Fix merge failure: Remove files deleted on trunk already.  Fix trivial thinko.

From-SVN: r218414

Remove accidentally committed files which were removed on trunk already.

From-SVN: r218204

* Add -mlra option.

From-SVN: r218194

* Split QI/HImode displacement addressing load/store via R0.

* Use std::swap in the sh-lra change.

From-SVN: r218193

* Add legitimize_address_displacement target macto.

From-SVN: r218192

Merge from trunk revision 218173.

From-SVN: r218191

Fix merge failure: add new directories and files on trunk.

From-SVN: r217981

merge from trunk revision 217978.

From-SVN: r217980

Make *addsi3_compact insn_and_split and add an alternative which will split to...

Make *addsi3_compact insn_and_split and add an alternative which
will split to (set a c) and (set (plus a b)) after reload.
* config/sh/predicates.md (arith_or_int_operand): New.
* config/sh/sh.md (addsi3): Use arith_or_int_operand for 2nd operand.
  Check register overlap for non I08 constant 2nd operand.
  (*addsi3_compact): Make it insn_and_split and add an alternative.

From-SVN: r216535

Define SH specific HARD_REGNO_CALLER_SAVE_MODE target macro.

* config/sh/sh.h (HARD_REGNO_CALLER_SAVE_MODE): Define.
* config/sh/sh-protos.h (sh_hard_regno_caller_save_mode): Declare.
* config/sh/sh.c (sh_hard_regno_caller_save_mode): New function.

From-SVN: r216534

Introduce target hook which disables the substitution of pseudos with equivalent memory values.

Introduce target hook which disables the substitution of pseudos
with equivalent memory values.
* target.def (cannot_substitute_mem_equiv_p): New hook.
* lra-constraints.c (get_equiv): Check targetm.cannot_substitute_mem_equiv_p.
* config/sh/sh.c (TARGET_CANNOT_SUBSTITUTE_MEM_EQUIV_P): Define.
  (sh_cannot_substitute_mem_equiv_p): New function.
* doc/tm.texi.in (TARGET_CANNOT_SUBSTITUTE_MEM_EQUIV_P): New @hook.
* doc/tm.texi: Regenerate.

From-SVN: r216533

lra-constraints.c (process_address_1): Fixup the result of...

lra-constraints.c (process_address_1): Fixup the result of
decompose_mem_address when INDEX_REG_CLASS is a single register class.

From-SVN: r216532

merge from trunk revision 216447.

From-SVN: r216529

* merge from trunk.

From-SVN: r216025

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* lra.c (lra_update_insn_regno_info): Handle mem args in calls.

From-SVN: r215744

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* config/sh/sh.md (untyped_call): Clobber the function value registers
  before the call.

From-SVN: r215677

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* config/sh/sh-protos.h (sh_movsf_ie_ra_split_p): New prototype.
* config/sh/sh.c (sh_movsf_ie_ra_split_p): New function.
* config/sh/sh.md (movsi_ie): Use constraint "mr" instead of "m".
  (movsf_ie_ra): New insn_and_split.
  (movsf): Use movsf_ie_ra when lra_in_progress is set.

From-SVN: r215676

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* config/sh/predicates.md (general_movsrc_operand): Take
  subregs into account for plus address expression.
  (general_movdst_operand): Likewise.
  (t_reg_operand): Allow (zero_extend (reg t)).

From-SVN: r215341

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* lra-constraints.c: Print insn when hitting MAX_RELOAD_INSNS_NUMBER.

From-SVN: r215247

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* Apply patch from https://gcc.gnu.org/ml/gcc/2014-04/msg00273.html

From-SVN: r215246

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* Remove LEGITIMIZE_RELOAD_ADDRESS and sh_legitimize_reload_address.
  They are not used by LRA.

From-SVN: r215244

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* predicates.md (general_movsrc_operand,  general_movdst_operand):
  Allow only valid plus address expressions.

From-SVN: r215243

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* predicates.md (general_movsrc_operand,  general_movdst_operand):
  Don't legitimate QI/HImode  mem displacement while LRA is running.

From-SVN: r215241

re PR target/55212 ([SH] Switch to LRA)

PR target/55212
* lra.c: fix comments.  print insn before lra_assert.

From-SVN: r215240

* merge from trunk.

From-SVN: r215239

* MAINTAINERS: Move myself to reviewers (Fortran).

From-SVN: r215237

re PR fortran/63252 (tree_class_check_failed)

2014-09-13  Tobias Burnus  <burnus@net-b.de>

        PR fortran/63252
        * trans-decl.c (gfc_build_builtin_function_decls): Fix
        caf_unlock declaration.

From-SVN: r215236

re PR c++/60862 (bad location in invalid conversion error)

PR c++/60862
* parser.c (cp_parser_postfix_expression) <case CPP_OPEN_PAREN>: Set
location of a call expression.

* g++.dg/diagnostic/pr60862.C: New test.

From-SVN: r215235

tree.c (protected_set_expr_location): Don't check whether T is non-null here.

* tree.c (protected_set_expr_location): Don't check whether T is
non-null here.

From-SVN: r215234

Daily bump.

From-SVN: r215233

msp430.md (extendhipsi2): Use 20-bit form of RLAM/RRAM.

* config/msp430/msp430.md (extendhipsi2): Use 20-bit form of RLAM/RRAM.
(extend_and_shift1_hipsi2): Likewise.
(extend_and_shift2_hipsi2): Likewise.

From-SVN: r215229

re PR c++/63201 (Full specialization of a member variable template of a class template does not work)

PR c++/63201
* decl.c (start_decl): Handle specialization of member variable
template.
* pt.c (check_explicit_specialization): Adjust error.

From-SVN: r215226

deque (__gnu_debug::deque): Make base class C++11 allocator aware.

* include/debug/deque (__gnu_debug::deque): Make base class C++11
allocator aware.

From-SVN: r215223

params 2 and 3 of reg_set_between_p

gcc/ChangeLog:
* config/alpha/alpha.c (alpha_ra_ever_killed): Replace NULL_RTX
with NULL when dealing with an insn.
* config/sh/sh.c (sh_reorg): Strengthen local "last_float_move"
from rtx to rtx_insn *.
* rtl.h (reg_set_between_p): Strengthen params 2 and 3 from
const_rtx to const rtx_insn *.
* rtlanal.c (reg_set_between_p): Likewise, removing a checked cast.

From-SVN: r215222

re PR libstdc++/59603 (std::random_shuffle tries to swap element with itself)

PR libstdc++/59603
* include/bits/stl_algo.h (random_shuffle): Prevent self-swapping.
* testsuite/25_algorithms/random_shuffle/59603.cc: New.

From-SVN: r215219

fix ChangeLog typo

From-SVN: r215218

fix assert in hash_table pch routines

gcc/ChangeLog:

2014-09-12  Trevor Saunders  <tsaunders@mozilla.com>

* hash-table.h (gt_pch_nx): don't call gt_pch_note_object within an
assert.

From-SVN: r215216

Remove LIBGCC2_HAS_?F_MODE target macros.

This patch removes the LIBGCC2_HAS_{SF,DF,XF,TF}_MODE target macros,
replacing them by predefines with -fbuilding-libgcc, together with a
target hook that can influence those predefines when needed.

The new default is that a floating-point mode is supported in libgcc
if (a) it passes the scalar_mode_supported_p hook (otherwise it's not
plausible for it to be supported in libgcc) and (b) it's one of those
four modes (since those are the modes for which libgcc hardcodes the
possibility of support).  The target hook can override the default
choice (in either direction) for modes that pass
scalar_mode_supported_p (although overriding in the direction of
returning true when the default would return false only makes sense if
all relevant functions are specially defined in libgcc for that
particular target).

The previous default settings depended on various settings such as
LIBGCC2_LONG_DOUBLE_TYPE_SIZE, as well as targets defining the above
target macros if the default wasn't correct.

The default scalar_mode_supported_p only declares a floating-point
mode to be supported if it matches one of float / double / long
double.  This means that in most cases where a mode is only supported
conditionally in libgcc (TFmode only supported if it's the mode of
long double, most commonly), the default gets things right.  Overrides
were needed in the following cases:

* SFmode would always have been supported in libgcc (the condition was
  BITS_PER_UNIT == 8, true for all current targets), but pdp11
  defaults to 64-bit float, and in that case SFmode would fail
  scalar_mode_supported_p.  I don't know if libgcc actually built for
  pdp11 (and the port may well no longer be being used), but this
  patch adds a scalar_mode_supported_p hook to it to ensure SFmode is
  treated as supported.

* Certain i386 and ia64 targets need the new hook to match the
  existing cases for when XFmode or TFmode support is present in
  libgcc.  For i386, the hook can always declare XFmode to be
  supported - the cases where it's not are the cases where long double
  is TFmode, in which case XFmode fails scalar_mode_supported_p[*] -
  but TFmode support needs to be conditional.  (And of the targets not
  defining LIBGCC2_HAS_TF_MODE before this patch, some defined
  LONG_DOUBLE_TYPE_SIZE to 64, so ensuring LIBGCC2_HAS_TF_MODE would
  always be false, while others did not define it, so allowing it to
  be true in the -mlong-double-128 case.  This patch matches that
  logic, although I suspect all the latter targets would have been
  broken if you tried to enable -mlong-double-128 by default, for lack
  of the soft-fp TFmode support in libgcc, which is separately
  configured.)

  [*] I don't know if it's deliberate not to support __float80 at all
  with -mlong-double-128.

In order to implement the default version of the new hook,
insn-modes.h was made to contain macros such as HAVE_TFmode for each
machine mode, so the default hook can contain conditionals on whether
XFmode and TFmode exist (to match the hardcoding of a list of modes in
libgcc).  This is also used in fortran/trans-types.c; previously it
had a conditional on defined(LIBGCC2_HAS_TF_MODE) (a bit dubious,
since it ignored the value of the macro), which is replaced by testing
defined(HAVE_TFmode), in conjunction with requiring
targetm.libgcc_floating_mode_supported_p.

(Fortran is testing something stronger than that hook: not only is
libgcc support required, but also libm or equivalent.  Thus, it has a
test for ENABLE_LIBQUADMATH_SUPPORT in the case that the mode is
TFmode and that's not the same as any of the three standard types.
The old and new tests are intended to accept exactly the same set of
modes for all targets.)

Apart from the four target macros eliminated by this patch, it gets us
closer to eliminating LIBGCC2_LONG_DOUBLE_TYPE_SIZE as well, though a
few more places using that macro need changing first.

Bootstrapped with no regressions on x86_64-unknown-linux-gnu; also
built cc1 for crosses to ia64-elf and pdp11-none as a minimal test of
changes for those targets.

gcc:
* target.def (libgcc_floating_mode_supported_p): New hook.
* targhooks.c (default_libgcc_floating_mode_supported_p): New
function.
* targhooks.h (default_libgcc_floating_mode_supported_p): Declare.
* doc/tm.texi.in (LIBGCC2_HAS_DF_MODE, LIBGCC2_HAS_XF_MODE)
(LIBGCC2_HAS_TF_MODE): Remove.
(TARGET_LIBGCC_FLOATING_MODE_SUPPORTED_P): New @hook.
* doc/tm.texi: Regenerate.
* genmodes.c (emit_insn_modes_h): Define HAVE_%smode for each
machine mode.
* system.h (LIBGCC2_HAS_SF_MODE, LIBGCC2_HAS_DF_MODE)
(LIBGCC2_HAS_XF_MODE, LIBGCC2_HAS_TF_MODE): Poison.
* config/i386/cygming.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/i386/darwin.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/i386/djgpp.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/i386/dragonfly.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/i386/freebsd.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/i386/gnu-user-common.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/i386/i386-interix.h (IX86_NO_LIBGCC_TFMODE): Define.
* config/i386/i386.c (ix86_libgcc_floating_mode_supported_p): New
function.
(TARGET_LIBGCC_FLOATING_MODE_SUPPORTED_P): Define.
* config/i386/i386elf.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/i386/lynx.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/i386/netbsd-elf.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/i386/netbsd64.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/i386/nto.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/i386/openbsd.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/i386/openbsdelf.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/i386/rtemself.h (IX86_NO_LIBGCC_TFMODE): Define.
* config/i386/sol2.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/i386/vx-common.h (IX86_MAYBE_NO_LIBGCC_TFMODE): Define.
* config/ia64/elf.h (IA64_NO_LIBGCC_TFMODE): Define.
* config/ia64/freebsd.h (IA64_NO_LIBGCC_TFMODE): Define.
* config/ia64/hpux.h (LIBGCC2_HAS_XF_MODE, LIBGCC2_HAS_TF_MODE):
Remove.
* config/ia64/ia64.c (TARGET_LIBGCC_FLOATING_MODE_SUPPORTED_P):
New macro.
(ia64_libgcc_floating_mode_supported_p): New function.
* config/ia64/linux.h (LIBGCC2_HAS_TF_MODE): Remove.
* config/ia64/vms.h (IA64_NO_LIBGCC_XFMODE)
(IA64_NO_LIBGCC_TFMODE): Define.
* config/msp430/msp430.h (LIBGCC2_HAS_DF_MODE): Remove.
* config/pdp11/pdp11.c (TARGET_SCALAR_MODE_SUPPORTED_P): New
macro.
(pdp11_scalar_mode_supported_p): New function.
* config/rl78/rl78.h (LIBGCC2_HAS_DF_MODE): Remove.
* config/rx/rx.h (LIBGCC2_HAS_DF_MODE): Remove.

gcc/c-family:
* c-cppbuiltin.c (c_cpp_builtins): Define __LIBGCC_HAS_%s_MODE__
macros for floating-point modes.

gcc/fortran:
* trans-types.c (gfc_init_kinds): Check
targetm.libgcc_floating_mode_supported_p for floating-point
modes.  Check HAVE_TFmode instead of LIBGCC2_HAS_TF_MODE.

libgcc:
* libgcc2.h (LIBGCC2_HAS_SF_MODE): Define using
__LIBGCC_HAS_SF_MODE__.
(LIBGCC2_HAS_DF_MODE): Define using __LIBGCC_HAS_DF_MODE__.
(LIBGCC2_HAS_XF_MODE): Define using __LIBGCC_HAS_XF_MODE__.
(LIBGCC2_HAS_TF_MODE): Define using __LIBGCC_HAS_TF_MODE__.
* config/libbid/bid_gcc_intrinsics.h
(LIBGCC2_LONG_DOUBLE_TYPE_SIZE): Do not define.
(LIBGCC2_HAS_XF_MODE): Define using __LIBGCC_HAS_XF_MODE__.
(LIBGCC2_HAS_TF_MODE): Define using __LIBGCC_HAS_TF_MODE__.
* fixed-bit.h (LIBGCC2_LONG_DOUBLE_TYPE_SIZE): Do not define.
(LIBGCC2_HAS_SF_MODE): Define using __LIBGCC_HAS_SF_MODE__.
(LIBGCC2_HAS_DF_MODE): Define using __LIBGCC_HAS_DF_MODE__.

From-SVN: r215215

re PR middle-end/63237 (error: invalid operand in unary operation)

2014-09-12  Richard Biener  <rguenther@suse.de>

PR middle-end/63237
* gimple-fold.c (get_maxval_strlen): Gimplify string length.

* g++.dg/torture/pr63237.C: New testcase.

From-SVN: r215212

tree.c (integer_each_onep): New function.

2014-09-12  Marc Glisse  <marc.glisse@inria.fr>

gcc/
* tree.c (integer_each_onep): New function.
* tree.h (integer_each_onep): Declare it.
* fold-const.c (fold_binary_loc): Use it for ~A + 1 to -A and
-A - 1 to ~A.  Disable (X & 1) ^ 1, (X ^ 1) & 1 and ~X & 1 to
(X & 1) == 0 for vector and complex.
gcc/testsuite/
* gcc.dg/vec-andxor1.c: New file.

From-SVN: r215209

[AArch64] Add regmove_costs for Cortex-A57 and A53

2014-09-12  Wilco Dijkstra  <wdijkstr@arm.com>

  * gcc/config/aarch64/aarch64.c: (cortexa57_regmove_cost): New cost table for
  A57.
  (cortexa53_regmove_cost): New cost table for A53.  Increase GP2FP/FP2GP cost
  to spilling from integer to FP registers.

From-SVN: r215208

[AArch64] Fix cost for Q register moves

2014-09-12  Wilco Dijkstra  <wdijkstr@arm.com>

  * gcc/config/aarch64/aarch64.c (aarch64_register_move_cost): Fix Q register
  move handling.
  (generic_regmove_cost): Undo raised FP2FP move cost as Q register moves are
  now handled correctly.

From-SVN: r215207

[AArch64] Add cost handling of CALLER_SAVE_REGS and POINTER_REGS

2014-09-12  Wilco Dijkstra  <wdijkstr@arm.com>

  * gcc/config/aarch64/aarch64.c (aarch64_register_move_cost): Add cost handling
  of CALLER_SAVE_REGS and POINTER_REGS.

From-SVN: r215206

[Ree] Ensure inserted copy don't change the number of hard registers

2014-09-12  Wilco Dijkstra  <wilco.dijkstra@arm.com>

  gcc/
    * ree.c (combine_reaching_defs): Ensure inserted copy don't change the
    number of hard registers.

From-SVN: r215205

AVX-512. Extend vpternlog, valign, vrotate insns.

gcc/
* config/i386/sse.md
(define_mode_iterator VI48_AVX512VL): New.
(define_expand "<avx512>_vternlog<mode>_maskz"): Rename from
"avx512f_vternlog<mode>_maskz" and update mode iterator.
(define_insn "<avx512>_vternlog<mode><sd_maskz_name>"): Rename
from "avx512f_vternlog<mode><sd_maskz_name>" and update mode iterator.
(define_insn "<avx512>_vternlog<mode>_mask"): Rename from
"avx512f_vternlog<mode>_mask" and update mode iterator.
(define_insn "<mask_codefor><avx512>_align<mode><mask_name>"): Rename
from "<mask_codefor>avx512f_align<mode><mask_name>" and update mode
iterator.
(define_insn "<avx512>_<rotate>v<mode><mask_name>"): Rename from
"avx512f_<rotate>v<mode><mask_name>" and update mode iterator.
(define_insn "<avx512>_<rotate><mode><mask_name>"): Rename from
"avx512f_<rotate><mode><mask_name>" and update mode iterator.
(define_insn "clz<mode>2<mask_name>"): Use VI48_AVX512VL mode iterator.
(define_insn "<mask_codefor>conflict<mode><mask_name>"): Ditto.

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215203

AVX-512.  Extend max/min insn patterns.

gcc/
* config/i386/sse.md (VI128_256): Delete.
(define_mode_iterator VI124_256): New.
(define_mode_iterator VI124_256_AVX512F_AVX512BW): Ditto.
(define_expand "<code><mode>3<mask_name><round_name>"): Delete.
(define_expand "<code><VI124_256_AVX512F_AVX512BW:mode>3"): New.
(define_insn "*avx2_<code><VI124_256:mode>3"): Rename from
"*avx2_<code><mode>3<mask_name><round_name>" and update mode iterator.
(define_expand "<code><VI48_AVX512VL:mode>3_mask"): New.
(define_insn "*avx512bw_<code><VI48_AVX512VL:mode>3<mask_name>"): Ditto.
(define_insn "<mask_codefor><code><mode>3<mask_name>"): Update mode
iterator.
(define_expand "<code><VI8_AVX2:mode>3"): Update pettern generation
in presence of AVX-512.

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215202

AVX-512. Extend gather insn patterns.

gcc/
* config/i386/sse.md
(define_expand "<avx512>_gathersi<mode>"): Rename from
"avx512f_gathersi<mode>".
(define_insn "*avx512f_gathersi<mode>"): Use VI48F.
(define_insn "*avx512f_gathersi<mode>_2"): Ditto.
(define_expand "<avx512>_gatherdi<mode>"): Rename from
"avx512f_gatherdi<mode>".
(define_insn "*avx512f_gatherdi<mode>"): Use VI48F.
(define_insn "*avx512f_gatherdi<mode>_2"): Use VI48F, add 128/256-bit
wide versions.
(define_expand "<avx512>_scattersi<mode>"): Rename from
"avx512f_scattersi<mode>".
(define_insn "*avx512f_scattersi<mode>"): Use VI48F.
(define_expand "<avx512>_scatterdi<mode>"): Rename from
"avx512f_scatterdi<mode>".
(define_insn "*avx512f_scatterdi<mode>"): Use VI48F.

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215201

ira.h (ira_finish_once): Delete.

gcc/
* ira.h (ira_finish_once): Delete.
* ira-int.h (target_ira_int::~target_ira_int): Declare.
(target_ira_int::free_ira_costs): Likewise.
(target_ira_int::free_register_move_costs): Likewise.
(ira_finish_costs_once): Delete.
* ira.c (free_register_move_costs): Replace with...
(target_ira_int::free_register_move_costs): ...this new function.
(target_ira_int::~target_ira_int): Define.
(ira_init): Call free_register_move_costs as a member function rather
than a global function.
(ira_finish_once): Delete.
* ira-costs.c (free_ira_costs): Replace with...
(target_ira_int::free_ira_costs): ...this new function.
(ira_init_costs): Call free_ira_costs as a member function rather
than a global function.
(ira_finish_costs_once): Delete.
* target-globals.c (target_globals::~target_globals): Call the
target_ira_int destructor.
* toplev.c: Include lra.h.
(finalize): Call lra_finish_once rather than ira_finish_once.

From-SVN: r215200

Daily bump.

From-SVN: r215199

common.opt (flto-odr-type-merging): New flag.

* common.opt (flto-odr-type-merging): New flag.
* ipa-deivrt.c (hash_type_name): Use ODR names for hasing if availale.
(types_same_for_odr): Likewise.
(odr_subtypes_equivalent_p): Likewise.
(add_type_duplicate): Do not walk type variants.
(register_odr_type): New function.
* ipa-utils.h (register_odr_type): Declare.
(odr_type_p): New function.
* langhooks.c (lhd_set_decl_assembler_name): Do not compute
TYPE_DECLs
* doc/invoke.texi (-flto-odr-type-merging): Document.
* tree.c (need_assembler_name_p): Compute ODR names when asked
for it.
* tree.h (DECL_ASSEMBLER_NAME): Update comment.

* lto.c (lto_read_decls): Register ODR types.

From-SVN: r215196

Also turn off OPTION_MASK_ABI_X32 for -m16

PR target/63228
* config/i386/i386.c (ix86_option_override_internal): Also turn
off OPTION_MASK_ABI_X32 for -m16.

From-SVN: r215194

rs6000.md (rs6000_mftb_<mode>): Use mode iterator GPR instead of P.

2014-09-11  Segher Boessenkool  <segher@kernel.crashing.org>

* config/rs6000/rs6000.md (rs6000_mftb_<mode>): Use mode iterator
GPR instead of P.

From-SVN: r215193

re PR target/58757 (Advertise the lack of denormal support on alpha without -mieee)

2014-09-11  Marc Glisse  <marc.glisse@inria.fr>

PR target/58757
gcc/c-family/
* c-cppbuiltin.c (builtin_define_float_constants): Correct
__*_DENORM_MIN__ without denormals.
gcc/
* ginclude/float.h (FLT_TRUE_MIN, DBL_TRUE_MIN, LDBL_TRUE_MIN):
Directly forward to __*_DENORM_MIN__.
gcc/testsuite/
* gcc.dg/c11-true_min-1.c: New testcase.

From-SVN: r215191

Introduce LABEL_REF_LABEL

gcc/ChangeLog:
2014-09-11  David Malcolm  <dmalcolm@redhat.com>

* rtl.h (LABEL_REF_LABEL): New macro.

* alias.c (rtx_equal_for_memref_p): Use LABEL_REF_LABEL in place
of XEXP (, 0), where we know that we have a LABEL_REF.
* cfgbuild.c (make_edges): Likewise.
(purge_dead_tablejump_edges): Likewise.
* cfgexpand.c (convert_debug_memory_address): Likewise.
* cfgrtl.c (patch_jump_insn): Likewise.
* combine.c (distribute_notes): Likewise.
* cse.c (hash_rtx_cb): Likewise.
(exp_equiv_p): Likewise.
(fold_rtx): Likewise.
(check_for_label_ref): Likewise.
* cselib.c (rtx_equal_for_cselib_1): Likewise.
(cselib_hash_rtx): Likewise.
* emit-rtl.c (mark_label_nuses): Likewise.
* explow.c (convert_memory_address_addr_space): Likewise.
* final.c (output_asm_label): Likewise.
(output_addr_const): Likewise.
* gcse.c (add_label_notes): Likewise.
* genconfig.c (walk_insn_part): Likewise.
* genrecog.c (validate_pattern): Likewise.
* ifcvt.c (cond_exec_get_condition): Likewise.
(noce_emit_store_flag): Likewise.
(noce_get_alt_condition): Likewise.
(noce_get_condition): Likewise.
* jump.c (maybe_propagate_label_ref): Likewise.
(mark_jump_label_1): Likewise.
(redirect_exp_1): Likewise.
(rtx_renumbered_equal_p): Likewise.
* lra-constraints.c (operands_match_p): Likewise.
* reload.c (operands_match_p): Likewise.
(find_reloads): Likewise.
* reload1.c (set_label_offsets): Likewise.
* reorg.c (get_branch_condition): Likewise.
* rtl.c (rtx_equal_p_cb): Likewise.
(rtx_equal_p): Likewise.
* rtlanal.c (reg_mentioned_p): Likewise.
(rtx_referenced_p): Likewise.
(get_condition): Likewise.
* sched-vis.c (print_value): Likewise.
* varasm.c (const_hash_1): Likewise.
(compare_constant): Likewise.
(const_rtx_hash_1): Likewise.
(output_constant_pool_1): Likewise.

From-SVN: r215190

htm.md (tabort, [...]): Use xor instead of minus.

2014-09-11  Segher Boessenkool  <segher@kernel.crashing.org>

* config/rs6000/htm.md (tabort, tabortdc, tabortdci, tabortwc,
tabortwci, tbegin, tcheck, tend, trechkpt, treclaim, tsr): Use xor
instead of minus.
* config/rs6000/vector.md (cr6_test_for_zero_reverse,
cr6_test_for_lt_reverse): Ditto.

From-SVN: r215187

re PR c++/61489 (Wrong warning with -Wmissing-field-initializers.)

2014-09-11  Paolo Carlini  <paolo.carlini@oracle.com>

PR c++/61489
* doc/invoke.texi ([-Wmissing-field-initializers]): Update.

/cp
2014-09-11  Paolo Carlini  <paolo.carlini@oracle.com>

PR c++/61489
* typeck2.c (process_init_constructor_record): Do not warn about
missing field initializer if EMPTY_CONSTRUCTOR_P (init).

/testsuite
2014-09-11  Paolo Carlini  <paolo.carlini@oracle.com>

PR c++/61489
* g++.dg/warn/Wmissing-field-initializers-1.C: New.
* g++.old-deja/g++.other/warn5.C: Adjust.

From-SVN: r215186

[AArch64] Simplify vreinterpret for float64x1_t using casts.

* config/aarch64/aarch64-builtins.c (aarch64_types_unop_su_qualifiers,
TYPES_REINTERP_SU, aarch64_types_unop_sp_qualifiers, TYPE_REINTERP_SP,
aarch64_types_unop_us_qualifiers, TYPES_REINTERP_US,
aarch64_types_unop_ps_qualifiers, TYPES_REINTERP_PS, BUILTIN_VD):
Delete.

(aarch64_fold_builtin): Remove all reinterpret cases.

* config/aarch64/aarch64-protos.h (aarch64_simd_reinterpret): Delete.

* config/aarch64/aarch64-simd-builtins.def (reinterpret*) : Delete.

* config/aarch64/aarch64-simd.md (aarch64_reinterpretv8qi<mode>,
aarch64_reinterpretv4hi<mode>, aarch64_reinterpretv2si<mode>,
aarch64_reinterpretv2sf<mode>, aarch64_reinterpretdi<mode>,
aarch64_reinterpretv1df<mode>, aarch64_reinterpretv16qi<mode>,
aarch64_reinterpretv8hi<mode>, aarch64_reinterpretv4si<mode>,
aarch64_reinterpretv4sf<mode>, aarch64_reinterpretv2di<mode>,
aarch64_reinterpretv2df<mode>): Delete.

* config/aarch64/aarch64.c (aarch64_simd_reinterpret): Delete.

* config/aarch64/arm_neon.h (vreinterpret_p8_f64,
vreinterpret_p16_f64, vreinterpret_f32_f64, vreinterpret_f64_f32,
vreinterpret_f64_p8, vreinterpret_f64_p16, vreinterpret_f64_s8,
vreinterpret_f64_s16, vreinterpret_f64_s32, vreinterpret_f64_u8,
vreinterpret_f64_u16, vreinterpret_f64_u32, vreinterpret_s64_f64,
vreinterpret_u64_f64, vreinterpret_s8_f64, vreinterpret_s16_f64,
vreinterpret_s32_f64, vreinterpret_u8_f64, vreinterpret_u16_f64,
vreinterpret_u32_f64): Use cast.

* config/aarch64/iterators.md (VD_RE): Delete.

From-SVN: r215180

[AArch64] Replace temporary inline assembler for vset_lane

* config/aarch64/arm_neon.h (aarch64_vset_lane_any): New (*2).
(vset_lane_f32, vset_lane_f64, vset_lane_p8, vset_lane_p16,
vset_lane_s8, vset_lane_s16, vset_lane_s32, vset_lane_s64,
vset_lane_u8, vset_lane_u16, vset_lane_u32, vset_lane_u64,
vsetq_lane_f32, vsetq_lane_f64, vsetq_lane_p8, vsetq_lane_p16,
vsetq_lane_s8, vsetq_lane_s16, vsetq_lane_s32, vsetq_lane_s64,
vsetq_lane_u8, vsetq_lane_u16, vsetq_lane_u32, vsetq_lane_u64):
Replace inline assembler with __aarch64_vset_lane_any.

From-SVN: r215179

[AArch64 Testsuite] Add execution test of vset(q?)_lane intrinsics.

* gcc.target/aarch64/vset_lane_1.c: New test.

From-SVN: r215177

[AArch64] Cheap fix for argument types of vmull_high_lane_{us}{16,32}

gcc/

* config/aarch64/arm_neon.h (vmull_high_lane_s16): Fix argument
types.
(vmull_high_lane_s32): Likewise.
(vmull_high_lane_u16): Likewise.
(vmull_high_lane_u32): Likewise.

From-SVN: r215176

re PR c++/63139 (Class-scope typedef overwrites typedef of previously defined class)

PR c++/63139
* pt.c (tsubst_pack_expansion): Simplify substitution into T....
(tsubst): Don't throw away PACK_EXPANSION_EXTRA_ARGS.

From-SVN: r215171

re PR c++/58678 (pykde4-4.11.2 link error (devirtualization too trigger happy))

PR c++/58678
* ipa-devirt.c (ipa_devirt): Don't check DECL_COMDAT.

From-SVN: r215168

re PR libstdc++/63219 (Superfluous template parameter in match_result::format overload)

PR libstdc++/63219
* include/bits/regex.h (match_results::format): Remove stray template
parameter.
* include/bits/regex_compiler.h (_RegexTranslator::_RegexTranslator):
Remove parameter name to avoid -Wunused-parameter warning.
* include/bits/regex_executor.h (_State_info::_State_info): Reorder
mem-initializers to avoid -Wreorder warning.
* include/bits/regex_executor.tcc (_Executor::_M_word_boundary):
Remove parameter name to avoid -Wunused-parameter warning.
* include/bits/regex_scanner.tcc (_Scanner::_M_advance): Add braces
to avoid -Wempty-body warning when not in debug mode.

From-SVN: r215160

Fix declarations in some tests.

* gcc.dg/compat/struct-by-value-13_main.c (struct_by_value_13_x):
Fix declaration.
* gcc.dg/compat/struct-by-value-16a_main.c (struct_by_value_16a_x):
Fix declaration.
* gcc.dg/compat/struct-by-value-17a_main.c (struct_by_value_17a_x):
Fix declaration.
* gcc.dg/compat/struct-by-value-18a_main.c (struct_by_value_18a_x):
Fix declaration.

From-SVN: r215158

linux64.S: Emit .note.GNU-stack even when POWERPC64 is not defined.

* src/powerpc/linux64.S: Emit .note.GNU-stack even when
POWERPC64 is not defined.
* src/powerpc/linux64_closure.S: Likewise.  Also test _CALL_ELF == 2.

From-SVN: r215155

re PR target/63223 ([avr] Make jumptables work with -Wl,--section-start,.text=)

gcc/
PR target/63223
* config/avr/avr.md (*tablejump.3byte-pc): New insn.
(*tablejump): Restrict to !AVR_HAVE_EIJMP_EICALL.  Add void clobber.
(casesi): Expand to *tablejump.3byte-pc if AVR_HAVE_EIJMP_EICALL.
libgcc/
PR target/63223
* config/avr/libgcc.S (__tablejump2__): Rewrite to use RAMPZ, ELPM
and R24 as needed.  Make work for all devices and .text locations.
(__do_global_ctors, __do_global_dtors): Use word addresses.
(__tablejump__, __tablejump_elpm__): Remove functions.
* t-avr (LIB1ASMFUNCS): Remove _tablejump, _tablejump_elpm.
Add _tablejump2.
(XICALL, XIJMP): New macros.

From-SVN: r215152

AVX-512. Add vperm[it]2 insns support.

gcc/
* config/i386/sse.md
(define_expand "<avx512>_vpermi2var<VI48F:mode>3_maskz"): Rename from
"avx512f_vpermi2var<mode>3_maskz" and update mode iterator.
(define_expand "<avx512>_vpermi2var<VI2_AVX512VL:mode>3_maskz"):
New.
(define_insn "<avx512>_vpermi2var<VI48F:mode>3<sd_maskz_name>"): Rename
from "avx512f_vpermi2var<mode>3<sd_maskz_name>" and update mode
iterator.
(define_insn "<avx512>_vpermi2var<VI2_AVX512VL:mode>3<sd_maskz_name>"):
New.
(define_insn "<avx512>_vpermi2var<VI48F:mode>3_mask"): Rename from
"avx512f_vpermi2var<mode>3_mask" and update mode iterator.
(define_insn "<avx512>_vpermi2var<VI2_AVX512VL:mode>3_mask"): New.
(define_expand "<avx512>_vpermt2var<VI48F:mode>3_maskz"): Rename from
"avx512f_vpermt2var<mode>3_maskz" and update mode iterator.
(define_expand "<avx512>_vpermt2var<VI2_AVX512VL:mode>3_maskz"): New.
(define_insn "<avx512>_vpermt2var<VI48F:mode>3<sd_maskz_name>"): Rename
from "avx512f_vpermt2var<mode>3<sd_maskz_name>" and update mode
iterator.
(define_insn "<avx512>_vpermt2var<VI2_AVX512VL:mode>3<sd_maskz_name>"):
New.
(define_insn "<avx512>_vpermt2var<VI48F:mode>3_mask"): Rename from
"avx512f_vpermt2var<mode>3_mask" and update mode iterator.
(define_insn "<avx512>_vpermt2var<VI2_AVX512VL:mode>3_mask"): New.

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215151

varpool.c (varpool_node::ctor_useable_for_folding_p): Do not try to access removed nodes.

* varpool.c (varpool_node::ctor_useable_for_folding_p): Do not try
to access removed nodes.

From-SVN: r215150

re PR middle-end/63186 (Undefined .L* symbols because of fnsplit)

PR tree-optimization/63186
* ipa-split.c (test_nonssa_use): Skip nonforced labels.
(mark_nonssa_use): Likewise.
(verify_non_ssa_vars): Verify all header blocks for label
definitions.

* gcc.dg/pr63186.c: New testcase.

From-SVN: r215149

AVX-512. Extend vpermvar insn patterns.

gcc/
* config/i386/sse.md
(define_mode_attr avx2_avx512): Rename from avx2_avx512bw.
(define_mode_iterator VI48F_256_512): Extend to AVX-512VL.
(define_insn "<avx2_avx512>_permvar<mode><mask_name>"): Rename from
"<avx2_avx512f>_permvar<mode><mask_name>".
(define_insn "<avx512>_permvar<mode><mask_name>"): New.
(define_insn "<avx2_avx512>_ashrv<VI48_AVX512F_AVX512VL:mode><mask_name>"):
Rename from "<avx2_avx512f>_ashrv<mode><mask_name>".
(define_insn "<avx2_avx512>_ashrv<VI2_AVX512VL:mode><mask_name>"):
Ditto.
(define_insn "<avx2_avx512>_<shift_insn>v<VI48_AVX512F:mode><mask_name>"):
Rename from "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>".
(define_insn "<avx2_avx512>_<shift_insn>v<VI2_AVX512VL:mode><mask_name>"):
Rename from "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>".

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215148

Daily bump.

From-SVN: r215147

sh.c (sh_secondary_reload): return NO_REGS instead of FPUL_REGS in one case.

* config/sh/sh.c (sh_secondary_reload): return NO_REGS
  instead of FPUL_REGS in one case.

* lra-constraints.c: print some info before bailing out
  in some cases.

From-SVN: r215140

vsx.md (vsx_fmav4sf4): Use correct constraints for V2DF, V4SF, DF, and DI modes.

2014-09-10  Michael Meissner  <meissner@linux.vnet.ibm.com>

* config/rs6000/vsx.md (vsx_fmav4sf4): Use correct constraints for
V2DF, V4SF, DF, and DI modes.
(vsx_fmav2df2): Likewise.
(vsx_float_fix_<mode>2): Likewise.
(vsx_reduc_<VEC_reduc_name>_v2df_scalar): Likewise.

From-SVN: r215138

Fix PR target/63209.

From-SVN: r215136

i386.exp: Only run vect-args.c tests if runtest_file_p says they should be run.

* gcc.target/i386/i386.exp: Only run vect-args.c tests
if runtest_file_p says they should be run.

From-SVN: r215135

re PR ipa/61659 (Extra undefined symbol because of devirtualization)

PR c++/61659
* decl.c (grokfndecl): Don't set DECL_COMDAT on static inlines.
(duplicate_decls, start_decl): Likewise.
* pt.c (check_explicit_specialization): Likewise.
(push_template_decl_real): Or static templates.

From-SVN: r215134

error_for_asm and warning_for_asm take const rtx_insn *

gcc/ChangeLog:
2014-09-10  David Malcolm  <dmalcolm@redhat.com>

* final.c (this_is_asm_operands): Strengthen this variable from
rtx to const rtx_insn *.
* output.h (this_is_asm_operands): Likewise.
* rtl-error.c (location_for_asm): Strengthen param "insn" from
const_rtx to const rtx_insn *.
(diagnostic_for_asm): Likewise.
* rtl-error.h (error_for_asm): Likewise.
(warning_for_asm): Likewise.

From-SVN: r215133

insn_extract takes an rtx_insn

gcc/ChangeLog:
2014-09-10  David Malcolm  <dmalcolm@redhat.com>

* genextract.c (print_header): When writing out insn_extract to
insn-extract.c, strengthen the param "insn" from rtx to rtx_insn *.
* recog.h (insn_extract): Strengthen the param from rtx to
rtx_insn *.

From-SVN: r215132

[AArch64 Testsuite] Add a test of the vst[234](q?) intrinics

        * gcc.target/aarch64/vstN_1.c: New test.

From-SVN: r215129

hook.in: Load the xmethods.

2014-09-10  Siva Chandra Reddy  <sivachandra@google.com>

* python/hook.in: Load the xmethods.
* python/Makefile.am (nobase_python_DATA): Add xmethods.py.
* python/Makefile.in: Regenerated.
* python/libstdcxx/v6/xmethods.py: New file.
* testsuite/lib/gdb-test.exp (gdb_version_check_xmethods): New
function.
(gdb-test): New optional argument LOAD_XMETHODS.  Load xmethods
python script if LOAD_XMETHODS is true.
* testsuite/libstdc++-xmethods/unique_ptr.cc: New file.
* testsuite/libstdc++-xmethods/vector.cc: New file.
* testsuite/libstdc++-xmethods/xmethods.exp: New file.

From-SVN: r215128

install.texi (Prerequisites): Note Tcl 8.6 bug fixed in 8.6.1.

* doc/install.texi (Prerequisites): Note Tcl 8.6 bug fixed in
8.6.1.

From-SVN: r215127

[AArch64 Testsuite] Add a test of the vldN_lane intrinsic

        * gcc.target/aarch64/vldN_lane_1.c: New test.

From-SVN: r215126

cgraphunit.c (expand_thunk): If not expanding, set analyzed flag.

2014-09-10  Martin Jambor  <mjambor@suse.cz>

* cgraphunit.c (expand_thunk): If not expanding, set analyzed flag.
(analyze): Do not set analyze flag if expand_thunk returns false;.
(create_wrapper): Likewise.
* cgraphclones.c (duplicate_thunk_for_node): Likewise.

From-SVN: r215123

re PR middle-end/61654 (ICE in release_function_body, at cgraph.c:1699)

2014-09-10  Martin Jambor  <mjambor@suse.cz>

PR ipa/61654
* cgraphclones.c (duplicate_thunk_for_node): Copy arguments of the
new decl properly.  Analyze the new thunk if it is expanded.

gcc/testsuite/
* g++.dg/ipa/pr61654.C: New test.

From-SVN: r215122

coretypes.h (struct _dont_use_rtx_insn_here_, rtx_insn): Define.

* coretypes.h (struct _dont_use_rtx_insn_here_, rtx_insn)
[USED_FOR_TARGET]: Define.

From-SVN: r215121

MIPS: Do not reload unallocated FP_REGS pseudos via GR_REGS

gcc/

* config/mips/mips.c (mips_secondary_reload_class): Handle
regno < 0 case.

From-SVN: r215120

Fix ICE in bitmap routines with LRA and inline assembly language

gcc/

* lra-lives.c (process_bb_lives): Replace assignment with bitwise OR
assignment.

From-SVN: r215119

flag-types.h (enum sanitize_code): Add SANITIZE_NONNULL_ATTRIBUTE and SANITIZE_RETURNS_NONNULL_ATTRIBUTE...

gcc/
* flag-types.h (enum sanitize_code): Add SANITIZE_NONNULL_ATTRIBUTE
and SANITIZE_RETURNS_NONNULL_ATTRIBUTE, or them into SANITIZE_UNDEFINED.
* opts.c (common_handle_option): Handle SANITIZE_NONNULL_ATTRIBUTE and
SANITIZE_RETURNS_NONNULL_ATTRIBUTE and disable
flag_delete_null_pointer_checks for them.
* sanitizer.def (BUILT_IN_UBSAN_HANDLE_NONNULL_ARG,
BUILT_IN_UBSAN_HANDLE_NONNULL_ARG_ABORT,
BUILT_IN_UBSAN_HANDLE_NONNULL_RETURN,
BUILT_IN_UBSAN_HANDLE_NONNULL_RETURN_ABORT): New.
* ubsan.c (instrument_bool_enum_load): Set *gsi back to
stmt's iterator.
(instrument_nonnull_arg, instrument_nonnull_return): New functions.
(pass_ubsan::gate): Return true even for SANITIZE_NONNULL_ATTRIBUTE
or SANITIZE_RETURNS_NONNULL_ATTRIBUTE.
(pass_ubsan::execute): Call instrument_nonnull_{arg,return}.
* doc/invoke.texi (-fsanitize=nonnull-attribute,
-fsanitize=returns-nonnull-attribute): Document.
gcc/testsuite/
* c-c++-common/ubsan/attrib-3.c: New test.
* c-c++-common/ubsan/nonnull-1.c: New test.
* c-c++-common/ubsan/nonnull-2.c: New test.
* c-c++-common/ubsan/nonnull-3.c: New test.
* c-c++-common/ubsan/nonnull-4.c: New test.
* c-c++-common/ubsan/nonnull-5.c: New test.
libsanitizer/
* ubsan/ubsan_handlers.cc, ubsan/ubsan_handlers.h: Cherry pick
upstream r215485, r217389, r217391 and r217400.

From-SVN: r215118

ubsan.h (struct ubsan_mismatch_data): Removed.

* ubsan.h (struct ubsan_mismatch_data): Removed.
(ubsan_create_data): Remove MISMATCH argument, add LOCCNT argument.
* ubsan.c (ubsan_source_location): For unknown locations,
pass { NULL, 0, 0 } instead of { "<unknown>", x, y }.
(ubsan_create_data): Remove MISMATCH argument, add LOCCNT argument.
Allow more than one location and arbitrary extra arguments passed
in ... instead of through MISMATCH pointer.
(ubsan_instrument_unreachable, ubsan_expand_bounds_ifn,
ubsan_expand_null_ifn, ubsan_build_overflow_builtin,
instrument_bool_enum_load, ubsan_instrument_float_cast): Adjust
callers.
c-family/
* c-ubsan.c (ubsan_instrument_division, ubsan_instrument_shift,
ubsan_instrument_vla, ubsan_instrument_return): Adjust
ubsan_create_data callers.
(ubsan_instrument_bounds): Don't emit UBSAN_BOUNDS at all if
index is constant or BIT_AND_EXPR with constant mask and is
small enough for the bound.
* c-gimplify.c (ubsan_walk_array_refs_r): For ADDR_EXPR of
ARRAY_REF, make sure the inner ARRAY_REF is not walked again.

From-SVN: r215117

AVX-512. Add patterns for compress, expand.

gcc/
* config/i386/sse.md
(define_mode_iterator VI48F): New.
(define_insn "<avx512>_compress<mode>_mask"): Rename from
"avx512f_compress<mode>_mask" and update mode iterator.
(define_insn "<avx512>_compressstore<mode>_mask"): Rename from
"avx512f_compressstore<mode>_mask" and update mode iterator.
(define_expand "<avx512>_expand<mode>_maskz"): Rename from
"avx512f_expand<mode>_maskz" and update mode iterator.
(define_insn "<avx512>_expand<mode>_mask"): Rename from
"avx512f_expand<mode>_mask" and update mode iterator.

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215108

AVX-512. Add reduce, range, fpclass insn patterns.

gcc/
* config/i386/i386.c
(ix86_expand_args_builtin): Handle avx512dq_rangepv8df_mask_round,
avx512dq_rangepv16sf_mask_round, avx512dq_rangepv4df_mask,
avx512dq_rangepv8sf_mask, avx512dq_rangepv2df_mask,
avx512dq_rangepv4sf_mask.
* config/i386/sse.md
(define_c_enum "unspec"): Add UNSPEC_REDUCE, UNSPEC_FPCLASS,
UNSPEC_RANGE.
(define_insn "<mask_codefor>reducep<mode><mask_name>"): New.
(define_insn "reduces<mode>"): Ditto.
(define_insn "avx512dq_rangep<mode><mask_name><round_saeonly_name>"):
Ditto.
(define_insn "avx512dq_ranges<mode><round_saeonly_name>"): Ditto.
(define_insn "avx512dq_fpclass<mode><mask_scalar_merge_name>"): Ditto.
(define_insn "avx512dq_vmfpclass<mode>"): Ditto..

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215107

AVX-512. Update float unspecs: storeu, rcp14, rsqrt14, scalef, getexp, fixupimm, rndscale, getmant.

gcc/
* config/i386/i386.c
(avx512f_vgetmantv2df_round): Rename from "avx512f_getmantv2df_round".
(avx512f_vgetmantv4sf_round): Rename from "avx512f_vgetmantv4sf_round".
(ix86_expand_args_builtin): Handle avx512vl_getmantv8sf_mask,
avx512vl_getmantv4df_mask, avx512vl_getmantv4sf_mask,
avx512vl_getmantv2df_mask.
(ix86_expand_round_builtin): Handle avx512f_vgetmantv2df_round,
avx512f_vgetmantv4sf_round.
* config/i386/sse.md
(define_insn "<avx512>_storeu<ssemodesuffix><avxsizesuffix>_mask"):
Rename from "avx512f_storeu<ssemodesuffix>512_mask" and update
mode iterator.
(define_insn "<mask_codefor>rcp14<mode><mask_name>"): Use VF_AVX512VL.
(define_insn "<mask_codefor>rsqrt14<mode><mask_name>"): Ditto.
(define_insn "<avx512>_scalef<mode><mask_name><round_name>"): Rename
from "avx512f_scalef<mode><mask_name><round_name>" and update mode
iterator..
(define_insn "<avx512>_getexp<mode><mask_name><round_saeonly_name>"):
Rename from "avx512f_getexp<mode><mask_name><round_saeonly_name>" and
update mode iterator.
(define_expand
"<avx512>_fixupimm<mode>_maskz<round_saeonly_expand_name>"): Rename from
"avx512f_fixupimm<mode>_maskz<round_saeonly_expand_name>" and update
mode iterator.
(define_insn
"<avx512>_fixupimm<mode><sd_maskz_name><round_saeonly_name>"): Rename
from "avx512f_fixupimm<mode><sd_maskz_name><round_saeonly_name>" and
update mode iterator.
(define_insn "<avx512>_fixupimm<mode>_mask<round_saeonly_name>"): Rename
from "avx512f_fixupimm<mode>_mask<round_saeonly_name>" and update mode
iterator..
(define_insn
"<avx512>_rndscale<mode><mask_name><round_saeonly_name>"): rename from
"avx512f_rndscale<mode><mask_name><round_saeonly_name>" and update
mode iterator..
(define_insn "<avx512>_getmant<mode><mask_name><round_saeonly_name>"):
Rename from "avx512f_getmant<mode><mask_name><round_saeonly_name>" and
update mode iterator.
(define_insn "avx512f_vgetmant<mode><round_saeonly_name>"): Rename from
"avx512f_getmant<mode><round_saeonly_name>".

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215106

re PR lto/63166 (ICE (LTO): ipa_intraprocedural_devirtualization, at ipa-prop.c:2611)

PR ipa/63166
* ipa-prop.c (compute_known_type_jump_func): Fix conditional.

* g++.dg/lto/pr63166_0.ii: New testcase.
* g++.dg/lto/pr63166_1.ii: New testcase.

From-SVN: r215105

AVX-512. Extend FMA patterns.

gcc/
* config/i386/sse.md (define_mode_iterator VF_AVX512VL): New.
(define_mode_iterator FMAMODEM): Allow 128/256bit EVEX version.
(define_mode_iterator FMAMODE_AVX512): New.
(define_mode_iterator FMAMODE): Remove conditions.
(define_expand "fma4i_fmadd_<mode>"): Use FMAMODE_AVX512 mode iterator.
(define_expand "<avx512>_fmadd_<mode>_maskz<round_expand_name>"): Rename
from "<avx512>_fmadd_<mode>_maskz<round_expand_name>" and use VF_AVX512VL
mode iterator.
(define_mode_iterator FMAMODE_NOVF512): Remove.
(define_insn "*fma_fmadd_<mode>"): Rename from
"<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name><round_name>" and use
FMAMODE mode iterator.
(define_mode_iterator VF_SF_AVX512VL): New.
(define_insn "<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name><round_name>"):
Use VF_SF_AVX512VL mode iterator.
(define_insn "<avx512>_fmadd_<mode>_mask<round_name>"): Rename from
"avx512f_fmadd_<mode>_mask<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "<avx512>_fmadd_<mode>_mask3<round_name>"): Rename from
"avx512f_fmadd_<mode>_mask3<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "*fma_fmsub_<mode>"): Rename from
"<sd_mask_codefor>fma_fmsub_<mode><sd_maskz_name><round_name>" and use
FMAMODE mode iterator.
(define_insn "<sd_mask_codefor>fma_fmsub_<mode><sd_maskz_name><round_name>"):
Use VF_SF_AVX512VL mode iterator.
(define_insn "<avx512>_fmsub_<mode>_mask<round_name>"): Rename from
"avx512f_fmsub_<mode>_mask<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "<avx512>_fmsub_<mode>_mask3<round_name>"): Rename from
"avx512f_fmsub_<mode>_mask3<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "*fma_fnmadd_<mode>"): Rename from
"<sd_mask_codefor>fma_fnmadd_<mode><sd_maskz_name><round_name>" and
use FMAMODE mode iterator.
(define_insn "<sd_mask_codefor>fma_fnmadd_<mode><sd_maskz_name><round_name>"):
Use VF_SF_AVX512VL mode iterator.
(define_insn "<avx512>_fnmadd_<mode>_mask<round_name>"): Rename from
"avx512f_fnmadd_<mode>_mask<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "<avx512>_fnmadd_<mode>_mask3<round_name>"): Rename from
"avx512f_fnmadd_<mode>_mask3<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "*fma_fnmsub_<mode>"): Rename from
"<sd_mask_codefor>fma_fnmsub_<mode><sd_maskz_name><round_name>" and use
FMAMODE mode iterator.
(define_insn "<sd_mask_codefor>fma_fnmsub_<mode><sd_maskz_name><round_name>"):
Use VF_SF_AVX512VL mode iterator.
(define_insn "<avx512>_fnmsub_<mode>_mask<round_name>"): Rename from
"avx512f_fnmsub_<mode>_mask<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "<avx512>_fnmsub_<mode>_mask3<round_name>"): Rename from
"avx512f_fnmsub_<mode>_mask3<round_name>" and use VF_AVX512VL mode
iterator.
(define_expand "<avx512>_fmaddsub_<mode>_maskz<round_expand_name>"):
Rename from "avx512f_fmaddsub_<mode>_maskz<round_expand_name>" and
use VF_AVX512VL mode iterator.
(define_insn "*fma_fmaddsub_<mode>"): Rename from
"<sd_mask_codefor>fma_fmaddsub_<mode><sd_maskz_name><round_name>" and
remove subst usage.
(define_insn "<sd_mask_codefor>fma_fmaddsub_<mode><sd_maskz_name><round_name>"):
Use VF_SF_AVX512VL mode iterator.
(define_insn "<avx512>_fmaddsub_<mode>_mask<round_name>"): Rename from
"avx512f_fmaddsub_<mode>_mask<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "<avx512>_fmaddsub_<mode>_mask3<round_name>"): Rename from
"avx512f_fmaddsub_<mode>_mask3<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "*fma_fmsubadd_<mode>"): Rename from
"<sd_mask_codefor>fma_fmsubadd_<mode><sd_maskz_name><round_name>" and
remove usage of subst.
(define_insn "<sd_mask_codefor>fma_fmsubadd_<mode><sd_maskz_name><round_name>"):
Use VF_SF_AVX512VL mode iterator.
(define_insn "<avx512>_fmsubadd_<mode>_mask<round_name>"): Rename from
"avx512f_fmsubadd_<mode>_mask<round_name>" and use VF_AVX512VL mode
iterator.
(define_insn "<avx512>_fmsubadd_<mode>_mask3<round_name>"): Rename from
"avx512f_fmsubadd_<mode>_mask3<round_name>" and use VF_AVX512VL mode
iterator.

Co-Authored-By: Andrey Turetskiy <andrey.turetskiy@intel.com>
Co-Authored-By: Anna Tikhonova <anna.tikhonova@intel.com>
Co-Authored-By: Ilya Tocar <ilya.tocar@intel.com>
Co-Authored-By: Ilya Verbin <ilya.verbin@intel.com>
Co-Authored-By: Kirill Yukhin <kirill.yukhin@intel.com>
Co-Authored-By: Maxim Kuznetsov <maxim.kuznetsov@intel.com>
Co-Authored-By: Michael Zolotukhin <michael.v.zolotukhin@intel.com>
From-SVN: r215104

re PR libgcc/56846 (_Unwind_Backtrace on ARM and noexcept)

2014-09-10  Tony Wang  <tony.wang@arm.com>

    libstdc++-v3/
    PR target/56846
    * libsupc++/eh_personality.cc (PERSONALITY_FUNCTION):
    Return with CONTINUE_UNWINDING when the state pattern
    contains: _US_VIRTUAL_UNWIND_FRAME | _US_FORCE_UNWIND

From-SVN: r215101

revert: calls.c (precompute_arguments): Check promoted_for_signed_and_unsigned_p and set the promoted mode.

gcc/ChangeLog:

2014-09-10  Kugan Vivekanandarajah  <kuganv@linaro.org>

Revert r213751:
* calls.c (precompute_arguments): Check
 promoted_for_signed_and_unsigned_p and set the promoted mode.
(promoted_for_signed_and_unsigned_p): New function.
(expand_expr_real_1): Check promoted_for_signed_and_unsigned_p
and set the promoted mode.
* expr.h (promoted_for_signed_and_unsigned_p): New function definition.
* cfgexpand.c (expand_gimple_stmt_1): Call emit_move_insn if
SUBREG is promoted with SRP_SIGNED_AND_UNSIGNED.

From-SVN: r215100

Daily bump.

From-SVN: r215099

opth-gen.awk: Generate mapping from cpp message reasons to the options that enable them.

gcc/ChangeLog:

2014-09-09  Manuel López-Ibáñez  <manu@gcc.gnu.org>

* opth-gen.awk: Generate mapping from cpp message reasons to the
options that enable them.
* doc/options.texi (CppReason): Document.

gcc/c-family/ChangeLog:

2014-09-09  Manuel López-Ibáñez  <manu@gcc.gnu.org>

* c.opt: Add CppReason to various flags.
(Wdate-time): Re-sort.
* c-common.c: Include c-common.h earlier.
(struct reason_option_codes_t): Delete.
(c_option_controlling_cpp_error): Prefix global type and struct
with cpp_.

From-SVN: r215095

invoke.texi (Wnormalized=): Update.

gcc/ChangeLog:

2014-09-09  Manuel López-Ibáñez  <manu@gcc.gnu.org>

* doc/invoke.texi (Wnormalized=): Update.

libcpp/ChangeLog:

2014-09-09  Manuel López-Ibáñez  <manu@gcc.gnu.org>

* include/cpplib.h (struct cpp_options): Declare warn_normalize as
int instead of enum.

gcc/c-family/ChangeLog:

2014-09-09  Manuel López-Ibáñez  <manu@gcc.gnu.org>

* c.opt (Wnormalized): New.
(Wnormalized=): Use Enum and Reject Negative.
* c-opts.c (c_common_handle_option): Do not handle Wnormalized here.

gcc/testsuite/ChangeLog:

2014-09-09  Manuel López-Ibáñez  <manu@gcc.gnu.org>

* gcc.dg/cpp/warn-normalized-3.c: Delete useless dg-prune-output.

From-SVN: r215093

Fix pad status check.

2014-09-10  Janne Blomqvist  <jb@gcc.gnu.org>

* io/transfer.c (read_block_form): Fix pad status check (found by
Thomas Schwinge with -Wlogical-not-parentheses).

From-SVN: r215092