H.J. Lu [Mon, 10 Dec 2012 14:43:54 +0000 (14:43 +0000)]
Record the global variables if WPA isn't enabled
PR lto/55466
* lto-symtab.c (lto_symtab_merge_decls_1): Don't record the
prevailing variable.
* lto.c (lto_register_var_decl_in_symtab): Don't record static
variables.
(lto_main): Record the global variables if WPA isn't enabled.
Richard Biener [Mon, 10 Dec 2012 14:00:25 +0000 (14:00 +0000)]
re PR tree-optimization/55107 (GCC in an infinite loop in PRE)
2012-12-10 Richard Biener <rguenther@suse.de>
PR tree-optimization/55107
* tree-ssa-pre.c (struct pre_stats): Remove constified field.
(bitmap_set_replace_value): Add gcc_unreachable.
(do_regular_insertion): Re-write all_same handling. Insert
an assignment instead of a PHI in this case.
(eliminate_bb): Record availability also for SSA names defined
by a constant.
(do_pre): Do not record constified events.
(execute_fre): Likewise.
* gcc.dg/torture/pr55107.c: New testcase.
* gcc.dg/tree-ssa/ssa-pre-5.c: Adjust.
Jakub Jelinek [Mon, 10 Dec 2012 12:14:36 +0000 (13:14 +0100)]
asan.c (asan_init_shadow_ptr_types): Move earlier in the file.
* asan.c (asan_init_shadow_ptr_types): Move earlier in the file.
Call initialize_sanitizer_builtins at the end.
(asan_pp_string): Use TREE_TYPE (shadow_ptr_types[0])
as character type instead of char_type_node.
(asan_emit_stack_protection): Call asan_init_shadow_ptr_types
if shadow_ptr_types isn't initialized.
(asan_protect_global): Return true for STRING_CSTs except those
created by asan_pp_string.
(count_string_csts, add_string_csts): New functions.
(struct asan_add_string_csts_data): New type.
(asan_finish_file): Clear flag_asan at the beginning, restore at the
end. Traverse constant_pool_htab () to look for protected
STRING_CSTs. Don't call initialize_sanitizer_builtins,
instead call asan_init_shadow_ptr_types if shadow_ptr_types isn't
initialized yet.
(asan_instrument): Don't call initialize_sanitizer_builtins.
* varasm.c (output_constant_def_contents): If STRING_CST should be
asan protected, align it sufficiently and emit padding after it.
(categorize_decl_for_section): If flag_asan, don't put STRING_CSTs
that should be asan protected into mergeable sections. For
-fmerge-all-constants, ignore it for -fmudflap or if decl is
asan protected.
Thomas Koenig [Sun, 9 Dec 2012 09:15:36 +0000 (09:15 +0000)]
re PR fortran/55593 (Bogus error on passing DO LOOP variable)
2012-12-09 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/55593
* frontend-passes.c (doloop_code): Use resolved_sym
instead of n.sym->formal for formal argument list
to get the correct version for all generic subroutines.
2012-12-09 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/55593
* gfortran.dg/do_check_8.f90: New test.
Steven Bosscher [Sat, 8 Dec 2012 12:12:50 +0000 (12:12 +0000)]
re PR rtl-optimization/55158 (segfault in schedule_region at -O3)
PR rtl-optimization/55158
* sched-rgn.c (bb_state_array, bb_state): Add some explaining
comment, and initialize to NULL explicitly.
(realloc_bb_state_array): New function.
(free_bb_state_array): New function.
(schedule_region): Call realloc_bb_state_array after schedule_block.
(sched_rgn_init): Use realloc_bb_state_array to initialize bb_state.
(sched_rgn_finish): Use free_bb_state_array to free it.
Dodji Seketeli [Fri, 7 Dec 2012 17:05:19 +0000 (17:05 +0000)]
PR c++/54401 - Confusing diagnostics about type-alias at class scope
Consider this invalid example given in the PR, where T is not defined:
1 template<typename>
2 struct X {
3 using type = T;
4 };
g++ yields the confusing diagnostics:
test.cc:3:10: error: expected nested-name-specifier before 'type'
using type = T;
^
test.cc:3:10: error: using-declaration for non-member at class scope
test.cc:3:15: error: expected ';' before '=' token
using type = T;
^
test.cc:3:15: error: expected unqualified-id before '=' token
I think this is because in cp_parser_member_declaration we tentatively
parse an alias declaration; we then have a somewhat meaningful
diagnostic which alas is not emitted because we are parsing
tentatively. As the parsing didn't succeed (because the input is
invalid) we try to parse a using declaration, which fails as well; but
then the diagnostic emitted is the one for the failed attempt at
parsing a using declaration, not an alias declaration. Oops.
The idea of this patch is to commit the tentative parse when we see
the '=' token in the alias-declaration. That way any error encounter
after that token is reported to the user.
We are now getting the following output:
test.cc:3:18: erreur: expected type-specifier before ‘T’
using type = T;
^
test.cc:3:18: erreur: ‘T’ does not name a type
I don't really like the "before 'T'" there, but I think we maybe could
revisit the format of what cp_parser_error emits in general, now that
we have caret diagnostics; We could maybe do away with the "before T"
altogether?
In the mean time, it seems to me that this patch brings an improvement
over what we already have in trunk, and the issue above could be
addressed separately.
Tested on x86_64-unknown-linux-gnu against trunk.
gcc/cp/
* parser.c (cp_parser_alias_declaration): Commit to tentative
parse when see the '=' token. Get out if the type-id is invalid.
Update function comment.
(cp_parser_member_declaration): Don't try to parse a using
declaration if we know that we expected an alias declaration; that
is, if we see the '=' token after the identifier.
gcc/testsuite/
* g++.dg/cpp0x/alias-decl-28.C: New test.
* g++.dg/cpp0x/alias-decl-16.C: Update.
Martin Jambor [Fri, 7 Dec 2012 13:05:52 +0000 (14:05 +0100)]
re PR middle-end/55078 (FAIL: g++.dg/torture/pr46154.C)
2012-12-07 Martin Jambor <mjambor@suse.cz>
PR middle-end/55078
* ipa-inline-transform.c (inline_call): Turn #if 0 to #ifdef
ENABLE_CHECKING.
* ipa-prop.c (try_make_edge_direct_simple_call): Use
ipa_value_from_jfunc.
(try_make_edge_direct_virtual_call): Likewise.
(update_indirect_edges_after_inlining): Lookup new_root_info and pass
it to the functions above.
Jason Merrill [Thu, 6 Dec 2012 20:21:08 +0000 (15:21 -0500)]
re PR c++/54744 (internal compiler error: Segmentation fault, by dependent base, member typedef and ctor-initializer)
PR c++/54744
* pt.c (resolve_typename_type): Check TYPENAME_IS_RESOLVING_P on scope.
* init.c (expand_member_init): Check for being in a template first.
* parser.c (cp_parser_mem_initializer_list): Only check class types
for equivalence to the current class.
Jakub Jelinek [Thu, 6 Dec 2012 14:37:59 +0000 (15:37 +0100)]
re PR middle-end/43631 (var-tracking inserts notes with non-NULL BLOCK_FOR_INSN in between basic blocks)
PR middle-end/43631
* var-tracking.c (emit_note_insn_var_location, emit_notes_in_bb):
Clear BLOCK_FOR_INSN on notes emitted in between basic blocks,
don't adjust BB_END when inserting note after BB_END of some bb.
Jakub Jelinek [Thu, 6 Dec 2012 14:37:09 +0000 (15:37 +0100)]
re PR c++/55137 (Unexpected static structure initialization)
PR c++/55137
* fold-const.c (fold_binary_loc) <associate>: Don't introduce
TREE_OVERFLOW through reassociation. If type doesn't have defined
overflow, but one or both of the operands do, use the wrapping type
for reassociation and only convert to type at the end.
* g++.dg/opt/pr55137.C: New test.
* gcc.c-torture/execute/pr55137.c: New test.
Aldy Hernandez [Wed, 5 Dec 2012 21:10:18 +0000 (21:10 +0000)]
re PR target/54160 (gcc should not define __OBJC2__ when lang is not set to ObjC (gcc 4.6 and later))
PR target/54160
* config/t-darwin (darwin-c.o): Depend on $(CPP_INTERNAL_H).
* config/darwin-c.c (darwin_cpp_builtins): Only set __OBJC2__
when in obj mode.
Include libcpp/internal.h.
Arnaud Charlet [Wed, 5 Dec 2012 11:21:32 +0000 (12:21 +0100)]
[multiple changes]
2012-12-05 Thomas Quinot <quinot@adacore.com>
* par_sco.adb (Traverse_Aspects): Ensure we always have
an entry in the sloc -> SCO map for invariants, since
Set_SCO_Pragma_Enabled is called with that sloc when checks
are enabled.
* par-prag.adb: Checks and processing of pragma Assume are
carried out by Sem_Prag.
* sem_prag.adb (Analyze_Pragma): Check the legality of pragma
Assume.
* snames.ads-tmpl: Add new name Assume. Add a pragma identifier
for Assume.
Ed Schonberg [Wed, 5 Dec 2012 11:20:13 +0000 (11:20 +0000)]
aspects.ads, [...]: Add aspect Relative_Deadline.
2012-12-05 Ed Schonberg <schonberg@adacore.com>
* aspects.ads, aspects.adb: Add aspect Relative_Deadline.
* sem_ch13.adb (Analyze_Aspect_Specifications): Process aspect
Relative_Deadline, and introduce the corresponding pragma within
the task definition of the task type to which it applies.
(Check_Aspect_At_Freeze_Point): Expression in a Relative_Deadline
aspect is of type Time_Span.
* sem_prag.adb (Check_Loop_Invariant_Variant_Placement): When pragma
Loop_[In]variant does not appear immediately within the statements
of a loop, it must appear in a chain of nested blocks.
2012-12-05 Thomas Quinot <quinot@adacore.com>
* sem_ch13.adb: Minor reformatting.
Remove redundant assertion.
Arnaud Charlet [Wed, 5 Dec 2012 11:15:35 +0000 (12:15 +0100)]
[multiple changes]
2012-12-05 Thomas Quinot <quinot@adacore.com>
* par_sco.adb, scos.ads, put_scos.adb, put_scos.ads,
get_scos.adb: Generation of SCOs for aspects.
2012-12-05 Thomas Quinot <quinot@adacore.com>
* sem_prag.adb (Check_Precondition_Postcondition): Remove
redundant call to Set_SCO_Pragma_Enabled (the pragma will be
rewritten into a pragma Check later on, and the call will be
made when processing the rewritten pragma).
(Analyze_Pragma, case Pragma_Check): Omit call to
Set_SCO_Pragma_Enabled if Split_PPC is set.
Arnaud Charlet [Wed, 5 Dec 2012 11:11:28 +0000 (12:11 +0100)]
[multiple changes]
2012-12-05 Robert Dewar <dewar@adacore.com>
* atree.ads, par-ch4.adb, sem_attr.adb, sem_ch13.adb: Minor
reformatting.
2012-12-05 Eric Botcazou <ebotcazou@adacore.com>
* gnat_ugn.texi (Switches for gcc): Document -flto and reorder.
2012-12-05 Ed Schonberg <schonberg@adacore.com>
* par-ch5.adb (P_Condition): Handle properly Ada 2012
expressions that must be parenthesized before checking for
redundant parentheses when that check is enabled.
Arnaud Charlet [Wed, 5 Dec 2012 11:09:27 +0000 (12:09 +0100)]
[multiple changes]
2012-12-05 Ed Schonberg <schonberg@adacore.com>
* par-ch4.adb (P_Allocator): In Ada 2012 (AI05-0104) an
uninitialized allocator cannot carry an explicit not null
indicator.
* sem_ch4.adb (Analyze_Allocator): Remove code that implements
the check for AI05-0104, the check is syntactic and performed
in the parser.
2012-12-05 Geert Bosch <bosch@adacore.com>
* sem_attr.adb (Analyze_Attribute): Use base type for floating
point attributes.
2012-12-05 Ed Schonberg <schonberg@adacore.com>
* sem_ch13.adb (Analyze_Aspect_Specifications,
Ahalyze_Aspect_Default_Value): For a scalar type attach default
value to base type as well, because it is a type-specific aspect
even though it can be specified on a first subtype.
Yufeng Zhang [Wed, 5 Dec 2012 11:08:43 +0000 (11:08 +0000)]
aarch64.c (aarch64_simd_mangle_map_entry): New typedef.
gcc/
2012-12-05 Yufeng Zhang <yufeng.zhang@arm.com>
* config/aarch64/aarch64.c (aarch64_simd_mangle_map_entry): New
typedef.
(aarch64_simd_mangle_map): New table.
(aarch64_mangle_type): Locate and return the mangled name for
a given AdvSIMD vector type.
Arnaud Charlet [Wed, 5 Dec 2012 11:06:35 +0000 (12:06 +0100)]
[multiple changes]
2012-12-05 Yannick Moy <moy@adacore.com>
* urealp.ads: Minor rewording.
2012-12-05 Yannick Moy <moy@adacore.com>
* aspects.ads (No_Duplicates_Allowed): Forbid use of duplicate
Contract_Cases aspects.
* sem_prag.adb (Analyze_Pragma/Pragma_Contract_Case): Rename
POST_CASE into CONTRACT_CASE in both grammar and code, to be
consistent with current language definition. Issue a more precise
error message when the pragma duplicates another pragma or aspect.
* exp_attr.adb (Expand_N_Attribute_Reference): Add processing
for attribute Update.
(Expand_Update_Attribute): New routine.
* par-ch4.adb (P_Name): The sole expression of attribute Update
is an aggregate, parse it accordingly.
* sem_attr.adb (Analyze_Attribute): Verify the legality of
attribute Update.
(Eval_Attribute): Attribute Update does not
need evaluation because it is never static.
* snames.ads-tmpl: Add Name_Update to the list of special names
recognized by the compiler. Add an Attribute_Id for Update.
2012-12-05 Ed Schonberg <schonberg@adacore.com>
* exp_util.adb (Remove_Side_Effects): For purposes of removing
side effects, qualified expressions do not receive a special
treatment, even though in Ada 2012 they are defined as object
references.
Thomas Quinot [Wed, 5 Dec 2012 11:03:15 +0000 (11:03 +0000)]
exp_dist.adb (Build_From_Any_Call, [...]): For a used-defined subtype, always go to the first subtype of the base type.
2012-12-05 Thomas Quinot <quinot@adacore.com>
* exp_dist.adb (Build_From_Any_Call, Build_To_Any_Call,
Build_TypeCode_Call): For a used-defined subtype, always go to
the first subtype of the base type.
projects / thirdparty / gcc.git / log
