Daily bump.

From-SVN: r242941

i386.md (*and<mode>_1): Merge insn pattern from *andsi_1 and *andhi_1 using SWI24 mode iterator.

* config/i386/i386.md (*and<mode>_1): Merge insn pattern from
*andsi_1 and *andhi_1 using SWI24 mode iterator.  Use multi-line
output template string.
(*anddi_1): Use multi-line output template string.
(*andqi_1): Ditto.

From-SVN: r242938

re PR middle-end/78540 (ICE: in df_refs_verify, at df-scan.c:4062 with -O -march=core2)

PR middle-end/78540
* rtl.h (remove_reg_equal_equiv_notes): Return bool instead of void.
* rtlanal.c (remove_reg_equal_equiv_notes): Return true if any
note has been removed.
* postreload.c (reload_combine_recognize_pattern): If
remove_reg_equal_equiv_notes returns true, call df_notes_rescan.

* gcc.dg/pr78540.c: New test.

From-SVN: r242937

PR middle-end/78521 - [7 Regression] incorrect byte count in -Wformat-length...

PR middle-end/78521 - [7 Regression] incorrect byte count in -Wformat-length warning with non-constant width or precision
PR middle-end/78520 - missing warning for snprintf with size greater than INT_MAX

gcc/ChangeLog:

PR middle-end/78520
* gimple-ssa-sprintf.c (target_max_value): Remove.
(target_int_max, target_size_max): Use TYPE_MAX_VALUE.
(get_width_and_precision): New function.
(format_integer, format_floating, get_string_length, format_string):
Correct handling of width and precision with unknown value.
(format_directive): Add warning.
(pass_sprintf_length::compute_format_length): Allow for precision
to consist of a sole period with no asterisk or digits after it.

gcc/testsuite/ChangeLog:

PR middle-end/78520
* gcc.dg/tree-ssa/builtin-sprintf-5.c: Add test cases.
* gcc.dg/tree-ssa/builtin-sprintf-6.c: New test.
* gcc.dg/tree-ssa/builtin-sprintf-warn-1.c: Add test cases.
* gcc.dg/tree-ssa/builtin-sprintf-warn-3.c: Add test cases.

From-SVN: r242935

re PR target/74748 (libgcc_s.so.1 isn't created correctly for Blackfin FDPIC)

PR gcc/74748
* libgcc/config/bfin/libgcc-glibc.ver, libgcc/config/bfin/t-linux:
use generic linker version information on Blackfin.

2016-11-27  Iain Sandoe  <iain@codesourcery.com>

From-SVN: r242934

re PR target/74748 (libgcc_s.so.1 isn't created correctly for Blackfin FDPIC)

PR gcc/74748
* libgcc/mkmap-symver.awk: add support for skip_underscore

From-SVN: r242933

Fix typo.

From-SVN: r242931

re PR c++/72808 (ICE on valid c++ code in verify_type (gcc/tree.c:14047))

PR c++/72808
* decl.c (finish_enum_value_list): Call fixup_type_variants on
current_class_type after
insert_late_enum_def_into_classtype_sorted_fields.

* g++.dg/debug/pr72808.C: New test.

Co-Authored-By: Jason Merrill <jason@redhat.com>
From-SVN: r242930

re PR rtl-optimization/78546 (wrong code at -O2 and above)

PR rtl-optimization/78546
* simplify-rtx.c (neg_const_int): When negating most negative
number in mode wider than HOST_BITS_PER_WIDE_INT, use
simplify_const_unary_operation to produce CONST_DOUBLE or
CONST_WIDE_INT.
(simplify_plus_minus): Hanlde the case where neg_const_int
doesn't return a CONST_INT.

* gcc.dg/torture/pr78546-1.c: New test.
* gcc.dg/torture/pr78546-2.c: New test.

From-SVN: r242929

Fix PR78556 - left shift of negative values

Running bootstrap-ubsan on ppc64le shows many instances of e.g.:
 config/rs6000/rs6000.c:6217:36: runtime error: left shift of negative value -12301

        PR target/78556
* config/rs6000/rs6000.c (vspltis_constant): Add casts to avoid
left shifting of negative values.

From-SVN: r242928

re PR fortran/78298 (ICE in lookup_decl_in_outer_ctx, bei omp-low.c:4115)

PR fortran/78298
* tree-nested.c (convert_local_reference_stmt): After adding
shared (FRAME.NN) clause to omp parallel, task or target,
add it also to all outer omp parallel, task or target constructs.

* gfortran.dg/gomp/pr78298.f90: New test.

From-SVN: r242926

i386.md (UNSPEC_KMASKOP): New.

* config/i386/i386.md (UNSPEC_KMASKOP): New.
(UNSPEC_KMOV): Remove.
(kmovw): Expand to plain HImode move.
(k<any_logic:code><mode>): Rename from *k<logic><mode>. Use
register_operand predicates.  Tag pattern with UNSPEC_KMASKOP.
Remove corresponding clobber-removing splitter.
(*anddi_1): Remove mask register alternatives.
(*andsi_1): Ditto.
(*andhi_1): Ditto.
(*andqi_1): Ditto.
(*<any_or:code><mode>_1): Ditto.
(*<any_or:code>qi_1): Ditto.
(kandn<mode>): Use SWI1248_AVX512BW mode iterator.  Remove
general register alternatives.  Tag pattern with UNSPEC_KMASKOP.
Remove corresponding splitter to operation with general registers.
(*andn<SWI38:mode>): Rename from *bmi_andn_<mode>.
(*andn<SWI12:mode>): New pattern.
(*kxnor<mode>): Remove general register alternatives.  Tag pattern
with UNSPEC_KMASKOP.  Remove corresponding splitter to operation
with general registers.
(knot<mode>): New insn pattern.
(*one_cmpl<mode>2_1): Remove mask register alternatives.
(one_cmplqi2_1): Ditto.
(*k<any_lshift:code><mode>): Rename from *k<mshift><mode>3.
Tag pattern with UNSPEC_KMASKOP. Add mode attribute.
* config/i386/predicates.md (mask_reg_operand): Remove predicate.
* config/i386/sse.md (vec_unpacks_hi_hi): Update pattern
to generate kmaskop shift.
(vec_unpacks_hi_<mode>): Ditto.
* config/i386/i386-builtin.def (__builtin_ia32_kandhi):
Use CODE_FOR_kandhi.
(__builtin_ia32_knothi): Use CODE_FOR_knothi.
(__builtin_ia32_korhi): Use CODE_FOR_kiorhi.
(__builtin_ia32_kxorhi): Use CODE_FOR_kxorhi.

testsuite/ChangeLog:

* gcc.target/i386/bmi-andn-1a.c (dg-final): Update scan string.
* gcc.target/i386/bmi-andn-2a.c (dg-final): Ditto.

From-SVN: r242925

re PR c++/77591 (decltype(auto) and ternary operator allow returning local reference without a warning)

PR c++/77591
* typeck.c (maybe_warn_about_returning_address_of_local): Optimize
whats_returned through fold_for_warn.

* g++.dg/cpp1y/pr77591.C: New test.

From-SVN: r242924

* MAINTAINERS (nvptx): Remove self.

From-SVN: r242923

* gcc.dg/torture/pr78515.c: Ignore ABI extension warning.

From-SVN: r242922

tree-vrp.c (vrp_visit_assignment_or_call): Handle simplifications to SSA names via extract_range_from_ssa_name if allowed.

2016-11-28  Richard Biener  <rguenther@suse.de>

* tree-vrp.c (vrp_visit_assignment_or_call): Handle
simplifications to SSA names via extract_range_from_ssa_name
if allowed.

From-SVN: r242921

re PR tree-optimization/78542 (wrong code at -Og results in endless loop)

2016-11-28  Richard Biener  <rguenther@suse.de>

PR tree-optimization/78542
* tree-ssa-ccp.c (evaluate_stmt): Only valueize simplification
if allowed.

* gcc.dg/torture/pr78542.c: New testcase.

From-SVN: r242920

libgomp: remove config/nvptx/critical.c

* config/nvptx/critical.c: Delete to use generic implementation.

From-SVN: r242919

libgomp: fix OpenMP offloading to NVPTX and correct location of generic affinity.c

2016-11-28  Jonas Hahnfeld  <Hahnfeld@itc.rwth-aachen.de>

* config/linux/affinity.c [!HAVE_PTHREAD_AFFINITY_NP]: Include
../../affinity.c as fallback.
* config/nvptx/affinity.c: Delete to use fallback implementation.

From-SVN: r242918

combine.c (simplify_if_then_else): Simplify IF_THEN_ELSE that isolates a single bit...

gcc:
* combine.c (simplify_if_then_else): Simplify IF_THEN_ELSE
that isolates a single bit, even if the condition involves
subregs.

From-SVN: r242917

2016-11-28  Tamar Christina  <tamar.christina@arm.com>

* config/aarch64/aarch64-simd-builtins.def
(BSL_P): Added di and v2di mode.
* config/aarch64/arm_neon.h
(vsriq_n_p64, vsri_n_p64): Added poly type.
(vextq_p64, vext_p64): Likewise.
(vceq_p64, vbslq_p64, vbsl_p64): Likewise.

From-SVN: r242916

aarch64-builtins.c (TYPES_SETREGP): Added poly type.

2016-11-28  Tamar Christina  <tamar.christina@arm.com>

* config/aarch64/aarch64-builtins.c (TYPES_SETREGP): Added poly type.
(TYPES_GETREGP): Likewise.
(TYPES_SHIFTINSERTP): Likewise.
(TYPES_COMBINEP): Likewise.
(TYPES_STORE1P): Likewise.
* config/aarch64/aarch64-simd-builtins.def
(combine): Added poly generator.
(get_dregoi): Likewise.
(get_dregci): Likewise.
(get_dregxi): Likewise.
(ssli_n): Likewise.
(ld1): Likewise.
(st1): Likewise.
* config/aarch64/arm_neon.h
(poly64x1x2_t, poly64x1x3_t): New.
(poly64x1x4_t, poly64x2x2_t): Likewise.
(poly64x2x3_t, poly64x2x4_t): Likewise.
(poly64x1_t): Likewise.
(vcreate_p64, vcombine_p64): Likewise.
(vdup_n_p64, vdupq_n_p64): Likewise.
(vld2_p64, vld2q_p64): Likewise.
(vld3_p64, vld3q_p64): Likewise.
(vld4_p64, vld4q_p64): Likewise.
(vld2_dup_p64, vld3_dup_p64): Likewise.
(vld4_dup_p64, vsli_n_p64): Likewise.
(vsliq_n_p64, vst1_p64): Likewise.
(vst1q_p64, vst2_p64): Likewise.
(vst3_p64, vst4_p64): Likewise.
(__aarch64_vdup_lane_p64, __aarch64_vdup_laneq_p64): Likewise.
(__aarch64_vdupq_lane_p64, __aarch64_vdupq_laneq_p64): Likewise.
(vget_lane_p64, vgetq_lane_p64): Likewise.
(vreinterpret_p8_p64, vreinterpretq_p8_p64): Likewise.
(vreinterpret_p16_p64, vreinterpretq_p16_p64): Likewise.
(vreinterpret_p64_f16, vreinterpret_p64_f64): Likewise.
(vreinterpret_p64_s8, vreinterpret_p64_s16): Likewise.
(vreinterpret_p64_s32, vreinterpret_p64_s64): Likewise.
(vreinterpret_p64_f32, vreinterpret_p64_u8): Likewise.
(vreinterpret_p64_u16, vreinterpret_p64_u32): Likewise.
(vreinterpret_p64_u64, vreinterpret_p64_p8): Likewise.
(vreinterpretq_p64_f64, vreinterpretq_p64_s8): Likewise.
(vreinterpretq_p64_s16, vreinterpretq_p64_s32): Likewise.
(vreinterpretq_p64_s64, vreinterpretq_p64_f16): Likewise.
(vreinterpretq_p64_f32, vreinterpretq_p64_u8): Likewise.
(vreinterpretq_p64_u16, vreinterpretq_p64_u32): Likewise.
(vreinterpretq_p64_u64, vreinterpretq_p64_p8): Likewise.
(vreinterpret_f16_p64, vreinterpretq_f16_p64): Likewise.
(vreinterpret_f32_p64, vreinterpretq_f32_p64): Likewise.
(vreinterpret_f64_p64, vreinterpretq_f64_p64): Likewise.
(vreinterpret_s64_p64, vreinterpretq_s64_p64): Likewise.
(vreinterpret_u64_p64, vreinterpretq_u64_p64): Likewise.
(vreinterpret_s8_p64, vreinterpretq_s8_p64): Likewise.
(vreinterpret_s16_p64, vreinterpret_s32_p64): Likewise.
(vreinterpretq_s32_p64, vreinterpret_u8_p64): Likewise.
(vreinterpret_u16_p64, vreinterpretq_u16_p64): Likewise.
(vreinterpret_u32_p64, vreinterpretq_u32_p64): Likewise.
(vset_lane_p64, vsetq_lane_p64): Likewise.
(vget_low_p64, vget_high_p64): Likewise.
(vcombine_p64, vst2_lane_p64): Likewise.
(vst3_lane_p64, vst4_lane_p64): Likewise.
(vst2q_lane_p64, vst3q_lane_p64): Likewise.
(vst4q_lane_p64, vget_lane_p64): Likewise.
(vget_laneq_p64, vset_lane_p64): Likewise.
(vset_laneq_p64, vcopy_lane_p64): Likewise.
(vcopy_laneq_p64, vdup_n_p64): Likewise.
(vdupq_n_p64, vdup_lane_p64): Likewise.
(vdup_laneq_p64, vld1_p64): Likewise.
(vld1q_p64, vld1_dup_p64): Likewise.
(vld1q_dup_p64, vld1q_dup_p64): Likewise.
(vmov_n_p64, vmovq_n_p64): Likewise.
(vst3q_p64, vst4q_p64): Likewise.
(vld1_lane_p64, vld1q_lane_p64): Likewise.
(vst1_lane_p64, vst1q_lane_p64): Likewise.
(vcopy_laneq_p64, vcopyq_laneq_p64): Likewise.
(vdupq_laneq_p64): Likewise.

From-SVN: r242915

arm_neon.h (vget_lane_p64): New.

* config/arm/arm_neon.h (vget_lane_p64): New.

From-SVN: r242914

re PR tree-optimization/78343 (Loop is not eliminated)

2016-11-28  Richard Biener  <rguenther@suse.de>

PR tree-optimization/78343
* gcc.dg/tree-ssa/20030808-1.c: Fix dump to generate.
* gcc.dg/tree-ssa/20040305-1.c: Likewise.

From-SVN: r242913

[Darwin, config] Fix version number extraction to portable method

The method used in the applied patch caused configuration errors on freeBSD and hppa.

2016-11-28  Iain Sandoe  <iain@codesourcery.com>

PR target/71767
* configure.ac (with_ld64): Use portable method to extract the
major part of the version number.
* configure: Regenerated.

From-SVN: r242912

gimple-ssa-sprintf.c (build_intmax_type_nodes): Look at UINTMAX_TYPE rather than SIZE_TYPE.

* gimple-ssa-sprintf.c (build_intmax_type_nodes): Look at
UINTMAX_TYPE rather than SIZE_TYPE.  Add gcc_unreachable if
intmax_t couldn't be determined.
(format_integer): Make {,u}intmax_type_node no longer static,
initialize them only when needed.  For z and t use
signed_or_unsigned_type_for instead of assuming size_t and
ptrdiff_t have the same precision.

From-SVN: r242911

re PR lto/78211 (-fcompare-debug failure with -flto -fno-use-linker-plugin)

PR lto/78211
* ipa-icf.h (sem_item_optimizer): Add m_classes_vec member.
* ipa-icf.c (sem_item_optimizer::sem_item_optimizer): Initialize
it.
(sem_item_optimizer::~sem_item_optimizer): Traverse m_classes_vec
vector instead of traversing m_classes hash table.  Release
m_classes_vec.
(sem_item_optimizer::read_section, sem_item_optimizer::add_class):
Formatting fixes.
(sem_item_optimizer::get_group_by_hash): When inserting a new group,
add it also to m_classes_vec vector.
(sem_item_optimizer::remove_symtab_node,
sem_item_optimizer::build_hash_based_classes,
sem_item_optimizer::parse_nonsingleton_classes): Formatting fixes.
(sem_item_optimizer::subdivide_classes_by_equality,
sem_item_optimizer::subdivide_classes_by_sensitive_refs,
sem_item_optimizer::verify_classes): Traverse m_classes_vec vector
instead of traversing m_classes hash table.  Formatting fixes.
(sem_item_optimizer::traverse_congruence_split,
sem_item_optimizer::do_congruence_step_for_index,
sem_item_optimizer::do_congruence_step): Formatting fixes.
(sem_item_optimizer::process_cong_reduction): Traverse m_classes_vec
vector instead of traversing m_classes hash table.
(sem_item_optimizer::dump_cong_classes): Likewise.  Formatting fixes.
(sem_item_optimizer::merge_classes): Traverse m_classes_vec vector
instead of traversing m_classes hash table.

* g++.dg/ipa/pr78211.C: New test.

From-SVN: r242910

avr.c (out_movhi_r_mr): Only SBIW if X is not unused after.

* config/avr/avr.c (out_movhi_r_mr) [REG_X + PLUS]: Only SBIW if
X is not unused after.

From-SVN: r242909

re PR rtl-optimization/78120 (If conversion no longer performed)

PR rtl-optimization/78120
* rtlanal.c (insn_rtx_cost): Revert previous change.

From-SVN: r242908

re PR target/41076 ([avr] pessimal code for logical OR of 8-bit fields)

PR 41076
* config/avr/avr.md (SPLIT34): New mode iterator.
(bitop): New code iterator.
(*iorhi3.ashift8-*). New insn-and-split patterns.
(*movhi): Post-reload split reg = 0.
[!MOVW]: Post-reload split reg = reg.
(*mov<mode>) [SI,SF,PSI,SQ,USQ,SA,USA]: Post-reload split reg = reg.
(andhi3, andpsi3, andsi3): Post-reload split reg-reg operations.
(iorhi3, iorpsi3, iorsi3): Same.
(xorhi3, xorpsi3, xorsi3): Same.
* config/avr/avr.c (avr_rtx_costs_1) [IOR && HImode]: Adjust rtx
costs to *iorhi3.ashift8-* patterns.

From-SVN: r242907

PR c++/77907 - static init and PMF

* constexpr.c (cxx_eval_constant_expression): Use cp_fold_convert.

From-SVN: r242906

Daily bump.

From-SVN: r242905

re PR ada/78524 (failure of ACATS c41104a at -O0)

PR ada/78524
* gcc-interface/utils.c (max_size) <tcc_reference>: Add missing
conversion to original type in the PLACEHOLDER_EXPR case.

From-SVN: r242902

re PR ada/78531 (bootstrap broken with _FORTIFY_SOURCE enabled)

PR ada/78531
* namet.h (Max_Line_Length): Define.
(struct Bounded_String): Declare Chars with exact size.
(namet__get_decoded_name_string): Delete.
(Get_Decoded_Name_String): Likewise.
(casing__set_all_upper_case): Likewise.

From-SVN: r242901

re PR fortran/78474 ([F08] gfortran accepts invalid submodule syntax)

2016-11-27  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/78474
* module.c (gfc_match_submodule): If there is more than one
colon, it is a syntax error.

PR fortran/78331
* module.c (gfc_use_module): If an smod file does not exist it
is either because the module does not have a module procedure
interface or there is an error in the module.

2016-11-27  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/78474
* gfortran.dg/submodule_22.f08: New test.

PR fortran/78331
* gfortran.dg/submodule_21.f08: New test.

From-SVN: r242900

pr65655.C: Use dg-timeout-factor 2.

* g++.dg/torture/pr65655.C: Use dg-timeout-factor 2.

From-SVN: r242899

[Darwin] fix PR67710 by updating 'as' specs to handle newer assembler versions.

A/ Newer versions of ld64 check the min_version command, and newer versions of
   the system assembler inserts this in response to "-mmacosx-version-min=" on
   the assembler line. Unless one makes sensible versions, some object is bound
   to conflict.

B/ Additionally, there's a difference in behaviour between "as" and "ld" when
   presented with xx.yy.zz (ld truncates to xx.yy, as doesn't); net result is
   that one needs to pass a truncated version to "as".

So (if the assembler supports minversion commands)
(a) provide a truncated minversion (as asm_macosx_version_min, which is a
    driver-only var).
(b) pass this to "as"
(c) Update tests to determine 'HAVE_AS_MMACOSX_VERSION_MIN_OPTION'
    (Rainer's patch)
(d) For some reason the testcases are "run" (it's not obvious they need to be,
    they are checking compile-time issues)
  - anyway, to preserve the status quo, I've left them as exec.  However, the
    minimum version that can be code-gened for is target-dependent (there are no
    released x86 versions before 10.4, for example).  To avoid conflicts where
    the "as" is assuming some minimum, I've set the testversion to 10.5 (which
    is supported by all the archs we have)
(e) We need to ensure that libgcc and crts are generated with a sufficiently
    old minversion not to conflict.

gcc/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>
    Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>

PR target/67710
* config.in: Regenerate
* config/darwin-driver.c (darwin_driver_init): Emit a version string
for the assembler.
* config/darwin.h(ASM_MMACOSX_VERSION_MIN_SPEC): New, new tests.
* config/darwin.opt(asm_macosx_version_min): New.
* config/i386/darwin.h: Handle ASM_MMACOSX_VERSION_MIN_SPEC.
* configure: Regenerate
* configure.ac: Check for mmacosx-version-min handling.

gcc/testsuite/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>
    Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
    Dominique d'Humieres  <dominiq@lps.ens.fr>

PR target/67710
*  gcc.dg/darwin-minversion-1.c: Update min version check.
*  gcc.dg/darwin-minversion-2.c: Likewise.
*  gcc.dg/darwin-minversion-3.c: Likewise.

libgcc/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>
    Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>

PR target/67710
*  config/t-darwin: Default builds to 10.5 codegen.

Co-Authored-By: Dominique d'Humieres <dominiq@lps.ens.fr>
Co-Authored-By: Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
From-SVN: r242898

[Darwin] Fix PR57438 by avoiding empty function bodies and trailing labels.

A.
Empty function bodies causes two problems for Darwin's linker (i) zero-length
FDEs and (ii) coincident label addresses that might point to items of
differing weakness.

B.
Trailing local labels can be problematic when they end a function because
similarly they might apparently point to a following weak function, leading
to the linker concluding that there's a pointer-diff to a weak symbol
(which is not allowed).

Both conditions arise from __builtin_unreachable() lowering to a barrier.

The solution for both is to emit some finite amount of code; in the case of A
a trap is emitted, in the case of B a nop.

gcc/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>

PR target/57438
* config/i386/i386.c (ix86_code_end): Note that we emitted code
where the function might otherwise appear empty for picbase thunks.
(ix86_output_function_epilogue): If we find a zero-sized function
assume that reaching it is UB and trap.  If we find a trailing label
append a nop.
* config/rs6000/rs6000.c (rs6000_output_function_epilogue): If we
find a zero-sized function assume that reaching it is UB and trap.
If we find a trailing label, append a nop.

gcc/testsuite/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>

PR target/57438
* gcc.dg/pr57438-1.c: New Test.
* gcc.dg/pr57438-2.c: New Test.

From-SVN: r242897

[Testsuite] pr71767 Changes.

 Update Darwin tests to reflect the differences in symbol names and section
 usage.

gcc/testsuite/

2016-11-27  Dominique d'Humieres  <dominiq@lps.ens.fr>
    Iain Sandoe  <iain@codesourcery.com>

PR target/71767
* g++.dg/abi/key2.C: Adjust for changed Darwin sections and
linker-visible symbols.
* g++.dg/torture/darwin-cfstring-3.C: Likewise.
* gcc.dg/const-uniq-1.c: Likewise.
* gcc.dg/torture/darwin-cfstring-3.c: Likewise.
* gcc.target/i386/pr70799-1.c: Likewise.

Co-Authored-By: Iain Sandoe <iain@codesourcery.com>
From-SVN: r242896

[Darwin] Fix PR71767 - adjust the sections used where necessary.

 (much) Older Darwin linkers needed separate sections marked "coalesce" to
 allow for weak symbol coalescing.  This has not been needed for some time
 and is now deprecated, newer assemblers warn if the old coalesced sections
 are used.

gcc/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>

PR target/71767
* config/darwin-sections.def (picbase_thunk_section): New.
* config/darwin.c (darwin_init_sections): Set up picbase thunk
section. (darwin_rodata_section, darwin_objc2_section,
machopic_select_section, darwin_asm_declare_constant_name,
darwin_emit_weak_or_comdat, darwin_function_section): Don’t use
coalesced with newer linkers.
(darwin_override_options): Decide on usage of coalesed sections
on the basis of the target linker version.
* config/darwin.h (MIN_LD64_NO_COAL_SECTS): New.
* config/darwin.opt  (mtarget-linker): New.
* config/i386/i386.c (ix86_code_end): Do not force the thunks into
a coalesced section, instead use a thunks section.

From-SVN: r242895

[Darwin, config] Arrange for ld64 to be detected as Darwin's linker.

 This is an initial patch in a series that converts Darwin's configury
 to detect ld64 features, rather than the current process of hard-coding
 them on target system version.

 A ld64-compatible linker is currently required and assumed by Darwin.
 If a DEFAULT_LINKER is set via --with-ld= then this will be tested to
 see if it is ld64.

 The ld64 version is determined for the chosen ld and this is exported for
 use in setting a default value for -mtarget-linker (needed for run-time
 code-gen changes to section choices).
 The support for -rdynamic is converted to be detected at config time, or
 by the ld64 version if that is found.

gcc/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>

PR target/71767
* configure.ac (with-ld64): New var, set for Darwin, set on
detection of ld64, gcc_cv_ld64_export_dynamic: New, New test.
* config/darwin.h: Use LD64_HAS_DYNAMIC export. DEF_LD64: New,
define.
* config/darwin10.h(DEF_LD64): Update for this target version.
* config/darwin12.h(LINK_GCC_C_SEQUENCE_SPEC): Remove rdynamic
test.  (DEF_LD64): Update for this target version.
* configure: Regenerated.
* config.in: Regenerated.

From-SVN: r242894

[Darwin, ld64] Make PIC indirections and constant labels linker-visible.

Indirections:

  If we have a situation like:

global_weak_symbol:
  ....
Lnon_weak_local:
  ....

  ld64 will be unable to split this into two atoms (because the "L" makes
  the second symbol 'invisible').  This means that legitimate direct accesses
  to the second symbol will appear to be non-allowed direct accesses to an
  atom of type weak, global which are not allowed.

  To avoid this, we make the indirections have a leading 'l' (lower-case L)
  which has a special meaning: linker can see this and use it to determine
  atoms, but it is not placed into the final symbol table.

  The implementation here is somewhat heavy-handed in that it will also mark
  indirections to the __IMPORT,__pointers section the same way which is
  really unnecessary, since ld64 _can_ split those into atoms as they are
  fixed size.  FIXME: determine if this is a penalty worth extra code to
  fix.

Similarly, with:

  .const
weak_global_constant:
  ....

LCxx:
  ...

 ld64 can't split the second, causing a warning when it's directly
accessed.

gcc/

2016-11-27  Iain Sandoe  <iain@codesourcery.com>

PR target/71767
* config/darwin.c (imachopic_indirection_name): Make data
section indirections linker-visible.
* config/darwin.h (ASM_GENERATE_INTERNAL_LABEL): Make local
constant labels linker-visible.

From-SVN: r242893

re PR libstdc++/78441 ([variant] variant_alternative doesn't allow cv qualifiers)

PR libstdc++/78441
* include/std/variant: Propagate cv qualifications to types returned
by variant_alternative.
* testsuite/20_util/variant/compile.cc: Tests.

From-SVN: r242892

Daily bump.

From-SVN: r242891

tree.c (build_common_tree_nodes): Initialize ptrdiff_type_node.

2016-11-26  Prathamesh Kulkarni  <prathamesh.kulkarni@linaro.org>

* tree.c (build_common_tree_nodes): Initialize ptrdiff_type_node.
(free_lang_data): Remove assignment to ptrdiff_type_node.
c-family/
* c-common.c (c_common_nodes_and_builtins): Remove initialization of
ptrdiff_type_node.
lto/
* lto-lang.c (lto_init): Remove initialization of ptrdiff_type_node.

From-SVN: r242888

Daily bump.

From-SVN: r242887

For typo in documentation wrt. C_PTRDIFF_T constant

2016-11-25  Janne Blomqvist  <jb@gcc.gnu.org>

       * intrinsic.texi: Fix ptrdiff_t typo in ISO_C_BINDING constants
       table.

From-SVN: r242884

re PR rtl-optimization/78438 (incorrect comparison optimization)

PR rtl-optimization/78438
PR rtl-optimization/78477
* gcc.c-torture/execute/pr78438.c: New test.
* gcc.c-torture/execute/pr78477.c: New test.

From-SVN: r242883

re PR rtl-optimization/78526 (ICE: in decompose, at rtl.h:2117 with -g -mavx512bw)

PR rtl-optimization/78526
* simplify-rtx.c (simplify_immed_subreg): Don't use wi::extract_uhwi
beyond val's precision.

* gcc.dg/pr78526.c: New test.

From-SVN: r242882

re PR rtl-optimization/77541 (wrong code with 512bit vectors of int128 @ -O1)

2016-11-25  Vladimir Makarov  <vmakarov@redhat.com>

PR rtl-optimization/77541
* gcc.target/i386/pr77541.c: Change target to int128.

From-SVN: r242881

re PR fortran/60853 ([OOP] Failure to disambiguate generic with unlimited polymorphic)

2016-11-25  Janus Weil  <janus@gcc.gnu.org>

PR fortran/60853
* interface.c (gfc_compare_interfaces): Remove bad special case for
unlimited polymorphism. Refactor for loop.

2016-11-25  Janus Weil  <janus@gcc.gnu.org>

PR fortran/60853
* gfortran.dg/typebound_assignment_8.f90: New test case.

From-SVN: r242880

re PR rtl-optimization/78527 (ice on valid C code at -O3 in both 32-bit and 64-bit modes on x86_64-linux-gnu (internal compiler error: in smallest_mode_for_size, at stor-layout.c:364))

PR rtl-optimization/78527
* combine.c (make_compound_operation_int): Ignore LSHIFTRT with
out of bounds shift count.

* gcc.c-torture/compile/pr78527.c: New test.

From-SVN: r242879

Fix documentation reference (PR web/71666)

PR web/71666
* doc/invoke.texi (-fprofile-use): Fix reference to a section
where -fprofile-generate is documented.

From-SVN: r242878

Don't use priority {cd}tors if not supported by a target (PR

PR gcov-profile/78086
* g++.dg/gcov/pr16855.C: Clean up the test case.
* g++.dg/gcov/pr16855-priority.C: New test.
* coverage.c (build_init_ctor): Don't use priority {cd}tors if
not supported by a target.  Set priority to 100 if possible.
(build_gcov_exit_decl): Likewise.

From-SVN: r242877

re PR ipa/78515 (ICE: in fold_binary_loc, at fold-const.c:8999 with -Os -mavx512bw)

2016-11-25  Richard Biener  <rguenther@suse.de>

PR ipa/78515
* ipa-prop.c (compute_complex_assign_jump_func): Properly identify
unary, binary and single RHSs.
* tree.def (BIT_INSERT_EXPR): Adjust tree code name.

* gcc.dg/torture/pr78515.c: New testcase.

From-SVN: r242876

[multiple changes]

2016-11-25  Andre Vehreschild  <vehre@gcc.gnu.org>
Paul Thomas  <pault@gcc.gnu.org>

PR fortran/78293
* trans-expr.c (gfc_conv_procedure_call): Prepend deallocation
of alloctable components to post, rather than adding to
se->post.
* trans-stmt.c (gfc_trans_allocate): Move deallocation of expr3
allocatable components so that all expr3s are visited.

2016-11-25  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/78293
* gfortran.dg/allocatable_function_10.f90: New test.
* gfortran.dg/class_array_15.f03: Increase builtin_free count
from 11 to 12.

From-SVN: r242875

re PR middle-end/78507 (ICE: tree check: expected integer_type or enumeral_type or boolean_type or real_type or fixed_point_type, have pointer_type in int_fits_type_p, at tree.c:9153)

PR middle-end/78507
PR middle-end/78510
PR middle-end/78517
* match.pd ((cond (cmp (convert1? @1) @3) (convert2? @1) @2)): Use
cmp directly, rather than cmp_code.  Initialize code to ERROR_MARK
and set it to result code if transformation is valid.  Use code EQ
directly in last simplification case.

gcc/testsuite
PR middle-end/78507
PR middle-end/78510
PR middle-end/78517
* g++.dg/torture/pr78507.C: New test.
* gcc.dg/torture/pr78510.c: New test.
* gcc.dg/torture/pr78517.c: New test.

From-SVN: r242874

gimple-fold.c (fold_stmt_1): Check may_propagate_copy before valueizing return stmts.

2016-11-25  Richard Biener  <rguenther@suse.de>

* gimple-fold.c (fold_stmt_1): Check may_propagate_copy
before valueizing return stmts.

From-SVN: r242873

re PR tree-optimization/78343 (Loop is not eliminated)

2016-11-24  Richard Biener  <rguenther@suse.de>

PR tree-optimization/78343
* passes.def: Add CD-DCE pass after loop splitting.
* tree-ssa-dce.c (find_obviously_necessary_stmts): Move
SCEV init/finalize ...
(perform_tree_ssa_dce): ... here.  Deal with being
executed inside the loop pipeline in aggressive mode.

* gcc.dg/tree-ssa/sccp-2.c: New testcase.
* gcc.dg/autopar/uns-outer-6.c: Adjust.
* gcc.dg/tree-ssa/20030808-1.c: Likewise.
* gcc.dg/tree-ssa/20040305-1.c: Likewise.
* gcc.dg/vect/pr38529.c: Likewise.

From-SVN: r242872

Remove conflict marker

From-SVN: r242871

Improve comment for struct symbolic_number in bswap pass

2016-11-25  Thomas Preud'homme  <thomas.preudhomme@arm.com>

    gcc/
    * tree-ssa-math-opts.c (struct symbolic_number): Improve comment.

From-SVN: r242870

Fix PR77673: bswap loads passed end of object

2016-11-25  Thomas Preud'homme  <thomas.preudhomme@arm.com>

    gcc/
    PR tree-optimization/77673
    * tree-ssa-math-opts.c (struct symbolic_number): Add new src field.
    (init_symbolic_number): Initialize src field from src parameter.
    (perform_symbolic_merge): Select most dominated statement as the
    source statement.  Set src field of resulting n structure from the
    input src with the lowest address.
    (find_bswap_or_nop): Rename source_stmt into ins_stmt.
    (bswap_replace): Rename src_stmt into ins_stmt.  Initially get source
    of load from src field rather than insertion statement.  Cancel
    optimization if statement analyzed is not dominated by the insertion
    statement.
    (pass_optimize_bswap::execute): Rename src_stmt to ins_stmt.  Compute
    dominance information.

    gcc/testsuite/
    PR tree-optimization/77673
    * gcc.dg/pr77673.c: New test.

From-SVN: r242869

re PR ada/67205 (eliminate No_Implicit_Dynamic_Code restriction violations)

PR ada/67205
* config/mips/mips.c (TARGET_CUSTOM_FUNCTION_DESCRIPTORS): Define.

From-SVN: r242868

[PR 70965] Schedule extra rebuild_cgraph_edges

2016-11-25  Martin Jambor  <mjambor@suse.cz>

PR tree-optimization/70965
* passes.def (pass_build_ssa_passes): Add pass_rebuild_cgraph_edges.

gcc/testsuite/
* g++.dg/pr70965.C: New test.

From-SVN: r242867

[Patch i386] PR78509 - TARGET_C_EXCESS_PRECISION should not return
 "unpredictable" for EXCESS_PRECISION_TYPE_STANDARD

gcc/

PR target/78509
* config/i386/i386.c (i386_excess_precision): Do not return
FLT_EVAL_METHOD_UNPREDICTABLE when "type" is
EXCESS_PRECISION_TYPE_STANDARD.
* target.def (excess_precision): Document that targets should
not return FLT_EVAL_METHOD_UNPREDICTABLE when "type" is
EXCESS_PRECISION_TYPE_STANDARD or EXCESS_PRECISION_TYPE_FAST.
Fix typo in first sentence.
* doc/tm.texi: Regenerate.

From-SVN: r242866

re PR tree-optimization/78396 (gcc.dg/vect/bb-slp-cond-1.c FAILs after fix for PR77848)

2016-11-25  Richard Biener  <rguenther@suse.de>

PR tree-optimization/78396
* tree-vectorizer.c (vectorize_loops): When the if-converted
body contains masked loads or stores do not attempt to
basic-block-vectorize it.

From-SVN: r242865

re PR gcov-profile/78467 (gcc.dg/tree-prof/comp-goto-1.c FAILs)

PR gcov-profile/78467
* gcc.dg/tree-prof/comp-goto-1.c (insn_t): Change offset to
signed int.

Co-Authored-By: Andreas Schwab <schwab@linux-m68k.org>
From-SVN: r242864

Tweak LRA handling of shared spill slots

The previous code processed the users of a stack slot in order of
decreasing size and allocated the slot based on the first user.
This seems a bit dangerous, since the ordering is based on the
mode of the biggest reference while the allocation is based also
on the size of the register itself (which I think could be larger).

That scheme doesn't scale well to polynomial sizes, since there's
no guarantee that the order of the sizes is known at compile time.
This patch instead records an upper bound on the size required
by all users of a slot.  It also records the maximum alignment
requirement.

gcc/
2016-11-15  Richard Sandiford  <richard.sandiford@arm.com>
    Alan Hayward  <alan.hayward@arm.com>
    David Sherwood  <david.sherwood@arm.com>

* function.h (spill_slot_alignment): Declare.
* function.c (spill_slot_alignment): New function.
* lra-spills.c (slot): Add align and size fields.
(assign_mem_slot): Use them in the call to assign_stack_local.
(add_pseudo_to_slot): Update the fields.
(assign_stack_slot_num_and_sort_pseudos): Initialise the fields.

Co-Authored-By: Alan Hayward <alan.hayward@arm.com>
Co-Authored-By: David Sherwood <david.sherwood@arm.com>
From-SVN: r242863

Set mode of decimal floats before calling layout_type

Previously decimal floating-point types were created and laid
out as binary floating-point types, then the caller changed
the mode to a decimal mode later.  The problem with that
approach is that not all targets support an equivalent binary
floating-point mode.  When they didn't, we would give the
type BLKmode and lay it out as a zero-sized type.

This probably had no effect in practice.  If a target doesn't
support a binary mode then it's unlikely to support the decimal
equivalent either.  However, with the stricter mode checking
added by later patches, we would assert if a scalar floating-
point type didn't have a scalar floating-point mode.

gcc/
2016-11-16  Richard Sandiford  <richard.sandiford@arm.com>
    Alan Hayward  <alan.hayward@arm.com>
    David Sherwood  <david.sherwood@arm.com>

* stor-layout.c (layout_type): Allow the caller to set the mode of
a float type.  Only choose one here if the mode is still VOIDmode.
* tree.c (build_common_tree_nodes): Set the type mode of decimal
floats before calling layout_type.
* config/rs6000/rs6000.c (rs6000_init_builtins): Likewise.

Co-Authored-By: Alan Hayward <alan.hayward@arm.com>
Co-Authored-By: David Sherwood <david.sherwood@arm.com>
From-SVN: r242862

Add run tests for recent sibcall patches

gcc/testsuite/
* gcc.dg/tree-ssa/tailcall-7-run.c: New test.
* gcc.dg/tree-ssa/tailcall-8-run.c: Likewise.

From-SVN: r242861

Tighten check for whether sibcall references local variables

This loop:

      /* Make sure the tail invocation of this function does not refer
         to local variables.  */
      FOR_EACH_LOCAL_DECL (cfun, idx, var)
        {
          if (TREE_CODE (var) != PARM_DECL
              && auto_var_in_fn_p (var, cfun->decl)
              && (ref_maybe_used_by_stmt_p (call, var)
                  || call_may_clobber_ref_p (call, var)))
            return;
        }

triggered even for local variables that are passed by value.
This meant that we didn't allow local aggregates to be passed
to a sibling call but did (for example) allow global aggregates
to be passed.

I think the loop is really checking for indirect references,
so should be able to skip any variables that never have their
address taken.

gcc/
* tree-tailcall.c (find_tail_calls): Allow calls to reference
local variables if all references are known to be direct.

gcc/testsuite/
* gcc.dg/tree-ssa/tailcall-8.c: New test.

From-SVN: r242860

Fix bogus pr64277.c failure for avr

The smaller int size for the avr target breaks the test's
expectation on the number of iterations. The failure goes
away if 32 bit ints are used in place of a plain int.

Fix by conditionally typedef int32_t to __INT32_TYPE__ for targets
with int size < 4,  and then use int32_t everywhere.

gcc/testsuite
016-11-25  Senthil Kumar Selvaraj  <senthil_kumar.selvaraj@atmel.com>

* gcc.dg/pr64277.c: Use __INT32_TYPE__ for targets
with sizeof(int) < 4.

From-SVN: r242859

re PR middle-end/78501 (SEGV in vrp_val_max)

2016-11-25  Jakub Jelinek  <jakub@redhat.com>
    Prathamesh Kulkarni  <prathamesh.kulkarni@linaro.org>

PR middle-end/78501
* tree-vrp.c (extract_range_basic): Check for ptrdiff_type_node to be
non null and it's precision matches precision of lhs's type.

Co-Authored-By: Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
From-SVN: r242858

Daily bump.

From-SVN: r242857

PR tree-optimization/78476 - snprintf(0, 0, ...) with known arguments not optimized away

gcc/testsuite/ChangeLog:

PR tree-optimization/78476
* gcc.dg/tree-ssa/builtin-sprintf-5.c: New test.

gcc/ChangeLog:

PR tree-optimization/78476
* gimple-ssa-sprintf.c (struct pass_sprintf_length::call_info):
Add a member.
(handle_gimple_call): Adjust signature.
(try_substitute_return_value): Remove calls to bounded functions
with zero buffer size whose result is known.
(pass_sprintf_length::execute): Adjust call to handle_gimple_call.

From-SVN: r242854

Fix bootstrap with !ASM_OUTPUT_MAX_SKIP_ALIGN

* varasm.c (assemble_start_function): Wrap align_log definition in
ASM_OUTPUT_MAX_SKIP_ALIGN.

From-SVN: r242853

i386.md (wide AND insn to QImode splitter): Use explicit mode macros.

* config/i386/i386.md (wide AND insn to QImode splitter): Use
explicit mode macros.
(wide OR insn to QImode splitter): Ditto.

From-SVN: r242852

config.guess: Import latest version.

* config.guess: Import latest version.
* config.sub: Likewise.

From-SVN: r242851

* config/i386/i386.md: Move some more patterns around.

From-SVN: r242850

re PR rtl-optimization/77541 (wrong code with 512bit vectors of int128 @ -O1)

2016-11-24  Vladimir Makarov  <vmakarov@redhat.com>

PR rtl-optimization/77541
* lra-constraints.c (struct input_reload): Add field match_p.
(get_reload_reg): Check modes of input reloads to generate unique
value reload pseudo.
(match_reload): Add input reload pseudo for the current insn.

2016-11-24  Vladimir Makarov  <vmakarov@redhat.com>

PR rtl-optimization/77541
* gcc.target/i386/pr77541.c: New.

From-SVN: r242848

re PR fortran/78500 (ICE in gfc_check_vardef_context, at fortran/expr.c:5289)

2016-11-24  Steven G. Kargl  <kargl@gcc.gnu.org>

PR fortran/78500
* expr.c (gfc_check_vardef_contextm): Fix NULL pointer dereference.
* interface.c (matching_typebound_op): Ditto.

2016-11-24  Steven G. Kargl  <kargl@gcc.gnu.org>

PR fortran/78500
* gfortran.dg/class_result_4.f90: New test.

From-SVN: r242846

[Patch AArch64 13/17] Enable _Float16 for AArch64

gcc/

* config/aarch64/aarch64-c.c (aarch64_update_cpp_builtins): Update
__FLT_EVAL_METHOD__ and __FLT_EVAL_METHOD_C99__ when we switch
architecture levels.
* config/aarch64/aarch64.c (aarch64_promoted_type): Only promote
the aarch64_fp16_type_node, not all HFmode types.
(aarch64_libgcc_floating_mode_supported_p): Support HFmode.
(aarch64_scalar_mode_supported_p): Likewise.
(aarch64_excess_precision): New.
(TARGET_LIBGCC_FLOATING_MODE_SUPPORTED_P): Define.
(TARGET_SCALAR_MODE_SUPPORTED_P): Likewise.
(TARGET_C_EXCESS_PRECISION): Likewise.

gcc/testsuite/

* gcc.target/aarch64/_Float16_1.c: New.
* gcc.target/aarch64/_Float16_2.c: Likewise.
* gcc.target/aarch64/_Float16_3.c: Likewise.

From-SVN: r242845

[Patch libgcc AArch64 12/17] Enable hfmode soft-float conversions and truncations

gcc/

* config/aarch64/aarch64-c.c (aarch64_scalar_mode_supported_p): New.
(TARGET_SCALAR_MODE_SUPPORTED_P): Define.

libgcc/

* config/aarch64/sfp-machine.h (_FP_NANFRAC_H): Define.
(_FP_NANSIGN_H): Likewise.
* config/aarch64/t-softfp (softfp_extensions): Add hftf.
(softfp_truncations): Add tfhf.
(softfp_extras): Add required conversion functions.

From-SVN: r242844

[Patch AArch64 11/17] Add floatdihf2 and floatunsdihf2 patterns

gcc/

* config/aarch64/aarch64.md (<optab>sihf2): Convert to expand.
(<optab>dihf2): Likewise.
(aarch64_fp16_<optab><mode>hf2): New.

gcc/testsuite/

* gcc.target/aarch64/floatdihf2_1.c: New.

From-SVN: r242843

Allow -fopenmp in NVPTX mkoffload

PR target/67822
* config/nvptx/mkoffload.c (main): Allow -fopenmp.

From-SVN: r242842

sparc-common.c (sparc_option_optimization_table): Enable REE at -O2 and higher.

* common/config/sparc/sparc-common.c (sparc_option_optimization_table):
Enable REE at -O2 and higher.
* config/sparc/sparc.c (sparc_option_override): Disable it by default
in 32-bit mode.

From-SVN: r242841

[TER] PR target/48863 : Don't replace expressions across local register variable definitions

PR target/48863
PR inline-asm/70184
* tree-ssa-ter.c (temp_expr_table): Add reg_vars_cnt field.
(new_temp_expr_table): Initialise reg_vars_cnt.
(free_temp_expr_table): Release reg_vars_cnt.
(process_replaceable): Add reg_vars_cnt argument, set reg_vars_cnt
field of TAB.
(find_replaceable_in_bb): Use the above to record register variable
write occurrences and cancel replacement across them.

* gcc.target/arm/pr48863.c: New test.

From-SVN: r242840

re PR rtl-optimization/78437 (invalid sign-extend conversion in REE pass)

PR rtl-optimization/78437
* ree.c (get_uses): New function.
(combine_reaching_defs): When a copy is needed, return false if any
reaching use of the source register reads it in a mode larger than
the mode it is set in and WORD_REGISTER_OPERATIONS is true.

From-SVN: r242839

Dump probability for edges a frequency for BBs

* gimple-pretty-print.c (dump_edge_probability): New function.
(dump_gimple_switch): Dump label edge probabilities.
(dump_gimple_cond): Likewise.
(dump_gimple_label): Dump
(dump_gimple_bb_header): Dump basic block frequency.
(pp_cfg_jump): Replace e->dest argument with e.
(dump_implicit_edges): Likewise.
* tree-ssa-loop-ivopts.c (get_scaled_computation_cost_at):
Use gimple_bb (at) instead of at->bb.
* gcc.dg/builtin-unreachable-6.c: Update test to not to scan
parts for frequencies/probabilities.
* gcc.dg/pr34027-1.c: Likewise.
* gcc.dg/strict-overflow-2.c: Likewise.
* gcc.dg/tree-ssa/20040703-1.c: Likewise.
* gcc.dg/tree-ssa/builtin-sprintf-2.c: Likewise.
* gcc.dg/tree-ssa/pr32044.c: Likewise.
* gcc.dg/tree-ssa/vector-3.c: Likewise.
* gcc.dg/tree-ssa/vrp101.c: Likewise.
* gcc.dg/tree-ssa/dump-2.c: New test.

From-SVN: r242837

common.opt (flimit-function-alignment): New.

gcc/
* common.opt (flimit-function-alignment): New.
* doc/invoke.texi (-flimit-function-alignment): Document.
* emit-rtl.h (struct rtl_data): Add max_insn_address field.
* final.c (shorten_branches): Set it.
* varasm.c (assemble_start_function): Limit alignment if
requested.

gcc/testsuite/
* gcc.target/i386/align-limit.c: New test.

From-SVN: r242836

re PR tree-optimization/71595 (ICE on valid code at -O2 and -O3 on x86_64-linux-gnu: in check_loop_closed_ssa_use, at tree-ssa-loop-manip.c:704)

2016-11-24  Richard Biener  <rguenther@suse.de>

PR tree-optimization/71595
* cfgloopmanip.h (remove_path): Add irred_invalidated and
loop_closed_ssa_invalidated parameters, defaulted to NULL.
* cfgloopmanip.c (remove_path): Likewise, pass them along to
called functions.  Only fix irred flags if the caller didn't
request state.
* tree-ssa-loop-ivcanon.c (unloop_loops): Use add_bb_to_loop.
(unloop_loops): Pass irred_invalidated and loop_closed_ssa_invalidated
to remove_path.

* gcc.dg/torture/pr71595.c: New testcase.

From-SVN: r242835

re PR rtl-optimization/78120 (If conversion no longer performed)

PR rtl-optimization/78120
* ifcvt.c (noce_conversion_profitable_p): Check original cost in all
cases, and additionally test against max_seq_cost for speed
optimization.
(noce_process_if_block): Compute an estimate for the original cost when
optimizing for speed, using the minimum of then and else block costs.

testsuite/
PR rtl-optimization/78120
* gcc.target/i386/pr78120.c: New test.

From-SVN: r242834

re PR rtl-optimization/78120 (If conversion no longer performed)

PR rtl-optimization/78120
* rtlanal.c (insn_rtx_cost): Use set_rtx_cost.

From-SVN: r242833

re PR rtl-optimization/78120 (If conversion no longer performed)

PR rtl-optimization/78120
* config/i386/i386.c (ix86_rtx_costs): Fully handle SETs.

From-SVN: r242832

match.pd: Refine type conversion in result expr for below pattern: (cond (cmp (convert1?

* match.pd: Refine type conversion in result expr for below pattern:
(cond (cmp (convert1? x) c1) (convert2? x) c2) -> (minmax (x c)).

From-SVN: r242831

re PR middle-end/78429 (ICE in set_value_range, at tree-vrp.c on non-standard boolean)

PR middle-end/78429
* tree.h (wi::fits_to_boolean_p): New predicate.
(wi::fits_to_tree_p): Use it for boolean types.
* tree.c (int_fits_type_p): Likewise.

From-SVN: r242829

cp_parser_range_for: use safe_push instead of quick_push (PR

PR bootstrap/78493
* parser.c (cp_parser_range_for): Use safe_push instead of quick_push.
PR bootstrap/78493
* g++.dg/cpp1z/decomp18.C: New test.

From-SVN: r242828

ldp_stp_1.c: Add -mcpu=generic.

2016-11-23  Naveen H.S  <Naveen.Hurugalawadi@caviumnetworks.com>

* gcc.target/aarch64/ldp_stp_1.c : Add -mcpu=generic.
* gcc.target/aarch64/store-pair-1.c : Likewise.

From-SVN: r242827

re PR middle-end/71762 (~X & Y to X < Y doesn't work for uninitialized values)

2016-11-23  Richard Biener  <rguenther@suse.de>

PR middle-end/71762
* match.pd ((~X & Y) -> X < Y, (X & ~Y) -> Y < X,
(~X | Y) -> X <= Y, (X | ~Y) -> Y <= X): Remove.

* gcc.dg/torture/pr71762-1.c: New testcase.
* gcc.dg/torture/pr71762-2.c: Likewise.
* gcc.dg/torture/pr71762-3.c: Likewise.
* gcc.dg/tree-ssa/forwprop-28.c: XFAIL.

From-SVN: r242822

fmaxmin.c: Add -fno-vect-cost-model.

2016-11-23  Naveen H.S  <Naveen.Hurugalawadi@caviumnetworks.com>

* gcc.target/aarch64/fmaxmin.c : Add -fno-vect-cost-model.
* gcc.target/aarch64/fmul_fcvt_2.c : Likewise.
* gcc.target/aarch64/vect-abs-compile.c : Likewise.
* gcc.target/aarch64/vect-clz.c : Likewise.
* gcc.target/aarch64/vect-fcm-eq-d.c : Likewise.
* gcc.target/aarch64/vect-fcm-ge-d.c : Likewise.
* gcc.target/aarch64/vect-fcm-gt-d.c : Likewise.
* gcc.target/aarch64/vect-fmovd-zero.c : Likewise.
* gcc.target/aarch64/vect-fmovd.c : Likewise.
* gcc.target/aarch64/vect-fmovf-zero.c : Likewise.
* gcc.target/aarch64/vect-fmovf.c : Likewise.
* gcc.target/aarch64/vect_ctz_1.c : Likewise.

From-SVN: r242821