tree.h (BLOCK_IN_COLD_SECTION_P): Define.

* tree.h (BLOCK_IN_COLD_SECTION_P): Define.
* final.c (final_scan_insn): Set BLOCK_IN_COLD_SECTION_P.
* dwarf2out.c (rnglist_idx): New variable.
(struct dw_ranges): Add label, idx and maybe_new_sec fields.
(DEBUG_RNGLISTS_SECTION): Define.
(ranges_base_label): New variable.
(size_of_die) <case dw_val_class_range_list>: If using
DW_FORM_rnglistx, count size of uleb128 of range list index.
(value_format) <case dw_val_class_range_list>: For
-gdwarf-5 -gsplit-dwarf return DW_FORM_rnglistx.
(output_range_list_offset): Handle -gdwarf-5 .debug_rnglists
offsets.  Multiply dwarf < 5 offsets by 2 * DWARF_ADDR_SIZE.
(add_ranges_num): Remove useless prototype.  Don't multiply
by 2 * DWARF2_ADDR_SIZE.  Add maybe_new_sec argument, adjust
for new fields added to dw_ranges struct.
(add_ranges): Add maybe_new_sec argument and pass it
through to add_ranges_num.
(note_rnglist_head): New function.
(add_ranges_by_labels): Pass true as maybe_new_sec to
add_ranges_num, call note_rnglist_head on the head of the list.
(output_ranges): Add function comment.  Switch to
.debug_ranges section here and emit .Ldebug_ranges0 label.
(index_rnglists, output_rnglists): New functions.
(gen_subprogram_die): Formatting fixes.
(add_high_low_attributes): Don't divide offsets
by 2 * DWARF2_ADDR_SIZE.  Call note_rnglist_head on the
first list element or when pointing into the middle of
a list.  Pass true as second argument to add_ranges on the
first block fragment after cold/hot section switch.
(init_sections_and_labels): For -gdwarf-5 use .debug_rnglists
section instead of .debug_ranges.  Initialize
ranges_base_label if -gdwarf-5 -gsplit-dwarf.
(dwarf2out_finish): For -gdwarf-5 -gsplit-dwarf call
index_rnglists and add DW_AT_rnglists_base attr.  Don't switch
to dwarf_ranges_section here or emit .Ldebug_ranges0 label.
Call output_rnglists for -gdwarf-5.
(dwarf2out_c_finalize): Clear rnglist_idx.

From-SVN: r241747

hashtab.h: Use standard GPLv3 with runtime exception boilerplate.

* hashtab.h: Use standard GPLv3 with runtime exception
boilerplate.

From-SVN: r241746

re PR fortran/69544 (Internal compiler error with -Wall and where)

2016-11-01  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/69544
* match.c (gfc_match_where):  Fill in locus for assigment
in simple WHERE statement.

2016-11-01  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/69544
* gfortran.dg/where_5.f90:  New test.

From-SVN: r241745

Mark fallthrough cases with gcc_fallthrough.

* gcc/combine.c (simplify_compare_const): Add gcc_fallthrough.

From-SVN: r241744

runtime: recreate function called by cgo -gccgo

    When using cgo -gccgo calls to C.GoString, C.GoStringN, and C.GoBytes
    are turned into calls to __go_byte_array_to_string and
    __go_string_to_byte_array.  Those functions were removed when the string
    code was copied from Go 1.7, but we still need them for cgo.  While cgo
    should be updated, old versions will exist for some time.

    Reviewed-on: https://go-review.googlesource.com/32474

From-SVN: r241743

libgo: only rebuild package if dependent .gox has changed

    Use mvifdiff and stamp files to track whether a .gox file has changed.
    When package A depends on package B, and we rebuild package B, only
    rebuild package A if package B's .gox file changes.  This is safe
    because when package A imports package B it only reads package B's .gox
    file.  This means that changes that do not affect export information
    will not cause dependent packages to be recompiled.

    Reviewed-on: https://go-review.googlesource.com/32476

From-SVN: r241742

libgo: minor Makefile improvements

    Suggested by Ralph Corderoy.

    Reviewed-on: https://go-review.googlesource.com/32414

From-SVN: r241741

re PR go/78145 (Several go.test tests fail with error: integer constant overflow on 32bit targets)

PR go/78145
    compiler: don't put print/println constants into temporaries

    It's not necessary, and it breaks setting their type to int64/uint64
    when appropriate.

    This fixes GCC PR 78145.

    Reviewed-on: https://go-review.googlesource.com/32475

From-SVN: r241740

* class.c (declared_access): Split out from handle_using_decl.

From-SVN: r241739

* es.po: Update.

From-SVN: r241737

arm-c.c (arm_cpu_builtins): New macro definition.

gcc/

2016-09-30  Bilyan Borisov  <bilyan.borisov@arm.com>
    Tamar Christina <tamar.christina@arm.com>

* config/arm/arm-c.c (arm_cpu_builtins): New macro definition.
* config/arm/arm_neon.h (vmaxnm_f32): New intrinsinc.
(vmaxnmq_f32): Likewise.
(vminnm_f32): Likewise.
(vminnmq_f32): Likewise.
* config/arm/arm_neon_builtins.def (vmaxnm): New builtin.
(vminnm): Likewise.
* config/arm/neon.md (neon_<fmaxmin_op><mode>, VCVTF): New
expander.

gcc/testsuite/

2016-09-30  Bilyan Borisov  <bilyan.borisov@arm.com>

* gcc.target/arm/simd/vmaxnm_f32_1.c: New.
* gcc.target/arm/simd/vmaxnmq_f32_1.c: Likewise.
* gcc.target/arm/simd/vminnm_f32_1.c: Likewise.
* gcc.target/arm/simd/vminnmq_f32_1.c: Likewise.

Co-Authored-By: Tamar Christina <tamar.christina@arm.com>
From-SVN: r241736

Minor tweaks

From-SVN: r241734

target-supports.exp: Normalize order of i?86 and x86_64 targets.

* lib/target-supports.exp: Normalize order of i?86 and x86_64 targets.
Whitespace fixes.
(check_effective_target_vect_cmdline_needed): Check
is-effective-target ia32 for x86 targets.
(check_effective_target_vect_simd_clones): Simplify condition.
(check_effective_target_vect_double): Ditto.
(check_effective_target_vect_aligned_arrays): Check
is-effective-target ia32 for x86 targets.  Simplify condition.
(check_effective_target_vect_multiple_sizes): Simplify condition.
(check_effective_target_sqrt_insn): Add i?86-*-* target.
(check_effective_target_sync_int_128): Simplify condition.
(check_effective_target_sync_int_128_runtime): Ditto.
(check_effective_target_sync_long_long_runtime): Ditto.
(check_effective_target_divmod): Add i?86-*-* target.

From-SVN: r241733

vsx.md (VSX_EXTRACT_FL): New iterator for all binary floating point types supported by the hardware...

[gcc]
2016-10-31  Michael Meissner  <meissner@linux.vnet.ibm.com>

* config/rs6000/vsx.md (VSX_EXTRACT_FL): New iterator for all
binary floating point types supported by the hardware except for
double.
(vsx_xvcvsxwdp_df): Provide scalar result alternative to the
vector instruction for optimizing extracting a SImode from a
V4SImode vector and converting it to floating point.
(vsx_xvcvuxwdp_df): Likewise.
(vsx_extract_si): On ISA 3.0, allow extract target and temporary
registers to be any VSX register.  Move stores to the end of the
constraints.
(vsx_extract_si_<uns>float_df): New combiner pattern and splitter
to optimize extracting a SImode from a V4SImode vector and
converting it to a binary floating point type supported by the
hardware.  Use the vector converts instead of extracting the
element, sign extending it, and then converting it to double.
Other floating point types  than double first convert to double,
then the double is converted to that type.
(vsx_extract_si_<uns>float_<mode>): Likewise.

[gcc/testsuite]
2016-10-31  Michael Meissner  <meissner@linux.vnet.ibm.com>

* gcc.target/powerpc/vsx-extract-4.c: New test.
* gcc.target/powerpc/vsx-extract-5.c: Likewise.

From-SVN: r241731

Daily bump.

From-SVN: r241730

driver-aarch64.c (host_detect_local_cpu): Rewrite handling of part num to handle the case where multiple implementers...

2016-10-31  Andrew Pinski  <apinski@cavium.com>

        * config/aarch64/driver-aarch64.c (host_detect_local_cpu):
        Rewrite handling of part num to handle the case where
        multiple implementers share the same part num.

From-SVN: r241726

dwarf2out.c (DWARF_COMPILE_UNIT_HEADER_SIZE): Adjust for -gdwarf-5.

* dwarf2out.c (DWARF_COMPILE_UNIT_HEADER_SIZE): Adjust for -gdwarf-5.
(DWARF_COMDAT_TYPE_UNIT_HEADER_SIZE): Likewise.

Co-Authored-By: Jakub Jelinek <jakub@redhat.com>
From-SVN: r241725

dwarf2out.c (dwarf_AT): Handle DW_AT_dwo_name.

* dwarf2out.c (dwarf_AT): Handle DW_AT_dwo_name.
(use_debug_types): Adjust comment for DWARF5 DW_UT_type units.
(new_die): Handle DW_TAG_skeleton_unit like DW_TAG_compile_unit.
(is_cu_die, is_unit_die): Likewise.
(should_move_die_to_comdat, break_out_comdat_types): Adjust
comments for DWARF5 DW_UT_type units.
(output_compilation_unit_header): Add UT argument, output
start of DWARF5 .debug_info section header.
(output_comp_unit): Add dwo_id argument.  Adjust
output_compilation_unit_header caller, for DW_UT_split_compile
emit dwo_id field, otherwise padding1.  Emit padding2 field.
(add_top_level_skeleton_die_attrs): Add DW_AT_dwo_name
rather than DW_AT_GNU_dwo_name attr for -gdwarf-5.
(output_skeleton_debug_sections): Add dwo_id argument, for
-gdwarf-5 emit DWARF 5 DW_UT_skeleton header.
(output_comdat_type_unit): For -gdwarf-5 emit .debug_info
DW_UT_type or DW_UT_split_type units rather than .debug_types.
(dwarf2out_finish): Use DW_TAG_skeleton_unit rather than
DW_TAG_compile_unit for skeleton unit die.  Don't add
DW_AT_GNU_dwo_id attributes for -gdwarf-5, instead pass checksum
address to output_comp_unit and output_skeleton_debug_sections.

From-SVN: r241724

dwarf2out.c (debug_line_str_section): New variable.

* dwarf2out.c (debug_line_str_section): New variable.
(debug_line_str_hash): Likewise.
(DEBUG_LINE_STR_SECTION): Define.
(set_indirect_string): Handle DW_FORM_line_strp like
DW_FORM_strp.
(find_string_form): Fix up formatting.
(size_of_die): Handle DW_FORM_line_strp like DW_FORM_strp.
Fix up indentation.
(output_die): Handle DW_FORM_line_strp.
(DWARF5_USE_DEBUG_LINE_STR): Define.
(output_line_string): New function.
(output_file_names): Add -gdwarf-5 support.
(output_line_info): Likewise.
(init_sections_and_labels): Initialize debug_line_str_section.
(output_indirect_string): Change 2nd argument from void *
to enum dwarf_form form, compare with form rather than
DW_FORM_strp.
(output_indirect_strings): Pass DW_FORM_strp to
output_indirect_string traversion.
(dwarf2out_finish): Output .debug_line_str strings.
(dwarf2out_c_finalize): Clear debug_line_str_section and
debug_line_str_hash.

From-SVN: r241723

re PR debug/77315 (emit DW_OP_form_tls_address)

PR debug/77315:
* dwarf2out.c (mem_loc_descriptor): Use DW_OP_form_tls_address.
(resolve_args_picking_1): Move DW_OP_form_tls_address case next to
DW_OP_GNU_push_tls_address case.
(loc_list_from_tree_1): Use DW_OP_form_tls_address.

From-SVN: r241721

re PR fortran/54679 (Erroneous "Expected P edit descriptor" in conjunction with L descriptor)

2016-10-31  Jerry DeLisle  <jvdelisle@gcc.gnu.org>

PR fortran/54679
* io.c (check_format): Adjust checks for FMT_L to treat a zero
width as an extension, giving warnings or error as appropriate.
Improve messages.
PR libgfortran/54679
* io/format.c (parse_format_list): Adjust checks for FMT_L to
treat a zero width as an extension, giving warnings or error
as appropriate. Improve messages.
PR fortran/54679
* gfortran.dg/fmt_l.f90: Update test.
* gfortran.dg/fmt_l0.f90: New test.

From-SVN: r241720

dwarf2out.h (struct dw_loc_descr_node): Adjust comment for frame_offset_rel bit.

* dwarf2out.h (struct dw_loc_descr_node): Adjust comment
for frame_offset_rel bit.
(struct array_descr_info): Add rank field.
* dwarf2out.c (struct loc_descr_context): Add placeholder_arg
and placeholder_seen fields.
(resolve_args_picking_1): Handle also frame_offset_rel DW_OP_dup
and DW_OP_over.  Optimize DW_OP_pick 0 into DW_OP_dup and
DW_OP_pick 1 into DW_OP_over.
(function_to_dwarf_procedure, type_byte_size, field_byte_offset,
gen_variant_part): Clear placeholder_{arg,seen}.
(loc_list_from_tree_1): Drop const from context argument.
Handle integral PLACEHOLDER_EXPR if context->placeholder_arg.
(loc_list_for_address_of_addr_expr_of_indirect_ref,
loc_list_from_tree, loc_descriptor_from_tree): Drop const from
context argument.
(add_scalar_info): Drop const from context argument.  Handle
context->placeholder_arg.
(add_bound_info): Drop const from context argument.
(gen_descr_array_type_die): Drop const from ctx variable.
Initialize placeholder_arg and placeholder_seen.  Add DW_AT_rank
attribute and use a single DW_TAG_generic_subrange instead of
7 DW_TAG_subrange_type for assumed rank arrays.
fortran/
* trans-types.c (gfc_get_array_descr_info): For -gdwarf-5 or
-gno-strict-dwarf, handle assumed rank arrays the way dwarf2out
expects.
ada/
* gcc-interface/misc.c (gnat_get_array_descr_info): Clear rank
field.

From-SVN: r241719

dwarf2out.h (enum dw_val_class): Add dw_val_class_loclistsptr.

* dwarf2out.h (enum dw_val_class): Add dw_val_class_loclistsptr.
* dwarf2out.c (struct dw_loc_list_struct): Change emitted field
from bool to 1-bit uchar bitfield.  Add num_assigned and
offset_emitted bitfields.
(dw_val_equal_p): Compare v.val_lbl_id rather than v.val_unsigned
for dw_val_class_lineptr and dw_val_class_macptr.  Handle
dw_val_class_loclistsptr.
(new_addr_loc_descr): Fix up formatting.
(DEBUG_LOCLISTS_SECTION, DEBUG_DWO_LOCLISTS_SECTION): Define.
(add_AT_low_high_pc): Fix up formatting.
(add_AT_loclistsptr): New function.
(AT_lbl): Allow dw_val_class_loclistsptr.
(print_dw_val, attr_checksum, attr_checksum_ordered, same_dw_val_p):
Handle dw_val_class_loclistsptr.
(loc_list_idx): New variable.
(output_loclists_offsets, assign_location_list_indexes): New
functions.
(size_of_die): For dw_val_class_loc_list -gsplit-dwarf -gdwarf-5
add size_of_uleb128 of the index.  Drop never used
dwarf_split_debug_info AT_index handling.  Handle
dw_val_class_loclistsptr.
(value_format): Return DW_FORM_loclistsx for dw_val_class_loc_list
if -gsplit-dwarf -gdwarf-5.  Handle dw_val_class_loclistsptr.
(output_loc_list): Handle DWARF 5 .debug_loclists* format.
(output_loc_list_offset): Handle -gsplit-dwarf -gdwarf-5
DW_FORM_loclistx indexes.
(output_attr_index_or_value): Fix up formatting.  Don't handle
dw_val_class_loc_list here.
(output_die): Formatting fixes.  Handle dw_val_class_loclistsptr.
For dw_val_class_loc_list call output_loc_list_offset rather than
output_attr_index_or_value.
(init_sections_and_labels): For -gdwarf-5 use .debug_loclists
or .debug_loclists.dwo section name for debug_loc_section.
(resolve_addr_in_expr): Formatting fix.
(index_location_lists): Likewise.
(dwarf2out_finish): If there are any location lists, for
-gsplit-dwarf -gdwarf-5 add DW_AT_loclists_base attribute.  Call
index_location_lists only if have_location_lists.  Call
assign_location_list_indexes for -gsplit-dwarf -gdwarf-5.  Emit
.debug_loclists{,.dwo} section header for -gdwarf-5, for -gdwarf-5
-gsplit-dwarf also emit offset table.

From-SVN: r241718

dwarf2out.c (DWARF_LARGEST_DATA_FORM_BITS): Define.

* dwarf2out.c (DWARF_LARGEST_DATA_FORM_BITS): Define.
(size_of_die, value_format, output_die): Use
DW_FORM_data16 for 128-bit dw_val_class_const_double or
dw_val_class_wide_int.

From-SVN: r241714

dwarf2out.c (dwarf_op): Renamed to ...

* dwarf2out.c (dwarf_op): Renamed to ...
(dwarf_OP): ... this.
(convert_descriptor_to_mode, scompare_loc_descriptor,
minmax_loc_descriptor, typed_binop, mem_loc_descriptor,
implicit_ptr_descriptor, optimize_one_addr_into_implicit_ptr): Adjust
callers.
(dwarf_AT, dwarf_TAG): New functions.
(check_die): Disallow DW_AT_call_all_calls next to
DW_AT_GNU_all_call_sites.
(gen_call_site_die): Use dwarf_TAG and dwarf_AT with DWARF 5 tag
and attributes instead of the corresponding GNU tag and attributes.
(gen_subprogram_die): Likewise.  Emit call site information even
for -gdwarf-5 -gstrict-dwarf.  Replace DW_AT_GNU_defaulted with
DW_AT_defaulted in comment.
(resolve_addr): Handle DW_AT_call_origin attribute on
DW_TAG_call_site DIE like DW_AT_abstract_origin on
DW_TAG_GNU_call_site DIE.

From-SVN: r241713

dwarf2out.c (dwarf_op): New function.

* dwarf2out.c (dwarf_op): New function.
(size_of_loc_descr): Handle DW_OP_{implicit_pointer,entry_value},
DW_OP_{const,regval,deref}_type and DW_OP_{convert,reinterpret}.
(output_loc_operands, output_loc_operands_raw): Likewise.
(resolve_args_picking_1, prune_unused_types_walk_loc_descr,
mark_base_types, hash_loc_operands, compare_loc_operands): Likewise.
(resolve_addr_in_expr): Likewise.  Only punt for !dwarf_strict
if dwarf_version < 5.
(convert_descriptor_to_mode): Use dwarf_op (DW_OP_xxx) instead of
DW_OP_GNU_xxx.
(scompare_loc_descriptor, ucompare_loc_descriptor,
minmax_loc_descriptor, typed_binop, mem_loc_descriptor,
implicit_ptr_descriptor, optimize_one_addr_into_implicit_ptr,
optimize_location_into_implicit_ptr): Likewise.  Only punt for
!dwarf_strict if dwarf_version < 5.
(string_cst_pool_decl): Adjust comment.
(non_dwarf_expression): Handle DW_OP_implicit_pointer.

From-SVN: r241712

re PR c++/77886 (-Wimplicit-fallthrough: breaks duff's device (in function templates))

PR c++/77886
* pt.c (tsubst_expr) <case CASE_LABEL_EXPR> Copy over
FALLTHROUGH_LABEL_P flag to the new LABEL_DECL.
(tsubst_expr) <case LABEL_EXPR>: Likewise.

* g++.dg/warn/Wimplicit-fallthrough-2.C: New test.

From-SVN: r241711

re PR c++/78089 (__builtin_shuffle parsing bug)

PR c++/78089
* parser.c (cp_parser_postfix_expression): Replace return statement in
the first switch with setting postfix_expression to the return
expression and break;.

* c-c++-common/builtin-shuffle-1.c: New test.
* g++.dg/cpp0x/addressof3.C: New test.

From-SVN: r241710

dwarf2out.h (enum dw_val_class): Add dw_val_class_const_implicit...

* dwarf2out.h (enum dw_val_class): Add dw_val_class_const_implicit,
dw_val_class_unsigned_const_implicit and dw_val_class_file_implicit.
(struct dw_val_node): Add val_file_implicit field.
* dwarf2out.c (dw_val_equal_p, print_dw_val, attr_checksum,
attr_checksum_ordered, same_dw_val_p, size_of_die, value_format,
output_die): Handle dw_val_class_const_implicit,
dw_val_class_unsigned_const_implicit and dw_val_class_file_implicit.
(abbrev_die_table): Change into va_gc vec.
(abbrev_die_table_allocated, abbrev_die_table_in_use,
ABBREV_DIE_TABLE_INCREMENT): Remove.
(AT_int, AT_unsigned, AT_file): Allow dw_val_class_*_implicit.
(abbrev_opt_start, abbrev_usage_count, sorted_abbrev_dies): New
variables.
(build_abbrev_table): Adjust for abbrev_die_table being a va_gc vec.
If abbrev_opt_start, fill in abbrev_usage_count and abbrev_dies
vectors.
(die_abbrev_cmp, optimize_implicit_const, optimize_abbrev_table): New
functions.
(output_die_abbrevs): For DW_FORM_implicit_const emit sleb128 with
the implicit value.
(output_abbrev_section): Adjust for abbrev_die_table being a va_gc
vec.
(output_comp_unit): Initialize abbrev_opt_start if emitting the main
unit.  Call optimize_abbrev_table.
(dwarf2out_init, dwarf2out_finish, dwarf2out_c_finalize): Adjust for
abbrev_die_table being a va_gc vec.

From-SVN: r241709

Add tests for a const member and a reference member for launder.

* g++.dg/cpp1z/launder3.C: New.
* g++.dg/cpp1z/launder4.C: Likewise.
* g++.dg/cpp1z/launder5.C: Likewise.
* g++.dg/cpp1z/launder5.cc: Likewise.
* g++.dg/cpp1z/launder5.h: Likewise.
* g++.dg/cpp1z/launder6.C: Likewise.
* g++.dg/cpp1z/launder6.cc: Likewise.
* g++.dg/cpp1z/launder6.h: Likewise.

From-SVN: r241708

re PR c++/77948 (Option processing of -std=c++11 -std=gnu++11 doesn't reset ext_numeric_literals)

PR c++/77948
* c.opt (fext-numeric-literals): Add Var and Init.
* c-opts.c (c_common_handle_option): Don't clear
cpp_opts->ext_numeric_literals for -std=c++{11,14,1z}.
(c_common_post_options): Clear it here if not set
explicitly.

* g++.dg/cpp0x/pr77948-1.C: New test.
* g++.dg/cpp0x/pr77948-2.C: New test.
* g++.dg/cpp0x/pr77948-3.C: New test.
* g++.dg/cpp0x/pr77948-4.C: New test.
* g++.dg/cpp0x/pr77948-5.C: New test.
* g++.dg/cpp0x/pr77948-6.C: New test.

From-SVN: r241707

re PR tree-optimization/77860 (ICE in gimple_build_assign_1, at gimple.c:420)

PR tree-optimization/77860
* tree-ssa-reassoc.c (eliminate_using_constants): Handle
also integral complex and vector constants.

* gcc.dg/pr77860.c: New test.

From-SVN: r241706

dwarf2out.c (dwarf2out_define, [...]): Replace DW_MACRO_GNU_* constants with corresponding DW_MACRO_* constants.

* dwarf2out.c (dwarf2out_define, dwarf2out_undef, output_macinfo_op,
optimize_macinfo_range, save_macinfo_strings): Replace
DW_MACRO_GNU_* constants with corresponding DW_MACRO_* constants.
(output_macinfo): Likewise.  Emit .debug_macro* rather than
.debug_macinfo* even for -gstrict-dwarf -gdwarf-5.
(init_sections_and_labels): Use .debug_macro* labels rather than
.debug_macinfo* labels even for -gstrict-dwarf -gdwarf-5.
(dwarf2out_finish): Use DW_AT_macros instead of DW_AT_macro_info
or DW_AT_GNU_macros for -gdwarf-5.

From-SVN: r241705

linux.h (UCLIBC_DYNAMIC_LINKER): Define.

2016-10-31  Waldemar Brodkorb  <wbx@openadk.org>

* config/microblaze/linux.h (UCLIBC_DYNAMIC_LINKER): Define.

From-SVN: r241704

Index...

Index: ChangeLog
===================================================================
--- ChangeLog (revision 241702)
+++ ChangeLog (working copy)
@@ -1,5 +1,10 @@
 2016-10-27  Carl Love  <cel@us.ibm.com>

+ * MAINTAINERS (Write After Approval): Fix my entry in the Write After
+ Approval list to make it alphabetical.
+
+2016-10-27  Carl Love  <cel@us.ibm.com>
+
  * MAINTAINERS (Write After Approval): Add myself.

 2016-10-27  Andrew Burgess  <andrew.burgess@embecosm.com>
Index: MAINTAINERS
===================================================================
--- MAINTAINERS (revision 241702)
+++ MAINTAINERS (working copy)
@@ -477,9 +477,9 @@
 Gabor Loki <loki@inf.u-szeged.hu>
 Sandra Loosemore <sandra@codesourcery.com>
 Manuel López-Ibáñez <manu@gcc.gnu.org>
+Carl Love <cel@us.ibm.com>
 Martin v. Löwis <loewis@informatik.hu-berlin.de>
 H.J. Lu <hjl.tools@gmail.com>
-Carl Love <cel@us.ibm.com>
 Christophe Lyon <christophe.lyon@st.com>
 Luis Machado <luisgpm@br.ibm.com>
 Ziga Mahkovec <ziga.mahkovec@klika.si>

From-SVN: r241703

alpha-passes.def: Swap pass_convert_to_eh_region_ranges and pass_shorten_branches positions.

* config/alpha/alpha-passes.def: Swap pass_convert_to_eh_region_ranges
  and pass_shorten_branches positions.

From-SVN: r241702

Skip gcc.dg/lto/pr60449_0.c for avr

The testcase requires gettimeofday to be available for the target. The avr
target doesn't have an implementation, so the test always fails with a
linker error.

gcc/testsuite

2016-10-31  Senthil Kumar Selvaraj  <senthil_kumar.selvaraj@atmel.com>

* gcc.dg/lto/pr60449_0.c: Skip for avr.

From-SVN: r241701

re PR c++/77886 (-Wimplicit-fallthrough: breaks duff's device (in function templates))

PR c++/77886
* pt.c (tsubst_expr) <case CASE_LABEL_EXPR> Copy over
FALLTHROUGH_LABEL_P flag to the new LABEL_DECL.
(tsubst_expr) <case LABEL_EXPR>: Likewise.

* g++.dg/warn/Wimplicit-fallthrough-2.C: New test.

From-SVN: r241700

Add -Wshadow=global -Wshadow=local and -Wshadow=compatible-local.

This patch from Le-Chun Wu adds two new shadow warning flags for
C and C++:

  -Wshadow=local which warns if a local variable shadows another local
  variable or parameter,

  -Wshadow=compatible-local which warns if a local variable shadows
  another local variable or parameter whose type is compatible with
  that of the shadowing variable.

It is already on the google/main branch (Google ref 39127) and was
previously submitted by Diego Novillo and reviewed on
http://codereview.appspot.com/4452058

I addressed the review comments and made the following changes:
- Add -Wshadow=global (the default alias for -Wshadow).
- Make the documented options -Wshadow=global, -Wshadow=local
  and -Wshadow=compatible-local (with hidden undocumented aliases
  -Wshadow-local and -Wshadow-compatible-local for compatibility).
- The -Wshadow=global, -Wshadow=local and -Wshadow=compatible-local
  relationships are expressed in common.opt instead of in opts.c
  and documented in invoke.texi.
- The "previous declaration" warnings were turned into notes and use
  the (now) existing infrastructure instead of duplicating the warnings.
  The testcases have been adjusted to expect the notes.
- The conditional change in name-lookup.c for non-locals (where we
  don't want to change the warnings, but just check the global ones)
  has been dropped.
- Use warning_at in c-decl.c (warn_if_shadowing).

gcc/ChangeLog:
2016-10-30  Le-Chun Wu  <lcwu@google.com>
            Mark Wielaard  <mjw@redhat.com>

       * doc/invoke.texi: Document Wshadow-local and Wshadow-compatible-local.
       * common.opt (Wshadow=global): New option. Default for -Wshadow.
       (Wshadow=local): New option.
       (Wshadow-local): Hidden alias for -Wshadow=local.
       (Wshadow=compatible-local): New option.
       (Wshadow-compatible-local): Hidden alias for
       -Wshadow=compatible-local.
       * doc/invoke.texi: Document Wshadow=global, Wshadow=local and
       Wshadow=compatible-local.

gcc/c/ChangeLog:
2016-10-30  Le-Chun Wu  <lcwu@google.com>
            Mark Wielaard  <mjw@redhat.com>

       * c-decl.c (warn_if_shadowing): Use the warning code corresponding
       to the given -Wshadow= variant. Use warning_at.

gcc/cp/ChangeLog:
2016-10-30  Le-Chun Wu  <lcwu@google.com>
            Mark Wielaard  <mjw@redhat.com>

       * name-lookup.c (pushdecl_maybe_friend): When emitting a
       shadowing warning, use the code corresponding to the
       given -Wshadow= variant.

gcc/testsuite/ChangeLog
2016-10-30  Le-Chun Wu  <lcwu@google.com>
            Mark Wielaard  <mjw@redhat.com>

       * gcc.dg/Wshadow-compatible-local-1.c: New test.
       * gcc.dg/Wshadow-local-1.c: Likewise.
       * gcc.dg/Wshadow-local-2.c: Likewise.
       * g++.dg/warn/Wshadow-compatible-local-1.C: Likewise.
       * g++.dg/warn/Wshadow-local-1.C: Likewise.
       * g++.dg/warn/Wshadow-local-2.C: Likewise.

Co-Authored-By: Mark Wielaard <mjw@redhat.com>
From-SVN: r241699

re PR lto/78129 (-Werror=suggest-final-types leads to -ENOSPC.)

2016-10-31  Richard Biener  <rguenther@suse.de>

PR lto/78129
* lto.c (do_whole_program_analysis): Bail out after errors
from WPA analysis.

From-SVN: r241698

tree-vect-slp.c (vect_get_and_check_slp_defs): New parameter SWAP.

* tree-vect-slp.c (vect_get_and_check_slp_defs): New parameter SWAP.
Check slp defs for COND_EXPR by swapping/inverting operands if the
new parameter SWAP indicates so.
(vect_build_slp_tree_1): New parameter SWAP.  Check COND_EXPR stmt
is isomorphic to the first stmt via swapping/inverting.  Store swap
information in the new parameter SWAP.
(vect_build_slp_tree): New local array SWAP and pass it to function
vect_build_slp_tree_1.  Cleanup result handling code for function
call to vect_get_and_check_slp_defs.  Skip operand swapping if the
order of operands has been fixed as indicated by SWAP[i].

From-SVN: r241697

tree-vect-data-refs.c (vect_slp_analyze_node_dependences): Skip unnecessary data dependence check after visited store stmt.

* tree-vect-data-refs.c (vect_slp_analyze_node_dependences): Skip
unnecessary data dependence check after visited store stmt.

From-SVN: r241696

re PR tree-optimization/71915 (A missed opportunity for SLSR)

[gcc]

2016-10-30  Bill Schmidt  <wschmidt@linux.vnet.ibm.com>

PR tree-optimization/71915
PR tree-optimization/71490
* gimple-ssa-strength-reduction.c (struct slsr_cand_d): Add
stride_type field.
(find_basis_for_base_expr): Require stride types to match when
seeking a basis.
(alloc_cand_and_find_basis): Record the stride type.
(slsr_process_phi): Pass stride type to alloc_cand_and_find_basis.
(backtrace_base_for_ref): Pass types to legal_cast_p_1 rather than
the expressions having those types.
(slsr_process_ref): Pass stride type to alloc_cand_and_find_basis.
(create_mul_ssa_cand): Likewise.
(create_mul_imm_cand): Likewise.
(create_add_ssa_cand): Likewise.
(create_add_imm_cand): Likewise.
(legal_cast_p_1): Change interface to accept types rather than the
expressions having those types.
(legal_cast_p): Pass types to legal_cast_p_1.
(slsr_process_cast): Pass stride type to
alloc_cand_and_find_basis.
(slsr_process_copy): Likewise.
(dump_candidate): Display stride type when a cast exists.
(create_add_on_incoming_edge): Introduce a cast when necessary for
the stride type.
(analyze_increments): Change the code checking for invalid casts
to rely on the stride type, and update the documentation and
example.  Change the code checking for pointer multiplies to rely
on the stride type.
(insert_initializers): Introduce a cast when necessary for the
stride type.  Use the stride type for the type of the initializer.

[gcc/testsuite]

2016-10-30  Bill Schmidt  <wschmidt@linux.vnet.ibm.com>

PR tree-optimization/71915
PR tree-optimization/71490
* gcc.dg/tree-ssa/pr54245.c: Delete.
* gcc.dg/tree-ssa/slsr-8.c: Adjust for new optimization and
document why.

From-SVN: r241695

Daily bump.

From-SVN: r241694

re PR libfortran/78123 (Short reads with T edit descriptor not padding correctly)

2016-10-30  Jerry DeLisle  <jvdelisle@gcc.gnu.org>

PR fortran/78123
* io/transfer.c (formatted_transfer_scalar_read): Clear seen_eor
only if we have tabbed to left of current position.

* gfortran.dg/fmt_t_9.f: New test.

From-SVN: r241691

re PR fortran/67219 (Incorrect conversion warning)

2016-10-30  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/67219
* arith.c (gfc_int2real):  Change gfc_warning_now
to gfc_warning.
* primary.c (match_complex_constant):  If there
is no comma, throw away any warning which might have
been issued by gfc_int2real.

2016-10-30  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/67219
* gfortran.dg/pr67219.f90:  New test.

From-SVN: r241689

compiler: fix slice/array evaluation order bug

    There was a phase ordering issue in the handling of "keyed" array
    literal expressions: the lowering phase was canonicalizing the
    indices/vals before the phase that fixed evaluation order, meaning that
    the evaluation order was incorrect. The fix is to capture the orginal
    ordering and use that ordering when doing traversals (there is already
    something similar being done for struct literal expressions).

    Fixes golang/go#17640

    Reviewed-on: https://go-review.googlesource.com/32296

From-SVN: r241688

libgo: eliminate a lot of duplication in Makefile.am

    I read through the GNU make manual.  I knew there had to be a way to do it.

    Remove the special netgo library.  The essential feature--using the Go
    DNS resolver--is now available by setting GODEBUG=netdns=go.

    Reviewed-on: https://go-review.googlesource.com/32333

From-SVN: r241687

arm.c (arm_const_not_ok_for_debug_p): Use VAR_P.

2016-10-30  Prathamesh Kulkarni  <prathamesh.kulkarni@linaro.org>

* config/arm/arm.c (arm_const_not_ok_for_debug_p): Use VAR_P.

From-SVN: r241686

Add missing name

From-SVN: r241685

Daily bump.

From-SVN: r241684

re PR rtl-optimization/77919 (ICE converting DC to V2DF mode)

PR rtl-optimization/77919
* expr.c (expand_expr_real_1) <normal_inner_ref>: Only avoid forcing
into memory if both modes are complex and their inner modes have the
same precision.  If the two modes are different complex modes, convert
each part separately and generate a new CONCAT.

* g++.dg/torture/pr77919-2.C: New test.

From-SVN: r241681

pa64-hpux.h (FINI_SECTION_ASM_OP): Define to null string.

* config/pa/pa64-hpux.h (FINI_SECTION_ASM_OP): Define to null string.

From-SVN: r241680

re PR tree-optimization/78148 (r241649 causes -fcompare-debug failure on ppc64le)

PR target/78148
* gimple-ssa-store-merging.c
(imm_store_chain_info::output_merged_store): Use build_aligned_type
instead of SET_TYPE_ALIGN on shared integral type.

* gcc.dg/pr78148.c: New test.

From-SVN: r241679

pa.h (BIGGEST_ALIGNMENT): Adjust comment.

* config/pa/pa.h (BIGGEST_ALIGNMENT): Adjust comment.
(MALLOC_ABI_ALIGNMENT): Define to 128 on all targets except SOM.
Adjust comment.

From-SVN: r241676

vax.h (REGNO_REG_CLASS): Access the REGNO argument.

* config/vax/vax.h (REGNO_REG_CLASS): Access the REGNO argument.
* config/spu/spu.h (REGNO_REG_CLASS): Likewise.

From-SVN: r241675

* eo.po: Update.

From-SVN: r241673

Daily bump.

From-SVN: r241672

re PR fortran/71891 (fortran/symbol.c:4864: suspicious if ?)

2016-10-28  Steven G. Kargl <kargl@gcc.gnu.org>

PR fortran/71891
* symbol.c (gfc_type_compatible): Fix typo.

From-SVN: r241668

compiler, runtime: copy slice code from Go 1.7 runtime

    Change the compiler handle append as the gc compiler does: call a
    function to grow the slice, but otherwise assign the new elements
    directly to the final slice.

    For the current gccgo memory allocator the slice code has to call
    runtime_newarray, not mallocgc directly, so that the allocator sets the
    TypeInfo_Array bit in the type pointer.

    Rename the static function cnew to runtime_docnew, so that the stack
    trace ignores it when ignoring runtime functions.  This was needed to
    fix the runtime/pprof tests on 386.

    Reviewed-on: https://go-review.googlesource.com/32218

From-SVN: r241667

* doc/sourcebuild.texi (Ada Tests): Remove mention of gcc chapter.

From-SVN: r241666

target.def (min_arithmetic_precision): New hook.

* target.def (min_arithmetic_precision): New hook.
* doc/tm.texi.in (Misc): Add TARGET_MIN_ARITHMETIC_PRECISION.
* doc/tm.texi: Regenerate.
* internal-fn.c (expand_arith_overflow): Adjust handling of target
dependent support by means of TARGET_MIN_ARITHMETIC_PRECISION.
* targhooks.c (default_min_arithmetic_precision): New function.
* targhooks.h (default_min_arithmetic_precision): Declare.
* config/sparc/sparc.c (TARGET_MIN_ARITHMETIC_PRECISION): Define.
(sparc_min_arithmetic_precision): New function.

From-SVN: r241665

combine: Improve change_zero_ext (fixes PR71847)

This improves a few things in change_zero_ext.  Firstly, it should use
the passed in pattern in recog_for_combine, not the pattern of the insn
(they are not the same if the whole pattern was replaced).  Secondly,
it handled zero_ext of a subreg, but with hard registers we do not get
a subreg, instead the mode of the reg is changed.  So this handles that.
Thirdly, after changing a zero_ext to an AND, the resulting RTL may become
non-canonical, like (ior (ashift ..) (and ..)); the AND should be first,
it is commutative.  And lastly, zero_extract as a set_dest wasn't handled
at all, but now it is.

This fixes the testcase in PR71847, and improves code generation in some
other edge cases too.

PR target/71847
* combine.c (change_zero_ext): Handle zero_ext of hard registers.
Swap commutative operands in new RTL if needed.  Handle zero_ext
in the set_dest.
(recog_for_combine): Pass *pnewpat to change_zero_ext instead of
PATTERN (insn).

From-SVN: r241664

re PR go/78144 (FAIL: time on systems with tzdata2016g installed)

PR go/78144
    libgo: incorporate fix for timezone test

    This brings over the test-only fix for issue 17276 into gccgo/libgo
    (with tzdata-2016g there is a new zone abbreviation).  This is a
    copy of https://golang.org/cl/29995.

    Reviewed-on: https://go-review.googlesource.com/32182

From-SVN: r241661

re PR tree-optimization/43721 (Failure to optimise (a/b) and (a%b) into single __aeabi_idivmod call)

2016-10-28  Prathamesh Kulkarni  <prathamesh.kulkarni@linaro.org>
    Kugan Vivekanandarajah  <kuganv@linaro.org>
    Jim Wilson  <jim.wilson@linaro.org>

PR tree-optimization/43721
* target.def: New hook expand_divmod_libfunc.
* doc/tm.texi.in: Add hook for TARGET_EXPAND_DIVMOD_LIBFUNC
* doc/tm.texi: Regenerate.
* internal-fn.def: Add new entry for DIVMOD ifn.
* internal-fn.c (expand_DIVMOD): New.
* tree-ssa-math-opts.c: Include optabs-libfuncs.h, tree-eh.h,
targhooks.h.
(widen_mul_stats): Add new field divmod_calls_inserted.
(target_supports_divmod_p): New.
(divmod_candidate_p): Likewise.
(convert_to_divmod): Likewise.
(pass_optimize_widening_mul::execute): Call
calculate_dominance_info(), renumber_gimple_stmt_uids() at
beginning of function. Call convert_to_divmod()
and record stats for divmod.
* config/arm/arm.c (arm_expand_divmod_libfunc): Override hook
TARGET_EXPAND_DIVMOD_LIBFUNC.
* doc/sourcebuild.texi: Add items for arm_divmod_simode, divmod,
divmod_simode.

testsuite/
* lib/target-supports.exp (check_effective_target_divmod): New.
(check_effective_target_divmod_simode): Likewise.
(check_effective_target_arm_divmod_simode): Likewise.
* gcc.dg/divmod-1-simode.c: New test.
* gcc.dg/divmod-1.c: Likewise.
* gcc.dg/divmod-2-simode.c: Likewise.
* gcc.dg/divmod-2.c: Likewise.
* gcc.dg/divmod-3-simode.c: Likewise.
* gcc.dg/divmod-3.c: Likewise.
* gcc.dg/divmod-4-simode.c: Likewise.
* gcc.dg/divmod-4.c: Likewise.
* gcc.dg/divmod-5.c: Likewise.
* gcc.dg/divmod-6-simode.c: Likewise.
* gcc.dg/divmod-6.c: Likewise.
* gcc.dg/divmod-7.c: Likewise.

Co-Authored-By: Jim Wilson <jim.wilson@linaro.org>
Co-Authored-By: Kugan Vivekanandarajah <kuganv@linaro.org>
From-SVN: r241660

re PR go/78143 (bootstrap broken in libgo on powerpc-linux-gnu)

PR go/78143
    runtime: build lfstack_32bit.go on ppc

    Missed a build tag.  This is GCC PR 78143.

    Reviewed-on: https://go-review.googlesource.com/32295

From-SVN: r241659

Make filesystem::path work with basic_string_view (P0392R0)

* include/experimental/bits/fs_path.h (__is_path_src)
(_S_range_begin, _S_range_end): Overload to treat string_view as a
Source object.
(path::operator+=, path::compare): Overload for basic_string_view.
* testsuite/experimental/filesystem/path/construct/string_view.cc:
New test.
* testsuite/experimental/filesystem/path/construct/
string_view_cxx17.cc: New test.

From-SVN: r241658

dojump.c (do_jump_by_parts_greater_rtx): Invert probability when swapping the arms of the branch.

* dojump.c (do_jump_by_parts_greater_rtx): Invert probability when
swapping the arms of the branch.
* internal-fn.c (expand_addsub_overflow): Use a straight-line code
sequence for the generic signed-signed-signed case.

From-SVN: r241656

libgo: redirect grep output in mkrsysinfo.sh to /dev/null

    I noticed a stray useless output line when building libgo.

    Reviewed-on: https://go-review.googlesource.com/32294

From-SVN: r241655

Fix filesystem::path for iterators with const value_type

* include/experimental/bits/fs_path.h
(path::_S_convert<_Iter>(_Iter, _Iter)): Remove cv-qualifiers from
iterator's value_type.
(path::_S_convert<_Iter>(_Iter __first, __null_terminated)): Likewise.
Do not use operation not supported by input iterators.
(path::__is_path_iter_src): Add partial specialization for const
encoded character types.
* testsuite/experimental/filesystem/path/construct/range.cc: Test
construction from input iterators with const value types.

From-SVN: r241654

re PR debug/77773 (Segfault when compiling __simd64_float16_t using arm-none-eabi-g++ with debug information)

PR debug/77773
* c-pretty-print.c (simple_type_specifier): Do not dereference `t'
if NULL.

From-SVN: r241653

bfin.c (bfin_legitimate_address_p): Add missing fallthru comment.

* config/bfin/bfin.c (bfin_legitimate_address_p): Add missing
fallthru comment.
* config/bfin/bfin.h (TARGET_CPU_CPP_BUILTINS): Likewise.

From-SVN: r241651

sched: Do not mix prologue and epilogue insns

This patch makes scheduling not reorder prologue insns relative to
epilogue insns and vice versa.  This fixes PR78029.

The problem in that PR:
We have two insns, in this order:

(insn/f 300 299 267 8 (set (reg:DI 65 lr)
        (reg:DI 0 0)) 579 {*movdi_internal64}
     (expr_list:REG_DEAD (reg:DI 0 0)
        (expr_list:REG_CFA_RESTORE (reg:DI 65 lr)
            (nil))))
...
(insn/f 310 268 134 8 (set (mem/c:DI (plus:DI (reg/f:DI 1 1)
                (const_int 144 [0x90])) [6  S8 A8])
        (reg:DI 0 0)) 579 {*movdi_internal64}
     (expr_list:REG_DEAD (reg:DI 0 0)
        (expr_list:REG_CFA_OFFSET (set (mem/c:DI (plus:DI (reg/f:DI 1 1)
                        (const_int 144 [0x90])) [6  S8 A8])
                (reg:DI 65 lr))
            (nil))))

and sched swaps them (when compiling for power6, it tries to put memory
stores together, so insn 310 is moved up past 300 to go together with
some other store).  But the REG_CFA_RESTORE and REG_CFA_OFFSET cannot be
swapped (they both say where the orig value of LR now lives).

PR rtl-optimization/78029
* function.c (prologue_contains, epilogue_contains): New functions.
(record_prologue_seq, record_epilogue_seq): New functions.
* function.h (prologue_contains, epilogue_contains,
record_prologue_seq, record_epilogue_seq): New declarations.
* sched-deps.c (sched_analyze_insn): Make dependencies to prevent
mixing prologue and epilogue insns.
(init_deps): Initialize the new fields in struct deps_desc.
* sched-int.h (struct deps_desc): New fields last_prologue,
last_epilogue, and last_logue_was_epilogue.
* shrink-wrap.c (emit_common_heads_for_components): Record all
emitted prologue and epilogue insns.
(emit_common_tails_for_components): Ditto.
(insert_prologue_epilogue_for_components): Ditto.

From-SVN: r241650

GIMPLE store merging pass

2016-10-28  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>

PR middle-end/22141
* Makefile.in (OBJS): Add gimple-ssa-store-merging.o.
* common.opt (fstore-merging): New Optimization option.
* opts.c (default_options_table): Add entry for
OPT_ftree_store_merging.
* fold-const.h (can_native_encode_type_p): Declare prototype.
* fold-const.c (can_native_encode_type_p): Define.
* params.def (PARAM_STORE_MERGING_ALLOW_UNALIGNED): Define.
(PARAM_MAX_STORES_TO_MERGE): Likewise.
* timevar.def (TV_GIMPLE_STORE_MERGING): New timevar.
* passes.def: Insert pass_tree_store_merging.
* tree-pass.h (make_pass_store_merging): Declare extern
prototype.
* gimple-ssa-store-merging.c: New file.
* doc/invoke.texi (Optimization Options): Document
-fstore-merging.
(--param documentation): Document store-merging-allow-unaligned
and max-stores-to-merge.

2016-10-28  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
            Jakub Jelinek  <jakub@redhat.com>
            Andrew Pinski  <pinskia@gmail.com>

PR middle-end/22141
PR rtl-optimization/23684
* gcc.c-torture/execute/pr22141-1.c: New test.
* gcc.c-torture/execute/pr22141-2.c: Likewise.
* gcc.target/aarch64/ldp_stp_1.c: Adjust for -fstore-merging.
* gcc.target/aarch64/ldp_stp_4.c: Likewise.
* gcc.dg/store_merging_1.c: New test.
* gcc.dg/store_merging_2.c: Likewise.
* gcc.dg/store_merging_3.c: Likewise.
* gcc.dg/store_merging_4.c: Likewise.
* gcc.dg/store_merging_5.c: Likewise.
* gcc.dg/store_merging_6.c: Likewise.
* gcc.dg/store_merging_7.c: Likewise.
* gcc.target/i386/pr22141.c: Likewise.
* gcc.target/i386/pr34012.c: Add -fno-store-merging to dg-options.
* g++.dg/init/new17.C: Likewise.

Co-Authored-By: Andrew Pinski <pinskia@gmail.com>
Co-Authored-By: Jakub Jelinek <jakub@redhat.com>
From-SVN: r241649

Implement std::launder for C++17

* doc/xml/manual/status_cxx2017.xml: Update status.
* doc/html/*: Regenerate.
* include/std/type_traits (has_unique_object_representations): Guard
with __has_builtin check.
* libsupc++/new (launder): Define for C++17.
* testsuite/18_support/launder/1.cc: New test.
* testsuite/18_support/launder/requirements.cc: New test.
* testsuite/18_support/launder/requirements_neg.cc: New test.

From-SVN: r241648

re PR middle-end/72747 (powerpc: wrong code generated for vec_splats in cascading assignment)

gcc:
2016-10-26  Will Schmidt <will_schmidt@vnet.ibm.com>

        PR middle-end/72747
        * gimplify.c (gimplify_init_constructor): Move emit of constructor
        assignment to earlier in the if/else logic.

testsuite:
2016-10-26  Will Schmidt <will_schmidt@vnet.ibm.com>

        PR middle-end/72747
        * c-c++-common/pr72747-1.c: New test.
        * c-c++-common/pr72747-2.c: Likewise.

From-SVN: r241647

re PR middle-end/78128 (fortran/resolve.c:resolve_operator miscompiled at -O2)

2016-10-28  Richard Biener  <rguenther@suse.de>

PR middle-end/78128
PR middle-end/71002
* fold-const.c (make_bit_field_ref): Only adjust alias set
when the original alias set was zero.

From-SVN: r241645

S/390: Add static OSC breaker if necessary.

This patch adds a magic OSC (operand store compare) break instruction
which is necessary if a store is followed closely by a load with same
base+indx+displ while either base or index get modified in between.

The patch improves several SpecCPU testcases running on IBM z13.

gcc/testsuite/ChangeLog:

2016-10-28  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

* gcc.target/s390/oscbreak-1.c: New test.

gcc/ChangeLog:

2016-10-28  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

* config/s390/s390.c (s390_adjust_loop_scan_osc): New function.
(s390_adjust_loops): New function.
(s390_reorg): Invoke s390_adjust_loops.
* config/s390/s390.md: (UNSPEC_OSC_BREAK): New constant.
("osc_break"): New insn definition.

From-SVN: r241644

S/390: Add support for arch<n> arch/tune options.

This patch adds an alternate CPU level naming following the
architecture level number in the Principles of Operations manual.  So
instead of having z196, zEC12, and z13 you can use arch9, arch10, and
arch11.  The old cpu names stay valid and should preferably be used.

The alternate names are supposed to improve compatibility with the IBM
XL compiler toolchain which uses the arch numbering.

gcc/testsuite/ChangeLog:

2016-10-28  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

* gcc.target/s390/target-attribute/tattr-m64-33.c: New test.

gcc/ChangeLog:

2016-10-28  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

* config/s390/s390.opt: Support alternate cpu level naming (archXX).
* config.gcc: Support alternate archXX cpu levels with
--with-arch= and --with-tune=.
* config/s390/linux.h: Translate new archXX cpu levels to the
original names when calling GAS.
* config/s390/tpf.h: Likewise.
* doc/invoke.texi: Document the alternate cpu level names.

From-SVN: r241643

re PR rtl-optimization/77919 (ICE converting DC to V2DF mode)

PR rtl-optimization/77919
* expr.c (expand_expr_real_1) <normal_inner_ref>: Force CONCAT into
MEM if mode1 is not a complex mode.

* g++.dg/torture/pr77919.C: New test.

From-SVN: r241642

re PR rtl-optimization/78132 (GCC produces invalid instruction (kmovd and kmovq) for KNL.)

PR rtl-optimization/78132
* ree.c (combine_reaching_defs): Give up if copy_needed and
!HARD_REGNO_MODE_OK (REGNO (src_reg), dst_mode).

* gcc.target/i386/pr78132.c: New test.

From-SVN: r241641

Daily bump.

From-SVN: r241640

Index...

Index: ChangeLog
===================================================================
--- ChangeLog (revision 241636)
+++ ChangeLog (working copy)
@@ -1,3 +1,7 @@
+2016-10-27  Carl Love  <cel@us.ibm.com>
+
+ * MAINTAINERS (Write After Approval): Add myself.
+
 2016-10-27  Andrew Burgess  <andrew.burgess@embecosm.com>

  * MAINTAINERS (Reviewers): Add myself.
Index: MAINTAINERS
===================================================================
--- MAINTAINERS (revision 241636)
+++ MAINTAINERS (working copy)
@@ -479,6 +479,7 @@
 Manuel López-Ibáñez <manu@gcc.gnu.org>
 Martin v. Löwis <loewis@informatik.hu-berlin.de>
 H.J. Lu <hjl.tools@gmail.com>
+Carl Love <cel@us.ibm.com>
 Christophe Lyon <christophe.lyon@st.com>
 Luis Machado <luisgpm@br.ibm.com>
 Ziga Mahkovec <ziga.mahkovec@klika.si>

From-SVN: r241637

* gcc.dg/vect/pr71264.c: XFAIL on SPARC.

From-SVN: r241634

* config/sparc/sparc.md (<*vlop:code><VL:mode>3): Remove leading '*'.

From-SVN: r241632

constraints.md (wH constraint): Add new constraints for allowing 32-bit integers (and eventually 8/16-bit...

[gcc]
2016-10-27  Michael Meissner  <meissner@linux.vnet.ibm.com>

* config/rs6000/constraints.md (wH constraint): Add new
constraints for allowing 32-bit integers (and eventually 8/16-bit
integers) into the vector registers.
(wI constraint): Likewise.
(wJ constraint): Likewise.
(wK constraint): Likewise.
* config/rs6000/rs6000-cpus.def (ISA_2_7_MASKS_SERVER): Add
-mvsx-small-integer as a default option for ISA 2.07
(i.e. power8).
(POWERPC_MASKS): Likewise.
* config/rs6000/rs6000.opt (-mvsx-small-integer): Add new debug
switch to turn off small integer support in vector registers.
* config/rs6000/rs6000.c (rs6000_hard_regno_mode_ok): Eliminate
test for -mupper-regs-di, since it is already done with the
reg_add[mode].scalar_in_vsx_p.  Add support for the switch
-mvsx-small-integer.
(rs6000_debug_reg_global): Add support for wH, wI, wJ, and wK
constraints.
(rs6000_setup_reg_addr_masks): Likewise.
(rs6000_init_hard_regno_mode_ok): Likewise.
(rs6000_option_override_internal): Add consistency checks for
-mvsx-small-integer.
(rs6000_secondary_reload_simple_move): SImode is a simple move if
-mvsx-small-integer.
(rs6000_secondary_reload): Use std::swap.
(rs6000_preferred_reload_class): Don't prefer FLOAT_REGS over
VSX_REGS for small integers in vector registers, since there is no
D-FORM address mode for such types.
(rs6000_register_move_cost): Use FIRST_FPR_REGNO instead of 32.
(rs6000_opt_masks): Add -mvsx-small-integer.
* config/rs6000/vsx.md (VSINT_84): Add SImode for small integer
support.
(VSX_EXTRACT_I2): Clone VSX_EXTRACT_I, but drop V4SI since SImode
extracts can be done on ISA 2.07.
(vsx_extract_<mode>): Add support for small integers in vsx
registers.
(vsx_extract_<mode>_p9): Use 'v' instead of VSX_EX, since we no
longer support V4SImode in this pattern.
(vsx_extract_si): New insn to support extraction of SImode in ISA
2.07 using either xxextractuw or vspltw.
(vsx_extract_<mode>_p8): Use 'v' instead of VSX_EX, since we no
longer support V4SImode in this pattern.
* config/rs6000/rs6000.h (enum rs6000_reg_class_enum): Add wH, wI,
wJ, and wK constraints.
* config/rs6000/rs6000.md (f32_sv): Use correct instruction for
storing SDmode with VSX instructions.
(zero_extendsi<mode>2): Reorder pattern, so RLDICL comes after the
GPR load and before the FPR and VSX loads.  Remove ??, ! from the
constraints.  Add MFVSRWZ and XXEXTRACTUW instructions to support
small integers in vector registers.
(extendsi<mode>2): Reorder pattern, so EXTSW comes after the GPR
load and before the FPR and VSX loads.  Remove ??, ! from the
constraints.  Add VEXTSW2D support for small integers in vector
registers.
(lfiwax): Remove ! constraint.  Add VEXTSW2D support for small
integers in vector registers.
(floatsi<mode>2_lfiwax): If -mvsx-small-integer issue a normal
move instead of using an UNSPEC.
(lfiwzx): Remove ! constraint.  Add XXEXTRACTUW support for small
integers in vector registers.
(floatunssi<mode>2_lfiwzx): If -mvsx-small-integer issue a normal
move instead of using an UNSPEC.
(movsi_internal1): Add support for -mvsx-small-integer.  Align
columns so that it is more readable.
(SImode splitter for ISA 3.0 constants): Add splitter for
-128..127 constants that can easily be constructed on ISA 3.0.
* doc/md.texi (PowerPC Constraints): Document wH, wI, wJ, and wK
constraints.

[gcc/testsuite]
2016-10-27  Michael Meissner  <meissner@linux.vnet.ibm.com>

* gcc.target/powerpc/vsx-simode.c: New test.
* gcc.target/powerpc/vsx-simode2.c: Likewise.
* gcc.target/powerpc/vsx-simode3.c: Likewise.

From-SVN: r241631

re PR fortran/78026 (ICE in gfc_resolve_omp_declare_simd, at fortran/openmp.c:5190)

PR fortran/78026
* parse.c (decode_statement): Don't create namespace for possible
select type here and destroy it afterwards.
(parse_select_type_block): Set gfc_current_ns to new_st.ext.block.ns.
(parse_executable, gfc_parse_file): Formatting fixes.
* match.c (gfc_match_select_type): Create namespace for select type
here, only after matching select type.  Formatting fixes.  Free that
namespace if not returning MATCH_YES, after gfc_undo_symbols,
otherwise remember it in new_st.ext.block.ns and switch to parent
namespace anyway.

* gfortran.dg/gomp/pr78026.f03: New test.
* gfortran.dg/select_type_38.f03: New test.

From-SVN: r241630

PR70975 Pass valid offset argument to sendfile

PR libstdc++/70975
* src/filesystem/ops.cc (do_copy_file) [_GLIBCXX_USE_SENDFILE]:
Pass non-null pointer to sendfile for offset argument.

From-SVN: r241629

re PR middle-end/78025 (ICE in simd_clone_adjust, at omp-simd-clone.c:1126)

PR middle-end/78025
* omp-simd-clone.c (simd_clone_adjust): Handle noreturn declare simd
functions.

* g++.dg/gomp/declare-simd-7.C: New test.

From-SVN: r241628

oacc-init.c (goacc_new_thread): Use sizeof of the appropriate size when allocating new thread.

* oacc-init.c (goacc_new_thread): Use sizeof of the appropriate
size when allocating new thread.

From-SVN: r241627

Fix initialization of UNIONs with -finit-derived.

gcc/fortran/
* expr.c (generate_union_initializer, get_union_initializer): New.
* expr.c (component_initializer): Consider BT_UNION specially.
* resolve.c (resolve_structure_cons): Hack for BT_UNION.
* trans-expr.c (gfc_trans_subcomponent_assign): Ditto.
* trans-expr.c (gfc_conv_union_initializer): New.
* trans-expr.c (gfc_conv_structure): Replace UNION handling code with
new function gfc_conv_union_initializer.

gcc/testsuite/gfortran.dg/
* dec_init_1.f90, dec_init_2.f90: Remove -fdump-tree-original.
* dec_init_3.f90, dec_init_4.f90: New tests.

From-SVN: r241626

builtins.c (expand_builtin_nonlocal_goto): Avoid evaluating PIC_OFFSET_TABLE_REGNUM twice.

* builtins.c (expand_builtin_nonlocal_goto): Avoid evaluating
PIC_OFFSET_TABLE_REGNUM twice.

From-SVN: r241625

match.pd ((convert (op:s (convert@2 @0) (convert?@3 @1)))): Add support for constant operand for OP.

* match.pd ((convert (op:s (convert@2 @0) (convert?@3 @1)))): Add
support for constant operand for OP.

gcc/testsuite
* gcc.dg/fold-narrowbopcst-1.c: New test.

From-SVN: r241624

Fix some DEC I/O testcases.

gcc/testsuite/gfortran.dg/
* dec_io_5.f90, dec_io_6.f90: Don't use "test.txt", and use
dg-shouldfail/dg-output instead of XFAIL.

From-SVN: r241623

dwarf2out.c (gen_member_die): Only reparent_child instead of splice_child_die if...

* dwarf2out.c (gen_member_die): Only reparent_child instead of
splice_child_die if child doesn't have DW_AT_specification attribute.

From-SVN: r241622

Fix target selectors in uniform_inside_sphere_distribution tests

* testsuite/ext/random/uniform_inside_sphere_distribution/cons/
default.cc: Fix effective target selector.
* testsuite/ext/random/uniform_inside_sphere_distribution/cons/
parms.cc: Likewise.
* testsuite/ext/random/uniform_inside_sphere_distribution/operators/
equal.cc: Likewise.
* testsuite/ext/random/uniform_inside_sphere_distribution/operators/
generate.cc: Likewise.
* testsuite/ext/random/uniform_inside_sphere_distribution/operators/
inequal.cc: Likewise.
* testsuite/ext/random/uniform_inside_sphere_distribution/operators/
serialize.cc: Likewise.

From-SVN: r241621

* class.c (add_method): Allow using-declarations to coexist.

From-SVN: r241620

Add myself to the MAINTAINERS file

* MAINTAINERS (Reviewers): Add myself.
(Write After Approval): Add myself.

From-SVN: r241619

Adjust precision of filesystem::last_write_time tests

* testsuite/experimental/filesystem/iterators/directory_iterator.cc:
Use end() function to get end iterator.
* testsuite/experimental/filesystem/iterators/pop.cc: Remove printf
statements that were present for debugging.
* testsuite/experimental/filesystem/iterators/
recursive_directory_iterator.cc: Use end() function to get end
iterator.
* testsuite/experimental/filesystem/operations/last_write_time.cc:
Only require file timestamps to be accurate to one second.

From-SVN: r241616

Enable ARMv8-M atomic and synchronization support for ARMv8-M Baseline

2016-10-27  Thomas Preud'homme  <thomas.preudhomme@arm.com>

    gcc/
    * config/arm/arm.h (TARGET_HAVE_LDREX): Define for ARMv8-M Baseline.
    (TARGET_HAVE_LDREXBH): Likewise.
    (TARGET_HAVE_LDACQ): Likewise.

    gcc/testsuite/
    * gcc.target/arm/atomic-comp-swap-release-acquire-3.c: New test.
    * gcc.target/arm/atomic-op-acq_rel-3.c: Likewise.
    * gcc.target/arm/atomic-op-acquire-3.c: Likewise.
    * gcc.target/arm/atomic-op-char-3.c: Likewise.
    * gcc.target/arm/atomic-op-consume-3.c: Likewise.
    * gcc.target/arm/atomic-op-int-3.c: Likewise.
    * gcc.target/arm/atomic-op-relaxed-3.c: Likewise.
    * gcc.target/arm/atomic-op-release-3.c: Likewise.
    * gcc.target/arm/atomic-op-seq_cst-3.c: Likewise.
    * gcc.target/arm/atomic-op-short-3.c: Likewise.

From-SVN: r241615