +0000 (19:28 +0000)]
fixed comments

[r22052]

16 years ago- BugFix: typeof operator must set the in_type_prop attribute
Michael Beck [Wed, 17 Sep 2008 19:21:22 +0000 (19:21 +0000)]
- BugFix: typeof operator must set the in_type_prop attribute
- __extension__ now sets/resets the in_gcc_extension attribute
- use new GNU_MODE macro instead of (c_mode & _GNUC) which checks for in_gcc_extension

[r22051]

16 years agoReduce anchor scope.
Christoph Mallon [Wed, 17 Sep 2008 17:14:29 +0000 (17:14 +0000)]
Reduce anchor scope.

[r22050]

16 years agoadd support for gcc nested functions
Michael Beck [Wed, 17 Sep 2008 15:09:24 +0000 (15:09 +0000)]
add support for gcc nested functions

[r22043]

16 years agoadd missing print_indent() (now correctly prints nsted functions)
Michael Beck [Wed, 17 Sep 2008 15:08:48 +0000 (15:08 +0000)]
add missing print_indent() (now correctly prints nsted functions)

[r22042]

16 years agoeat_statement() is dead, long live eat_until_anchor()!
Christoph Mallon [Wed, 17 Sep 2008 14:56:33 +0000 (14:56 +0000)]
eat_statement() is dead, long live eat_until_anchor()!

[r22041]

16 years agonested function example
Michael Beck [Wed, 17 Sep 2008 14:52:31 +0000 (14:52 +0000)]
nested function example

[r22040]

16 years agoSmall simplification.
Christoph Mallon [Wed, 17 Sep 2008 14:47:40 +0000 (14:47 +0000)]
Small simplification.

[r22039]

16 years agodo not record empty declarations by default
Michael Beck [Wed, 17 Sep 2008 12:51:33 +0000 (12:51 +0000)]
do not record empty declarations by default

[r22036]

16 years ago- type property expressions are NOT self-initializer
Michael Beck [Wed, 17 Sep 2008 12:40:52 +0000 (12:40 +0000)]
- type property expressions are NOT self-initializer

[r22035]

16 years agoAlways return a non-null pointer from parse_array_declarator() to avoid dereferencing...
Christoph Mallon [Wed, 17 Sep 2008 08:55:51 +0000 (08:55 +0000)]
Always return a non-null pointer from parse_array_declarator() to avoid dereferencing a null pointer later on.

[r22029]

16 years agoThe element type of an array shall not be an incomplete or function type.
Christoph Mallon [Wed, 17 Sep 2008 07:50:51 +0000 (07:50 +0000)]
The element type of an array shall not be an incomplete or function type.

[r22028]

16 years agoStill record the element type of arrays and the return type of functions, even if...
Christoph Mallon [Wed, 17 Sep 2008 07:29:23 +0000 (07:29 +0000)]
Still record the element type of arrays and the return type of functions, even if they are invalid.

[r22027]

16 years agoWarn about type qualifiers in the return type of function types.
Christoph Mallon [Tue, 16 Sep 2008 18:38:09 +0000 (18:38 +0000)]
Warn about type qualifiers in the return type of function types.

[r22020]

16 years agoDo not warn about the return type of main(), if it is the error type.
Christoph Mallon [Tue, 16 Sep 2008 18:11:08 +0000 (18:11 +0000)]
Do not warn about the return type of main(), if it is the error type.

[r22017]

16 years agoUse the real error type instead of ATOMIC_TYPE_INVALID, when parsing type specifiers...
Christoph Mallon [Tue, 16 Sep 2008 18:06:30 +0000 (18:06 +0000)]
Use the real error type instead of ATOMIC_TYPE_INVALID, when parsing type specifiers, to avoid strange errors/warnings later on.

[r22016]

16 years agoAvoid segfault when encountering a parse error while parsing __declspec in type speci...
Christoph Mallon [Tue, 16 Sep 2008 18:02:07 +0000 (18:02 +0000)]
Avoid segfault when encountering a parse error while parsing __declspec in type specifiers.

[r22015]

16 years agoDo not warn about excess initialisers, when trying to initialise an incomplete compou...
Christoph Mallon [Tue, 16 Sep 2008 17:54:43 +0000 (17:54 +0000)]
Do not warn about excess initialisers, when trying to initialise an incomplete compound variable.

[r22014]

16 years agoSlightly simplify printing binary operators.
Christoph Mallon [Tue, 16 Sep 2008 17:53:24 +0000 (17:53 +0000)]
Slightly simplify printing binary operators.

[r22013]

16 years agoUse fputc() instead of fputs() to print a single char.
Christoph Mallon [Tue, 16 Sep 2008 12:38:38 +0000 (12:38 +0000)]
Use fputc() instead of fputs() to print a single char.

[r22006]

16 years agoPrint no " " between the ")" and the "{" of a compound literal.
Christoph Mallon [Tue, 16 Sep 2008 12:37:10 +0000 (12:37 +0000)]
Print no " " between the ")" and the "{" of a compound literal.

[r22005]

16 years agoabort() instead of exit(1) in case of anchor token leaks.
Christoph Mallon [Tue, 16 Sep 2008 10:48:48 +0000 (10:48 +0000)]
abort() instead of exit(1) in case of anchor token leaks.

[r22001]

16 years agoWhen printing AST for asm statements first print outputs, then inputs.
Christoph Mallon [Tue, 16 Sep 2008 10:09:34 +0000 (10:09 +0000)]
When printing AST for asm statements first print outputs, then inputs.

[r21998]

16 years agoImprove parse_va_start():
Christoph Mallon [Tue, 16 Sep 2008 09:39:36 +0000 (09:39 +0000)]
Improve parse_va_start():
- reference expressions never have no declaration, at least there is an error declaration
- do not return an invalid expression, when the second argument is a reference, but not the last parameter
- if the second parameter is invalid, still expect the ).

[r21997]

16 years agoMake parsing of statements more robust in case of parser errors by avoiding endless...
Christoph Mallon [Tue, 16 Sep 2008 09:06:10 +0000 (09:06 +0000)]
Make parsing of statements more robust in case of parser errors by avoiding endless loops, because no token gets consumed.

[r21996]

16 years agoPlug anchor token leak in parse_statement_expression().
Christoph Mallon [Tue, 16 Sep 2008 07:50:06 +0000 (07:50 +0000)]
Plug anchor token leak in parse_statement_expression().

[r21995]

16 years agoPlug two more anchor token underflows.
Christoph Mallon [Tue, 16 Sep 2008 07:19:52 +0000 (07:19 +0000)]
Plug two more anchor token underflows.

[r21994]

16 years agoFix the first anchor token leak (actually an underflow).
Christoph Mallon [Tue, 16 Sep 2008 07:16:38 +0000 (07:16 +0000)]
Fix the first anchor token leak (actually an underflow).

[r21993]

16 years agoAdd code to detect anchor token leaks.
Christoph Mallon [Tue, 16 Sep 2008 07:15:53 +0000 (07:15 +0000)]
Add code to detect anchor token leaks.

[r21992]

16 years agoCreate a few less invalid expressions on parse errors.
Christoph Mallon [Tue, 16 Sep 2008 06:38:51 +0000 (06:38 +0000)]
Create a few less invalid expressions on parse errors.

[r21991]