Jakub Jelinek [Wed, 30 Jan 2013 18:05:53 +0000 (19:05 +0100)]
re PR sanitizer/55374 ([asan] -static-libasan -static-libstdc++ doesn't work)
PR sanitizer/55374
* gcc.c (LIBASAN_SPEC): Define just to ADD_STATIC_LIBASAN_LIBS if
LIBASAN_EARLY_SPEC is defined.
(LIBASAN_EARLY_SPEC): Define to empty string if not already defined.
(LINK_COMMAND_SPEC): Add LIBASAN_EARLY_SPEC for -fsanitize=address,
before %o.
* config/gnu-user.h (LIBASAN_EARLY_SPEC): Define.
* g++.dg/asan/large-func-test-1.C: Allow both _Zna[jm] in addition
to _Znw[jm] in the backtrace. Allow _Zna[jm] to be the first frame
printed in backtrace.
* g++.dg/asan/deep-stack-uaf-1.C: Use malloc instead of operator new
to avoid errors about mismatched allocation vs. deallocation.
Jakub Jelinek [Wed, 30 Jan 2013 18:04:34 +0000 (19:04 +0100)]
re PR c++/55742 (__attribute__ in class function declaration cause "prototype does not match" errors.)
PR c++/55742
* config/i386/i386.c (ix86_valid_target_attribute_inner_p): Diagnose
invalid args instead of ICEing on it.
(ix86_valid_target_attribute_tree): Return error_mark_node if
ix86_valid_target_attribute_inner_p failed.
(ix86_valid_target_attribute_p): Return false only if
ix86_valid_target_attribute_tree returned error_mark_node. Allow
target("default") attribute.
(sorted_attr_string): Change argument from const char * to tree,
merge in all target attribute arguments rather than just one.
Formatting fix. Use XNEWVEC instead of xmalloc and XDELETEVEC
instead of free. Avoid using strcat.
(ix86_mangle_function_version_assembler_name): Mangle
target("default") as if no target attribute is present. Adjust
sorted_attr_string caller. Avoid leaking memory. Use XNEWVEC
instead of xmalloc and XDELETEVEC instead of free.
(ix86_function_versions): Don't return true if one of the decls
doesn't have target attribute. If they don't and one of the decls
is DECL_FUNCTION_VERSIONED, report an error. Adjust
sorted_attr_string caller. Use XDELETEVEC instead of free.
(ix86_supports_function_versions): Remove.
(make_name): Fix up formatting.
(make_dispatcher_decl): Remove resolver_name and its initialization.
Avoid leaking memory.
(is_function_default_version): Return true if there is
target("default") attribute rather than no target attribute at all.
(make_resolver_func): Avoid leaking memory.
(ix86_generate_version_dispatcher_body): Likewise.
(TARGET_OPTION_SUPPORTS_FUNCTION_VERSIONS): Remove.
* target.def (supports_function_versions): Remove.
* doc/tm.texi.in (SUPPORTS_FUNCTION_VERSIONS): Remove.
* doc/tm.texi: Regenerated.
Richard Biener [Wed, 30 Jan 2013 15:42:49 +0000 (15:42 +0000)]
sparseset.h (sparseset_bit_p): Use gcc_checking_assert.
2013-01-30 Richard Biener <rguenther@suse.de>
* sparseset.h (sparseset_bit_p): Use gcc_checking_assert.
(sparseset_pop): Likewise.
* cfganal.c (compute_idf): Likewise. Increase work-stack size
to be able to use quick_push in the worker loop.
Greta Yorsh [Tue, 29 Jan 2013 18:39:43 +0000 (18:39 +0000)]
arm-protos.h (arm_mac_accumulator_is_result): New declaration.
2013-01-29 Greta Yorsh <Greta.Yorsh@arm.com>
* config/arm/arm-protos.h (arm_mac_accumulator_is_result): New
declaration.
* config/arm/arm.c (arm_mac_accumulator_is_result): New function.
* config/arm/cortex-a7.md: New bypasses using
arm_mac_accumulator_is_result.
Greta Yorsh [Tue, 29 Jan 2013 18:34:12 +0000 (18:34 +0000)]
arm.md (type): Add ffmas and ffmad to "type" attribute.
2013-01-29 Greta Yorsh <Greta.Yorsh@arm.com>
* config/arm/arm.md (type): Add ffmas and ffmad to "type" attribute.
* config/arm/vfp.md (fma,fmsub,fnmsub,fnmadd): Change type
from fmac to ffma.
* config/arm/vfp11.md (vfp_farith): Use ffmas.
(vfp_fmul): Use ffmad.
* config/arm/cortex-r4f.md (cortex_r4_fmacs): Use ffmas.
(cortex_r4_fmacd): Use ffmad.
* config/arm/cortex-m4-fpu.md (cortex_m4_fmacs): Use ffmas.
* config/arm/cortex-a9.md (cortex_a9_fmacs): Use ffmas.
(cortex_a9_fmacd): Use ffmad.
* config/arm/cortex-a8-neon.md (cortex_a8_vfp_macs): Use ffmas.
(cortex_a8_vfp_macd): Use ffmad.
* config/arm/cortex-a5.md (cortex_a5_fpmacs): Use ffmas.
(cortex_a5_fpmacd): Use ffmad.
* config/arm/cortex-a15-neon.md (cortex_a15_vfp_macs) Use ffmas.
(cortex_a15_vfp_macd): Use ffmad.
* config/arm/arm1020e.md (v10_fmul): Use ffmas and ffmad.
Jason Merrill [Tue, 29 Jan 2013 17:24:51 +0000 (12:24 -0500)]
re PR libstdc++/54314 (undefined references to 'construction vtable for std::ostream-in-std::basic_ostringstream<char, std::char_traits<char>, std::allocator<char> >')
PR libstdc++/54314
gcc/
* varasm.c (default_assemble_visibility): Don't warn about
visibility on artificial decls.
gcc/cp/
* class.c (build_ctor_vtbl_group): Give construction vtables
hidden visibility.
libstdc++-v3/
* config/abi/pre/gnu.ver: Don't export construction vtables.
* config/abi/pre/gnu-versioned-namespace.ver: Likewise.
Arnaud Charlet [Tue, 29 Jan 2013 14:33:47 +0000 (15:33 +0100)]
[multiple changes]
2013-01-29 Ed Schonberg <schonberg@adacore.com>
* exp_ch5.adb (Expand_Iterator_Loop_Over_Array): Preserve loop
identifier only if it comes from source.
(Expand_N_Loop_Statement): If the domain of iteration is an
enumeration type with a representation clause, remove from
visibility the loop identifier before rewriting the loop as a
block with a declaration for said identifier.
* sem_util.adb (Remove_Homonym): Handle properly the default case.
Arnaud Charlet [Tue, 29 Jan 2013 14:26:54 +0000 (15:26 +0100)]
[multiple changes]
2013-01-29 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb (Analyze_Expression_Function): An expression
function declaration is not a subprogram declaration, and thus
cannot appear in a protected definition.
* exp_util.adb (Insert_Actions): When new
actions come from the expression of the expression with actions,
then they must be added to the list of existing actions.
2013-01-29 Eric Botcazou <ebotcazou@adacore.com>
* sem_ch3.adb (Analyze_Subtype_Declaration) <Private_Kind>: For
the subtype of a constrained private type with discriminants
that has got a full view, show that the completion is a clone
of the full view.
Richard Biener [Tue, 29 Jan 2013 14:22:47 +0000 (14:22 +0000)]
re PR c/56113 (out of memory when compiling a function with many goto labels (50k > ))
2013-01-29 Richard Biener <rguenther@suse.de>
PR tree-optimization/56113
* tree-ssa-structalias.c (equiv_class_lookup): Also return
the bitmap leader.
(label_visit): Free duplicate bitmaps and record the leader instead.
(perform_var_substitution): Adjust.
Arnaud Charlet [Tue, 29 Jan 2013 14:17:44 +0000 (15:17 +0100)]
[multiple changes]
2013-01-29 Robert Dewar <dewar@adacore.com>
* a-calend-vms.adb: Minor comment fix.
2013-01-29 Robert Dewar <dewar@adacore.com>
* mlib-utl.adb, gnatlink.adb: Avoid reference to ASCII.Back_Slash
because of casing issues.
* sem_util.ads: Minor comment fix.
* style.adb (Check_Identifier): Set proper casing for entities
in ASCII.
* styleg.adb: Minor comment improvement.
* stylesw.ads (Style_Check_Standard): Fix bad comments.
* sem_prag.adb: Add the grammar for pragmas Abstract_State and Global.
(Analyze_Pragma): Push the scope of the related subprogram and install
its formals once before starting the analysis of the [moded] global
list.
2013-01-29 Pascal Obry <obry@adacore.com>
* prj-proc.adb (Process_Expression_Variable_Decl): Always handle
relative paths in Project_Path as relative to the aggregate
project location. Note that this was what was documented.
2013-01-29 Vincent Celier <celier@adacore.com>
* gnatcmd.adb: For "gnat stub -P ...", do not check the naming
scheme for Ada, when Ada is not a language for the project.
Ed Schonberg [Tue, 29 Jan 2013 14:14:48 +0000 (14:14 +0000)]
sem_ch3.adb (Analyze_Subtype_Declaration): Inherit Is_Generic_Actual_Type flag in a nested instance.
2013-01-29 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Analyze_Subtype_Declaration): Inherit
Is_Generic_Actual_Type flag in a nested instance.
* sem_ch12.adb (Restore_Private_Views): Preserve
Is_Generic_Actual_Type flag if actual is a Generic_Actual_Type
of an enclosing instance.
* sem_util.adb (Corresponding_Generic_Type): Handle generic actual
which is an actual of an enclosing instance.
* sem_type.adb (Real_Actual): If a generic_actual_type is the
formal of an enclosing generic and thus renames the corresponding
actual, use the actual of the enclosing instance to resolve
spurious ambiguities in instantiations when two formals are
instantiated with the same actual.
Arnaud Charlet [Tue, 29 Jan 2013 14:07:21 +0000 (15:07 +0100)]
[multiple changes]
2013-01-29 Robert Dewar <dewar@adacore.com>
* par-ch6.adb (No_Constraint_Maybe_Expr_Func): New procedure.
* par-util.adb (No_Constraint): Undo special handling, moved
to par-ch6.adb.
2013-01-29 Robert Dewar <dewar@adacore.com>
* aspects.ads: Aspect Warnings is implementation defined Add
some other missing entries to impl-defined list Mark Warnings
as GNAT pragma in main list.
* sem_ch8.adb: Process aspects for all cases of renaming
declarations.
2013-01-29 Robert Dewar <dewar@adacore.com>
* sem_ch6.adb (Analyze_Function_Call): Set In_Assertion flag.
* sem_elab.adb (Check_Internal_Call_Continue): Do not issue
warnings about possible elaboration error if call is within
an assertion.
* sinfo.ads, sinfo.adb (In_Assertion): New flag in N_Function_Call node.
* opt.ads (Ignore_Style_Checks_Pragmas): New flag.
* par-prag.adb (Par, case Style_Checks): Recognize
Ignore_Style_Checks_Pragmas.
* sem_prag.adb (Analyze_Pragma, case Style_Checks): Recognize
Ignore_Style_Checks_Pragmas.
* switch-c.adb: Recognize -gnateY switch.
* usage.adb: Add documentation for "-gnateY".
* vms_data.ads: Add IGNORE_STYLE_CHECKS_PRAGMAS (-gnateY).
2013-01-29 Vincent Celier <celier@adacore.com>
* clean.adb (Clean_Executables): Add Sid component when calling
Queue.Insert.
* make.adb: When inserting in the Queue, add the Source_Id
(Sid) when it is known (Start_Compile_If_Possible): When the
Source_Id is known (Sid), get the path name of the ALI file
(Full_Lib_File) from it, to avoid finding old ALI files in other
object directories.
* makeutl.ads (Source_Info): New Source_Id component Sid in
Format_Gnatmake variant.
Arnaud Charlet [Tue, 29 Jan 2013 14:01:21 +0000 (15:01 +0100)]
[multiple changes]
2013-01-29 Robert Dewar <dewar@adacore.com>
* atree.ads, atree.adb (Node30): New function.
(Set_Node30): New procedure.
(Num_Extension_Nodes): Change to 5 (activate new fields/flags).
* atree.h: Add macros for Field30 and Node30.
* einfo.ads, einfo.adb: Move some fields to avoid duplexing.
* treepr.adb (Print_Entity_Information): Print fields 30-35.
2013-01-29 Robert Dewar <dewar@adacore.com>
* sem_prag.adb (Analyze_Pragma, case Interface): Consider to
be a violation of No_Obsolescent_Features even in Ada 95. Also
generates a warning in -gnatwj mode.
(Analyze_Pragma, case Interface_Name): Generates a warning in -gnatwj
mode.
* gnat_ugn.texi: Additional documentation on -gnatwj and pragma
Interface[_Name].
2013-01-29 Vincent Celier <celier@adacore.com>
* snames.ads-tmpl: Add new standard name Trailing_Switches.
2013-01-29 Ed Schonberg <schonberg@adacore.com>
* sem_disp.adb (Check_Controlling_Type): If a designated type T
of an anonymous access type is a limited view of a tagged type,
it can be a controlling type only if the subprogram is in the
same scope as T.
2013-01-29 Vincent Celier <celier@adacore.com>
* gnatcmd.adb: Use the project where the config pragmas file is
declared to get its path.
2013-01-29 Vincent Celier <celier@adacore.com>
* prj-attr.adb: New attribute Linker'Trailing_Switches.
Jonathan Wakely [Mon, 28 Jan 2013 23:07:35 +0000 (23:07 +0000)]
re PR libstdc++/56112 (cannot create unordered_map from range of types convertible to value_type)
PR libstdc++/56112
* include/bits/hashtable_policy.h (insert(_Pair&&)): Use _M_emplace
to construct value_type explicitly before trying to extract the key.
* testsuite/23_containers/unordered_map/cons/56112.cc: New.
François Dumont [Mon, 28 Jan 2013 20:52:13 +0000 (20:52 +0000)]
hashtable_policy.h (_Local_iterator_base): Use _Hashtable_ebo_helper to embed functors into the local_iterator when necessary.
2013-01-28 François Dumont <fdumont@gcc.gnu.org>
* include/bits/hashtable_policy.h (_Local_iterator_base): Use
_Hashtable_ebo_helper to embed functors into the local_iterator
when necessary. Pass information about functors involved in hash
code by copy.
* include/bits/hashtable.h (__cache_default): Do not cache for
builtin integral types unless the hash functor is not noexcept
qualified or is not default constructible. Adapt static assertions
and local iterator instantiations.
* include/debug/unordered_set
(std::__debug::unordered_set<>::erase): Detect local iterators to
invalidate using contained node rather than generating a dummy
local_iterator instance.
(std::__debug::unordered_multiset<>::erase): Likewise.
* include/debug/unordered_map
(std::__debug::unordered_map<>::erase): Likewise.
(std::__debug::unordered_multimap<>::erase): Likewise.
* testsuite/performance/23_containers/insert_erase/41975.cc: Test
std::tr1 and std versions of unordered_set regardless of any
macro. Add test on default cache behavior.
* testsuite/performance/23_containers/insert/54075.cc: Likewise.
* testsuite/23_containers/unordered_set/instantiation_neg.cc:
Adapt line number.
* testsuite/23_containers/unordered_set/
not_default_constructible_hash_neg.cc: New.
* testsuite/23_containers/unordered_set/buckets/swap.cc: New.
Leif Ekblad [Mon, 28 Jan 2013 20:42:55 +0000 (20:42 +0000)]
config.gcc (i[34567]86-*-rdos*, [...]): New targets.
* config.gcc (i[34567]86-*-rdos*, x86_64-*-rdos*): New targets.
* config/i386/i386.h (TARGET_RDOS): New macro.
(DEFAULT_LARGE_SECTION_THRESHOLD): New macro.
* config/i386/i386.c (ix86_option_override_internal): For 64bit
TARGET_RDOS, set ix86_cmodel to CM_MEDIUM_PIC and flag_pic to 1.
* config/i386/i386.opt (mlarge-data-threshold): Initialize to
DEFAULT_LARGE_SECTION_THRESHOLD.
* config/i386/i386.md (R14_REG, R15_REG): New constants.
* config/i386/rdos.h: New file.
* config/i386/rdos64.h: New file.
Jakub Jelinek [Mon, 28 Jan 2013 16:50:22 +0000 (17:50 +0100)]
re PR rtl-optimization/56117 (ICE: in cselib_subst_to_values, at cselib.c:1853 with -O2 -fsched2-use-superblocks and __builtin_prefetch())
PR rtl-optimization/56117
* sched-deps.c (sched_analyze_2) <case PREFETCH>: For use_cselib
call cselib_lookup_from_insn on the MEM before calling
add_insn_mem_dependence.
Richard Biener [Mon, 28 Jan 2013 14:50:57 +0000 (14:50 +0000)]
tree-inline.c (remap_gimple_stmt): Do not assing a BLOCK to a stmt that didn't have one.
2013-01-28 Richard Biener <rguenther@suse.de>
* tree-inline.c (remap_gimple_stmt): Do not assing a BLOCK
to a stmt that didn't have one.
(copy_phis_for_bb): Likewise for PHI arguments.
(copy_debug_stmt): Likewise for debug stmts.
Richard Biener [Mon, 28 Jan 2013 14:45:46 +0000 (14:45 +0000)]
re PR tree-optimization/56034 (ICE: verify_gimple failed (invalid PHI argument) with -ftree-loop-distribution)
2013-01-28 Richard Biener <rguenther@suse.de>
PR tree-optimization/56034
* tree-loop-distribution.c (enum partition_kind): Add
PKIND_REDUCTION.
(partition_builtin_p): Adjust.
(generate_code_for_partition): Handle PKIND_REDUCTION. Assert
it is the last partition.
(rdg_flag_uses): Check SSA_NAME_IS_DEFAULT_DEF before looking
up the vertex for the definition.
(classify_partition): Classify whether a partition is a
PKIND_REDUCTION, thus has uses outside of the loop.
(ldist_gen): Inherit PKIND_REDUCTION when merging partitions.
Merge all PKIND_REDUCTION partitions into the last partition.
(tree_loop_distribution): Seed partitions from reductions as well.
Jakub Jelinek [Mon, 28 Jan 2013 14:43:03 +0000 (15:43 +0100)]
re PR tree-optimization/56125 (-O2 -ffast-math generates bad code when dividing a double by the square of another double.)
PR tree-optimization/56125
* tree-ssa-math-opts.c (gimple_expand_builtin_pow): Don't optimize
pow(x,c) into sqrt(x) * powi(x, n/2) or
1.0 / (sqrt(x) * powi(x, abs(n/2))) if c is an integer or when
optimizing for size.
Don't optimize pow(x,c) into powi(x, n/3) * powi(cbrt(x), n%3) or
1.0 / (powi(x, abs(n)/3) * powi(cbrt(x), abs(n)%3)) if 2c is an
integer.
Tobias Burnus [Mon, 28 Jan 2013 14:37:20 +0000 (15:37 +0100)]
re PR fortran/53537 (Explicit IMPORT of renamed USE-associated symbol fails)
2013-01-28 Tobias Burnus <burnus@net-b.de>
Mikael Morin <mikael@gcc.gnu.org>
PR fortran/53537
* symbol.c (gfc_find_sym_tree): Don't look for the symbol outside an
interface block.
(gfc_get_ha_symtree): Let gfc_find_sym_tree lookup the parent namespace.
* decl.c (gfc_match_data_decl): Ditto.
(variable_decl): Remove undeclared type error.
(gfc_match_import): Use renamed instead of original name.
2013-01-28 Tobias Burnus <burnus@net-b.de>
Mikael Morin <mikael@gcc.gnu.org>
Jakub Jelinek [Mon, 28 Jan 2013 14:28:16 +0000 (15:28 +0100)]
re PR testsuite/56053 (FAIL: c-c++-common/asan/(global|stack)-overflow-1.c)
PR testsuite/56053
* c-c++-common/asan/heap-overflow-1.c: Don't include stdlib.h and
string.h. Provide memset, malloc and free prototypes, adjust line
numbers in dg-output.
* c-c++-common/asan/stack-overflow-1.c: Don't include string.h.
Provide memset prototype and adjust line numbers in dg-output.
* c-c++-common/asan/global-overflow-1.c: Likewise.
Paul Thomas [Sun, 27 Jan 2013 07:09:06 +0000 (07:09 +0000)]
[multiple changes]
2013-01-27 Paul Thomas <pault@gcc.gnu.org>
PR fortran/55789
PR fortran/56047
* gfortran.h : Add associate_var to symbol_attr.
* resolve.c (resolve_assoc_var): Set associate_var attribute.
If the target class_ok is set, set it for the associate
variable.
* check.c (allocatable_check): Associate variables should not
have the allocatable attribute even if their symbols do.
* class.c (gfc_build_class_symbol): Symbols with associate_var
set will always have a good class container.
2013-01-27 Paul Thomas <pault@gcc.gnu.org>
PR fortran/55789
* gfortran.dg/associate_14.f90: New test.
PR fortran/56047
* gfortran.dg/associate_13.f90: New test.
David Holsgrove [Sat, 26 Jan 2013 16:53:45 +0000 (16:53 +0000)]
gcc: PR target/54663
gcc: PR target/54663
* config.gcc (microblaze*-linux*): Add tmake_file to allow building
of microblaze-c.o.
libgcc:
* config.host(microblaze*-linux*): tmake_file: Remove
t-slibgcc-nolc-override, add t-slibgcc-libgcc.
* config/microblaze/t-microblaze: Set LIB2FUNCS_EXCLUDE
to exclude functions from being built with libgcc.c and use
the microblaze assembly.
Jakub Jelinek [Fri, 25 Jan 2013 20:03:54 +0000 (21:03 +0100)]
re PR middle-end/56098 (conditional write through volatile pointer produces unintended read)
PR tree-optimization/56098
* tree-ssa-phiopt.c (nt_init_block): Don't call add_or_mark_expr
for stmts with volatile ops.
(cond_store_replacement): Don't optimize if assign has volatile ops.
(cond_if_else_store_replacement_1): Don't optimize if either
then_assign or else_assign have volatile ops.
(hoist_adjacent_loads): Don't optimize if either def1 or def2 have
volatile ops.
* gcc.dg/pr56098-1.c: New test.
* gcc.dg/pr56098-2.c: New test.
Tejas Belagod [Fri, 25 Jan 2013 11:35:03 +0000 (11:35 +0000)]
aarch64-simd-builtins.def: Separate sq<r>dmulh_lane entries into lane and laneq entries.
2013-01-25 Tejas Belagod <tejas.belagod@arm.com>
* config/aarch64/aarch64-simd-builtins.def: Separate sq<r>dmulh_lane
entries into lane and laneq entries.
* config/aarch64/aarch64-simd.md (aarch64_sq<r>dmulh_lane<mode>): Remove
AdvSIMD scalar modes.
(aarch64_sq<r>dmulh_laneq<mode>): New.
(aarch64_sq<r>dmulh_lane<mode>): New RTL pattern for Scalar AdvSIMD
modes.
* config/aarch64/arm_neon.h: Fix all the vq<r>dmulh_lane* intrinsics'
builtin implementations to relfect changes in RTL in aarch64-simd.md.
* config/aarch64/iterators.md (VCOND): New.
(VCONQ): New.
projects / thirdparty / gcc.git / log
