GCC Administrator [Thu, 17 Apr 2008 00:17:31 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134372
Andreas Tobler [Wed, 16 Apr 2008 21:30:11 +0000 (21:30 +0000)]
re PR libstdc++/35915 (atomic.cc:31:20: error: stdint.h: No such file)
2008-04-16 Andreas Tobler <a.tobler@schweiz.org>
PR libstdc++/35915
* configure.ac: Use GCC_HEADER_STDINT.
* src/atomic.cc: Include gstdint.h
* configure: Regenerate.
* Makefile.in: Likewise.
* doc/Makefile.in: Likewise.
* include/Makefile.in: Likewise.
* libmath/Makefile.in: Likewise.
* libsupc++/Makefile.in: Likewise.
* po/Makefile.in: Likewise.
* src/Makefile.in: Likewise.
* testsuite/Makefile.in: Likewise.
* aclocal.m4: Likewise.
From-SVN: r134368
Tom Tromey [Wed, 16 Apr 2008 21:10:37 +0000 (21:10 +0000)]
Makefile.in (TAGS_SOURCES): New variable.
* Makefile.in (TAGS_SOURCES): New variable.
(TAGS): New target.
From-SVN: r134367
Paul Thomas [Wed, 16 Apr 2008 20:53:07 +0000 (20:53 +0000)]
re PR fortran/35932 (ICE: CHAR with array arg and also a KIND arg)
2008-04-16 Paul Thomas <pault@gcc.gnu.org>
PR fortran/35932
* trans-intrinsic.c (gfc_conv_intrinsic_char): Even though KIND
is not used, the argument must be converted.
2008-04-16 Paul Thomas <pault@gcc.gnu.org>
PR fortran/35932
* gfortran.dg/intrinsic_char_1.f90: New test.
From-SVN: r134364
Uros Bizjak [Wed, 16 Apr 2008 18:39:08 +0000 (20:39 +0200)]
re PR target/35944 (wrong result for MOD with kind=10 for some array argument values)
PR target/35944
* gfortran.dg/pr35944-1.f90: New test.
* gfortran.dg/pr35944-2.f90: Ditto.
From-SVN: r134359
Andrew Haley [Wed, 16 Apr 2008 18:20:16 +0000 (18:20 +0000)]
libgcj_bc.c: (_Jv_LookupInterfaceMethod...
2008-04-16 Andrew Haley <aph@littlepinkcloud.com>
* libgcj_bc.c: (_Jv_LookupInterfaceMethod, _Jv_MonitorExit,
_Jv_RegisterResource, _Jv_JNI_PopSystemFrame): Add.
From-SVN: r134358
Andrew Haley [Wed, 16 Apr 2008 16:49:11 +0000 (16:49 +0000)]
libgcj_bc.c: Sort lines.
2008-04-16 Andrew Haley <aph@redhat.com>
* libgcj_bc.c: Sort lines.
From-SVN: r134356
Jakub Jelinek [Wed, 16 Apr 2008 16:01:57 +0000 (18:01 +0200)]
re PR c/35739 (ICE with _Decimal128 and va_list)
PR c/35739
* tree-nrv.c (tree_nrv): Don't optimize if result_type is GIMPLE
reg type.
* gcc.dg/dfp/pr35739.c: New test.
From-SVN: r134351
Jakub Jelinek [Wed, 16 Apr 2008 16:00:34 +0000 (18:00 +0200)]
re PR tree-optimization/35899 (ICE on filesystem code)
PR tree-optimization/35899
* tree-inline.c (expand_call_inline): Use GIMPLE_STMT_OPERAND
rather than TREE_OPERAND.
* gcc.dg/pr35899.c: New test.
From-SVN: r134350
Jakub Jelinek [Wed, 16 Apr 2008 15:56:58 +0000 (17:56 +0200)]
re PR target/35662 (gfortran interfaces badly with glibc sincosf() causing breakage)
PR target/35662
* f95-lang.c (gfc_init_builtin_functions): Make sure
BUILT_IN_SINCOS{,F,L} types aren't varargs.
* gfortran.dg/pr35662.f90: New test.
From-SVN: r134349
Uros Bizjak [Wed, 16 Apr 2008 15:53:59 +0000 (17:53 +0200)]
re PR target/35944 (wrong result for MOD with kind=10 for some array argument values)
PR target/35944
* config/i386/i386.md (fmodxf3): Copy operand 1 and operand 2 into
temporary registers. Change operand predicate to general_operand.
(remainderxf3): Ditto.
From-SVN: r134348
Richard Guenther [Wed, 16 Apr 2008 15:16:42 +0000 (15:16 +0000)]
Makefile.in (tree-affine.o): Add $(FLAGS_H) dependency.
2008-04-16 Richard Guenther <rguenther@suse.de>
* Makefile.in (tree-affine.o): Add $(FLAGS_H) dependency.
* tree-affine.c (aff_combination_expand): Look through some
conversions.
* gcc.dg/tree-ssa/loop-35.c: New testcase.
From-SVN: r134346
Samuel Tardieu [Wed, 16 Apr 2008 12:37:38 +0000 (12:37 +0000)]
re PR ada/29015 (Ada 2005 observer pattern with mutually dependent packages and containers produces compiler error)
gcc/ada/
PR ada/29015
* sem_ch12.adb (Instantiate_Type): Check whether the full view of
the type is known instead of the underlying type.
gcc/testsuite/
PR ada/29015
* gnat.dg/incomplete1.ads, gnat.dg/incomplete2.ads,
gnat.dg/incomplete2.adb: New.
From-SVN: r134345
Hans-Peter Nilsson [Wed, 16 Apr 2008 02:44:01 +0000 (02:44 +0000)]
re PR tree-optimization/35629 (gcc.dg/tree-ssa/loop-25.c scan-tree-dump-times profile fails)
PR tree-optimization/35629
* gcc.dg/tree-ssa/loop-25.c: Xfail all but the first
scan-tree-dump-times.
From-SVN: r134339
GCC Administrator [Wed, 16 Apr 2008 00:17:12 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134337
Doug Kwan [Tue, 15 Apr 2008 19:56:20 +0000 (19:56 +0000)]
dwarf2asm.c (dw2_assemble_integer): Cast to unsigned HOST_WIDE_INT for hex printing.
2008-04-15 Doug Kwan <dougkwan@google.com>
* dwarf2asm.c (dw2_assemble_integer): Cast to unsigned HOST_WIDE_INT
for hex printing.
* tree-pretty-print.c (dump_generic_node): Ditto.
* final.c (output_addr_const): Ditto.
* dwarf2out.c (output_cfi): Ditto.
* c-pretty-print.c (pp_c_integer_constant): Ditto.
* print-rtl.c (print_rtx): Ditto.
* print-tree.c (print_node_brief, print_node): Ditto.
* c-common.c (match_case_to_enum_1): Ditto.
* sched-vis.c (print_value): Ditto.
* config/i386/i386.c (print_operand): Cast to long unsigned int
for hex printing.
From-SVN: r134331
Danny Smith [Tue, 15 Apr 2008 19:44:16 +0000 (19:44 +0000)]
libgcc2.c [L_trampoline]: Remove unnecessary prototype for
Windows VirtualProtect function.
From-SVN: r134330
Paul Thomas [Tue, 15 Apr 2008 19:40:33 +0000 (19:40 +0000)]
re PR fortran/35864 (Revision 133965 broke gfortran.dg/initialization_1.f90)
2008-04-15 Paul Thomas <pault@gcc.gnu.org>
PR fortran/35864
* expr.c (scalarize_intrinsic_call): Reorder identification of
array argument so that if one is not found a segfault does not
occur. Return FAILURE if all scalar arguments.
From-SVN: r134329
Jan Hubicka [Tue, 15 Apr 2008 19:37:22 +0000 (21:37 +0200)]
gengtype.c (write_root): Param_is argument is OK.
* gengtype.c (write_root): Param_is argument is OK.
* expr.c (expand_expr_real_1): Update call of get_exception_*.
* function.h: Include varray.h
(rtl_eh): New stucture based on except.c one.
(call_site_record): New forward declaration and vector type.
* calls.c (emit_call_1): Do not call
note_current_region_may_contain_throw.
* except.c (eh_status): Remove cur_region, try_region since they are
unused.
Move filter, exc_ptr, ttype_data, ehspec_data, action_record_data and
exception_handler_label_map, ehr_stackadj, ehr_handler, ehr_label,
sjlj_fc, sjlj_exit_after to rth_eh in function.h.
Remove call_site_data_used, call_site_data_size.
Turn call_site_record into vector in function.h.
(note_current_region_may_contain_throw): Remove.
(get_exception_pointer, get_exception_filter): Do not take struct
function argument; update.
(add_ehl_entry, find_exception_handler_labels, ehspec_filter_hash,
add_ttypes_entry, add_ehspec_entry, assign_filter_values,
build_post_landing_pads, dw2_build_landing_pads,
sjlj_assign_call_site_values, sjlj_mark_call_sites,
sjlj_emit_function_enter, sjlj_emit_function_enter,
sjlj_emit_function_exit, sjlj_emit_dispatch_table,
sjlj_build_landing_pads, finish_eh_generation,
remove_exception_handler_label, remove_eh_handler,
maybe_remove_eh_handler, add_reachable_handler,
reachable_handlers, expand_builtin_eh_return,
expand_eh_return, add_action_record, collect_one_action_chain,
add_call_site, convert_to_eh_region_ranges, sjlj_size_of_call_site_table,
sjlj_output_call_site_table, output_function_exception_table,
* except.h (note_current_region_may_contain_throw): Remove
(get_exception_pointer, get_exception_filter): Do not take struct
function argument.
* Makefile.in (GTFILES): Put varargs before struct function.
From-SVN: r134328
Samuel Tardieu [Tue, 15 Apr 2008 19:05:29 +0000 (19:05 +0000)]
[multiple changes]
2008-04-15 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
PR ada/22387
* exp_ch5.adb (Expand_Assign_Record): Within an initialization
procedure for a derived type retrieve the discriminant values from
the parent using the corresponding discriminant.
2008-04-15 Samuel Tardieu <sam@rfc1149.net>
gcc/testsuite/
PR ada/22387
* gnat.dg/specs/corr_discr.ads: New.
From-SVN: r134326
Eric Botcazou [Tue, 15 Apr 2008 18:22:51 +0000 (18:22 +0000)]
tree-ssa-structalias.c (get_constraint_for_component_ref): Do not punt for STRING_CST.
* tree-ssa-structalias.c (get_constraint_for_component_ref): Do not
punt for STRING_CST.
(get_constraint_for): Deal with STRING_CST here instead.
From-SVN: r134325
Richard Guenther [Tue, 15 Apr 2008 15:54:26 +0000 (15:54 +0000)]
tree-ssa-propagate.c (substitute_and_fold): Substitute statements in a basic-block with a backward walk.
2008-04-15 Richard Guenther <rguenther@suse.de>
* tree-ssa-propagate.c (substitute_and_fold): Substitute
statements in a basic-block with a backward walk. Do not
substitute into dead statements but instead remove those.
* gcc.dg/fold-compare-2.c: Adjust testcase.
* gcc.dg/tree-ssa/pr21086.c: Likewise.
From-SVN: r134322
Richard Guenther [Tue, 15 Apr 2008 15:41:52 +0000 (15:41 +0000)]
params.def (PARAM_MAX_FIELDS_FOR_FIELD_SENSITIVE): Set default to zero, thus disable creation of SFTs.
2008-04-15 Richard Guenther <rguenther@suse.de>
* params.def (PARAM_MAX_FIELDS_FOR_FIELD_SENSITIVE): Set default
to zero, thus disable creation of SFTs.
* gcc.dg/tree-ssa/salias-1.c: Remove.
* gcc.dg/tree-ssa/pr26421.c: Adjust pattern.
* gcc.dg/tree-ssa/alias-15.c: Likewise.
* gcc.dg/tree-ssa/ssa-lim-3.c: Run at -O2.
From-SVN: r134321
Eric Botcazou [Tue, 15 Apr 2008 15:39:42 +0000 (15:39 +0000)]
tree-predcom.c (suitable_reference_p): Return false if the reference can throw.
* tree-predcom.c (suitable_reference_p): Return false if the
reference can throw.
From-SVN: r134319
Jakub Jelinek [Tue, 15 Apr 2008 15:09:42 +0000 (17:09 +0200)]
re PR c/35751 (ICE with invalid variable after #pragma omp parallel)
PR c/35751
* c-decl.c (finish_decl): If extern or static var has variable
size, set TREE_TYPE (decl) to error_mark_node.
* decl.c (layout_var_decl): If extern or static var has variable
size, set TREE_TYPE (decl) to error_mark_node.
* gcc.dg/gomp/pr35751.c: New test.
* g++.dg/gomp/pr35751.C: New test.
From-SVN: r134317
Rafael Espindola [Tue, 15 Apr 2008 13:50:28 +0000 (13:50 +0000)]
fix file name on ChangeLog
From-SVN: r134316
Rafael Avila de Espindola [Tue, 15 Apr 2008 13:48:52 +0000 (13:48 +0000)]
tree-const.c (tree_call_nonnegative_warnv_p): Remove local variable arg1.
2008-04-15 Rafael Espindola <espindola@google.com>
* gcc/tree-const.c (tree_call_nonnegative_warnv_p): Remove local
variable arg1.
From-SVN: r134315
Richard Guenther [Tue, 15 Apr 2008 13:28:08 +0000 (13:28 +0000)]
tree-ssa-sccvn.h (vn_reference_lookup): Adjust prototype.
2008-04-15 Richard Guenther <rguenther@suse.de>
* tree-ssa-sccvn.h (vn_reference_lookup): Adjust prototype.
* tree-ssa-sccvn.c (vn_reference_lookup): New parameter maywalk.
(visit_reference_op_load): Do walk vuse-vdef chains on
vn_reference_lookup.
(visit_reference_op_store): But do not here.
* tree-vn.c (vn_lookup): Do not walk vuse-vdef chains on
vn_reference_lookup.
(vn_lookup_with_vuses): But do so here.
From-SVN: r134314
Samuel Tardieu [Tue, 15 Apr 2008 11:59:39 +0000 (11:59 +0000)]
re PR ada/28733 (GNAT crash while compiling Ada-2005 code)
2008-04-15 Samuel Tardieu <sam@rfc1149.net>
Gary Dismukes <dismukes@adacore.com>
gcc/ada/
PR ada/28733
* sem_ch8.adb (Analyze_Use_Package): Do not allow "use" of something
which is not an entity (and hence not a package).
(End_Use_Package): Ditto.
Co-Authored-By: Gary Dismukes <dismukes@adacore.com>
From-SVN: r134313
Samuel Tardieu [Tue, 15 Apr 2008 11:02:58 +0000 (11:02 +0000)]
[multiple changes]
2008-04-15 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
PR ada/16086
* sem_ch12.adb (Analyze_Formal_Subprogram): The default can be any
protected operation that matches the signature, not only an entry, a
regular subprogram or a literal.
2008-04-15 Samuel Tardieu <sam@rfc1149.net>
gcc/testsuite/
PR ada/16086
* gnat.dg/prot_def.adb: New.
From-SVN: r134312
Eric Botcazou [Tue, 15 Apr 2008 07:18:21 +0000 (07:18 +0000)]
ada-tree.h (DECL_BY_COMPONENT_PTR_P): Use DECL_LANG_FLAG_3.
* ada-tree.h (DECL_BY_COMPONENT_PTR_P): Use DECL_LANG_FLAG_3.
* decl.c (gnat_to_gnu_entity) <object>: Call maybe_pad_type only
if a size or alignment is specified. Do not take into account
alignment promotions for the computation of the object's size.
<type>: Call maybe_pad_type only if a size or alignment is specified.
(maybe_pad_type): Really reuse the RM_Size of the original type if
requested.
* trans.c (Attribute_to_gnu): Fix a couple of nits.
* utils2.c (build_binary_op) <MODIFY_EXPR>: Merge related conditional
statements. Use the padded view of the type when copying between
padded objects of the same underlying type.
From-SVN: r134310
Ian Lance Taylor [Tue, 15 Apr 2008 05:57:00 +0000 (05:57 +0000)]
* fold-const.c (fold_overflow_warning): Remove assertion.
From-SVN: r134307
GCC Administrator [Tue, 15 Apr 2008 00:17:41 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134303
Ben Elliston [Tue, 15 Apr 2008 00:03:46 +0000 (10:03 +1000)]
Fix bootstrap on Alpha.
* config/alpha/alpha.c (alpha_initialize_trampoline): Remove temp,
temp1 local variables.
From-SVN: r134299
Danny Smith [Mon, 14 Apr 2008 23:53:54 +0000 (23:53 +0000)]
re PR target/35921 (Con/de-structor definition fails to override dllimport declaration)
PR target/35921
* optimize.c (maybe_clone_body): Copy DECL_DLLIMPORT_P flag
to clone.
From-SVN: r134298
Benjamin Kosnik [Mon, 14 Apr 2008 23:53:15 +0000 (23:53 +0000)]
re PR libstdc++/35816 (libstdc++ build during bootstrap is optimizing away atomic builtin check)
2008-04-14 Benjamin Kosnik <bkoz@redhat.com>
PR libstdc++/35816
* acinclude.m4 (GLIBCXX_ENABLE_ATOMIC_BUILTINS): Set CXXFLAGS to
-O0 when checking for atomic builtins.
* configure: Regenerate.
From-SVN: r134297
Zuxy Meng [Mon, 14 Apr 2008 23:47:39 +0000 (23:47 +0000)]
re PR target/35661 (__attribute__((cold)) generates wrong code)
2008-04-16 Zuxy Meng <zuxy.meng@gmail.com>
PR target/35661
* config/i386/winnt.c (i386_pe_section_type_flags): Mark
".text.unlikely" section as executable.
From-SVN: r134296
James E Wilson [Mon, 14 Apr 2008 23:38:40 +0000 (16:38 -0700)]
ia64.c (rtx_needs_barrier): Handle UNSPEC_FR_SQRT_RECIP_APPROX_RES.
* config/ia64/ia64.c (rtx_needs_barrier): Handle
UNSPEC_FR_SQRT_RECIP_APPROX_RES.
* config/ia64/ia64.c (UNSPEC_FR_SQRT_RECIP_APPROX_RES): Define.
(divsi3_internal, divdi3_internal_lat, divdi3_internal_thr,
divsf3_internal_lat, sqrt_approx, sqrtsf2_internal_thr,
divdf3_internal_lat, sqrtdf2_internal_thr, divxf3_internal_lat,
divxf3_internal_thr, sqrtxf2_internal_thr, recip_approx): Use it.
From-SVN: r134295
Ralf Wildenhues [Mon, 14 Apr 2008 21:07:59 +0000 (21:07 +0000)]
vms_data.ads: Fix typo in constant.
gcc/ada/
* vms_data.ads: Fix typo in constant.
* gen-soccon.c: Fix typo in error string.
* gnat_rm.texi (Pragma Optimize_Alignment, Pragma Postcondition):
Fix typos.
* a-calcon.ads, a-calend-vms.adb, a-calend.adb, a-crdlli.ads,
bcheck.adb, checks.adb, einfo.ads, errout.adb, erroutc.adb,
erroutc.ads, exp_attr.adb, exp_ch11.adb, exp_ch2.adb,
exp_ch5.adb, exp_ch9.adb, exp_ch9.ads, exp_pakd.adb,
exp_util.adb, fmap.adb, g-soccon-linux-mips.ads,
g-soccon-rtems.ads, g-timsta.adb, g-timsta.ads, lib-writ.ads,
mlib-tgt-specific-linux.adb, mlib-tgt-specific-tru64.adb,
s-interr-vxworks.adb, s-interr.adb, s-osinte-lynxos.ads,
s-rident.ads, s-taprop-solaris.adb, s-tassta.adb, s-win32.ads,
sem_aggr.adb, sem_attr.ads, sem_ch10.adb, sem_ch13.ads,
sem_ch3.adb, sem_ch6.adb, sem_ch7.adb, sem_ch8.adb, sem_ch9.adb,
sem_prag.ads, sem_res.adb, sem_util.adb, sem_util.ads,
sinfo.ads: Fix typos in comments.
From-SVN: r134291
Benjamin Kosnik [Mon, 14 Apr 2008 19:22:27 +0000 (19:22 +0000)]
c++config: Guard _LDBL_ macros with __cplusplus.
2008-04-14 Benjamin Kosnik <bkoz@redhat.com>
* include/bits/c++config: Guard _LDBL_ macros with __cplusplus.
From-SVN: r134288
Ian Lance Taylor [Mon, 14 Apr 2008 19:18:31 +0000 (19:18 +0000)]
flags.h (POINTER_TYPE_OVERFLOW_UNDEFINED): Define.
gcc/:
* flags.h (POINTER_TYPE_OVERFLOW_UNDEFINED): Define.
* fold-const.c (fold_comparison): If appropriate, test
POINTER_TYPE_OVERFLOW_UNDEFINED, and issue an overflow warning.
(fold_binary): Test POINTER_TYPE_OVERFLOW_UNDEFINED when
reassociating a pointer type.
* doc/invoke.texi (Optimize Options): Document that
-fstrict-overflow applies to pointer wraparound.
gcc/testsuite/:
* gcc.dg/strict-overflow-6.c: New.
* gcc.dg/no-strict-overflow-7.c: New.
* gcc.dg/Wstrict-overflow-22.c: New.
From-SVN: r134287
Thomas Koenig [Mon, 14 Apr 2008 18:50:57 +0000 (18:50 +0000)]
re PR libfortran/32972 (performance of pack/unpack)
2008-04-14 Thomas Koenig <tkoenig@gcc.gnu.org>
PR libfortran/32972
* intrinsics/spread_generic.c (spread): Use spread_i2()
for GFC_DTYPE_DERIVED_2 (fix typo from previous commit).
From-SVN: r134286
Samuel Tardieu [Mon, 14 Apr 2008 16:41:58 +0000 (16:41 +0000)]
storage.ads: Fix expected error message.
gcc/testsuite/
* gnat.dg/specs/storage.ads: Fix expected error message.
From-SVN: r134281
Robert Dewar [Mon, 14 Apr 2008 16:38:16 +0000 (16:38 +0000)]
sem_prag.adb (Analyze_Pragma, [...]): Extend error to every non-object and change error message.
2008-04-14 Robert Dewar <dewar@adacore.com>
gcc/ada/
* sem_prag.adb (Analyze_Pragma, Linker_Section case): Extend error
to every non-object and change error message.
gcc/testsuite/
* gnat.dg/specs/linker_section.ads: Check for more error conditions.
From-SVN: r134280
Bernd Schmidt [Mon, 14 Apr 2008 15:55:04 +0000 (15:55 +0000)]
Sigh.
Sigh. Fix the license of the new file from the previous commit to be
GPLv3, and fix the copyright year.
From-SVN: r134277
Bernd Schmidt [Mon, 14 Apr 2008 15:51:51 +0000 (15:51 +0000)]
Add missing file from my previous commit...
From-SVN: r134276
Robert Dewar [Mon, 14 Apr 2008 15:01:42 +0000 (15:01 +0000)]
sem_util.ads, [...] (In_Subprogram): Remove.
2008-04-14 Robert Dewar <dewar@adacore.com>
gcc/ada/
* sem_util.ads, sem_util.adb (In_Subprogram): Remove.
* sem_attr.adb (Anayze_Attribute): Check for Current_Subprogram
directly.
From-SVN: r134273
Samuel Tardieu [Mon, 14 Apr 2008 13:41:25 +0000 (13:41 +0000)]
re PR ada/18680 (pragma Linker_Section problems)
gcc/ada/
PR ada/18680
* sem_prag.adb (Analyze_Pragma, Linker_Section case): Refuse to
apply pragma Linker_Section on type.
gcc/testsuite/
PR ada/18680
* gnat.dg/specs/linker_section.ads: New.
From-SVN: r134266
Samuel Tardieu [Mon, 14 Apr 2008 12:11:06 +0000 (12:11 +0000)]
re PR ada/16098 (Illegal program not detected, RM 13.1(6))
gcc/ada/
PR ada/16098
* sem_prag.adb (Error_Pragma_Ref): New.
(Process_Convention): Specialized message for non-local
subprogram renaming. Detect the problem in homonyms as well.
gcc/testsuite/
PR ada/16098
* gnat.dg/specs/renamings.ads: New.
From-SVN: r134262
Samuel Tardieu [Mon, 14 Apr 2008 12:10:16 +0000 (12:10 +0000)]
re PR ada/15915 (Illegal program not detected, RM 13.11(15))
gcc/ada/
PR ada/15915
* sem_util.ads, sem_util.adb (Denotes_Variable): New function.
* sem_ch12.adb (Instantiate_Object): Use it.
* sem_ch13.adb (Analyze_Attribute_Definition_Clause): Ensure that
storage pool denotes a variable as per RM 13.11(15).
gcc/testsuite/
PR ada/15915
* gnat.dg/specs/storage.ads: New.
From-SVN: r134261
Samuel Tardieu [Mon, 14 Apr 2008 12:08:31 +0000 (12:08 +0000)]
sem_util.ads, [...] (In_Subprogram): New function.
gcc/ada/
* sem_util.ads, sem_util.adb (In_Subprogram): New function.
* sem_attr.adb (Analyze_Attribute, Attribute_Old case): Use it.
gcc/testsuite/
* gnat.dg/deep_old.adb: New.
From-SVN: r134260
Eric Botcazou [Mon, 14 Apr 2008 11:34:05 +0000 (11:34 +0000)]
loop_address2.adb: New test.
* gnat.dg/loop_address2.adb: New test.
From-SVN: r134258
Rolf Ebert [Mon, 14 Apr 2008 09:39:39 +0000 (09:39 +0000)]
re PR ada/20822 (makeinfo cannot process gnat_ugn_unw.texi)
2008-04-14 Rolf Ebert <rolf.ebert.gcc@gmx.de>
gcc/ada/
PR ada/20822
* xgnatugn.adb (Put_Line): New procedure, ensuring Unix
line endings even on non-Unix platforms.
From-SVN: r134257
Samuel Tardieu [Mon, 14 Apr 2008 09:38:34 +0000 (09:38 +0000)]
re PR ada/35050 (renames entities not in symbol table)
gcc/ada/
PR ada/35050
* xref_lib.adb (Parse_Identifier_Info): Correctly parse and ignore the
renaming information.
From-SVN: r134256
Ben Elliston [Mon, 14 Apr 2008 09:31:06 +0000 (09:31 +0000)]
* config.sub, config.guess: Update from upstream sources.
From-SVN: r134255
Jerry DeLisle [Mon, 14 Apr 2008 00:47:13 +0000 (00:47 +0000)]
re PR fortran/35882 (Miscounted continuation lines when interspersed with data)
2008-04-13 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/35882
* gfortran.dg/continuation_3.f90: Update test.
* gfortran.dg/continuation_5.f: Update test.
* gfortran.dg/continuation_10.f90: New test.
From-SVN: r134252
Jerry DeLisle [Mon, 14 Apr 2008 00:43:32 +0000 (00:43 +0000)]
re PR fortran/35882 (Miscounted continuation lines when interspersed with data)
2008-04-13 Jerry DeLisle <jvdelisle@gcc.gnu.org>
Tobias Burnus <burnus@net-b.de>
PR fortran/35882
* options.c (gfc_init_options): Set the default maximum continuation
lines to 255 for both free and fixed form source for warnings.
(gfc_handle_option): Set -std=f95 fixed form max continuations to 19 and
the -std=f95 free form max continuations to 39 for warnings.
* scanner.c (gfc_next_char_literal): Adjust the current_line number only
if it is less than the current locus.
Co-Authored-By: Tobias Burnus <burnus@net-b.de>
From-SVN: r134251
GCC Administrator [Mon, 14 Apr 2008 00:17:43 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134249
Thomas Koenig [Sun, 13 Apr 2008 20:15:58 +0000 (20:15 +0000)]
re PR libfortran/32972 (performance of pack/unpack)
2008-04-13 Thomas Koenig <tkoenig@gcc.gnu.org>
Francois-Xavier Coudert <fxcoudert@gcc.gnu.org>
PR libfortran/32972
PR libfortran/32512
configure.ac: Add test for uintptr_t.
configure: Regenerated.
config.h.in: Regenerated.
* libgfortran.h: GFC_DTYPE_DERIVED_1: New macro.
GFC_DTYPE_DERIVED_2: New macro.
GFC_DTYPE_DERIVED_4: New macro.
GFC_DTYPE_DERIVED_8: New macro.
GFC_DTYPE_DERIVED_16: New macro.
GFC_UNALIGNED_2: New macro.
GFC_UNALIGNED_4: New macro.
GFC_UNALIGNED_8: New macro.
GFC_UNALIGNED_16: New macro.
intptr_t: Define if we don't have it.
uintptr_t: Likewise.
* runtime/backtrace.c (show_backtrace): Use intptr_t.
* intrinsics/signal.c (signal_sub): Likewise.
(signal_sub_int): Likewise.
(alarm_sub_int_i4): Likewise.
* intrinsics/spread_generic.c (spread): Use the integer
routines for handling derived types of sizes 1, 2, 4, 8 and 16
if the alignment of all pointers is correct.
(spread_scalar): Likewise.
* intrinsics/pack_generic.c (pack): Likewise.
Use GFD_DTYPE_TYPE_SIZE to avoid nested switch statements.
* intrinsics/unpack_generic.c (unpack1): Likewise.
(unpack0): Likewise.
* runtime/in_pack_generic.c (internal_pack): Likewise.
* runtime/in_unpack_generic.c (internal_unpack): Likewise.
2008-04-13 Thomas Koenig <tkoenig@gcc.gnu.org>
PR libfortran/32972
PR libfortran/32512
* gfortran.dg/internal_pack_1.f90: Add test for derived type.
* gfortran.dg/intrinsic_spread_1.f90: Likewise.
* gfortran.dg/intrinsic_pack_1.f90: Likewise.
* gfortran.dg/intrinsic_unpack_1.f90: Likewise.
Co-Authored-By: Francois-Xavier Coudert <fxcoudert@gcc.gnu.org>
From-SVN: r134245
Samuel Tardieu [Sun, 13 Apr 2008 18:15:20 +0000 (18:15 +0000)]
re PR ada/17985 (GNAT accepts extension aggregate where expexted type is not extension)
gcc/ada/
PR ada/17985
* sem_aggr.adb (Valid_Ancestor_Type): A type is not an ancestor of
itself.
gcc/testsuite/
PR ada/17985
* gnat.dg/ancestor_type.ads, gnat.dg/ancestor_type.adb: New test.
From-SVN: r134244
Ralf Wildenhues [Sun, 13 Apr 2008 18:03:09 +0000 (18:03 +0000)]
sfn_scan.adb, [...]: Fix comment typos.
gcc/ada/
* sfn_scan.adb, sfn_scan.ads, sinfo.ads,
sinput-d.ads, sinput-l.adb, sinput-l.ads, sinput.ads,
snames.ads, sprint.adb, stand.ads, stringt.ads,
styleg.adb, styleg.ads, stylesw.adb, stylesw.ads,
switch.ads, sysdep.c, table.adb, table.ads,
targparm.ads, tb-gcc.c, tbuild.ads, tracebak.c,
trans.c, tree_io.adb, treepr.adb, types.adb, types.ads,
uintp.adb, uintp.ads, utils.c, utils2.c, validsw.ads,
vms_conv.adb, vms_conv.ads, vms_data.ads, widechar.adb,
widechar.ads, xeinfo.adb, xgnatugn.adb, xr_tabls.adb,
xr_tabls.ads, xref_lib.adb, xref_lib.ads, xsinfo.adb:
Fix comment typos.
From-SVN: r134243
Ralf Wildenhues [Sun, 13 Apr 2008 17:41:15 +0000 (17:41 +0000)]
sem_ch10.adb, [...]: Fix comment typos.
gcc/ada/
* sem_ch10.adb, sem_ch10.ads,
sem_ch12.adb, sem_ch12.ads, sem_ch13.adb, sem_ch13.ads,
sem_ch3.adb, sem_ch4.adb, sem_ch5.adb, sem_ch6.adb,
sem_ch6.ads, sem_ch8.adb, sem_ch8.ads, sem_ch9.adb,
sem_elab.adb, sem_elab.ads, sem_elim.ads, sem_eval.adb,
sem_eval.ads, sem_intr.adb, sem_mech.adb, sem_mech.ads,
sem_prag.adb, sem_prag.ads, sem_res.adb, sem_res.ads,
sem_type.adb, sem_util.adb, sem_util.ads, sem_warn.adb,
sem_warn.ads: Fix comment typos.
From-SVN: r134242
Ralf Wildenhues [Sun, 13 Apr 2008 17:25:22 +0000 (17:25 +0000)]
s-secsta.adb, [...]: Fix comment typos.
gcc/ada/
* s-secsta.adb, s-sequio.ads, s-shasto.ads,
s-soflin.ads, s-stalib.ads, s-stausa.adb,
s-stausa.ads, s-strxdr.adb, s-taenca.adb, s-taenca.ads,
s-taprob.adb, s-taprop-hpux-dce.adb, s-taprop-irix.adb,
s-taprop-linux.adb, s-taprop-mingw.adb, s-taprop-posix.adb,
s-taprop-solaris.adb, s-taprop-tru64.adb, s-taprop-vms.adb,
s-taprop-vxworks.adb, s-taprop.ads, s-tarest.adb,
s-tarest.ads, s-tasini.adb, s-tasini.ads, s-taskin.ads,
s-tasque.ads, s-tassta.adb, s-tassta.ads, s-tasuti.ads,
s-tpoben.adb, s-tpoben.ads, s-tpobop.adb,
s-tpopsp-posix.adb, s-tpopsp-rtems.adb, s-tposen.adb,
s-tposen.ads, s-traceb-hpux.adb, s-traces.ads,
s-trafor-default.ads, s-unstyp.ads, s-utf_32.ads,
s-vaflop.adb, s-vaflop.ads, s-valrea.adb, s-valuti.adb,
s-wchstw.ads, s-wchwts.adb, s-wchwts.ads, scans.ads,
scn.adb, scng.adb, seh_init.c, sem.ads, sem_aggr.adb,
sem_attr.adb, sem_attr.ads, sem_case.adb, sem_case.ads,
sem_cat.adb, sem_cat.ads: Fix comment typos.
From-SVN: r134241
Jan Hubicka [Sun, 13 Apr 2008 14:05:31 +0000 (16:05 +0200)]
* m32.c (m32c_pushm_popm): Use crtl->retrun_rtx.
From-SVN: r134240
Andrew Pinski [Sun, 13 Apr 2008 03:17:01 +0000 (20:17 -0700)]
rs6000.c (compute_save_world_info): Set lr_save_p if we are going to "save the world".
2008-04-12 Andrew Pinski <pinskia@gmail.com>
* config/rs6000/rs6000.c (compute_save_world_info): Set lr_save_p if
we are going to "save the world".
2008-04-12 Andrew Pinski <pinskia@gmail.com>
* gcc.target/powerpc/darwin-save-world-1.c: New test.
From-SVN: r134237
Hans-Peter Nilsson [Sun, 13 Apr 2008 00:51:51 +0000 (00:51 +0000)]
* config/cris/cris.md ("*andhi_lowpart_non_v32", "*andhi_lowpart_v32")
("*andqi_lowpart_non_v32", "*andqi_lowpart_v32"): Use "+" for the
operand 0 constraint, not "=".
From-SVN: r134236
GCC Administrator [Sun, 13 Apr 2008 00:17:35 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134234
Joel Sherrill [Sat, 12 Apr 2008 21:55:35 +0000 (21:55 +0000)]
re PR ada/35825 (Update g-soccon-rtems.ads)
2008-04-12 Joel Sherrill <joel@oarcorp.com>
gcc/ada/
PR ada/35825
* g-soccon-rtems.ads: Add IP_PKTINFO as unsupported.
From-SVN: r134230
Arnaud Charlet [Sat, 12 Apr 2008 19:08:18 +0000 (19:08 +0000)]
re PR ada/35917 (s-linux.ads:106:41: "int" is not visible)
PR ada/35917
* s-linux-hppa.ads: Fix syntax errors.
From-SVN: r134226
Hans-Peter Nilsson [Sat, 12 Apr 2008 18:40:10 +0000 (18:40 +0000)]
Makefile.tpl: Error early unless at least GNU make 3.80.
* Makefile.tpl <gcc>: Error early unless at least GNU make 3.80.
* Makefile.in: Regenerate.
From-SVN: r134225
James E Wilson [Sat, 12 Apr 2008 07:13:44 +0000 (00:13 -0700)]
Fix bug reported by Andrew Pinski.
* system.h: Change ASSERT_CHECKING to ENABLE_ASSERT_CHECKING.
From-SVN: r134219
GCC Administrator [Sat, 12 Apr 2008 00:17:55 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134213
H.J. Lu [Fri, 11 Apr 2008 17:24:24 +0000 (17:24 +0000)]
dse.c (record_store): Use HOST_BITS_PER_WIDE_INT instead of size of positions_needed * CHAR_BIT.
2008-04-11 H.J. Lu <hongjiu.lu@intel.com>
* dse.c (record_store): Use HOST_BITS_PER_WIDE_INT instead
of size of positions_needed * CHAR_BIT.
From-SVN: r134203
H.J. Lu [Fri, 11 Apr 2008 15:52:19 +0000 (15:52 +0000)]
re PR middle-end/35897 (DSE doesn't support targets with wide registers)
2008-04-11 H.J. Lu <hongjiu.lu@intel.com>
PR middle-end/35897
* dse.c (store_info): Change positions_needed to unsigned
HOST_WIDE_INT.
(lowpart_bitmask): New.
(record_store): Cast to unsigned HOST_WIDE_INT for
positions_needed. Assert width <= size of positions_needed *
CHAR_BIT. Call lowpart_bitmask to initialize positions_needed.
(check_mem_read_rtx): Use unsigned HOST_WIDE_INT on mask. Call
lowpart_bitmask to set mask.
From-SVN: r134199
Bernd Schmidt [Fri, 11 Apr 2008 14:40:55 +0000 (14:40 +0000)]
constraints.md: New file.
* config/bfin/constraints.md: New file.
* config/bfin/bfin.md: Include it.
(adddi3): Use satisfies_constraint functions instead of the old macros.
* config/bfin/bfin.h (REG_CLASS_FROM_LETTER, CONSTRAINT_LEN,
CONST_18UBIT_IMM_P, CONST_16BIT_IMM_P, CONST_16UBIT_IMM_P,
CONST_7BIT_IMM_P, CONST_7NBIT_IMM_P, CONST_5UBIT_IMM_P,
CONST_4BIT_IMM_P, CONST_4UBIT_IMM_P, CONST_3BIT_IMM_P,
CONST_3UBIT_IMM_P, CONST_OK_FOR_K, CONST_OK_FOR_P, CONST_OK_FOR_M,
CONST_OK_FOR_CONSTRAINT_P, CONST_DOUBLE_OK_FOR_LETTER,
EXTRA_CONSTRAINT): Delete.
* config/bfin/predicates.md (highbits_operand, reg_or_7bit_operand,
reg_or_neg7bit_operand): Use satisfies_constraint functions instead
of the old macros.
* config/bfin/bfin.c: Include "tm-constrs.h".
(bfin_secondary_reload, split_load_immediate, bfin_rtx_costs):
Use satisfies_constraint functions instead of the old macros.
* doc/md.texi (Blackfin Constraints): Update file name reference.
From-SVN: r134198
Richard Guenther [Fri, 11 Apr 2008 14:14:04 +0000 (14:14 +0000)]
re PR tree-optimization/35869 (ICE in calc_dfs_tree at -O2 -gnatp after VRP optimization)
2008-04-11 Richard Guenther <rguenther@suse.de>
PR tree-optimization/35869
* tree-vrp.c (execute_vrp): Move switch statement update after
jump threading. Schedule another cfg cleanup run.
* gcc.c-torture/compile/pr35869.c: New testcase.
From-SVN: r134197
H.J. Lu [Fri, 11 Apr 2008 13:30:52 +0000 (13:30 +0000)]
sse-13.c: Move __builtin_ia32_round* to new /* mmintrin-common.h */ section.
2008-04-11 H.J. Lu <hongjiu.lu@intel.com>
* gcc.target/i386/sse-13.c: Move __builtin_ia32_round* to new
/* mmintrin-common.h */ section.
* gcc.target/i386/sse-14.c: Add tests for mmintrin-common.h.
From-SVN: r134195
Eric B. Weddington [Fri, 11 Apr 2008 12:38:49 +0000 (12:38 +0000)]
configure.ac: Do not build libssp for the AVR.
* configure.ac: Do not build libssp for the AVR.
* configure: Regenerate.
From-SVN: r134194
Volker Reichelt [Fri, 11 Apr 2008 06:55:38 +0000 (06:55 +0000)]
re PR c/35744 (ICE attributes for invalid types)
PR c/35744
* attribs.c (decl_attributes): Return early on errorneous node.
* gcc.dg/attr-error-1.c: New test.
From-SVN: r134193
Oleg Ryjkov [Fri, 11 Apr 2008 05:30:33 +0000 (05:30 +0000)]
tree.h (struct tree_base): Added a new flag default_def_flag.
2008-04-10 Oleg Ryjkov <olegr@google.com>
* tree.h (struct tree_base): Added a new flag default_def_flag.
(SSA_NAME_IS_DEFAULT_DEF): Changed to use the new flag.
From-SVN: r134192
Kaz Kojima [Fri, 11 Apr 2008 04:21:18 +0000 (04:21 +0000)]
config.gcc (need_64bit_hwint): Need 64bit hwint for sh-*-*.
* config.gcc (need_64bit_hwint): Need 64bit hwint for sh-*-*.
[libcpp]
* configure.ac: (need_64bit_hwint): Need 64bit hwint for sh-*-*
and shbe-*-*.
* configure: Rebuilt.
From-SVN: r134190
GCC Administrator [Fri, 11 Apr 2008 00:17:30 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134188
John David Anglin [Thu, 10 Apr 2008 22:50:49 +0000 (22:50 +0000)]
re PR target/35768 (gcc.c-torture/compile/
20010226-1.c:22: ICE: in do_output_reload, at reload1.c:7331)
PR target/35768
* pa.md: Define mode iterator P. Define mode attribute dwc.
(dcacheflush): Update pattern to use iterator P and attribute dwc.
(icacheflush): Likewise.
* pa.h (INITIALIZE_TRAMPOLINE): Use dcacheflushsi/icacheflushsi if
!TARGET_64BIT, and dcacheflushdi/icacheflushdi if TARGET_64BIT.
From-SVN: r134182
Ben Elliston [Thu, 10 Apr 2008 22:43:35 +0000 (22:43 +0000)]
* config/spu/spu.c (spu_init_builtins): Mark builtins as nothrow.
From-SVN: r134181
Benjamin Kosnik [Thu, 10 Apr 2008 22:14:17 +0000 (22:14 +0000)]
[multiple changes]
2008-04-10 Benjamin Kosnik <bkoz@redhat.com>
* doc/html/*: Regenerate.
2008-04-10 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
* doc/xml/manual/bitmap_allocator.xml: Improve wording a bit.
* doc/xml/authors.xml: Fix typos.
* doc/xml/faq.xml: Likewise.
* doc/xml/manual/abi.xml: Likewise.
* doc/xml/manual/allocator.xml: Likewise.
* doc/xml/manual/appendix_contributing.xml: Likewise.
* doc/xml/manual/backwards_compatibility.xml: Likewise.
* doc/xml/manual/build_hacking.xml: Likewise.
* doc/xml/manual/codecvt.xml: Likewise.
* doc/xml/manual/concurrency.xml: Likewise.
* doc/xml/manual/ctype.xml: Likewise.
* doc/xml/manual/debug_mode.xml: Likewise.
* doc/xml/manual/diagnostics.xml: Likewise.
* doc/xml/manual/evolution.xml: Likewise.
* doc/xml/manual/extensions.xml: Likewise.
* doc/xml/manual/locale.xml: Likewise.
* doc/xml/manual/messages.xml: Likewise.
* doc/xml/manual/parallel_mode.xml: Likewise.
* doc/xml/manual/status_cxx200x.xml: Likewise.
* doc/xml/manual/strings.xml: Likewise.
* doc/xml/manual/support.xml: Likewise.
* doc/xml/manual/test.xml: Likewise.
* doc/xml/manual/using.xml: Likewise.
2008-04-10 Benjamin Kosnik <bkoz@redhat.com>
Johannes Singler <singler@ira.uka.de>
* doc/xml/manual/parallel_mode.xml: Remove map/set bulk
insertors. Correct omp_set_num_threads example.
From-SVN: r134178
Ralf Wildenhues [Thu, 10 Apr 2008 21:44:46 +0000 (21:44 +0000)]
gnat_ugn.texi: Fix typos.
gcc/ada/
* gnat_ugn.texi: Fix typos.
* raise-gcc.c, repinfo.adb, repinfo.ads, restrict.adb,
restrict.ads, rtsfind.adb, rtsfind.ads, s-arit64.ads,
s-asthan-vms-alpha.adb, s-auxdec.ads, s-casuti.ads,
s-fatflt.ads, s-fatgen.adb, s-fatlfl.ads,
s-fatllf.ads, s-fatsfl.ads, s-filofl.ads,
s-finimp.adb, s-finroo.ads, s-fishfl.ads,
s-fvadfl.ads, s-fvaffl.ads, s-fvagfl.ads,
s-hibaen.ads, s-htable.ads, s-imgcha.adb,
s-imgenu.ads, s-imgint.adb, s-imgrea.adb,
s-inmaop-dummy.adb, s-inmaop.ads, s-interr-vms.adb,
s-interr-vxworks.adb, s-interr.adb, s-interr.ads,
s-intman-vxworks.ads, s-intman.ads, s-mastop-irix.adb,
s-os_lib.adb, s-os_lib.ads, s-osinte-aix.ads,
s-osinte-darwin.ads, s-osinte-freebsd.ads,
s-osinte-hpux.ads, s-osinte-lynxos-3.adb,
s-osinte-lynxos-3.ads, s-osinte-lynxos.ads,
s-osinte-rtems.ads, s-osinte-solaris-posix.ads,
s-osprim-mingw.adb, s-osprim-vms.adb, s-parame-ae653.ads,
s-parame-hpux.ads, s-parame-vms-alpha.ads,
s-parame-vms-ia64.ads, s-parame-vms-restrict.ads,
s-parame-vxworks.ads, s-parame.ads, s-parint.adb,
s-parint.ads, s-poosiz.adb, s-proinf-irix-athread.ads,
s-proinf.ads, s-regexp.adb, s-regpat.adb, s-regpat.ads,
s-rident.ads: Fix comment typos.
From-SVN: r134177
Rafael Avila de Espindola [Thu, 10 Apr 2008 20:59:50 +0000 (20:59 +0000)]
tree-vrp.c (extract_range_from_binary_expr): Don't handle TRUTH_ANDIF_EXPR or TRUTH_ORIF_EXPR.
2008-04-10 Rafael Espindola <espindola@google.com>
* gcc/tree-vrp.c (extract_range_from_binary_expr): Don't handle
TRUTH_ANDIF_EXPR or TRUTH_ORIF_EXPR.
(extract_range_from_expr): The same.
From-SVN: r134176
Benjamin Kosnik [Thu, 10 Apr 2008 20:28:38 +0000 (20:28 +0000)]
Changelog: Fix merge bit.
2008-04-10 Benjamin Kosnik <bkoz@redhat.com>
* Changelog: Fix merge bit.
From-SVN: r134175
Benjamin Kosnik [Thu, 10 Apr 2008 20:26:45 +0000 (20:26 +0000)]
[multiple changes]
2008-04-10 Benjamin Kosnik <bkoz@redhat.com>
* include/c_global/cstdatomic: C++ types only, format, doxygen markup.
* include/c_compatibility/stdatomic.h: "C" only, format, doxygen markup.
* src/atomic.c: Format, doxygen markup, edits.
* include/Makefile.am (cstdatomic, stdatomic.h): Add.
* src/Makefile.am (atomic.c): Add.
* acinclude.m4 (GLIBCXX_CHECK_STANDARD_LAYOUT): New.
* configure.ac: Call it.
* include/Makefile.in: Regenerate.
* src/Makefile.in: Same.
* Makefile.in: Same.
* doc/Makefile.in: Same.
* po/Makefile.in: Same.
* libmath/Makefile.in: Same.
* include/Makefile.in: Same.
* testsuite/Makefile.in: Same.
* libsupc++/Makefile.in: Same.
* config.h.in: Same.
* configure: Same.
* config/abi/pre/gnu.ver: Add atomics exports.
* include/ext/typelist.h (apply_generator): New.
* include/bits/c++config (_GLIBCXX_BEGIN_EXTERN_C): New.
(_GLIBCXX_END_EXTERN_C): New.
Deal with _GLIBCXX_BEGIN_NAMESPACE, _GLIBCXX_END_NAMESPACE when
compiling as C.
* testsuite/lib/libstdc++.exp (v3_target_compile_as_c): Add -L
flag for libstdc++.
* testsuite/util/testsuite_common_types.h (integral_types): New
typelist.
(atomics, atomics_tl): Same.
(assignable, default_constructible, copy_constructible,
explicit_value_constructible): New.
* testsuite/17_intro/headers/c++200x/all.cc: Test new headers.
* testsuite/29_atomics: New.
* testsuite/29_atomics/atomic_address: Same.
* testsuite/29_atomics/atomic_address/cons: Same.
* testsuite/29_atomics/atomic_address/cons/explicit_value.cc: Same.
* testsuite/29_atomics/atomic_address/cons/assign_neg.cc: Same.
* testsuite/29_atomics/atomic_address/cons/copy_neg.cc: Same.
* testsuite/29_atomics/atomic_address/cons/default.cc: Same.
* testsuite/29_atomics/atomic_address/cons/aggregate.cc: Same.
* testsuite/29_atomics/atomic_flag: Same.
* testsuite/29_atomics/atomic_flag/cons: Same.
* testsuite/29_atomics/atomic_flag/cons/assign_neg.cc: Same.
* testsuite/29_atomics/atomic_flag/cons/copy_neg.cc: Same.
* testsuite/29_atomics/atomic_flag/cons/default.cc: Same.
* testsuite/29_atomics/atomic_flag/cons/aggregate.cc: Same.
* testsuite/29_atomics/atomic_flag/requirements: Same.
* testsuite/29_atomics/atomic_flag/requirements/standard_layout.cc:
Same.
* testsuite/29_atomics/atomic_flag/
atomic_global_fence_compatibility.cc: Same.
* testsuite/29_atomics/atomic_flag/test_and_set: Same.
* testsuite/29_atomics/atomic_flag/test_and_set/explicit.c: Same.
* testsuite/29_atomics/atomic_flag/test_and_set/explicit.cc: Same.
* testsuite/29_atomics/atomic_flag/test_and_set/implicit.c: Same.
* testsuite/29_atomics/atomic_flag/test_and_set/implicit.cc: Same.
* testsuite/29_atomics/headers: Same.
* testsuite/29_atomics/headers/cstdatomic: Same.
* testsuite/29_atomics/headers/cstdatomic/types_std_c++0x.cc: Same.
* testsuite/29_atomics/headers/cstdatomic/functions_std_c++0x.cc: Same.
* testsuite/29_atomics/headers/cstdatomic/std_c++0x_neg.cc: Same.
* testsuite/29_atomics/headers/cstdatomic/macros.cc: Same.
* testsuite/29_atomics/headers/stdatomic.h: Same.
* testsuite/29_atomics/headers/stdatomic.h/macros.c: Same.
* testsuite/29_atomics/headers/stdatomic.h/types.c: Same.
* testsuite/29_atomics/headers/stdatomic.h/functions.c: Same.
* testsuite/29_atomics/atomic: Same.
* testsuite/29_atomics/atomic/cons: Same.
* testsuite/29_atomics/atomic/cons/explicit_value.cc: Same.
* testsuite/29_atomics/atomic/cons/assign_neg.cc: Same.
* testsuite/29_atomics/atomic/cons/copy_neg.cc: Same.
* testsuite/29_atomics/atomic/cons/default.cc: Same.
* testsuite/29_atomics/atomic/requirements: Same.
* testsuite/29_atomics/atomic/requirements/explicit_instantiation: Same.
* testsuite/29_atomics/atomic/requirements/explicit_instantiation/
1.cc: Same.
2008-04-10 Lawrence Crowl <crowl@google.com>
* include/c_global/cstdatomic: New.
* include/c_compatibility/stdatomic.h: New.
* src/atomic.c: New.
From-SVN: r134174
Benjamin Kosnik [Thu, 10 Apr 2008 19:02:57 +0000 (19:02 +0000)]
tuple: Consistency check for include guard, doxygen file markup.
2008-04-10 Benjamin Kosnik <bkoz@redhat.com>
* include/std/tuple: Consistency check for include guard, doxygen
file markup.
* include/std/date_time: Same.
* include/std/type_traits: Same.
* include/std/regex: Same.
* include/std/array: Same.
* include/std/system_error: Same, add error_code ctor.
From-SVN: r134171
Adam Nemet [Thu, 10 Apr 2008 18:28:45 +0000 (18:28 +0000)]
mips.md (GPR2): New mode iterator.
* config/mips/mips.md (GPR2): New mode iterator.
(seq): Add comment.
(*seq_<mode>, *seq_<mode>_mips16, *sne_<mode>, *sgt<u>_<mode>,
*sgt<u>_<mode>_mips16, *sge<u>_<mode>, *slt<u>_<mode>,
*slt<u>_<mode>_mips16 *sle<u>_<mode>, *sle<u>_<mode>_mips16):
Rewrite these to take two modes, the mode of comparison and the
mode of the destination.
* config/mips/mips.c (mips_expand_scc): Instead of having
paradoxical subreg as destination, expand "narrowing" scc if mode
of comparison is SI and target is requested in DI mode.
(mips_emit_int_order_test): Update comment. Make mode of
comparison match CMP0 rather than TARGET. When creating inverse
target use mode of TARGET.
testsuite/
* gcc.target/mips/scc-2.c: New test.
* gcc.target/mips/scc-3.c: New test.
* gcc.target/mips/scc-4.c: New test.
From-SVN: r134167
Adam Nemet [Thu, 10 Apr 2008 18:21:26 +0000 (18:21 +0000)]
gcov-dump.c (tag_summary): Only print summaries for the first GCOV_COUNTERS_SUMMABLE counters.
* gcov-dump.c (tag_summary): Only print summaries for the first
GCOV_COUNTERS_SUMMABLE counters.
From-SVN: r134166
Uros Bizjak [Thu, 10 Apr 2008 17:58:54 +0000 (19:58 +0200)]
i386.md (absneg): New code iterator.
* config/i386/i386.md (absneg): New code iterator.
(absnegprefix): New code attribute.
(<code><mode>2): Macroize expander from abs<mode>2 and neg<mode>2
patterns using absneg code iterator.
(<code>tf2): Macroize expander from abstf2 and negtf2 patterns
using absneg code iterator.
(*<code><mode>2_1): Macroize insn pattern from *abs<mode>2_1 and
*neg<mode>2 patterns using absneg code iterator.
(*<code>extendsfdf2): Macroize insn pattern from *absextendsfdf2 and
*negextendsfdf2 patterns using absneg code iterator.
(*<code>extendsfxf2): Macroize insn pattern from *absextendsfxf2 and
*negextendsfxf2 patterns using absneg code iterator.
(*<code>extendsfdf2): Macroize insn pattern from *absextendsfdf2 and
*negextendsfdf2 patterns using absneg code iterator.
* config/i386/sse.md (<code><mode>2): Macroize expander from
abs<mode>2 and neg<mode>2 patterns using absneg code iterator.
From-SVN: r134165
Andreas Krebbel [Thu, 10 Apr 2008 16:14:51 +0000 (16:14 +0000)]
s390.h: Remove the remains of the recent search & replace action of...
2008-04-10 Andreas Krebbel <krebbel1@de.ibm.com>
* config/s390/s390.h: Remove the remains of the recent search
& replace action of current_function_outgoing_args_size.
From-SVN: r134164
Ira Rosen [Thu, 10 Apr 2008 05:51:59 +0000 (05:51 +0000)]
re PR tree-optimization/35821 (Internal compiler error: segmentation fault)
PR tree-optimization/35821
* tree-vect-transform.c (vect_create_data_ref_ptr): Add check that
NEW_STMT_LIST is not NULL.
From-SVN: r134162
GCC Administrator [Thu, 10 Apr 2008 00:17:52 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r134157
Andy Hutchinson [Wed, 9 Apr 2008 23:58:39 +0000 (23:58 +0000)]
re PR testsuite/34894 (Some compile tests require trampolines even when the target sets no_trampolines)
2008-04-09 Andy Hutchinson <hutchinsonandy@aim.com>
PR testsuite/34894
PR testsuite/33782
* lib/target-supports.dg: Add check_effective_target_trampolines.
Disable profiling for avr-*-*.
* gcc.c-torture/compile/pr27889.c: dg-requires trampolines.
* gcc.c-torture/compile/nested-1.c: Ditto.
* gcc.c-torture/compile/
20050122-2.c: Ditto.
* gcc.c-torture/compile/
20010226-1.c: Ditto.
* gcc.c-torture/compile/
20010327-1.c: Skip for avr-*-*.
* gcc.c-torture/compile/980506-1.c: Ditto.
* gcc.c-torture/compile/
20020604-1.c: Ditto.
* gcc.c-torture/compile/limits-stringlit.c: Ditto
* gcc.c-torture/compile/
20001226-1.c: Ditto
From-SVN: r134153
Jason Merrill [Wed, 9 Apr 2008 16:15:53 +0000 (12:15 -0400)]
re PR c++/35708 (jump to label enters catch block)
PR c++/35708
* semantics.c (finish_compound_literal): Return a TARGET_EXPR,
not a pushed variable.
From-SVN: r134146
David Edelsohn [Wed, 9 Apr 2008 14:41:07 +0000 (14:41 +0000)]
re PR libstdc++/35597 (libstdc++ -ffunction-sections -fdata-sections disabled on AIX)
PR libstdc++/35597
* toplev.c (process_options): Remove -ffunction-sections debugging
warning.
From-SVN: r134143
Samuel Tardieu [Wed, 9 Apr 2008 14:21:18 +0000 (14:21 +0000)]
re PR ada/28305 (GNAT bug when inlining instance of a generic subprogram)
gcc/ada/
PR ada/28305
* sem_ch6.adb (Build_Body_To_Inline): Do not save and restore
environment if generic instance is a top-level one.
gcc/testsuite/
PR ada/28305
* gnat.dg/specs/fe_inlining.ads, gnat.dg/specs/fe_inlining_helper.ads,
gnat.dg/specs/fe_inlining_helper.adb: New test.
From-SVN: r134142
projects / gcc.git / log