Fix PR82488 - signed integer overflow in expr.c

bootstrap-ubsan shows:
 gcc/expr.c:4103:17: runtime error: signed integer overflow: 0 - -9223372036854775808 cannot be represented in type 'long int'

Fix by handling the saw_unknown case earlier.

PR rtl-optimization/82488
* expr.c (fixup_args_size_notes): Avoid signed integer overflow.

diff --git a/gcc/expr.c b/gcc/expr.c
index ee07de5aaa44..e9d8555c9452 100644
--- a/gcc/expr.c
+++ b/gcc/expr.c
@@ -4100,10 +4100,13 @@ fixup_args_size_notes (rtx_insn *prev, rtx_insn *last, int end_args_size)
       if (STACK_GROWS_DOWNWARD)
  this_delta = -(unsigned HOST_WIDE_INT) this_delta;

-      args_size -= this_delta;
+      if (saw_unknown)
+ args_size = INT_MIN;
+      else
+ args_size -= this_delta;
     }

-  return saw_unknown ? INT_MIN : args_size;
+  return args_size;
 }

 #ifdef PUSH_ROUNDING
--
Markus

From-SVN: r255159

Daily bump.

From-SVN: r255158

i386.c (processor_target_table): Add skylake_cost for skylake-avx512.

* config/i386/i386.c (processor_target_table): Add skylake_cost for
skylake-avx512.
* config/i386/x86-tune-costs.h (skylake_memcpy, skylake_memset,
skylake_cost): New.

* config/i386/driver-i386.c (host_detect_local_cpu):
Detect skylake-avx512.

* config.gcc: Add -march=cannonlake.
* config/i386/driver-i386.c (host_detect_local_cpu): Detect cannonlake.
* config/i386/i386-c.c (ix86_target_macros_internal): Handle cannonlake.
* config/i386/i386.c (processor_costs): Add m_CANNONLAKE.
(PTA_CANNONLAKE): New.
(processor_target_table): Add cannonlake.
(ix86_option_override_internal): Ditto.
(fold_builtin_cpu): Ditto.
(get_builtin_code_for_version): Handle cannonlake.
(M_INTEL_COREI7_CANNONLAKE): New.
* config/i386/i386.h (TARGET_CANNONLAKE, PROCESSOR_CANNONLAKE): New.
* doc/invoke.texi: Add -march=cannonlake.

gcc/testsuite/

* gcc.target/i386/funcspec-56.inc: Handle new march.
* g++.dg/ext/mv16.C: Ditto.

libgcc/

* config/i386/cpuinfo.c (get_intel_cpu): Handle cannonlake.
* config/i386/cpuinfo.h (processor_subtypes): Add
INTEL_COREI7_CANNONLAKE.

From-SVN: r255155

Plugin support on Windows/MinGW

config/ChangeLog:
2017-11-14 Boris Kolpackov  <boris@codesynthesis.com>

* gcc-plugin.m4: Add support for MinGW.

gcc/ChangeLog:
2017-11-14 Boris Kolpackov  <boris@codesynthesis.com>

* plugin.c (add_new_plugin): Use platform-specific library extensions.
(try_init_one_plugin): Alternative implementation for MinGW.
* Makefile.in (plugin_implib): New.
(gengtype-lex.c): Fix broken AIX workaround.
* configure: Regenerate.
* doc/plugins.texi: Document support for MinGW.

gcc/c/ChangeLog:
2017-11-14 Boris Kolpackov  <boris@codesynthesis.com>

* Make-lang.in (c.install-plugin): Install backend import library.

gcc/cp/ChangeLog:
2017-11-14 Boris Kolpackov  <boris@codesynthesis.com>

* Make-lang.in (c++.install-plugin): Install backend import library.

libcc1/ChangeLog:
2017-11-14 Boris Kolpackov  <boris@codesynthesis.com>

* configure: Regenerate.

From-SVN: r255154

Daily bump.

From-SVN: r255153

re PR rtl-optimization/81553 (ICE in immed_wide_int_const, at emit-rtl.c:607)

PR rtl-optimization/81553
* combine.c (simplify_if_then_else): In (if_then_else COND (OP Z C1) Z)
to (OP Z (mult COND (C1 * STORE_FLAG_VALUE))) optimization, if OP
is a shift where C1 has different mode than the whole shift, use C1's
mode for MULT rather than the shift's mode.

* gcc.c-torture/compile/pr81553.c: New test.

From-SVN: r255150

structret1.C (FrameworkObject::action): Return a value.

* g++.dg/abi/structret1.C (FrameworkObject::action): Return a
value.

From-SVN: r255149

re PR target/82848 (ICE in expand_expr_real_2, at expr.c:8715)

PR target/82848
* config/rs6000/rs6000.c (rs6000_gimple_fold_builtin): Don't fold
builtins not enabled in the currently selected ISA.

* gcc.target/powerpc/pr82848.c: New test.

From-SVN: r255148

Daily bump.

From-SVN: r255147

re PR fortran/81304 (Bogus warning with -Wsurprising and -fopenmp: Type specified for intrinsic function 'min' / 'max')

PR fortran/81304
* trans-openmp.c (gfc_trans_omp_array_reduction_or_udr): Set
attr.implicit_type in intrinsic_sym to avoid undesirable warning.

* testsuite/libgomp.fortran/pr81304.f90: New test.

From-SVN: r255144

Factor out division by squares

This patch implements the some of the division optimizations discussed in
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71026.

The division reciprocal optimization now handles divisions by squares:

     x / (y * y) -> x  * (1 / y) * (1 / y)

This requires at least one more division by y before it triggers - the
3 divisions of (1/ y) are then CSEd into a single division.  Overall
this changes 1 division into 1 multiply, which is generally much faster.

2017-11-24  Jackson Woodruff  <jackson.woodruff@arm.com>

    gcc/
PR tree-optimization/71026
* tree-ssa-math-opts (is_division_by_square, is_square_of): New.
(insert_reciprocals): Change to insert reciprocals before a division
by a square and to insert the square of a reciprocal.
(execute_cse_reciprocals_1): Change to consider division by a square.
(register_division_in): Add importance parameter.

    testsuite/
PR tree-optimization/71026
* gfortran.dg/extract_recip_1.f: New test.
* gcc.dg/extract_recip_3.c: New test.
* gcc.dg/extract_recip_4.c: New test.

From-SVN: r255141

re PR tree-optimization/82402 (error: SSA_NAME_OCCURS_IN_ABNORMAL_PHI should be set)

2017-11-24  Richard Biener  <rguenther@suse.de>

PR tree-optimization/82402
* tree-vect-loop-manip.c (create_lcssa_for_virtual_phi): Properly
set SSA_NAME_OCCURS_IN_ABNORMAL_PHI.

* gcc.dg/torture/pr82402.c: New testcase.

From-SVN: r255140

Simplify 0 - ptr.

2017-11-24  Marc Glisse  <marc.glisse@inria.fr>

* match.pd (0-ptr): New transformation.

From-SVN: r255139

re PR bootstrap/83015 (bootstrap comparison failure on ia64)

PR bootstrap/83015
* ipa-inline.c (inline_small_functions): Set current badnes correctly
when skipping checking.

From-SVN: r255138

Whitespace fix for libgfortran/ChangeLog

From-SVN: r255137

re PR tree-optimization/83128 (Unable to optimize {m,c}alloc when strings builtin are used)

2017-11-24  Richard Biener  <rguenther@suse.de>

PR tree-optimization/83128
* tree-ssa-sccvn.c (fully_constant_vn_reference_p): Handle STRING_CSTs.
(vn_reference_lookup_3): Likewise.

* gcc.dg/tree-ssa/ssa-fre-62.c: New testcase.

From-SVN: r255136

PR 36313 Replace int with gfc_charlen_type, take 3

Still some prototypes that didn't match the implementation.

2017-11-24  Janne Blomqvist  <jb@gcc.gnu.org>
        PR fortran/36313
        * m4/maxloc2s.m4: Replace int with gfc_charlen_type, take 3.
        * m4/minloc2s.m4: Likewise.
        * generated/maxloc2_16_s1.c: Regenerated.
        * generated/maxloc2_16_s4.c: Regenerated.
        * generated/maxloc2_4_s1.c: Regenerated.
        * generated/maxloc2_4_s4.c: Regenerated.
        * generated/maxloc2_8_s1.c: Regenerated.
        * generated/maxloc2_8_s4.c: Regenerated.
        * generated/minloc2_16_s1.c: Regenerated.
        * generated/minloc2_16_s4.c: Regenerated.
        * generated/minloc2_4_s1.c: Regenerated.
        * generated/minloc2_4_s4.c: Regenerated.
        * generated/minloc2_8_s1.c: Regenerated.
        * generated/minloc2_8_s4.c: Regenerated.

From-SVN: r255135

re PR sanitizer/83014 (ICE in pretty-print with -fsanitize=bounds)

PR sanitizer/83014
* ubsan.c (ubsan_type_descriptor): Use pp_unsigned_wide_integer
instead of pp_printf with HOST_WIDE_INT_PRINT_DEC.  Avoid calling
tree_to_uhwi twice.

* gcc.dg/ubsan/pr83014.c: New test.

From-SVN: r255134

tree-object-size.c (pass_through_call): Use gimple_call_return_flags ERF_RETURN*ARG* for builtins other than...

* tree-object-size.c (pass_through_call): Use gimple_call_return_flags
ERF_RETURN*ARG* for builtins other than BUILT_IN_ASSUME_ALIGNED,
check for the latter with gimple_call_builtin_p.  Do not handle
BUILT_IN_STPNCPY_CHK which is not a pass through call.

* gcc.dg/builtin-object-size-18.c: New test.

From-SVN: r255133

Add missing gcc/config.gcc hunks from r255121.

From-SVN: r255132

Add missing hunks from r255121.

From-SVN: r255131

Daily bump.

From-SVN: r255130

Add [[nodiscard]] attribute to std::launder

* libsupc++/new (launder): Add nodiscard attribute.
* testsuite/18_support/launder/nodiscard.cc: New test.

From-SVN: r255127

[ARM] Fix Dot Product NEON intrinsics

2017-11-24  Christophe Lyon  <christophe.lyon@linaro.org>

* config/arm/arm_neon.h: Fix pragma GCC push_options before
vdot_u32.

From-SVN: r255126

re PR ada/83091 (ICE with -g in arm-eabi compilation)

PR ada/83091
* gcc-interface/decl.c (gnat_to_gnu_entity): Do not build a variant
type for the implementation type of a packed array.

From-SVN: r255125

Add [[nodiscard]] attribute to C++17 components

* include/bits/fs_path.h (path::empty): Add nodiscard attribute.
* include/bits/range_access.h (empty): Likewise.
* include/std/string_view (basic_string_view::empty): Likewise.
* testsuite/21_strings/basic_string_view/capacity/empty_neg.cc: New
test.
* testsuite/24_iterators/range_access_cpp17_neg.cc: New test.
* testsuite/27_io/filesystem/path/query/empty_neg.cc: New test.

From-SVN: r255124

PR libstdc++/83134 Ensure std::__not_ converts B::value to bool

PR libstdc++/83134
* include/std/type_traits (__not_): Explicitly convert to bool.
* testsuite/20_util/declval/requirements/1_neg.cc: Adjust dg-error.
* testsuite/20_util/logical_traits/83134.cc: New test.
* testsuite/20_util/make_signed/requirements/typedefs_neg.cc: Adjust
dg-error.
* testsuite/20_util/make_unsigned/requirements/typedefs_neg.cc:
Likewise.

From-SVN: r255122

Enable VBMI2 support [3/7]

gcc/
config/i386/avx512vbmi2intrin.h (_mm512_mask_expand_epi8,
_mm512_maskz_expand_epi8, _mm512_mask_expandloadu_epi8,
_mm512_maskz_expandloadu_epi8, _mm512_mask_expand_epi16,
_mm512_maskz_expand_epi16, _mm512_mask_expandloadu_epi16,
_mm512_maskz_expandloadu_epi16): New intrinsics.
config/i386/avx512vbmi2vlintrin.h (_mm_mask_expand_epi8,
_mm_maskz_expand_epi8, _mm_mask_expandloadu_epi8,
_mm_maskz_expandloadu_epi8, _mm_mask_expand_epi16,
_mm_maskz_expand_epi16, _mm_mask_expandloadu_epi16,
_mm_maskz_expandloadu_epi16, _mm256_mask_expand_epi16,
_mm256_maskz_expand_epi16, _mm256_mask_expandloadu_epi16,
_mm256_maskz_expandloadu_epi16, _mm256_mask_expand_epi8,
_mm256_maskz_expand_epi8, _mm256_mask_expandloadu_epi8,
_mm256_maskz_expandloadu_epi8): New intrinsics.
config/i386/i386-builtin-types.def (V64QI_FTYPE_PCV64QI_V64QI_UDI,
V32HI_FTYPE_PCV32HI_V32HI_USI, V32QI_FTYPE_PCV32QI_V32QI_USI,
V16HI_FTYPE_PCV16HI_V16HI_UHI, V16QI_FTYPE_PCV16QI_V16QI_UHI,
V8HI_FTYPE_PCV8HI_V8HI_UQI): New types.
config/i386/i386.c (ix86_expand_special_args_builtin): Use new types.
config/i386/sse.md (VI248_VLBW): New iterator.
(expand<mode>_mask, expand<mode>_maskz): New patterns.

gcc/testsuite/
gcc.target/i386/avx512f-vpexpandb-1.c: New test.
gcc.target/i386/avx512f-vpexpandb-2.c: Ditto.
gcc.target/i386/avx512f-vpexpandw-1.c: Ditto.
gcc.target/i386/avx512f-vpexpandw-2.c: Ditto.
gcc.target/i386/avx512vl-vpexpandb-1.c: Ditto.
gcc.target/i386/avx512vl-vpexpandb-2.c: Ditto.
gcc.target/i386/avx512vl-vpexpandw-1.c: Ditto.
gcc.target/i386/avx512vl-vpexpandw-2.c: Ditto.

From-SVN: r255121

Enable VBMI2 support [2/7]

gcc/
config.gcc (avx512vbmi2intrin.h, avx512vbmi2vlintrin): New headers.
config/i386/avx512vbmi2intrin.h (_mm512_mask_compress_epi8,
_mm512_maskz_compress_epi8, _mm512_mask_compressstoreu_epi8,
_mm512_mask_compress_epi16, _mm512_maskz_compress_epi16,
_mm512_mask_compressstoreu_epi16): New.
config/i386/avx512vbmi2vlintrin.h (_mm_mask_compress_epi8,
_mm_maskz_compress_epi8, _mm256_mask_compressstoreu_epi16,
_mm_mask_compress_epi16, _mm_maskz_compress_epi16,
_mm256_mask_compress_epi16, _mm256_maskz_compress_epi16,
_mm_mask_compressstoreu_epi8, _mm_mask_compressstoreu_epi16,
_mm256_mask_compress_epi8, _mm256_maskz_compress_epi8,
_mm256_mask_compressstoreu_epi8): New.
config/i386/i386-builtin-types.def (VOID_FTYPE_PV64QI_V64QI_UDI,
VOID_FTYPE_PV32HI_V32HI_USI, VOID_FTYPE_PV32QI_V32QI_USI,
VOID_FTYPE_PV16QI_V16QI_UHI, VOID_FTYPE_PV16HI_V16HI_UHI,
VOID_FTYPE_PV8HI_V8HI_UQI): New types.
config/i386/i386-builtin.def (__builtin_ia32_compressqi512_mask,
__builtin_ia32_compresshi512_mask, __builtin_ia32_compressqi256_mask,
__builtin_ia32_compressqi128_mask, __builtin_ia32_compresshi256_mask,
__builtin_ia32_compresshi128_mask,
__builtin_ia32_compressstoreuqi512_mask,
__builtin_ia32_compressstoreuhi512_mask,
__builtin_ia32_compressstoreuqi256_mask,
__builtin_ia32_compressstoreuqi128_mask,
__builtin_ia32_compressstoreuhi256_mask,
__builtin_ia32_compressstoreuhi128_mask): New builtins.
config/i386/i386.c (ix86_init_mmx_sse_builtins): Create special args
array for flags2.
(ix86_expand_special_args_builtin): Handle new types.
(s4fma_expand): Handle new builtin array.
config/i386/immintrin.h: Include new headers.
config/i386/sse.md (VI12_AVX512VLBW): New iterator.
(compress<mode>_mask, compressstore<mode>_mask): New patterns.

gcc/testsuite/
gcc.target/i386/avx512-check.h: Handle AVX512VBMI2 bit.
gcc.target/i386/avx512f-vpcompressb-1.c: New test.
gcc.target/i386/avx512f-vpcompressb-2.c: Ditto.
gcc.target/i386/avx512f-vpcompressw-1.c: Ditto.
gcc.target/i386/avx512f-vpcompressw-2.c: Ditto.
gcc.target/i386/avx512vl-vpcompressb-1.c: Ditto.
gcc.target/i386/avx512vl-vpcompressb-2.c: Ditto.
gcc.target/i386/avx512vl-vpcompressw-1.c: Ditto.
gcc.target/i386/avx512vl-vpcompressw-2.c: Ditto.
gcc.target/i386/i386.exp (check_effective_target_avx512vbmi2): New.

From-SVN: r255119

rs6000.c (rs6000_gimple_fold_builtin): Formatting fixes.

* config/rs6000/rs6000.c (rs6000_gimple_fold_builtin): Formatting
fixes.  Declare temp and g variables at the top in order to avoid
{} in most of the cases.

From-SVN: r255118

parser.c (cp_parser_omp_declare): Change return type to bool from void, return true for declare simd.

* parser.c (cp_parser_omp_declare): Change return type to bool from
void, return true for declare simd.
(cp_parser_pragma): Return cp_parser_omp_declare returned value
rather than always false.

From-SVN: r255117

c-parser.c (c_parser_omp_declare_simd): Reject declare simd in pragma_stmt context.

* c-parser.c (c_parser_omp_declare_simd): Reject declare simd in
pragma_stmt context.

* gcc.dg/gomp/declare-simd-1.c (f9): Remove.
* gcc.dg/gomp/declare-simd-5.c: New test.

From-SVN: r255116

re PR fortran/81841 (THREADPRIVATE (OpenMP) wrongly rejected in BLOCK DATA)

PR fortran/81841
* parse.c (parse_spec): Allow ST_OMP_THREADPRIVATE inside of
BLOCK DATA.

* libgomp.fortran/pr81841.f90: New test.

From-SVN: r255115

pr65240-1.C: Add -Wno-return-type to dg-options.

* g++.dg/pr65240-1.C: Add -Wno-return-type to dg-options.
* g++.dg/pr65240-2.C: Likewise.
* g++.dg/pr65240-3.C: Likewise.
* g++.dg/pr65240-4.C: Likewise.

From-SVN: r255114

Simplify ptr - 0

2017-11-23  Marc Glisse  <marc.glisse@inria.fr>

* match.pd (ptr-0): New transformation.

From-SVN: r255113

This patch adds support for modelling the varying costs of different addressing modes.

This patch adds support for modelling the varying costs of
different addressing modes. The generic cost table treats
all addressing modes as having equal cost.

gcc/ChangeLog:

2017-11-23  Charles Baylis  <charles.baylis@linaro.org>

        * config/arm/arm-protos.h (enum arm_addr_mode_op): New.
        (struct addr_mode_cost_table): New.
        (struct tune_params): Add field addr_mode_costs.
        * config/arm/arm.c (generic_addr_mode_costs): New.
        (arm_slowmul_tune): Initialise addr_mode_costs field.
        (arm_fastmul_tune): Likewise.
        (arm_strongarm_tune): Likewise.
        (arm_xscale_tune): Likewise.
        (arm_9e_tune): Likewise.
        (arm_marvell_pj4_tune): Likewise.
        (arm_v6t2_tune): Likewise.
        (arm_cortex_tune): Likewise.
        (arm_cortex_a8_tune): Likewise.
        (arm_cortex_a7_tune): Likewise.
        (arm_cortex_a15_tune): Likewise.
        (arm_cortex_a35_tune): Likewise.
        (arm_cortex_a53_tune): Likewise.
        (arm_cortex_a57_tune): Likewise.
        (arm_exynosm1_tune): Likewise.
        (arm_xgene1_tune): Likewise.
        (arm_cortex_a5_tune): Likewise.
        (arm_cortex_a9_tune): Likewise.
        (arm_cortex_a12_tune): Likewise.
        (arm_cortex_a73_tune): Likewise.
        (arm_v7m_tune): Likewise.
        (arm_cortex_m7_tune): Likewise.
        (arm_v6m_tune): Likewise.
        (arm_fa726te_tune): Likewise.
        (arm_mem_costs): Use table lookup to calculate cost of addressing
        mode.

From-SVN: r255112

[ARM] Refactor costs calculation for MEM.

This patch moves the calculation of costs for MEM into a
separate function, and reforms the calculation into two
parts. Firstly any additional cost of the addressing mode
is calculated, and then the cost of the memory access itself
is added.

In this patch, the calculation of the cost of the addressing
mode is omitted, to be added in a subsequent patch.

gcc/ChangeLog:

<date>  Charles Baylis  <charles.baylis@linaro.org>

        * config/arm/arm.c (arm_mem_costs): New function.
        (arm_rtx_costs_internal): Use arm_mem_costs.

gcc/testsuite/ChangeLog:

<date>  Charles Baylis  <charles.baylis@linaro.org>

        * gcc.target/arm/addr-modes-float.c: New test.
        * gcc.target/arm/addr-modes-int.c: New test.
        * gcc.target/arm/addr-modes.h: New header.

From-SVN: r255111

re PR fortran/36313 ([F03] {MIN,MAX}{LOC,VAL} should accept character arguments)

2017-11-23  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/36313
* m4/maxloc2s.m4: Replace int for character length by
gfc_charlen_type.
* m4/minloc2s.m4: Likewise.
* generated./maxloc2_16_s1.c: Regenerated.
* generated./maxloc2_16_s4.c: Regenerated.
* generated./maxloc2_4_s1.c: Regenerated.
* generated./maxloc2_4_s4.c: Regenerated.
* generated./maxloc2_8_s1.c: Regenerated.
* generated./maxloc2_8_s4.c: Regenerated.
* generated./minloc2_16_s1.c: Regenerated.
* generated./minloc2_16_s4.c: Regenerated.
* generated./minloc2_4_s1.c: Regenerated.
* generated./minloc2_4_s4.c: Regenerated.
* generated./minloc2_8_s1.c: Regenerated.
* generated./minloc2_8_s4.c: Regenerated.

From-SVN: r255109

Fix .debug_rnglists generation with -gdwarf-5 -gsplit-dwarf.

Early debug broke generation of .debug_rnglists when using both -gdwarf5
and -gsplit-dwarf. It introduces a generation for init_sections_and_labels,
but doesn't account for the generation of up to 4 unique ranges labels,
two created in init_sections_and_labels and two in output_rnglists.
Fix this by passing generation to output_rnglists and creating 4 unique
labels per generation.

Without this fix using -gdwarf-5 -gsplit-dwarf could result in:
  Error: symbol `.Ldebug_ranges2' is already defined

gcc/ChangeLog:

       * dwarf2out.c (init_sections_and_labels): Use generation to create
       unique ranges_section_label and ranges_base_label. Return generation.
       (output_rnglists): Add generation argument. Use generation to create
       unique ranges labels.
       (dwarf2out_finish): Get generation from init_sections_and_labels
       and pass generation to output_rnglists.

From-SVN: r255107

generic.texi (ANNOTATE_EXPR): Document 3rd operand.

* doc/generic.texi (ANNOTATE_EXPR): Document 3rd operand.
* cfgloop.h (struct loop): Add unroll field.
* function.h (struct function): Add has_unroll bitfield.
* gimplify.c (gimple_boolify) <ANNOTATE_EXPR>: Deal with unroll kind.
(gimplify_expr) <ANNOTATE_EXPR>: Propagate 3rd operand.
* loop-init.c (pass_loop2::gate): Return true if cfun->has_unroll.
(pass_rtl_unroll_loops::gate): Likewise.
* loop-unroll.c (decide_unrolling): Tweak note message.  Skip loops
for which loop->unroll==1.
(decide_unroll_constant_iterations): Use note for consistency and
take loop->unroll into account.  Return early if loop->unroll is set.
Fix thinko in existing test.
(decide_unroll_runtime_iterations): Use note for consistency and
take loop->unroll into account.
(decide_unroll_stupid): Likewise.
* lto-streamer-in.c (input_cfg): Read loop->unroll.
* lto-streamer-out.c (output_cfg): Write loop->unroll.
* tree-cfg.c (replace_loop_annotate_in_block) <annot_expr_unroll_kind>:
        New case.
(replace_loop_annotate) <annot_expr_unroll_kind>: Likewise.
(print_loop): Print loop->unroll if set.
* tree-core.h (enum annot_expr_kind): Add annot_expr_unroll_kind.
* tree-inline.c (copy_loops): Copy unroll and set cfun->has_unroll.
* tree-pretty-print.c (dump_generic_node) <annot_expr_unroll_kind>:
New case.
* tree-ssa-loop-ivcanon.c (try_unroll_loop_completely): Bail out if
loop->unroll is set and smaller than the trip count.  Otherwise bypass
entirely the heuristics if loop->unroll is set.  Remove dead note.
Fix off-by-one bug in other note.
(try_peel_loop): Bail out if loop->unroll is set.  Fix formatting.
(tree_unroll_loops_completely_1): Force unrolling if loop->unroll
is greater than 1.
(tree_unroll_loops_completely): Make static.
(pass_complete_unroll::execute): Use correct type for variable.
(pass_complete_unrolli::execute): Fix formatting.
* tree.def (ANNOTATE_EXPR): Add 3rd operand.
ada/
* gcc-interface/trans.c (gnat_gimplify_stmt) <LOOP_STMT>: Pass 3rd
operand to ANNOTATE_EXPR and also pass unrolling hints.
c/
* c-parser.c (c_parser_while_statement): Pass 3rd operand to
ANNOTATE_EXPR.
(c_parser_do_statement): Likewise.
(c_parser_for_statement): Likewise.
cp/
* pt.c (tsubst_expr) <ANNOTATE_EXPR>: Recurse on 3rd operand.
* semantics.c (finish_while_stmt_cond): Pass 3rd operand to
ANNOTATE_EXPR.
(finish_do_stmt): Likewise.
(finish_for_cond): Likewise.
fortran/
* trans-stmt.c (gfc_trans_forall_loop): Pass 3rd operand to
ANNOTATE_EXPR.

From-SVN: r255106

i386.h (TARGET_PREFER_AVX256): Also enable when TARGET_PREFER_AVX128 is set.

* config/i386/i386.h (TARGET_PREFER_AVX256): Also
enable when TARGET_PREFER_AVX128 is set.

From-SVN: r255105

ipa-profile.c (ipa_propagate_frequency_1): Use count instead of frequency.

* ipa-profile.c (ipa_propagate_frequency_1): Use count instead of
frequency.
* cgraph.c (cgraph_edge::dump_edge_flags): Dump sreal frequencies.
(cgraph_edge::maybe_hot_p): Use sreal frequencies.

From-SVN: r255104

ipa-fnsummary.c (record_modified_bb_info): Use sreal frequencies.

* ipa-fnsummary.c (record_modified_bb_info): Use sreal
frequencies. Fix estimation of aggregate parameters.

* gcc.dg/ipa/inline-1.c: Update template.

From-SVN: r255103

cgraphclones.c (cgraph_node::create_clone): Fix updating of profile when inlining.

* cgraphclones.c (cgraph_node::create_clone): Fix updating of profile
when inlining.

From-SVN: r255102

[libgccjit] Wrap RETURN_NULL_IF_FAIL_NONNULL_NUMERIC_TYPE in JIT_{BEGIN,END}_STMT.

2017-11-23  Tom de Vries  <tom@codesourcery.com>

* libgccjit.c (RETURN_NULL_IF_FAIL_NONNULL_NUMERIC_TYPE): Wrap in
JIT_{BEGIN,END}_STMT.

From-SVN: r255101

[sh] Wrap ASM_OUTPUT_ADDR_VEC_ELT in do {} while (0)

2017-11-23  Tom de Vries  <tom@codesourcery.com>

* config/sh/sh.h (ASM_OUTPUT_ADDR_VEC_ELT): Wrap in "do {} while (0)".

From-SVN: r255100

[libobjc] Wrap CLASS_TABLE_HASH in do {} while (0)

2017-11-23  Tom de Vries  <tom@codesourcery.com>

* class.c (CLASS_TABLE_HASH): Wrap in "do {} while (0)".

From-SVN: r255099

[ft32] Remove semicolon after ASM_OUTPUT_ADDR_VEC_ELT

2017-11-23  Tom de Vries  <tom@codesourcery.com>

* config/ft32/ft32.h (ASM_OUTPUT_ADDR_VEC_ELT): Remove semicolon after
macro.

From-SVN: r255098

re PR target/83111 ([sh] stack smashing detected in gen_udivsi3)

gcc/
PR target/83111
* config/sh/sh.md (udivsi3, divsi3, sibcall_value_pcrel,
sibcall_value_pcrel_fdpic): Use local variable instead of
operands[3].
(calli_tbr_rel): Add missing operand 2.
(call_valuei_tbr_rel): Add missing operand 3.

From-SVN: r255096

re PR middle-end/82253 (ICE in convert_move, at expr.c:604)

PR middle-end/82253
* expr.c (expand_assignment): For CONCAT to_rtx, complex type from and
bitpos/bitsize covering the whole destination, use store_expr only if
the complex mode is the same.  Otherwise, use expand_normal and if
it returns CONCAT, subreg each part separately instead of trying to
subreg the whole result.

* gfortran.dg/pr82253.f90: New test.

From-SVN: r255095

re PR fortran/82814 (ICE from submodule character function)

2017-11-23  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/82814
* trans-types.c (gfc_sym_type): If a character function result
is missing the charlen backend_decl, use the one from the name-
space procedure symbol, if present.

2017-11-23  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/82814
* gfortran.dg/submodule_31.f08: New test.

From-SVN: r255094

re PR tree-optimization/23094 (store ccp, or store copy prop misses an optimization)

2017-11-23  Richard Biener  <rguenther@suse.de>

PR tree-optimization/23094
* tree-ssa-sccvn.c (vuse_ssa_val): Handle VN_TOP when we
come here from walking over backedges in the first iteration.
(vn_reference_lookup_3): Skip clobbers that store the same value.

* gcc.dg/tree-ssa/ssa-fre-61.c: New testcase.

From-SVN: r255093

re PR tree-optimization/81403 (wrong code at -O3)

2017-11-23  Richard Biener  <rguenther@suse.de>

PR tree-optimization/81403
* tree-ssa-pre.c (get_representative_for): Add parameter specifying
a block we need a leader relative to.
(phi_translate_1): For nary processing require a leader from
get_representative_for given we run expression simplification
using match-and-simplify.  Remove previous fix.

From-SVN: r255092

re PR tree-optimization/83089 (ICE: Segmentation fault (in instantiate_scev_name))

2017-11-23  Richard Biener  <rguenther@suse.de>

PR tree-optimization/83089
* gcc.dg/pr83089.c: Require pthread.

From-SVN: r255091

cmd/go, go/internal/gccgoimporter: pass -X to ar on AIX

    Reviewed-on: https://go-review.googlesource.com/72930

From-SVN: r255090

Daily bump.

From-SVN: r255089

gimple-ssa-evrp-analyze.c (evrp_range_analyzer::try_find_new_range): Use new method allocate_value_range rather than accessing the...

* gimple-ssa-evrp-analyze.c (evrp_range_analyzer::try_find_new_range):
Use new method allocate_value_range rather than accessing the
vrp_value_range_pool data member directly.
* tree-vrp.c (simplify_stmt_for_jump_threading): Tweak slightly
to use extract_range_from_stmt method to avoid need for
extract_range_from_assignment method.
(vrp_prop::vrp_finalize): Use set_lattice_propagation_complete
method rather than setting values_propgated data member directly.
* vr-values.h (class vr_values): Privatize vrp_value_range_pool,
and values propagated data members and extract_range_from_assignment
method.  Reorder private data members to conform to standards.
Add new methods set_lattice_propagation_complete and
allocate_value_range.

From-SVN: r255086

* g++.dg/abi/pr68355.C: Don't expect tail call on ia32.

From-SVN: r255085

re PR c++/82401 (error: qsort comparator non-negative on sorted output: 1 in insert_late_enum_def_bindings on an invalid code)

PR c++/82401
* name-lookup.c (member_name_cmp): Return 0 if a == b.

* g++.dg/cpp0x/pr82401.C: New test.

From-SVN: r255084

re PR rtl-optimization/83030 (ICE in create_pseudo_cfg, at dwarf2cfi.c:2840)

PR rtl-optimization/83030
* doc/rtl.texi (Flags in an RTL Expression): Alphabetize, add entry
for CROSSING_JUMP_P and mention usage of 'jump' for JUMP_INSNs.
(Insns): Delete entry for REG_CROSSING_JUMP in register notes.
* bb-reorder.c (update_crossing_jump_flags): Do not test whether the
CROSSING_JUMP_P flag is already set before setting it.
* cfgrtl.c (fixup_partition_crossing): Likewise.
* reorg.c (relax_delay_slots): Do not consider a CROSSING_JUMP_P insn
as useless.

From-SVN: r255083

jit: update test suite for change to -Warray-bounds

jit.dg/test-error-array-bounds.c is a test of how libgccjit
handles errors coming from the middle-end, using -Warray-bounds as
an example.

It started failing as of r254830 (aka 987b3316c4e7c908e3c7d2151c20971f9707ad66),
due to that commit changing the wording of the message.

This commit fixes the jit testcase by updating it to use the new
wording.

gcc/testsuite/ChangeLog:
PR tree-optimization/82588
PR tree-optimization/82583
* jit.dg/test-error-array-bounds.c (verify_code): Update expected
error message from -Warray-bounds.

From-SVN: r255082

* g++.dg/graphite/pr41562.C: Add -Wno-return-type.

From-SVN: r255081

re PR libgomp/83106 (libgomp/target.c:2671:2: error: ‘strncat’ specified bound 5 equals source length [-Werror=stringop-overflow=])

PR libgomp/83106
* target.c (gomp_target_init): Compute lengths just once and
use them in both malloc size and subsequent copying.

From-SVN: r255080

simplify-rtx.c (simplify_binary_operation_1): Handle the case where both arguments are using gen_const_vec_series.

* simplify-rtx.c (simplify_binary_operation_1) <case VEC_SERIES>:
Handle the case where both arguments are using gen_const_vec_series.

From-SVN: r255079

Make shift argument to eoshift0 and eoshift2 be of type index_type

Test case that requires 4 GB RAM that exercises eoshift0:

program main
  implicit none
  integer(1) :: line(2_8**31+10)
  line = 42
  line = eoshift(line, 2_8**31+7)
  print *, line(1:4)
  print *, line((ubound(line, 1, kind=8)-4):ubound(line, 1, kind=8))
end program main

Before patch:
❯ ./a.out
    0    0    0    0
    42   42   42   42   42

With patch:
❯ ./a.out
   42   42   42    0
   0    0    0    0    0

Regtested on x86_64-pc-linux-gnu.

libgfortran/ChangeLog:

2017-11-22  Janne Blomqvist  <jb@gcc.gnu.org>

        * intrinsics/eoshift0.c (eoshift0): Make shift an index_type.
        * intrinsics/eoshift2.c (eoshift2): Make shift an index_type.

From-SVN: r255077

C/C++: fix quoting of "aka" typedef information (PR 62170)

PR 62170 describes a problem with how the quoting in pp_format
interacts with the "aka" information for typedefs in %qT for
the C family of frontends, and also now for %qH and %qI in the
C++ frontend: we print:

  'Py_ssize_t* {aka int*}'
   ^^^^^^^^^^^^^^^^^^^^^^ colorized as "quote"

i.e.
  '[START_COLOR]Py_ssize_t* {aka int*}[END_COLOR]'

when we should print:

  'Py_ssize_t*' {aka 'int*'}
   ^^^^^^^^^^^        ^^^^ colorized as "quote"

i.e.
  '[START_COLOR]Py_ssize_t*[END_COLOR]' {aka '[START_COLOR]int*[END_COLOR]'}

where the opening and closing quote characters and colorization are
added by the 'q' handling within pp_format.

This patch fixes the quoting by updating the %T handling in C and C++
and the %H/%I handling in C++ to insert the quoting appropriately.
It converts the "quote" param of the pp_format_decoder callback from
bool to bool *, allowing for the %T and %H/%I handlers to write
false back to it, to avoid printing the closing quote for the cases
like the above where the trailing closing quote isn't needed.

It introduces pp_begin_quote/pp_end_quote to simplify this.  These
take a "bool show_color", rather than using "pp_show_color (pp)"
since cxx_pp's pp_show_color isn't currently initialized (since
cxx_initialize_diagnostics happens before diagnostic_color_init).

gcc/c/ChangeLog:
PR c++/62170
* c-objc-common.c (c_tree_printer): Convert penultimate param from
bool to bool *.  Within '%T' handling, if showing an "aka", use
"quoted" param to add appropriate quoting.

gcc/cp/ChangeLog:
PR c++/62170
* error.c (type_to_string): Add leading comment.  Add params
"postprocessed", "quote", and "show_color", using them to fix
quoting of the "aka" for types involving typedefs.
(arg_to_string): Update for new params to type_to_string.
(cxx_format_postprocessor::handle): Likewise.
(cp_printer): Convert penultimate param from bool to bool *.
Update call to type_to_string and calls to
defer_phase_2_of_type_diff.

gcc/fortran/ChangeLog:
PR c++/62170
* error.c (gfc_notify_std): Convert "quoted" param from bool to
bool *.

gcc/ChangeLog:
PR c++/62170
* pretty-print.c (pp_format): Move quoting implementation to
pp_begin_quote and pp_end_quote.  Update pp_format_decoder call
to pass address of "quote" local.
(pp_begin_quote): New function.
(pp_end_quote): New function.
* pretty-print.h (printer_fn): Convert penultimate param from bool
to bool *.
(pp_begin_quote): New decl.
(pp_end_quote): New decl.
* tree-diagnostic.c (default_tree_printer): Convert penultimate
param from bool to bool *.
* tree-diagnostic.h (default_tree_printer): Likewise.

gcc/testsuite/ChangeLog:
PR c++/62170
* g++.dg/diagnostic/aka1.C: Update expected error messages to
reflect fixes to quoting.
* g++.dg/diagnostic/aka2.C: New test case.
* g++.dg/parse/error55.C: Update expected error messages to
reflect fixes to quoting.
* gcc.dg/diag-aka-1.c: Likewise.
* gcc.dg/diag-aka-2.c: New test case.
* gcc.dg/pr13804-1.c: Update expected error messages to reflect
fixes to quoting.
* gcc.dg/pr56980.c: Likewise.
* gcc.dg/pr65050.c: Likewise.
* gcc.dg/redecl-14.c: Likewise.
* gcc.dg/utf16-4.c Likewise.
* gcc.target/i386/sse-vect-types.c (__m128d): Likewise.
* obj-c++.dg/invalid-type-1.mm: Likewise.
* objc.dg/proto-lossage-4.m: Likewise.

From-SVN: r255076

target-supports.exp (check_effective_target_bswap16): Remove

* lib/target-supports.exp (check_effective_target_bswap16): Remove
(check_effective_target_bswap32): Ditto.
(check_effective_target_bswap64): Ditto.
* gcc.dg/optimize-bswaphi-1.c: Use bswap effective
target instead of bswap16.
* gcc.dg/optimize-bswapsi-1.c: Use bswap effective
target instead of bswap32.
* gcc.dg/optimize-bswapsi-2.c: Ditto.
* gcc.dg/optimize-bswapsi-3.c: Ditto.
* gcc.dg/optimize-bswapsi-4.c: Ditto.
* gcc.dg/optimize-bswapsi-5.c: Ditto.
* gcc.dg/pr77673.c: Ditto.
* gcc.dg/optimize-bswapdi-1.c: Use bswap effective
target instead of bswap64.
* gcc.dg/optimize-bswapdi-2.c: Ditto.
* gcc.dg/optimize-bswapdi-3.c: Ditto.

From-SVN: r255075

* gimple-ssa-evrp-analyze.c (evrp_range_analyzer::evrp_range_analyzer)
Initialize vr_values.
(evrp_range_analyzer::try_find_new_range): Call methods attached to
vr_values via vr_values class instance rather than delegators.
(evrp_range_analyzer::record_ranges_from_phis): Likewise.
(evrp_range_analyzer::record_ranges_from_stmt): Likewise.
(evrp_range_analyzer::push_value_range): Likewise.
(evrp_range_analyzer::pop_value_range): Likewise.
* gimple-ssa-evrp-analyze.h (class evrp_range_analyzer): Remove
most delegators.  Those remaining are exposed as public interfaces.
Make vr_values a pointer and private.
(evrp_range_analyzer::~evrp_range_analyzer): Delete the attached
vr_values.
(evrp_range_analyzer::get_vr_value): New method.
* gimple-ssa-evrp.c (class evrp_folder): Use DISABLE_COPY_AND_ASSIGN.
(evrp_folder::evrp_folder): New ctor to initialize vr_values.
(class evrp_dom_walker): Attach evrp_folder class, initialize
it in the ctor.  Remove temporary delegators.
(evrp_dom_walker::before_dom_children): Call methods in attached
evrp_range_analyzer class via class instance pointer.  Use
free value_range_constant_singleton to remove need for
op_with_constant_singleton_value delegator method.  Do not
create a vrp_prop class instance for every call!  Narrow
scope of a couple variables.
(evrp_dom_walker::cleanup): Call methods in attached
evrp_range_analyzer class via class instance pointer.
* vr-values.h (class vr_values): Privatize many methods and
data members.

From-SVN: r255074

Replace REDUC_*_EXPRs with internal functions.

This patch replaces the REDUC_*_EXPR tree codes with internal functions.
This is needed so that the upcoming in-order reductions can also use
internal functions without too much complication.

2017-11-22  Richard Sandiford  <richard.sandiford@linaro.org>

gcc/
* tree.def (REDUC_MAX_EXPR, REDUC_MIN_EXPR, REDUC_PLUS_EXPR): Delete.
* cfgexpand.c (expand_debug_expr): Remove handling for them.
* expr.c (expand_expr_real_2): Likewise.
* fold-const.c (const_unop): Likewise.
* optabs-tree.c (optab_for_tree_code): Likewise.
* tree-cfg.c (verify_gimple_assign_unary): Likewise.
* tree-inline.c (estimate_operator_cost): Likewise.
* tree-pretty-print.c (dump_generic_node): Likewise.
(op_code_prio): Likewise.
(op_symbol_code): Likewise.
* internal-fn.def (DEF_INTERNAL_SIGNED_OPTAB_FN): Define.
(IFN_REDUC_PLUS, IFN_REDUC_MAX, IFN_REDUC_MIN): New internal functions.
* internal-fn.c (direct_internal_fn_optab): New function.
(direct_internal_fn_array, direct_internal_fn_supported_p
(internal_fn_expanders): Handle DEF_INTERNAL_SIGNED_OPTAB_FN.
* fold-const-call.c (fold_const_reduction): New function.
(fold_const_call): Handle CFN_REDUC_PLUS, CFN_REDUC_MAX and
CFN_REDUC_MIN.
* tree-vect-loop.c: Include internal-fn.h.
(reduction_code_for_scalar_code): Rename to...
(reduction_fn_for_scalar_code): ...this and return an internal
function.
(vect_model_reduction_cost): Take an internal_fn rather than
a tree_code.
(vect_create_epilog_for_reduction): Likewise.  Build calls rather
than assignments.
(vectorizable_reduction): Use internal functions rather than tree
codes for the reduction operation.  Update calls to the functions
above.
* config/aarch64/aarch64-builtins.c (aarch64_gimple_fold_builtin):
Use calls to internal functions rather than REDUC tree codes.
* config/aarch64/aarch64-simd.md: Update comment accordingly.

From-SVN: r255073

PR 83097 Use __BYTE_ORDER__ predefined macro instead of runtime check

By using the __BYTE_ORDER__ predefined macro we don't need the
determine_endianness function anymore.

Regtested on x86_64-pc-linux-gnu.

libgfortran/ChangeLog:

2017-11-22  Janne Blomqvist  <jb@gcc.gnu.org>

        PR libfortran/83097
* io/inquire.c (inquire_via_unit): Use __BYTE_ORDER__ predefined
macro.
* io/open.c (st_open): Likewise.
* io/transfer.c (data_transfer_init): Likewise.
* io/write.c (btoa_big): Likewise.
(otoa_big): Likewise.
(ztoa_big): Likewise.
* libgfortran.h (big_endian): Remove variable.
(GFOR_POINTER_TO_L1): Use __BYTE_ORDER__ macro.
* runtime/main.c (determine_endianness): Remove function.
(init): Remove call to determine_endianness.
* runtime/minimal.c: Remove setting big_endian variable.

From-SVN: r255072

[PATCH] Add myself as GCC maintainer

*** ChangeLog ***

2017-11-22  Sudakshina Das  <sudi.das@arm.com>

* MAINTAINERS (Write After Approval): Add myself.

From-SVN: r255071

re PR fortran/36313 ([F03] {MIN,MAX}{LOC,VAL} should accept character arguments)

2017-11-22  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/36313
* Makefile.am: Add i_maxloc0s_c, i_maxloc1s_c, i_maxloc2s_c,
i_minloc0s_c, i_minloc1s_c and i_minloc2s_c.
* Makefile.in: Regenerated.
        * generated/maxloc0_16_s1.c: New file.
        * generated/maxloc0_16_s4.c: New file.
        * generated/maxloc0_4_s1.c: New file.
        * generated/maxloc0_4_s4.c: New file.
        * generated/maxloc0_8_s1.c: New file.
        * generated/maxloc0_8_s4.c: New file.
        * generated/maxloc1_16_s1.c: New file.
        * generated/maxloc1_16_s4.c: New file.
        * generated/maxloc1_4_s1.c: New file.
        * generated/maxloc1_4_s4.c: New file.
        * generated/maxloc1_8_s1.c: New file.
        * generated/maxloc1_8_s4.c: New file.
        * generated/maxloc2_16_s1.c: New file.
        * generated/maxloc2_16_s4.c: New file.
        * generated/maxloc2_4_s1.c: New file.
        * generated/maxloc2_4_s4.c: New file.
        * generated/maxloc2_8_s1.c: New file.
        * generated/maxloc2_8_s4.c: New file.
        * generated/minloc0_16_s1.c: New file.
        * generated/minloc0_16_s4.c: New file.
        * generated/minloc0_4_s1.c: New file.
        * generated/minloc0_4_s4.c: New file.
        * generated/minloc0_8_s1.c: New file.
        * generated/minloc0_8_s4.c: New file.
        * generated/minloc1_16_s1.c: New file.
        * generated/minloc1_16_s4.c: New file.
        * generated/minloc1_4_s1.c: New file.
        * generated/minloc1_4_s4.c: New file.
        * generated/minloc1_8_s1.c: New file.
        * generated/minloc1_8_s4.c: New file.
        * generated/minloc2_16_s1.c: New file.
        * generated/minloc2_16_s4.c: New file.
        * generated/minloc2_4_s1.c: New file.
        * generated/minloc2_4_s4.c: New file.
        * generated/minloc2_8_s1.c: New file.
        * generated/minloc2_8_s4.c: New file.
        * m4/iforeach-s.m4: New file.
        * m4/ifunction-s.m4: New file.
        * m4/maxloc0s.m4: New file.
        * m4/maxloc1s.m4: New file.
        * m4/maxloc2s.m4: New file.
        * m4/minloc0s.m4: New file.
        * m4/minloc1s.m4: New file.
        * m4/minloc2s.m4: New file.
* gfortran.map: Add new functions.
* libgfortran.h: Add gfc_array_s1 and gfc_array_s4.

2017-11-22  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/36313
* check.c (int_or_real_or_char_check_f2003): New function.
* iresolve.c (gfc_resolve_maxloc): Add number "2" for
character arguments and rank-zero return value.
(gfc_resolve_minloc): Likewise.
* trans-intrinsic.c (gfc_conv_intrinsic_minmaxloc): Handle case of
character arguments and rank-zero return value by removing
unneeded arguments and calling the library function.

2017-11-22  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR fortran/36313
* gfortran.dg/maxloc_string_1.f90: New test.
* gfortran.dg/minloc_string_1.f90: New test.

From-SVN: r255070

vxworks.c (vxworks_override_options): Pick default dwarf version from macro value, VXWORKS_DWARF_VERSION_DEFAULT.

2017-11-22  Olivier Hainque  <hainque@adacore.com>

       * config/vxworks.c (vxworks_override_options): Pick default
dwarf version from macro value, VXWORKS_DWARF_VERSION_DEFAULT.
       * config/vxworks.h: Define VXWORKS_DWARF_VERSION_DEFAULT and
       DWARF_GNAT_ENCODINGS_DEFAULT.
       * config/vxworksae.h: Likewise.

From-SVN: r255069

Avoid two_valued_val_range_p for pointers.

2017-11-22  Marc Glisse  <marc.glisse@inria.fr>

PR tree-optimization/83104
gcc/
* vr-values.c (simplify_stmt_using_ranges): Check integral argument,
not result.

gcc/testsuite/
* gcc.c-torture/compile/pr83104.c: New file.

From-SVN: r255068

[ARM] Fix more -Wreturn-type fallout (gcc.target/arm/pr56184.C).

2017-11-22  Christophe Lyon  <christophe.lyon@linaro.org>

* gcc.target/arm/pr56184.C: Add -Wno-return-type to dg-options.

From-SVN: r255067

re PR c++/60336 (empty struct value is passed differently in C and C++)

PR c++/60336
PR middle-end/67239
PR target/68355
* c-decl.c (grokdeclarator): Set DECL_PADDING_P on unnamed bit-fields.

* class.c (layout_class_type): Set DECL_PADDING_P on padding.
* decl.c (cxx_init_decl_processing): Set TRANSLATION_UNIT_WARN_EMPTY_P.
(grokdeclarator): Set DECL_PADDING_P on unnamed bit-fields.

* lto.c (compare_tree_sccs_1): Compare TYPE_EMPTY_P and DECL_PADDING_P.

* calls.c (initialize_argument_information): Call
warn_parameter_passing_abi target hook.
(store_one_arg): Use 0 for empty record size.  Don't push 0 size
argument onto stack.
(must_pass_in_stack_var_size_or_pad): Return false for empty types.
* common.opt: Update -fabi-version description.
* config/i386/i386.c (init_cumulative_args): Set cum->warn_empty.
(ix86_gimplify_va_arg): Call arg_int_size_in_bytes instead of
int_size_in_bytes.
(ix86_is_empty_record): New function.
(ix86_warn_parameter_passing_abi): New function.
(TARGET_EMPTY_RECORD_P): Redefine.
(TARGET_WARN_PARAMETER_PASSING_ABI): Redefine.
* config/i386/i386.h (CUMULATIVE_ARGS): Add warn_empty.
* doc/tm.texi: Regenerated.
* doc/tm.texi.in (TARGET_EMPTY_RECORD_P,
TARGET_WARN_PARAMETER_PASSING_ABI): Add.
* dwarf2out.c (get_ultimate_context): Move to tree.c.
* explow.c (hard_function_value): Call arg_int_size_in_bytes
instead of int_size_in_bytes.
* expr.c (copy_blkmode_to_reg): Likewise.
* function.c (aggregate_value_p): Return 0 for empty types.
(assign_parm_find_entry_rtl): Call warn_parameter_passing_abi target hook.
(locate_and_pad_parm): Call arg size_in_bytes instead
size_in_bytes.
* lto-streamer-out.c (hash_tree): Hash TYPE_EMPTY_P and DECL_PADDING_P.
* stor-layout.c (finalize_type_size): Set TYPE_EMPTY_P.
* target.def (empty_record_p, warn_parameter_passing_abi): New target
hooks.
* targhooks.c (hook_void_CUMULATIVE_ARGS_tree): New hook.
(std_gimplify_va_arg_expr): Skip empty records.  Call
arg_size_in_bytes instead size_in_bytes.
* targhooks.h (hook_void_CUMULATIVE_ARGS_tree): Declare.
* tree-core.h (tree_type_common): Add empty_flag.
(tree_decl_common): Update comments.
* tree-streamer-in.c (unpack_ts_decl_common_value_fields): Stream
DECL_PADDING_P.
(unpack_ts_type_common_value_fields): Stream TYPE_EMPTY_P.
* tree-streamer-out.c (pack_ts_decl_common_value_fields): Stream
DECL_PADDING_P.
(pack_ts_type_common_value_fields): Stream TYPE_EMPTY_P.
* tree.c (default_is_empty_type): New function.
(default_is_empty_record): New function.
(arg_int_size_in_bytes): New function.
(arg_size_in_bytes): New function.
(get_ultimate_context): New function.
* tree.h: Define TYPE_EMPTY_P, DECL_PADDING_P and
TRANSLATION_UNIT_WARN_EMPTY_P.
(default_is_empty_record, arg_int_size_in_bytes,
arg_size_in_bytes, get_ultimate_context): Declare.

* g++.dg/abi/empty12.C: New test.
* g++.dg/abi/empty12.h: New test.
* g++.dg/abi/empty12a.c: New test.
* g++.dg/abi/empty13.C: New test.
* g++.dg/abi/empty13.h: New test.
* g++.dg/abi/empty13a.c: New test.
* g++.dg/abi/empty14.C: New test.
* g++.dg/abi/empty14.h: New test.
* g++.dg/abi/empty14a.c: New test.
* g++.dg/abi/empty15.C: New test.
* g++.dg/abi/empty15.h: New test.
* g++.dg/abi/empty15a.c: New test.
* g++.dg/abi/empty16.C: New test.
* g++.dg/abi/empty16.h: New test.
* g++.dg/abi/empty16a.c: New test.
* g++.dg/abi/empty17.C: New test.
* g++.dg/abi/empty17.h: New test.
* g++.dg/abi/empty17a.c: New test.
* g++.dg/abi/empty18.C: New test.
* g++.dg/abi/empty18.h: New test.
* g++.dg/abi/empty18a.c: New test.
* g++.dg/abi/empty19.C: New test.
* g++.dg/abi/empty19.h: New test.
* g++.dg/abi/empty19a.c: New test.
* g++.dg/abi/empty20.C: New test.
* g++.dg/abi/empty21.C: New test.
* g++.dg/abi/empty22.C: New test.
* g++.dg/abi/empty22.h: New test.
* g++.dg/abi/empty22a.c: New test.
* g++.dg/abi/empty23.C: New test.
* g++.dg/abi/empty24.C: New test.
* g++.dg/abi/empty25.C: New test.
* g++.dg/abi/empty25.h: New test.
* g++.dg/abi/empty25a.c: New test.
* g++.dg/abi/empty26.C: New test.
* g++.dg/abi/empty26.h: New test.
* g++.dg/abi/empty26a.c: New test.
* g++.dg/abi/empty27.C: New test.
* g++.dg/abi/empty28.C: New test.
* g++.dg/abi/pr60336-1.C: New test.
* g++.dg/abi/pr60336-10.C: New test.
* g++.dg/abi/pr60336-11.C: New test.
* g++.dg/abi/pr60336-12.C: New test.
* g++.dg/abi/pr60336-2.C: New test.
* g++.dg/abi/pr60336-3.C: New test.
* g++.dg/abi/pr60336-4.C: New test.
* g++.dg/abi/pr60336-5.C: New test.
* g++.dg/abi/pr60336-6.C: New test.
* g++.dg/abi/pr60336-7.C: New test.
* g++.dg/abi/pr60336-8.C: New test.
* g++.dg/abi/pr60336-9.C: New test.
* g++.dg/abi/pr68355.C: New test.
* g++.dg/lto/pr60336_0.C: New test.

Co-Authored-By: H.J. Lu <hongjiu.lu@intel.com>
Co-Authored-By: Jason Merrill <jason@redhat.com>
From-SVN: r255066

[ARM] Factor out CMSE register clearing code

Functions cmse_nonsecure_call_clear_caller_saved and
cmse_nonsecure_entry_clear_before_return both contain very similar code
to clear registers. What's worse, they differ slightly at times so if a
bug is found in one careful thoughts is needed to decide whether the
other function needs fixing too.

This commit addresses the situation by factoring the two pieces of code
into a new function. In doing so the code generated to clear VFP
registers in cmse_nonsecure_call now uses the same sequence as
cmse_nonsecure_entry functions. Tests expectation are thus updated
accordingly.

2017-11-22  Thomas Preud'homme  <thomas.preudhomme@arm.com>

    gcc/
    * config/arm/arm.c (cmse_clear_registers): New function.
    (cmse_nonsecure_call_clear_caller_saved): Replace register clearing
    code by call to cmse_clear_registers.
    (cmse_nonsecure_entry_clear_before_return): Likewise.

    gcc/testsuite/
    * gcc.target/arm/cmse/mainline/hard-sp/cmse-13.c: Adapt expectations
    to vmov instructions now generated.
    * gcc.target/arm/cmse/mainline/hard-sp/cmse-7.c: Likewise.
    * gcc.target/arm/cmse/mainline/hard-sp/cmse-8.c: Likewise.
    * gcc.target/arm/cmse/mainline/hard/cmse-13.c: Likewise.
    * gcc.target/arm/cmse/mainline/hard/cmse-7.c: Likewise.
    * gcc.target/arm/cmse/mainline/hard/cmse-8.c: Likewise.

From-SVN: r255065

2017-11-22  Tamar Christina  <tamar.christina@arm.com>

* config/arm/arm_neon.h (vdot_u32, vdotq_u32)
(vdot_s32, vdotq_s32): New.
(vdot_lane_u32, vdotq_lane_u32): New.
(vdot_lane_s32, vdotq_lane_s32): New.

gcc/testsuite/
2017-11-22  Tamar Christina  <tamar.christina@arm.com>

* gcc.target/arm/simd/vdot-compile.c: New.
* gcc.target/arm/simd/vect-dot-qi.h: New.
* gcc.target/arm/simd/vect-dot-s8.c: New.
* gcc.target/arm/simd/vect-dot-u8.c: New

From-SVN: r255064

target-supports.exp (check_effective_target_bswap16): Do not skip on alpha and x86 targets.

* lib/target-supports.exp (check_effective_target_bswap16):
Do not skip on alpha and x86 targets.

From-SVN: r255063

compiler: make comparison operator() methods const

    This is required for new versions of libstdc++ in C++17 mode.

    Fixes GCC PR 83102.

    Reviewed-on: https://go-review.googlesource.com/79396

From-SVN: r255062

[ARM] Fix test armv8_2-fp16-move-1.c

2017-11-22  Sudakshina Das  <sudi.das@arm.com>

* gcc.target/arm/armv8_2-fp16-move-1.c: Add -mfloat-abi=hard option.

From-SVN: r255061

PR82547: Undetected overflow for UNSIGNED wide_ints

wi::add_large and wi::sub_large weren't setting the overflow bit
correctly for unsigned operations if the result needed fewer HWIs
than the precision.

2017-11-22  Richard Sandiford  <richard.sandiford@linaro.org>

gcc/
PR middle-end/82547
* wide-int.cc (wi::add_large, wi::sub_large): Fix overflow detection
for unsigned values with fewer HWIs than the precision.
(test_overflow): New function.
(wide_int_cc_tests): Call it.

From-SVN: r255059

PR83004: Accidental change to pr81136.c for VECTOR_BITS==128

r254589 was supposed to leave tests unchanged for the default
VECTOR_BITS==128, but I must have got my sums wrong on pr81136.c.

2017-11-22  Richard Sandiford  <richard.sandiford@linaro.org>

gcc/testsuite/
PR testsuite/83004
* gcc.dg/vect/pr81136.c: Restore previous alignment of 32
in the default case.

From-SVN: r255058

Fix calculation of ptr_mode for MODE_PARTIAL_INT Pmode

This patch fixes a regression caused by r251469, where I'd incorrectly
converted a call to mode_for_size that sometimes needs MODE_PARTIAL_INTs.

2017-11-22  Richard Sandiford  <richard.sandiford@linaro.org>

gcc/
* emit-rtl.c (init_derived_machine_modes): Make sure ptr_mode
has the same mode class as Pmode.

From-SVN: r255057

arm.c (cmse_nonsecure_call_clear_caller_saved): Get rid of padding_bits_to_clear_ptr.

2017-11-22  Thomas Preud'homme  <thomas.preudhomme@arm.com>

    gcc/
    * config/arm/arm.c (cmse_nonsecure_call_clear_caller_saved): Get rid of
    padding_bits_to_clear_ptr.
    (cmse_nonsecure_entry_clear_before_return): Likewise.

From-SVN: r255056

[ARM] Use bitmap to control cmse_nonsecure_call register clearing

As part of r253256, cmse_nonsecure_entry_clear_before_return has been
rewritten to use auto_sbitmap instead of an integer bitfield to control
which register needs to be cleared. This commit continues this work in
cmse_nonsecure_call_clear_caller_saved.

2017-11-22  Thomas Preud'homme  <thomas.preudhomme@arm.com>

    gcc/
    * config/arm/arm.c (cmse_nonsecure_call_clear_caller_saved): Use
    auto_sbitap instead of integer bitfield to control register needing
    clearing.

From-SVN: r255055

re PR tree-optimization/83044 (ice in contains_struct_check)

PR tree-optimization/83044
* tree-vrp.c (vrp_prop::check_array_ref): If eltsize is not
INTEGER_CST or is 0, clear up_bound{,_p1} and later ignore tests
that need the upper bound.  Subtract offset from
get_addr_base_and_unit_offset only if positive and subtract it
before division by eltsize rather than after it.

* gcc.dg/pr83044.c: New test.
* c-c++-common/Warray-bounds.c (fb): Fix up MAX value.

From-SVN: r255054

re PR fortran/79072 (ICE with class(*) pointer function result and character value)

2017-11-22  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/79072
* trans-stmt.c (trans_associate_var): Weaken the over strong
condition for using the fake result decl.

2017-11-22  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/79072
* gfortran.dg/class_result_6.f90: New test for comment 10 of
the PR.

From-SVN: r255053

PR go/83102 relax std::set checks for invocable comparison object

PR go/83102
* include/bits/stl_tree.h (_Rb_tree): Relax invocable checks for
comparison object pre-C++17.

From-SVN: r255052

re PR debug/83084 (-fcompare-debug failure on ppc64le)

PR debug/83084
* valtrack.c (propagate_for_debug_subst, propagate_for_debug): Reset
debug insns if they would contain UNSPEC_VOLATILE or volatile asm.
(dead_debug_insert_temp): Likewise, but also ignore even non-volatile
asm.

* g++.dg/opt/pr83084.C: New test.

From-SVN: r255051

re PR middle-end/82875 (ICE at -Os on valid code on x86_64-linux-gnu: in find_widening_optab_handler_and_mode, at optabs-query.c:414)

PR middle-end/82875
* optabs.c (expand_doubleword_mult, expand_binop): Before calling
expand_binop with *mul_widen_optab, make sure at least one of the
operands doesn't have VOIDmode.

* gcc.dg/pr82875.c: New test.
* gcc.c-torture/compile/pr82875.c: New test.

From-SVN: r255050

re PR debug/83034 (ice in mem_loc_descriptor, at dwarf2out.c :15669)

PR debug/83034
* dwarf2out.c (mem_loc_descriptor): Handle VEC_SERIES.

* gcc.dg/pr83034.c: New test.

From-SVN: r255049

re PR rtl-optimization/82044 (runtime signed integer overflow in check_mem_read_rtx() and all_positions_needed_p() in dse.c)

PR rtl-optimization/82044
PR tree-optimization/82042
* dse.c (record_store): Check for overflow.
(check_mem_read_rtx): Properly check for overflow if width == -1, call
add_wild_read instead of clear_rhs_from_active_local_stores on
overflow and log it into dump_file.

From-SVN: r255048

gimple-iterator.c (gimple_find_edge_insert_loc): Ignore fake edges to exit when looking for a place to insert.

2017-11-22  Richard Biener  <rguenther@suse.de>

* gimple-iterator.c (gimple_find_edge_insert_loc): Ignore
fake edges to exit when looking for a place to insert.
* tree-ssa-pre.c (clear_expression_ids): Inline into callers
and remove.
(insert_into_preds_of_block): Commit edge insertion immediately,
assert that doesn't require new BBs.
(fini_pre): Release expressions.
(pass_pre::execute): Shuffle things around a bit, if the fn
is too large do not compute AVAIL either as this is really the
quadratic bit.

From-SVN: r255047

PR 83070 Fix -Wsign-compare warning

2017-11-22  Janne Blomqvist  <jb@gcc.gnu.org>

        PR libfortran/83070
        * intrinsics/eoshift0.c (eoshift0): Fix -Wsign-compare warning by
        making a_ex and r_ex index_type instead of size_t.

From-SVN: r255045

re PR tree-optimization/83089 (ICE: Segmentation fault (in instantiate_scev_name))

2017-11-22  Richard Biener  <rguenther@suse.de>

PR tree-optimization/83089
* tree-if-conv.c (pass_if_conversion::execute): If anything
changed reset SCEV and free the number of iteration estimates.

* gcc.dg/pr83089.c: New testcase.

From-SVN: r255044

libgo: don't use grep -q in mksigtab.sh

    Solaris grep does not support the -q option.

    Reviewed-on: https://go-review.googlesource.com/79239

From-SVN: r255042

Daily bump.

From-SVN: r255041

C: don't suggest names that came from earlier failures (PR c/83056)

PR c/83056 reports an issue affecting trunk and gcc-7 in which
the C frontend's implementation of lookup_name_fuzzy uses undeclared
identifiers as suggestions when encountering subsequent undeclared
identifiers.

The fix is to filter out the names bound to error_mark_node
in lookup_name_fuzzy.

The C++ frontend is unaffected, as it already does this.

gcc/c/ChangeLog:
PR c/83056
* c-decl.c (lookup_name_fuzzy): Don't suggest names that came from
earlier failed lookups.

gcc/testsuite/ChangeLog:
PR c/83056
* gcc.dg/spellcheck-pr83056.c: New test case.

From-SVN: r255038

re PR tree-optimization/82945 (add warning for passing non-strings to functions that expect string arguments)

gcc/ChangeLog:

PR tree-optimization/82945
* calls.h (warn_nonstring_bound): Remove unused function.

From-SVN: r255036