uweigand [Wed, 18 Jun 2008 20:49:12 +0000 (20:49 +0000)]
* targhooks.h (struct gcc_target): New member unwind_word_mode.
(default_unwind_word_mode): Add prototype.
* targhooks.c (default_unwind_word_mode): New function.
(default_eh_return_filter_mode): Return targetm.unwind_word_mode ()
instead of word_mode.
* target-def.h (TARGET_UNWIND_WORD_MODE): New macro.
(TARGET_INITIALIZER): Use it.
* c-common.c (handle_mode_attribute): Support "unwind_word"
mode attribute.
* unwind-generic.h (_Unwind_Word, _Unwind_Sword): Use it.
* except.c (init_eh): Use targetm.unwind_word_mode () instead of
word_mode to access SjLj_Function_Context member "data".
(sjlj_emit_dispatch_table): Likewise. Also, perform type
conversion from targetm.eh_return_filter_mode () to
targetm.unwind_word_mode () if they differ.
* builtin-types.def (BT_UNWINDWORD): New primitive type.
(BT_FN_UNWINDWORD_PTR): New function type.
(BT_FN_WORD_PTR): Remove.
* builtins.def (BUILT_IN_EXTEND_POINTER): Use BT_FN_UNWINDWORD_PTR.
* except.c (expand_builtin_extend_pointer): Convert pointer to
targetm.unwind_word_mode () instead of word_mode.
* config/spu/spu-protos.h (spu_eh_return_filter_mode): Remove.
* config/spu/spu.c (spu_eh_return_filter_mode): Remove.
(spu_unwind_word_mode): New function.
(TARGET_EH_RETURN_FILTER_MODE): Do not define.
(TARGET_UNWIND_WORD_MODE): Define.
* config/spu/t-spu-elf (TARGET_LIBGCC2_CFLAGS): Remove -D__word__=SI.
hubicka [Wed, 18 Jun 2008 19:07:51 +0000 (19:07 +0000)]
* c-opts.c (c_common_post_options): PCH is not compatible with
no-unit-at-a-time.
* opts.c (handle_options): Enable unit-at-a-time at O0 along with
-fno-toplevel-reorder by default now.
* gcc.dg/weak/weak-2.c: We no longer complain about incompatibilty.
* gcc.dg/weak/weak-3.c: We no longer complain about incompatibilty.
* gcc.dg/weak/weak-4.c: We no longer complain about incompatibilty.
* gcc.dg/weak/weak-5.c: We no longer complain about incompatibilty.
* gcc.dg/weak/weak-6.c: Fix thinko in previous change.
* gcc.dg/weak/weak-7.c: Likewise.
uros [Wed, 18 Jun 2008 17:16:05 +0000 (17:16 +0000)]
PR rtl-optimization/35604
* jump.c (redirect_exp_1): Skip the condition of an IF_THEN_ELSE. We
only want to change jump destinations, not eventual label comparisons.
hubicka [Wed, 18 Jun 2008 14:07:00 +0000 (14:07 +0000)]
* cgraphunit.c (cgraph_expand_pending_functions): Give up at
syntax errors.
(cgraph_analyze_function): Likewise.
* gcc.dg/asm-wide-1.c: Do not require extra errors we output
confused by earlier errors.
* gcc.target/i386/sseregparm-2.c: Remove markers for errors not
output at parsing time.
* gcc.target/i386/sseregparm-8.c: Move here compile time errors
from gcc.target/i386/sseregparm-2.c.
* gcc.dg/weak/weak-6.c: Do not require extra errors we output
confused by earlier errors.
* gcc.dg/weak/weak-7.c: Likewise.
domob [Wed, 18 Jun 2008 13:53:32 +0000 (13:53 +0000)]
2008-06-18 Daniel Kraft <d@domob.eu>
PR fortran/36517, fortran/36492
* gfortran.dg/array_constructor_25.f03: New test.
* gfortran.dg/array_constructor_26.f03: New test.
* gfortran.dg/array_constructor_27.f03: New test.
* gfortran.dg/array_constructor_28.f03: New test.
* gfortran.dg/array_constructor_29.f03: New test.
* gfortran.dg/array_constructor_30.f03: New test.
* gfortran.dg/array_constructor_type_19.f03: New test.
* gfortran.dg/array_constructor_type_20.f03: New test.
* gfortran.dg/array_constructor_type_21.f03: New test.
2008-06-18 Daniel Kraft <d@domob.eu>
PR fortran/36517, fortran/36492
* array.c (gfc_resolve_character_array_constructor): Call
gfc_set_constant_character_len with changed length-chec argument.
* decl.c (gfc_set_constant_character_len): Changed array argument to
be a generic length-checking argument that can be used for correct
checking with typespec and in special cases where the should-be length
is different from the target length.
(build_struct): Call gfc_set_constant_character_len with changed length
checking argument and introduced additional checks for exceptional
conditions on invalid code.
(add_init_expr_to_sym), (do_parm): Call gfc_set_constant_character_len
with changed argument.
* match.h (gfc_set_constant_character_len): Changed third argument to
int for the should-be length rather than bool.
mkuvyrkov [Wed, 18 Jun 2008 07:39:45 +0000 (07:39 +0000)]
* config/mips/mips.h (ISA_HAS_CONDMOVE): Slice ISA_HAS_FP_CONDMOVE
from it.
(ISA_HAS_FP_CONDMOVE): New macro.
(ISA_HAS_FP_MADD4_MSUB4, ISA_HAS_FP_MADD3_MSUB3): New macros.
(ISA_HAS_NMADD_NMSUB): Rename to ISA_HAS_NMADD4_NMSUB4.
(ISA_HAS_NMADD3_NMSUB3): New macro.
* config/mips/mips.c (mips_rtx_costs): Update.
* config/mips/mips.md (MOVECC): Don't use FP conditional moves when
compiling for ST Loongson 2E/2F.
(madd<mode>): Rename to madd4<mode>. Update.
(madd3<mode>): New pattern.
(msub<mode>): Rename to msub4<mode>. Update.
(msub3<mode>): New pattern.
(nmadd<mode>): Rename to nmadd4<mode>. Update.
(nmadd3<mode>): New pattern.
(nmadd<mode>_fastmath): Rename to nmadd4<mode>_fastmath. Update.
(nmadd3<mode>_fastmath): New pattern.
(nmsub<mode>): Rename to nmsub4<mode>. Update.
(nmsub3<mode>): New pattern.
(nmsub<mode>_fastmath): Rename to nmsub4<mode>_fastmath. Update.
(nmsub3<mode>_fastmath): New pattern.
(mov<SCALARF:mode>_on_<MOVECC:mode>, mov<mode>cc): Update.
steven [Wed, 18 Jun 2008 05:42:23 +0000 (05:42 +0000)]
* df.h (struct df_ref): Replace 'insn' field with 'insn_info' field.
(DF_REF_INSN_INFO): New.
(DF_REF_INSN, DF_REF_INSN_UID): Rewrite macros using DF_REF_INSN_INFO.
(DF_REF_IS_ARTIFICIAL): Artificial refs are now identified as refs
with a NULL DF_REF_INSN_INFO.
(DF_INSN_INFO_GET, DF_INSN_INFO_SET): Renamed from DF_INSN_GET and
DF_INSN_SET.
(DF_INSN_INFO_LUID, DF_INSN_INFO_DEFS, DF_INSN_INFO_USES,
DF_INSN_INFO_EQ_USES): New.
(DF_INSN_LUID, DF_INSN_DEFS, DF_INSN_USES, DF_INSN_EQ_USES,
DF_INSN_UID_LUID, DF_INSN_UID_DEFS, DF_INSN_UID_USES,
DF_INSN_UID_EQ_USES): Rewrite using DF_INSN_INFO_* macros.
* df-core.c: Update comment for above changes.
(df_insn_debug_regno): Use DF_INSN_INFO_GET instead of INSN_UID and
DF_INSN_UID_* macros.
(df_ref_debug): Check for NULL DF_REF_INSN_INFO.
* df-scan.c (df_ref_record): Take a df_insn_info instead of an
insn rtx. Update all callers.
(df_def_record_1, df_defs_record, df_uses_record, df_get_call_refs,
df_ref_create_structure, df_insn_refs_collect): Likewise.
(df_ref_equal_p): Compare DF_REF_INSN_INFO pointers for the refs.
* df-problems.c (df_chain_dump): Test for non-NULL DF_REF_INSN_INFO.
(df_live_bb_local_compute): Retrieve DF_INSN_INFO, use DF_INSN_INFO_*
macros to access the insn refs.
(df_chain_top_dump, df_chain_bottom_dump, df_byte_lr_alloc): Likewise.
* fwprop.c (use_killed_between): Use DF_REF_INSN accessor macro.
(all_uses_available): Retrieve DF_INSN_INFO for def_insn, and use it
for accessing the refs.
(try_fwprop_subst): Likewise.
* ddg.c (add_cross_iteration_register_deps): Use DF_REF_INSN macro.
* web.c (union_defs): Retrieve DF_INSN_INFO for def_insn, and use it
for accessing the refs.
* loop-invariant.c (invariant_for_use): Use DF_REF_BB macro.
(check_dependencies): Use DF_INSN_INFO_GET, use DF_INSN_INFO_* macros
to look at the insn refs.
(record_uses): Likewise.
* dce.c (deletable_insn_p): Don't tolerate artificial DEFs in this
function anymore.
(mark_artificial_uses): Don't mark_insn for artificial refs.
(mark_reg_rependencies): Likewise.
* doc/rtl.texi: Remove documentation of ADDRESSOF.
config/
* override.m4: Use m4_version_prereq throughout.
(_AC_ARG_VAR_VALIDATE, AC_MSG_FAILURE): Backport from git
Autoconf: output pwd along with fatal errors, so the right
config.log file is hinted at more prominently.
(PARSE_ARGS): Push setting of ac_pwd in this diversion.
(_GCC_AUTOCONF_VERSION): New, define to 2.59 if not defined.
(_GCC_AUTOCONF_VERSION_CHECK): New macro, require use of Autoconf
version _GCC_AUTOCONF_VERSION throughout the tree.
(m4_wrap): New override, fix for Posix semantics of m4wrap.
domob [Tue, 17 Jun 2008 20:24:20 +0000 (20:24 +0000)]
2008-06-17 Daniel Kraft <d@domob.eu>
PR fortran/36112
* array.c (gfc_resolve_character_array_constructor): Check that all
elements with constant character length have the same one rather than
fixing it if no typespec is given, emit an error if they don't. Changed
return type to "try" and return FAILURE for the case above.
(gfc_resolve_array_constructor): Removed unneeded call to
gfc_resolve_character_array_constructor in this function.
* gfortran.h (gfc_resolve_character_array_constructor): Returns try.
* trans-array.c (get_array_ctor_strlen): Return length of first element
rather than last element.
* resolve.c (gfc_resolve_expr): Handle FAILURE return from
gfc_resolve_character_array_constructor.
2008-06-17 Daniel Kraft <d@domob.eu>
PR fortran/36112
* gfortran.dg/bounds_check_array_ctor_1.f90: New test.
* gfortran.dg/bounds_check_array_ctor_2.f90: New test.
* gfortran.dg/bounds_check_array_ctor_3.f90: New test.
* gfortran.dg/bounds_check_array_ctor_4.f90: New test.
* gfortran.dg/bounds_check_array_ctor_5.f90: New test.
* gfortran.dg/bounds_check_array_ctor_6.f90: New test.
* gfortran.dg/bounds_check_array_ctor_7.f90: New test.
* gfortran.dg/bounds_check_array_ctor_8.f90: New test.
* gfortran.dg/arrayio_0.f90: Fixed invalid array constructor.
* gfortran.dg/char_cons_len.f90: Ditto.
* gfortran.dg/char_initializer_actual.f90: Ditto.
* gfortran.dg/pr15959.f90: Ditto.
* gfortran.dg/transfer_simplify_2.f90: Ditto.
* gfortran.dg/char_length_1.f90: Changed expected error messages.
pault [Tue, 17 Jun 2008 18:08:24 +0000 (18:08 +0000)]
2008-06-17 Paul Thomas <pault@gcc.gnu.org>
PR fortran/34396
* resolve.c (add_dt_to_dt_list): New function.
(resolve_fl_derived): Call new function for pointer components
and when derived type resolved.
2008-06-17 Paul Thomas <pault@gcc.gnu.org>
PR fortran/36366
* gfortran.dg/used_types_20.f90: New test.
irar [Mon, 16 Jun 2008 10:39:07 +0000 (10:39 +0000)]
PR tree-optimization/36493
* tree-vect-transform.c (vect_create_data_ref_ptr): Remove TYPE from
the arguments list. Use VECTYPE to create vector pointer.
(vectorizable_store): Fail if accesses through a pointer to vectype
do not alias the original memory reference operands.
Call vect_create_data_ref_ptr without the removed argument.
(vectorizable_load): Likewise.
(vect_setup_realignment): Call vect_create_data_ref_ptr without the
removed argument.
simartin [Sat, 14 Jun 2008 05:21:30 +0000 (05:21 +0000)]
gcc/cp/
2008-06-14 Simon Martin <simartin@users.sourceforge.net>
PR c++/35320
* decl2.c (grokbitfield): Receive the list of attributes, pass it to
grokdeclarator and apply it to the created declaration.
* cp-tree.h (grokbitfield): Update prototype.
* parser.c (cp_parser_member_declaration): Don't apply the attributes
since they are now applied in grokbitfield. Adjusted the call to
grokbitfield.
(cp_parser_objc_class_ivars): Likewise.
gcc/testsuite/
2008-06-14 Simon Martin <simartin@users.sourceforge.net>
PR c++/35320
* g++.dg/parse/bitfield3.C: New test.
hainque [Fri, 13 Jun 2008 22:07:28 +0000 (22:07 +0000)]
2008-06-13 Olivier Hainque <hainque@adacore.com>
ada/
* decl.c (FORCE_PREFERRED_STACK_BOUNDARY_IN_MAIN): Define to 0
if undefined.
(gnat_to_gnu_entity) <case E_Function/Procedure>: Request stack
realignment with force_align_arg_pointer attribute on foreign
convention subprograms if FORCE_PREFERRED_STACK_BOUNDARY_IN_MAIN.
testsuite/
* gnat.dg/task_stack_align.adb: New test.
jvdelisle [Fri, 13 Jun 2008 20:35:12 +0000 (20:35 +0000)]
2008-06-13 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/35863
* gfortran.dg/widechar_IO_1.f90: New test.
* gfortran.dg/widechar_IO_2.f90: New test.
* gfortran.dg/widechar_IO_3.f90: New test.
* gfortran.dg/widechar_IO_4.f90: New test.
jvdelisle [Fri, 13 Jun 2008 20:30:48 +0000 (20:30 +0000)]
2008-06-13 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/35863
* trans-io.c (gfc_build_io_library_fndecls): Build declaration for
transfer_character_wide which includes passing in the character kind to
support wide character IO. (transfer_expr): If the kind == 4, create the
argument and build the call.
* gfortran.texi: Fix typo.
jvdelisle [Fri, 13 Jun 2008 20:28:08 +0000 (20:28 +0000)]
2008-06-13 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/35863
* libgfortran.h: Change l8_to_l4_offset to big_endian and add endian_off.
* runtime/main.c: Fix error in comment. Change l8_to_l4_offset to
big_endian. (determine_endianness): Add endian_off and set its value
according to big_endian.
* gfortran.map: Add symbol for new _gfortran_transfer_character_wide.
* io/io.h: Add prototype declarations for new functions.
* io/list_read.c (list_formatted_read_scalar): Modify to handle kind=4.
(list_formatted_read): Calculate stride based on kind for character type
and use it when calling list_formatted_read_scalar.
* io/inquire.c (inquire_via_unit): Change l8_to_l4_offset to big_endian.
* io/open.c (st_open): Change l8_to_l4_offset to big_endian.
* io/read.c (read_a_char4): New function to handle formatted read.
* io/write.c: Define GFC_CHAR4(x) to improve readability of code.
(write_a_char4): New function to handle formatted write.
(write_character): Modify to accept the kind parameter and adjust for
endianess of the machine. (list_formatted_write): Calculate the stride
resulting from the kind and adjust the list_formatted_write_scalar call
accordingly. (nml_write_obj): Adjust calls to write_character.
(namelist_write): Likewise.
* io/transfer.c (formatted_transfer_scaler): Rename 'len' argument to
'kind' argument to better describe what it is. Add calls to new
functions for kind == 4. (formatted_transfer): Modify to handle the case
of type character and kind equals 4 to pass in the kind to the transfer
routines. (transfer_character_wide): Add this new function.
(transfer_array): Don't set kind to the character string length. Adjust
strides bases on character kind.
(unformatted_read): Adjust size based on kind for character types.
(unformatted_write): Likewise. (data_transfer_init): Change
l8_to_l4_offset to big_endian.
hainque [Fri, 13 Jun 2008 10:27:29 +0000 (10:27 +0000)]
2008-06-13 Olivier Hainque <hainque@adacore.com>
* utils.c (rest_of_record_type_compilation): When computing
encodings for the components of a variable size type, early
strip conversions on the current position expression to make
sure it's shape is visible. Use remove_conversions for this
purpose.
paolo [Fri, 13 Jun 2008 07:35:22 +0000 (07:35 +0000)]
2008-06-13 Paolo Carlini <paolo.carlini@oracle.com>
* include/tr1_impl/array (array<>::assign): Do not define in
C++0x mode, per DR 776...
(array<>::fill): ... define instead.
* doc/xml/manual/intro.xml: Add an entry for DR 776.
* testsuite/23_containers/array/requirements/fill.cc: New.
ebotcazou [Thu, 12 Jun 2008 13:19:06 +0000 (13:19 +0000)]
* decl.c (gnat_to_gnu_entity) <E_Record_Subtype>: In the case of a
constrained subtype of a discriminated type, discard the fields that
are beyond its limits according to its size.
paolo [Thu, 12 Jun 2008 10:17:53 +0000 (10:17 +0000)]
2008-06-12 Paolo Carlini <paolo.carlini@oracle.com>
* include/std/complex (pow(const complex<>&, int)): Do not define in
C++0x mode, per DR 844.
* include/tr1/complex (pow(const complex<>&, int)): Remove.
* doc/xml/manual/intro.xml: Add an entry for DR 844.
* testsuite/26_numerics/complex/dr844.cc: New.
* testsuite/tr1/8_c_compatibility/complex/overloads_int.cc: Adjust.
ktietz [Thu, 12 Jun 2008 09:19:35 +0000 (09:19 +0000)]
2008-06-11 Kai Tietz <kai.tietz@onevision.com>
* config/i386/i386.c (ix86_compute_frame_layout): Disable red zone for
w64 abi.
(ix86_expand_prologue): Likewise.
(ix86_force_to_memory): Likewise.
(ix86_free_from_memory): Likewise.
jakub [Wed, 11 Jun 2008 21:43:45 +0000 (21:43 +0000)]
* libgomp.h (struct gomp_task): Add in_tied_task field.
* task.c (gomp_init_task): Initialize it.
(GOMP_task): Likewise. Call gomp_team_barrier_set_task_pending
unconditionally. Don't call gomp_team_barrier_wake if
current task is implicit or if(0) from implicit and number of
running tasks is equal to nthreads - 1.
jsm28 [Wed, 11 Jun 2008 20:02:55 +0000 (20:02 +0000)]
2008-06-11 Edmar Wienskoski <edmar@freescale.com>
PR target/36425
* config/rs6000/rs6000.c (rs6000_override_options): Set
rs6000_isel conditionally to the absence of comand line
override.
* config/rs6000/linuxspe.h (SUBSUBTARGET_OVERRIDE_OPTIONS):
Remove duplicate rs6000_isel setting.
* config/rs6000/eabispe.h: Ditto.
rguenth [Wed, 11 Jun 2008 18:19:28 +0000 (18:19 +0000)]
2008-06-11 Richard Guenther <rguenther@suse.de>
* alias.c (get_alias_set): Use the element alias-set for arrays.
(record_component_aliases): For arrays and vectors do nothing.
* c-common.c (strict_aliasing_warning): Handle the cases
of alias set zero explicitly.
* Makefile.in (dfp.o-warn): Add -Wno-error.