2020-06-03 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* exp_util.adb (Remove_Side_Effects): For a type conversion, do
not remove the side effects of the expression only if it is of
universal integer type.
2020-06-03 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* exp_ch4.adb (Expand_N_Unchecked_Type_Conversion): Also fold the
conversion for literals of enumeration types.
2020-06-03 Yannick Moy <moy@adacore.com>
gcc/ada/
* rtsfind.adb (Load_RTU): Correctly set/reset global variable to
ignore SPARK_Mode in instances around loading.
* sem_ch6.adb (Analyze_Subprogram_Body_Helper): Accept Off
without prior On.
* sem_ch7.adb (Analyze_Package_Body_Helper): Likewise.
* sem_prag.adb (Analyze_Pragma): Always take into account
SPARK_Mode Off.
2020-06-03 Justin Squirek <squirek@adacore.com>
gcc/ada/
* libgnat/a-cborse.adb, libgnat/a-cihase.adb,
libgnat/a-ciorse.adb, libgnat/a-coorse.adb: Modified to use
'Unrestricted_Access in certain cases where static accessibility
errors were triggered.
* exp_ch6.adb (Expand_Simple_Return_Statement): Add generation
of dynamic accessibility checks as determined by
Is_Special_Aliased_Formal_Access.
* sem_attr.adb (Resolve_Attribute): Add call to
Is_Special_Aliased_Formal_Access to avoid performing static
checks where dynamic ones are required.
* sem_ch6.adb (Check_Return_Obj_Accessibility): Handle renamed
objects within component associations requiring special
accessibility checks.
* sem_util.adb, sem_util.ads (Is_Special_Aliased_Formal_Access):
Created to detect the special case where an aliased formal is
being compared against the level of an anonymous access return
object.
(Object_Access_Level): Remove incorrect condition leading to
overly permissive accessibility levels being returned on
explicitly aliased parameters.
2020-06-03 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* einfo.ads (Enum_Pos_To_Rep): Adjust description.
* exp_attr.adb (Expand_N_Attribute_Reference) <Pred>:
Reimplement in the case of an enumeration type with non-standard
but contiguous representation.
<Succ>: Likewise.
<Val>: Likewise.
* exp_ch3.adb (Expand_Freeze_Enumeration_Type): Count the
literals in the first loop. If the representation is
contiguous, just build the index type of the array type and set
Enum_Pos_To_Rep to it.
2020-06-03 Gary Dismukes <dismukes@adacore.com>
gcc/ada/
* exp_attr.adb (Expand_Loop_Entry_Attribute): Revise loop that
resets the scopes of entities associated with Loop_Id to the
scope of the new function, so the resetting is not restricted to
itypes, but excludes loop parameters and the function entity
itself. However, this fix is believed to be incomplete and a ???
comment is added to indicate that.
2020-06-03 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* exp_attr.adb (Expand_N_Attribute_Reference) <Enum_Rep>:
In the case of an enumeration type, do an intermediate
conversion to a small integer type. Remove useless stuff.
<Finalization_Size>: Do not hardcode Universal_Integer and
fix a type mismatch in the assignment to the variable.
<Max_Size_In_Storage_Elements>: Likewise.
<From_Any>: Do not redefine the Ptyp local variable.
<To_Any>: Likewise.
<TypeCode>: Likewise.
<Pos>: Small tweaks.
<Val>: For an enumeration type with standard representation,
apply the range check to the expression of a convertion to
Universal_Integer, if any. For an integer type, expand to
a mere conversion.
2020-06-03 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* libgnat/s-rannum.adb (Random_Discrete): In the 32-bit case,
use the same linear implementation as in the 64-bit case when
the type has a contiguous representation.
2020-06-03 Gary Dismukes <dismukes@adacore.com>
gcc/ada/
* sem_util.adb (Build_Class_Wide_Clone_Decl): Call
Set_Debug_Info_Needed to set the Needs_Debug_Info flag on
Clone_Id if the flag is set on Spec_Id.
2020-06-03 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* exp_attr.adb (Attribute_Valid): Use Standard_Long_Long_Integer
in lieu of Universal_Integer as large integer type.
* exp_ch3.adb (Expand_Freeze_Enumeration_Type): Likewise.
2020-06-03 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* checks.adb (Is_Single_Attribute_Reference): New predicate.
(Generate_Range_Check): Do not force the evaluation if the
node is a single attribute reference.
* exp_util.adb (Side_Effect_Free_Attribute): New predicate.
(Side_Effect_Free) <N_Attribute_Reference>: Call it.
(Remove_Side_Effects): Remove the side effects of the prefix
for an attribute reference whose prefix is not a name.
We must guard used atomic builtins with GCOV_SUPPORTS_ATOMIC.
The patch is tested on AIX and I'm going to push it.
libgcc/ChangeLog:
PR gcov-profile/95480
* libgcov-profiler.c (GCOV_SUPPORTS_ATOMIC): Move to...
* libgcov.h (GCOV_SUPPORTS_ATOMIC): ...here.
(gcov_counter_add): Use GCOV_SUPPORTS_ATOMIC guard.
(gcov_counter_set_if_null): Likewise.
PR jit/95426 reports a crash deep inside "expand" when using
__builtin_unreachable via gcc_jit_context_get_builtin_function,
due to BLOCK_FOR_INSN being erroneously used on a barrier within
rtl_verify_bb_pointers.
The root cause turns out to be that I didn't implement
LANG_HOOKS_COMMON_ATTRIBUTE_TABLE and LANG_HOOKS_FORMAT_ATTRIBUTE_TABLE
for the jit "frontend". When building a decl for the builtin, the
libgccjit frontend generates a chain of attributes names, but when
this is passed to decl_attributes and the attributes are looked up by
namespace and name within lookup_scoped_attribute_spec, attributes_table
is empty. Hence no attributes were being associated with the fndecl,
and so ECF_NORETURN was not set on the gimple_call (along with various
other flags missing on the decl, etc), and so the call is treated as
not terminating its BB, and so the CFG rapidly diverges from the
equivalent created by the C frontend.
This patch fixes things by implementing these langhooks, copying the
minimal attribute-handling code from LTO. I stepped through the
creation of the fndecl and verified that with this fix it has the same
attributes as the equivalent created by the C frontend.
gcc/jit/ChangeLog:
PR jit/95426
* dummy-frontend.c: Include "options.h", "stringpool.h", and
"attribs.h".
(ATTR_EXCL): New, copied from lto/lto-lang.c.
(attr_noreturn_exclusions): Likewise.
(attr_returns_twice_exclusions): Likewise.
(attr_const_pure_exclusions): Likewise.
(jit_attribute_table): Likewise, copied from lto_attribute_table.
(jit_format_attribute_table): Likewise, copied from
lto_format_attribute_table.
(handle_noreturn_attribute): New, copied from lto/lto-lang.c.
(handle_leaf_attribute): Likewise.
(handle_const_attribute): Likewise.
(handle_malloc_attribute): Likewise.
(handle_pure_attribute): Likewise.
(handle_novops_attribute): Likewise.
(get_nonnull_operand): Likewise.
(handle_nonnull_attribute): Likewise.
(handle_nothrow_attribute): Likewise.
(handle_sentinel_attribute): Likewise.
(handle_type_generic_attribute): Likewise.
(handle_transaction_pure_attribute): Likewise.
(handle_returns_twice_attribute): Likewise.
(handle_patchable_function_entry_attribute): Likewise.
(ignore_attribute): Likewise.
(handle_format_attribute): Likewise.
(handle_format_arg_attribute): Likewise.
(handle_fnspec_attribute): Likewise.
(LANG_HOOKS_COMMON_ATTRIBUTE_TABLE): Define.
(LANG_HOOKS_FORMAT_ATTRIBUTE_TABLE): Define.
gcc/testsuite/ChangeLog:
PR jit/95426
* jit.dg/all-non-failing-tests.h: Add note about...
* jit.dg/test-builtin-unreachable.c: New test.
Here, the capture proxy for *this is const, but its DECL_VALUE_EXPR is not.
Don't ICE on this; it's a reasonable difference, since in C++ an rvalue of
scalar type does not have cv-qualifiers.
gcc/cp/ChangeLog:
PR c++/95193
* pt.c (tsubst_decl): Relax assert.
gcc/testsuite/ChangeLog:
PR c++/95193
* g++.dg/cpp1z/lambda-this7.C: New test.
The ISA manual specifies that divide by zero always returns -1 as the result.
We were failing to do that when the dividend was negative.
Original patch from Virginie Moser.
libgcc/
* config/riscv/div.S (__divdi3): For negative arguments, change bgez
to bgtz.
Remove occurrences of auxbase that remained in comments.
for gcc/ada/ChangeLog
* lib.ads (Compilation_Switches): Remove -auxbase from
comments.
* switch.ads (Is_Internal_GCC_Switch): Likewise.
Since the co_await expression is mostly opaque to the existing
machinery, we were hiding the details of the await_resume return
value. If that needs to be wrapped in a target expression, then
emulate this with the whole co_await. Similarly, if the await
expression we build in response to co_await p.yield_value (e)
is wrapped in a target expression, then we need to transfer that
wrapper to the resultant CO_YIELD_EXPR (which is, itself, just
a proxy for the underlying co_await).
gcc/cp/ChangeLog:
PR c++/95050
* coroutines.cc (build_co_await): Wrap the co_await expression
in a TARGET_EXPR, where needed.
(finish_co_yield_expr): Likewise.
gcc/testsuite/ChangeLog:
PR c++/95050
* g++.dg/coroutines/pr95050.C: New test.
In my recent changes to outputs.exp and gcc-defs.exp, I misreferenced
dejagnu board property ldscript, singular, as ldscripts, plural.
This probably didn't have much impact on gcc-defs.exp: the code there
would just prefix with -Wl, any options that amounted to file names,
and linker scripts probably wouldn't be named without a -T or even an
-Wl,-T, prefix.
The visible effects were in outputs.exp, that also intended to add the
ldscript, if present, to the set of options to be passed to the
compiler driver for linking. Using the wrong option name, that
wouldn't work.
for gcc/testsuite/ChangeLog
* gcc.misc-tests/outputs.exp: Spell ldscript correctly.
* lib/gcc-defs.exp (gcc_adjust_linker_flags): Likewise.
In aarch64_short_vector_p, we are simply checking whether a type (and a mode)
is a 64/128-bit short vector or not. This should not be affected by the value
of TARGET_SVE. Simply leave later code to report an error if SVE is disabled.
2020-06-02 Felix Yang <felix.yang@huawei.com>
gcc/
PR target/95459
* config/aarch64/aarch64.c (aarch64_short_vector_p):
Leave later code to report an error if SVE is disabled.
gcc/testsuite/
PR target/95459
* gcc.target/aarch64/mgeneral-regs_6.c: New test.
When given a type which can convert to any container-like type, the
C(const C&) copy constructor and C(const C::_Base&) converting
constructor are ambiguous. This change replaces the converting
constructor's parameter with a reference_wrapper-like type so that
calling that constructor requires an additional user-defined conversion.
This gives it a lower rank than the copy constructor, avoiding the
ambiguity.
While testing this change I discovered that __gnu_debug::forward_list
doesn't have a convering constructor from the std::forward_list base, so
this adds it.
We should probably consider whether the converting constructors should
be 'explicit' but I'm not changing that now.
libstdc++-v3/ChangeLog:
PR libstdc++/90102
* include/debug/deque (deque(const _Base&)): Replace parameter
with a struct that wraps a const _Base&.
* include/debug/forward_list (forward_list(_Base_ref)): New
constructor.
* include/debug/list (list(const _Base&)): Replace parameter
with a struct that wraps a const _Base&.
* include/debug/map.h (map(const _Base&)): Likewise.
* include/debug/multimap.h (multimap(const _Base&)): Likewise.
* include/debug/multiset.h (multiset(const _Base&)): Likewise.
* include/debug/set.h (set(const _Base&)): Likewise.
* include/debug/unordered_map (unordered_map(const _Base&))
(unordered_multimap(const _Base&)): Likewise.
* include/debug/unordered_set (unordered_set(const _Base&))
(unordered_multiset(const _Base&)): Likewise.
* testsuite/23_containers/vector/cons/destructible_debug_neg.cc:
Adjust dg-error line number.
* include/debug/vector (vector(const _Base&)): Likewise.
* testsuite/23_containers/deque/debug/90102.cc: New test.
* testsuite/23_containers/forward_list/debug/90102.cc: New test.
* testsuite/23_containers/list/debug/90102.cc: New test.
* testsuite/23_containers/map/debug/90102.cc: New test.
* testsuite/23_containers/multimap/debug/90102.cc: New test.
* testsuite/23_containers/multiset/debug/90102.cc: New test.
* testsuite/23_containers/set/debug/90102.cc: New test.
* testsuite/23_containers/unordered_map/debug/90102.cc: New test.
* testsuite/23_containers/unordered_multimap/debug/90102.cc: New test.
* testsuite/23_containers/unordered_multiset/debug/90102.cc: New test.
* testsuite/23_containers/unordered_set/debug/90102.cc: New test.
* testsuite/23_containers/vector/debug/90102.cc: New test.
This patch adds support for the Arm Zeus CPU.
Bootstrapped and tested on aarch64-none-linux-gnu.
gcc/
2020-06-02 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
* config/aarch64/aarch64-cores.def (zeus): Define.
* config/aarch64/aarch64-tune.md: Regenerate.
* doc/invoke.texi (AArch64 Options): Document zeus -mcpu option.
Because reg_to_non_prefixed() only looks at the register being used, it
doesn't get the right answer for stfs, which leads to us not seeing
that it has a PCREL symbol ref. This patch works around this by
introducing a helper function that inspects the insn to see if it is in
fact a stfs. Then if we use NON_PREFIXED_DEFAULT, address_to_insn_form()
can see that it has the PCREL symbol ref.
gcc/ChangeLog:
PR target/95347
* config/rs6000/rs6000.c (prefixed_store_p): Add special case
for stfs.
(is_stfs_insn): New helper function.
This patch removes the obsolete -mlocal-symbol-id option. This was used to
control mangling of local symbol names in order to work around a ROCm runtime
bug, but that has not been needed in some time, and the mangling was removed
already.
gcc/ChangeLog:
* config/gcn/gcn-hsa.h (CC1_SPEC): Delete.
* config/gcn/gcn.opt (-mlocal-symbol-id): Delete.
* config/gcn/mkoffload.c (main): Don't use -mlocal-symbol-id.
gcc/testsuite/ChangeLog:
* gcc.dg/intermod-1.c: Don't use -mlocal-symbol-id.
contrib/ChangeLog:
* gcc-changelog/git_commit.py: Support foo/bar/*: patterns in
wildcard_prefixes locations.
* gcc-changelog/test_email.py: Test it.
* gcc-changelog/test_patches.txt: Add 3 new patches.
The calloc was in the original tested version of the patch
and I made accidental last minute change.
Installed to master as obvious.
libgcc/ChangeLog:
* libgcov.h (gcov_topn_add_value): Use xcalloc instead
of xmalloc.
The files in this subdirectory are part of the D2 test suite maintained
in the upstream DMD code repository.
contrib/ChangeLog:
* gcc-changelog/git_commit.py (ignored_prefixes): Add
gcc/testsuite/gdc.test/.
Like a similarly named function in the visitor class for statements,
this ensures that the current input_location is set to the correct
source file location of the decl.
It is likely that there are a number of cases where declarations have
ended up with no location without this.
gcc/d/ChangeLog:
* decl.cc (DeclVisitor::build_dsymbol): New function.
(DeclVisitor::visit (TupleDeclaration *)): Use build_dsymbol to
traverse AST instead of accept.
(DeclVisitor::visit (AttribDeclaration *)): Likewise.
(DeclVisitor::visit (Nspace *)): Likewise.
(DeclVisitor::visit (TemplateDeclaration *)): Likewise.
(DeclVisitor::visit (TemplateInstance *)): Likewise.
(DeclVisitor::visit (TemplateMixin *)): Likewise.
(DeclVisitor::visit (StructDeclaration *)): Likewise.
(DeclVisitor::visit (ClassDeclaration *)): Likewise.
(DeclVisitor::visit (InterfaceDeclaration *)): Likewise.
(DeclVisitor::visit (VarDeclaration *)): Likewise.
(build_decl_tree): Likewise.
This error is wrong, the line is what exceeds LINE_LIMIT characters, the
limit doesn't exceed itself.
contrib/ChangeLog:
* gcc-changelog/git_commit.py (GitCommit.parse_changelog): Fix
grammar.
If a user installs this script as .git/hooks/prepare-commit-msg and then
works on an old branch which doesn't have the mklog.py script, trying to
commit will fail with an error like:
environment: /.../gcc/contrib/mklog.py: No such file or directory
This makes it exit cleanly so it's possible to commit.
contrib/ChangeLog:
* prepare-commit-msg: Do nothing if the mklog.py script isn't
present.
gcc/ChangeLog:
* coverage.c (get_coverage_counts): Skip sanity check for TOP N counters
as they have variable number of counters.
* gcov-dump.c (main): Add new option -r.
(print_usage): Likewise.
(tag_counters): All new raw format.
* gcov-io.h (struct gcov_kvp): New.
(GCOV_TOPN_VALUES): Remove.
(GCOV_TOPN_VALUES_COUNTERS): Likewise.
(GCOV_TOPN_MEM_COUNTERS): New.
(GCOV_TOPN_DISK_COUNTERS): Likewise.
(GCOV_TOPN_MAXIMUM_TRACKED_VALUES): Likewise.
* ipa-profile.c (ipa_profile_generate_summary): Use
GCOV_TOPN_MAXIMUM_TRACKED_VALUES.
(ipa_profile_write_edge_summary): Likewise.
(ipa_profile_read_edge_summary): Likewise.
(ipa_profile): Remove usage of GCOV_TOPN_VALUES.
* profile.c (sort_hist_values): Sort variable number
of counters.
(compute_value_histograms): Special case for TOP N counters
that have dynamic number of key-value pairs.
* value-prof.c (dump_histogram_value): Dump variable number
of key-value pairs.
(stream_in_histogram_value): Stream in variable number
of key-value pairs for TOP N counter.
(get_nth_most_common_value): Deal with variable number
of key-value pairs.
(dump_ic_profile): Use GCOV_TOPN_MAXIMUM_TRACKED_VALUES
for loop iteration.
(gimple_find_values_to_profile): Set GCOV_TOPN_MEM_COUNTERS
to n_counters.
* doc/gcov-dump.texi: Document new -r option.
libgcc/ChangeLog:
* libgcov-driver.c (prune_topn_counter): Remove.
(prune_counters): Likewise.
(merge_one_data): Special case TOP N counters
as they have variable length.
(write_top_counters): New.
(write_one_data): Special case TOP N.
(dump_one_gcov): Do not prune TOP N counters.
* libgcov-merge.c (merge_topn_values_set): Remove.
(__gcov_merge_topn): Use gcov_topn_add_value.
* libgcov-profiler.c (__gcov_topn_values_profiler_body):
Likewise here.
* libgcov.h (gcov_counter_add): New.
(gcov_counter_set_if_null): Likewise.
(gcov_topn_add_value): New.