Fix PR tree-optimization/72810
gcc/ChangeLog:
PR tree-optimization/72810
* tree-vrp.c (simplify_switch_using_ranges): Avoid changing
the type of the case labels when truncating.
gcc/testsuite/ChangeLog:
PR tree-optimization/72810
* gcc.dg/tree-ssa/vrp110.c: New test.
From-SVN: r239174
Fix PR Target/72819.
gcc/
PR Target/72819
* config/aarch64/aarch64.h (aarch64_fp16_type_node): Declare.
(aarch64_fp16_ptr_type_node): Likewise.
* config/aarch64/aarch64-simd-builtins.c
(aarch64_fp16_ptr_type_node): Define.
(aarch64_init_fp16_types): New, refactored out of...
(aarch64_init_builtins): ...here, update to call
aarch64_init_fp16_types.
* config/aarch64/aarch64.c (aarch64_gimplify_va_arg_expr): Handle
HFmode.
(aapcs_vfp_sub_candidate): Likewise.
gcc/testsuite/
PR Target/72819
* gcc.target/aarch64/aapcs64/abitest-common.h: Define half-precision
registers.
* gcc.target/aarch64/aapcs64/abitest.S (dumpregs): Add assembly for
saving the half-precision registers.
* gcc.target/aarch64/aapcs64/func-ret-1.c: Test that an __fp16
value is returned in h0.
* gcc.target/aarch64/aapcs64/test_2.c: Check that __FP16 arguments
are passed in FP/SIMD registers.
* gcc.target/aarch64/aapcs64/test_27.c: New, test that __fp16 HFA
passing works corrcetly.
* gcc.target/aarch64/aapcs64/type-def.h (hfa_f16x1_t): New.
(hfa_f16x2_t): Likewise.
(hfa_f16x3_t): Likewise.
* gcc.target/aarch64/aapcs64/va_arg-1.c: Check that __fp16 values
are promoted to double and passed in a double register.
* gcc.target/aarch64/aapcs64/va_arg-2.c: Check that __fp16 values
are promoted to double and stacked.
* gcc.target/aarch64/aapcs64/va_arg-4.c: Check stacking of HFA of
__fp16 data types.
* gcc.target/aarch64/aapcs64/va_arg-5.c: Likewise.
* gcc.target/aarch64/aapcs64/va_arg-16.c: New, check HFAs of
__fp16 first get passed in FP/SIMD registers, then stacked.
From-SVN: r239173
* include/std/functional (__callable_functor): Remove.
(_Function_handler::_M_invoke): Use __invoke instead of
__callable_functor or mem_fn.
(function::_Callable): Use lvalue in result_of expression.
(function): Remove TODO comments about allocators.
* testsuite/20_util/function/cons/refqual.cc: New test.
From-SVN: r239166
2016-08-05 Richard Biener <rguenther@suse.de>
* tree-ssa-threadupdate.c (thread_block_1): Remove unnecessary
restriction on threading to a loop header.
From-SVN: r239164
2016-08-05 Richard Biener <rguenther@suse.de>
* tree-cfgcleanup.c (tree_forwarder_block_p): Use bb_loop_header_p.
* cfghooks.c (force_nonfallthru): If we ended up splitting a latch
adjust loop info accordingly.
From-SVN: r239163
* config/rs6000/rs6000.c (rs6000_preferred_reload_class): Delete
code accidentally committed 2016-05-02 providing class when given
NO_REGS.
From-SVN: r239161
gcc/ChangeLog:
* tree-vrp.c (simplify_switch_using_ranges): Try to truncate
the case label ranges that partially overlap with OP's value
range.
gcc/testsuite/ChangeLog:
* gcc.dg/tree-ssa/vrp107.c: New test.
* gcc.dg/tree-ssa/vrp108.c: New test.
* gcc.dg/tree-ssa/vrp109.c: New test.
From-SVN: r239157
gcc/ChangeLog:
* gimple.c (preprocess_case_label_vec_for_gimple): When the case
labels are exhaustive, designate the label with the widest
range to be the default label.
gcc/testsuite/ChangeLog:
* gcc.dg/switch-10.c: New test.
From-SVN: r239146
2016-08-04 Marcel Böhme <boehme.marcel@gmail.com>
PR c++/71696
* cplus-dem.c: Prevent infinite recursion when there is a cycle
in the referencing of remembered mangled types.
(work_stuff): New stack to keep track of the remembered mangled
types that are currently being processed.
(push_processed_type): New method to push currently processed
remembered type onto the stack.
(pop_processed_type): New method to pop currently processed
remembered type from the stack.
(work_stuff_copy_to_from): Copy values of new variables.
(delete_non_B_K_work_stuff): Free stack memory.
(demangle_args): Push/Pop currently processed remembered type.
(do_type): Do not demangle a cyclic reference and push/pop
referenced remembered type.
From-SVN: r239143
When using gccgo on systems without full support for split stacks a
recursive template can overrun the available stack space. Reduce the
limit from 100000 to 10000 to make this less likely. It's still high
enough that real uses will work.
Reviewed-on: https://go-review.googlesource.com/25467
From-SVN: r239141
2016-08-04 Andrew Pinski <apinski@cavium.com>
* config/aarch64/aarch64.c (thunderx_vector_cost): New variable.
(thunderx_tunings): Use thunderx_vector_cost instead of
generic_vector_cost.
From-SVN: r239135
PR rtl-optimization/71779
* emit-rtl.c (set_reg_attrs_from_value): Only propagate REG_POINTER,
if the value was sign-extended according to POINTERS_EXTEND_UNSIGNED
or if it was truncated.
From-SVN: r239123
* testsuite/20_util/enable_shared_from_this/members/reinit.cc: Use
effective target not dg-options. Move check for feature-test macro to:
* testsuite/20_util/enable_shared_from_this/members/weak_from_this.cc:
New test.
From-SVN: r239121
gcc/ChangeLog:
2016-08-04 Kugan Vivekanandarajah <kuganv@linaro.org>
* tree-vrp.c (set_value_range): Use vrp_equiv_obstack with
BITMAP_ALLOC.
(add_equivalence): Likewise.
(get_value_range): Allocate value range with vrp_value_range_pool.
(vrp_initialize): Initialize vrp_equiv_obstack for equiv allocation.
(vrp_finalize): Relase vrp_equiv_obstack and vrp_value_range_pool.
From-SVN: r239113
Patrick Palka
James Greenhalgh
Jonathan Wakely
Martin Liska
Nathan Sidwell
Richard Biener
Kugan Vivekanandarajah
Alan Modra
GCC Administrator
Uros Bizjak
David Malcolm
Prathamesh Kulkarni
Paolo Carlini
Ian Lance Taylor
Marcel Böhme
Jason Merrill
Andrew Pinski
Thomas Schwinge
Bernd Edlinger
Georg-Johann Lay
Marek Polacek