This fixes the sorting of to copy symbols in lto_output introduced
with GCC 10 to not depend on the actual values of the randomness
we append to LTO section names but instead on the order they appear
in the unsorted array.
This fixed observed debug info differences due to tree merging
prevailing different early debug pointers.
2020-07-06 Richard Biener <rguenther@suse.de>
* lto-streamer-out.c (cmp_symbol_files): Use the computed
order map to sort symbols from the same sub-file together.
(lto_output): Compute a map of sub-file to an order number
it appears in the symbol output array.
This makes the formatting in <optional> consistent and also removes
redundant && tokens from template arguments for traits like
is_constructible and is_convertible.
libstdc++-v3/ChangeLog:
* include/std/optional (_Optional_payload_base, _Optional_base)
(optional, __optional_hash_call_base): Adjust whitespace and
other formatting. Remove redundant && tokens on template
arguments to type traits.
The standard rquires that std::make_optional is constrained similarly to
the std::optional constructors, which our implementation fails to do.
As a conforming extension this also adds a noexcept-specifier to each
std::make_optional overload.
libstdc++-v3/ChangeLog:
* include/std/optional (make_optional): Add enable_if
constraints and noexcept-specifier to each overload.
* testsuite/20_util/optional/make_optional-2.cc: New test.
libstdc++-v3/ChangeLog:
PR libstdc++/96036
* include/std/optional (optional): Add noexcept-specifier to
every constructor, assignment operator, emplace function and
dereference operator.
* testsuite/20_util/optional/assignment/noexcept.cc: New test.
* testsuite/20_util/optional/cons/noexcept.cc: New test.
The ABI exception helpers like __throw were being created by first
looking for them, and then adding if not found. Primarily because
libitm wasn't declaring them with the correct exception specifiers. I
fixed libitm a while back, so let's just use push_library_fn and let
the symbol table machinery deal with duplicates. push_library_fn was
making the assumtion there wasn't already a decl available, by always
returning the new decl. Bad things would happen if there was a
duplicate, because duplicate_decls explicitly gcc_frees the new decl.
Fixed by having it return whatever pushdecl returns.
gcc/cp/
* decl.c (push_library_fn): Return the decl pushdecl_toplevel returns.
* except.c (verify_library_fn): Replace with ...
(declare_library_fn_1): ... this fn. Always push the fn.
(declare_library_fn): Call it.
(build_throw): Call declare_library_fn_1.
gcc/testsuite/
* g++.dg/eh/builtin10.C: Adjust expected errors.
* g++.dg/eh/builtin11.C: Likewise.
* g++.dg/eh/builtin5.C: Likewise.
* g++.dg/eh/builtin6.C: Likewise.
* g++.dg/eh/builtin7.C: Likewise.
* g++.dg/eh/builtin9.C: Likewise.
* g++.dg/parse/crash55.C: Likewise.
In SELECT TYPE, the argument may be an incorrectly specified unlimited
polymorphic variable. Avoid a NULL pointer dereference for clean error
recovery.
gcc/fortran/
PR fortran/95980
* match.c (copy_ts_from_selector_to_associate, build_class_sym):
Distinguish between unlimited polymorphic and ordinary variables
to avoid NULL pointer dereference.
* resolve.c (resolve_select_type):
Distinguish between unlimited polymorphic and ordinary variables
to avoid NULL pointer dereference.
Since C++11 a semim-colon on its own at namespace scope is not invalid,
so do not give a pedantic diagnostic about it.
gcc/cp/ChangeLog:
PR c++/96068
* parser.c (cp_parser_toplevel_declaration): Only do pedwarn for
empty-declaration in C++98.
gcc/testsuite/ChangeLog:
* g++.old-deja/g++.bugs/900404_04.C: Add c++98_only selector to
dg-error for extra ';'.
* g++.old-deja/g++.law/missed-error2.C: Likewise.
This fixes bogus misalignment calculation for negative steps
since an assertion a previous comment indicated no longer holds:
/* DR_STEP(dr) is the same as -TYPE_SIZE of the scalar type,
otherwise we wouldn't be here. */
Thus the following replaces DR_STEP by -TYPE_SIZE.
2020-07-06 Richard Biener <rguenther@suse.de>
PR tree-optimization/96075
* tree-vect-data-refs.c (vect_compute_data_ref_alignment): Use
TYPE_SIZE_UNIT of the vector component type instead of DR_STEP
for the misalignment calculation for negative step.
* gcc.dg/vect/slp-46.c: New testcase.
gcc/ada/
* libgnat/g-catiio.ads: Document newly supported format. Add
ISO_Time constant, for convenience.
* libgnat/g-catiio.adb (Image_Helper): New helper function to do
all the formatting work, called by the two exported Image
functions. Add support for "%:::z" here. Add a Time_Zone
parameter used by the "%:::z" processing. This parameter is not
used for the actual time zone computations; local time is always
used for that, for ease of implementation reasons. It would
make sense to use Append throughout this function, but that's a
cleanup for another day.
(Image): Modify these to pass the local time zone, or the
specified time zone, as appropriate.
gcc/ada/
* libgnat/g-catiio.ads: Change the regular expression that
documents the allowed format to match what ISO-8601 allows.
* libgnat/g-catiio.adb (Scan_Subsecond): Rewrite so it doesn't
assume the subsecond comes last.
(Parse_ISO_8601): Parse an optional subsecond, followed by an
optional time zone, rather than making these alternatives to
each other.
gcc/ada/
* libgnat/g-catiio.adb (Parse_ISO_8601): Minor cleanups:
Give some objects clearer names.
Make some objects more local to where they are used.
Remove some validity checks that can't fail, because some of
the variables tested have been moved so they're not visible here.
Anyway, Wrong_Syntax is about errors in the input, not data
validity.
Use Time_Zone_Seen instead of Local_Sign = ' ' to determine
that a time zone indication was seen.
We don't need to distinguish two different kinds of
syntax error (End_Of_Source_Reached and Wrong_Syntax),
so use Wrong_Syntax always.
Remove comment, "Certain scanning scenarios may handle
this exception on their own."; there are no such scenarios.
gcc/ada/
* libgnat/g-catiio.ads, libgnat/g-catiio.adb (Image): New
function. It might seem like the local-time Image should be
implemented in terms of the Time_Zone Image, but that would be
far harder to implement, so we do it the other way around.
gcc/ada/
* libgnat/g-sechas.ads, libgnat/g-sechas.adb: Refactor to use
Stream_Element_Array as the internal buffer type.
* libgnat/g-shshco.adb: Adjust to use Stream_Element_Offset
instead of Integer as the index in the internal state buffer.
gcc/ada/
* sem_util.ads (Interval_Lists.Aggregate_Intervals): New
subprogram.
* sem_util.adb (Has_Null_Others_Choice,
Non_Static_Or_Null_Range, Interval_Lists.Aggregate_Intervals):
New subprograms.
(Is_Potentially_Unevaluated): Adding support to detect
potentially unevaluated components of array aggregates.
gcc/ada/
* libgnat/g-catiio.adb (Parse_ISO_8601): New name for
Parse_ISO_8861_UTC. 8601 is the correct ISO standard number.
Also, "UTC" was confusing. All Time values are represented in
UTC, but the ISO 8601 date strings include a time zone.
If a time zone was specified, call
Ada.Calendar.Formatting.Time_Of instead of
GNAT.Calendar.Time_Of, because the latter adjusts to the current
time zone, whereas we want to use (just) the time zone specified
in the ISO string. This allows us to pass Time_Zone instead to
Time_Of, instead of adjusting by Local_Disp by hand.
If no time zone was specified, call GNAT.Calendar.Time_Of as
before.
Use expanded names to clarify which Time_Of is being called.
Remove redundant comment, and move nonredundant part of the
commment to the spec.
(Value): Minor: use "not in" instead of "or else".
* libgnat/g-catiio.ads: Comment moved here. Correct the ISO
standard number.
* libgnat/g-calend.adb: Add ??? comments.
* libgnat/a-calend.ads, libgnat/a-calend.adb: Update obsolete
comments regarding the representation of type Time. Move the
information about the epoch (year 2150) to the spec, and avoid
uttering "2150" more than once.
* libgnat/a-catizo.ads (Time_Offset): Add comment.
gcc/ada/
* exp_ch5.adb (Find_Component): If the target type is a derived
record type and the required component is a discriminant that is
renamed in the derived type declaration, use the name of the
original discriminant to locate the intended target component.
gcc/ada/
* sinfo.ads: Indicate that Field 1 of an N_Delta_Aggregate is
unused. Previously it was erroneously labelled as holding an
Expressions list, in analogy with other aggregate constructs,
but there are no Expressions attached to this node syntactically
or semantically.
gcc/ada/
* exp_spark.adb (Expand_SPARK_N_Attribute_Reference): Apply
scalar range checks.
* sem_attr.adb (Resolve_Attribute): Do not set scalar range
checks when resolving attribute Update.
gcc/ada/
* sem_attr.adb (Analyze_Attribute, Attribute_Constrained): Issue
a warning if the attribute prefix is a current instance
reference within an aspect of a type or subtype.
(Address_Checks): Replace test of Is_Object (Ent) with
Is_Object_Reference (P) so that testing for current instances
will be done.
(Eval_Attribute): Add test for current instance reference, to
ensure that we still fold array attributes when current
instances are involved, since value prefixes are allowed for
array attributes, and will now be excluded by
Is_Object_Reference.
* sem_util.ads (Is_Current_Instance_Reference_In_Type_Aspect):
New exported query function.
* sem_util.adb (Is_Object_Reference): Return False for the case
where N is a current instance reference within an
aspect_specification of a type or subtype (basically if the
reference occurs within a predicate, invariant, or DIC aspect
expression).
(Is_Current_Instance_Reference_In_Type_Aspect): New function
that tests whether a node is a reference to a current instance
formal of a predicate, invariant, or
Default_Initial_Condition (DIC) subprogram.
gcc/ada/
* sem_res.adb (Resolve_Declare_Expression): New subprogram, to
install the scope holding local declarations of the expression,
before completing its resolution.
(Resolve): For an Expression_With_Actions that comes from a
source occurrence of a Declare_Expression, call new subprogram.
gcc/ada/
* sem_ch5.adb (Analyze_Loop_Parameter_Specification): Propagate
exception when switch -gnatdk is used and no previous errors are
present.
* sem_eval.adb (Compile_Time_Known_Value, Is_In_Range):
Likewise.
* sem_warn.adb (Operand_Has_Warnings_Suppressed): Likewise.
gcc/ada/
* exp_unst.adb (Needs_Fat_Pointer): Don't check for formal.
(Unnest_Subprogram): Use 'Unchecked_Access instead of 'Access
when populating activation record.
gcc/ada/
* exp_ch6.adb (Add_Simple_Call_By_Copy_Code): Rename parameter
from Bit_Packed_Array to Force.
(Expand_Actuals): Do not apply BIP processing if the subprogram
is intrinsic. Adjust calls to Add_Simple_Call_By_Copy_Code and
add one for In parameters whose actual is a CPP constructor call.
gcc/ada/
* sem_ch3.adb (Analyze_Object_Declaration): If the type is an
Unchecked_Union, and the expression is an aggregate. complete
the analysis and resolution of the aggregate, and treat like a
regular object declaration, instead of as a renaming
declarattion.
gcc/ada/
* exp_ch9.adb (Is_Potentially_Large_Family): Add documentation.
(Actual_Index_Expression): Use Entry_Index_Type.
(Build_Entry_Count_Expression): Likewise.
(Build_Find_Body_Index): Likewise.
(Collect_Entry_Families): Likewise. Use directly the bounds of
the index type to find out whether the family is large.
(Entry_Index_Expression): Likewise.
gcc/ada/
* exp_aggr.adb (Aggr_Assignment_OK_For_Backend): Move to library
level and use a new predicate Is_OK_Aggregate to recognize the
aggregates suitable for direct assignment by the back-end.
(Convert_Array_Aggr_In_Allocator): If neither in CodePeer mode nor
generating C code, generate a direct assignment instead of further
expanding if Aggr_Assignment_OK_For_Backend returns true.
gcc/ada/
* sem_aux.adb: Add a with clause for Nlists.
(Nearest_Ancestor): Test for the case of concurrent
types (testing for both Is_Concurrent_Type and
Is_Concurrent_Record_Type), and return the first ancestor in the
Interfaces list if present (otherwise will return Empty if no
interfaces).
* sem_ch13.adb (Build_Predicate_Functions): Add a ??? comment
about missing handling for adding predicates when they can be
inherited from multiple progenitors.
The following patch adds support for three-input addition instructions to
the nvptx backend. The PTX ISA's "vadd.u32.u32.u32.add d, a, b, c"
instruction effectively implements 32-bit d = a+b+c, and the
"vsub.u32.u32.u32 d,a,b,c" instruction that provides 32-bit d = (a-b)+c.
The hope is that these mnemonics help ptxas generate the low-level
hardware's IADD3 instruction.
2020-07-06 Roger Sayle <roger@nextmovesoftware.com>
gcc/ChangeLog:
* config/nvptx/nvptx.md (*vadd_addsi4): New instruction.
(*vsub_addsi4): New instruction.
gcc/testsuite/ChangeLog:
* gcc.target/nvptx/vadd_add.c: New test.
* gcc.target/nvptx/vsub_add.c: New test.
Combine likes to change a zero-extension / and + shift as seen
in the test-case source to a logical shift followed by an and of
the shifted mask, like:
lsrq 1,r0
and.d 0x7f,r0
This was observed in the hot loop of coremark crcu16 and crcu32,
when doing other changes affecting instruction selection. While
fixable by other means (like instruction costs or combine
patches), I wanted to break this out from those "other means".
The similarity to extant peephole optimizations is not
deliberate.
I noticed some paths to other peephole2 test-cases have changed
due to moves and renaming, so I updated them.
gcc:
* config/cris/cris.md (movulsr): New peephole2.
gcc/testsuite:
* gcc.target/cris/peep2-movulsr.c: New test.
Yet another misnumbering of operands: the asserted non-overlap
would be the only benign operands overlap. "Suddenly" exposed
by g++.dg/cpp0x/pr81325.C when testing unrelated changes
affecting register allocation.
To wit, operands 2 and 1 are the only ones that are safe for
overlap, it's only that it doesn't seem to make much sense to
write the address of the atomic data as the atomic data.
gcc:
* config/cris/sync.md ("cris_atomic_fetch_<atomic_op_name><mode>_1"):
Correct gcc_assert of overlapping operands.
The code in cris_select_cc_mode for selecting CC_NZmode was
partly inconsistent with the comment and partly seemed
ambiguous. I couldn't find a reason why I qualified selection
of CC_NZmode on the setting operation once a matching user was
spotted, so I just removed that. The cris.c update was due to
observing the new test-case failing; the CC_NZmode compare
wasn't eliminated.
The recently re-instated adds/addu/subs/subu/bound patterns are
rewritten to replace the use of match_operator with iterators.
gcc:
* config/cris/cris.c (cris_select_cc_mode): Always return
CC_NZmode for matching comparisons. Clarify comments.
* config/cris/cris-modes.def: Clarify mode comment.
* config/cris/cris.md (plusminus, plusminusumin, plusumin): New
code iterators.
(addsub, addsubbo, nd): New code iterator attributes.
("*<addsub><su>qihi"): Rename from "*extopqihi". Use code
iterator constructs instead of match_operator constructs.
("*<addsubbo><su><nd><mode>si<setnz>"): Similar from
"*extop<mode>si<setnz>".
("*add<su>qihi_swap"): Similar from "*addxqihi_swap".
("*<addsubbo><su><nd><mode>si<setnz>_swap"): Similar from
"*extop<mode>si<setnz>_swap".
gcc/testsuite:
* gcc.target/cris/pr93372-39.c: New test.
When cleaning out the multitude of patterns with unknown
coverage, this one went the way of the bathwater. It's use is
barely common enough to mark when diffing libgcc, and has a
minimal impact on performance-testsuites. Anyway, reinstated
with a couple of test-cases. It's suboptimal of gcc-core not to
make use of the SImode pattern when performing HImode; see the
FIXME (which is actually also reinstated).
This version uses match_operator, for continuity but will be
replaced with a version making use of iterators (like it does
for the mode).
gcc:
* config/cris/cris.md ("*extopqihi", "*extop<mode>si<setnz>_swap")
("*extop<mode>si<setnz>", "*addxqihi_swap"): Reinstate.
gcc/testsuite:
* gcc.target/cris/pr93372-36.c, gcc.target/cris/pr93372-37.c,
gcc.target/cris/pr93372-38.c: New tests.
Apart from calling gfc_compare_interfaces to check interfaces against
global identifiers, this also sets and check a few sym->error flags
to avoid duplicate error messages. I thought about issuing errors
on mismatched interfaces, but when the procedure is not invoked,
a warning should be enough to alert the user.
gcc/fortran/ChangeLog:
PR fortran/27318
* frontend-passes.c (check_against_globals): New function.
(gfc_check_externals): Split; also invoke check_against_globals
via gfc_traverse_ns.
(gfc_check_externals0): Recursive part formerly in
gfc_check_externals.
* resolve.c (resolve_global_procedure): Set sym->error on
interface mismatch.
* symbol.c (ambiguous_symbol): Check for, and set sym->error.
gcc/testsuite/ChangeLog:
PR fortran/27318
* gfortran.dg/error_recovery_1.f90: Adjust test case.
* gfortran.dg/use_15.f90: Likewise.
* gfortran.dg/interface_47.f90: New test.
