Add whitespace to some dejagnu directives in libstdc++ tests

* testsuite/20_util/add_rvalue_reference/requirements/alias_decl.cc:
Add whitespace to dejagnu directive.
* testsuite/23_containers/array/element_access/at_neg.cc: Likewise.

From-SVN: r262245

tree-vect-data-refs.c (vect_analyze_data_ref_dependences): Assert compute_all_dependences succeeds.

2018-06-29  Richard Biener  <rguenther@suse.de>

* tree-vect-data-refs.c (vect_analyze_data_ref_dependences): Assert
compute_all_dependences succeeds.
* tree-vect-loop.c (vect_get_datarefs_in_loop): Fail early if we
exceed --param loop-max-datarefs-for-datadeps.

From-SVN: r262244

* config/rs6000/t-rs6000: Append rs6000-modes.h to TM_H.

From-SVN: r262241

Daily bump.

From-SVN: r262240

runtime: don't stat a NULL filename

    Noticed in https://gcc.gnu.org/PR86331.

    Reviewed-on: https://go-review.googlesource.com/121417

From-SVN: r262234

C++: less verbose error-recovery for version conflict markers

gcc/cp/ChangeLog:
* parser.c (cp_parser_error_1): After issuing a conflict marker
error, consume tokens until the end of the source line.

gcc/testsuite/ChangeLog:
* g++.dg/conflict-markers-2.C: New test.

From-SVN: r262232

PR c++/86342 - -Wdeprecated-copy and system headers.

* decl2.c (cp_warn_deprecated_use): Don't warn about declarations
in system headers.

From-SVN: r262231

p9-extract-1.c: Add test case.

gcc/testsuite/ChangeLog:

2018-06-28  Carl Love  <cel@us.ibm.com>

* gcc.target/p9-extract-1.c: Add test case.
* gcc.target/builtins-3-p9-runnable.c: Add test case to match
name in ABI.

From-SVN: r262230

re PR target/86348 (ICE: in curr_insn_transform, at lra-constraints.c:3919: unable to generate reloads for: vec_extractv4si_0_zext_sse4)

PR target/86348
* config/i386/sse.md (*vec_extractv4si_0_zext_sse4): Use
alternative 0 in preferred_for_speed attribute.

testsuite/ChangeLog:

PR target/86348
* gcc.target/i386/pr86348.c: New test.

From-SVN: r262228

Fix insn length for pdp11 shift patterns.

* config/pdp11/pdp11-protos.h (pdp11_shift_length): New function.
* config/pdp11/pdp11.c (pdp11_shift_length): New function.
* config/pdp11/pdp11.h (ADJUST_INSN_LENGTH): Remove.
* config/pdp11/pdp11.md: Correct "length" attribute calculation
for shift insn patterns.

From-SVN: r262227

re PR go/86343 (types built by GO share TYPE_FIELDS in unsupported way)

PR go/86343
* go-gcc.cc (Gcc_backend::set_placeholder_struct_type): Call
build_variant_type_copy rather than build_distinct_type_copy.

From-SVN: r262225

[testsuite] Fix f16_mov_immediate_3.c

Fix and simplify the testcase so it generates dup even on latest trunk.

    testsuite/
* gcc.target/aarch64/f16_mov_immediate_3.c: Fix testcase.

From-SVN: r262222

re PR fortran/82865 (Option -fdec collides with PDT)

2018-06-28  Fritz Reese  <fritzoreese@gmail.com>

    gcc/fortran/ChangeLog:

PR fortran/82865
* decl.c (gfc_match_type): Refactor and check for PDT declarations.

    gcc/testsuite/ChangeLog:

PR fortran/82865
* gfortran.dg/dec_type_print_2.f03: New testcase.

From-SVN: r262221

Hide alt_dump_file within dumpfile.c

This patch removes alt_dump_file from dumpfile.h, making it static
within dumpfile.c.  This allows for changing how -fopt-info is
implemented, and potentially adding other kinds of dump target, such
as remarks or optimization records.

Doing so requires changing the implementation of dump_enabled_p, so
the patch changes this to a simple lookup of a boolean global, which
is updated any time dump_file or alt_dump_file change.

gcc/ChangeLog:
* cgraph.c (cgraph_node::get_body): Replace assignments to
"dump_file" with calls to set_dump_file.
* dumpfile.c (alt_dump_file): Make static, and group with...
(alt_flags): ...this definition.
(dumps_are_enabled): New variable.
(refresh_dumps_are_enabled): New function.
(set_dump_file): New function.
(set_alt_dump_file): New function.
(gcc::dump_manager::dump_start): Replace assignments to
"dump_file" and "alt_dump_file" with calls to set_dump_file and
set_alt_dump_file.
(gcc::dump_manager::dump_finish): Likewise.
* dumpfile.h (alt_dump_file): Delete decl.
(dumps_are_enabled): New variable decl.
(set_dump_file): New function decl.
(dump_enabled_p): Rewrite in terms of new "dumps_are_enabled"
global.
* tree-nested.c (lower_nested_functions): Replace assignments to
"dump_file" with calls to set_dump_file.

From-SVN: r262220

tree-cfg.c (verify_gimple_in_cfg): Call verify_location on the goto_locus of each outgoing edge of each basic block.

* tree-cfg.c (verify_gimple_in_cfg): Call verify_location on the
goto_locus of each outgoing edge of each basic block.

From-SVN: r262219

dwarf2out.c (decl_scope_table): Remove.

2018-06-28  Richard Biener  <rguenther@suse.de>

* dwarf2out.c (decl_scope_table): Remove.
(push_decl_scope): Likewise.
(pop_decl_scope): Likewise.
(gen_type_die_for_member): Do not call push/pop_decl_scope.
(gen_struct_or_union_type_die): Likewise.
(gen_tagged_type_die): Likewise.
(dwarf2out_init): Do not initialize decl_scope_table.
(dwarf2out_c_finalize): Do not free it.

From-SVN: r262218

dwarf2out.c (gen_subprogram_die): Use is_unit_die when deciding whether to not re-use a DIE.

2018-06-28  Richard Biener  <rguenther@suse.de>

* dwarf2out.c (gen_subprogram_die): Use is_unit_die when
deciding whether to not re-use a DIE.

From-SVN: r262217

* MAINTAINERS (write after approval): Add myself.

From-SVN: r262216

re PR c/55976 (-Werror=return-type should error on returning a value from a void function)

2018-06-28  David Pagan  <dave.pagan@oracle.com>

PR c/55976
* gcc.dg/noncompile/pr55976-1.c: Add dg-prune-output for extraneous
message causing unexpected test FAIL.

From-SVN: r262215

Add missing header file inclusion.

2018-06-28  Martin Liska  <mliska@suse.cz>

* brigspec.c: Add missing header file inclusion.

From-SVN: r262214

dwarf2out.c (gen_subprogram_die): Always re-use DIEs with an DW_AT_abstract_origin attribute.

2018-06-28  Richard Biener  <rguenther@suse.de>

* dwarf2out.c (gen_subprogram_die): Always re-use DIEs with an
DW_AT_abstract_origin attribute.

From-SVN: r262213

Come up with jump_table ratio constants used in jump_table_cluster.

2018-06-28  Martin Liska  <mliska@suse.cz>

* tree-switch-conversion.c (jump_table_cluster::can_be_handled):
        Use newly introduced constants.
* tree-switch-conversion.h (struct jump_table_cluster):
        Define max_ratio_for_size and max_ratio_for_speed.

From-SVN: r262212

Fix clustering algorithm in switch expansion.

2018-06-28  Martin Liska  <mliska@suse.cz>

* tree-switch-conversion.c (jump_table_cluster::find_jump_tables):
        Add new checking assert to catch invalid state.
(jump_table_cluster::can_be_handled): Handle single case
        clusters.
(jump_table_cluster::is_beneficial): Bail out for such case.
(bit_test_cluster::find_bit_tests):
        Add new checking assert to catch invalid state.
(bit_test_cluster::can_be_handled): Handle single case
        clusters.
(bit_test_cluster::is_beneficial): Bail out for such case.
(switch_decision_tree::analyze_switch_statement):
        Fix comment.
2018-06-28  Martin Liska  <mliska@suse.cz>

* gcc.dg/tree-ssa/switch-1.c: New test.

From-SVN: r262211

Come up with new --completion option.

2018-06-28  Martin Liska  <mliska@suse.cz>

* common.opt: Introduce -completion option.
* gcc.c (driver_handle_option): Handle it.
(driver::main): Print completions if completion
        is set.
* opt-suggestions.c (option_proposer::get_completions):
        New function.
(option_proposer::suggest_completion): Likewise.
(option_proposer::find_param_completions): Likewise.
(verify_autocompletions): Likewise.
(test_completion_valid_options): Likewise.
(test_completion_valid_params): Likewise.
(in_completion_p): Likewise.
(empty_completion_p): Likewise.
(test_completion_partial_match): Likewise.
(test_completion_garbage): Likewise.
(opt_proposer_c_tests): Likewise.
* opt-suggestions.h: Declare new functions.
* opts.c (common_handle_option): Handle OPT__completion_.
* selftest-run-tests.c (selftest::run_tests): Add
        opt_proposer_c_tests.
* selftest.c (assert_str_startswith): New.
* selftest.h (assert_str_startswith): Likewise.
(opt_proposer_c_tests): New.
(ASSERT_STR_STARTSWITH): Likewise.

From-SVN: r262210

Makefile.in: Add opt-suggestions.o.

.
2018-06-28  Martin Liska  <mliska@suse.cz>

* Makefile.in: Add opt-suggestions.o.
* gcc-main.c: Include opt-suggestions.h.
* gcc.c (driver::driver): Likewise.
(driver::~driver): Remove m_option_suggestions.
(driver::build_option_suggestions): Moved to option_proposer.
(driver::suggest_option): Likewise.
(driver::handle_unrecognized_options): Use option_proposer.
* gcc.h (class driver): Add new memver m_option_proposer.
* opt-suggestions.c: New file.
* opt-suggestions.h: New file.
2018-06-28  Martin Liska  <mliska@suse.cz>

* cppspec.c: Include opt-suggestions.h.
2018-06-28  Martin Liska  <mliska@suse.cz>

* gfortranspec.c: Include opt-suggestions.h.
2018-06-28  Martin Liska  <mliska@suse.cz>

* jit-playback.c: Include opt-suggestions.h.

From-SVN: r262209

Introduce auto_string_vec class.

2018-06-28  Martin Liska  <mliska@suse.cz>

* vec.h (class auto_string_vec): New (moved from auto_argvec).
(auto_string_vec::~auto_string_vec): Likewise.
2018-06-28  Martin Liska  <mliska@suse.cz>

* jit-playback.c (class auto_argvec): Moved to vec.h.
(auto_argvec::~auto_argvec): Likewise.
(compile): Use the renamed name.
(invoke_driver): Likewise.

From-SVN: r262208

tree-inline.c (remap_gimple_stmt): Force input_location on the new statement if id->reset_location is true.

* tree-inline.c (remap_gimple_stmt): Force input_location on the new
statement if id->reset_location is true.
(copy_edges_for_bb): Do not set goto_locus on the new edges if
id->reset_location is true.
(copy_phis_for_bb): Force input_location on the arguments if
id->reset_location is true.
(expand_call_inline): Set id->reset_location if DECL_IGNORED_P
is set on the function to be inlined.
* tree-inline.h (struct copy_body_data): Move remapping_type_depth and
prevent_decl_creation_for_types fields up and add reset_location field.

From-SVN: r262207

* doc/invoke.texi (Debugging Options): Add -gsplit-dwarf.

From-SVN: r262206

lra-eliminations.c (update_reg_eliminate): Mark all spanning hard registers for Pmode.

2018-06-23  Dimitar Dimitrov  <dimitar@dinux.eu>

* lra-eliminations.c (update_reg_eliminate): Mark all spanning hard
registers for Pmode.
* lra-lives.c (check_pseudos_live_through_calls): Mark all spanning
hard registers for the clobbered pseudo.

From-SVN: r262205

Daily bump.

From-SVN: r262203

C++: don't offer bogus "._0" suggestions (PR c++/86329)

PR c++/86329 reports that the C++ frontend can offer bogus suggestions like:

#include <string>

int compare()
{
  return __n1 - __n2;
}

suggested.cc: In function 'int compare()':
suggested.cc:5:10: error: '__n1' was not declared in this scope
   return __n1 - __n2;
          ^~~~
suggested.cc:5:10: note: suggested alternative: '._61'
   return __n1 - __n2;
          ^~~~
          ._61
suggested.cc:5:17: error: '__n2' was not declared in this scope
   return __n1 - __n2;
                 ^~~~
suggested.cc:5:17: note: suggested alternative: '._72'
   return __n1 - __n2;
                 ^~~~
                 ._72

The dot-prefixed names are an implementation detail of how we implement
anonymous enums found in the header files, generated via
anon_aggrname_format in make_anon_name.

This patch uses anon_aggrname_p to filter them out when considering
which names to suggest.

gcc/cp/ChangeLog:
PR c++/86329
* name-lookup.c (consider_binding_level): Filter out names that
match anon_aggrname_p.

gcc/testsuite/ChangeLog:
PR c++/86329
* g++.dg/lookup/pr86329.C: New test.

From-SVN: r262199

Convert pdp11 back end to CCmode.

* common/config/pdp11/pdp11-common.c (pdp11_handle_option): Handle
mutually exclusive options.
* config/pdp11/constraints.md (h): New constraint.
(O): Update definition to match shift code generation.
(D): New constraint.
* config/pdp11/pdp11-modes.def (CCNZ): Define mode.
(CCFP): Remove.
* config/pdp11/pdp11-protos.h (int_no_side_effect_operand): New
function.
(output_jump): Change arguments.
(pdp11_fixed_cc_regs): New function.
(pdp11_cc_mode): Ditto.
(pdp11_expand_shift): Ditto.
(pdp11_assemble_shift): Ditto.
(pdp11_small_shift): Ditto.
(pdp11_branch_cost): Remove.
* config/pdp11/pdp11.c (pdp11_assemble_integer): Remove comments
from output.
(pdp11_register_move_cost): Update for CC registers.
(pdp11_rtx_costs): Add case for LSHIFTRT.
(pdp11_output_jump): Add CCNZ mode conditional branches.
(notice_update_cc_on_set): Remove.
(pdp11_cc_mode): New function.
(simple_memory_operand): Correct pre/post decrement case.
(no_side_effect_operand): New function.
(pdp11_regno_reg_class): Add CC_REGS class.
(pdp11_fixed_cc_regs): New function.
(pdp11_small_shift): New function.
(pdp11_expand_shift): New function to expand shift insns.
(pdp11_assemble_shift): New function to output shifts.
(pdp11_branch_cost): Remove.
(pdp11_modes_tieable_p): Make QI/HI modes tieable.
* config/pdp11/pdp11.h (SIZE_TYPE): Ensure 16-bit type.
(WCHAR_TYPE): Ditto.
(PTRDIFF_TYPE): Ditto.
(ADJUST_INSN_LENGTH): New macro.
(FIXED_REGISTERS): Add CC registers.
(CALL_USED_REGISTERS): Ditto.
(reg_class): Ditto.
(REG_CLASS_NAMES): Ditto.
(REG_CLASS_CONTENTS): Ditto.
(SELECT_CC_MODE): Use new function.
(TARGET_FLAGS_REGNUM): New macro.
(TARGET_FIXED_CONDITION_CODE_REGS): Ditto.
(cc0_reg_rtx): Remove.
(CC_STATUS_MDEP): Remove.
(CC_STATUS_MDEFP_INIT): Remove.
(CC_IN_FPU): Remove.
(NOTICE_UPDATE_CC): Remove.
(REGISTER_NAMES): Add CC registers.
(BRANCH_COST): Change to constant 1.
* config/pdp11/pdp11.md: Rewrite for CCmode condition code
handling.
* config/pdp11/pdp11.opt (mbcopy): Remove.
(mbcopy-builtin): Remove.
(mbranch-cheap): Remove.
(mbranch-expensive): Remove.
* config/pdp11/predicates.md (expand_shift_operand): Update to
match shift code generation.
(ccnz_operator): New predicate.
* doc/invoke.texi (PDP-11 Options): Remove deleted options
-mbcopy, -mbcopy-builtin, -mbranch-cheap, -mbranch-expensive.
Remove non-existent option -mabshi, -mno-abshi.  Document mutually
exclusive options.
* doc/md.texi (PDP-11): Document new D and h constraints.  Update
description of O constraint.

From-SVN: r262198

Add test case that was supposed to be added in commit 255556 on 2017-12-11.

gcc/testsuite/ChangeLog:

2018-06-27  Carl Love  <cel@us.ibm.com>

Add test case that was supposed to be added in commit 255556 on 2017-12-11.

* gcc.target/vsx-vector-abss.c: New file to test vec_abss.

From-SVN: r262196

Fix typo in libgcc/Makefile.in

* Makefile.in (install_leaf): Use enable_gcov instead of
enable_libgcov.

From-SVN: r262195

stl_vector.h (struct _Vector_base<>::_Vector_impl_data): New.

2018-06-27  François Dumont  <fdumont@gcc.gnu.org>

* include/bits/stl_vector.h
(struct _Vector_base<>::_Vector_impl_data): New.
(struct _Vector_base<>::_Vector_impl): Inherit from latter.
(_Vector_base<>::_Vector_impl::_M_swap_data): Move...
(_Vector_base<>::_Vector_impl_data::_M_swap_data): ...here.
(_Vector_base<>::_Vector_impl()): Add noexcept qualification.
(_Vector_base<>::_Vector_impl(_Vector_impl&&)): New.
(_Vector_base<>::_Vector_impl(_Tp_alloc_type&&, _Vector_impl&&)): New.
(_Vector_base(const allocator_type&, _Vector_base&&)): New, use latter.
(_Vector_base()): Default.
(_Vector_base(_Vector_base&&)): Default.
(_Vector_base(size_t)) [_GLIBCXX_INLINE_VERSION]: Delete.
(_Vector_base(_Tp_alloc_type&&)) [_GLIBCXX_INLINE_VERSION]: Delete.
(_Vector_base::_M_create_storage(size_t)): Make protected.
(vector()): Default.
(vector(vector&&)): Default.
(vector(vector&&, const allocator_type&, true_type)): New.
(vector(vector&&, const allocator_type&, false_type)): New.
(vector(vector&&, const allocator_type&)): Use latters.
(vector(_InputIte, _InputIte, const allocator_type&)): Call
_M_range_initialize directly.
* include/debug/vector
(vector(vector&&, const allocator_type&)): Add noexcept qualification.
* testsuite/23_containers/vector/allocator/default_init.cc: New.
* testsuite/23_containers/vector/cons/noexcept_move_construct.cc: Add
static assertions.

From-SVN: r262194

v850.md (addsi3_set_flags): New pattern.

* config/v850/v850.md (addsi3_set_flags): New pattern.
(subsi3_set_flags, negsi2_set_flags, andsi3_set_flags): Likewise.
(iorsi3_set_flags, xorsi3_set_flags, one_cmplsi2_set_flags): Likewise.
(zero_extendhisi2_v850_set_flags): Likewise.
(zero_extendqisi2_v850_set_flags): Likewise.
(ashlsi3_set_flags, ashlsi3_v850e2_set_flags): Likewise.
(lshrsi3_set_flags, lshrsi3_v850e2_set_flags): Likewise.
(ashrsi3_set_flags, ashrsi3_v850e2_set_flags): Likewise.

Co-Authored-By: Austin Law <austinklaw@gmail.com>
From-SVN: r262192

v850-protos.h (notice_update_cc): Remove.

* config/v850/v850-protos.h (notice_update_cc): Remove.
* config/v850/v850.c (v850_compare_op0, v850_compare_op1): Remove.
(v850_print_operand): Handle 'D' and "d".
(v850_select_cc_mode): Remove ATTRIBUTE_UNUSED for last argument.
Add handling of arithmetic/logical operations compared against zero.
(v850_gen_float_compare): Remove ATTRIBUTE_UNUSED for last argument.
Do not look at v850_compare_op, instead get mode from last argument.
(v850_gen_compare): Remove
(increment_stack): Use addsi3_clobber_flags to avoid splitting failure
after reload for prologue insns.
(expand_prologue): Account for CLOBBER of CC_REGNUM in various
patterns.
(construct_save_jarl): Likewise.
(TARGET_FLAGS_REGNUM): Define.
* config/v850/v850.h (v850_compare_op0, v850_compare_op1): Remove.
(NOTICE_UPDATE_CC): Remove.
* config/v850/v850.md (v850_tst1): Use (reg:CCZ CC_REGNUM) rather
than cc0.  Conditionalize on reload_completed.
(cmpsi_insn, setfcc_insn): Likewise.
(tst1 splitter): Turn into define_and_split which sets the flags
after reload.
(cstoresi4, cbranchsf4, cbranchdf4, cbranchsi4_insn): Likewise.
(cbranchsi4, branch_normal, branch_invert): Do not expose cc0 here.
(cstoresf4, cstoredf4): Clobber the flags.
(cmpsi, cmpsf, cmpdf): Remove expanders.
(setf_insn): Remove pattern.
(addsi3): Turn into define_and_split which clobbers the flags after
reload and a suitable pattern (addsi3_clobber_flags) for use after
reload.
(subsi3, negsi2, andsi3, iorsi3, xorsi3, one_cmplsi2) Likewise.
(ashlsi3, ashlsi3_v850e2, lshrsi3, lsh4si3_v850e2): Likewise.
(ashrsi3, ashrsi3_v850e2): Likewise.
(bins): Clobber the flags.
(movsicc_normal_cc, movsicc_normal, movsicc_tst1): Likewise.
(movsicc_tst1_revesed, sasf, swap and rotate patterns): Likewise.
(fix_loop_counter, call_internal_short, call_internal_long): Likewise.
(call_value_internal_short, call_value_internal_long): Likewise.
(callt_save_interrupt, callt_return_interrupt): Likewise.
(save_interrupt, return_interrupt): Likewise.
(callt_save_all_interrupt, save_all_interrupt): Likewise.
(_save_all_interrupt, callt_restore_all_interrupt): Likewise.
(restore_all_interrupt, _restore_all_interrupt): Likewise.
(All FP comparisons): Only allow after reload has completed.
(trfsr): Likewise.
(divh, divhu): Tweak output template.
(branch_z_normal, branch_z_invert): Remove
(branch_nz_normal, branch_nz_invert): Likewise.
(extendhisi_insn, extendqisi_insn): Do not clobber flags.

Co-Authored-By: Austin Law <austinklaw@gmail.com>
From-SVN: r262190

v850-modes.def (CCZ, CCNZ): Add new modes.

* config/v850/v850-modes.def (CCZ, CCNZ): Add new modes.
* config/v850/v850.c (notice_update_cc): Remove.
* config/v850/v850.h  (CC_OVERFLOW_UNUSABLE): Remove
(CC_NO_CARRY): Likewise.
(NOTICE_UPDATE_CC): Define to nothing.
* config/v850/v850.md: Remove block comment on cc0 handling
Remove "cc" attribute from all patterns.  Remove cc_status handling
from all patterns.  Minor formatting fixes.

Co-Authored-By: Austin Law <austinklaw@gmail.com>
From-SVN: r262189

Avoid crash on friend in nested class template.

* name-lookup.c (do_pushtag): If we skip a class level, also skip
its template level.

From-SVN: r262188

[AArch64] Add support for Arm Cortex-A76

The Cortex-A76 is an Armv8.2-A processor with dotproduct and FP16 support.
It can be paired with the Cortex-A55 and hence the option
-mcpu/-mtune=cortex-a76.cortex-a55 is also introduced.

Bootstrapped and tested on aarch64-none-linux-gnu.

* config/aarch64/aarch64-cores.def (cortex-a76): New entry.
(cortex-a76.cortex-a55): Likewise.
* config/aarch64/aarch64-tune.md: Regenerate.
* doc/invoke.texi (AArch64 Options): Document cortex-a76 and
cortex-a76.cortex-a55.

From-SVN: r262186

t-v850 (MULTILIB_OPTIONS): Remove 8byte-align.

* config/v850/t-v850 (MULTILIB_OPTIONS): Remove 8byte-align.
(MULTILIB_DIRNAMES): Similarly.

From-SVN: r262185

Add std::__is_byte<std::byte> specialization

* include/bits/cpp_type_traits.h [__cplusplus >= 201703]
(__is_byte<byte>): Define specialization for std::byte.

From-SVN: r262182

gimple.h (gimple_return_retbnd): Delete.

* gimple.h (gimple_return_retbnd): Delete.
(gimple_return_set_retbnd): Likewise.
* cgraphunit.c (cgraph_node::expand_thunk): Remove call to
gimple_return_set_retbnd.
* gimple-pretty-print.c (dump_gimple_return): Remove call to
gimple_return_retbnd and adjust.
* tree-inline.h (struct copy_body_data): Remove retbnd field.
* tree-inline.c (remap_gimple_stmt): Remove handling of retbnd.
Explicitly return NULL in a couple more cases.  Move assertion
on debug statements and remove unreachable code.
(reset_debug_binding): Do not test id->retbnd.
(expand_call_inline): Do not set it.

From-SVN: r262181

add support for --disable-gcov

For some targets (in my case VxWorks 5.5), libgcov does not compile due
to missing functions and macros such as getpid() and F_OK.

Incidentally, gcc/Makefile.in already contains comments such as

# Install gcov if it was compiled.

but there is no logic in place to actually allow gcov to not be
compiled.

So add an option for disabling build and install of libgcov and the
related host tools.

From-SVN: r262180

[arm] Add support for Arm Cortex-A76

The Cortex-A76 is an Armv8.2-A processor with dotproduct and FP16 support.
It can be paired with the Cortex-A55 and hence the option
-mcpu/-mtune=cortex-a76.cortex-a55 is also introduced.

Bootstrapped and tested on arm-none-linux-gnueabihf.

* config/arm/arm-cpus.in (cortex-a76): New entry.
(cortex-a76.cortex-a55): Likewise.
* config/arm/arm-tables.opt: Regenerate.
* config/arm/arm-tune.md: Likewise.
* config/arm/driver-arm.c (arm_cpu_table): Add Cortex-A76 entry.
* doc/invoke.texi (ARM Options): Document cortex-a76 and
cortex-a76.cortex-a55.

From-SVN: r262179

Add SIMD to REG pattern for movhf without armv8.2-a support for AArch64

This fixes a regression where we don't have an instruction for pre Armv8.2-a
to do a move of an fp16 value from a GP reg to a SIMD reg.

This patch adds that pattern to movhf_aarch64 using a dup and only selectes it
using a very low priority.

This fixes an ICE at -O0.

gcc/
2018-06-20  Tamar Christina  <tamar.christina@arm.com>

PR target/85769
* config/aarch64/aarch64.md (*movhf_aarch64): Add dup v0.4h pattern.

gcc/testsuite/
2018-06-20  Tamar Christina  <tamar.christina@arm.com>

PR target/85769
* gcc.target/aarch64/f16_mov_immediate_3.c: New.

From-SVN: r262178

fixincludes: Add missing hunk to tests/base/ioLib.h

When adding the vxworks_iolib_include_unistd hack I failed to add the
appropriate hunk to the tests/base/ioLib.h file, causing "make
check-fixincludes" to fail.

From-SVN: r262177

[aarch64] Fix obsolete comment about X30

r217431 changed X30 as caller-saved in CALL_USE_REGISTERS because of
which this comment about X30 not being marked as call-clobbered is no
longer accurate.  Fixed to describe the current state more accurately.

* config/aarch64/aarch64.h (CALL_USE_REGISTERS): Fix obsolete
comment.
(EPILOGUE_USES): Likewise.

From-SVN: r262176

PR c++/86320 - memory-hog with std::array of pair

* typeck2.c (process_init_constructor_array): Only compute a
constant initializer once.

In this PR, we have a large std::array of pairs.  Since the C array is
wrapped in a class we don't go to build_vec_init, so we end up with
digest_init wanting to build up the element initializer for each element of
the array.

In the more general case, like 80272, we have a data structure problem: we
don't currently have a good way of expressing the same dynamic
initialization of many elements within a CONSTRUCTOR.  RANGE_EXPR probably
ought to work, but will need more work at genericize or gimplify time.

But in this case, the initialization for each element reduces to constant
0, so we don't even need to add anything to the CONSTRUCTOR.  We just need
to realize that if the initializer for one element is 0, the others will be
as well, and we don't need to iterate over the whole array.

For the trunk, I also use a RANGE_EXPR to handle constant initialization by
a value other than 0.

void foo ()
{
  std::array<std::pair<int, int>, 1024 * 1024> arr {};
}

From-SVN: r262173

PR c++/80290 - memory-hog with std::pair.

* pt.c (fn_type_unification): Add convs parameter.
(check_non_deducible_conversion): Remember conversion.
(check_non_deducible_conversions): New.  Do checks here.
(type_unification_real): Not here.  Remove flags parm.
* call.c (add_function_candidate): Make convs a parameter.
Don't recalculate the conversion if it's already set.
(add_template_candidate_real): Allocate convs here.
(good_conversion, conv_flags): New.

When the std::pair constructors got more complex to handle, it aggravated a
preexisting algorithmic problem in template overload resolution:

As part of template argument deduction in a call, once we've deduced all
the template arguments we can but before we substitute them to form an
actual declaration, for any function parameters that don't involve template
parameters we need to check that it's possible to convert the argument to
the parameter type (wg21.link/cwg1391).

As a result, we end up calculating the conversion twice: once here, and
then again in add_function_candidate as part of normal overload resolution.
Normally this isn't a big deal, but when the argument is a multiply-nested
initializer list, doubling the conversion processing at each level leads to
combinatorial explosion.

The patch for trunk avoids the duplication by remembering the conversion we
calculate at deduction time and then reusing it in overload resolution
rather than calculating it again.

From-SVN: r262172

Daily bump.

From-SVN: r262171

Declare some explicit instantiations for strings in Debug Mode

The empty reps and the I/O functions do not need to be implicitly
instantiated to enable assertions, so declare the explicit
instantiations when _GLIBCXX_EXTERN_TEMPLATE == -1 (i.e. when
_GLIBCXX_ASSERTIONS is defined).

PR libstdc++/86138
* include/bits/basic_string.tcc: [_GLIBCXX_EXTERN_TEMPLATE < 0]
Declare explicit instantiations of COW empty reps and I/O functions.

From-SVN: r262167

v850-common.c (TARGET_DEFAULT_TARGET_FLAGS): Turn on -mbig-switch by default.

* common/config/v850/v850-common.c (TARGET_DEFAULT_TARGET_FLAGS): Turn
on -mbig-switch by default.

From-SVN: r262166

tree-inline.c (remap_location): New function extracted from...

* tree-inline.c (remap_location): New function extracted from...
(copy_edges_for_bb): Add ID parameter.  Remap goto_locus.
(copy_phis_for_bb): ...here.  Call remap_location.
(copy_cfg_body): Adjust call to copy_edges_for_bb.

From-SVN: r262165

sample-2.cc: Add TLS DejaGNU directives.

        * testsuite/experimental/algorithm/sample-2.cc: Add TLS DejaGNU
        directives.
        * testsuite/experimental/algorithm/shuffle.cc: Likewise.

From-SVN: r262163

builtins-1.c: Correct a comment.

gcc/testsuite/ChangeLog:

2018-06-26  Kelvin Nilsen  <kelvin@gcc.gnu.org>

* gcc.target/powerpc/builtins-1.c: Correct a comment.

From-SVN: r262162

rs6000-string.c (expand_block_clear): Don't use unaligned vsx for 16B memset.

2018-06-26  Aaron Sawdey  <acsawdey@linux.ibm.com>

* config/rs6000/rs6000-string.c (expand_block_clear): Don't use
unaligned vsx for 16B memset.

From-SVN: r262158

I typoed the PR numnber, correct is:

PR target/86285
* config/rs6000/rs6000.c (rs6000_init_builtins): Do not set
ieee128_float_type_node to long_double_type_node unless
TARGET_LONG_DOUBLE_128 is set.

From-SVN: r262156

fold-vec-neg-int.p8.c: Specify powerpc_p8vector_ok requirement for this test.

[testsuite]

2018-06-12  Will Schmidt  <will_schmidt@vnet.ibm.com>

* gcc.target/powerpc/fold-vec-neg-int.p8.c: Specify powerpc_p8vector_ok
requirement for this test.
* gcc.target/powerpc/fold-vec-neg-int.c: Specify powerpc_p8vector_ok
requirement, and -mpower8-vector compile option.

From-SVN: r262153

rs6000: Set up ieee128_float_type_node correctly (PR82625)

We shouldn't init __ieee128 to be the same as long double if the
latter is not even a 128-bit type.

This also reorders the nearby __ibm128 code so both types use similar
logic.

PR target/82625
* config/rs6000/rs6000.c (rs6000_init_builtins): Do not set
ieee128_float_type_node to long_double_type_node unless
TARGET_LONG_DOUBLE_128 is set.

From-SVN: r262152

Introduce dump_location_t

gcc/ChangeLog:
* cfgloop.c (get_loop_location): Convert return type from
location_t to dump_user_location_t, replacing INSN_LOCATION lookups
by implicit construction from rtx_insn *, and using
dump_user_location_t::from_function_decl for the fallback case.
* cfgloop.h (get_loop_location): Convert return type from
location_t to dump_user_location_t.
* cgraphunit.c (walk_polymorphic_call_targets): Update call to
dump_printf_loc to pass in a dump_location_t rather than a
location_t, via the gimple stmt.
* coverage.c (get_coverage_counts): Update calls to
dump_printf_loc to pass in dump_location_t rather than a
location_t.
* doc/optinfo.texi (Dump types): Convert example of
dump_printf_loc from taking "locus" to taking "insn".  Update
description of the "_loc" calls to cover dump_location_t.
* dumpfile.c: Include "backend.h", "gimple.h", "rtl.h", and
"selftest.h".
(dump_user_location_t::dump_user_location_t): New constructors,
from gimple *stmt and rtx_insn *.
(dump_user_location_t::from_function_decl): New function.
(dump_loc): Make static.
(dump_gimple_stmt_loc): Convert param "loc" from location_t to
const dump_location_t &.
(dump_generic_expr_loc): Delete.
(dump_printf_loc): Convert param "loc" from location_t to
const dump_location_t &.
(selftest::test_impl_location): New function.
(selftest::dumpfile_c_tests): New function.
* dumpfile.h: Include "profile-count.h".
(class dump_user_location_t): New class.
(struct dump_impl_location_t): New struct.
(class dump_location_t): New class.
(dump_printf_loc): Convert 2nd param from source_location to
const dump_location_t &.
(dump_generic_expr_loc): Delete.
(dump_gimple_stmt_loc): Convert 2nd param from source_location to
const dump_location_t &.
* gimple-fold.c (fold_gimple_assign): Update call to
dump_printf_loc to pass in a dump_location_t rather than a
location_t, via the gimple stmt.
(gimple_fold_call): Likewise.
* gimple-loop-interchange.cc
(loop_cand::analyze_iloop_reduction_var): Update for change to
check_reduction_path.
(tree_loop_interchange::interchange): Update for change to
find_loop_location.
* graphite-isl-ast-to-gimple.c (scop_to_isl_ast): Update for
change in return-type of find_loop_location.
(graphite_regenerate_ast_isl): Likewise.
* graphite-optimize-isl.c (optimize_isl): Likewise.
* graphite.c (graphite_transform_loops): Likewise.
* ipa-devirt.c (ipa_devirt): Update call to dump_printf_loc to
pass in a dump_location_t rather than a location_t, via the
gimple stmt.
* ipa-prop.c (ipa_make_edge_direct_to_target): Likewise.
* ipa.c (walk_polymorphic_call_targets): Likewise.
* loop-unroll.c (report_unroll): Convert "locus" param from
location_t to dump_location_t.
(decide_unrolling): Update for change to get_loop_location's
return type.
* omp-grid.c (struct grid_prop): Convert field "target_loc" from
location_t to dump_user_location_t.
(grid_find_single_omp_among_assignments_1): Updates calls to
dump_printf_loc to pass in a dump_location_t rather than a
location_t, via the gimple stmt.
(grid_parallel_clauses_gridifiable): Convert "tloc" from
location_t to dump_location_t.  Updates calls to dump_printf_loc
to pass in a dump_location_t rather than a location_t, via the
gimple stmt.
(grid_inner_loop_gridifiable_p): Likewise.
(grid_dist_follows_simple_pattern): Likewise.
(grid_gfor_follows_tiling_pattern): Likewise.
(grid_target_follows_gridifiable_pattern): Likewise.
(grid_attempt_target_gridification): Convert initialization
of local "grid" from memset to zero-initialization; FIXME: does
this require C++11?  Update call to dump_printf_loc to pass in a
optinfo_location rather than a location_t, via the gimple stmt.
* profile.c (read_profile_edge_counts): Updates call to
dump_printf_loc to pass in a dump_location_t rather than a
location_t
(compute_branch_probabilities): Likewise.
* selftest-run-tests.c (selftest::run_tests): Call
dumpfile_c_tests.
* selftest.h (dumpfile_c_tests): New decl.
* tree-loop-distribution.c (pass_loop_distribution::execute):
Update for change in return type of find_loop_location.
* tree-parloops.c (parallelize_loops): Likewise.
* tree-ssa-loop-ivcanon.c (try_unroll_loop_completely): Convert
"locus" from location_t to dump_user_location_t.
(canonicalize_loop_induction_variables): Likewise.
* tree-ssa-loop-ivopts.c (tree_ssa_iv_optimize_loop): Update
for change in return type of find_loop_location.
* tree-ssa-loop-niter.c (number_of_iterations_exit): Update call
to dump_printf_loc to pass in a dump_location_t rather than a
location_t, via the stmt.
* tree-ssa-sccvn.c (eliminate_dom_walker::before_dom_children):
Likewise.
* tree-vect-loop-manip.c (find_loop_location): Convert return
type from source_location to dump_user_location_t.
(vect_do_peeling): Update for above change.
(vect_loop_versioning): Update for change in type of
vect_location.
* tree-vect-loop.c (check_reduction_path): Convert "loc" param
from location_t to dump_user_location_t.
(vect_estimate_min_profitable_iters): Update for change in type
of vect_location.
* tree-vect-slp.c (vect_print_slp_tree): Convert param "loc" from
location_t to dump_location_t.
(vect_slp_bb): Update for change in type of vect_location.
* tree-vectorizer.c (vect_location): Convert from source_location
to dump_user_location_t.
(try_vectorize_loop_1): Update for change in vect_location's type.
(vectorize_loops): Likewise.
(increase_alignment): Likewise.
* tree-vectorizer.h (vect_location): Convert from source_location
to dump_user_location_t.
(find_loop_location): Convert return type from source_location to
dump_user_location_t.
(check_reduction_path): Convert 1st param from location_t to
dump_user_location_t.
* value-prof.c (check_counter): Update call to dump_printf_loc to
pass in a dump_user_location_t rather than a location_t; update
call to error_at for change in type of "locus".
(check_ic_target): Update call to dump_printf_loc to
pass in a dump_user_location_t rather than a location_t, via the
call_stmt.

From-SVN: r262149

Add missing noexcept on definition to match declaration

* include/bits/regex.tcc (regex_iterator::operator==): Add missing
noexcept.

From-SVN: r262146

S/390: Fix mtune default.

When building with --with-tune=zEC12 and calling the resulting gcc
with --march=z13 (no extra -mtune), the binary would unexpectedly be compiled
with -march=z13 -mtune=zEC12.  This patch avoids using the default tune
parameter if -march is specified as argument but the user can still
explicitly state -march=z13 -mtune=zEC12.

gcc/ChangeLog:

2018-06-26  Robin Dapp  <rdapp@linux.vnet.ibm.com>

* config/s390/s390.h (enum processor_flags): Do not use
default tune parameter when -march was specified.

From-SVN: r262143

re PR target/86314 (GCC 7.x and 8.x zero out "eax" before using "rax" in "lock bts")

PR target/86314
* config/i386/i386.md (setcc + movzbl to xor + setcc peephole2s):
Check reg_overlap_mentioned_p in addition to reg_set_p with the same
operands.

* gcc.dg/pr86314.c: New test.

From-SVN: r262141

re PR c++/86291 (OpenMP incorrect for-loop collapsing with iterators and at least 5 nested loops)

PR c++/86291
* parser.c (cp_parser_omp_for_loop_init): Change for_block argument
type from vec<tree, va_gc> * to vec<tree, va_gc> *&.

* testsuite/libgomp.c++/pr86291.C: New test.

From-SVN: r262137

re PR debug/86257 (Program compiled with fPIC crashes while stepping over thread-local variable GDB)

PR debug/86257
* gcc.target/i386/pr86257.c: Add -mtls-dialect=gnu to dg-options.

From-SVN: r262136

Require 64-bit in gcc.target/i386/pr86257.c

* gcc.target/i386/pr86257.c: Require 64-bit.

From-SVN: r262135

Require IBM1047 support in g++.dg/pr86082.C

* g++.dg/pr86082.C: Require IBM1047 support.

From-SVN: r262134

Add myself to MAINTAINERS file

From-SVN: r262133

re PR tree-optimization/86287 (AddressSanitizer: heap-use-after-free on bootstrap with -O3)

2018-06-26  Richard Biener  <rguenther@suse.de>

PR tree-optimization/86287
PR bootstrap/86316
* tree-vect-loop.c (vect_transform_loop_stmt): Fix read-after-free.
(vect_analyze_loop): Initialize n_stmts.

From-SVN: r262132

re PR ipa/86271 (ICE due to size mismatch when inlining)

2018-06-26  Richard Biener  <rguenther@suse.de>

PR middle-end/86271
* fold-const.c (fold_convertible_p): Pointer extension
isn't valid.

* gcc.dg/pr86271.c: New testcase.

From-SVN: r262131

[PR86064] split single cross-partition range with nonzero locviews

We didn't split cross-partition ranges in loclists to output a
whole-function location expression, but with nonzero locviews, we
force loclists, and then we have to split to avoid cross-partition
list entries.

for  gcc/ChangeLog

PR debug/86064
* dwarf2out.c (loc_list_has_views): Adjust comments.
(dw_loc_list): Split single cross-partition range with
nonzero locview.

for  gcc/testsuite/ChangeLog

PR debug/86064
* gcc.dg/pr86064.c: New.

From-SVN: r262130

* lib/target-supports.exp
(check_effective_target_logical_op_short_circuit): Add v850.

From-SVN: r262129

Daily bump.

From-SVN: r262127

predicates.md (const_float_1_operand): Fix match_code test.

* config/v850/predicates.md (const_float_1_operand): Fix match_code
test.
(const_float_0_operand): Remove unused predicate.
* config/v850/v850.md (define_constants): Remove UNSPEC_LOOP.
(define_c_enum unspec): Add LOOP, RCP and RSQRT constants.
(recipsf2): New expander.  Original pattern now called
(recipsf2_insn).
(recipdf2, recipdf2_insn): Similarly.
(rsqrtsf2, rsqrtsf2_insn): Similarly
(rsqrtdf2, rsqrtdf2_insn): Similarly

From-SVN: r262123

stddef.h: Remove an obsolete comment on FreeBSD 5.

* ginclude/stddef.h: Remove an obsolete comment on FreeBSD 5.
Simplify logic for FreeBSD (twice).

From-SVN: r262121

compiler: improve escape analysis

    This CL ports recent enhancements of the escape analysis in the
    gc compiler to gofrontend.

    - CL 99335: unnamed receiver should not escape.

    - CL 105257: propagate loop depth to field. This prevents it
      from escaping when a field's address is taken inside a loop
      (but not otherwise escape).

    - CL 107597: use element type for "indirection" of slice/string.
      This prevents the slice/string from escaping when only the
      element, in case that it is pointerless, flows to outer scope.

    Reviewed-on: https://go-review.googlesource.com/120760

From-SVN: r262120

PR libstdc++/86112 fix printers for Python 2.6

Dict comprehensions are only supported since Python 2.7, so use an
alternative syntax that is backwards compatible.

PR libstdc++/86112
* python/libstdcxx/v6/printers.py (add_one_template_type_printer):
Replace dict comprehension.

From-SVN: r262115

PR tree-optimization/86204 -  wrong strlen result after prior strnlen

gcc/ChangeLog:

PR tree-optimization/86204
* tree-ssa-strlen.c (handle_builtin_strlen): Avoid storing
a strnlen result if it's less than the length of the string.

gcc/testsuite/ChangeLog:

PR tree-optimization/86204
* gcc.dg/strlenopt-46.c: New test.

From-SVN: r262114

Update powerpc64-linux-gnu/baseline_symbols.txt

PR libstdc++/81092
* config/abi/post/powerpc64-linux-gnu/baseline_symbols.txt: Update.

From-SVN: r262112

PR tree-optimization/85700 - Spurious -Wstringop-truncation warning with strncat

gcc/ChangeLog:

PR tree-optimization/85700
* gimple-fold.c (gimple_fold_builtin_strncat): Adjust comment.
* tree-ssa-strlen.c (is_strlen_related_p): Handle integer subtraction.
(maybe_diag_stxncpy_trunc): Distinguish strncat from strncpy.

gcc/testsuite/ChangeLog:

PR tree-optimization/85700
* gcc.dg/Wstringop-truncation-4.c: New test.

From-SVN: r262110

extend.texi (Zero-length arrays): Update and clarify.

gcc/ChangeLog:

* doc/extend.texi (Zero-length arrays): Update and clarify.

From-SVN: r262109

config.gcc (powerpc64le*): Revert January 16th...

2018-06-25  Michael Meissner  <meissner@linux.ibm.com>

* config.gcc (powerpc64le*): Revert January 16th, 2018 patch that
added IEEE/IBM long double multilib support on PowerPC little
endian Linux systems.
* config/rs6000/linux64.h (MULTILIB_DEFAULTS_IEEE): Likewise.
(MULTILIB_DEFAULTS): Likewise.
* config/rs6000/rs6000.c (rs6000_option_override_internal):
Likewise.
* config/rs6000/rs6000.h (TARGET_IEEEQUAD_MULTILIB): Likewise.
* config/rs6000/t-ldouble-linux64le-ibm: Delete, no longer used.
* config/rs6000/t-ldouble-linux64le-ieee: Delete, no longer used.

From-SVN: r262105

Fix -finit-derived for c_ptr and c_funptr in programs which use
    iso_c_binding.

    gcc/fortran/ChangeLog:

    2018-06-25  Fritz Reese  <fritzoreese@gmail.com>

PR fortran/82972
        PR fortran/83088
        PR fortran/85851
* expr.c (component_initializer): Assign init expr to c->initializer.
(generate_isocbinding_initializer): New.
(gfc_generate_initializer): Call generate_isocbinding_initializer to
generate initializers for c_ptr and c_funptr with -finit-derived.

    gcc/testsuite/ChangeLog:

    2018-06-25  Fritz Reese  <fritzoreese@gmail.com>

PR fortran/82972
        PR fortran/83088
        PR fortran/85851
* gfortran.dg/init_flag_17.f90: New testcase.

From-SVN: r262104

gcc_qsort: avoid overlapping memcpy (PR 86311)

PR middle-end/86311
* sort.cc (REORDER_23): Avoid memcpy with same destination and source.
(REORDER_45): Likewise.

From-SVN: r262092

PR libstdc++/86292 fix exception safety of std::vector<InputIterator> constructor

PR libstdc++/86292
* include/bits/stl_vector.h (vector::_M_range_initialize<InputIter>):
Add try-catch block.
* testsuite/23_containers/vector/cons/86292.cc: New.

From-SVN: r262029

* doc/xml/manual/status_cxx2017.xml: Document N4531 status.

From-SVN: r262025

Add experimental::sample and experimental::shuffle from N4531

The additions to <experimental/random> were added in 2015 but the new
algorithms in <experimental/algorithm> were not. This adds them.

* include/experimental/algorithm (sample, shuffle): Add new overloads
using per-thread random number engine.
* testsuite/experimental/algorithm/sample.cc: Simpify and reduce
dependencies by using __gnu_test::test_container.
* testsuite/experimental/algorithm/sample-2.cc: New.
* testsuite/experimental/algorithm/shuffle.cc: New.

From-SVN: r262024

p9-extract-1.c: Make second argument of vec_extract a signed int.

gcc/testsuite/ChangeLog:

2018-06-25  Carl Love  <cel@us.ibm.com>

* gcc.target/powerpc/p9-extract-1.c: Make second argument of
  vec_extract a signed int. Add vec_extract tests for bool char
and bool int.
* gcc.target/powerpc/p9-extract-4.c:  New test file for long long
vec_extract tests.

From-SVN: r262023

v850.md (divmodhi4): Make sure to sign extend the dividend to 32 bits.

* config/v850/v850.md (divmodhi4): Make sure to sign extend the
dividend to 32 bits.  Adjust length.
(udivmodhi4): Cleanup output template.  Fix length.

From-SVN: r262022

[testsuite/guality] Fix tcl error on gdb -v failure

2018-06-25  Tom de Vries  <tdevries@suse.de>

* lib/gcc-gdb-test.exp (report_gdb): Handle gdb -v failure.

From-SVN: r262021

vsx.md: Change word selector to prefered location.

gcc/ChangeLog:

2018-06-25  Carl Love  <cel@us.ibm.com>

* config/rs6000/vsx.md: Change word selector to prefered location.

Signed-off-by: Carl Love <cel@us.ibm.com>
---
 gcc/config/rs6000/vsx.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/rs6000/vsx.md b/gcc/config/rs6000/vsx.md
index a528ef2e8..6e7a4277f 100644
--- a/gcc/config/rs6000/vsx.md
+++ b/gcc/config/rs6000/vsx.md
@@ -3934,7 +3934,7 @@

   operands[5] = gen_rtx_REG (V4SFmode, tmp_regno);
   operands[6] = gen_rtx_REG (V4SImode, tmp_regno);
-  operands[7] = GEN_INT (BYTES_BIG_ENDIAN ? 1 : 2);
+  operands[7] = GEN_INT (BYTES_BIG_ENDIAN ? 0 : 3);
   operands[8] = gen_rtx_REG (V4SImode, reg_or_subregno (operands[0]));
 }
   [(set_attr "type" "vecperm")
--
2.17.1

From-SVN: r262020

re PR regression/86304 ([MIPS] FAIL: gcc.dg/vect/pr79887.c  -mmsa (internal compiler error) starting with r261740)

2018-06-25  Richard Biener  <rguenther@suse.de>

PR tree-optimization/86304
* tree-vectorizer.c (vectorize_loops): Walk over new possibly
epilogue-if-converted loops as well.

From-SVN: r262017

re PR target/84786 ([miscompilation] vunpcklpd accessing xmm16-22 targeting KNL)

PR target/84786
* config/i386/sse.md (vshift_count): New mode attr.
(<shift_insn><mode>3<mask_name>): Use <vshift_count>N instead of vN
as last operand's constraint for VI2_AVX2_AVX512BW shifts.  Use YvN
instead of vN as last operand's constraint for VI48_AVX2 shifts.

* gcc.target/i386/avx512f-pr84786-3.c: New test.

From-SVN: r262015

lto-section-out.c (lto_begin_section): Do not print section name for noaddr and unnumbered dumps.

* lto-section-out.c (lto_begin_section): Do not print section
name for noaddr and unnumbered dumps.

From-SVN: r262013

tree-vectorizer.h (struct vec_info_shared): New structure with parts split out from struct vec_info and loop_nest from...

2018-06-25  Richard Biener  <rguenther@suse.de>

* tree-vectorizer.h (struct vec_info_shared): New structure
with parts split out from struct vec_info and loop_nest from
struct _loop_vec_info.
(struct vec_info): Adjust accordingly.
(struct _loop_vec_info): Likewise.
(LOOP_VINFO_LOOP_NEST): Adjust.
(LOOP_VINFO_DATAREFS): Likewise.
(LOOP_VINFO_DDRS): Likewise.
(struct _bb_vec_info): Likewise.
(BB_VINFO_DATAREFS): Likewise.
(BB_VINFO_DDRS): Likewise.
(struct _stmt_vec_info): Add dr_aux member.
(DR_VECT_AUX): Adjust to refer to member of DR_STMTs vinfo.
(DR_MISALIGNMENT_UNINITIALIZED): New.
(set_dr_misalignment): Adjust.
(dr_misalignment): Assert misalign isn't DR_MISALIGNMENT_UNINITIALIZED.
(vect_analyze_loop): Adjust prototype.
(vect_analyze_loop_form): Likewise.
* tree-vect-data-refs.c (vect_analyze_data_ref_dependences):
Compute dependences lazily.
(vect_record_base_alignments): Use shared datarefs/ddrs.
(vect_verify_datarefs_alignment): Likewise.
(vect_analyze_data_refs_alignment): Likewise.
(vect_analyze_data_ref_accesses): Likewise.
(vect_analyze_data_refs): Likewise.
* tree-vect-loop.c (_loop_vec_info::_loop_vec_info): Add
constructor parameter for shared part.
(vect_analyze_loop_form): Pass in shared part and adjust.
(vect_analyze_loop_2): Pass in storage for the number of
stmts.  Move loop nest finding to the caller.  Compute
datarefs lazily.
(vect_analyze_loop): Pass in shared part.
(vect_transform_loop): Verify shared datarefs are unchanged.
* tree-vect-slp.c (_bb_vec_info::_bb_vec_info): Add
constructor parameter for shared part.
(vect_slp_analyze_bb_1): Pass in shared part and adjust.
(vect_slp_bb): Verify shared datarefs are unchanged before
transform.
* tree-vect-stmts.c (ensure_base_align): Adjust for DR_AUX
change.
(new_stmt_vec_info): Initialize DR_AUX misalignment to
DR_MISALIGNMENT_UNINITIALIZED.
* tree-vectorizer.c (vec_info::vec_info): Add constructor
parameter for shared part.
(vec_info::~vec_info): Adjust.
(vec_info_shared::vec_info_shared): New.
(vec_info_shared::~vec_info_shared): Likewise.
(vec_info_shared::save_datarefs): Likewise.
(vec_info_shared::check_datarefs): Likewise.
(try_vectorize_loop_1): Construct shared part live for analyses
of a single loop for multiple vector sizes.
* tree-parloops.c (gather_scalar_reductions): Adjust.

From-SVN: r262009

tree-vect-data-refs.c (vect_find_stmt_data_reference): Modify DR for SIMD lane accesses here and mark DR with (void *)-1 aux.

2018-06-25  Richard Biener  <rguenther@suse.de>

* tree-vect-data-refs.c (vect_find_stmt_data_reference): Modify
DR for SIMD lane accesses here and mark DR with (void *)-1 aux.
(vect_analyze_data_refs): Remove similar code from here and
simplify accordingly.

From-SVN: r262008

tree-vect-data-refs.c (vect_check_gather_scatter): Fail for reverse storage order accesses rather than asserting they cannot...

2018-06-25  Richard Biener  <rguenther@suse.de>
        Eric Botcazou  <ebotcazou@libertysurf.fr>

* tree-vect-data-refs.c (vect_check_gather_scatter): Fail
for reverse storage order accesses rather than asserting
they cannot happen here.

* gnat.dg/sso9.adb: New testcase.
* gnat.dg/sso9_pkg.ads: Likewise.

From-SVN: r262007

[i386/debug] Fix insn prefix in tls_global_dynamic_64_<mode>

2018-06-25  Tom de Vries  <tdevries@suse.de>

PR debug/86257
* config/i386/i386.md (define_insn "*tls_global_dynamic_64_<mode>"):
Use data16 instead of .byte for insn prefix.

* gcc.target/i386/pr86257.c: New test.

From-SVN: r262006

C++: Fix PR86082

When turning a user-defined numerical literal into an operator
invocation the literal needs to be translated to the execution
character set.

gcc/cp/ChangeLog:

2018-06-25  Andreas Krebbel  <krebbel@linux.ibm.com>

PR C++/86082
* parser.c (make_char_string_pack): Pass this literal chars
through cpp_interpret_string.
(cp_parser_userdef_numeric_literal): Check the result of
make_char_string_pack.

gcc/testsuite/ChangeLog:

2018-06-25  Andreas Krebbel  <krebbel@linux.ibm.com>

PR C++/86082
* g++.dg/pr86082.C: New test.

From-SVN: r262003