marxin [Thu, 2 May 2019 08:43:25 +0000 (08:43 +0000)]
Enhance target and target_clone error messages.
2019-05-02 Martin Liska <mliska@suse.cz>
* cgraphclones.c: Call valid_attribute_p with 1 for
target_clone.
* config/i386/i386-c.c (ix86_pragma_target_parse): Use 0 as
it's for target attribute.
* config/i386/i386-protos.h (ix86_valid_target_attribute_tree):
Add new boolean argument.
* config/i386/i386.c (ix86_valid_target_attribute_inner_p):
Likewise.
(ix86_valid_target_attribute_tree): Pass target_clone_attr
to ix86_valid_target_attribute_inner_p.
(ix86_valid_target_attribute_p): Pass flags argument to
ix86_valid_target_attribute_inner_p.
(get_builtin_code_for_version): Use 0 as it's target attribute.
2019-05-02 Martin Liska <mliska@suse.cz>
marxin [Thu, 2 May 2019 08:16:12 +0000 (08:16 +0000)]
Postpone print of --help=* option.
2019-05-02 Martin Liska <mliska@suse.cz>
* gcc.c (process_command): Add dummy file only
if n_infiles == 0.
* opts-global.c (decode_options): Pass lang_mask.
* opts.c (print_help): New function.
(finish_options): Print --help if help_option_argument
is set.
(common_handle_option): Factor out content of OPT__help_
into print_help.
* opts.h (finish_options): Add new argument.
marxin [Thu, 2 May 2019 07:57:38 +0000 (07:57 +0000)]
Prefer to use strlen call instead of inline expansion (PR target/88809).
2019-05-02 Martin Liska <mliska@suse.cz>
PR target/88809
* config/i386/i386.c (ix86_expand_strlen): Use strlen call.
With -minline-all-stringops use inline expansion using 4B loop.
* doc/invoke.texi: Document the change of
-minline-all-stringops.
2019-05-02 Martin Liska <mliska@suse.cz>
PR target/88809
* gcc.target/i386/pr88809.c: New test.
* gcc.target/i386/pr88809-2.c: New test.
ian [Wed, 1 May 2019 21:34:16 +0000 (21:34 +0000)]
compiler,runtime: do more direct interfaces
A direct interface is an interface whose data word contains the
actual data value, instead of a pointer to it. The gc toolchain
creates a direct interface if the value is pointer shaped, that
includes pointers (including unsafe.Pointer), functions, channels,
maps, and structs and arrays containing a single pointer-shaped
field. In gccgo, we only do this for pointers. This CL unifies
direct interface types with gc. This reduces allocations when
converting such types to interfaces.
Our method functions used to always take pointer receivers, to
make interface calls easy. Now for direct interface types, their
value methods will take value receivers. For a pointer to those
types, when converted to interface, the interface data contains
the pointer. For that interface to call a value method, it will
need a wrapper method that dereference the pointer and invokes
the value method. The wrapper method, instead of the actual one,
is put into the itable of the pointer type.
In the runtime, adjust funcPC for the new layout of interfaces of
functions.
ian [Wed, 1 May 2019 20:27:36 +0000 (20:27 +0000)]
runtime: persistentalloc and cache itabs
Previously, each time we do an interface conversion for which the
method table is not known at compile time, we allocate a new
method table.
This CL ports the mechanism of itab caching from the gc runtime,
adapted to our itab representation and method finding mechanism.
With the cache, we reuse the same itab for the same (interface,
concrete) type pair. This reduces allocations in interface
conversions.
Unlike the gc runtime, we don't prepopulate the cache with
statically allocated itabs, as currently we don't have a way to
find them. This means we don't deduplicate run-time allocated
itabs with compile-time allocated ones. But that is not too bad
-- it is just a cache anyway.
As now itabs are never freed, it is also possible to drop the
write barrier for writing the first word of an interface header.
I'll leave this optimization for the future.
PR fortran/60144
* match.c (gfc_match_parens): Change the location for missing ')'.
(gfc_match_if): Detect a missing '('. Remove the spurious named
constant error. Change the wording of some errors.
(gfc_match_else): Change the wording of an error.
(gfc_match_elseif): Detect a missing '('. Improve the matching
process to get a better syntax analysis.
redi [Wed, 1 May 2019 00:08:36 +0000 (00:08 +0000)]
PR libstdc++/61761 fix std::proj for targets without C99 cproj
The current generic implementation of __complex_proj used when cproj is
not available calculates the wrong projection, giving a different result
than given by C99's cproj.
When C99 cproj is not available but isinf and copysign are, use those to
give correct results for float, double and long double. Otherwise, and
for other specializations of std::complex, just use a generic version
that returns its argument, and so doesn't support infinities.
We might want to consider adding additional overloads of __complex_proj
to support extended types such as _Float64x, _Float128 etc.
PR libstdc++/61761
* include/std/complex (__complex_proj): Return parameter unchanged.
[_GLIBCXX_USE_C99_COMPLEX] (__complex_proj): Change overloads for
floating-point types to take std::complex arguments.
[_GLIBCXX_USE_C99_MATH_TR1] (__complex_proj): Add overloads for
floating-point types.
* testsuite/26_numerics/complex/proj.cc: New test.
This keeps coming up repeatedly and the ACLE has finally added
__ARM_FEATURE_ATOMICS for the LSE feature in GCC. This is now part of
the latest ACLE release
(https://developer.arm.com/docs/101028/latest/5-feature-test-macros)
I know it's late for GCC-9 but this is a simple macro which need not
wait for another year.
Ok for trunk and to backport to all release branches ?
The root_path.cc test had some debugging macros left in accidentally, so
didn't FAIL correctly if an assertion failed.
The string-char8_t.cc tests didn't compile on Windows.
* testsuite/27_io/filesystem/path/decompose/root_path.cc: Remove
macros accidentally left in.
* testsuite/27_io/filesystem/path/native/string-char8_t.cc: Remove
unnecessary -lstdc++fs option. Fix test for mingw.
* testsuite/experimental/filesystem/path/native/string-char8_t.cc:
Fix test for mingw.
* gimple-fold.c (gimple_fold_builtin_memory_op): Change endp
into built_in_function enum. Remove code for endp == 2 and
use BUILT_IN_* constants.
(gimple_fold_builtin): Call the function with fcode.
If a noexcept function calls a function that might throw, doing the tail
call optimization means that an exception thrown in the called function
will propagate out, breaking the noexcept specification. So we need to
prevent the optimization in that case.
* tree-tailcall.c (find_tail_calls): Don't turn a call from a
nothrow function to a might-throw function into a tail call.
...since it was only ever zero in practice. I wondered about making
it a symbolic constant instead, but it didn't seem worth it when
there was only one user (and that user was part of the internal
tree-data-ref.c implementation).
2019-04-29 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* tree-data-ref.h (data_dependence_relation::inner_loop): Delete.
(DDR_INNER_LOOP): Likewise.
* tree-data-ref.c (dump_data_dependence_relation): Update accordingly.
(initialize_data_dependence_relation): Likewise.
(insert_innermost_unit_dist_vector): Use 0 instead of DDR_INNER_LOOP.
PR libstdc++/87982 Fix generate_n and fill_n use of _Size parameter
The standard only requires that _Size can be converted to an integral
type, not that it can be used for arithmetic. Add a new set of
__size_to_integer helper functions to do the conversion (which will be
ambiguous if there is no one conversion that is better than any others).
Also add tests for DR 426 which requires these algorithms and search_n
to handle negative values of n.
PR libstdc++/87982
* include/bits/stl_algo.h (generate_n): Convert _Size parameter to
an integral type.
* include/bits/stl_algobase.h (__size_to_integer): New overloaded
functions to convert a value to an integral type.
(__fill_n_a, __fill_n_a): Assert that __n is already an integral type.
(fill_n): Convert _Size parameter to an integral type.
* testsuite/25_algorithms/fill_n/dr426.cc: New test.
* testsuite/25_algorithms/generate_n/87982.cc: New test.
* testsuite/25_algorithms/generate_n/dr426.cc: New test.
* libdruntime/gcc/sections/elf_shared.d (IBMZ_Any): Define when
version S390 or SystemZ. Use condition instead of SystemZ.
(getTLSRange): Return null on GNU_EMUTLS targets.
PR target/89261
* config/i386/i386-protos.h (ix86_data_alignment): Change
the second argument type to unsigned int.
* config/i386/i386.c (ix86_data_alignment): Change "align"
argument type to unsigned int.
testsuite/ChangeLog:
PR target/89261
* gcc.target/i386/pr89261.c: New test.
wilson [Sat, 27 Apr 2019 00:46:20 +0000 (00:46 +0000)]
RISC-V: Promode modes of constant loads for store insns.
This improves optimization of code storing constants to memory. Given this
testcase:
void sub1 (int *a, long long *b) { *a = 1; *b = 1; }
an unpatched rv64 compiler emits two li instructions, one for an SImode pseudo
and one for a DImode pseudo. With the patch, we get a single DImode li insn.
gcc/
* config/riscv/riscv-protos.h (riscv_move_integer): Add machine_mode
parameter.
* config/riscv/riscv.c (riscv_move_integer): New parameter orig_mode.
Pass orig_mode to riscv_build_integer.
(riscv_split_integer): Pass mode to riscv_move_integer.
(riscv_legitimize_const_move): Likewise.
(riscv_legitimize_move): For MEM dest and CONST_INT src case, new local
promoted_mode. Replace force_reg call with code to load constant into
promoted reg and then subreg it for the store.
* config/riscv/riscv.md (low<mode>+1): Pass <GPR:MODE>mode to
riscv_move_integer.
ian [Fri, 26 Apr 2019 17:20:55 +0000 (17:20 +0000)]
runtime: fix TestPhysPageSize on AIX
AIX doesn't allow to mmap an address range which is already mmap.
Therefore, once the region has been allocated, it must munmap before
being able to play with it.
The corresponding Go Toolchain patch is CL 174059.
Remove redundant step in experimental::filesystem::path construction
* include/experimental/bits/fs_path.h
(path::_S_convert_loc<_InputIterator>): Create const std::string to
avoid redundant call to _S_convert_loc with non-const pointers.
The pmr::string typedef isn't available for the old COW string, so some
tests that use it were failing.
* testsuite/20_util/variant/run.cc: Use a new Hashable type to test
hashing, because pmr::string depends on _GLIBCXX_USE_CXX11_ABI==1.
* testsuite/21_strings/basic_string/hash/hash.cc
[!_GLIBCXX_USE_CXX11_ABI]: Don't test pmr strings.
* testsuite/21_strings/basic_string/hash/hash_char8_t.cc
[!_GLIBCXX_USE_CXX11_ABI]: Likewise.
Reduce code instantiated by filesystem::path::_S_convert_loc
Jakub noted in https://gcc.gnu.org/ml/libstdc++/2019-04/msg00140.html
that an unwanted std::wstring::_M_replace_dispatch symbol has started to
be exported from the Fedora shared library. This symbol is triggered by
the instantiation of std::wstring::assign(const char*, const char*) from
std::__str_codecvt_in which is called from path::_S_convert_loc. The
branch that triggers that instantiation can't actually happen in that
case, because codecvt facets will only return noconv when the input and
output types are the same. Guarding the assign call with an if-constexpr
check that the types are the same avoids instantiating template
specializations that will never actually be needed.
* config/abi/pre/gnu.ver (GLIBCXX_3.4): Replace wildcard that matches
wstring::_M_replace_dispatch with more specific patterns.
* include/bits/fs_path.h (path::_S_convert_loc<_InputIterator>):
Create const std::string to avoid redundant call to _S_convert_loc
with non-const pointers.
* include/bits/locale_conv.h (__do_str_codecvt): Use if-constexpr to
avoid unnecessary basic_string::assign instantiations.
The 20_util/scoped_allocator/69293_neg.cc test was failing in C++2a mode
because the expected static_assert text wasn't matched. The code is
still ill-formed in C++2a, but uses the new __uses_alloc_args function
and so fails a different static_assert. This patch adds the same string
to the new static_assert, so the test passes.
Now that G++ allows concepts to be declared without 'bool' we can use
the correct C++2a syntax for the _Std_pair concept used to constrain the
uses-allocator construction utilities.
Also add a new test to verify that pmr::polymorphic_allocator correctly
performs recursive uses-allocator construction for nested pairs in
C++2a.
* include/std/memory (__uses_alloc_args): Add string-literal to
static_assert, to match the one in __uses_alloc.
[__cpp_concepts] (_Std_pair): Use C++2a syntax for concept.
* testsuite/20_util/polymorphic_allocator/construct_c++2a.cc: Check
for recursive uses-allocator construction of nested pairs.
* testsuite/20_util/scoped_allocator/construct_pair_c++2a.cc:: Add
comment.
Fix use of COMPLETE_TYPE_P for -Wstrict-aliasing=1
The handling of -Wstrict-aliasing=1 applied COMPLETE_TYPE_P to the
pointer type rather than the pointer target, so missed the warnings
for "struct incomplete" in the testcase.
I couldn't find any existing C tests for -Wstrict-aliasing=1,
so I added a few extra tests besides the ones fixed by the patch.
I'm sure there's lots more we could test -- this is just supposed
to be better than the status quo (i.e. nothing).
2019-04-26 Richard Sandiford <richard.sandiford@arm.com>
gcc/c-family/
* c-warn.c (strict_aliasing_warning): Apply COMPLETE_TYPE_P to
the pointer target rather than the pointer itself.
PR libstdc++/90239 use uses_allocator_construction_args in <scoped_allocator>
PR libstdc++/90239
* doc/xml/manual/status_cxx2020.xml: Amend P0591R4 status.
* include/std/scoped_allocator [__cplusplus > 201703L]
(scoped_allocator_adaptor::construct): Define in terms of
uses_allocator_construction_args, as per P0591R4.
* testsuite/20_util/scoped_allocator/construct_pair_c++2a.cc: New test.
* testsuite/util/testsuite_allocator.h: Remove name of unused
parameter.
* config/i386/freebsd64.h: Add bits for 32-bit multilib support.
* config/i386/t-freebsd64: New file.
* config.gcc: Add the t-freebsd64 for multilib support.
x86: Update message for target_clones and unsupported ISAs
Before AVX512F, processors with the newer ISAs also support the older
ISAs, i.e., AVX2 processors also support AVX and SSE4, SSE4 processors
also support SSSE3, ... After AVX512F, an AVX512XX processor may not
support AVX512YY. It means AVX512XX features, except for AVX512F, can't
be used to decide priority in target_clones.
This patch updates error message for ISAs with P_ZERO priority. It also
merges _feature_list into _isa_names_table and marks ISAs, which have
unknown priority, with P_ZERO so that we only need to update one place
to add a new ISA feature.
gcc/
2019-04-25 H.J. Lu <hongjiu.lu@intel.com>
PR target/89929
* config/i386/i386.c (feature_priority): Moved to file scope.
(processor_features): Likewise.
(processor_model): Likewise.
(_arch_names_table): Likewise.
(arch_names_table): Likewise.
(_feature_list): Removed.
(feature_list): Likewise.
(_isa_names_table): Moved to file scope. Add priority.
(isa_names_table): Likewise.
(get_builtin_code_for_version): Replace feature_list with
isa_names_table. Update error message for P_ZERO priority.
gcc/testsuite/
2019-04-25 Martin Liska <mliska@suse.cz>
H.J. Lu <hongjiu.lu@intel.com>
projects / thirdparty / gcc.git / log
