Steven G. Kargl [Wed, 18 Nov 2015 22:55:17 +0000 (22:55 +0000)]
re PR fortran/59910 (ICE in gfc_conv_array_initializer, at fortran/trans-array.c:5327)
2015-11-18 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/59910
* primary.c (gfc_match_structure_constructor): Reduce a structure
constructor in a DATA statement.
2015-11-18 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/59910
* gfortran.dg/pr59910.f90: New test.
From-SVN: r230580
Steven G. Kargl [Wed, 18 Nov 2015 22:51:39 +0000 (22:51 +0000)]
re PR fortran/43996 (ICE in gfc_conv_array_initializer due to incomplete simplification of init expressions)
2015-11-18 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/43996
* simplify.c (gfc_simplify_spread): Issue error for too large array
constructor in a PARAMETER statement.
2015-11-18 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/43996
* gfortran.dg/pr43996.f90: New test.
From-SVN: r230579
Joseph Myers [Wed, 18 Nov 2015 22:13:44 +0000 (22:13 +0000)]
Add out-of-line versions of some <stdatomic.h> functions (PR c/65083).
PR c/65083 notes that some functions in <stdatomic.h> are normal
functions, not generic functions, and so need to have out-of-line
copies that can be called when macro expansion is suppressed (unlike
the generic functions where DR#419 makes it undefined if you suppress
a macro expansion).
This patch adds such out-of-line definitions in libatomic for those
six functions, at a new LIBATOMIC_1.2 symbol version, as trivial
wrappers to the <stdatomic.h> macros, along with declarations of those
functions in <stdatomic.h>. Tests are added that are based on the
corresponding tests for the macros, but with parentheses around the
function names to force the out-of-line functions to be used.
Bootstrapped with no regressions on x86_64-pc-linux-gnu.
gcc:
* ginclude/stdatomic.h (atomic_thread_fence, atomic_signal_fence)
(atomic_flag_test_and_set, atomic_flag_test_and_set_explicit)
(atomic_flag_clear, atomic_flag_clear_explicit): Declare as
functions before defining as macros.
gcc/testsuite:
* gcc.dg/atomic/stdatomic-fence-2.c,
gcc.dg/atomic/stdatomic-flag-2.c: New tests.
libatomic:
* fence.c, flag.c: New files.
* Makefile.am (libatomic_la_SOURCES): Add fence.c and flag.c.
* Makefile.in: Regenerate.
* configure.ac (libtool_VERSION): Change to 3:0:2.
* configure: Regenerate.
* libatomic.map (LIBATOMIC_1.2): New symbol version.
From-SVN: r230578
Eric Botcazou [Wed, 18 Nov 2015 21:59:30 +0000 (21:59 +0000)]
trans.c (elaborate_all_entities_for_package): New function extracted from...
* gcc-interface/trans.c (elaborate_all_entities_for_package): New
function extracted from... Recurse on packages.
(elaborate_all_entities): ...here. Call it on packages.
From-SVN: r230576
Eric Botcazou [Wed, 18 Nov 2015 21:55:11 +0000 (21:55 +0000)]
ada-tree.h (DECL_INVARIANT_P): New macro.
* gcc-interface/ada-tree.h (DECL_INVARIANT_P): New macro.
* gcc-interface/gigi.h (enum standard_datatypes): Remove
ADT_longjmp_decl and add ADT_not_handled_by_others_decl.
(longjmp_decl): Delete.
(not_handled_by_others_decl): New macro.
(build_simple_component_ref): Delete.
(build_component_ref): Adjust prototype.
* gcc-interface/decl.c (gnat_to_gnu_entity): Adjust calls to
build_component_ref.
(gnat_to_gnu_field): Set DECL_INVARIANT_P on discriminants
without default value.
* gcc-interface/trans.c (gigi): Reorder initialization sequence
and add not_handled_by_others_decl.
(Attribute_to_gnu): Adjust calls to build_component_ref.
(Subprogram_Body_to_gnu): Likewise.
(Call_to_gnu): Likewise.
(Exception_Handler_to_gnu_sjlj): Likewise.
(gnat_to_gnu): Likewise.
(range_check_info_d): Add inserted_cond field.
(Loop_Statement_to_gnu): Make two passes on the recorded range checks.
(build_noreturn_cond): New static function.
(Raise_Error_to_gnu): Record range checks in loops at -O1 and above.
(make_invariant): New static function.
(Loop_Statement_to_gnu): Use it to compute invariant expressions for
the loop bounds if possible, but do not require it if loop unswitching
is enabled.
* gcc-interface/utils.c (convert_to_fat_pointer): Likewise.
(convert): Likewise.
(maybe_unconstrained_array): Likewise. Call it instead of
build_simple_component_ref and add guard for CONSTRUCTORs.
(unchecked_convert): Likewise.
* gcc-interface/utils2.c (compare_fat_pointers): Likewise.
(build_simple_component_ref): Remove COMPONENT parameter, unify
code dealing with VIEW_CONVERT_EXPR and make it more general,
remove special treatment for CONSTRUCTORs of template types.
(build_component_ref): Remove COMPONENT parameter and adjust call
to build_simple_component_ref.
(maybe_wrap_malloc): Likewise.
(build_allocator): Likewise.
(gnat_invariant_expr): Look through overflow checks, deal with
addition and subtraction of constants and take into account
DECL_INVARIANT_P for the COMPONENT_REF case.
From-SVN: r230575
Nathan Sidwell [Wed, 18 Nov 2015 21:33:57 +0000 (21:33 +0000)]
nvptx.c (nvptx_process_pars): Fix whitespace.
* config/nvptx/nvptx.c (nvptx_process_pars): Fix whitespace.
(nvptx_record_offload_symbol): Remove code compensating for lack
of default dimension handling.
(nvptx_goacc_validate_dims): Remove incorrect ARG_UNUSED markers.
From-SVN: r230571
Eric Botcazou [Wed, 18 Nov 2015 21:17:51 +0000 (21:17 +0000)]
misc.c: Move global variables to the top of the file.
* gcc-interface/misc.c: Move global variables to the top of the file.
(gnat_handle_option): Remove obsolete ATTRIBUTE_UNUSED markers.
(gnat_init_options): Minor tweak.
(gnat_post_options): Set -fstrict-overflow if not done by the user.
(internal_error_function): Minor reformatting.
From-SVN: r230569
Aditya Kumar [Wed, 18 Nov 2015 21:08:40 +0000 (21:08 +0000)]
Enable condegen in case of cond phis.
The codegen of conditional PHIs inside the scop where one predecessor dominates
the other was difficult so it wasn't enabled in the previous patch. After a
couple of bug-fixes this has been enabled in this patch. Not all the cases
could be handled in this case because it becomes difficult to map the basic
block back to original code in some cases.
Bug-fixes:
1. The vec_find returns -1 when no element was found. This wasn't checked.
2. When the arguments to pending phis could not be resolved in the second pass,
the codegen would fail so the new code should be cleaned up.
This patch passes regtest and bootstrap on linux-x86-64 with
BOOT_CFLAGS='-O2 -fgraphite-identity -floop-nest-optimize'
2015-11-14 hiraditya <hiraditya@msn.com>
* graphite-isl-ast-to-gimple.c (copy_loop_phi_args): Change the return
type to bool for early exit.
(translate_isl_ast_to_gimple::copy_loop_phi_nodes): Early return in
case of error.
(translate_isl_ast_to_gimple::copy_loop_close_phi_args): Same.
(add_phi_arg_for_new_expr): Enable codegen for if-block where one predecessor
dominates the other.
(translate_isl_ast_to_gimple::copy_cond_phi_args): Fix. When the element is not
found it returns -1.
(translate_isl_ast_to_gimple::translate_pending_phi_nodes): Bail out early when
codegen fails.
(graphite_regenerate_ast_isl): Remove codegen region when pending phis
could not be generated.
From-SVN: r230567
Aditya Kumar [Wed, 18 Nov 2015 21:08:33 +0000 (21:08 +0000)]
Move codegen related functions to graphite-isl-ast-to-gimple.c
No functional changes intended.
This patch passes regtest and bootstrap on linux-x86-64 with
BOOT_CFLAGS='-O2 -fgraphite-identity -floop-nest-optimize'
gcc/ChangeLog:
2015-11-14 hiraditya <hiraditya@msn.com>
* graphite-isl-ast-to-gimple.c (struct ast_build_info): Remove semicolon.
(class translate_isl_ast_to_gimple): Indentation.
(translate_pending_phi_nodes): Comment.
(add_parameters_to_ivs_params): Moved from sese.c inside class translate_isl_ast_to_gimple.
(get_max_schedule_dimensions): Same.
(generate_isl_context): Same.
(extend_schedule): Same.
(generate_isl_schedule): Same.
(set_options): Same.
(scop_to_isl_ast): Same.
(is_valid_rename): Same.
(get_rename): Same.
(get_rename_from_scev): Same.
(get_def_bb_for_const): Same.
(get_new_name): Same.
(collect_all_ssa_names): Same.
(copy_loop_phi_args): Same.
(copy_loop_phi_nodes): Same.
(copy_loop_close_phi_args): Same.
(copy_loop_close_phi_nodes): Same.
(copy_cond_phi_args): Same.
(copy_cond_phi_nodes): Same.
(graphite_copy_stmts_from_block): Same.
(copy_bb_and_scalar_dependences): Same.
(add_phi_arg_for_new_expr): Same.
(rename_uses): Same.
(set_rename): Same.
(set_rename_for_each_def): Same.
(gsi_insert_earliest): Same.
(rename_all_uses): Same.
(codegen_error_p): Same.
(print_isl_ast_node): Same.
(translate_isl_ast_for_loop): Call function codegen_error_p.
(translate_isl_ast_to_gimple::translate_isl_ast): Same.
(translate_isl_ast_node_user): Make nb_loops const and release iv_map before exit.
(get_true_edge_from_guard_bb): Move all free-functions early.
(get_false_edge_from_guard_bb): Same.
(bb_contains_loop_close_phi_nodes): Same.
(bb_contains_loop_phi_nodes): Same.
(is_loop_closed_ssa_use): Same.
(number_of_phi_nodes): Same.
(phi_uses_name): Same.
(later_of_the_two): Same.
(substitute_ssa_name):
(get_edges): Same.
(get_loc): Same.
(get_loop_init_value): Same.
(find_init_value): Same.
(find_init_value_close_phi): Same.
(ast_build_before_for): Same.
(graphite_regenerate_ast_isl): Formatting changes.
* graphite-scop-detection.c (build_cross_bb_scalars_use): Same.
* sese.c (get_rename): Move to graphite-isl-ast-to-gimple.c
(set_rename): Same.
(gsi_insert_earliest): Same.
(collect_all_ssa_names): Same.
(rename_all_uses): Same.
(rename_uses): Same.
(get_def_bb_for_const): Same.
(copy_loop_phi_nodes): Same.
(copy_loop_close_phi_args): Same.
(copy_loop_close_phi_nodes): Same.
(copy_cond_phi_args): Same.
(copy_cond_phi_nodes): Same.
(set_rename_for_each_def): Same.
(graphite_copy_stmts_from_block): Same.
(copy_bb_and_scalar_dependences): Same.
(if_region_set_false_region): Same.
(scev_analyzable_p): Same.
* sese.h: Delete extern functions moved to graphite-isl-ast-to-gimple.c
From-SVN: r230566
Andreas Tobler [Wed, 18 Nov 2015 20:48:18 +0000 (21:48 +0100)]
freebsd.h (ASM_OUTPUT_MAX_SKIP_ALIGN): Bring in the commit from r125920 for FreeBSD.
2015-11-18 Andreas Tobler <andreast@gcc.gnu.org>
* config/i386/freebsd.h (ASM_OUTPUT_MAX_SKIP_ALIGN): Bring in the
commit from r125920 for FreeBSD.
From-SVN: r230565
Jason Merrill [Wed, 18 Nov 2015 20:11:32 +0000 (15:11 -0500)]
Support GGC finalizers with PCH.
* ggc-page.c (ggc_globals): Change finalizers and vec_finalizers
to be vecs of vecs.
(add_finalizer): Split out from ggc_internal_alloc.
(ggc_handle_finalizers): Run finalizers for the current depth.
(init_ggc, ggc_pch_read): Reserve space for finalizers.
From-SVN: r230564
Sandra Loosemore [Wed, 18 Nov 2015 19:45:47 +0000 (14:45 -0500)]
re PR target/68410 (config/nios2/nios2.c: 4123: duplicates in expression)
2015-11-18 Sandra Loosemore <sandra@codesourcery.com>
PR target/68410
* config/nios2/nios2.c (cdx_and_immed): Remove duplicate tests
from || expression.
From-SVN: r230563
Pitchumani Sivanupandi [Wed, 18 Nov 2015 19:15:31 +0000 (19:15 +0000)]
avr-mcus.def: Add new avr4 devices atmega48pb and atmega88pb.
* config/avr/avr-mcus.def: Add new avr4 devices atmega48pb and
atmega88pb. Add new avr5 devices ata5791, ata8210, ata8510, atmega168pb
and atmega328pb.
* doc/avr-mmcu.texi: Regenerate.
From-SVN: r230562
Nathan Sidwell [Wed, 18 Nov 2015 18:33:38 +0000 (18:33 +0000)]
nvptx.c (bb_pair_t, [...]): New types.
gcc/
* config/nvptx/nvptx.c (bb_pair_t, bb_pair_vec_t): New types.
(pseudo_node_t, struct bracket, bracket_vec_t): New types.
(struct bb_sese): New struct.
(bb_sese::~bb_sese, bb_sese::append, bb_sese::remove): New.
(BB_GET_SESE, BB_SET_SESE): Define.
(nvptx_sese_number, nvptx_sese_pseudo, nvptx_sese_color): New.
(nvptx_find_sese): New.
(nvptx_neuter_pars): Find SESE regions when optimizing.
gcc/testsuite/
* gcc.dg/goacc/nvptx-sese-1.c: New.
From-SVN: r230561
Eric Botcazou [Wed, 18 Nov 2015 18:31:42 +0000 (18:31 +0000)]
decl.c (gnat_to_gnu_entity): Use case #1 for the renaming of an address.
* gcc-interface/decl.c (gnat_to_gnu_entity) <E_Variable>: Use case #1
for the renaming of an address.
From-SVN: r230560
Eric Botcazou [Wed, 18 Nov 2015 18:19:57 +0000 (18:19 +0000)]
trans.c (fold_constant_decl_in_expr): If the index is not itself constant then bail out.
* gcc-interface/trans.c (fold_constant_decl_in_expr) <ARRAY_REF>: If
the index is not itself constant then bail out.
From-SVN: r230558
Jeff Law [Wed, 18 Nov 2015 16:55:41 +0000 (09:55 -0700)]
Add PR marker to last change
From-SVN: r230556
Ajit Agarwal [Wed, 18 Nov 2015 16:54:16 +0000 (16:54 +0000)]
split-path-1.c: Fix the testcase to avoid undefined behaviour.
* gcc.dg/tree-ssa/split-path-1.c: Fix the testcase
to avoid undefined behaviour.
From-SVN: r230555
Cesar Philippidis [Wed, 18 Nov 2015 16:31:04 +0000 (08:31 -0800)]
cp-gimplify.c (cp_fold_r): Add support for OACC_LOOP.
gcc/cp/
* cp-gimplify.c (cp_fold_r): Add support for OACC_LOOP.
From-SVN: r230554
Alan Modra [Wed, 18 Nov 2015 15:31:18 +0000 (02:01 +1030)]
rs6000.c (use_toc_relative_ref): Ignore type-limits warning.
2015-11-18 Alan Modra <amodra@gmail.com>
* config/rs6000/rs6000.c (use_toc_relative_ref): Ignore
type-limits warning.
From-SVN: r230551
Richard Sandiford [Wed, 18 Nov 2015 14:31:10 +0000 (14:31 +0000)]
PR 68406: Avoid problem with older host compilers
Older compilers require the Key typedef to be expanded for the traversal
function templates. This patch uses the same approach as we did for
hash_map.
Tested on x86_64-linux-gnu.
gcc/
PR bootstrap/68406
* hash-set.h (hash_set::traverse): Expand Key typedef.
From-SVN: r230549
Martin Liska [Wed, 18 Nov 2015 14:18:11 +0000 (15:18 +0100)]
Finalize func_options in arm target in
* config/arm/arm.c (arm_valid_target_attribute_p): Finalize
options struct.
From-SVN: r230548
Ed Schonberg [Wed, 18 Nov 2015 13:55:42 +0000 (13:55 +0000)]
sem_ch4.adb (Try_Container_Indexing): When constructing the parameter list for the potentially overloaded calls to...
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_ch4.adb (Try_Container_Indexing): When constructing the
parameter list for the potentially overloaded calls to indexing
functions, do not propagate overloadings if the actual is a named
association: overloadings appear directly on the expression in
the association.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_type.adb (Interface_Present_In_Ancestor): If the type is
incomplete, the list of ancestors is found either on the full
view or the non-limited view of the type.
From-SVN: r230547
Arnaud Charlet [Wed, 18 Nov 2015 13:53:58 +0000 (14:53 +0100)]
[multiple changes]
2015-11-18 Nicolas Roche <roche@adacore.com>
* sysdep.c (__gnat_localtime_tzoff): On Windows platform
GetTimeZoneInformation function is thread-safe. Thus there
is no need to lock the runtime in the implementation of
__gnat_localtime_tzoff on that platform.
2015-11-18 Eric Botcazou <ebotcazou@adacore.com>
* s-arit64.adb (To_Neg_Int): Add a special case for 2**63 input.
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* contracts.adb (Analyze_Contracts): New routine.
(Analyze_Enclosing_Package_Body_Contract): Removed.
(Analyze_Entry_Or_Subprogram_Contract): Add formal parameter
Freeze_Id. Propagate the entity of the freezing body to vaious
analysis routines.
(Analyze_Initial_Declaration_Contract): Removed.
(Analyze_Object_Contract): Add formal parameter
Freeze_Id. Propagate the entity of the freezing body to vaious
analysis routines.
(Analyze_Previous_Contracts): New routine.
* contracts.ads (Analyze_Enclosing_Package_Body_Contract): Removed.
(Analyze_Contracts): New routine.
(Analyze_Entry_Or_Subprogram_Contract): Add formal
parameter Freeze_Id and update the comment on usage.
(Analyze_Initial_Declaration_Contract): Removed.
(Analyze_Object_Contract): Add formal parameter Freeze_Id and
update the comment on usage.
(Analyze_Previous_Contracts): New routine.
* sem_ch3.adb (Analyze_Declarations): Use Analyze_Contracts to
analyze all contracts of eligible constructs.
* sem_ch6.adb (Analyze_Generic_Subprogram_Body):
A body no longer freezes the contract of its initial
declaration. This effect is achieved through different means.
(Analyze_Subprogram_Body_Helper): A body now freezes the contracts
of all eligible constructs that precede it. A body no longer
freezes the contract of its initial declaration. This effect is
achieved through different means.
* sem_ch7.adb (Analyze_Package_Body_Helper): A body now freezes
the contracts of all eligible constructs that precede it. A body
no longer freezes the contract of its initial declaration. This
effect is achieved through different means.
* sem_ch9.adb (Analyze_Entry_Body): A body now freezes
the contracts of all eligible constructs that precede
it. A body no longer freezes the contract of its initial
declaration. This effect is achieved through different means.
(Analyze_Protected_Body): A body now freezes the contracts
of all eligible constructs that precede it. A body no longer
freezes the contract of its initial declaration. This effect
is achieved through different means.
(Analyze_Task_Body): A
body now freezes the contracts of all eligible constructs that
precede it. A body no longer freezes the contract of its initial
declaration. This effect is achieved through different means.
* sem_prag.adb (Add_Item_To_Name_Buffer): Single protected/task
objects now output their respective current instance of xxx
type messages. (Analyze_Contract_Cases_In_Decl_Part): Add
formal parameter Freeze_Id. Emit a clarification message
when an undefined entity may the byproduct of contract
freezing.
(Analyze_Part_Of_In_Decl_Part): Add formal
parameter Freeze_Id. Emit a clarification message when an
undefined entity may the byproduct of contract freezing.
(Analyze_Pre_Post_Condition_In_Decl_Part): Add formal
parameter Freeze_Id. Emit a clarification message when an
undefined entity may the byproduct of contract freezing.
(Analyze_Refined_State_In_Decl_Part): Do not report unused body
states as constituents of single protected/task types may not
bave been identified yet.
(Collect_Subprogram_Inputs_Outputs):
Reimplemented. (Contract_Freeze_Error): New routine.
(Process_Overloadable): Use predicate Is_Single_Task_Object.
* sem_prag.ads (Analyze_Contract_Cases_In_Decl_Part):
Add formal parameter Freeze_Id and update the comment
on usage.
(Analyze_Part_Of_In_Decl_Part): Add formal
parameter Freeze_Id and update the comment on usage.
(Analyze_Pre_Post_Condition_In_Decl_Part): Add formal parameter
Freeze_Id and update the comment on usage.
* sem_util.adb (Check_Unused_Body_States): Remove global
variable Legal_Constits. The routine now reports unused
body states regardless of whether constituents are
legal or not.
(Collect_Body_States): A constituent of a
single protected/task type is not a visible state of a
package body.
(Collect_Visible_States): A constituent
of a single protected/task type is not a visible
state of a package body.
(Has_Undefined_Reference): New routine.
(Is_Single_Concurrent_Object): Reimplemented.
(Is_Single_Protected_Object): New routine.
(Is_Single_Task_Object): New routine.
(Is_Visible_Object): New routine.
(Report_Unused_Body_States): Moved to Check_Unused_Body_States.
* sem_util.ads (Check_Unused_Body_States): Update the comment on usage.
(Has_Undefined_Reference): New routine.
(Is_Single_Protected_Object): New routine.
(Is_Single_Task_Object): New routine.
(Report_Unused_Body_States): Moved to Check_Unused_Body_States.
2015-11-18 Pierre-Marie de Rodat <derodat@adacore.com>
* Makefile.rtl, impunit.adb: Add g-strhas.ads.
* g-strhas.ads: New file.
* s-strhas.ads: Add a comment to redirect users to g-strhas.ads.
2015-11-18 Bob Duff <duff@adacore.com>
* sem_elab.adb (Check_Internal_Call_Continue): Fix the case
where the call in question is to a renaming of a subprogram that
can be safely called without ABE.
* checks.adb: Minor edits.
From-SVN: r230546
Nathan Sidwell [Wed, 18 Nov 2015 13:49:17 +0000 (13:49 +0000)]
nvptx.c (global_lock_var): New.
gcc/
* config/nvptx/nvptx.c (global_lock_var): New.
(nvptx_global_lock_addr): New.
(nvptx_lockless_update): Recomment and adjust for clarity.
(nvptx_lockfull_update): New.
(nvptx_reduction_update): New.
(nvptx_goacc_reduction_fini): Call it.
libgcc/
* config/nvptx/reduction.c: New.
* config/nvptx/t-nvptx (LIB2ADD): Add it.
libgomp/
* testsuite/libgomp.oacc-c-c++-common/reduction-cplx-flt.c: Add
worker & gang cases.
* testsuite/libgomp.oacc-c-c++-common/reduction-cplx-dbl.c: Likewise.
From-SVN: r230545
Bernd Schmidt [Wed, 18 Nov 2015 12:26:43 +0000 (12:26 +0000)]
regrename.h (struct du_head): Add target_data_1 and target_data_2 fields.
* regrename.h (struct du_head): Add target_data_1 and target_data_2
fields.
* regrename.c (create_new_chain): Clear entire struct after allocating.
* config/i386/i386.opt (mmitigate-rop): New option.
* doc/invoke.texi (mmitigate-rop): Document.
* config/i386/i386.c: Include "regrename.h".
(ix86_rop_should_change_byte_p, reg_encoded_number,
ix86_get_modrm_for_rop, set_rop_modrm_reg_bits, ix86_mitigate_rop): New
static functions.
(ix86_reorg): Call ix86_mitigate_rop if -fmitigate-rop.
* config/i386/i386.md (attr "modrm_class"): New.
(cmp<mode>_ccno_1, mov<mode>_xor, movstrict<mode>_xor,
x86_mov<mode>cc_0_m1. x86_mov<mode>cc_0_m1_se)
(x86_mov<mode>cc_0_m1_neg): Override modrm_class attribute.
From-SVN: r230543
Ilya Enkovich [Wed, 18 Nov 2015 11:51:58 +0000 (11:51 +0000)]
re PR target/68405 (trunk/gcc/config/i386/i386.c:22951: possible missing break ?)
gcc/
PR target/68405
* config/i386/i386.c (ix86_expand_mask_vec_cmp): Add missing
break.
From-SVN: r230541
Arnaud Charlet [Wed, 18 Nov 2015 10:53:39 +0000 (11:53 +0100)]
[multiple changes]
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* atree.adb (Elist11): New routine.
(Set_Elist11): New routine.
* atree.ads (Elist11): New routine.
(Set_Elist11): New routine.
* atree.h: Define Elist11.
* contracts.adb (Analyze_Object_Contract): Verify the legality
of all references to a variable given that the variable is a
constituent of a single protected/task type.
* einfo.adb: Part_Of_References now utilizes Elist11.
(Part_Of_References): New routine.
(Set_Part_Of_References): New routine.
(Write_Field11_Name): Add output for Part_Of_References.
* einfo.ads New attribute Part_Of_References along with usage
in entities.
(Part_Of_References): New routine along with
pragma Inline.
(Set_Part_Of_References): New routine along with pragma Inline.
* sem_prag.adb (Analyze_Constituent): Record a possible
reference to a concurrent constituent.
(Analyze_Global_Item): Record a possible reference to a concurrent
constituent.
(Analyze_Input_Output): Record a possible reference to a
concurrent constituent.
* sem_res.adb (Resolve_Entity_Name): Record a possible reference
to a concurrent constituent.
* sem_util.adb (Check_Part_Of_Reference): New routine.
(Record_Possible_Part_Of_Reference): New routine.
* sem_util.ads (Check_Part_Of_Reference): New routine.
(Record_Possible_Part_Of_Reference): New routine.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* checks.adb (Apply_Arithmetic_Overflow_Minimized_Eliminated):
An if_expression is the proper place to apply the overflow
minimization procedure if its context is not an enclosing
arithmetic expression.
From-SVN: r230540
Arnaud Charlet [Wed, 18 Nov 2015 10:51:46 +0000 (11:51 +0100)]
Regenerate doc.
From-SVN: r230539
Arnaud Charlet [Wed, 18 Nov 2015 10:50:40 +0000 (10:50 +0000)]
gnat_project_manager.rst, [...]: Update doc.
* gnat_ugn/gnat_project_manager.rst,
gnat_ugn/building_executable_programs_with_gnat.rst,
gnat_ugn/gnat_and_program_execution.rst,
gnat_ugn/the_gnat_compilation_model.rst,
gnat_rm/implementation_defined_pragmas.rst,
gnat_rm/standard_and_implementation_defined_restrictions.rst,
gnat_ugn.texi, gnat_rm.texi: Update doc.
From-SVN: r230538
Arnaud Charlet [Wed, 18 Nov 2015 10:48:32 +0000 (11:48 +0100)]
Minor code clean ups.
From-SVN: r230537
Arnaud Charlet [Wed, 18 Nov 2015 10:42:27 +0000 (11:42 +0100)]
[multiple changes]
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* contracts.adb (Add_Contract_Item): Chain pragmas Attach_Handler
and Interrupt_Handler on the classifications list of a [generic]
procedure N_Contract node.
* contracts.ads (Add_Contract_Item): Update the comment on usage.
* einfo.adb (Get_Pragma): Pragmas Attach_Handler and
Interrupt_Handler are found on the classifications list of
N_Contract nodes.
* einfo.ads (Get_Pragma): Update the comment on usage.
* sem_prag.adb (Process_Interrupt_Or_Attach_Handler): Code
reformatting. Store the pragma as a contract item.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_ch8.adb (Available_Subtype): Use only in GNATprove
mode. When generating code it may be necessary to create itypes
at the point of use of a selected component, for example in the
expansion of a record equality operation.
2015-11-18 Vincent Celier <celier@adacore.com>
* s-os_lib.adb (Normalize_Pathname.Get_Directory): When
invoking Normalize_Pathname, use the same values for parameters
Resolve_Links and Case_Sensitive as the parent Normalize_Pathname.
2015-11-18 Vincent Celier <celier@adacore.com>
* a-direct.adb (Containing_Directory): Return "." when the result
is the current directory, not specified as an absolute path name.
From-SVN: r230536
Arnaud Charlet [Wed, 18 Nov 2015 10:40:47 +0000 (11:40 +0100)]
[multiple changes]
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* exp_aggr.adb (Is_Completely_Hidden_Discriminant): New routine.
(Init_Hidden_Discriminants): Code reformatting. Do not initialize
a completely hidden discriminant.
* a-interr.ads (Get_CPU): Added SPARK_Mode, Volatile_Function
and Global aspects on the function.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* exp_intr.adb (Expand_Unc_Deallocation): If the designated
type is a concurrent type, the deallocation applies to the
corresponding record type, or to its class-wide type if the type
is tagged.
From-SVN: r230535
Doug Rupp [Wed, 18 Nov 2015 10:39:37 +0000 (10:39 +0000)]
s-parame-vxworks.adb: Reduce default stack size for stack limit check to a reasonable value
2015-11-18 Doug Rupp <rupp@adacore.com>
* s-parame-vxworks.adb: Reduce default stack size for stack
limit check to a reasonable value
* s-tpopsp-vxworks.adb: Make Stack_Limit a task variable for vx5 and
vxmils.
From-SVN: r230534
Arnaud Charlet [Wed, 18 Nov 2015 10:31:53 +0000 (11:31 +0100)]
[multiple changes]
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_ch5.adb (Analyze_Assignment): Diagnose assignment where
left-hand side has a limited view of a class-wide type.
* sem_ch6.adb (Detected_And_Exchange): Do not install a non-limited
view if the scope of the type of the formal is visible through
a limited_with_clause, even if the non-limited view is available.
2015-11-18 Emmanuel Briot <briot@adacore.com>
* s-pooglo.ads: Make Global_Pool_Object aliased.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_attr.adb (Analyze_Attribute, case 'Access): If the prefix
is an expression function, do not analyze the generated body if
the current scope is not the scope of the expression function,
as may happen if the attribute is an actual in an instantiation.
* sem_ch10.adb, sem_ch12.adb, a-stzmap.adb, s-soflin.adb,
a-stwima.adb: Fix typos.
2015-11-18 Arnaud Charlet <charlet@adacore.com>
* exp_ch11.adb (Expand_N_Exception_Declaration): No-op when
generating C code.
2015-11-18 Vincent Celier <celier@adacore.com>
* g-comlin.adb (Find_Longest_Matching_Switch): Do not take into
account a switch ending with '=' when the actual command line
switch is not followed with a space or '='.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_res.adb (Resolve_Explicit_Dereference): Do not add an access
check if the type of the prefix is known to exclude null.
* a-coinve.adb (Get_Element_Access): The function returns an
access type that excludes null, so add an explicit check for
a null element pointer, which is created when Insert_Spaces is
called on an indefinite vector container.
From-SVN: r230533
Hristian Kirtchev [Wed, 18 Nov 2015 10:30:12 +0000 (11:30 +0100)]
re PR ada/66242 (Front-end error if exception propagation disabled)
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
PR ada/66242
* exp_ch3.adb (Default_Initialize_Object): Reimplemented. Abort
defer / undefer pairs are now encapsulated in a block with
an AT END handler. Partial finalization now takes restriction
No_Exception_Propagation into account when generating blocks.
* exp_ch7.adb Various reformattings.
(Create_Finalizer): Change
the generation of abort defer / undefer pairs and explain the
lack of an AT END handler.
(Process_Transient_Objects): Add generation of abort defer/undefer
pairs.
* exp_ch9.adb Various reformattings.
(Build_Protected_Subprogram_Body): Use
Build_Runtime_Call to construct a call to Abort_Defer.
(Build_Protected_Subprogram_Call_Cleanup): Use
Build_Runtime_Call to construct a call to Abort_Undefer.
(Expand_N_Asynchronous_Select): Use Build_Runtime_Call to
construct a call to Abort_Defer.
* exp_intr.adb (Expand_Unc_Deallocation): Abort defer
/ undefer pairs are now encapsulated in a block with
an AT END handler. Finalization now takes restriction
No_Exception_Propagation into account when generating blocks.
* exp_util.ads, exp_util.adb (Wrap_Cleanup_Procedure): Removed.
From-SVN: r230531
Jakub Jelinek [Wed, 18 Nov 2015 10:23:21 +0000 (11:23 +0100)]
re PR tree-optimization/68157 (internal compiler error: in reassoc_stmt_dominates_stmt_p, at tree-ssa-reassoc.c:1287)
PR tree-optimization/68157
* tree-ssa-reassoc.c (attempt_builtin_powi): Set uid of
pow_stmt or mul_stmt from stmt's uid.
(reassociate_bb): Set uid of mul_stmt from stmt's uid.
* g++.dg/opt/pr68157.C: New test.
From-SVN: r230530
Arnaud Charlet [Wed, 18 Nov 2015 10:23:05 +0000 (11:23 +0100)]
[multiple changes]
2015-11-18 Arnaud Charlet <charlet@adacore.com>
* switch-c.adb (Scan_Front_End_Switches): Add a check requiring
-gnatc with -gnateg.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_ch7.adb (Uninstall_Declarations): Before swapping private
and full views, ensure that the potential use visbility of the
two views is consistent.
2015-11-18 Doug Rupp <rupp@adacore.com>
* s-parame-vxworks.adb: Increase default stack size for targets
using stack limit checking.
2015-11-18 Jose Ruiz <ruiz@adacore.com>
* a-interr.adb, a-interr.ads (Ada.Interrupts.Get_CPU): Add this
function that was added to Ada 2005 returning the processor where
interrupt handlers execute.
2015-11-18 Tristan Gingold <gingold@adacore.com>
* exp_ch9.adb (Is_Pure_Barrier): Handle expanded component
declaration of a protected definition.
From-SVN: r230529
Arnaud Charlet [Wed, 18 Nov 2015 10:20:28 +0000 (11:20 +0100)]
[multiple changes]
2015-11-18 Ed Falis <falis@adacore.com>
* s-soflin.adb, s-stchop-vxworks.adb: Code clean ups.
2015-11-18 Gary Dismukes <dismukes@adacore.com>
* bcheck.adb: Minor editorial changes.
From-SVN: r230528
Arnaud Charlet [Wed, 18 Nov 2015 10:12:27 +0000 (11:12 +0100)]
[multiple changes]
2015-11-18 Arnaud Charlet <charlet@adacore.com>
* interfac.ads (Unsigned_24): New type.
* sem_ch13.adb (Validate_Unchecked_Conversion): Disable error on
unconstrained arrays in Relaxed_RM_Semantics mode.
2015-11-18 Vincent Celier <celier@adacore.com>
* bcheck.adb (Check_Consistent_Dynamic_Elaboration_Checking): No
warning is withed unit is the interface of a Stand-Alone Library.
From-SVN: r230527
Arnaud Charlet [Wed, 18 Nov 2015 10:10:46 +0000 (11:10 +0100)]
[multiple changes]
2015-11-18 Thomas Quinot <quinot@adacore.com>
* s-os_lib.ads, s-os_lib.adb(Normalize_Pathname): Support the case of
an unresolved Directory argument, by recursively resolving it
against the current dir.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb (Process_Formals): A function declaration that
returns a class-wide type must have freeing deferred, so that it
is not frozen before the class-wide type and its root type are
frozen. This is significant when there may be a limited view of
the class_wide type in another package.
From-SVN: r230526
Arnaud Charlet [Wed, 18 Nov 2015 10:08:00 +0000 (11:08 +0100)]
[multiple changes]
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* einfo.adb (Has_Non_Null_Refinement): Rename to
Has_Non_Null_Visible_Refinement.
(Has_Null_Refinement): Rename to Has_Null_Visible_Refinement.
* einfo.ads Update the documentation of
attribute Has_Non_Null_Refinement and attribute Has_Null_Refinement.
(Has_Non_Null_Refinement): Rename to Has_Non_Null_Visible_Refinement
and update occurrences in entities.
(Has_Null_Refinement): Rename to Has_Null_Visible_Refinement and update
occurrences in entities.
* sem_prag.adb (Check_In_Out_States): Update the calls to
Has_[Non_]Null_Refinement.
(Check_Input_States): Update the
calls to Has_[Non_]Null_Refinement.
(Check_Output_States): Update the calls to Has_[Non_]Null_Refinement.
(Check_Proof_In_States): Update the calls to Has_[Non_]Null_Refinement.
(Collect_Global_Item): Update the calls to Has_[Non_]Null_Refinement.
(Is_Null_Refined_State): Update the calls to Has_[Non_]Null_Refinement.
(Match_Item): Update the calls to Has_[Non_]Null_Refinement.
* sem_util.adb (Has_Non_Null_Refinement): New routine.
(Has_Null_Refinement): New routine.
* sem_util.ads (Has_Non_Null_Refinement): New routine.
(Has_Null_Refinement): New routine.
2015-11-18 Gary Dismukes <dismukes@adacore.com>
* exp_util.adb: Minor reformatting and typo fixes.
From-SVN: r230525
Hristian Kirtchev [Wed, 18 Nov 2015 10:05:58 +0000 (10:05 +0000)]
sem_ch4.adb: Minor reformatting.
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* sem_ch4.adb: Minor reformatting.
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* exp_util.adb (Expand_Subtype_From_Expr): Add new formal
parameter Related_Id and propagate it to Make_Subtype_From_Expr.
(Make_Subtype_From_Expr): Add new formal parameter
Related_Id. Create external entities when requested by the caller.
* exp_util.ads (Expand_Subtype_From_Expr): Add new formal
parameter Related_Id. Update the comment on usage.
(Make_Subtype_From_Expr): Add new formal parameter
Related_Id. Update the comment on usage.
* sem_ch3.adb (Analyze_Object_Declaration): Add local variable
Related_Id. Generate an external constrained subtype when the
object is a public symbol.
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* sem_prag.adb (Analyze_Pragma): Update the grammars of pragmas
Abstract_State, Depends, Global, Initializes, Refined_Depends,
Refined_Global and Refined_State.
From-SVN: r230524
Arnaud Charlet [Wed, 18 Nov 2015 10:03:44 +0000 (11:03 +0100)]
[multiple changes]
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* sem_util.adb (Has_Full_Default_Initialization):
Perform the test for the presence of pragma
Default_Initial_Condition prior to the specialized type
checks. Add a missing case where the lack of a pragma argument
yields full default initialization.
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* sem_res.adb (Resolve_Entity_Name): Do not check
for elaboration issues when a variable appears as the name of
an object renaming declaration as this constitutes an aliasing,
not a read.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* checks.adb (Overlap_Check): An actual that is an aggregate
cannot overlap with another actual, and no check should be
generated for it.
* targparm.ads: Fix typos.
2015-11-18 Pascal Obry <obry@adacore.com>
* adaint.c: Routine __gnat_killprocesstree only implemented on
Linux and Windows.
2015-11-18 Pascal Obry <obry@adacore.com>
* g-ctrl_c.adb: Minor style fixes.
From-SVN: r230523
Arnaud Charlet [Wed, 18 Nov 2015 09:38:46 +0000 (10:38 +0100)]
[multiple changes]
2015-11-18 Pascal Obry <obry@adacore.com>
* adaint.c, s-os_lib.adb, s-os_lib.ads (Kill_Process_Tree): New.
2015-11-18 Hristian Kirtchev <kirtchev@adacore.com>
* sem_util.adb (Check_Nonvolatile_Function_Profile): Place the error
message concerning the return type on the result definition.
(Is_Volatile_Function): A function with a parameter of a protected
type is a protected function if it is defined within a protected
definition.
2015-11-18 Ed Schonberg <schonberg@adacore.com>
* sem_ch4.adb (Try_Container_Indexing): When building the
parameter list for the function call on indexing functions,
preserve overloading of the parameters, which may themselves be
generalized indexing operations.
From-SVN: r230522
Martin Liska [Wed, 18 Nov 2015 09:06:50 +0000 (09:06 +0000)]
Add missing test-case for PR ipa/68035
From-SVN: r230521
Martin Liska [Wed, 18 Nov 2015 09:01:35 +0000 (10:01 +0100)]
Release memory in haifa-sched
* haifa-sched.c (haifa_finish_h_i_d): Release reg_set_list.
From-SVN: r230520
Martin Liska [Wed, 18 Nov 2015 09:00:58 +0000 (10:00 +0100)]
Fix memory leaks in tree-ssa-uninit.c
* tree-ssa-uninit.c (convert_control_dep_chain_into_preds):
Fix GNU coding style.
(find_def_preds): Use auto_vec.
(destroy_predicate_vecs): Change signature of the function.
(prune_uninit_phi_opnds_in_unrealizable_paths): Use the
new signature.
(simplify_preds_4): Use destroy_predicate_vecs instread of
just releasing preds vector.
(normalize_preds): Likewise.
(is_use_properly_guarded): Use new signature of
destroy_predicate_vecs.
(find_uninit_use): Likewise.
From-SVN: r230519
Richard Biener [Wed, 18 Nov 2015 08:31:32 +0000 (08:31 +0000)]
re PR middle-end/67790 (verify_ssa failed: definition in block 20 follows the use)
2015-11-18 Richard Biener <rguenther@suse.de>
PR tree-optimization/67790
* tree-vect-loop.c (vect_is_simple_reduction_1): Remove
IL rewrite for MINUS_EXPR reductions, rename back to ...
(vect_is_simple_reduction): ... this, removing the wrapper.
(vect_force_simple_reduction): Adjust.
(vectorizable_reduction): Adjust reduc_index for MINUS_EXPR
reductions and make use if reduc_index in all places. For
the final reduction of MINUS_EXPR use PLUS_EXPR.
* gcc.dg/vect/pr67790.c: New testcase.
From-SVN: r230518
David Edelsohn [Wed, 18 Nov 2015 00:43:59 +0000 (00:43 +0000)]
* g++.dg/cpp1y/pr58708.C: Define wfoo based on sizeof wchar_t.
From-SVN: r230516
GCC Administrator [Wed, 18 Nov 2015 00:16:14 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r230515
Alan Modra [Tue, 17 Nov 2015 23:24:30 +0000 (09:54 +1030)]
A default of -z relro (as on Fedora) defeats the deliberate mis-alignment of .got in the testcase.
POWERPC64_TOC_POINTER_ALIGNMENT
A default of -z relro (as on Fedora) defeats the deliberate
mis-alignment of .got in the testcase.
* configure.ac (POWERPC64_TOC_POINTER_ALIGNMENT): Pass -z norelro
to ld.
* configure: Regenerate.
From-SVN: r230509
Jason Merrill [Tue, 17 Nov 2015 21:49:23 +0000 (16:49 -0500)]
re PR bootstrap/68346 (Bootstrap failure on i686-linux)
PR bootstrap/68346
* typeck.c (build_static_cast_1): Force a NOP when converting to
the same type.
From-SVN: r230508
Jason Merrill [Tue, 17 Nov 2015 21:49:18 +0000 (16:49 -0500)]
LITERAL_ZERO_P obsoleted by delayed folding.
* cp-tree.h (LITERAL_ZERO_P): Remove.
* parser.c (cp_parser_postfix_expression, literal_zeros)
(cp_parser_parenthesized_expression_list): Don't mess with it.
From-SVN: r230507
Jason Merrill [Tue, 17 Nov 2015 21:44:08 +0000 (16:44 -0500)]
Don't fold -(constant) or -0.
* parser.c (cp_parser_unary_expression): Fold -constant here.
* typeck.c (cp_build_unary_op): Not here.
From-SVN: r230506
Tom de Vries [Tue, 17 Nov 2015 21:42:36 +0000 (21:42 +0000)]
Improve comments in pass_tree_loop_init::execute
2015-11-17 Tom de Vries <tom@codesourcery.com>
* tree-ssa-loop.c (pass_tree_loop_init::execute): Improve comments.
From-SVN: r230505
Tom de Vries [Tue, 17 Nov 2015 21:42:27 +0000 (21:42 +0000)]
Make fdump-tree-sccp-details more complete
2015-11-17 Tom de Vries <tom@codesourcery.com>
* tree-scalar-evolution.c (scev_const_prop): Dump details if replacing
uses of ssa_name with constant.
From-SVN: r230504
Tom de Vries [Tue, 17 Nov 2015 21:42:18 +0000 (21:42 +0000)]
Clear LOOP_CLOSED_SSA after pass_ccp
2015-11-17 Tom de Vries <tom@codesourcery.com>
* tree-ssa-ccp.c (do_ssa_ccp): Clear LOOP_CLOSED_SSA in loops state if
something changed.
From-SVN: r230503
Tom de Vries [Tue, 17 Nov 2015 21:42:09 +0000 (21:42 +0000)]
Add in_oacc_kernels_region in struct loop
2015-11-17 Tom de Vries <tom@codesourcery.com>
* cfgloop.h (struct loop): Add in_oacc_kernels_region field.
* omp-low.c (mark_loops_in_oacc_kernels_region): New function.
(expand_omp_target): Call mark_loops_in_oacc_kernels_region.
From-SVN: r230502
Bernd Schmidt [Tue, 17 Nov 2015 21:32:14 +0000 (21:32 +0000)]
Preliminary regrename patch for i386 ROP patch
* regrename.c (regrename_find_superclass): New function, code moved
from ...
(rename_chains): ... here. Call it.
* regrename.h (regrename_find_superclass): Declare.
From-SVN: r230501
Bernd Schmidt [Tue, 17 Nov 2015 21:24:38 +0000 (21:24 +0000)]
Ensure earlyclobbered operands go through a terminate_write stage.
* regrename.c (record_out_operands): Terminate earlyclobbered
operands here.
From-SVN: r230500
Bernd Schmidt [Tue, 17 Nov 2015 21:20:20 +0000 (21:20 +0000)]
re PR target/66785 (internal compiler error in record_operand_use)
PR target/66785
* regrename.c (record_operand_use): Keep track of failed operands
and stop appending if we see any.
* regrename.h (struct operand_rr_info): Add a failed field and shrink
n_chains to short.
From-SVN: r230499
Sandra Loosemore [Tue, 17 Nov 2015 21:07:15 +0000 (16:07 -0500)]
re PR c/48568 (Missing documentation for __attribute__((visibility ("protected"))) on variables.)
2015-11-17 Sandra Loosemore <sandra@codesourcery.com>
PR 48568
* doc/extend.texi (Common Function Attributes) [visibility]:
Add cross-references to matching variable and type attributes.
(Common Variable Attributes) [visibility]: Add missing entry.
From-SVN: r230498
David Malcolm [Tue, 17 Nov 2015 20:09:56 +0000 (20:09 +0000)]
Fix uninitialized src_range values for c_expr
gcc/c/ChangeLog:
* c-parser.c (set_c_expr_source_range): Bulletproof both
overloaded implementations against NULL expr->value.
(c_parser_braced_init): Set src_range for "ret" to a sane pair of
values.
(c_parser_unary_expression): Likewise when handling addresses of
labels.
(c_parser_postfix_expression): Likewise for statement expressions,
for __FUNCTION__, __PRETTY_FUNCTION_ and __func__ keywords, for
__builtin_va_arg, and for __builtin_offset_of.
(c_parser_postfix_expression_after_paren_type): Initialize expr's
src_range using the range of the braced initializer.
(c_parser_transaction_expression): Set src_range for "ret" to a
sane pair of values.
gcc/testsuite/ChangeLog:
* gcc.dg/plugin/diagnostic-test-expressions-1.c (vector): New
macro.
(test_braced_init): New function.
(test_statement_expression): New function.
(test_address_of_label): New function.
(test_transaction_expressions): New function.
(test_keywords): New function.
(test_builtin_va_arg): New function.
(test_builtin_offsetof): New function.
* lib/multiline.exp (_build_multiline_regex): Escape braces.
From-SVN: r230497
Jonathan Wakely [Tue, 17 Nov 2015 19:54:33 +0000 (19:54 +0000)]
PR libstdc++/66059 optimise _Build_index_tuple
PR libstdc++/66059
* include/std/utility (_Build_index_tuple): Optimise.
From-SVN: r230496
Steven G. Kargl [Tue, 17 Nov 2015 19:51:10 +0000 (19:51 +0000)]
primary.c (gfc_match_structure_constructor): Fix whitespace.
2015-11-17 Steven G. Kargl <kargl@gcc.gnu.org>
* primary.c (gfc_match_structure_constructor): Fix whitespace.
From-SVN: r230495
Richard Sandiford [Tue, 17 Nov 2015 18:56:52 +0000 (18:56 +0000)]
Simplify rs6000_builtin_vectorized_function
After the previous patches it's no longer necessary for
TARGET_BUILTIN_VECTORIZED_FUNCTION to return functions that
map to the vector optab of the original operation. We'll use
a vector form of the internal function instead.
Tested on powerpc64-linux-gnu.
gcc/
* config/rs6000/rs6000.c (rs6000_builtin_vectorized_function): Remove
entries that map directly to optabs.
From-SVN: r230494
Richard Sandiford [Tue, 17 Nov 2015 18:56:21 +0000 (18:56 +0000)]
Simplify ix86_builtin_vectorized_function
After the previous patches it's no longer necessary for
TARGET_BUILTIN_VECTORIZED_FUNCTION to return functions that
map to the vector optab of the original operation. We'll use
a vector form of the internal function instead.
Tested on x86_64-linux-gnu.
gcc/
* config/i386/i386.c (ix86_builtin_vectorized_function): Remove
entries that map directly to optabs.
From-SVN: r230493
Richard Sandiford [Tue, 17 Nov 2015 18:55:55 +0000 (18:55 +0000)]
Vectorize internal functions
This patch tries to vectorize built-in and internal functions as
internal functions first, falling back on the current built-in
target hooks otherwise.
This means that we'll automatically pick up vector versions of optabs
without the target having to implement any special hooks. E.g. we'll
use V4SF sqrt if the target defines a "sqrtv4sf2" optab. As well as
being simpler, it means that the target-independent code has more
idea what the vectorized function does.
Tested on x86_64-linux-gnu, aarch64-linux-gnu, arm-linux-gnu and
powerpc64-linux-gnu.
gcc/
* internal-fn.h (direct_internal_fn_info): Add vectorizable flag.
* internal-fn.c (direct_internal_fn_array): Update accordingly.
* tree-vectorizer.h (vectorizable_function): Delete.
* tree-vect-stmts.c: Include internal-fn.h.
(vectorizable_internal_function): New function.
(vectorizable_function): Inline into...
(vectorizable_call): ...here. Explicitly reject calls that read
from or write to memory. Try using an internal function before
falling back on the old vectorizable_function behavior.
From-SVN: r230492
Richard Sandiford [Tue, 17 Nov 2015 18:55:13 +0000 (18:55 +0000)]
Make builtin_vectorized_function take a combined_fn
This patch replaces the fndecl argument to builtin_vectorized_function
with a combined_fn and gets the vectoriser to call it for internal
functions too. The patch also moves vectorisation of machine-specific
built-ins to a new hook, builtin_md_vectorized_function.
Tested on x86_64-linux-gnu, aarch64-linux-gnu, arm-linux-gnu and
powerpc64-linux-gnu.
gcc/
* target.def (builtin_vectorized_function): Take a combined_fn (in
the form of an unsigned int) rather than a function decl.
(builtin_md_vectorized_function): New.
* targhooks.h (default_builtin_vectorized_function): Replace the
fndecl argument with an unsigned int.
(default_builtin_md_vectorized_function): Declare.
* targhooks.c (default_builtin_vectorized_function): Replace the
fndecl argument with an unsigned int.
(default_builtin_md_vectorized_function): New function.
* doc/tm.texi.in (TARGET_VECTORIZE_BUILTIN_MD_VECTORIZED_FUNCTION):
New hook.
* doc/tm.texi: Regenerate.
* tree-vect-stmts.c (vectorizable_function): Update call to
builtin_vectorized_function, also passing internal functions.
Call builtin_md_vectorized_function for target-specific builtins.
* config/aarch64/aarch64-protos.h
(aarch64_builtin_vectorized_function): Replace fndecl argument
with an unsigned int.
* config/aarch64/aarch64-builtins.c: Include case-cfn-macros.h.
(aarch64_builtin_vectorized_function): Update after above changes.
Use CASE_CFN_*.
* config/arm/arm-protos.h (arm_builtin_vectorized_function): Replace
fndecl argument with an unsigned int.
* config/arm/arm-builtins.c: Include case-cfn-macros.h
(arm_builtin_vectorized_function): Update after above changes.
Use CASE_CFN_*.
* config/i386/i386.c: Include case-cfn-macros.h
(ix86_veclib_handler): Take a combined_fn rather than a
built_in_function.
(ix86_veclibabi_svml, ix86_veclibabi_acml): Likewise. Use
mathfn_built_in rather than calling builtin_decl_implicit directly.
(ix86_builtin_vectorized_function) Update after above changes.
Use CASE_CFN_*.
* config/rs6000/rs6000.c: Include case-cfn-macros.h
(rs6000_builtin_vectorized_libmass): Replace fndecl argument
with a combined_fn. Use CASE_CFN_*. Use mathfn_built_in rather
than calling builtin_decl_implicit directly.
(rs6000_builtin_vectorized_function): Update after above changes.
Use CASE_CFN_*. Move BUILT_IN_MD to...
(rs6000_builtin_md_vectorized_function): ...this new function.
(TARGET_VECTORIZE_BUILTIN_MD_VECTORIZED_FUNCTION): Define.
From-SVN: r230491
Richard Sandiford [Tue, 17 Nov 2015 18:54:36 +0000 (18:54 +0000)]
Use IFN_SQRT in tree-vect-patterns.c
In practice all targets that can vectorise sqrt define the appropriate
sqrt<mode>2 optab. The only case where this isn't immediately obvious
is the libmass support in rs6000.c, but Mike Meissner said that it shouldn't
be exercised for sqrt.
This patch therefore uses the internal function interface instead of
going via the target hook.
Tested on x86_64-linux-gnu, aarch64-linux-gnu, arm-linux-gnueabi and
powerpc64-linux-gnu.
gcc/
* tree-vect-patterns.c: Include internal-fn.h.
(vect_recog_pow_pattern): Use IFN_SQRT instead of BUILT_IN_SQRT*.
From-SVN: r230490
Richard Sandiford [Tue, 17 Nov 2015 18:53:43 +0000 (18:53 +0000)]
Remove dead macros
Nothing uses these macros and removing them makes it more likely
that future code will use CASE_CFN_* instead.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* tree.h (BUILTIN_EXP10_P, BUILTIN_EXPONENT_P, BUILTIN_SQRT_P)
(BUILTIN_CBRT_P, BUILTIN_ROOT_P): Delete.
From-SVN: r230489
Richard Sandiford [Tue, 17 Nov 2015 18:51:55 +0000 (18:51 +0000)]
Extend tree-call-cdce to calls whose result is used
For -fmath-errno, builtins.c currently expands calls to sqrt to:
y = sqrt_optab (x);
if (y != y)
[ sqrt (x); or errno = EDOM; ]
The drawbacks of this are:
- the call to sqrt is protected by the result of the optab rather
than the input. It would be better to check __builtin_isless (x, 0),
like tree-call-cdce.c does.
- the branch isn't exposed at the gimple level and so gets little
high-level optimisation.
- we do this for log too, but for log a zero input produces
-inf rather than a NaN, and sets errno to ERANGE rather than EDOM.
This patch moves the code to tree-call-cdce.c instead, with the optab
operation being represented as an internal function. This means that
we can use the existing argument-based range checks rather than the
result-based checks and that we get more gimple optimisation of
the branch.
Previously the pass was only enabled by default at -O2 or above,
but the old builtins.c code was enabled at -O. The patch therefore
enables the pass at -O as well.
The previous patch to cfgexpand.c handled cases where functions
don't (or are assumed not to) set errno, so this patch makes
the builtins.c code dead.
Tested on x86_64-linux-gnu, aarch64-linux-gnu, arm-linux-gnueabi
and visium-elf (for the EDOM stuff).
gcc/
* builtins.c (expand_errno_check, expand_builtin_mathfn)
(expand_builtin_mathfn_2): Delete.
(expand_builtin): Remove handling of functions with
internal function equivalents.
* internal-fn.def (SET_EDOM): New internal function.
* internal-fn.h (set_edom_supported_p): Declare.
* internal-fn.c (expand_SET_EDOM): New function.
(set_edom_supported_p): Likewise.
* tree-call-cdce.c: Include builtins.h and internal-fn.h.
Rewrite comment at head of file.
(is_call_dce_candidate): Rename to...
(can_test_argument_range): ...this. Don't check gimple_call_lhs
or gimple_call_builtin_p here.
(edom_only_function): New function.
(shrink_wrap_one_built_in_call_with_conds): New function, split out
from...
(shrink_wrap_one_built_in_call): ...here.
(can_use_internal_fn, use_internal_fn): New functions.
(shrink_wrap_conditional_dead_built_in_calls): Call use_internal_fn
for calls that have an lhs.
(pass_call_cdce::gate): Remove optimize_function_for_speed_p check.
(pass_call_cdce::execute): Skip blocks that are optimized for size.
Check gimple_call_builtin_p here. Use can_use_internal_fn for
calls with an lhs.
* opts.c (default_options_table): Enable -ftree-builtin-call-cdce
at -O and above.
From-SVN: r230488
Richard Sandiford [Tue, 17 Nov 2015 18:51:20 +0000 (18:51 +0000)]
Short-cut generation of simple built-in functions
This patch short-circuits the builtins.c expansion code for a particular
gimple call if:
- the function has an associated internal function
- the target implements that internal function
- the call has no side effects
This allows a later patch to remove the builtins.c code, once calls with
side effects have been handled.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* builtins.h (called_as_built_in): Declare.
* builtins.c (called_as_built_in): Make external.
* internal-fn.h (expand_internal_call): Define a variant that
specifies the internal function explicitly.
* internal-fn.c (expand_load_lanes_optab_fn)
(expand_store_lanes_optab_fn, expand_ANNOTATE, expand_GOMP_SIMD_LANE)
(expand_GOMP_SIMD_VF, expand_GOMP_SIMD_LAST_LANE)
(expand_GOMP_SIMD_ORDERED_START, expand_GOMP_SIMD_ORDERED_END)
(expand_UBSAN_NULL, expand_UBSAN_BOUNDS, expand_UBSAN_VPTR)
(expand_UBSAN_OBJECT_SIZE, expand_ASAN_CHECK, expand_TSAN_FUNC_EXIT)
(expand_UBSAN_CHECK_ADD, expand_UBSAN_CHECK_SUB)
(expand_UBSAN_CHECK_MUL, expand_ADD_OVERFLOW, expand_SUB_OVERFLOW)
(expand_MUL_OVERFLOW, expand_LOOP_VECTORIZED)
(expand_mask_load_optab_fn, expand_mask_store_optab_fn)
(expand_ABNORMAL_DISPATCHER, expand_BUILTIN_EXPECT, expand_VA_ARG)
(expand_UNIQUE, expand_GOACC_DIM_SIZE, expand_GOACC_DIM_POS)
(expand_GOACC_LOOP, expand_GOACC_REDUCTION, expand_direct_optab_fn)
(expand_unary_optab_fn, expand_binary_optab_fn): Add an internal_fn
argument.
(internal_fn_expanders): Update prototype.
(expand_internal_call): Define a variant that specifies the
internal function explicitly. Use it to implement the previous
interface.
* cfgexpand.c (expand_call_stmt): Try to expand calls to built-in
functions as calls to internal functions.
From-SVN: r230487
Richard Sandiford [Tue, 17 Nov 2015 18:49:10 +0000 (18:49 +0000)]
Replace match.pd DEFINE_MATH_FNs with auto-generated lists
This patch autogenerates the operator lists for maths functions
like SQRT, adding an additional entry for internal functions.
E.g.:
(define_operator_list SQRT
BUILT_IN_SQRTF
BUILT_IN_SQRT
BUILT_IN_SQRTL
IFN_SQRT)
and:
(define_operator_list CABS
BUILT_IN_CABSF
BUILT_IN_CABS
BUILT_IN_CABSL
null)
(since there's no internal function for CABS).
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* Makefile.in (MOSTLYCLEANFILES): Add cfn-operators.pd.
(generated_files): Likewise.
(s-cfn-operators, cfn-operators.pd): New rules.
(s-match): Depend on cfn-operators.pd.
* gencfn-macros.c: Expand comment to describe -o behavior.
(print_define_operator_list): New function.
(main): Accept -o. Call print_define_operator_list.
* genmatch.c (main): Add the current directory to the include path.
* match.pd (DEFINE_MATH_FN): Delete. Include cfn-operators.pd
instead.
From-SVN: r230486
Richard Sandiford [Tue, 17 Nov 2015 18:48:23 +0000 (18:48 +0000)]
Add null identifiers to genmatch
This patch adds a null identifier that can never match anything and
can never be generated. It is only valid in operator lists and fors.
Later patches will add uses of it.
The idea is to allow operator lists for maths functions that have
four entries:
- float built-in
- double built-in
- long double built-in
- internal function
Not all maths functions have an associated internal function,
and for those the final operator will be "null". Any simplification
that tries to use a null substitution will be skipped.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* doc/match-and-simplify.texi: Document the "null" identifier.
* genmatch.c (id_base::NULL_ID): New kind.
(null_id): New variable.
(get_operator): Add a parameter that says whether null identifiers
are allowed.
(contains_id): New function.
(lower_for): Skip substitutions that would have a null_id in
either the match or the result.
(parser::parse_for): Allow the null identifier to be used.
(parser::parse_operator_list): Likewise.
(main): Initialize null_id.
From-SVN: r230485
Richard Sandiford [Tue, 17 Nov 2015 18:47:44 +0000 (18:47 +0000)]
Add genmatch support for internal functions
This patch makes genmatch match calls based on combined_fn rather
than built_in_function and extends the matching to internal functions.
It also uses fold_const_call to fold the calls to a constant, rather
than going through fold_builtin_n.
In order to slightly simplify the code and remove potential
ambiguity, the patch enforces lower case for tree codes
(foo->FOO_EXPR), caps for functions (no built_in_hypot->BUILT_IN_HYPOT)
and requires an exact match for user-defined identifiers. The first two
were already met in practice but there were a couple of cases where
operator lists were defined in one case and used in another.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* match.pd: Use HYPOT and COS rather than hypot and cos.
Use CASE_CFN_* macros. Guard log/exp folds with
SCALAR_FLOAT_TYPE_P.
* genmatch.c (internal_fn): New enum.
(fn_id::fn): Change to an unsigned int.
(fn_id::fn_id): Accept internal_fn too.
(add_builtin): Rename to...
(add_function): ...this and turn into a template.
(get_operator): Only try one variation if the original name fails.
Only add _EXPR if the original name was all lower case.
Try converting internal and built-in function names to their
CFN equivalents.
(expr::gen_transform): Use maybe_build_call_expr_loc for generic.
(dt_simplify::gen_1): Likewise.
(dt_node::gen_kids_1): Use gimple_call_combined_fn for gimple
and get_call_combined_fn for generic.
(dt_simplify::gen): Use combined_fn as the type of fn_ids.
(decision_tree::gen): Likewise.
(main): Use lower case in the strings for {VIEW_,}CONVERT[012].
Use add_function rather than add_builtin. Register internal
functions too.
* generic-match-head.c: Include case-cfn-macros.h.
* gimple-fold.c (replace_stmt_with_simplification): Use
gimple_call_combined_fn to test whether we can keep an
existing call.
* gimple-match.h (code_helper): Replace built_in_function
with combined_fn.
* gimple-match-head.c: Include fold-const-call.h, internal-fn.h
and case-fn-macros.h.
(gimple_resimplify1): Use fold_const_call.
(gimple_resimplify2, gimple_resimplify3): Likewise.
(build_call_internal, build_call): New functions.
(maybe_push_res_to_seq): Use them.
(gimple_simplify): Use fold_const_call. Set *rcode to a combined_fn
rather than a built-in function.
* tree.h (build_call_expr_internal_loc): Declare.
(maybe_build_call_expr_loc): Likewise.
* tree.c (build_call_expr_internal_loc_array): New function.
(maybe_build_call_expr_loc): Likewise.
From-SVN: r230484
Richard Sandiford [Tue, 17 Nov 2015 18:47:02 +0000 (18:47 +0000)]
Extend mathfn_built_in to handle combined_fn
This patch extends mathfn_built_in to handle combined_fn, but keeps the
old built_in_function interface around since it's a common case.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* builtins.h (mathfn_built_in): Add a variant that takes
a combined_fn.
* builtins.c: Include case-cfn-macros.h.
(CASE_MATHFN): Use CASE_CFN_*.
(CASE_MATHFN_REENT): Use CFN_ codes.
(mathfn_built_in_2, mathfn_built_in_1): Replace built_in_function
argument with a combined_fn.
(mathfn_built_in): Add a variant that takes a combined_fn.
(expand_builtin_int_roundingfn_2): Update callers accordingly.
(fold_builtin_sincos, fold_builtin_classify): Likewise.
From-SVN: r230483
Richard Sandiford [Tue, 17 Nov 2015 18:46:22 +0000 (18:46 +0000)]
Use combined_fn in tree-vect-patterns.c
Another patch to extend uses of built_in_function to combined_fn,
this time in tree-vect-patterns.c. The old code didn't handle the
long double pow variants, but I think that's because noone had a target
that would benefit rather than because the code would mishandle them.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* tree-vect-patterns.c: Include case-cfn-macros.h.
(vect_recog_pow_pattern): Use combined_fn instead of built-in codes.
From-SVN: r230482
Richard Sandiford [Tue, 17 Nov 2015 18:45:41 +0000 (18:45 +0000)]
Use combined_fn in tree-ssa-math-opts.c
Another patch to extend uses of built_in_function to combined_fn, this time
in tree-ssa-math-opts.c.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* tree-ssa-math-opts.c: Include case-cfn-macros.h.
(execute_cse_sincos_1): Use combined_fn instead of built-in codes.
(pass_cse_sincos::execute): Likewise.
From-SVN: r230481
Richard Sandiford [Tue, 17 Nov 2015 18:45:20 +0000 (18:45 +0000)]
Use combined_fn in tree-ssa-reassoc.c
Another patch to extend uses of built_in_function to combined_fn, this time
in tree-ssa-reassoc.c.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* tree-ssa-reassoc.c: Include case-cfn-macros.h.
(stmt_is_power_of_op): Use combined_fn instead of built-in codes.
(decrement_power, acceptable_pow_call): Likewise.
(attempt_builtin_copysign): Likewise.
From-SVN: r230480
Richard Sandiford [Tue, 17 Nov 2015 18:44:22 +0000 (18:44 +0000)]
Use combined_fn in tree-vrp.c
Another patch to extend uses of built_in_function to combined_fn, this time
in tree-vrp.c.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* tree-vrp.c: Include case-cfn-macros.h.
(extract_range_basic): Switch on combined_fn rather than handling
built-in functions and internal functions separately.
From-SVN: r230479
Richard Sandiford [Tue, 17 Nov 2015 18:43:42 +0000 (18:43 +0000)]
Make more use of combined_fn
This patch generalises fold-const.[hc] routines to use combined_fn
instead of built_in_function. It also updates gimple-ssa-backprop,c
since the update is simple and it avoids churn on the call to
negate_mathfn_p.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* fold-const.h (negate_mathfn_p): Take a combined_fn rather
than a built_in_function.
(tree_call_nonnegative_warnv_p): Take a combined_fn rather than
a function decl.
(integer_valued_real_call_p): Likewise.
* fold-const.c: Include case-cfn-macros.h
(negate_mathfn_p): Take a combined_fn rather than a built_in_function.
(negate_expr_p): Update accordingly.
(tree_call_nonnegative_warnv_p): Take a combined_fn rather than
a function decl.
(integer_valued_real_call_p): Likewise.
(tree_invalid_nonnegative_warnv_p): Update accordingly.
(integer_valued_real_p): Likewise.
* gimple-fold.c (gimple_call_nonnegative_warnv_p): Update call
to tree_call_nonnegative_warnv_p.
(gimple_call_integer_valued_real_p): Likewise
integer_valued_real_call_p.
* gimple-ssa-backprop.c: Include case-cfn-macros.h.
(backprop::process_builtin_call_use): Extend to combined_fn.
(strip_sign_op_1): Likewise.
(backprop::process_use): Don't check for built-in calls here.
(backprop::execute): Likewise.
(backprop::optimize_builtin_call): Update call to negate_mathfn_p.
From-SVN: r230478
Richard Sandiford [Tue, 17 Nov 2015 18:42:48 +0000 (18:42 +0000)]
Extend fold_const_call to combined_fn
This patch extends fold_const_call so that it can handle internal
as well as built-in functions.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* fold-const-call.h (fold_const_call): Replace built_in_function
arguments with combined_fn arguments.
* fold-const-call.c: Include case-cfn-macros.h.
(fold_const_call_ss, fold_const_call_cs, fold_const_call_sc)
(fold_const_call_cc, fold_const_call_sss, fold_const_call_ccc)
(fold_const_call_ssss, fold_const_call_1, fold_const_call): Replace
built_in_function arguments with combined_fn arguments.
* builtins.c (fold_builtin_sincos, fold_builtin_1, fold_builtin_2)
(fold_builtin_3): Update calls to fold_const_call.
From-SVN: r230477
Richard Sandiford [Tue, 17 Nov 2015 18:41:55 +0000 (18:41 +0000)]
Add gencfn-macros.c
This patch automatically generates case macros such as:
CASE_CFN_SQRT
for each {F,,L} floating-point built-in function and each {,L,LL,IMAX}
integer built-in function. The macros match the same built-in
functions as CASE_FLT_FN and CASE_INT_FN but in addition include
the associated internal function, if any.
The idea is to make sure that users of combined_fn don't need to know
which built-in functions have internal-function equivalents. If we add
a new function to internal-fn.def, all combined_fn users should pick it
up automatically.
The generator wants to use "hash_set <nofree_string_hash>",
so the patch follows hash_map in using the types given by the
traits as the key. This is a no-op for current users of hash_set.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* Makefile.in (HASH_TABLE_H): Add GGC_H.
(MOSTLYCLEANFILES, generated_files): Add case-fn-macros.h.
(s-case-cfn-macros, case-cfn-macros.h, build/gencfn-macros.o)
(build/gencfn-macros$(build_exeext): New rules.
(genprogerr): Add cfn-macros.
* hash-set.h (hash_set): Use the traits value_type as the key.
* gencfn-macros.c: New file.
From-SVN: r230476
Richard Sandiford [Tue, 17 Nov 2015 18:40:31 +0000 (18:40 +0000)]
Add internal bitcount functions
This patch adds internal function equivalents of all the INT_FN functions.
Unlike the math functions, these functions never set errno and the internal
functions should be exactly equivalent to the built-in ones. The reason
for defining the internal functions is so that we can extend the
functionality to other modes, in particular vector modes.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* internal-fn.def (DEF_INTERNAL_INT_FN): New macro.
(CLRSB, CLZ, CTZ, FFS, PARITY, POPCOUNT): New functions.
* builtins.c (associated_internal_fn): Handle them.
From-SVN: r230475
Richard Sandiford [Tue, 17 Nov 2015 18:39:02 +0000 (18:39 +0000)]
Add internal math functions
This patch adds internal functions for simple FLT_FN built-in functions,
in cases where an associated optab already exists. Unlike some of the
built-in functions, these internal functions never set errno.
LDEXP is an odd-one out in that its second operand is an integer.
All the others operate on uniform types.
The patch also adds a function to query the internal function associated
with a built-in function (if any), and another to test whether a given
gcall could be replaced by a call to an internal function on the current
target (as long as the caller deals with errno appropriately).
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* builtins.h (associated_internal_fn): Declare.
(replacement_internal_fn): Likewise.
* builtins.c: Include internal-fn.h
(associated_internal_fn, replacement_internal_fn): New functions.
* internal-fn.def (DEF_INTERNAL_FLT_FN): New macro.
(ACOS, ASIN, ATAN, COS, EXP, EXP10, EXP2, EXPM1, LOG, LOG10, LOG1P)
(LOG2, LOGB, SIGNIFICAND, SIN, SQRT, TAN, CEIL, FLOOR, NEARBYINT)
(RINT, ROUND, TRUNC, ATAN2, COPYSIGN, FMOD, POW, REMAINDER, SCALB)
(LDEXP): New functions.
* internal-fn.c: Include recog.h.
(unary_direct, binary_direct): New macros.
(expand_direct_optab_fn): New function.
(expand_unary_optab_fn): New macro.
(expand_binary_optab_fn): Likewise.
(direct_unary_optab_supported_p): Likewise.
(direct_binary_optab_supported_p): Likewise.
From-SVN: r230474
Richard Sandiford [Tue, 17 Nov 2015 18:37:45 +0000 (18:37 +0000)]
Add basic support for direct_optab internal functions
This patch adds a concept of internal functions that map directly to an
optab (here called "direct internal functions"). The function can only
be used if the associated optab can be used.
We currently have four functions like that:
- LOAD_LANES
- STORE_LANES
- MASK_LOAD
- MASK_STORE
so the patch converts them to the new infrastructure. These four
all need different types of optabs, but future patches will add
regular unary and binary ones.
In general we need one or two modes to decide whether an optab is
supported, depending on whether it's a convert_optab or not.
This in turn means that we need up to two types to decide whether
an internal function is supported. The patch records which types
are needed for each internal function, using -1 if the return type
should be used and N>=0 if the type of argument N should be used.
(LOAD_LANES and STORE_LANES are unusual in that both optab modes
come from the same array type.)
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* coretypes.h (tree_pair): New type.
* internal-fn.def (DEF_INTERNAL_OPTAB_FN): New macro. Use it
for MASK_LOAD, LOAD_LANES, MASK_STORE and STORE_LANES.
* internal-fn.h (direct_internal_fn_info): New structure.
(direct_internal_fn_array): Declare.
(direct_internal_fn_p, direct_internal_fn): New functions.
(direct_internal_fn_types, direct_internal_fn_supported_p): Declare.
* internal-fn.c (not_direct, mask_load_direct, load_lanes_direct)
(mask_store_direct, store_lanes_direct): New macros.
(direct_internal_fn_array) New array.
(get_multi_vector_move): Return the optab handler without asserting
that it is available.
(expand_LOAD_LANES): Rename to...
(expand_load_lanes_optab_fn): ...this and add an optab argument.
(expand_STORE_LANES): Rename to...
(expand_store_lanes_optab_fn): ...this and add an optab argument.
(expand_MASK_LOAD): Rename to...
(expand_mask_load_optab_fn): ...this and add an optab argument.
(expand_MASK_STORE): Rename to...
(expand_mask_store_optab_fn): ...this and add an optab argument.
(direct_internal_fn_types, direct_optab_supported_p)
(multi_vector_optab_supported_p, direct_internal_fn_supported_p)
(direct_internal_fn_supported_p): New functions.
(direct_mask_load_optab_supported_p): New macro.
(direct_load_lanes_optab_supported_p): Likewise.
(direct_mask_store_optab_supported_p): Likewise.
(direct_store_lanes_optab_supported_p): Likewise.
From-SVN: r230473
Richard Sandiford [Tue, 17 Nov 2015 18:34:47 +0000 (18:34 +0000)]
Add a combined_fn enum
I'm working on a patch series that needs to be able to treat built-in
functions and internal functions in a similar way. This patch adds a
new enum, combined_fn, that combines the two together. It also adds
utility functions for seeing which combined_fn (if any) is called by
a given CALL_EXPR or gcall.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
gcc/
* tree-core.h (internal_fn): Move immediately after the definition
of built_in_function.
(combined_fn): New enum.
* tree.h (as_combined_fn, builtin_fn_p, as_builtin_fn)
(internal_fn_p, as_internal_fn): New functions.
(get_call_combined_fn, combined_fn_name): Declare.
* tree.c (get_call_combined_fn): New function.
(combined_fn_name): Likewise.
* gimple.h (gimple_call_combined_fn): Declare.
* gimple.c (gimple_call_combined_fn): New function.
From-SVN: r230472
Jason Merrill [Tue, 17 Nov 2015 18:16:35 +0000 (13:16 -0500)]
re PR bootstrap/68346 (Bootstrap failure on i686-linux)
PR bootstrap/68346
* c-common.c (warn_tautological_cmp): Fold before checking for
constants.
From-SVN: r230471
Jason Merrill [Tue, 17 Nov 2015 18:16:29 +0000 (13:16 -0500)]
re PR bootstrap/68361 (Bootstrap failure with --enable-checking=release)
PR bootstrap/68361
* cvt.c (cp_convert_and_check): Use warning_sentinel to suppress
-Wparentheses.
From-SVN: r230470
Sandra Loosemore [Tue, 17 Nov 2015 18:15:34 +0000 (13:15 -0500)]
Fix PR number in ChangeLog
From-SVN: r230469
Martin Sebor [Tue, 17 Nov 2015 18:09:36 +0000 (18:09 +0000)]
PR c++/68308 - [6 Regression] ICE: tree check: expected integer_cst,
have var_decl in decompose, at tree.h:5105
gcc/
* cp/init.c (build_new_1): Check for expression constness
the right way.
testsuite/
* g++.dg/init/new46.C: New test.
From-SVN: r230468
Sandra Loosemore [Tue, 17 Nov 2015 18:06:01 +0000 (13:06 -0500)]
re PR other/56036 (Wrong indentation in multiple -O options explaination)
2015-11-17 Sandra Loosemore <sandra@codesourcery.com>
PR target/56036
* doc/invoke.texi (Option Summary): Add -mms-bitfields to x86
option list.
(x86 Options): Add -mms-bitfields and -mno-ms-bitfields. Move
discussion of the Microsoft structure layout details here from
its former home in extend.texi.
* doc/extend.texi (x86 Variable Attributes): Replace detailed
discussion with pointer to its new location. Add cross-reference
to corresponding type attributes.
(x86 Type Attributes): Add cross-references to command-line options
and variable attributes.
From-SVN: r230467
Cesar Philippidis [Tue, 17 Nov 2015 17:29:34 +0000 (09:29 -0800)]
priority_queue.c: New file.
libgomp/
* config/nvptx/priority_queue.c: New file.
From-SVN: r230466
Dominique d'Humieres [Tue, 17 Nov 2015 16:29:45 +0000 (17:29 +0100)]
re PR fortran/65751 (Bogus &L in error message)
2015-11-17 Dominique d'Humieres <dominiq@lps.ens.fr>
PR fortran/65751
* expr.c (gfc_check_pointer_assign): Fix error message.
* gfortran.dg/unlimited_polymorphic_2.f03: Update test.
From-SVN: r230465
Ilya Enkovich [Tue, 17 Nov 2015 13:22:40 +0000 (13:22 +0000)]
re PR middle-end/68134 (float64x1_t comparison ICE on aarch64-none-elf)
gcc/
PR middle-end/68134
* targhooks.c (default_get_mask_mode): Filter out
scalar modes returned by mode_for_vector.
gcc/testsuite/
PR middle-end/68134
* gcc.dg/pr68134.c: New test.
From-SVN: r230463
Kyrylo Tkachov [Tue, 17 Nov 2015 13:20:08 +0000 (13:20 +0000)]
[ARM] PR 68143 Properly update memory offsets when expanding setmem
PR target/68143
* config/arm/arm.c (arm_block_set_unaligned_vect): Keep track of
offset from dstbase and use it appropriately in
adjust_automodify_address.
(arm_block_set_aligned_vect): Likewise.
* gcc.c-torture/execute/pr68143_1.c: New test.
From-SVN: r230462
Uros Bizjak [Tue, 17 Nov 2015 11:00:08 +0000 (12:00 +0100)]
* ChangeLog: Add missing entry.
From-SVN: r230460
Eric Botcazou [Tue, 17 Nov 2015 10:54:13 +0000 (10:54 +0000)]
t-visium (MULTILIB_OPTIONS): Add muser-mode.
* config/visium/t-visium (MULTILIB_OPTIONS): Add muser-mode.
(MULTILIB_DIRNAMES): Adjust accordingly.
From-SVN: r230459