virtual_operand_p handled SSA names by looking at the flags of the
underlying variable. This seems to be a relatively common source
of cache misses, mainly because virtual_operand_p is the first thing
tested by is_gimple_reg.
This patch caches the information in the SSA name itself. Several
flags seem to be free so the patch arbitrarily uses public_flag.
Tested on aarch64-linux-gnu and x86_64-linux-gnu. It improves
compile time by a small (<1%) but reproducable amount on the
tests I've tried.
gcc/
* tree-core.h (tree_base): Document the meaning of public_flag
for SSA names.
* tree.h (SSA_NAME_IS_VIRTUAL_OPERAND): New macro.
(SET_SSA_NAME_VAR_OR_IDENTIFIER): Record whether the variable
is a virtual operand.
* gimple-expr.h (virtual_operand_p): Use SSA_NAME_IS_VIRTUAL_OPERAND.
From-SVN: r243916
PR middle-end/78901
* except.c (nothrow_libfn_p): Expect libc_name_p to return
const struct libc_name_struct *, if it returns NULL, return 0,
otherwise check c_ver and use flag_isoc99 or flag_isoc11.
* cfns.gperf: Add %struct-type and libc_name_struct definition.
For all C89 C library functions add , 89 after the name, add
C99 C library functions with , 99 and C11 C library functions
with , 11 suffix.
* cfns.h: Regenerated.
From-SVN: r243914
gcc/testsuite/ChangeLog:
2016-12-23 Andre Vehreschild <vehre@gcc.gnu.org>
* gfortran.dg/class_assign_1.f08: New test.
gcc/fortran/ChangeLog:
2016-12-23 Andre Vehreschild <vehre@gcc.gnu.org>
* trans-expr.c (trans_class_assignment): Allocate memory of _vptr->size
before assigning an allocatable class object.
(gfc_trans_assignment_1): Flag that (re-)alloc of the class object
shall be done.
From-SVN: r243909
gcc/testsuite
* gcc.target/mips/oddspreg-2.c (dg-options): Remove dg-skip-if for
-fno-fat-lto-objects and add the -ffat-lto-objects option, along with
an explanation for its purpose.
* gcc.target/mips/oddspreg-3.c (dg-options): Likewise.
* gcc.target/mips/oddspreg-6.c (dg-options): Likewise.
* gcc.target/mips/no-dsp-1.c: Add an explanation for the purpose of
-ffat-lto-objects.
* gcc.target/mips/pr54240.c: Likewise.
* gcc.target/mips/r10k-cache-barrier-14.c: Likewise.
* gcc.target/mips/soft-float-1.c: Likewise.
From-SVN: r243908
Change Struct_type::do_mangled_name to incorporate the field
names even for hidden symbols. This is needed in cases where
a package imports a type "S" that has an anonymous struct, e.g.
// imported from some other package
type S struct {
X struct{ _ struct{} }
}
and then defines a local type that uses a structurally identical
anonymous struct, e.g.
// defined locally
type T struct {
U struct{ _ struct{} }
}
In the case above both types triggered the creation of hash/equal
methods, but the method names were clashing (since both structs
had the same mangled name).
Fixesgolang/go#18414
Reviewed-on: https://go-review.googlesource.com/34621
From-SVN: r243899
2016-12-22 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
* tree-pretty-print.c (dump_generic_node): Change dump format for
REALPART_EXPR and IMAGPART_EXPR with TDF_GIMPLE.
testsuite/
* gcc.dg/gimplefe-20.c: New test-case.
From-SVN: r243898
When pushing a value into the literal pool the resulting decl might
get a higher alignment than the original expression depending on how a
target defines CONSTANT_ALIGNMENT. Generating an RTX for the constant
pool access we currently use the alignment from the original
expression. Changed with the attached patch.
This fixes a GCC 6 regression for S/390. For arrays of string
constants as in the attached testcase encode_section_info is not able
to figure out that the constant pool slot is already properly aligned
since the mem_align field in the rtx is not set properly.
gcc/ChangeLog:
2016-12-22 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* varasm.c (build_constant_desc): Use the alignment of the var
decl instead of the original expression.
gcc/testsuite/ChangeLog:
2016-12-22 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* gcc.target/s390/litpool-str-1.c: New test.
From-SVN: r243888
PR tree-optimization/78886
* gcc.dg/tree-ssa/pr78886.c: New test.
PR tree-optimization/78886
* tree-ssa-strlen.c (handle_builtin_malloc): Return when LHS
is equal to NULL.
From-SVN: r243886
PR c++/72707
* name-lookup.c (pushdecl_maybe_friend_1): Do check shadowing of
artificial x if it is an anonymous union variable.
* g++.dg/warn/Wshadow-12.C: New test.
From-SVN: r243877
PR bootstrap/78817
* tree-pass.h (make_pass_post_ipa_warn): Declare.
* builtins.c (validate_arglist): Adjust get_nonnull_args call.
Check for NULL pointer argument to nonnull arg here.
(validate_arg): Revert 2016-12-14 changes.
* calls.h (get_nonnull_args): Remove declaration.
* tree-ssa-ccp.c: Include diagnostic-core.h.
(pass_data_post_ipa_warn): New variable.
(pass_post_ipa_warn): New class.
(pass_post_ipa_warn::execute): New method.
(make_pass_post_ipa_warn): New function.
* tree.h (get_nonnull_args): Declare.
* tree.c (get_nonnull_args): New function.
* calls.c (maybe_warn_null_arg): Removed.
(maybe_warn_null_arg): Removed.
(initialize_argument_information): Revert 2016-12-14 changes.
* passes.def: Add pass_post_ipa_warn after first ccp after IPA.
c-family/
* c-common.c (struct nonnull_arg_ctx): New type.
(check_function_nonnull): Return bool instead of void. Use
nonnull_arg_ctx as context rather than just location_t.
(check_nonnull_arg): Adjust for the new context type, set
warned_p to true if a warning has been diagnosed.
(check_function_arguments): Return bool instead of void.
* c-common.h (check_function_arguments): Adjust prototype.
c/
* c-typeck.c (build_function_call_vec): If check_function_arguments
returns true, set TREE_NO_WARNING on CALL_EXPR.
cp/
* typeck.c (cp_build_function_call_vec): If check_function_arguments
returns true, set TREE_NO_WARNING on CALL_EXPR.
* call.c (build_over_call): Likewise.
From-SVN: r243874
PR c++/77830
* constexpr.c (cxx_eval_array_reference): Perform out of bounds
verification even if lval is true, just allow one past the last
element in that case.
(cxx_eval_store_expression): Detect stores to out of bound
ARRAY_REF.
* g++.dg/cpp1y/pr77830.C: New test.
* g++.dg/cpp0x/pr65398.C: Adjust expected diagnostics.
From-SVN: r243873
gcc/c-family/
* c.opt (-fnew-ttp-matching): New flag.
* c-opts.c (c_common_post_options): Default on if -std=c++1z.
gcc/cp/
* pt.c (coerce_template_template_parms): Allow a template argument
that's less specialized than the parameter.
(unify_bound_ttp_args): Adjust parm's args to apply to arg's
template.
(coerce_template_args_for_ttp): Split out from
lookup_template_class_1.
(coerce_ttp_args_for_tta, store_defaulted_ttp)
(lookup_defaulted_ttp, add_defaults_to_ttp): New.
(process_partial_specialization): Set DECL_CONTEXT of
template template-parameters.
(coerce_template_parms): Only inform when complain.
(expand_template_argument_pack): Handle error_mark_node.
(convert_template_argument, template_args_equal, unify): Handle
any_targ_node.
* cp-tree.h (enum cp_tree_index): Add CPTI_ANY_TARG.
(any_targ_node): New.
* decl.c (cxx_init_decl_processing): Set it.
* name-lookup.c (consider_binding_level): Ignore names with embedded
spaces.
From-SVN: r243871
gcc/cp/
* pt.c (coerce_template_parms): Consider variadic_args_p before
complaining about too many template arguments.
libstdc++-v3/
* testsuite/util/testsuite_tr1.h (test_property): Don't define both
variadic and non-variadic overloads.
From-SVN: r243869
* pt.c (process_partial_specialization): Use
get_partial_spec_bindings to check that the partial specialization
is more specialized than the primary template.
From-SVN: r243868
* pt.c (convert_template_argument): Pass args to do_auto_deduction.
(mark_template_parm): Handle deducibility from type of non-type
argument here.
(for_each_template_parm_r): Not here.
From-SVN: r243867
All feedback from the earlier version has been taken into account now.
This patch adds the -fasynchronous-unwind-tables option to compilations when
the -fsanitize=address option is seen but not if any
-fasynchronous-unwind-tables options were already specified.
-fasynchronous-unwind-tables causes a full strack trace to be produced when
the sanitizer detects an error. Without the full trace several of the asan
test cases fail on powerpc.
See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65479 for more information.
Bootstrapped and tested on powerpc64le-unknown-linux-gnu,
powerpc64be-unknown-linux-gnu, and x86_64-pc-linux-gnu with no regressions.
Is this ok for trunk?
[gcc]
2016-12-21 Bill Seurer <seurer@linux.vnet.ibm.com>
PR sanitizer/65479
* config/rs6000/rs6000.c (rs6000_option_override_internal): Add
-fasynchronous-unwind-tables option when -fsanitize=address is
specified.
From-SVN: r243863
PR driver/78863
* gcc.c (driver::build_option_suggestions): Do not add
-fsanitize=all as a suggestion candidate.
PR driver/78863
* gcc.dg/spellcheck-options-13.c: New test.
From-SVN: r243857
Alexander Ivchenko
Andreas Schwab
Eric Botcazou
Sandra Loosemore
GCC Administrator
Richard Sandiford
François Dumont
Jakub Jelinek
Andre Vehreschild
Toma Tabacu
Joseph Myers
Ian Lance Taylor
Prathamesh Kulkarni
Jason Merrill
Georg-Johann Lay
Andreas Krebbel
Dominik Vogt
Martin Liska
Vladimir Makarov
Pat Haugen
Bill Seurer
Bernd Schmidt
David Edelsohn
Andre Vieira
Martin Liska
Alexander Monakov
Alexander Monakov