2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Analyze_Object_Declaration): New function
Has_Delayed_Aspect, used to defer resolution of an aggregate
expression when the object declaration carries aspects Address
and/or Alignment.
* freeze.adb (Freeze_Object_Declaration): New subsidiary procedure
to Freeze_Entity. In addition to the previous processing steps
at the freeze point of an object, this procedure also handles
aggregates in object declarations, when the declaration carries
delayed aspects that require that the initialization of the
object be attached to its freeze actions.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb (Analyze_Subprogram_Declaration): Following
AI12-0147, null procedures and expression functions are allowed
in protected bodies.
From-SVN: r223041
2015-05-12 Tristan Gingold <gingold@adacore.com>
* i-cpoint.adb (Copy_Terminated_Array): Copy nothing if Length is 0.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Complete_Private_Subtype): Propagate
Has_Delayed_Aspects flag from private to full view, to ensure
that predicate functions are constructed.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb (Process_Formals): If a tagged formal is an
incomplete class-wide type, the subprogram must have a delayed
freeze even though the opertation is not a primitive of the
type. THis ensures that the backend can recover the full view
when elaborating the subprogram declaration.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* exp_util.adb (Get_Current_Value_Condition): Nothing to be
done if an elsif part has been rewritten so that it is not part
of an enclosing if_statement.
From-SVN: r223040
2015-05-12 Robert Dewar <dewar@adacore.com>
* sem_type.adb, sem_ch10.adb, freeze.adb, sem_ch6.adb, exp_disp.adb:
Minor reformatting.
2015-05-12 Bob Duff <duff@adacore.com>
* exp_attr.adb (Size): Remove unnecessary check for types with
unknown discriminants. That was causing the compiler to build
a function call _size(T), where T is a type, not an object.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch4.adb (Extended_Primitive_Ops): Exclude overriding
primitive operations of a type extension declared in the package
body, to prevent duplicates in extended list.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Analyze_Component_Declaration): If the component is
an unconstrained synchronized type with discriminants, create a
constrained default subtype for it, so that the enclosing record
can be given the proper size.
* sem_util.adb (Build_Default_Subtype): If the subtype is created
for a record discriminant, do not analyze the declarztion at
once because it is added to the freezing actions of the enclosing
record type.
From-SVN: r223039
2015-05-12 Robert Dewar <dewar@adacore.com>
* exp_unst.adb (Get_Real_Subp): New subprogram.
(Unnest_Subprogram): Use Get_Real_Subp.
(Uplev_Refs_For_One_Subp): Skip if no ARECnU entity.
(Uplev_Refs_For_One_Subp): Use actual subtype in unconstrained case.
2015-05-12 Robert Dewar <dewar@adacore.com>
* a-reatim.adb ("/"): Add explicit check for Time_Span_First / -1.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch4.adb (Extended_Primitive_Ops): New subprogram,
auxiliary to Try_Primitive_Operation to handle properly prefixed
calls where the operation is not a primitive of the type, but
is declared in the package body that is in the immediate scope
of the type.
From-SVN: r223036
2015-05-12 Robert Dewar <dewar@adacore.com>
* sem_util.adb (Is_Variable): Allow X'Deref(Y) as a variable.
2015-05-12 Ed Schonberg <schonberg@adacore.com>
* sem_ch8.adb (Find_Expanded_Name): Handle properly a fully
qualified name for an instance of a generic grand-child unit in
the body its parent.
2015-05-12 Robert Dewar <dewar@adacore.com>
* exp_unst.adb (Upref_Name): New subprogram.
(Unnest_Subprogram): Use Upref_Name.
(Unnest_Subprogram): Use new Deref attribute.
* exp_unst.ads: Doc updates.
2015-05-12 Thomas Quinot <quinot@adacore.com>
* adaint.c: Enable Large File Support in adaint so that __gnat_readdir
can access files on filesystems mounted from servers that use large
NFS file handles.
From-SVN: r223035
2015-05-12 Thomas Preud'homme <thomas.preudhomme@arm.com>
* combine.c i(set_nonzero_bits_and_sign_copies): Split code updating
rsp->sign_bit_copies and rsp->nonzero_bits into ...
(update_rsp_from_reg_equal): This. Also use REG_EQUAL note on src if
present to get more accurate information about the number of sign bit
copies and non zero bits.
From-SVN: r223034
2015-05-12 Richard Biener <rguenther@suse.de>
* tree-vect-slp.c (vect_build_slp_tree_1): For BB vectorization
do not allow unrolling.
* gcc.dg/vect/bb-slp-35.c: New testcase.
From-SVN: r223033
If gimple_asm_volatile_p is correct, no point re-checking.
This is also done by the C and C++ front ends, but not Ada.
So we can't yet trust ASM_VOLATILE_P from the front end.
From-SVN: r223024
* class.c (fixup_type_variants): Do not copy TYPE_METHODS
(one_inheriting_sig): Assert tat we always set TYPE_METHODS of main variant.
* semantics.c (finish_member_declaration): Likewise.
* method.c (lazily_declare_fn): Allways add method to main variant list.
* dwarf2out.c (gen_member_die): Sanity check that we access TYPE_MAIN_VARIANT
for TYPE_METHODS.
* function.c (use_register_for_decl): Look for TYPE_MAIN_VARIANT when checking
TYPE_METHODS.
* tree.c (free_lang_data_in_type): See TYPE_METHODS to error_mark_node
if non-null.
(build_distinct_type_copy): Clear TYPE_METHODS.
(verify_type_variant): Verify that TYPE_METHODS is NULL for variants.
(verify_type): Allow TYPE_METHODS to be error_mark_node.
* tree.def: Update docs of YTPE_STUB_DECL and TYPE_METHODS.
From-SVN: r223021
gcc/
* genrecog.c (match_pattern_1): Expect the pattern to be a SEQUENCE
for define_peephole2s.
(get_peephole2_pattern): New function.
(main): Use it. Call validate_pattern.
From-SVN: r223015
* config/avr/avr.c (avr_hard_regno_call_part_clobbered): Use
LAST_CALLEE_SAVED_REG instead of hard-coded register number.
(Last callee saved reg is different for AVR_TINY architecture)
From-SVN: r223009
When libgo is not optimized the static function profilealloc
in malloc.goc shows up in the stack trace. Rename it to
runtime_profilealloc so that runtime/pprof.printStackRecord
ignores it.
From-SVN: r223006
Fix PR66105
2015-05-11 Markus Trippelsdorf <markus@trippelsdorf.de>
PR bootstrap/66105
* config/rs6000/option-defaults.h: Add space between string literal
and macro name.
From-SVN: r223002
2015-05-11 Thomas Preud'homme <thomas.preudhomme@arm.com>
gcc/
* gcc.target/arm/pr64616.c: Test dump rather than assembly to work
accross ARM targets.
From-SVN: r223001
gcc/
PR rtl-optimization/66076
* rtlanal.c (generic_subrtx_iterator <T>::add_single_to_queue):
Don't grow the heap array if it is already big enough from a
previous iteration.
gcc/testsuite/
PR rtl-optimization/66076
* gcc.dg/torture/pr66076.c: New test.
From-SVN: r222999
* ipa-devirt.c (warn_types_mismatch): Do not ICE when warning about
mismatch between C and C++ type; compoare correctly ARG_TYPES
for non-prototypes and output correctly parameter index for METHOD_TYPE.
(odr_types_equivalent_p): Fix wording of warning about attributes;
it is OK to match prototype and non-prototype.
From-SVN: r222985
2015-05-10 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/66041
* frontend-passes.c (scalarized_expr): Set correct dimension and
shape for the expression to be passed to lbound. Remove trailing
references after array refrence.
(inline_matmul_assign): Remove gfc_copy_expr from calls
to scalarized_expr().
2015-05-10 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/66041
* gfortran.dg/inline_matmul_7.f90: New test.
* gfortran.dg/inline_matmul_8.f90: New test.
* gfortran.dg/inline_matmul_9.f90: New test.
From-SVN: r222982
gcc/fortran/
* simplify.c (simplify_bound_dim): Don't check for emptyness
in the case of cobound simplification. Factor lower/upper
bound differenciation before the actual simplification.
(simplify_bound): Remove assumed shape specific simplification.
Don't give up early for the lbound of an assumed shape.
gcc/testsuite/
* gfortran.dg/bound_simplification_5.f90: New.
From-SVN: r222979
