My previous change to the Cortex-A53 scheduler resulted in a 13% regression on a proprietary benchmark.

My previous change to the Cortex-A53 scheduler resulted in a 13% regression on a
proprietary benchmark.  This turned out to be due to non-optimal scheduling of int
to float conversions.  This patch separates int to FP transfers from int to float
conversions based on experiments to determine the best schedule.  As a result of
these tweaks the performance of the benchmark improves by 20%.

    gcc/
* config/arm/cortex-a53.md: Add bypasses for
cortex_a53_r2f_cvt.
(cortex_a53_r2f): Only use for transfers.
(cortex_a53_f2r): Likewise.
(cortex_a53_r2f_cvt): Add reservation for conversions.
(cortex_a53_f2r_cvt): Likewise.

From-SVN: r244322

arm_neon.h: Add __artificial__ and gnu_inline to all inlined functions, change static to extern.

2017-01-09  Tamar Christina  <tamar.christina@arm.com>

* config/arm/arm_neon.h: Add __artificial__ and gnu_inline
          to all inlined functions, change static to extern.

From-SVN: r244321

[ARM] PR target/78253 Call weak function instead of strong when called through pointer.

2017-01-11  Christophe Lyon  <christophe.lyon@linaro.org>

PR target/78253
* config/arm/arm.c (legitimize_pic_address): Handle reference to
weak symbol.
(arm_assemble_integer): Likewise.

From-SVN: r244320

PR78134 fix return types of heterogeneous lookup functions

PR libstdc++/78134
* include/bits/stl_map.h (map::lower_bound, map::upper_bound)
(map::equal_range): Fix return type of heterogeneous overloads.
* include/bits/stl_multimap.h (multimap::lower_bound)
(multimap::upper_bound, multimap::equal_range): Likewise.
* include/bits/stl_multiset.h (multiset::lower_bound)
(multiset::upper_bound, multiset::equal_range): Likewise.
* include/bits/stl_set.h (set::lower_bound, set::upper_bound)
(set::equal_range): Likewise.
* testsuite/23_containers/map/operations/2.cc
* testsuite/23_containers/multimap/operations/2.cc
* testsuite/23_containers/multiset/operations/2.cc
* testsuite/23_containers/set/operations/2.cc

From-SVN: r244318

PR78273 fix count to work with partitioning function

PR libstdc++/78273
* include/bits/stl_map.h (map::count<_Kt>(const _Kt&)): Don't assume
the heterogeneous comparison can only find one match.
* include/bits/stl_set.h (set::count<_Kt>(const _Kt&)): Likewise.
* testsuite/23_containers/map/operations/2.cc: Test count works with
comparison function that just partitions rather than sorting.
* testsuite/23_containers/set/operations/2.cc: Likewise.

From-SVN: r244317

[arm] Replace command-line option .def files with single definition file

The files arm-cores.def, arm-fpus.def and arm-arches.def are parsed and
used in several places and the format is slightly awkward to maintain
as they must be parsable in C and by certain scripts.  Furthermore,
changes to the content that affects every entry is particularly awkward for
dealing with merges.

This patch replaces all three files with a single file that specifies all
the command-line related definitions in a new format that allows for better
checking for consistency as well as (hopefully) easier to merge changes.

The awk script used to parse it is relatively complicated, but should be
pretty portable.  It works by parsing in all the data and then operating
one of a number of possible sub-commands to generate the desired output.

The new method picked up one error.  The CPU descriptions referred to an
architecture ARMv5tej which was not supported by -march.  This has been
fixed by adding the relevant entry to the architecture list.

gcc:
* config.gcc: Use new awk script to check CPU, FPU and architecture
parameters for --with-... options.
* config/arm/parsecpu.awk: New file
* config/arm/arm-cpus.in: New file.
* config/arm/arm-opts.h: Include arm-cpu.h instead of processing .def
files.
* config/arm/arm.c: Include arm-cpu-data.h instead of processing .def
files.
* config/arm/t-arm: Update dependency rules.
* common/config/arm/arm-common.c: Include arm-cpu-cdata.h instead
of processing .def files.
* config/arm/genopt.sh: Deleted.
* config/arm/gentune.sh: Deleted.
* config/arm/arm-cores.def: Deleted.
* config/arm/arm-arches.def: Deleted.
* config/arm/arm-fpus.def: Deleted.
* config/arm/arm-tune.md: Regenerated.
* config/arm/arm-tables.opt: Regenerated.
* config/arm/arm-cpu.h: New generated file.
* config/arm/arm-cpu-data.h: New generated file.
* config/arm/arm-cpu-cdata.h: New generated file.

Contrib:
* gcc_update: Adjust touch list.

From-SVN: r244316

re PR lto/79042 (LTO doesn't propagate node->dynamically_initialized bit for varpool nodes.)

PR lto/79042
* lto-cgraph.c (lto_output_varpool_node): Pack dynamically_initialized
bit.
(input_varpool_node): Unpack dynamically_initialized bit.

From-SVN: r244314

Fix flatten-3.c test-case.

2017-01-11  Martin Liska  <mliska@suse.cz>

* gcc.dg/tree-ssa/flatten-3.c: Add -fno-ipa-icf to dg-options.

From-SVN: r244313

decomp4.C (test): Use 2 identifier decomposition instead of just 1 for the decomposition from...

* g++.dg/cpp1z/decomp4.C (test): Use 2 identifier decomposition
instead of just 1 for the decomposition from struct C.

From-SVN: r244312

re PR rtl-optimization/79032 (unaligned memory access generated with LRA and optimization)

PR rtl-optimization/79032
* lra-constraints.c (simplify_operand_subreg): In the MEM case, test
the alignment of the adjusted memory reference against that of MODE,
instead of the alignment of the original memory reference.

From-SVN: r244311

re PR bootstrap/79052 (bootstrap-ubsan failures due to warnings)

2017-01-11  Richard Biener  <rguenther@suse.de>

PR bootstrap/79052
* gimple-parser.c (c_parser_gimple_switch_stmt): Add missing
returns on parse errors.

From-SVN: r244310

Reduce the size of variant, it doesn't need an index of type size_t internally.

Reduce the size of variant, it doesn't need an index of
type size_t internally.
* include/std/variant (parse_numbers.h): New include.
(__select_index): New.
(_Variant_storage<false, _Types...>::_M_reset_impl): Use
_index_type for comparison with variant_npos.
(_Variant_storage<false, _Types...>::__index_type): New.
(_Variant_storage<false, _Types...>::_M_index): Change the
type from size_t to __index_type.
(_Variant_storage<true, _Types...>::__index_type): New.
(_Variant_storage<true, _Types...>::_M_index): Change the
type from size_t to __index_type.
(_Variant_base::_M_valid): Use _Storage::__index_type
for comparison with variant_npos.
(variant::index): Use _Base::_Storage::__index_type
for comparison with variant_npos.
* testsuite/20_util/variant/index_type.cc: New.

From-SVN: r244309

[hsa] Fix hsa function cloning test

2017-01-11  Martin Jambor  <mjambor@suse.cz>

* hsa.c (hsa_callable_function_p): Revert addition of DECL_ARTIFICIAL
test.
* ipa-hsa.c (process_hsa_functions): Only duplicate non-artificial
decorated functions.

From-SVN: r244308

tree-vrp.c (evrp_dom_walker::before_dom_children): Also set range/nonnull info for PHI results.

2017-01-11  Richard Biener  <rguenther@suse.de>

* tree-vrp.c (evrp_dom_walker::before_dom_children): Also
set range/nonnull info for PHI results.  Do not set it on
stmts marked for removal.

* gcc.dg/tree-ssa/pr61743-1.c: Adjust.

From-SVN: r244305

re PR middle-end/50199 (wrong code with -flto -fno-merge-constants)

PR middle-end/50199
* lto-lang.c (lto_post_options): Force flag_merge_constants = 1
if it was 0.

* gcc.dg/lto/pr50199_0.c: New test.

From-SVN: r244304

Daily bump.

From-SVN: r244302

expr.c (store_field): In the bitfield case...

* expr.c (store_field): In the bitfield case, fetch the return value
from the registers before applying a single big-endian adjustment.
Always do a final load for a BLKmode value not larger than a word.

From-SVN: r244299

PR testsuite/78960 - FAIL: gcc.dg/tree-ssa/builtin-sprintf.c execution test

PR testsuite/78960 - FAIL: gcc.dg/tree-ssa/builtin-sprintf.c execution test
PR testsuite/78959 - FAIL: gcc.c-torture/execute/pr78622.c
PR testsuite/78133 - Commit r241489 adds printf specifiers not supported by newlib

gcc/testsuite/ChangeLog:
PR testsuite/78960
PR testsuite/78959
PR testsuite/78133
* gcc.dg/tree-ssa/builtin-sprintf.c: Require C99 runtime.
* gcc.dg/tree-ssa/pr78622.c: Ditto.

From-SVN: r244298

PR middle-end/78138 - missing warnings on buffer overflow with non-constant source length

gcc/testsuite/ChangeLog:

PR middle-end/78138
* gcc.dg/pr78138.c: New test.

From-SVN: r244297

re PR c++/77598 (constexpr compilation failure on reference type casting)

PR c++/77598

* g++.dg/cpp0x/constexpr-ref11.C: New.

From-SVN: r244296

PR middle-end/78245 - missing -Wformat-length on an overflow of a dynamically allocated buffer

gcc/testsuite/ChangeLog:

PR middle-end/78245
* gcc.dg/tree-ssa/builtin-sprintf-warn-3.c: Add tests.

gcc/ChangeLog:

PR middle-end/78245
* gimple-ssa-sprintf.c (get_destination_size): Call
{init,fini}object_sizes.
* tree-object-size.c (addr_object_size): Adjust.
(pass_through_call): Adjust.
(pass_object_sizes::execute): Adjust.
* tree-object-size.h (fini_object_sizes): Declare.

From-SVN: r244294

PR middle-end/78245 - missing -Wformat-length on an overflow of a dynamically allocated buffer

gcc/testsuite/ChangeLog:

PR middle-end/78245
* gcc.dg/tree-ssa/builtin-sprintf-warn-3.c: Add tests.

gcc/ChangeLog:

PR middle-end/78245
* gimple-ssa-sprintf.c (get_destination_size): Call
{init,fini}object_sizes.
* tree-object-size.c (addr_object_size): Adjust.
(pass_through_call): Adjust.
(pass_object_sizes::execute): Adjust.
* tree-object-size.h (fini_object_sizes): Declare.

From-SVN: r244293

Fix issues with unrepresentable column numbers (PR c++/77949)

PR c++/77949 identifies an ICE when the C++ frontend attempts to emit a
fix-it hint inserting a missing semicolon at column 4097 of a source file.

This column value exceeds LINE_MAP_MAX_COLUMN_NUMBER and hence isn't
representable using a location_t.

Attempting to do so leads to these problems, which this patch fixes:

(a) when encountering a column number > LINE_MAP_MAX_COLUMN_NUMBER we
create a new linemap with m_column_and_range_bits == 0, but
linemap_position_for_column doesn't check for this, and hence can emit
a bogus location_t value that's calculated relative to the previous
linemap start, but which will be decoded relative to the new linemap,
leading to very large incorrect line values.

(b) when encountering a column number that can't be represented, and
for which the linemap was pre-existing, the code would hit this assertion:
  if (linemap_assert_fails (column < (1u << map->m_column_and_range_bits)))
around a bail-out condition.  The patch replaces this assertion with a
simple conditional, to stop the ICE when this occurs, and fixes the
bit count (effective column bits, vs column+range bits)

(c) the C++ frontend wasn't checking for failure of
linemap_position_for_loc_and_offset when considering emitting the fix-it
hint.  The patch adds a conditional, so that no fix-it hint is emitted
if the location is bogus.

gcc/cp/ChangeLog:
PR c++/77949
* parser.c (cp_parser_class_specifier_1): Only suggest inserting
a missing semicolon if we have a valid insertion location for
the fix-it hint.

gcc/ChangeLog:
PR c++/77949
* input.c (selftest::test_accessing_ordinary_linemaps): Verify
that we correctly handle column numbers greater than
LINE_MAP_MAX_COLUMN_NUMBER.

gcc/testsuite/ChangeLog:
PR c++/77949
* g++.dg/diagnostic/pr77949.C: New test case.

libcpp/ChangeLog:
PR c++/77949
* line-map.c (linemap_position_for_column): When calling
linemap_start_line, detect if a new linemap was created with
0 column bits, and bail out early if this is the case.
(linemap_position_for_loc_and_offset): Replace overzealous
linemap_assert_fails with a simple conditional; use correct
bit count.

From-SVN: r244292

runtime: copy more scheduler code from Go 1.7 runtime

    I looked at a diff of proc.go between Go 1.7 and gccgo, and copied
    over all the easy stuff.

    Reviewed-on: https://go-review.googlesource.com/35090

From-SVN: r244291

PR tree-optimization/78775 - [7 Regression] ICE in maybe_warn_alloc_args_overflow

gcc/ChangeLog:

PR tree-optimization/78775
* builtins.c (get_size_range): Move...
* calls.c: ...to here.
(alloc_max_size): Accept zero argument.
(operand_signed_p): Remove.
(maybe_warn_alloc_args_overflow): Call get_size_range.
* calls.h (get_size_range): Declare.

gcc/testsuite/ChangeLog:

PR tree-optimization/78775
* gcc.dg/attr-alloc_size-4.c: Add test cases.
* gcc.dg/attr-alloc_size-10.c: New test.
* gcc.dg/attr-alloc_size-11.c: New test.
* gcc.dg/builtin-stringop-chk-7.c: New test.
* gcc.dg/pr78775.c: New test.
* gcc.dg/pr78973-2.c: New test.
* gcc.dg/pr78973.c: New test.

From-SVN: r244290

driver-msp430.c (msp430_mcu_data): Sync with data from TI's devices.csv file as of September 2016.

2017-01-10  Joe Seymour  <joe.s@somniumtech.com>

* config/msp430/driver-msp430.c (msp430_mcu_data): Sync with data
from TI's devices.csv file as of September 2016.
* config/msp430/msp430.c (msp430_mcu_data): Likewise.

From-SVN: r244289

* da.po: Update.

From-SVN: r244288

re PR middle-end/77766 (wrong code at -O2 and -O3 in 64-bit mode on x86_64-linux-gnu (executable hangs))

PR tree-optimization/77766
PR tree-optimization/78856
* gcc.c-torture/execute/pr77766.c: New test.

From-SVN: r244287

extend.texi: Tweak formatting to fix overfull hbox warnings.

2017-01-10  Sandra Loosemore  <sandra@codesourcery.com>

gcc/
* doc/extend.texi: Tweak formatting to fix overfull hbox warnings.
* doc/invoke.texi: Likewise.
* doc/md.texi: Likewise.
* doc/objc.texi: Likewise.

From-SVN: r244286

* c-cppbuiltin.c (c_cpp_builtins): Define __cpp_variadic_using.

From-SVN: r244283

FI 20, decomposition declaration with parenthesized initializer.

* parser.c (cp_parser_decomposition_declaration): Use
cp_parser_initializer.

From-SVN: r244282

fuchsia-elf.h: New file.

2017-01-10  Joshua Conner  <joshconner@google.com>

        * config/arm/fuchsia-elf.h: New file.
        * config/fuchsia.h: New file.
        * config.gcc (*-*-fuchsia*): Set native_system_header_dir.
        (aarch64*-*-fuchsia*, arm*-*-fuchsia*, x86_64-*-fuchsia*): Add to
        targets.
        * config.host: (aarch64*-*-fuchsia*, arm*-*-fuchsia*): Add to hosts.

From-SVN: r244281

Adjust effective target selector to fix C++17 FAIL

* testsuite/18_support/exception_ptr/60612-unexpected.cc: Adjust
effective target selector to prevent running in C++17 mode.

From-SVN: r244280

PR77528 add default constructors for container adaptors

PR libstdc++/77528
* include/bits/stl_queue.h (queue::c): Add default member initializer.
(queue::queue()): Add constructor and define as defaulted.
(queue::queue(_Sequence&&)): Remove default argument.
(priority_queue::c, priority_queue::comp): Add default member
initializers.
(priority_queue::priority_queue()): Add constructor and define as
defaulted.
(priority_queue::priority_queue(const _Compare&, _Sequence&&)):
Remove default argument for first parameter.
* include/bits/stl_stack.h (stack::c): Add default member initializer.
(stack::stack()): Add constructor and define as defaulted.
(stack::stack(const _Sequence&)): Remove default argument.
* testsuite/23_containers/priority_queue/requirements/
explicit_instantiation/1.cc: Test explicit instantiation with
non-DefaultConstructible sequence.
* testsuite/23_containers/priority_queue/77528.cc: New test.
* testsuite/23_containers/priority_queue/requirements/
explicit_instantiation/1_c++0x.cc: Replace with 1_c++98.cc.
* testsuite/23_containers/queue/77528.cc: New test.
* testsuite/23_containers/queue/requirements/explicit_instantiation/
1.cc: Test explicit instantiation with non-DefaultConstructible
sequence.
* testsuite/23_containers/queue/requirements/explicit_instantiation/
1_c++0x.cc: Replace with 1_c++98.cc.
* testsuite/23_containers/stack/77528.cc: New test.
* testsuite/23_containers/stack/requirements/explicit_instantiation/
1.cc: Test explicit instantiation with non-DefaultConstructible
sequence.
* testsuite/23_containers/stack/requirements/explicit_instantiation/
1_c++0x.cc: Replace with 1_c++98.cc.

From-SVN: r244278

* ChangeLog: Fix whitespace.

From-SVN: r244276

re PR sanitizer/78992 (Incorrect sigaction definition on 32-bit sparc)

PR sanitizer/78992
* sanitizer_common/sanitizer_platform_limits_posix.h
(struct __sanitizer_sigaction): Cherry-pick upstream r291561.

From-SVN: r244275

re PR tree-optimization/79034 (error: missing PHI def in verify_gimple_in_cfg)

2016-01-10  Richard Biener  <rguenther@suse.de>

PR tree-optimization/79034
* tree-call-cdce.c (shrink_wrap_one_built_in_call_with_conds):
Propagate out degenerate PHIs in the joiner.

* g++.dg/torture/pr79034.C: New testcase.

From-SVN: r244274

IPA ICF: make algorithm stable to survive -fcompare-debug

2017-01-10  Martin Liska  <mliska@suse.cz>

* gcc.dg/ipa/ipa-icf-1.c: Change scanned pattern.
* gcc.dg/ipa/ipa-icf-10.c: Likewise.
* gcc.dg/ipa/ipa-icf-11.c: Likewise.
* gcc.dg/ipa/ipa-icf-12.c: Likewise.
* gcc.dg/ipa/ipa-icf-13.c: Likewise.
* gcc.dg/ipa/ipa-icf-16.c: Likewise.
* gcc.dg/ipa/ipa-icf-18.c: Likewise.
* gcc.dg/ipa/ipa-icf-2.c: Likewise.
* gcc.dg/ipa/ipa-icf-20.c: Likewise.
* gcc.dg/ipa/ipa-icf-21.c: Likewise.
* gcc.dg/ipa/ipa-icf-23.c: Likewise.
* gcc.dg/ipa/ipa-icf-25.c: Likewise.
* gcc.dg/ipa/ipa-icf-26.c: Likewise.
* gcc.dg/ipa/ipa-icf-27.c: Likewise.
* gcc.dg/ipa/ipa-icf-3.c: Likewise.
* gcc.dg/ipa/ipa-icf-35.c: Likewise.
* gcc.dg/ipa/ipa-icf-36.c: Likewise.
* gcc.dg/ipa/ipa-icf-37.c: Likewise.
* gcc.dg/ipa/ipa-icf-5.c: Likewise.
* gcc.dg/ipa/ipa-icf-7.c: Likewise.
* gcc.dg/ipa/ipa-icf-8.c: Likewise.
* gcc.dg/ipa/pr64307.c: Likewise.
* gcc.dg/ipa/pr77653.c: Likewise.
2017-01-10  Martin Liska  <mliska@suse.cz>

* ipa-icf.c (sort_sem_items_by_decl_uid): New function.
(sort_congruence_classes_by_decl_uid): Likewise.
(sort_congruence_class_groups_by_decl_uid): Likewise.
(sem_item_optimizer::merge_classes): Sort class, groups in these
classes and members in the groups by DECL_UID of declarations.
This would make merge operations stable.

From-SVN: r244273

Revert m_classes_vec introduction.

2017-01-10  Martin Liska  <mliska@suse.cz>

* ipa-icf.c (sem_item_optimizer::sem_item_optimizer): Remove
usage of m_classes_vec.
(sem_item_optimizer::~sem_item_optimizer):  Likewise.
(sem_item_optimizer::get_group_by_hash): Likewise.
(sem_item_optimizer::subdivide_classes_by_equality): Likewise.
(sem_item_optimizer::subdivide_classes_by_sensitive_refs): Likewise.
(sem_item_optimizer::verify_classes): Likewise.
(sem_item_optimizer::process_cong_reduction): Likewise.
(sem_item_optimizer::dump_cong_classes): Likewise.
(sem_item_optimizer::merge_classes): Likewise.
* ipa-icf.h (congruence_class_hash): Rename from
congruence_class_group_hash.  Remove declaration of
m_classes_vec.

From-SVN: r244272

Use temporary int objects to access struct tm members

Call _M_extract_* functions family through temporary int objects, so
it doesn't convert from lvalue to rvalue through a temporary in AVR
because of the incompatible types used in AVR-Libc.

This fixes compilation errors with AVR-Libc while compiling libstdc++
for AVR target.

2017-01-10  Felipe Magno de Almeida <felipe@expertisesolutions.com.br>

* include/bits/locale_facets_nonio.tcc
(time_get::_M_extract_via_format): Avoid compilation errors with
non-standard struct tm.

From-SVN: r244271

Enable AVX-512 VPOPCNTD/VPOPCNTQ instructions.

gcc/
* common/config/i386/i386-common.c (OPTION_MASK_ISA_AVX512VPOPCNTDQ_SET,
OPTION_MASK_ISA_AVX512VPOPCNTDQ_UNSET): New.
* config.gcc: Add avx512vpopcntdqintrin.h.
* config/i386/avx512vpopcntdqintrin.h: New.
* config/i386/cpuid.h (bit_AVX512VPOPCNTDQ): New.
* config/i386/i386-builtin-types.def: Add new types.
* config/i386/i386-builtin.def (__builtin_ia32_vpopcountd_v16si,
__builtin_ia32_vpopcountd_v16si_mask, __builtin_ia32_vpopcountq_v8di,
__builtin_ia32_vpopcountq_v8di_mask): New.
* config/i386/i386-c.c (ix86_target_macros_internal): Define
__AVX512VPOPCNTDQ__.
* config/i386/i386.c (ix86_target_string): Add -mavx512vpopcntdq.
(PTA_AVX512VPOPCNTDQ): Define.
* config/i386/i386.h (TARGET_AVX512VPOPCNTDQ,
TARGET_AVX512VPOPCNTDQ_P): Define.
* config/i386/i386.opt: Add mavx512vpopcntdq.
* config/i386/immintrin.h: Include avx512vpopcntdqintrin.h.
* config/i386/sse.md (define_insn "vpopcount<mode><mask_name>"): New.

libgcc/
* config/i386/cpuinfo.h (processor_features): Add
FEATURE_AVX512VPOPCNTDQ.
* config/i386/cpuinfo.c (get_available_features): Habdle new
feature.

gcc/testsuite/
* g++.dg/other/i386-2.C: Add -mavx512vpopcntdq.
* g++.dg/other/i386-3.C: Ditto.
* gcc.target/i386/sse-12.c: Ditto.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-22.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
* gcc.target/i386/builtin_target.c: Handle new option.
* gcc.target/i386/funcspec-56.inc: Test new attributes.
* gcc.target/i386/avx512vpopcntdq-vpopcntd.c: New test.
* gcc.target/i386/avx512vpopcntdq-vpopcntq.c: Ditto.

From-SVN: r244263

Make Python printers and xmethods work with versioned namespace

2017-01-10  François Dumont  <fdumont@gcc.gnu.org>
    Jonathan Wakely  <jwakely@redhat.com>

* python/libstdcxx/v6/printers.py (_versioned_namespace): Define.
(is_specialization, strip_versioned_namespace): New helpers functions
to work with symbols in the versioned namespace.
(Printer.add_version): Add second name using versioned namespace.
(add_one_template_type_printer, add_one_type_printer): Add second
type printers using versioned namespace.
(register_type_printers): Add template type printer for basic_string.
(build_libstdcxx_dictionary): Remove dead code.
* python/libstdcxx/v6/xmethods.py: Make all matchers look for
versioned namespace.
* testsuite/libstdc++-prettyprinters/48362.cc: Adjust expected
results.
* testsuite/libstdc++-prettyprinters/whatis.cc: Likewise.

Co-Authored-By: Jonathan Wakely <jwakely@redhat.com>
From-SVN: r244262

re PR middle-end/77484 (Static branch predictor causes ~6-8% regression of SPEC2000 GAP)

PR middle-end/77484
* predict.def (PRED_CALL): Set to 67.

From-SVN: r244260

compiler, runtime: drop size arguments to hash/equal functions

    Drop the size arguments for the hash/equal functions stored in type
    descriptors.  Types know what size they are.  To make this work,
    generate hash/equal functions for types that can use an identity
    comparison but are not a standard size and alignment.

    Drop the multiplications by 33 in the generated hash code and the
    reflect package hash code.  They are not necessary since we started
    passing a seed value around, as the seed includes the hash of the
    earlier values.

    Copy the algorithms for standard types from the Go 1.7 runtime,
    replacing the C functions.

    Reviewed-on: https://go-review.googlesource.com/34983

From-SVN: r244256

Daily bump.

From-SVN: r244255

PR testsuite/79036 - gcc.dg/tree-ssa/builtin-sprintf.c fails starting with r244037

PR testsuite/79036 - gcc.dg/tree-ssa/builtin-sprintf.c fails starting with r244037
gcc/testsuite/ChangeLog:
* gcc.dg/tree-ssa/builtin-sprintf.c: Handle extended formats
with different precisions.

From-SVN: r244250

expr.c (store_field): In the bitfield case...

* expr.c (store_field): In the bitfield case, if the value comes from
a function call and is of an aggregate type returned in registers, do
not modify the field mode; extract the value in all cases if the mode
is BLKmode and the size is not larger than a word.

From-SVN: r244249

re PR target/71017 (libgcc/config/i386/cpuinfo.c:346:17: runtime error: left shift of 1 by 31 places cannot be represented in type 'int')

PR target/71017
* config/i386/cpuid.h: Fix undefined behavior.

From-SVN: r244248

re PR tree-optimization/79007 (gcc.dg/tree-ssa/dse-points-to.c fails starting with r244067)

PR tree-optimization/79007
PR tree-optimization/67955
* tree-ssa-alias.c (same_addr_size_stores_p): Only need to be
conservative for pt.null when flag_non_call_exceptions is on.

From-SVN: r244247

Implement P0195R2, C++17 variadic using.

* parser.c (cp_parser_using_declaration): Handle ellipsis and comma.
* pt.c (tsubst_decl): Handle pack expansion in USING_DECL_SCOPE.
* error.c (dump_decl): Likewise.

From-SVN: r244246

re PR translation/79019 (translatable string typo in cif-code.def:141)

PR translation/79019
PR translation/79020
* params.def (PARAM_INLINE_MIN_SPEEDUP,
PARAM_IPA_CP_SINGLE_CALL_PENALTY,
PARAM_USE_AFTER_SCOPE_DIRECT_EMISSION_THRESHOLD): Fix typos
in descriptions.
* config/avr/avr.opt (maccumulate-args): Likewise.
* config/msp430/msp430.opt (mwarn-mcu): Likewise.
* common.opt (freport-bug): Likewise.
* cif-code.def (CIF_FINAL_ERROR): Likewise.
* doc/invoke.texi (ipa-cp-single-call-penalty): Likewise.
* config/s390/s390.c (s390_invalid_binary_op): Fix spelling in
translatable string.
* config/i386/i386.c (function_value_32): Likewise.
* config/nios2/nios2.c (nios2_valid_target_attribute_rec): Likewise.
* config/msp430/msp430.c (msp430_option_override, msp430_attr):
Likewise.
* config/msp430/driver-msp430.c (msp430_select_hwmult_lib): Likewise.
* common/config/msp430/msp430-common.c (msp430_handle_option):
Likewise.
* symtab.c (symtab_node::verify_base): Likewise.
* opts.c (set_debug_level): Likewise.
* tree.c (verify_type_variant): Likewise.  Fix typo in comment.
* config/rs6000/rs6000-c.c (altivec_resolve_overloaded_builtin): Add
missing whitespace to translatable strings.
* config/avr/avr.md (bswapsi2): Fix typo in comment.
* config/sh/superh.h: Likewise.
* config/i386/xopintrin.h: Likewise.
* config/i386/znver1.md: Likewise.
* config/rs6000/rs6000.c (struct rs6000_opt_mask): Likewise.
* ipa-inline-analysis.c (compute_inline_parameters): Likewise.
* double-int.h (struct double_int): Likewise.
* double-int.c (div_and_round_double): Likewise.
* wide-int.cc: Likewise.
* tree-ssa.c (non_rewritable_mem_ref_base): Likewise.
* tree-ssa-sccvn.c (vn_reference_lookup_3): Likewise.
* cfgcleanup.c (crossjumps_occured): Renamed to ...
(crossjumps_occurred): ... this.
(try_crossjump_bb, try_head_merge_bb, try_optimize_cfg, cleanup_cfg):
Adjust all uses.
cp/
* semantics.c (finish_omp_clauses): Add missing whitespace to
translatable strings.
* cp-cilkplus.c (cpp_validate_cilk_plus_loop_aux): Fix comment typo.
lto/
* lto-symtab.c (lto_symtab_merge_symbols): Fix comment typo.
fortran/
* decl.c (attr_decl1): Fix spelling in translatable string.
* intrinsic.texi: Fix spelling - invokation -> invocation.
* lang.opt (faggressive-function-elimination, gfc_convert): Fix
typos in descriptions.
* openmp.c (resolve_omp_clauses): Add missing whitespace to
translatable strings.
c-family/
* c.opt (Wnormalized=): Fix typo in description.
testsuite/
* c-c++-common/goacc/host_data-2.c (f): Adjust expected spelling of
diagnostics.
* gfortran.dg/initialization_17.f90: Likewise.

From-SVN: r244245

re PR testsuite/79026 (The tests changed by revision r244006 now fail on darwin)

testsuite/ChangeLog:

PR testsuite/79026
* gcc.target/i386/pr78904-2.c: Require nonpic targets.
* gcc.target/i386/pr78904-4.c: Likewise.
* gcc.target/i386/pr78904-6.c: Likewise.
* gcc.target/i386/pr78967-2.c: Likewise.

From-SVN: r244244

re PR libgomp/60670 (omp.h may differ between multilibs)

PR libgomp/60670
* Makefile.am: Make fincludedir multilib-aware.
* Makefile.in: Regenerate.

From-SVN: r244239

re PR tree-optimization/78899 (Vestorized loop with optmized mask stores motion is completely deleted after r242520.)

PR tree-optimization/78899
* tree-if-conv.c (version_loop_for_if_conversion): Instead of
returning bool return struct loop *, NULL for failure and the new
loop on success.
(versionable_outer_loop_p): Don't version outer loop if it has
dont_vectorized bit set.
(tree_if_conversion): When versioning outer loop, ensure
tree_if_conversion is performed also on the inner loop of the
non-vectorizable outer loop copy.
* tree-vectorizer.c (set_uid_loop_bbs): Formatting fix.  Fold
LOOP_VECTORIZED in inner loop of the scalar outer loop and
prevent vectorization of it.
(vectorize_loops): For outer + inner LOOP_VECTORIZED, ensure
the outer loop vectorization of the non-scalar version is attempted
before vectorization of the inner loop in scalar version.  If
outer LOOP_VECTORIZED guarded loop is not vectorized, prevent
vectorization of its inner loop.
* tree-vect-loop-manip.c (rename_variables_in_bb): If outer_loop
has 2 inner loops, rename also on edges from bb whose single pred
is outer_loop->header.  Fix typo in function comment.

* gcc.target/i386/pr78899.c: New test.
* gcc.dg/pr71077.c: New test.

From-SVN: r244238

PR bootstrap/79033 - asan.c not compiling with make BOOT_CFLAGS=-O0

gcc/ChangeLog:
* asan.c (asan_emit_stack_protection): Increase local buffer size
to avoid snprintf truncation warning.

From-SVN: r244237

runtime: copy more of scheduler from Go 1.7 runtime

    This started by moving procresize from C to Go so that we can pass the
    right type to the memory allocator when allocating a p, which forced
    the gomaxprocs variable to move from C to Go, and everything else
    followed from that.

    Reviewed-on: https://go-review.googlesource.com/34916

From-SVN: r244236

aarch64-cores.def: Add thunderx2t99.

2017-01-09  Andrew Pinski  <apinski@cavium.com>

        * config/aarch64/aarch64-cores.def: Add thunderx2t99.  Change vulcan
        to reference thunderx2t99 for the tuning structure
        * config/aarch64/aarch64-cost-tables.h (vulcan_extra_costs): Rename to ...
        (thunderx2t99_extra_costs): This.
        * config/aarch64/aarch64-tune.md: Regenerate.
        * config/aarch64/aarch64.c (vulcan_addrcost_table): Rename to ...
        (vulcan_addrcost_table): This.
        (vulcan_regmove_cost): Rename to ...
        (thunderx2t99_regmove_cost): This.
        (vulcan_vector_cost): Rename to ...
        (thunderx2t99_vector_cost): this.
        (vulcan_branch_cost): Rename to ...
        (thunderx2t99_branch_cost): This.
        (vulcan_tunings): Rename to ...
        (thunderx2t99_tunings): This and s/vulcan/thunderx2t99 .
        * doc/invoke.texi (AARCH64/mtune): Add thunderx2t99.

From-SVN: r244235

Fix formating.

From-SVN: r244234

[PR 78365] Prudent type handling in IPA VR-prop

2017-01-09  Martin Jambor  <mjambor@suse.cz>

PR ipa/78365
PR ipa/78599
* ipa-prop.h (ipa_jump_func): Swap positions of vr_known and m_vr.
* ipa-cp.c (ipa_vr_operation_and_type_effects): New function.
(propagate_vr_accross_jump_function): Use the above function for all
value range computations for pass-through jump functions and type
converasion from explicit value range values.
        (ipcp_propagate_stage): Do not attempt to deduce types of formal
        parameters from TYPE_ARG_TYPES.
* ipa-prop.c (ipa_write_jump_function): Remove trailing whitespace.
(ipa_write_node_info): Stream type of the actual argument.
(ipa_read_node_info): Likewise. Also remove trailing whitespace.

testsuite/
* gcc.dg/torture/pr78365.c: New test.

From-SVN: r244232

PR79017 workaround incomplete C99 math on darwin

PR libstdc++/79017
* acinclude.m4 (GLIBCXX_CHECK_C99_TR1): Check for llrint and llround
functions separately on darwin and if they're missing define
_GLIBCXX_NO_C99_ROUNDING_FUNCS.
* config.h.in: Regenerate.
* configure: Regenerate.
* include/c_global/cmath [_GLIBCXX_NO_C99_ROUNDING_FUNCS] (llrint)
(llrintf, llrintl, llround, llroundf, llroundl): Do not define.

From-SVN: r244231

* vi.po: Update.

From-SVN: r244230

* da.po, es.po, sv.po, vi.po: Update.

From-SVN: r244229

re PR c++/78948 ([C++17] constexpr if instantiating too eagerly)

PR c++/78948
* g++.dg/cpp1z/constexpr-if10.C: Fix PR number in comment.
(main): Use int* instead of long long.

From-SVN: r244228

Do not error when -E provided (PR pch/78970).

2017-01-09  Martin Liska  <mliska@suse.cz>

PR pch/78970
* gcc.c (driver_handle_option): Handle OPT_E and set
have_E.
(lookup_compiler): Do not show error message with have_E.

From-SVN: r244227

Define testsuite macro for correct thread_local destructors

* testsuite/30_threads/condition_variable/members/3.cc: Use new macro
to detect correct thread_local destructors.
* testsuite/util/testsuite_hooks.h (CORRECT_THREAD_LOCAL_DTORS):
Define.

From-SVN: r244226

PR66414 optimize std::string::find

2017-01-09  Jonathan Wakely  <jwakely@redhat.com>
    Aditya Kumar  <hiraditya@msn.com>

PR libstdc++/66414
* include/bits/basic_string.tcc
(basic_string::find(const CharT*, size_type, size_type)): Optimize.

Co-Authored-By: Aditya Kumar <hiraditya@msn.com>
From-SVN: r244225

einfo.ads, einfo.adb: Remove uses of flags Has_Default_Init_Cond...

2017-01-09  Hristian Kirtchev  <kirtchev@adacore.com>

* einfo.ads, einfo.adb: Remove uses of flags Has_Default_Init_Cond,
Is_Default_Init_Cond_Procedure, and
Has_Inherited_Default_Init_Cond.  Add uses of flags
Has_Own_DIC, Is_DIC_Procedure, and Has_Inherited_DIC.
(Default_Init_Cond_Procedure): Removed.
(DIC_Procedure): New routine.
(Has_Default_Init_Cond): Removed.
(Has_DIC): New routine.
(Has_Inheritable_Invariants): The attribute applies to the base type.
(Has_Inherited_Default_Init_Cond): Removed.
(Has_Inherited_DIC): New routine.
(Has_Inherited_Invariants): The attribute applies to the base type.
(Has_Own_DIC): New routine.
(Has_Own_Invariants): The attribute applies to the base type.
(Is_Default_Init_Cond_Procedure): Removed.
(Is_DIC_Procedure): New routine.
(Set_Default_Init_Cond_Procedure): Removed.
(Set_DIC_Procedure): New routine.
(Set_Has_Default_Init_Cond): Removed.
(Set_Has_Inheritable_Invariants): The attribute applies
to the base type.
(Set_Has_Inherited_Default_Init_Cond): Removed.
(Set_Has_Inherited_DIC): New routine.
(Set_Has_Inherited_Invariants): The attribute applies to the base type.
(Set_Has_Own_DIC): New routine.
(Set_Has_Own_Invariants): The attribute applies to the base type.
(Set_Is_Default_Init_Cond_Procedure): Removed.
(Set_Is_DIC_Procedure): New routine.
(Write_Entity_Flags): Update the output of all flags related to
default initial condition.
* exp_ch3.adb (Expand_N_Object_Declaration): Update the generation
of the call to the DIC procedure.
(Freeze_Type): Generate the body of the DIC procedure.
* exp_ch7.adb (Build_Invariant_Procedure_Body): Replace
all occurrences of Create_Append with Append_New_To. Do
not generate an invariant procedure for a class-wide type.
The generated body acts as a freeze action of the working type.
(Build_Invariant_Procedure_Declaration): Do not generate an
invariant procedure for a class-wide type.
(Create_Append): Removed.
* exp_util.adb: Add with and use clauses for Sem_Ch3, sem_ch6,
sem_Ch12, Sem_Disp, and GNAT.HTable. Move the handling of
class-wide pre/postcondition description and data structures from
Sem_Prag.
(Build_Class_Wide_Expression): Moved from Sem_Prag.
(Build_DIC_Call): New routine.
(Build_DIC_Procedure_Body): New routine.
(Build_DIC_Procedure_Declaration): New routine.
(Entity_Hash): Moved from Sem_Prag.
(Find_DIC_Type): New routine.
(Update_Primitives_Mapping): Reimplemented.
(Update_Primitives_Mapping_Of_Types): New routine.
* exp_util.ads (Build_Class_Wide_Expression): Moved from Sem_Prag.
(Build_DIC_Call): New routine.
(Build_DIC_Procedure_Body): New routine.
(Build_DIC_Procedure_Declaration): New routine.
(Update_Primitives_Mapping): Moved from Sem_Prag.
(Update_Primitives_Mapping_Of_Types): New routine.
* nlists.adb (Append_New): New routine.
(Append_New_To): New routine.
* nlists.ads (Append_New): New routine.
(Append_New_To): New routine.
* sem_ch3.adb (Analyze_Declarations): Do not generate the bodies
of DIC procedures here. This is now done at the end of the
visible declarations, private declarations, and at the freeze
point of a type.
(Analyze_Private_Extension_Declaration):
A private extension inherits the DIC pragma of a parent type.
(Analyze_Subtype_Declaration): No need to propagate invariant
attributes to a subtype as those apply to the base type.
(Build_Derived_Record_Type): No need to inherit invariants here
as this is now done in Build_Derived_Type.
(Build_Derived_Type): Inherit both the DIC pragma and invariants from
a parent type.
(Process_Full_View): Update the propagation of DIC attributes.
(Propagate_Default_Init_Cond_Attributes): Removed.
* sem_ch7.adb Add with and use clauses for Exp_Util.
(Analyze_Package_Specification): Create the body of the DIC
procedure at the end of the visible and private declarations.
(Preserve_Full_Attributes): Propagate DIC attributes.
* sem_ch9.adb (Analyze_Protected_Type_Declaration): Propagate
DIC attributes.
(Analyze_Task_Type_Declaration): Propagate DIC attributes.
* sem_elab.adb (Check_A_Call): Update the call to
Is_Nontrivial_Default_Init_Cond_Procedure.
* sem_prag.adb Remove the with and use clauses for
GNAT.HTable. Move the handling of class- wide pre/postcondition
description and data structures to Exp_Util.
(Analyze_Pragma): Create the declaration of the DIC procedure. There
is no need to propagate invariant-related attributes at this point
as this is done in Build_Invariant_Procedure_Declaration.
(Build_Class_Wide_Expression): Moved to Exp_Util.
(Entity_Hash): Moved to Exp_Util.
(Update_Primitives_Mapping): Moved to Exp_Util.
* sem_prag.ads (Build_Class_Wide_Expression): Moved to Exp_Util.
(Update_Primitives_Mapping): Moved to Exp_Util.
* sem_util.adb: Remove with and use clauses for Ghost
and Sem_Ch13.
(Build_Default_Init_Cond_Call): Removed.
(Build_Default_Init_Cond_Procedure_Bodies): Removed.
(Build_Default_Init_Cond_Procedure_Declaration): Removed.
(Get_Views): Reimplemented.
(Has_Full_Default_Initialization): Reimplement the section on DIC.
(Inherit_Default_Init_Cond_Procedure): Removed.
(Is_Nontrivial_Default_Init_Cond_Procedure): Removed.
(Is_Nontrivial_DIC_Procedure): New routine.
(Is_Verifiable_DIC_Pragma): New routine.
(Propagate_DIC_Attributes): New routine.
* sem_util.ads (Build_Default_Init_Cond_Call): Removed.
(Build_Default_Init_Cond_Procedure_Bodies): Removed.
(Build_Default_Init_Cond_Procedure_Declaration): Removed.
(Inherit_Default_Init_Cond_Procedure): Removed.
(Is_Nontrivial_Default_Init_Cond_Procedure): Removed.
(Is_Nontrivial_DIC_Procedure): New routine.
(Is_Verifiable_DIC_Pragma): New routine.
(Propagate_DIC_Attributes): New routine.
* sem_warn.adb (Is_OK_Fully_Initialized): Reimplement the section
on DIC.
* sinfo.ads, sinfo.adb: Add new attribute Expression_Copy along with
usage in nodes.
(Expression_Copy): New routine along with pragma Inline.
(Set_Expression_Copy): New routine along with pragma Inline.

From-SVN: r244224

re PR tree-optimization/78938 (ICE in expand_vec_cond_expr, at optabs.c:5636 w/ -mavx512bw -ftree-loop-vectorize -O1)

PR tree-optimization/78938
* tree-vect-stmts.c (vectorizable_condition): For non-masked COND_EXPR
where comp_vectype is VECTOR_BOOLEAN_TYPE_P, use
BIT_{NOT,XOR,AND,IOR}_EXPR on the comparison operands instead of
{EQ,NE,GE,GT,LE,LT}_EXPR directly inside of VEC_COND_EXPR.  Formatting
fixes.

* gcc.dg/vect/pr78938.c: New test.

From-SVN: r244223

[tree-ssa-address] Avoid creating non-canonical RTL in gen_addr_rtx

* tree-ssa-address.c (gen_addr_rtx): Don't handle index if it
is const0_rtx.

* gcc.dg/20161219.c: New test.

From-SVN: r244219

re PR tree-optimization/78997 (ICE on valid code at -O3 on x86_64-linux-gnu: verify_gimple failed)

2017-01-09  Richard Biener  <rguenther@suse.de>

PR tree-optimization/78997
* tree-vect-slp.c (vect_mask_constant_operand_p): Handle SSA
name condition properly.

* gcc.dg/torture/pr78997.c: New testcase.

From-SVN: r244218

re PR debug/79000 (ICE: in gen_member_die, at dwarf2out.c:23995)

2017-01-09  Richard Biener  <rguenther@suse.de>

PR debug/79000
* dwarf2out.c (is_cxx): New overload with context.
(is_naming_typedef_decl): Use it.

* g++.dg/lto/pr79000_0.C: New testcase.
* g++.dg/lto/pr79000_1.c: Likewise.

From-SVN: r244217

invoke.texi (Option Summary): Correct spacing in option lists and add line breaks to fix over-long lines.

2017-01-08  Sandra Loosemore  <sandra@codesourcery.com>

gcc/
* invoke.texi (Option Summary): Correct spacing in option lists
and add line breaks to fix over-long lines.

From-SVN: r244216

re PR middle-end/17660 (manual's description of __attribute__((mode)) is poor, arrangement is confusing)

2017-01-08  Sandra Loosemore  <sandra@codesourcery.com>

PR middle-end/17660

gcc/
* extend.texi (Common Variable Attributes): Add xref to GCC
Internals manual to explain mode attribute keywords.

From-SVN: r244215

re PR other/16519 (-pthread undocumented)

2017-01-08  Sandra Loosemore  <sandra@codesourcery.com>

PR other/16519

gcc/
* doc/invoke.texi (Option Summary): Move -pthread to Linker Options
and Preprocessor Options.
(Options for Linking): Document -pthread here....
(RS/6000 and PowerPC Options): ...not here.
(Solaris 2 Options): ...or here.
* doc/cppopts.texi: Document -pthread.

From-SVN: r244214

Daily bump.

From-SVN: r244213

PR tree-optimization/78913 - Probably misleading error reported by -Wformat-length

PR tree-optimization/78913 - Probably misleading error reported by -Wformat-length
PR middle-end/77708 - -Wformat-length %s warns for snprintf

gcc/ChangeLog:

PR middle-end/77708
* doc/invoke.texi (Warning Options): Document -Wformat-truncation.
* gimple-ssa-sprintf.c (call_info::reval_used, call_info::warnopt):
New member functions.
(format_directive): Used them.
(add_bytes): Same.
(pass_sprintf_length::handle_gimple_call): Same.
* graphite-sese-to-poly.c (tree_int_to_gmp): Increase buffer size
to avoid truncation for any argument.
(extract_affine_mul): Same.
* tree.c (get_file_function_name): Same.

gcc/c-family/ChangeLog:

PR middle-end/77708
* c.opt (-Wformat-truncation): New option.

gcc/fortran/ChangeLog:

PR tree-optimization/78913
PR middle-end/77708
* trans-common.c (build_equiv_decl): Increase buffer size to avoid
truncation for any argument.
* trans-types.c (gfc_build_logical_type): Same.

gcc/testsuite/ChangeLog:

PR middle-end/77708
* gcc.dg/tree-ssa/builtin-snprintf-warn-1.c: New test.
* gcc.dg/tree-ssa/builtin-snprintf-warn-2.c: New test.
* gcc.dg/tree-ssa/builtin-sprintf-warn-6.c: XFAIL test cases failing
due to bug 78969.
* gcc.dg/format/pr78569.c: Adjust.

From-SVN: r244210

* download_ecj: Remove.

From-SVN: r244209

re PR rtl-optimization/79003 (r238991 breaks ODR)

PR bootstrap/79003
* Makefile.in (CFLAGS, CPPFLAGS, LDFLAGS): Remove -fno-lto.
(NOLTO_FLAGS): New variable.
(ALL_CFLAGS): Use it.
* configure.ac (nolto_flags): New ACX_PROG_CC_WARNING_OPTS,
check for whether -fno-lto works.
* configure: Regenerated.

From-SVN: r244208

re PR middle-end/77484 (Static branch predictor causes ~6-8% regression of SPEC2000 GAP)

PR middle-end/77484
* predict.def (PRED_INDIR_CALL): Set to 86.

From-SVN: r244207

PR c++/78948 - instantiation from discarded statement

PR c++/78948 - instantiation from discarded statement
* parser.h (struct cp_parser): Remove in_discarded_stmt field.
* cp-tree.h (in_discarded_stmt): Declare it.
(struct saved_scope): Add discarded_stmt bitfield.
(in_discarded_stmt): New macro.
* decl2.c (mark_used): Check it.
* parser.c (cp_parser_selection_statement): Adjust.
(cp_parser_jump_statement): Adjust.

From-SVN: r244206

re PR preprocessor/54124 (GCC 4.7.1 manual:  -dM and similar options hard to find)

2017-01-07  Sandra Loosemore  <sandra@codesourcery.com>

PR preprocessor/54124

gcc/
* doc/cppopts.texi: Reformat -d subtable to list the full name
of the options.  Add cross-reference to the docs for the general
compiler -d options.
* doc/invoke.texi (Developer Options): Add cross-reference to the
preprocessor-specific -d option documentation.

From-SVN: r244205

Daily bump.

From-SVN: r244204

re PR preprocessor/13498 (CPP include search path documentation is incorrect)

2017-01-07  Sandra Loosemore  <sandra@codesourcery.com>

PR preprocessor/13498

gcc/
* doc/cpp.texi (Search Path): Rewrite to remove obsolete and
redudant material, and reflect new command-line options.
(System Headers): Likewise.

From-SVN: r244201

cppdiropts.texi: Merge documentation of -I, -iquote, -isystem, and -idirafter.

2017-01-07  Sandra Loosemore  <sandra@codesourcery.com>

gcc/
* doc/cppdiropts.texi: Merge documentation of -I, -iquote,
-isystem, and -idirafter.  Copy-edit.
* doc/cppopts.texi: Copy-edit.  Remove contradiction about
default for -ftrack-macro-expansion.  Delete obsolete and
badly-formatted implementation details about -fdebug-cpp output.
* doc/cppwarnopts.texi: Copy-edit.

From-SVN: r244200

Fix linemap corruption after very wide source lines (PR c++/72803)

PR c++/72803 describes an issue where a fix-it hint is to be emitted at
column 512 of a 511-column source line, leading to an ICE.

The root cause is a bug in linemap_line_start, when transitioning from
lines >= 512 in width to narrow lines.

The wide line in the reproducer has a line map with:
  m_column_and_range_bits = 15, m_range_bits = 5
giving 10 effective bits for representing columns, so that columns <= 1023
can be represented.

When parsing the following line,
  linemap_line_start (..., ..., max_column_hint=0);
is called.  This leads to the "add_map" logic, due to this condition:
      || (max_column_hint <= 80 && effective_column_bits >= 10)
i.e. the new line is sufficiently narrower than the old one to
potentially use a new linemap (so as to conserve values within the
location_t space).

It then attempts to avoid allocating a new line map.  Part of the logic
to determine if we really need a new line map is this condition:
   SOURCE_COLUMN (map, highest) >= (1U << column_bits)
The above condition is incorrect: we need to determine if the highest
column we've handed out will fit within the proposed *effective* column
bits, but "column_bits" here is the column plus the range bits, rather
than just the column bits.

Hence in this case linemap_line_start erroneously decides that we don't
need a new line map, and updates the column bits within the existing
line map, so any location_t values we've already handed out within it
that are offset from the start by
  >= (1<<new_column_and_range_bits)
effectively change meaning, leading to incorrect line&column information
when decoding them, and various "interesting" ways for the linemap
code to fail.

The fix is to use the effective column bits in the above conditional.

gcc/ChangeLog:
PR c++/72803
* input.c (selftest::test_accessing_ordinary_linemaps): Verify
that the transition from a max line width >= 1<<10 to narrower
lines works correctly.

gcc/testsuite/ChangeLog:
PR c++/72803
* g++.dg/diagnostic/pr72803.C: New test case.

libcpp/ChangeLog:
PR c++/72803
* line-map.c (linemap_line_start): When determining if the highest
column given out so far will fit into a proposed change to the
current map, use the effective number of column bits, rather than
the total number of column + range bits.

From-SVN: r244199

[-fcompare-debug] var tracking options are not optimization options

If we include them in the ICF hash, they may cause congruence_groups to
be processed in a different order due to different hashes, which in turn
causes different funcdef_nos to be assigned to functions.  Since these
numbers are included in -fcompare-debug dumps, they cause failures.

Since these options are not optimization options, in that they do not
(or should not, save for bugs like this) affect the executable code
output by the compiler, they should not be marked as such.

This patch replaces the Optimization flag in the var-tracking options
with the newly-introduced PerFunction flag, so that it can still be
controlled on a per-function basis, but that disregards it in the hash
computation used by ICF.

This fixes -fcompare-debug failures in numerous LTO testcases.

for  gcc/ChangeLog

* doc/options.texi (PerFunction): New.
* opt-functions.awk (switch_flags): Map both Optimization and
PerFunction to CL_OPTIMIZATION.
* opth-gen.awk: Test for PerFunction flag along with
Optimization.
* optc-save-gen.awk: Likewise.  Introduce var_opt_hash and set
it only when the latter is present.  Skip those that don't in
the hash function generator.
* common.opt (fvar-tracking): Mark as PerFunction instead of
Optimization.
(fvar-tracking-assignments): Likewise.
(fvar-tracking-assignments-toggle): Likewise.
(fvar-tracking-uninit): Likewise.

From-SVN: r244198

attr-alloc_size-3.c: Remove regex made redundant by r243461.

gcc/testsuite/ChangeLog:

* gcc.dg/attr-alloc_size-3.c: Remove regex made redundant by r243461.
* gcc.dg/attr-alloc_size-4.c: Same.

From-SVN: r244197

re PR fortran/78781 ([Coarray] ICE in gfc_deallocate_scalar_with_status, at fortran/trans.c:1588)

gcc/fortran/ChangeLog:

2017-01-07  Andre Vehreschild  <vehre@gcc.gnu.org>

PR fortran/78781
PR fortran/78935
* expr.c (gfc_check_pointer_assign): Return the same error message for
rewritten coarray pointer assignments like for plain ones.
* gfortran.h: Change prototype.
* primary.c (caf_variable_attr): Set attributes used ones only only
ones.  Add setting of pointer_comp attribute.
(gfc_caf_attr): Add setting of pointer_comp attribute.
* trans-array.c (gfc_array_allocate): Add flag that the component to
allocate is not an ultimate coarray component.  Add allocation of
pointer arrays.
(structure_alloc_comps): Extend nullify to treat pointer components in
coarrays correctly.  Restructure nullify to remove redundant code.
(gfc_nullify_alloc_comp): Allow setting caf_mode flags.
* trans-array.h: Change prototype of gfc_nullify_alloc_comp ().
* trans-decl.c (generate_coarray_sym_init): Call nullify_alloc_comp for
derived type coarrays with pointer components.
* trans-expr.c (gfc_trans_structure_assign): Also treat pointer
components.
(trans_caf_token_assign): Handle assignment of token of scalar pointer
components.
(gfc_trans_pointer_assignment): Call above routine.
* trans-intrinsic.c (conv_expr_ref_to_caf_ref): Add treating pointer
components.
(gfc_conv_intrinsic_caf_get): Likewise.
(conv_caf_send): Likewise.
* trans-stmt.c (gfc_trans_allocate): After allocating a derived type in
a coarray pre-register the tokens.
(gfc_trans_deallocate): Simply determining the coarray type (scalar or
array) and deregistering it correctly.
* trans-types.c (gfc_typenode_for_spec): Replace in_coarray flag by the
actual codim to allow lookup of array types in the cache.
(gfc_build_array_type): Likewise.
(gfc_get_array_descriptor_base): Likewise.
(gfc_get_array_type_bounds): Likewise.
(gfc_get_derived_type): Likewise.
* trans-types.h: Likewise.
* trans.c (gfc_deallocate_with_status): Enable deregistering of all kind
of coarray components.
(gfc_deallocate_scalar_with_status): Use free() in fcoarray_single mode
instead of caf_deregister.

libgfortran/ChangeLog:

2017-01-07  Andre Vehreschild  <vehre@gcc.gnu.org>

PR fortran/78781
PR fortran/78935
* caf/single.c (send_by_ref): Fix addressing of non-allocatable scalar
destination components.

gcc/testsuite/ChangeLog:

2017-01-07  Andre Vehreschild  <vehre@gcc.gnu.org>

* gfortran.dg/coarray/ptr_comp_1.f08: New test.
* gfortran.dg/coarray/ptr_comp_2.f08: New test.
* gfortran.dg/coarray/ptr_comp_3.f08: New test.
* gfortran.dg/coarray/ptr_comp_4.f08: New test.
* gfortran.dg/coarray_ptr_comp_1.f08: New test.
* gfortran.dg/coarray_ptr_comp_2.f08: New test.
* gfortran.dg/coarray_ptr_comp_3.f08: New test.

From-SVN: r244196

re PR translation/79018 (translatable string typo in params.def:1105)

PR translation/79018
* params.def (PARAM_MAX_STORES_TO_MERGE): Add missing space between
the and store.

From-SVN: r244195

Daily bump.

From-SVN: r244191

* fr.po: Update.

From-SVN: r244187

Add more tests for std::basic_string::find

* testsuite/21_strings/basic_string/operations/find/char/6.cc: New.
* testsuite/21_strings/basic_string/operations/find/wchar_t/6.cc: New.

From-SVN: r244185

re PR target/57583 (large switches with jump tables are horribly broken on m68k)

2017-01-06  Mikael Pettersson  <mikpelinux@gmail.com>

PR target/57583
* config/m68k/m68k.opt (LONG_JUMP_TABLE_OFFSETS): New option.
* config/m68k/linux.h (ASM_RETURN_CASE_JUMP): Handle
TARGET_LONG_JUMP_TABLE_OFFSETS.
* config/m68k/m68kelf.h (ASM_RETURN_CASE_JUMP): Likewise.
* config/m68k/netbsd-elf.h (ASM_RETURN_CASE_JUMP): Likewise.
* config/m68k/m68k.h (CASE_VECTOR_MODE): Likewise.
(ASM_OUTPUT_ADDR_DIFF_ELF): Likewise.
* config/m68k/m68k.md (tablejump expander): Likewise.
(*tablejump_pcrel_hi): Renamed from unnamed insn, reject
TARGET_LONG_JUMP_TABLE_OFFSETS.
(*tablejump_pcrel_si): New insn, handle TARGET_LONG_JUMP_TABLE_OFFSETS.
* doc/invoke.texi (M68K options): Add -mlong-jump-table-offsets.

From-SVN: r244184

Make MicroBlaze support DWARF EH (old Xilinx patch, needed for glibc build).

This patch, taken from
<https://git.busybox.net/buildroot/tree/package/gcc/5.4.0/840-microblaze-enable-dwarf-eh-support.patch>
and with a few formatting cleanups and an update for the removal of
gen_rtx_raw_REG, enables DWARF EH support for MicroBlaze.

This is needed for building glibc with a compiler that includes shared
libgcc; right now all glibc builds for MicroBlaze are failing with my
bot for lack of this support.  (It's dubious if we should have glibc
ports at all where required support is missing in FSF GCC.)

Tested building glibc with build-many-glibcs.py.  I have *not* done
any other testing or any execution testing for MicroBlaze.

2017-01-06  Edgar E. Iglesias <edgar.iglesias@xilinx.com>
    David Holsgrove <david.holsgrove@xilinx.com>

* common/config/microblaze/microblaze-common.c
(TARGET_EXCEPT_UNWIND_INFO): Remove.
* config/microblaze/microblaze-protos.h (microblaze_eh_return):
New prototype.
* config/microblaze/microblaze.c (microblaze_must_save_register)
(microblaze_expand_epilogue, microblaze_return_addr): Handle
calls_eh_return.
(microblaze_eh_return): New function.
* config/microblaze/microblaze.h (RETURN_ADDR_OFFSET)
(EH_RETURN_DATA_REGNO, MB_EH_STACKADJ_REGNUM)
(EH_RETURN_STACKADJ_RTX, ASM_PREFERRED_EH_DATA_FORMAT): New macros.
* config/microblaze/microblaze.md (eh_return): New pattern.

Co-Authored-By: David Holsgrove <david.holsgrove@xilinx.com>
From-SVN: r244183

system.h (GCC_DIAGNOSTIC_PUSH_IGNORED, [...]): Define.

* system.h (GCC_DIAGNOSTIC_PUSH_IGNORED, GCC_DIAGNOSTIC_POP,
GCC_DIAGNOSTIC_STRINGIFY): Define.

* simplify.c (simplify_transformation_to_array): Use
GCC_DIAGNOSTIC_PUSH_IGNORED and GCC_DIAGNOSTIC_POP instead of
#pragma GCC diagnostic {push,ignored,pop}.

From-SVN: r244179

read-rtl.c (rtx_reader::read_rtx_code): Avoid -Wsign-compare warning.

* read-rtl.c (rtx_reader::read_rtx_code): Avoid -Wsign-compare
warning.

From-SVN: r244178

memcmp-1.c: New.

2017-01-06  Aaron Sawdey  <acsawdey@linux.vnet.ibm.com>
* gcc.dg/memcmp-1.c: New.
* gcc.dg/strncmp-1.c: New.
New testcases to check that builtin expansion of memcmp and strncmp works properly.

From-SVN: r244177

Add missing <cassert> include to testcase

* testsuite/util/performance/priority_queue/mem_usage/pop_test.hpp:
Include <cassert> header.

From-SVN: r244176

[ARM] Implement support for ACLE Coprocessor MCRR and MRRC intrinsics

gcc/ChangeLog:
2017-01-06  Andre Vieira  <andre.simoesdiasvieira@arm.com>

* config/arm/arm.md (<mcrr>): New.
(<mrrc>): New.
* config/arm/arm.c (arm_arch5te): New.
(arm_option_override): Set arm_arch5te.
(arm_coproc_builtin_available): Add support for mcrr, mcrr2, mrrc
and mrrc2.
* config/arm/arm-builtins.c (MCRR_QUALIFIERS): Define to...
(arm_mcrr_qualifiers): ... this. New.
(MRRC_QUALIFIERS): Define to...
(arm_mrrc_qualifiers): ... this. New.
* config/arm/arm_acle.h (__arm_mcrr, __arm_mcrr2, __arm_mrrc,
__arm_mrrc2): New.
* config/arm/arm_acle_builtins.def (mcrr, mcrr2, mrrc, mrrc2): New.
* config/arm/iterators.md (MCRRI, mcrr, MCRR): New.
(MRRCI, mrrc, MRRC): New.
* config/arm/unspecs.md (VUNSPEC_MCRR, VUNSPEC_MCRR2, VUNSPEC_MRRC,
VUNSPEC_MRRC2): New.

gcc/testsuite/ChangeLog:
2017-01-06  Andre Vieira  <andre.simoesdiasvieira@arm.com>

* gcc.target/arm/acle/mcrr: New.
* gcc.target/arm/acle/mcrr2: New.
* gcc.target/arm/acle/mrrc: New.
* gcc.target/arm/acle/mrrc2: New.

From-SVN: r244175

[ARM] Implement support for ACLE Coprocessor MCR and MRC intrinsics

gcc/ChangeLog:
2017-01-06  Andre Vieira  <andre.simoesdiasvieira@arm.com>

* config/arm/arm.md (<mcr>): New.
(<mrc>): New.
* config/arm/arm.c (arm_coproc_builtin_available): Add
support for mcr, mrc, mcr2 and mrc2.
* config/arm/arm-builtins.c (MCR_QUALIFIERS): Define to...
(arm_mcr_qualifiers): ... this. New.
(MRC_QUALIFIERS): Define to ...
(arm_mrc_qualifiers): ... this. New.
(MCR_QUALIFIERS): Define to ...
(arm_mcr_qualifiers): ... this. New.
* config/arm/arm_acle.h (__arm_mcr, __arm_mrc, __arm_mcr2,
__arm_mrc2): New.
* config/arm/arm_acle_builtins.def (mcr, mcr2, mrc, mrc2): New.
* config/arm/iterators.md (MCRI, mcr, MCR, MRCI, mrc, MRC): New.
* config/arm/unspecs.md (VUNSPEC_MCR, VUNSPEC_MCR2, VUNSPEC_MRC,
VUNSPEC_MRC2): New.

gcc/testsuite/ChangeLog:
2017-01-06  Andre Vieira  <andre.simoesdiasvieira@arm.com>

* gcc.target/arm/acle/mcr.c: New.
* gcc.target/arm/acle/mrc.c: New.
* gcc.target/arm/acle/mcr2.c: New.
* gcc.target/arm/acle/mrc2.c: New.

From-SVN: r244174