Matthias Braun [Mon, 2 Jun 2008 20:25:34 +0000 (20:25 +0000)]
fix kr2.c
[r19947]
Matthias Braun [Mon, 2 Jun 2008 15:17:03 +0000 (15:17 +0000)]
precedence of shift was wrong
[r19944]
Christoph Mallon [Mon, 2 Jun 2008 15:07:02 +0000 (15:07 +0000)]
Fix incorrect format check warnings.
[r19943]
Matthias Braun [Mon, 2 Jun 2008 15:00:19 +0000 (15:00 +0000)]
builtin_nanf is const
[r19941]
Matthias Braun [Mon, 2 Jun 2008 14:44:04 +0000 (14:44 +0000)]
use size_t or ssize_t for array offsets
[r19940]
Matthias Braun [Mon, 2 Jun 2008 14:09:50 +0000 (14:09 +0000)]
more tricky kr things
[r19934]
Matthias Braun [Mon, 2 Jun 2008 13:53:20 +0000 (13:53 +0000)]
fix shouldfail part of testscript
[r19933]
Matthias Braun [Mon, 2 Jun 2008 13:45:39 +0000 (13:45 +0000)]
fix handling of kr style functions
[r19932]
Michael Beck [Mon, 2 Jun 2008 13:42:26 +0000 (13:42 +0000)]
added option -momit-leaf-frame-pointer
[r19931]
Matthias Braun [Mon, 2 Jun 2008 11:41:53 +0000 (11:41 +0000)]
fix parsing of some struct initializers
[r19924]
Michael Beck [Mon, 2 Jun 2008 11:32:06 +0000 (11:32 +0000)]
add option -Wreturn-type
[r19923]
Matthias Braun [Mon, 2 Jun 2008 11:14:26 +0000 (11:14 +0000)]
only build switch fallthrough block when needed
[r19922]
Michael Beck [Mon, 2 Jun 2008 11:13:50 +0000 (11:13 +0000)]
warn if missing return
[r19921]
Michael Beck [Mon, 2 Jun 2008 11:09:58 +0000 (11:09 +0000)]
add test for missing returns
[r19920]
Michael Beck [Mon, 2 Jun 2008 11:08:37 +0000 (11:08 +0000)]
checks for missing return are done in ast2firm yet
[r19919]
Matthias Braun [Mon, 2 Jun 2008 10:57:03 +0000 (10:57 +0000)]
only construct fallthrough block if needed for an if
[r19918]
Michael Beck [Mon, 2 Jun 2008 10:32:32 +0000 (10:32 +0000)]
add checks for missing return statement
[r19916]
Matthias Braun [Mon, 2 Jun 2008 10:31:42 +0000 (10:31 +0000)]
- make sure no expression is built twice
- rework creation of arithmetic assign statements
- rework handling of bitfields
[r19915]
Christoph Mallon [Mon, 2 Jun 2008 08:52:21 +0000 (08:52 +0000)]
- Add missing skip_typeref()
- Check for lvalue/const/incomplete type for assignments
[r19913]
Matthias Braun [Sun, 1 Jun 2008 22:44:06 +0000 (22:44 +0000)]
more fixes for bitfield type
[r19912]
Matthias Braun [Sun, 1 Jun 2008 22:28:08 +0000 (22:28 +0000)]
bitfield is a complete type
[r19911]
Michael Beck [Sun, 1 Jun 2008 19:55:45 +0000 (19:55 +0000)]
- add option -version (in contrast to --version)
- add option -m32
- add option -mprefered-stack-boundary
- renamed -mfpu to right -mfpmath option
[r19909]
Michael Beck [Sun, 1 Jun 2008 19:01:41 +0000 (19:01 +0000)]
- fixed typos
- fixed a warning
- added some doxygen comments
[r19908]
Matthias Braun [Sun, 1 Jun 2008 14:28:35 +0000 (14:28 +0000)]
testcase for sideeffects problems
[r19905]
Christoph Mallon [Sun, 1 Jun 2008 14:28:04 +0000 (14:28 +0000)]
Record the source position of alignof and sizeof. Disallow incomplete, function designator and bitfield types as operands.
[r19904]
Michael Beck [Sun, 1 Jun 2008 14:26:39 +0000 (14:26 +0000)]
BugFix: add erronously deleted break
[r19903]
Christoph Mallon [Sun, 1 Jun 2008 13:47:30 +0000 (13:47 +0000)]
Fix typo in parse_alignof(): s/EXPR_SIZEOF/EXPR_ALIGNOF/.
[r19901]
Christoph Mallon [Sun, 1 Jun 2008 13:44:44 +0000 (13:44 +0000)]
sizeof(function_designator) is not allowed.
[r19900]
Michael Beck [Sun, 1 Jun 2008 13:00:30 +0000 (13:00 +0000)]
- used MINGW instead of Win32 again
- alloca must be catched in ALL modes
- alloca must be immediate converted into a Alloc node
[r19899]
Michael Beck [Sun, 1 Jun 2008 12:57:42 +0000 (12:57 +0000)]
synchronize with eccp
[r19898]
Michael Beck [Sun, 1 Jun 2008 11:33:59 +0000 (11:33 +0000)]
return the status code of the preprocessor on -E
[r19896]
Michael Beck [Sun, 1 Jun 2008 11:32:25 +0000 (11:32 +0000)]
add option -E
[r19895]
Matthias Braun [Sat, 31 May 2008 12:44:21 +0000 (12:44 +0000)]
fix for handling of for scope handling
[r19890]
Matthias Braun [Sat, 31 May 2008 12:28:18 +0000 (12:28 +0000)]
rework local variable handling:
- they're now created when we are at the beginning of the compound statement
(maybe we should move that scope thing up to the function level...)
- they're still initialized when we are at the declaration_statement
- rework handling of implicitely defined functions. They're now created in the local
scope
[r19889]
Matthias Braun [Sat, 31 May 2008 12:26:16 +0000 (12:26 +0000)]
improve/cleanup some tests
[r19888]
Matthias Braun [Sat, 31 May 2008 10:40:54 +0000 (10:40 +0000)]
remove debug printfs
[r19886]
Matthias Braun [Sat, 31 May 2008 10:39:13 +0000 (10:39 +0000)]
fix rts_mapping code, add alloca mapping
[r19883]
Matthias Braun [Sat, 31 May 2008 09:09:20 +0000 (09:09 +0000)]
set alignment for long double
[r19880]
Matthias Braun [Sat, 31 May 2008 09:02:25 +0000 (09:02 +0000)]
fix cp_error032
[r19879]
Matthias Braun [Sat, 31 May 2008 09:02:11 +0000 (09:02 +0000)]
add bunch of testapps
[r19878]
Matthias Braun [Sat, 31 May 2008 07:55:33 +0000 (07:55 +0000)]
shouldfail tests for prototype checking
[r19877]
Matthias Braun [Sat, 31 May 2008 07:53:25 +0000 (07:53 +0000)]
do not print uninitialized stuff
[r19876]
Matthias Braun [Sat, 31 May 2008 07:52:22 +0000 (07:52 +0000)]
add shouldpass directory to testscript
[r19875]
Matthias Braun [Sat, 31 May 2008 07:49:41 +0000 (07:49 +0000)]
K&R style parameter lists are not a prototype...
[r19874]
Matthias Braun [Sat, 31 May 2008 06:23:42 +0000 (06:23 +0000)]
fix for firm changes, bugfix: reset saw_default_label
[r19873]
Matthias Braun [Fri, 30 May 2008 18:11:00 +0000 (18:11 +0000)]
fix extern inline handling, implement __builtin_huge_val
[r19862]
Matthias Braun [Fri, 30 May 2008 18:10:50 +0000 (18:10 +0000)]
hugeval test
[r19861]
Matthias Braun [Fri, 30 May 2008 17:35:30 +0000 (17:35 +0000)]
values can be bad, local static vars must have unique names
[r19860]
Matthias Braun [Fri, 30 May 2008 16:27:03 +0000 (16:27 +0000)]
improve creation of call nodes
[r19858]
Matthias Braun [Fri, 30 May 2008 16:26:34 +0000 (16:26 +0000)]
fix crash
[r19857]
Michael Beck [Fri, 30 May 2008 14:58:05 +0000 (14:58 +0000)]
synchronized with eccp
[r19854]
Matthias Braun [Fri, 30 May 2008 14:46:33 +0000 (14:46 +0000)]
adeclar printf
[r19851]
Matthias Braun [Fri, 30 May 2008 14:30:36 +0000 (14:30 +0000)]
add new shouldpass category
[r19850]
Matthias Braun [Fri, 30 May 2008 14:11:31 +0000 (14:11 +0000)]
we dont have an lvalue test, oops
[r19847]
Matthias Braun [Fri, 30 May 2008 13:52:50 +0000 (13:52 +0000)]
C can do such stupid things...
[r19845]
Matthias Braun [Fri, 30 May 2008 12:01:08 +0000 (12:01 +0000)]
add missing strict conv for variable initialisation
[r19836]
Matthias Braun [Fri, 30 May 2008 12:00:45 +0000 (12:00 +0000)]
fix for last commit
[r19835]
Matthias Braun [Fri, 30 May 2008 11:37:15 +0000 (11:37 +0000)]
union init test (copied from firm)
[r19834]
Matthias Braun [Fri, 30 May 2008 11:17:34 +0000 (11:17 +0000)]
reorganized some test programs
[r19832]
Matthias Braun [Fri, 30 May 2008 10:36:59 +0000 (10:36 +0000)]
better approach towards the parsing problems
[r19831]
Matthias Braun [Fri, 30 May 2008 08:57:45 +0000 (08:57 +0000)]
make parser more robust against invalid declarations
[r19830]
Matthias Braun [Fri, 30 May 2008 08:57:39 +0000 (08:57 +0000)]
make parser more robust against invalid declarations
[r19829]
Michael Beck [Thu, 29 May 2008 09:09:04 +0000 (09:09 +0000)]
- Moved cp_errot030.c to shouldfail/wrongtype.c, as it should fail
- now really crashes cparser
[r19826]
Michael Beck [Wed, 28 May 2008 13:24:15 +0000 (13:24 +0000)]
cparser crashes...
[r19811]
Michael Beck [Wed, 28 May 2008 12:59:20 +0000 (12:59 +0000)]
fixed handling of -b
[r19809]
Michael Beck [Wed, 28 May 2008 12:53:43 +0000 (12:53 +0000)]
fixed warning for 64bit build
[r19808]
Michael Beck [Tue, 27 May 2008 13:41:15 +0000 (13:41 +0000)]
non-existing entities must have peculiarity_description
[r19795]
Matthias Braun [Tue, 27 May 2008 12:50:28 +0000 (12:50 +0000)]
fix bug with big octal sequences
[r19793]
Matthias Braun [Tue, 27 May 2008 12:43:09 +0000 (12:43 +0000)]
one more
[r19792]
Matthias Braun [Tue, 27 May 2008 12:36:21 +0000 (12:36 +0000)]
make extensions starting with __ available for all configurations
[r19791]
Matthias Braun [Tue, 27 May 2008 12:31:30 +0000 (12:31 +0000)]
make extensions starting with __ available for all configurations
[r19790]
Matthias Braun [Tue, 27 May 2008 12:22:29 +0000 (12:22 +0000)]
rework struct/union type creation, fix error 26
[r19789]
Matthias Braun [Tue, 27 May 2008 12:21:57 +0000 (12:21 +0000)]
more errors
[r19788]
Matthias Braun [Tue, 27 May 2008 11:57:12 +0000 (11:57 +0000)]
we have to skip the expected type as well in some cases
[r19787]
Matthias Braun [Tue, 27 May 2008 11:33:15 +0000 (11:33 +0000)]
new error
[r19785]
Matthias Braun [Mon, 26 May 2008 08:19:03 +0000 (08:19 +0000)]
add a panic for unimplemented initializer creation
[r19767]
Matthias Braun [Mon, 26 May 2008 08:18:47 +0000 (08:18 +0000)]
improve mode attribute parsing/skipping
[r19766]
Matthias Braun [Sun, 25 May 2008 11:30:53 +0000 (11:30 +0000)]
implement null initializers for types on the stack
[r19763]
Matthias Braun [Sun, 25 May 2008 11:30:19 +0000 (11:30 +0000)]
add mode attribute parsing
[r19762]
Matthias Braun [Sun, 25 May 2008 11:30:08 +0000 (11:30 +0000)]
add mode attribute parsing
[r19761]
Michael Beck [Mon, 12 May 2008 16:29:03 +0000 (16:29 +0000)]
add -mcpu option
[r19587]
Michael Beck [Mon, 5 May 2008 11:13:21 +0000 (11:13 +0000)]
changed for new inline I/F
[r19514]
Michael Beck [Sat, 3 May 2008 17:38:04 +0000 (17:38 +0000)]
set the floating point model for the firm graph
[r19498]
Michael Beck [Sat, 3 May 2008 16:45:15 +0000 (16:45 +0000)]
added -march, -mtune, -mfpu
[r19497]
Michael Beck [Mon, 14 Apr 2008 11:17:31 +0000 (11:17 +0000)]
add -pg support, globals removed
[r19277]
Michael Beck [Fri, 11 Apr 2008 14:20:19 +0000 (14:20 +0000)]
synchronized with edgfe, using new inline heuristic
[r19246]
Christoph Mallon [Fri, 4 Apr 2008 11:43:18 +0000 (11:43 +0000)]
__attribute__((sentinel)) is first available in GCC 4.
[r19118]
Michael Beck [Thu, 3 Apr 2008 13:17:40 +0000 (13:17 +0000)]
we can safely enable set_opt_sel_based_null_check_elim() because we use Sel's only for compound member access
[r19091]
Michael Beck [Wed, 2 Apr 2008 23:22:55 +0000 (23:22 +0000)]
- fixed ld_name for Win32 and Mach-O
- add options -f mac, -f linux
- set the default mode depending on the host OS
[r19082]
Matthias Braun [Wed, 2 Apr 2008 22:05:05 +0000 (22:05 +0000)]
use new id_unique
[r19080]
Michael Beck [Wed, 2 Apr 2008 18:34:23 +0000 (18:34 +0000)]
switch off usage of global_pointer again: globals are only SymConsts. -fpic will be completely handled by the beabi
[r19074]
Michael Beck [Wed, 2 Apr 2008 14:25:36 +0000 (14:25 +0000)]
-fpic implemented
[r19070]
Matthias Braun [Tue, 25 Mar 2008 09:28:18 +0000 (09:28 +0000)]
fix wrong usage of idx_fmt (detected by a compiler warning)
[r19023]
Michael Beck [Sat, 22 Mar 2008 23:13:04 +0000 (23:13 +0000)]
improved is_type_arithmetic()
[r19022]
Michael Beck [Sat, 22 Mar 2008 22:54:50 +0000 (22:54 +0000)]
introduce distinct complex and imaginary types
[r19021]
Michael Beck [Sat, 22 Mar 2008 19:46:36 +0000 (19:46 +0000)]
fixed indentation
[r19020]
Michael Beck [Sat, 22 Mar 2008 19:44:18 +0000 (19:44 +0000)]
- removed useless symbol from reference expression
- improved format checker
[r19019]
Michael Beck [Sat, 22 Mar 2008 18:02:30 +0000 (18:02 +0000)]
- renamed modifier to decl_modifier
- fixed ld_name for win32 build
[r19018]
Matthias Braun [Sat, 22 Mar 2008 12:57:20 +0000 (12:57 +0000)]
another test
[r19017]
Matthias Braun [Sat, 22 Mar 2008 12:56:38 +0000 (12:56 +0000)]
another test
[r19016]