contrib/
2003-12-18 Kelley Cook <kcook@gcc.gnu.org>
* gcc_update (files_generated): Add in gcc/ada/stamp-xgnatug and update
gcc/ada/gnat_ug_* to use it.
gcc/ada/
2003-12-18 Kelley Cook <kcook@gcc.gnu.org>
* stamp-xgnatug: New stamp file.
* Make-lang.in (stamp-xgnatug): New stamp file and comment.
(ada/doctools/xgnatug): Add $(build_exeext).
(ada/gnat_ug_unx.texi, ada/gnat_ug_vwx.texi, ada/gnat_ug_vms.texi
ada/gnat_ug_wnt.texi): Update to depend on stamp-xgnatug.
From-SVN: r74808
2003-12-18 Andrew Pinski <pinskia@physics.uc.edu>
* g++.dg/abi/mangle18-1.C: Modify regexp to test for `[: \t\n]' at end
of label name and allow for USER_LABEL_PREFIX == "_" names.
* g++.dg/abi/mangle18-2.C: Likewise.
* g++.dg/abi/mangle19-1.C: Likewise.
* g++.dg/abi/mangle19-2.C: Likewise.
* g++.dg/abi/mangle20-1.C: Likewise.
* g++.dg/abi/mangle20-2.C: Likewise.
From-SVN: r74805
2003-12-18 Michael Koch <konqueror@gmx.de>
* mauve-libgcj: Removed the disabling of java.text.ACIAttribute and
java.text.CollationElementIterator tests as they compile again.
* testsuite/libjava.mauve/xfails: Added failing
java.text.CollationElementIterator tests.
From-SVN: r74796
* config/s390/s390.md (UNSPEC_SRST): New constant.
("strlendi", "strlensi"): New expanders.
("*strlendi", "*strlensi"): New insns.
Co-Authored-By: Mark Dettinger <dettinge@de.ibm.com>
From-SVN: r74794
* config/sol2.h (LINK_ARCH32_SPEC): Define in terms of ...
(LINK_ARCH32_SPEC_BASE): ... this new macro.
* config/sparc/sol2-bi.h (LINK_ARCH64_SPEC): Define in terms of
...
(LINK_ARCH64_SPEC_BASE): ... this new macro.
* config/sparc/sol2-gld-bi.h (LINK_ARCH32_SPEC): New macro.
(LINK_ARCH64_SPEC): Likewise.
From-SVN: r74793
PR c++/12253
* init.c (build_vec_init): Initialization of an element from
an initializer list is also a full-expression.
* parser.c, pt.c, semantics.c: Rename constant_expression_p
to integral_constant_expression_p.
From-SVN: r74790
PR other/12009
* g++.dg/compat/compat.exp: Do not set LD_LIBRARY_PATH when
testing a cross compiler, it causes spurious compile failures.
* lib/g++.exp: Likewise.
Co-Authored-By: Dan Kegel <dank@kegel.com>
From-SVN: r74789
* include/bits/demangle.h: Fix allocator type correctness,
i.e. make sure that when we instantiate a container with a value
type and an allocator, the allocator's value type matches the
container's.
* include/bits/stl_deque.h (_Deque_alloc_base): Eliminate.
(_Deque_base): inherit directly from the deque's allocator. Use
rebinding instead of _Alloc_traits. Pick up data members from
_Deque_alloc_base.
* include/bits/stl_list.h (_List_alloc_base): Eliminate.
(_List_base): Inherit directly from the list's allocator. Use
rebinding instead of _Alloc_traits. Pick up data members from
_List_alloc_base.
* include/bits/stl_vector.h (_Vector_alloc_base): Eliminate
(_Vector_base): Inherit directly from the vector's allocator. Use
rebinding instead of _Alloc_traits. Pick up data members from
_Vector_alloc_base.
* include/ext/hashtable.h: Fix allocator type correctness (the
vector of buckets must be passed an allocator for objects of
type _Node*). Use rebinding instead of _Alloc_traits.
From-SVN: r74787
PR c++/13262
* pt.c (instantiate_decl): Wrap push_nested_class and
pop_nested_class around cp_finish_decl call for static member
variable.
* g++.dg/template/access13.C: New test.
From-SVN: r74780
PR c++/9154
* parser.c (cp_parser_template_argument): A type-id followed by '>>'
is just an user typo, and should be accepted as last resort if any
other parsing fails.
(cp_parser_enclosed_template_argument_list): If the argument list is
parsed correctly, but the next token is '>>', emit a diagnostic.
(cp_parser_next_token_ends_template_argument): Accept '>>' as
delimiter of template argument, it will be later detected as a typo.
From-SVN: r74776
* config/frv/frv.c (frv_ifcvt_modify_insn): Don't leave alone
scratch insns of the then branch that clobber regs needed by the
else branch.
From-SVN: r74771
2003-12-17 Paolo Carlini <pcarlini@suse.de>
* include/bits/locale_facets.tcc (time_get::_M_extract_via_format):
Use narrow to get from __format[__i] a char __c representing a
formatting code, don't use it to compare *__beg to __format[__i].
* testsuite/22_locale/time_get/get_date/wchar_t/4.cc: New.
From-SVN: r74755
PR 12480
* c-pch.c (pch_init): Improve error message when precompiled
header can't be written.
PR 12606
* c-pch.c (pch_init): Make a PCH file appear invalid while it's
being written.
(c_common_write_pch): Make it valid once it's done.
From-SVN: r74754
PR optimization/10592
* caller-save.c (mark_referenced_regs): Don't short-circuit a reg
or subreg in SET_DEST if it isn't a hard register.
From-SVN: r74747
2003-12-16 Fernando Nasser <fnasser@redhat.com>
* gnu/java/awt/peer/gtk/GtkContainerPeer.java (handleEvent): Check for
null Graphics pointer returned by FileDialogPeer.
From-SVN: r74744
PR c++/10603
PR c++/12827
* parser.c (cp_parser_error): Help c_parse_error print good
messages if the next token is a keyword.
(cp_parser_parameter_declaration_list): When resynchronizing after
a bad parameter declaration, stop if a comma is found.
(cp_parser_parameter_declaration): Avoid backtracking.
PR c++/10603
* g++.dg/parse/error6.C: New test.
PR c++/12827
* g++.dg/parse/error7.C: New test.
From-SVN: r74743
2003-12-17 Ed Falis <falis@gnat.com>
* a-elchha.adb (Tailored_Exception_Information): made Info constant to
eliminate warning.
* a-exextr.adb: Add context clause for
Ada.Exceptions.Last_Chance_Handler.
2003-12-17 Sergey Rybin <rybin@act-europe.fr>
* cstand.adb (Create_Standard): Change the way how the declaration of
the Duration type is created (making it the same way as it is for all
the other standard types).
2003-12-17 Robert Dewar <dewar@gnat.com>
* s-crtl.ads: Fix header format
Change Pure to Preelaborate
2003-12-17 Ed Schonberg <schonberg@gnat.com>
* checks.adb (Selected_Length_Checks): Generate an Itype reference for
the expression type only if it is declared in the current unit.
* sem_ch3.adb (Constrain_Index): Handle properly a range whose bounds
are universal and already analyzed, as can occur in constrained
subcomponents that depend on discriminants, when one constraint is a
subtype mark.
* sem_res.adb (Resolve_Type_Conversion): Any arithmetic expression of
type Any_Fixed is legal as the argument of a conversion, if only one
fixed-point type is in context.
2003-12-17 GNAT Script <nobody@gnat.com>
* Make-lang.in: Makefile automatically updated
From-SVN: r74741
