Jason Merrill [Mon, 4 Jul 2011 21:44:11 +0000 (17:44 -0400)]
cp-demangle.c (d_expression): Handle 'this'.
* cp-demangle.c (d_expression): Handle 'this'.
(d_print_comp) [DEMANGLE_COMPONENT_FUNCTION_PARAM]: Likewise.
From-SVN: r175836
Jason Merrill [Mon, 4 Jul 2011 21:44:04 +0000 (17:44 -0400)]
DR 1207 PR c++/49589
DR 1207
PR c++/49589
* mangle.c (write_expression): Handle 'this'.
* parser.c (cp_parser_postfix_dot_deref_expression): Allow
incomplete *this.
* semantics.c (potential_constant_expression_1): Check that
DECL_CONTEXT is set on 'this'.
From-SVN: r175835
Jason Merrill [Mon, 4 Jul 2011 21:43:57 +0000 (17:43 -0400)]
error.c (dump_template_bindings): Don't print typenames for a partial instantiation.
* error.c (dump_template_bindings): Don't print typenames
for a partial instantiation.
(dump_function_decl): If we aren't printing function arguments,
print template arguments as <args> rather than [with ...].
(dump_expr): Don't print return type or template header.
[BASELINK]: Use BASELINK_FUNCTIONS rather than get_first_fn.
* pt.c (dependent_template_arg_p): Handle null arg.
From-SVN: r175834
Jason Merrill [Mon, 4 Jul 2011 21:43:49 +0000 (17:43 -0400)]
* error.c (type_to_string): Avoid redundant akas.
From-SVN: r175833
Jakub Jelinek [Mon, 4 Jul 2011 21:12:57 +0000 (23:12 +0200)]
re PR fortran/49623 (Segfault of "gfortran -L" if no argument follows the "-L")
PR fortran/49623
* gfortranspec.c (lang_specific_driver): Ignore options with
CL_ERR_MISSING_ARG errors.
From-SVN: r175830
Jakub Jelinek [Mon, 4 Jul 2011 21:04:54 +0000 (23:04 +0200)]
re PR rtl-optimization/49619 (ICE in simplify_subreg, at simplify-rtx.c:5362)
PR rtl-optimization/49619
* combine.c (combine_simplify_rtx): In PLUS -> IOR simplification
pass VOIDmode as op0_mode to recursive call, and return temp even
when different from tor, just if it is not IOR of the original
PLUS arguments.
* gcc.dg/pr49619.c: New test.
From-SVN: r175825
Jakub Jelinek [Mon, 4 Jul 2011 18:53:56 +0000 (20:53 +0200)]
re PR rtl-optimization/49472 (Compiler segfault on valid code)
PR rtl-optimization/49472
* simplify-rtx.c (simplify_unary_operation_1) <case NEG>: When
negating MULT, negate the second operand instead of first.
(simplify_binary_operation_1) <case MULT>: If one operand is
a NEG and the other is MULT, don't attempt to optimize by
negation of the MULT operand if it only moves the NEG operation
around.
* gfortran.dg/pr49472.f90: New test.
From-SVN: r175821
Jakub Jelinek [Mon, 4 Jul 2011 17:19:52 +0000 (19:19 +0200)]
re PR debug/49602 (verify_ssa failed (definition does not dominate use) with "-O2 -g")
PR debug/49602
* tree-into-ssa.c (rewrite_debug_stmt_uses): Disregard
get_current_def return value if it can't be trusted to be
the current value of the variable in the current bb.
* gcc.dg/pr49602.c: New test.
From-SVN: r175818
Uros Bizjak [Mon, 4 Jul 2011 16:26:08 +0000 (18:26 +0200)]
re PR target/49600 (Bad SSE2 int->float split in i386.md)
PR target/49600
* config/i386/i386.md (SSE2 int->float split): Push operand 1 in
general register to memory for !TARGET_INTER_UNIT_MOVES.
From-SVN: r175817
Georg-Johann Lay [Mon, 4 Jul 2011 12:48:04 +0000 (12:48 +0000)]
re PR target/34734 ([avr] attribute((progmem)) not handled properly in C++)
PR target/34734
PR target/44643
* gcc.target/avr/avr.exp: Run over cpp files, too.
* gcc.target/avr/torture/avr-torture.exp: Ditto.
* gcc.target/avr/progmem.h: New file.
* gcc.target/avr/exit-abort.h: New file.
* gcc.target/avr/progmem-error-1.c: New file.
* gcc.target/avr/progmem-error-1.cpp: New file.
* gcc.target/avr/progmem-warning-1.c: New file.
* gcc.target/avr/torture/progmem-1.c: New file.
* gcc.target/avr/torture/progmem-1.cpp: New file.
From-SVN: r175811
Georg-Johann Lay [Mon, 4 Jul 2011 12:28:02 +0000 (12:28 +0000)]
re PR target/44643 ([avr] ICE in c-typeck.c)
PR target/44643
* config/avr/avr.c (avr_insert_attributes): Use TYPE_READONLY
instead of TREE_READONLY.
From-SVN: r175809
Georg-Johann Lay [Mon, 4 Jul 2011 12:20:35 +0000 (12:20 +0000)]
extend.texi (AVR Built-in Functions): Update documentation of __builtin_avr_fmul*.
* doc/extend.texi (AVR Built-in Functions): Update documentation
of __builtin_avr_fmul*.
* config/avr/avr.c (avr_init_builtins): Don't depend on
AVR_HAVE_MUL.
* config/avr/avr-c.c (avr_cpu_cpp_builtins): Ditto.
* config/avr/avr.md (fmul): Rename to fmul_insn.
(fmuls): Rename to fmuls_insn.
(fmulsu): Rename to fmulsu_insn.
(fmul,fmuls,fmulsu): New expander.
(*fmul.call,*fmuls.call,*fmulsu.call): New Insn.
* config/avr/t-avr (LIB1ASMFUNCS): Add _fmul, _fmuls, _fmulsu.
* config/avr/libgcc.S (__fmul): New function.
(__fmuls): New function.
(__fmulsu,__fmulsu_exit): New function.
From-SVN: r175807
Richard Guenther [Mon, 4 Jul 2011 12:06:17 +0000 (12:06 +0000)]
re PR tree-optimization/49615 (internal compiler error: verify_stmts failed / LHS in noreturn call with pointer-to-never-returning-member)
2011-07-04 Richard Guenther <rguenther@suse.de>
PR tree-optimization/49615
* tree-cfgcleanup.c (split_bbs_on_noreturn_calls): Fix
basic-block index check.
* g++.dg/torture/pr49615.C: New testcase.
From-SVN: r175803
Georg-Johann Lay [Mon, 4 Jul 2011 10:02:34 +0000 (10:02 +0000)]
t-avr (intfuncs16): Add _clrsbXX2.
libgcc/
* config/avr/t-avr (intfuncs16): Add _clrsbXX2.
gcc/
* longlong.h (count_leading_zeros, count_trailing_zeros,
COUNT_LEADING_ZEROS_0): Define for target avr if W_TYPE_SIZE is 16
resp. 64.
From-SVN: r175800
Nicola Pero [Mon, 4 Jul 2011 08:02:50 +0000 (08:02 +0000)]
In gcc/objc/: 2011-07-04 Nicola Pero <nicola.pero@meta-innovation.com>
In gcc/objc/:
2011-07-04 Nicola Pero <nicola.pero@meta-innovation.com>
Refactored encoding code into objc-encoding.h and objc-encoding.c.
* objc-act.c (util_obstack, util_firstobj, encode_type_qualifiers,
encode_type, generating_instance_variables, objc_method_parm_type,
objc_encoded_type_size, encode_method_prototype,
objc_build_encode_expr, pointee_is_readonly, encode_pointer,
encode_array, encode_vector, encode_aggregate_fields,
encode_aggregate_within, encode_next_bitfield,
encode_gnu_bitfield, encode_field_decl,
objc_v2_encode_property_attr): Moved to objc-encoding.h and
objc-encoding.c. No change in the actual code.
Include objc-encoding.h.
(objc_init): Added TODO.
(objc_build_property_setter_name): Made non-static so it can be
called from objc-encoding.c.
* objc-act.h (OBJC_ENCODE_INLINE_DEFS,
OBJC_ENCODE_DONT_INLINE_DEFS): Moved to objc-encoding.h.
* objc-runtime-shared-support.h (objc_v2_encode_property_attr,
encode_method_prototype, encode_field_decl,
generating_instance_variables): Moved to objc-encoding.h.
(objc_build_property_setter_name): Declare.
* objc-encoding.c: New.
* objc-encoding.h: New.
* objc-gnu-runtime-abi-01.c: Include objc-encoding.h.
* objc-next-runtime-abi-01.c: Likewise.
* objc-next-runtime-abi-02.c: Likewise.
* objc-runtime-shared-support.c: Likewise.
* Make-lang.in (OBJC_OBJS): Added objc-encoding.o.
(objc/objc-lang.o): Reordered dependencies.
(objc/objc-runtime-shared-support.o): Reordered dependencies.
Added dependencies on objc-encoding.h and on $(GGC_H),
$(DIAGNOSTIC_CORE_H), $(FLAGS_H) and input.h.
(objc/objc-gnu-runtime-abi-01.o): Likewise.
(objc/objc-next-runtime-abi-01.o): Likewise.
(objc/objc-next-runtime-abi-02.o): Likewise.
(objc/objc-act.o): Reordered dependencies. Added dependency on
objc-encoding.h.
(objc/objc-encoding.o): New rule.
* objc-encoding.c (encode_type): Use "%<%T%>" format when printing
warning "unknown type %<%T%> found during Objective-C encoding"
instead of using gen_type_name.
In gcc/objcp/:
2011-07-04 Nicola Pero <nicola.pero@meta-innovation.com>
* Make-lang.in (OBJCXX_OBJS): Added objc-encoding.o.
(objcp/objcp-lang.o): Reordered dependencies. Depend on GGC_H.
(objcp/objcp-decl.o): Reordered dependencies.
(objcp/objc-runtime-shared-support.o): Reordered dependencies.
Updated them to be identical to the corresponding new objc/ ones,
with the addition of objcp-decl.h.
(objcp/objc-runtime-shared-support.o): Likewise.
(objcp/objc-gnu-runtime-abi-01.o): Likewise.
(objcp/objc-next-runtime-abi-01.o): Likewise.
(objcp/objc-next-runtime-abi-02.o): Likewise.
(objcp/objcp-act.o): Reordered dependencies. Added dependency on
objc-encoding.h.
(objcp/objc-encoding.o): New rule.
From-SVN: r175797
GCC Administrator [Mon, 4 Jul 2011 00:19:16 +0000 (00:19 +0000)]
Daily bump.
From-SVN: r175793
Janne Blomqvist [Sun, 3 Jul 2011 17:37:30 +0000 (20:37 +0300)]
PR 49296 Fix wrong PR number in ChangeLog
From-SVN: r175789
Ira Rosen [Sun, 3 Jul 2011 10:14:55 +0000 (10:14 +0000)]
re PR tree-optimization/49610 (Segfault with -ftree-vectorize (or -O3))
PR tree-optimization/49610
* tree-vect-loop.c (vect_is_slp_reduction): Check that DEF_STMT has
a basic block.
From-SVN: r175786
GCC Administrator [Sun, 3 Jul 2011 00:17:52 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r175785
Eric Botcazou [Sat, 2 Jul 2011 18:13:47 +0000 (18:13 +0000)]
tree.h (TYPE_ARTIFICIAL): New flag.
* tree.h (TYPE_ARTIFICIAL): New flag.
* dwarf2out.c (modified_type_die): Add a DW_AT_artificial attribute to
the DIE of the type if it is artificial.
(gen_array_type_die): Likewise.
(gen_enumeration_type_die): Likewise.
(gen_struct_or_union_type_die): Likewise.
* lto-streamer-in.c (unpack_ts_base_value_fields): Use TYPE_ARTIFICIAL.
* lto-streamer-out.c (pack_ts_base_value_fields): Likewise.
ada/
* gcc-interface/utils.c (record_builtin_type): Set TYPE_ARTIFICIAL on
the type according to the ARTIFICIAL_P parameter.
(create_type_decl): Likewise.
(create_type_stub_decl): Set TYPE_ARTIFICIAL on the type to 1.
Co-Authored-By: Nicolas Setton <setton@adacore.com>
Co-Authored-By: Olivier Hainque <hainque@adacore.com>
From-SVN: r175781
Janus Weil [Sat, 2 Jul 2011 11:08:41 +0000 (13:08 +0200)]
re PR fortran/49562 ([OOP] assigning value to type-bound function)
2011-07-02 Janus Weil <janus@gcc.gnu.org>
PR fortran/49562
* expr.c (gfc_check_vardef_context): Handle type-bound procedures.
2011-07-02 Janus Weil <janus@gcc.gnu.org>
PR fortran/49562
* gfortran.dg/typebound_proc_23.f90: New.
From-SVN: r175779
GCC Administrator [Sat, 2 Jul 2011 00:17:54 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r175775
Jonathan Wakely [Fri, 1 Jul 2011 22:24:42 +0000 (22:24 +0000)]
re PR c++/49605 (-Wdelete-non-virtual-dtor is not picky enough)
2011-07-01 Jonathan Wakely <jwakely.gcc@gmail.com>
PR c++/49605
* init.c (build_delete): Only warn for sfk_deleting_destructor.
From-SVN: r175771
Eric Botcazou [Fri, 1 Jul 2011 21:43:59 +0000 (21:43 +0000)]
Make-lang.in (gnat1): Prepend '+' to the command.
* gcc-interface/Make-lang.in (gnat1): Prepend '+' to the command.
(gnatbind): Likewise.
From-SVN: r175769
Jakub Jelinek [Fri, 1 Jul 2011 21:15:05 +0000 (23:15 +0200)]
tree-object-size.c (pass_through_call): Handle BUILT_IN_ASSUME_ALIGNED.
* tree-object-size.c (pass_through_call): Handle
BUILT_IN_ASSUME_ALIGNED.
From-SVN: r175768
Jakub Jelinek [Fri, 1 Jul 2011 21:14:14 +0000 (23:14 +0200)]
Make-lang.in (cp/decl.o): Depend on pointer-set.h.
* Make-lang.in (cp/decl.o): Depend on pointer-set.h.
(cp/class.o): Likewise.
(cp/error.o): Likewise.
(cp/name-lookup.o): Likewise.
(cp/decl2.o): Likewise. Don't depend on $(POINTER_SET_H).
From-SVN: r175767
Jason Merrill [Fri, 1 Jul 2011 20:24:38 +0000 (16:24 -0400)]
re PR c++/48261 (internal compiler error: in lookup_template_function, at cp/pt.c:6227)
PR c++/48261
* pt.c (lookup_template_function): Handle non-function.
From-SVN: r175766
Jason Merrill [Fri, 1 Jul 2011 20:24:25 +0000 (16:24 -0400)]
re PR c++/48593 (Wrong return type when applying address operator to inherited, template dependend member, using a typedef)
PR c++/48593
* pt.c (tsubst_qualified_id): Check PTRMEM_OK_P.
* tree.c (build_qualified_name): Set PTRMEM_OK_P.
* semantics.c (finish_parenthesized_expr): Clear PTRMEM_OK_P on
SCOPE_REF, too.
* cp-tree.h (PTRMEM_OK_P): Apply to SCOPE_REF, too.
(QUALIFIED_NAME_IS_TEMPLATE): Switch to lang flag 1.
From-SVN: r175765
Jason Merrill [Fri, 1 Jul 2011 20:24:08 +0000 (16:24 -0400)]
re PR c++/48883 (?: ternary operator fails in certain contexts - link error)
PR c++/48883
PR c++/49609
* pt.c (resolve_nondeduced_context): Call mark_used.
From-SVN: r175764
Joel Brobecker [Fri, 1 Jul 2011 17:51:05 +0000 (17:51 +0000)]
Darwin has case-insensitive filesystems
HFS+, the FS on Darwin, is case insensitive. So this patch adjusts
filename_cmp.c to ignore the casing when comparing filenames on Darwin.
include/ChangeLog:
* filenames.h (HAVE_CASE_INSENSITIVE_FILE_SYSTEM): Define
on Darwin, as well as on the systems that use a DOS-like
filesystem.
libiberty/ChangeLog:
* filename_cmp.c (filename_cmp, filename_ncmp): Add handling of
HAVE_CASE_INSENSITIVE_FILE_SYSTEM.
From-SVN: r175762
Jan Kratochvil [Fri, 1 Jul 2011 17:16:44 +0000 (17:16 +0000)]
re PR debug/49408 (member function template id not matching linkage name)
libiberty/
PR debug/49408
* cp-demangle.c (d_print_comp): Suppress argument list for function
references by the '&' unary operator. Keep also already processed
variant without the argument list. Suppress argument list types for
function call used in an expression.
* testsuite/demangle-expected: Fix excessive argument list types in
`test for typed function in decltype'. New testcase for no argument
list types printed. 3 new testcases for function references by the
'&' unary operator..
From-SVN: r175761
Martin Jambor [Fri, 1 Jul 2011 15:46:17 +0000 (17:46 +0200)]
tree-sra.c (tree_non_mode_aligned_mem_p): Also ignore MEM_REFs.
2011-07-01 Martin Jambor <mjambor@suse.cz>
* tree-sra.c (tree_non_mode_aligned_mem_p): Also ignore MEM_REFs.
From-SVN: r175760
Jason Merrill [Fri, 1 Jul 2011 13:36:17 +0000 (09:36 -0400)]
re PR c++/49085 (Crash with SIGSEGV during compilation.)
PR c++/49085
* semantics.c (finish_offsetof): Complain about incomplete type.
From-SVN: r175758
H.J. Lu [Fri, 1 Jul 2011 12:57:11 +0000 (12:57 +0000)]
Use proper mode for stack save area.
2011-07-01 H.J. Lu <hongjiu.lu@intel.com>
PR middle-end/48016
* explow.c (update_nonlocal_goto_save_area): Use proper mode
for stack save area.
* function.c (expand_function_start): Likewise.
From-SVN: r175756
Rainer Orth [Fri, 1 Jul 2011 11:27:00 +0000 (11:27 +0000)]
config-list.mk (LIST): Append OPT-enable-obsolete to alpha-dec-osf5.1, mips-sgi-irix6.5.
* config-list.mk (LIST): Append OPT-enable-obsolete to
alpha-dec-osf5.1, mips-sgi-irix6.5.
From-SVN: r175754
Richard Guenther [Fri, 1 Jul 2011 11:13:13 +0000 (11:13 +0000)]
re PR middle-end/49596 (FAIL: gcc.dg/torture/pr43879_1.c)
2011-07-01 Richard Guenther <rguenther@suse.de>
PR middle-end/49596
* cgraph.h (varpool_all_refs_explicit_p): Not analyzed nodes
may have unknown refs.
From-SVN: r175753
Kai Tietz [Fri, 1 Jul 2011 10:00:25 +0000 (12:00 +0200)]
tree-ssa-forwprop.c (simplify_bitwise_binary): Fix typo.
2011-07-01 Kai Tietz <ktietz@redhat.com>
* tree-ssa-forwprop.c (simplify_bitwise_binary): Fix typo.
2011-07-01 Kai Tietz <ktietz@redhat.com>
* gcc.dg/tree-ssa/bitwise-sink.c: New test.
From-SVN: r175752
Rainer Orth [Fri, 1 Jul 2011 09:25:29 +0000 (09:25 +0000)]
config.gcc: Obsolete alpha*-dec-osf5.1, mips-sgi-irix6.5.
* config.gcc: Obsolete alpha*-dec-osf5.1, mips-sgi-irix6.5.
* doc/install.texi (Specific, alpha*-dec-osf5.1): Document it.
(Specific, mips-sgi-irix6): Likewise.
From-SVN: r175751
Rainer Orth [Fri, 1 Jul 2011 08:59:20 +0000 (08:59 +0000)]
re PR libmudflap/49549 (Use of --noinhibit-exec is unportable)
libmudflap:
PR libmudflap/49549
* testsuite/lib/libmudflap.exp (load_gcc_lib): Load
target-supports.exp.
* testsuite/libmudflap.cth/cthfrags.exp: Only pass
--noinhibit-exec to GNU ld.
gcc:
PR libmudflap/49549
* doc/sourcebuild.texi (Effective-Target Keywords): Document gld.
gcc/testsuite:
PR libmudflap/49549
* lib/target-supports.exp (check_effective_target_gld): New proc.
From-SVN: r175749
Jakub Jelinek [Fri, 1 Jul 2011 08:21:09 +0000 (10:21 +0200)]
tree-pretty-print.c (dump_generic_code): Print [idx]= and [idx1 ...
* tree-pretty-print.c (dump_generic_code) <case CONSTRUCTOR>: Print
[idx]= and [idx1 ... idx2]= before initializers if needed for
array initializers.
From-SVN: r175748
Chen Liqin [Fri, 1 Jul 2011 07:51:38 +0000 (07:51 +0000)]
config.gcc (score-*-elf): Remove score7.o.
2011-07-01 Chen Liqin <liqin.gcc@gmail.com>
* config.gcc (score-*-elf): Remove score7.o.
* config/score/t-score-elf: Likewise.
* config/score/score.c: Merge score7 to score.c and remove forwarding functions.
* config/score/score7.c: Deleted.
* config/score/score7.h: Deleted.
From-SVN: r175747
Richard Guenther [Fri, 1 Jul 2011 06:56:18 +0000 (06:56 +0000)]
re PR middle-end/49603 (177.mesa in SPEC CPU 2000 failed to build)
2011-07-01 Richard Guenther <rguenther@suse.de>
PR tree-optimization/49603
* tree-vect-stmts.c (vectorizable_load): Remove unnecessary
assert.
* gcc.dg/torture/pr49603.c: New testcase.
From-SVN: r175746
Jason Merrill [Fri, 1 Jul 2011 00:54:27 +0000 (20:54 -0400)]
correct changelog
From-SVN: r175744
Jason Merrill [Fri, 1 Jul 2011 00:53:10 +0000 (20:53 -0400)]
re PR c++/49387 (t.cxx:140: error: too many initializers for ‘const __class_type_info_pseudo’)
PR c++/49387
* rtti.c (get_pseudo_ti_index): Call complete_type.
From-SVN: r175743
GCC Administrator [Fri, 1 Jul 2011 00:17:43 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r175742
Jason Merrill [Fri, 1 Jul 2011 00:03:58 +0000 (20:03 -0400)]
re PR c++/49569 (-std=gnu++0x causes segmentation fault)
PR c++/49569
* method.c (implicitly_declare_fn): Set DECL_PARM_LEVEL and
DECL_PARM_INDEX on rhs parm.
From-SVN: r175738
Jason Merrill [Fri, 1 Jul 2011 00:03:43 +0000 (20:03 -0400)]
* pt.c (iterative_hash_template_arg): Use cp_tree_operand_length.
From-SVN: r175737
Jason Merrill [Fri, 1 Jul 2011 00:03:34 +0000 (20:03 -0400)]
re PR c++/49355 (new T({""}) crashes G++ when struct T { std::string foobar };)
PR c++/49355
* tree.c (stabilize_init): Handle aggregate initialization.
From-SVN: r175736
Martin Jambor [Thu, 30 Jun 2011 21:54:29 +0000 (23:54 +0200)]
tree-sra.c (struct access): Rename total_scalarization to grp_total_scalarization
2011-06-30 Martin Jambor <mjambor@suse.cz>
* tree-sra.c (struct access): Rename total_scalarization to
grp_total_scalarization
(completely_scalarize_var): New function.
(sort_and_splice_var_accesses): Set total_scalarization in the
representative access.
(analyze_access_subtree): Propagate total scalarization accross the
tree, no holes in totally scalarized trees, simplify coverage
computation.
(analyze_all_variable_accesses): Call completely_scalarize_var instead
of completely_scalarize_record.
* testsuite/gcc.dg/tree-ssa/sra-12.c: New test.
From-SVN: r175733
Jason Merrill [Thu, 30 Jun 2011 21:10:03 +0000 (17:10 -0400)]
re PR c++/48481 (C++ overloading memory hog)
PR c++/48481
* name-lookup.c (struct arg_lookup): Add fn_set.
(add_function): Check it.
(lookup_arg_dependent_1): Initialize it.
From-SVN: r175732
Jason Merrill [Thu, 30 Jun 2011 21:09:55 +0000 (17:09 -0400)]
* g++.dg/cpp0x/named.C: Test subobject of xvalue.
From-SVN: r175731
Richard Henderson [Thu, 30 Jun 2011 21:08:54 +0000 (14:08 -0700)]
i386: Always use TARGET_DEEP_BRANCH_PREDICTION.
While it could be possible to output_set_got such that we can
individually annotate the instructions, it's simpler to simply
admit that all processors currently being manufactured do want
deep branch prediction. At which point all of the complication
simply goes away.
* config/i386/i386.h (X86_TUNE_DEEP_BRANCH_PREDICTION): Remove.
(TARGET_DEEP_BRANCH_PREDICTION): Remove.
* config/i386/i386.c: Don't include dwarf2out.h.
(initial_ix86_tune_features): Remove X86_TUNE_DEEP_BRANCH_PREDICTION.
(output_set_got): Don't test TARGET_DEEP_BRANCH_PREDICTION, delete
all code dead thereafter. Don't do dwarf2out_flush_queued_reg_saves.
(ix86_expand_prologue): Set REG_CFA_FLUSH_QUEUE on set_got insn.
(machopic_output_stub): Don't test TARGET_DEEP_BRANCH_PREDICTION.
From-SVN: r175730
Richard Henderson [Thu, 30 Jun 2011 21:04:23 +0000 (14:04 -0700)]
dwarf2out: Add REG_CFA_FLUSH_QUEUE.
* reg-notes.def (REG_CFA_FLUSH_QUEUE): New.
* dwarf2out.c (dwarf2out_frame_debug): Handle it.
* final.c (final_scan_insn): Look for it, and invoke
dwarf2out_frame_debug before the insn if found.
From-SVN: r175729
Richard Henderson [Thu, 30 Jun 2011 21:01:06 +0000 (14:01 -0700)]
dwarf2out: Handle pc_rtx as return column in REG_CFA_OFFSET too.
* dwarf2out.c (dwarf2out_frame_debug_cfa_offset): Allow PC_RTX
as a source, and interpret that as DWARF_FRAME_RETURN_COLUMN.
From-SVN: r175728
Richard Henderson [Thu, 30 Jun 2011 20:59:41 +0000 (13:59 -0700)]
dwarf2out: Convert regs_saved_in_regs to VEC.
* dwarf2out.c (struct reg_saved_in_data): Provide a typedef.
Define a vector of this type.
(regs_saved_in_regs): Use a VEC.
(num_regs_saved_in_regs): Remove.
(compare_reg_or_pc): New.
(record_reg_saved_in_reg): Split out from...
(dwarf2out_flush_queued_reg_saves): ... here.
(clobbers_queued_reg_save): Update for VEC.
(reg_saved_in): Likewise.
(dwarf2out_frame_debug_init): Likewise.
(dwarf2out_reg_save_reg): Use record_reg_saved_in_reg.
(dwarf2out_frame_debug_cfa_register): Likewise.
From-SVN: r175727
Eric Botcazou [Thu, 30 Jun 2011 19:32:23 +0000 (19:32 +0000)]
re PR tree-optimization/49572 (gcc.dg/tree-ssa/
20030709-2.c scan-tree-dump-times cddce2 ".rtmem" 0)
PR tree-optimization/49572
* tree-ssa-dom.c (initialize_hash_element) <GIMPLE_SINGLE_RHS>: Use the
type of the RHS instead of that of the LHS for the expression type.
From-SVN: r175721
Eric Botcazou [Thu, 30 Jun 2011 17:56:50 +0000 (17:56 +0000)]
df-scan.c (df_get_entry_block_def_set): Use INCOMING_REGNO macro unconditionally.
* df-scan.c (df_get_entry_block_def_set): Use INCOMING_REGNO macro
unconditionally.
From-SVN: r175718
Richard Guenther [Thu, 30 Jun 2011 14:03:25 +0000 (14:03 +0000)]
opts.c (finish_options): Do not disable IPA-PTA during ltrans.
2011-06-30 Richard Guenther <rguenther@suse.de>
* opts.c (finish_options): Do not disable IPA-PTA during ltrans.
* tree-ssa-structalias.c (create_variable_info_for): Do not
add initial constraints for non-var-decls. Properly handle
globals in other ltrans partitions.
(intra_create_variable_infos): Manually create constraints for
the fake no-alias parameter.
(ipa_pta_execute): Dump the cgraph, handle ltrans partitions properly
and assert there are no clones.
From-SVN: r175707
Richard Guenther [Thu, 30 Jun 2011 13:27:43 +0000 (13:27 +0000)]
re PR tree-optimization/46787 (Does not vectorize loop with load from scalar variable)
2011-06-30 Richard Guenther <rguenther@suse.de>
PR tree-optimization/46787
* tree-data-ref.c (dr_address_invariant_p): Remove.
(find_data_references_in_stmt): Invariant accesses are ok now.
* tree-vect-stmts.c (vectorizable_load): Handle invariant
loads.
* tree-vect-data-refs.c (vect_analyze_data_ref_access): Allow
invariant loads.
* gcc.dg/vect/vect-121.c: New testcase.
From-SVN: r175704
Martin Jambor [Thu, 30 Jun 2011 13:24:19 +0000 (15:24 +0200)]
re PR tree-optimization/49094 (ARM aligned(1) attribute is sometimes dropped)
2011-06-30 Martin Jambor <mjambor@suse.cz>
PR tree-optimization/49094
* tree-sra.c (tree_non_mode_aligned_mem_p): New function.
(build_accesses_from_assign): Use it.
* testsuite/gcc.dg/tree-ssa/pr49094.c: New test.
From-SVN: r175703
Joseph Myers [Thu, 30 Jun 2011 12:26:08 +0000 (13:26 +0100)]
* de.po: Update.
From-SVN: r175698
Jakub Jelinek [Thu, 30 Jun 2011 10:43:59 +0000 (12:43 +0200)]
tree-ssa-structalias.c (find_func_aliases_for_builtin_call): Fix handling of BUILT_IN_ASSUME_ALIGNED.
* tree-ssa-structalias.c (find_func_aliases_for_builtin_call): Fix
handling of BUILT_IN_ASSUME_ALIGNED.
From-SVN: r175695
Jakub Jelinek [Thu, 30 Jun 2011 10:26:32 +0000 (12:26 +0200)]
re PR debug/49364 ([4.7 regressions] Empty .debug_abbrev sections confuse SGI nm)
PR debug/49364
* dwarf2out.c (output_abbrev_section): Don't return early
if abbrev_die_table_in_use is 1.
(dwarf2out_finish): Instead don't call output_abbrev_section
nor emit abbrev_section_label in that case.
From-SVN: r175694
Jakub Jelinek [Thu, 30 Jun 2011 10:25:40 +0000 (12:25 +0200)]
re PR fortran/49540 (Memory-hog with large DATA stmt)
PR fortran/49540
* gfortran.h (gfc_constructor): Add repeat field.
* trans-array.c (gfc_conv_array_initializer): Handle repeat > 1.
* array.c (current_expand): Add repeat field.
(expand_constructor): Copy repeat.
* constructor.c (node_free, node_copy, gfc_constructor_get,
gfc_constructor_lookup): Handle repeat field.
(gfc_constructor_lookup_next, gfc_constructor_remove): New functions.
* data.h (gfc_assign_data_value): Add mpz_t * argument.
(gfc_assign_data_value_range): Removed.
* constructor.h (gfc_constructor_advance): Removed.
(gfc_constructor_lookup_next, gfc_constructor_remove): New prototypes.
* data.c (gfc_assign_data_value): Add REPEAT argument, handle it and
also handle overwriting a range with a single entry.
(gfc_assign_data_value_range): Removed.
* resolve.c (check_data_variable): Adjust gfc_assign_data_value
call. Use gfc_assign_data_value instead of
gfc_assign_data_value_expr.
* gfortran.dg/pr49540-1.f90: New test.
* gfortran.dg/pr49540-2.f90: New test.
From-SVN: r175693
Rainer Orth [Thu, 30 Jun 2011 10:04:52 +0000 (10:04 +0000)]
re PR ada/49511 (acats test setup fails on HP-UX using posix shell)
PR ada/49511
* ada/acats/run_acats (which): Use last field of type -p output.
From-SVN: r175690
Dominique d'Humieres [Thu, 30 Jun 2011 10:02:45 +0000 (12:02 +0200)]
re PR libobjc/36610 (objc_msg_sendv is broken for targets which pass argument via registers)
2011-06-28 Dominique d'Humieres <dominiq@lps.ens.fr>
Iain Sandoe <iains@gcc.gnu.org>
PR libobjc/36610
* objc.dg/torture/forward-1.m: Remove dg-xfail-run-if.
Only skip on 64-bit *-*-darwin8* && !objc2.
Co-Authored-By: Iain Sandoe <iains@gcc.gnu.org>
From-SVN: r175689
Nick Clifton [Thu, 30 Jun 2011 09:13:49 +0000 (09:13 +0000)]
v850.h (CPP_SPEC): Define __v850e__ when compiling for the V850E.
* config/v850/v850.h (CPP_SPEC): Define __v850e__ when compiling
for the V850E.
From-SVN: r175688
Richard Guenther [Thu, 30 Jun 2011 08:44:18 +0000 (08:44 +0000)]
re PR tree-optimization/38752 (ICE in set_uids_in_ptset, at tree-ssa-structalias.c:4817)
2011-06-30 Richard Guenther <rguenther@suse.de>
PR tree-optimization/38752
* gcc.c-torture/compile/pr38752.c: New testcase.
From-SVN: r175684
Jakub Jelinek [Thu, 30 Jun 2011 08:06:01 +0000 (10:06 +0200)]
tree-ssa-structalias.c (find_func_aliases_for_builtin_call): Fix handling of BUILT_IN_MEMPCPY_CHK and BUILT_IN_STPCPY_CHK.
* tree-ssa-structalias.c (find_func_aliases_for_builtin_call): Fix
handling of BUILT_IN_MEMPCPY_CHK and BUILT_IN_STPCPY_CHK.
From-SVN: r175682
Ira Rosen [Thu, 30 Jun 2011 06:37:41 +0000 (06:37 +0000)]
tree-vect-loop.c (vect_determine_vectorization_factor): Handle both pattern and original statements if necessary.
* tree-vect-loop.c (vect_determine_vectorization_factor): Handle
both pattern and original statements if necessary.
(vect_transform_loop): Likewise.
* tree-vect-patterns.c (vect_pattern_recog): Update documentation.
* tree-vect-stmts.c (vect_mark_relevant): Add new argument.
Mark the pattern statement only if the original statement doesn't
have its own uses.
(process_use): Call vect_mark_relevant with additional parameter.
(vect_mark_stmts_to_be_vectorized): Likewise.
(vect_get_vec_def_for_operand): Use vectorized pattern statement.
(vect_analyze_stmt): Handle both pattern and original statements
if necessary.
(vect_transform_stmt): Don't store vectorized pattern statement
in the original statement.
(vect_is_simple_use_1): Use related pattern statement only if the
original statement is irrelevant.
* tree-vect-slp.c (vect_get_and_check_slp_defs): Likewise.
From-SVN: r175681
GCC Administrator [Thu, 30 Jun 2011 00:18:02 +0000 (00:18 +0000)]
Daily bump.
From-SVN: r175678
Jason Merrill [Wed, 29 Jun 2011 22:28:15 +0000 (18:28 -0400)]
re PR c++/49216 ([C++0x] ICE on compiling new-expression with braced-init-list for arrays)
PR c++/49216
* init.c (build_new_1): Pass {} down to build_vec_init.
(build_vec_init): Handle it.
From-SVN: r175674
Jason Merrill [Wed, 29 Jun 2011 21:19:31 +0000 (17:19 -0400)]
DR 1207 PR c++/49003
DR 1207
PR c++/49003
* cp-tree.h (struct saved_scope): Add x_current_class_ptr,
x_current_class_ref.
(current_class_ptr, current_class_ref): Use them.
* decl.c (build_this_parm): Handle getting the class type.
* parser.c (cp_parser_late_return_type_opt): Set up 'this'
for use within the trailing return type.
From-SVN: r175671
Jason Merrill [Wed, 29 Jun 2011 21:19:22 +0000 (17:19 -0400)]
pt.c (tsubst_decl): In unevaluated operand, don't tsubst DECL_INITIAL unless our type use auto.
* pt.c (tsubst_decl) [VAR_DECL]: In unevaluated operand,
don't tsubst DECL_INITIAL unless our type use auto.
From-SVN: r175670
François Dumont [Wed, 29 Jun 2011 20:11:50 +0000 (22:11 +0200)]
set.h, [...]: Remove base class default constructor calls.
2011-06-29 François Dumont <francois.cppdevs@free.fr>
* include/debug/set.h, unordered_map, multiset.h, forward_list,
unordered_set, vector, deque, string, list, multimap.h: Remove base
class default constructor calls.
* include/debug/map.h: Likewise and cleanup several redefinition of
base iterator typedef.
From-SVN: r175665
Changpeng Fang [Wed, 29 Jun 2011 17:53:49 +0000 (17:53 +0000)]
Auto-vectorizer generates 128-bit AVX insns by default for bdver1.
* config/i386/i386.opt (mprefer-avx128): Redefine the flag as a Mask option.
* config/i386/i386.h (ix86_tune_indices): Add X86_TUNE_AVX128_OPTIMAL entry.
(TARGET_AVX128_OPTIMAL): New definition.
* config/i386/i386.c (initial_ix86_tune_features): Initialize
X86_TUNE_AVX128_OPTIMAL entry.
(ix86_option_override_internal): Enable the generation
of the 128-bit instructions when TARGET_AVX128_OPTIMAL is set.
(ix86_preferred_simd_mode): Use TARGET_PREFER_AVX128.
(ix86_autovectorize_vector_sizes): Use TARGET_PREFER_AVX128.
From-SVN: r175661
Jason Merrill [Wed, 29 Jun 2011 17:38:53 +0000 (13:38 -0400)]
re PR c++/49272 (FAIL: g++.old-deja/g++.jason/rfg10.C (test for errors, line 11))
PR c++/49272
* g++.old-deja/g++.jason/rfg10.C: Accept other overflow error too.
From-SVN: r175659
Jason Merrill [Wed, 29 Jun 2011 17:15:16 +0000 (13:15 -0400)]
re PR c++/49520 ([C++0x] using-declaration and operator&& confuses constexpr)
PR c++/49520
* semantics.c (constexpr_fn_retval): Handle CLEANUP_POINT_EXPR here.
(massage_constexpr_body): Not here.
From-SVN: r175658
Jason Merrill [Wed, 29 Jun 2011 17:15:06 +0000 (13:15 -0400)]
re PR c++/49554 ([C++0x] lambda capture causes "cannot call member function ... without object")
PR c++/49554
* semantics.c (lambda_proxy_type): New.
(build_capture_proxy): Use it.
* cp-tree.h (DECLTYPE_FOR_LAMBDA_PROXY): New.
* pt.c (tsubst) [DECLTYPE_TYPE]: Use them.
From-SVN: r175657
Ulrich Weigand [Wed, 29 Jun 2011 16:19:08 +0000 (16:19 +0000)]
re PR rtl-optimization/49114 ([x32] Reload failed to handle (set reg:X (plus:X (subreg:X (reg:Y) 0) (const_int))))
PR rtl-optimization/49114
* reload.c (push_reload): Define in_subreg_loc and out_subreg_loc
only if LIMIT_RELOAD_CLASS to avoid -Werror build breaks.
From-SVN: r175652
Eric Botcazou [Wed, 29 Jun 2011 16:02:56 +0000 (16:02 +0000)]
re PR tree-optimization/49539 (ICE building gnattools)
PR tree-optimization/49539
* tree-ssa-forwprop.c (can_propagate_from): Check for abnormal SSA
names by means of stmt_references_abnormal_ssa_name.
(associate_plusminus): Call can_propagate_from before propagating
from definition statements.
(ssa_forward_propagate_and_combine): Remove superfluous newline.
From-SVN: r175650
Jason Merrill [Wed, 29 Jun 2011 14:34:58 +0000 (10:34 -0400)]
re PR c++/45923 (constexpr diagnostics, more more)
PR c++/45923
* class.c (explain_non_literal_class): New.
(finalize_literal_type_property): Call it.
* cp-tree.h: Declare it.
* semantics.c (ensure_literal_type_for_constexpr_object): Call it.
(is_valid_constexpr_fn): Likewise.
(massage_constexpr_body): Split out from...
(register_constexpr_fundef): ...here.
(is_instantiation_of_constexpr): New.
(expand_or_defer_fn_1): Leave DECL_SAVED_TREE alone in that case.
(explain_invalid_constexpr_fn): New.
(cxx_eval_call_expression): Call it.
(potential_constant_expression_1): Likewise. Avoid redundant errors.
* method.c (process_subob_fn): Diagnose non-constexpr.
(walk_field_subobs): Likewise.
(synthesized_method_walk): Don't shortcut if we want diagnostics.
(explain_implicit_non_constexpr): New.
(defaulted_late_check): Use it.
* call.c (build_cxx_call): Remember location.
From-SVN: r175646
Jason Merrill [Wed, 29 Jun 2011 14:34:47 +0000 (10:34 -0400)]
method.c (maybe_explain_implicit_delete): Use pointer_set instead of htab.
* method.c (maybe_explain_implicit_delete): Use pointer_set
instead of htab.
From-SVN: r175645
Jason Merrill [Wed, 29 Jun 2011 14:34:39 +0000 (10:34 -0400)]
class.c (finalize_literal_type_property): Update conditions.
* class.c (finalize_literal_type_property): Update conditions.
* method.c (defaulted_late_check): Set TYPE_HAS_CONSTEXPR_CTOR.
From-SVN: r175644
Jason Merrill [Wed, 29 Jun 2011 14:34:22 +0000 (10:34 -0400)]
tree.c (build_vec_init_expr): Don't add TARGET_EXPR.
* tree.c (build_vec_init_expr): Don't add TARGET_EXPR.
* typeck2.c (digest_init_r): Handle VEC_INIT_EXPR.
* semantics.c (cxx_eval_vec_init_1): Correct type.
From-SVN: r175643
Richard Guenther [Wed, 29 Jun 2011 14:29:22 +0000 (14:29 +0000)]
invoke.texi: Document -scev dump modifier.
2011-06-29 Richard Guenther <rguenther@suse.de>
* doc/invoke.texi: Document -scev dump modifier.
* tree-pass.h (TDF_SCEV): New dump flag.
* tree-dump.c (dump_option_value_in): Add scev.
* tree-chrec.c: Replace all TDF_DETAILS checks with TDF_SCEV.
* tree-scalar-evolution.c: Likewise.
* gcc.dg/tree-ssa/loop-17.c: Adjust.
From-SVN: r175642
Nathan Sidwell [Wed, 29 Jun 2011 14:18:48 +0000 (14:18 +0000)]
unwind-arm.c (enum __cxa_type_match_result): New.
gcc/
* config/arm/unwind-arm.c (enum __cxa_type_match_result): New.
(cxa_type_match): Correct declaration.
(__gnu_unwind_pr_common): Reconstruct
additional indirection when __cxa_type_match returns
succeeded_with_ptr_to_base.
libstdc++/
* libsupc++/eh_arm.c (__cxa_type_match): Construct address of
thrown object here. Return succeded_with_ptr_to_base for all
pointer cases.
From-SVN: r175641
Jason Merrill [Wed, 29 Jun 2011 14:09:03 +0000 (10:09 -0400)]
init.c (build_value_init): Decide whether or not to zero-initialize based on user-providedness of default...
* init.c (build_value_init): Decide whether or not to zero-initialize
based on user-providedness of default ctor, not any ctor.
(build_value_init_noctor): Adjust assert.
From-SVN: r175640
Jason Merrill [Wed, 29 Jun 2011 14:08:55 +0000 (10:08 -0400)]
DR 990
DR 990
* call.c (convert_like_real) [ck_user]: Handle value-initialization.
(build_new_method_call_1): Likewise.
* init.c (expand_default_init): Handle direct list-initialization
of aggregates.
From-SVN: r175639
Rainer Orth [Wed, 29 Jun 2011 12:47:56 +0000 (12:47 +0000)]
cls_double_va.c: Move PR number to comment.
* testsuite/libffi.call/cls_double_va.c: Move PR number to comment.
* testsuite/libffi.call/cls_longdouble_va.c: Likewise.
From-SVN: r175632
Ulrich Weigand [Wed, 29 Jun 2011 12:15:08 +0000 (12:15 +0000)]
re PR rtl-optimization/49114 ([x32] Reload failed to handle (set reg:X (plus:X (subreg:X (reg:Y) 0) (const_int))))
PR rtl-optimization/49114
* reload.c (struct replacement): Remove SUBREG_LOC member.
(push_reload): Do not set it.
(push_replacement): Likewise.
(subst_reload): Remove dead code.
(copy_replacements): Remove assertion.
(copy_replacements_1): Do not handle SUBREG_LOC.
(move_replacements): Likewise.
(find_replacement): Remove dead code. Use reload_adjust_reg_for_mode.
Detect subregs via recursive descent instead of via SUBREG_LOC.
From-SVN: r175631
Georg-Johann Lay [Wed, 29 Jun 2011 11:20:06 +0000 (11:20 +0000)]
avr.c (avr_encode_section_info): Dispatch to default_encode_section_info.
* config/avr/avr.c (avr_encode_section_info): Dispatch to
default_encode_section_info.
From-SVN: r175630
Georg-Johann Lay [Wed, 29 Jun 2011 11:17:50 +0000 (11:17 +0000)]
avr.c (avr_encode_section_info): Dispatch to default_encode_section_info.
* config/avr/avr.c (avr_encode_section_info): Dispatch to
default_encode_section_info.
From-SVN: r175629
Richard Sandiford [Wed, 29 Jun 2011 10:05:43 +0000 (10:05 +0000)]
* MAINTAINERS (RTL optimizers): Add self.
From-SVN: r175628
Richard Sandiford [Wed, 29 Jun 2011 09:42:42 +0000 (09:42 +0000)]
re PR middle-end/49545 (New C++ test failures)
gcc/
PR tree-optimization/49545
* builtins.c (get_object_alignment_1): Update function comment.
Do not use DECL_ALIGN for functions, but test
TARGET_PTRMEMFUNC_VBIT_LOCATION instead.
* fold-const.c (get_pointer_modulus_and_residue): Don't check
for functions here.
* tree-ssa-ccp.c (get_value_from_alignment): Likewise.
gcc/testsuite/
* gcc.dg/torture/pr49169.c: Restrict to ARM and MIPS targets.
From-SVN: r175627
Rainer Orth [Wed, 29 Jun 2011 08:32:02 +0000 (08:32 +0000)]
inline_1.c: Remove dg-excess-errors.
* gcc.dg/inline_1.c: Remove dg-excess-errors.
* gcc.dg/inline_2.c: Likewise.
* gcc.dg/unroll_1.c: Likewise.
From-SVN: r175625
Rainer Orth [Wed, 29 Jun 2011 08:24:13 +0000 (08:24 +0000)]
* gcc-interface/Makefile.in (TOOLS_LIBS): Add $(LIBINTL).
From-SVN: r175624
Rainer Orth [Wed, 29 Jun 2011 08:19:08 +0000 (08:19 +0000)]
re PR libgcj/49451 (FileHandleGcTest FAILS on IRIX)
PR libgcj/49451
* include/private/gcconfig.h [IRIX5 && _MIPS_SZPTR == 64]
(ELF_CLASS): Define.
* dyn_load.c [IRIX5]: Include <elf.h>.
Include <obj.h> for O32, <objlist.h> for N32, N64.
[SUNOS5DL && USE_PROC_FOR_LIBRARIES] (obj_offset): Define.
[SUNOS4 && USE_PROC_FOR_LIBRARIES] (obj_offset): Define.
[IRIX5 && !USE_PROC_FOR_LIBRARIES] (struct link_map): Define.
(__rld_obj_head): Declare.
(l_next, l_addr, obj_offset): Define.
(GC_FirstDLOpenedLinkMap): New function.
[SUNOS4 || SUNOS5DL] (GC_register_dynamic_libraries): Also use on
IRIX5.
Use obj_offset.
[IRIX5 || (USE_PROC_FOR_LIBRARIES && !LINUX]
(GC_register_dynamic_libraries): Don't use /proc version on IRIX5.
From-SVN: r175623
Jakub Jelinek [Wed, 29 Jun 2011 08:15:00 +0000 (10:15 +0200)]
re PR debug/49567 (ICE in mem_loc_descriptor due to typed DWARF stack changes)
PR debug/49567
* dwarf2out.c (mem_loc_descriptor) <case ZERO_EXTEND>: Give up
for non-MODE_INT modes instead of asserting the mode has MODE_INT
class.
* gcc.target/i386/pr49567.c: New test.
From-SVN: r175622
Georg-Johann Lay [Wed, 29 Jun 2011 07:57:25 +0000 (07:57 +0000)]
re PR target/34734 ([avr] attribute((progmem)) not handled properly in C++)
PR target/34734
* config/avr/avr.c (avr_handle_progmem_attribute): Move warning
about uninitialized data attributed 'progmem' from here...
(avr_encode_section_info): ...to this new function.
(TARGET_ENCODE_SECTION_INFO): New define.
(avr_section_type_flags): For data in ".progmem.data", remove
section flag SECTION_WRITE.
From-SVN: r175621