gcc:
2014-01-29 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
* config/rs6000/rs6000.c (altivec_expand_vec_perm_const): Use
CODE_FOR_altivec_vmrg*_direct rather than CODE_FOR_altivec_vmrg*.
* config/rs6000/vsx.md (vsx_mergel_<mode>): Adjust for
-maltivec=be with LE targets.
(vsx_mergeh_<mode>): Likewise.
* config/rs6000/altivec.md (UNSPEC_VMRG[HL]_DIRECT): New
unspecs.
(mulv8hi3): Use gen_altivec_vmrg[hl]w_direct.
(altivec_vmrghb): Replace with define_expand and new
*altivec_vmrghb_internal insn; adjust for -maltivec=be with LE
targets.
(altivec_vmrghb_direct): New define_insn.
(altivec_vmrghh): Replace with define_expand and new
*altivec_vmrghh_internal insn; adjust for -maltivec=be with LE
targets.
(altivec_vmrghh_direct): New define_insn.
(altivec_vmrghw): Replace with define_expand and new
*altivec_vmrghw_internal insn; adjust for -maltivec=be with LE
targets.
(altivec_vmrghw_direct): New define_insn.
(*altivec_vmrghsf): Adjust for endianness.
(altivec_vmrglb): Replace with define_expand and new
*altivec_vmrglb_internal insn; adjust for -maltivec=be with LE
targets.
(altivec_vmrglb_direct): New define_insn.
(altivec_vmrglh): Replace with define_expand and new
*altivec_vmrglh_internal insn; adjust for -maltivec=be with LE
targets.
(altivec_vmrglh_direct): New define_insn.
(altivec_vmrglw): Replace with define_expand and new
*altivec_vmrglw_internal insn; adjust for -maltivec=be with LE
targets.
(altivec_vmrglw_direct): New define_insn.
(*altivec_vmrglsf): Adjust for endianness.
(vec_widen_umult_hi_v16qi): Use gen_altivec_vmrghh_direct.
(vec_widen_umult_lo_v16qi): Use gen_altivec_vmrglh_direct.
(vec_widen_smult_hi_v16qi): Use gen_altivec_vmrghh_direct.
(vec_widen_smult_lo_v16qi): Use gen_altivec_vmrglh_direct.
(vec_widen_umult_hi_v8hi): Use gen_altivec_vmrghw_direct.
(vec_widen_umult_lo_v8hi): Use gen_altivec_vmrglw_direct.
(vec_widen_smult_hi_v8hi): Use gen_altivec_vmrghw_direct.
(vec_widen_smult_lo_v8hi): Use gen_altivec_vmrglw_direct.
gcc/testsuite:
2014-01-29 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
* gcc.dg/vmx/merge-be-order.c: New.
* gcc.dg/vmx/merge.c: New.
* gcc.dg/vmx/merge-vsx-be-order.c: New.
* gcc.dg/vmx/merge-vsx.c: New.
From-SVN: r207262
2014-01-29 Hristian Kirtchev <kirtchev@adacore.com>
* sem_prag.adb (Check_Missing_Part_Of): List all values of
State_Space_Kind for readability reasons. Do not emit an error on
a private item when the enclosing package lacks aspect/pragma
Abstract_State. Do not emit an error on a private package
instantiation when the corresponding generic template lacks
visible state.
(Has_Visible_State): New routine.
* sem_util.adb (Find_Placement_In_State_Space): The visible
declarations of any kind of child units in general act as proper
placement location.
From-SVN: r207261
2014-01-29 Robert Dewar <dewar@adacore.com>
* expander.adb: Minor reformatting.
2014-01-29 Javier Miranda <miranda@adacore.com>
* exp_ch3.adb (Predefined_Primitive_Bodies): Adding documentation to
previous patch.
From-SVN: r207257
2014-01-29 Javier Miranda <miranda@adacore.com>
* exp_ch3.adb (Predefined_Primitive_Bodies): Complete the code
that checks if an interface types defines the predefined "="
function because the compiler was erroneously not generating the
predefined "=" primitive as soon as the name of some interface
primitive is "=" (formals were not checked).
2014-01-29 Ed Schonberg <schonberg@adacore.com>
* expander.adb (Expander): In GNATprove mode, do not process
transient scopes: they are in general not created in this mode,
and an attempt to examine them will lead to constraint errors when
processing configuration pragmas that have analyzable expressions.
From-SVN: r207256
2014-01-29 Vincent Celier <celier@adacore.com>
* clean.adb (Gnatclean): Fail if main project is an aggregate
project or if there is an aggregate library project in the
project tree.
* gnatcmd.adb: Fail if the main project is an aggregate project
or if there is an aggegate library project in the project tree.
* make.adb (Initialize): : Fail if main project is an aggregate
project or if there is an aggregate library project in the
project tree.
* makeutl.ads (Aggregate_Libraries_In): New Boolean function.
* prj-makr.adb (Initialize): Fail if the main project is an
aggregate project or an aggregate library project.
From-SVN: r207255
2014-01-29 Vincent Celier <celier@adacore.com>
* prj-part.adb (Check_Import_Aggregate): New procedure
to check if an imported project is an aggregate project.
(Parse_Single_Project): Call Check_Import_Aggregate
* projects.texi: Document that aggregate projects cannot be
extended or imported.
From-SVN: r207254
2014-01-29 Hristian Kirtchev <kirtchev@adacore.com>
* einfo.adb Flag264 is now unused.
(Has_Body_References): Removed.
(Set_Has_Body_References): Removed.
(Write_Entity_Flags): Remove the output for flag Has_Body_References.
* einfo.ads Update the comment on usage of attribute
Body_References. Remove attribute Has_Body_References and its
usage in nodes.
(Has_Body_References): Removed along with pragma Inline.
(Set_Has_Body_References): Removed along with pragma Inline.
* sem_prag.adb (Analyze_Global_Item): Move the call to
Record_Possible_Body_Reference in the state related checks
section. Add a comment intended function.
(Analyze_Input_Output): Move the call to Record_Possible_Body_Reference
in the state related checks section. Add a comment intended function.
(Analyze_Refinement_Clause): Cleanup the illegal body reference
reporting. Add a comment on timing of error reporting.
(Record_Possible_Body_Reference): Reimplement the routine.
2014-01-29 Vincent Celier <celier@adacore.com>
* makeutl.adb (Mains.Find_File_Add_Extension): Only look for specs for
unit-based languages.
(Mains.Complete_Mains.Do_Complete): Use the source file project
tree when calling Find_File_Add_Extension. Use the correct
project name when reporting an error.
From-SVN: r207252
* aspects.adb Add an entry for aspect Part_Of in table
Canonical_Aspect.
* aspects.ads Add an entry for aspect Part_Of in tables Aspect_Id,
Aspect_Argument, Aspect_Names and Aspect_Delay.
* atree.h Define Elist9.
* atree.adb (Elist9): New routine.
(Set_Elist9): New routine.
* atree.ads (Elist9): New routine.
(Set_Elist9): New routine.
* einfo.adb Add Part_Of_Constituents and Encapsulating_State to
the list of node usage. Remove Refined_State from the list of
node usage.
(Encapsulating_State): New routine.
(Get_Pragma):
Handle pragma Part_Of; (Part_Of_Constituents): New routine.
(Refined_State): Removed.
(Set_Encapsulating_State): New routine.
(Set_Part_Of_Constituents): New routine.
(Set_Refined_State): Removed.
(Write_Field9_Name): Add an entry
for Part_Of_Constituents (Write_Field10_Name): Add an entry for
Encapsulating_State. Remove the entry for Refined_State.
* einfo.ads Add new attributes Encapsulating_State
and Part_Of_Constituents alond with their usage in
entities. Remove attribute Refined_State along with its
usage in entities.
(Encapsulating_State): New routine and
pragma Inline. (Get_Pragma): Update the comment on usage.
(Part_Of_Constituents): New routine and pragma Inline.
(Refined_State): Removed along with pragma Inline.
(Set_Encapsulating_State): New routine and pragma Inline.
(Set_Part_Of_Constituents): New routine and pragma Inline.
(Set_Refined_State): Removed along with pragma Inline.
* par-prag.adb Pragma Part_Of does not need any special processing
by the parser.
* sem_ch3.adb (Analyze_Declarations): Remove local variables
Body_Id and Prag. Call separate routines to analyze the
contract of a package [body].
(Analyze_Object_Contract):
Update the comment on usage. Remove local variables
Items and Nam. Use Get_Pragma rather than traversing the
classification list. Verify whether the lack of indicator
Part_Of agrees with the placement of the variable in state space.
(Analyze_Object_Declaration): Initialize the encapsulating state
of a variable. (Requires_State_Refinement): Moved to sem_util.
* sem_ch7.adb (Analyze_Package_Body_Contract): New routine.
(Analyze_Package_Contract): New routine.
* sem_ch7.ads (Analyze_Package_Body_Contract): New routine.
(Analyze_Package_Contract): New routine.
* sem_ch10.adb (Decorate_State): Initialize the encapsulating
state and Part_Of constituents.
* sem_ch13.adb (Analyze_Aspect_Specifications):
Add processing for aspect Part_Of. Update all
calls to Decorate_Delayed_Aspect_And_Pragma.
(Check_Aspect_At_Freeze_Point): Aspect Part_Of does
not need any special processing at freeze time.
(Decorate_Delayed_Aspect_And_Pragma): Renamed to
Decorate_Aspect_And_Pragma. Add formal parameter Delayed and
update the associated comment.
* sem_prag.adb Add an entry for pragma Part_Of in table Sig_Flags.
(Analyze_Abstract_State): Add new global variable State_Id. Remove
local constants Errors and Loc. Remove local variables Is_Null
and State_Nam. Create the entity of the abstract state on the
spot, before all remaining checks are performed. Verify that a
missing Part_Of option agrees with the placement of the abstract
state within the state space.
(Analyze_Depends_In_Decl_Part):
Add new global variables Constits_Seen and States_Seen. Check
that a state and a corresponding constituent do not appear
in pragma [Refined_]Depends.
(Analyze_Global_In_Decl_Part):
Add new global variables Constits_Seen and States_Seen. Check
that a state and a corresponding constituent do not appear
in pragma [Refined_]Global.
(Analyze_Global_Item):
Remove the now obsolete code that deals with Part_Of.
Add the entity of the global item to the list of processed
items. (Analyze_Initializes_In_Decl_Part): Add new global
variables Constits_Seen and States_Seen. Check that a state
and a corresponding constituent do not appear in pragma
Initializes.
(Analyze_Initialization_Item): Add the entity
of the initialization item to the list of processed items.
(Analyze_Input_Item): Add the entity of the initialization
item to the list of processed items.
(Analyze_Input_Output):
Remove the now obsolete code that deals with Part_Of. Add the
entity of the input/output to the list of processed items.
(Analyze_Part_Of): New routine.
(Analyze_Part_Of_Option): Remove
local constant Par_State. Add local constant Encaps and local
variables Encaps_Id and Legal. Use Analyze_Part of to analyze
the option. Turn the related state into a Part_Of constituent
if the option is legal.
(Analyze_Pragma): Add processing
for pragma Part_Of.
(Analyze_Refined_State_In_Decl_Part):
Remove global constants Pack_Body and Spec_Id. Remove
global variables Abstr_States and Hidden_States. Add new
global variables Available_States, Body_Id, Body_States and
Spec_Id. Add new local constant Body_Decl. Reimplement the
logic that extracts the states available for refinement from
the related package and the body hidden states of the said
package.
(Analyze_Refinement_Clause): Add local variable Part_Of_Constits.
(Check_Applicable_Policy): Alphabetize body.
(Check_Dependency_Clause): Replace Refined_State
with Encapsulating_State.
(Check_Matching_Constituent):
Reimplement the logic that determines whether an item is a valid
/ invalid constituent of the current refined state. Return when
a construct does not denote a valid abstract state. Extract the
list of Part_Of constituents for further analysis. Check that all
Part_Of constituents of a state have been used in its refinement.
(Check_Matching_State): Update the comment on usage. Operate
on the list of available states.
(Check_Missing_Part_Of): New routine.
(Check_Refined_Global_Item): Replace Refined_State
with Encapsulating_State.
(Check_State_And_Constituent_Use): New routine.
(Create_Abstract_State): New routine.
(Is_Matching_Input): Replace Refined_State with Encapsulating_State.
(Is_Part_Of): Removed.
(Collect_Body_States): New routine.
(Collect_Constituent): Replace Refined_State with Encapsulating_State.
(Collect_Hidden_States): Removed.
(Report_Unrefined_States): Change the profile of the procedure along
with the comment on usage.
(Report_Unused_Constituents): New routine.
(Report_Unused_Hidden_States): Removed.
(Report_Unused_States): New routine.
* sem_prag.ads (Check_Missing_Part_Of): New routine.
* sem_util.adb (Add_Contract_Item): Pragma Part_Of can now
appear in the classification pragmas of a package instantiation
or a variable.
(Find_Placement_In_State_Space): New routine.
(Is_Child): Removed.
(Is_Child_Or_Sibling): Remove formal
parameter Private_Child. Remove the private child checks.
(Requires_State_Refinement): Moved from sem_ch3.
* sem_util.ads Add new type State_Space_Kind along with
comment on its usage and values.
(Add_Contract_Item): Update the comment on usage.
(Find_Body_Discriminal): Alphabetize spec.
(Find_Placement_In_State_Space): New routine.
(Is_Child_Or_Sibling): Remove formal parameter Private_Child
and update the comment on usage.
(Requires_State_Refinement): Moved from sem_ch3.
* sinfo.ads: Update the documentation of N_Contract.
* snames.ads-tmpl The predefined name for Part_Of is now used
to denote a pragma. Add Pragma_Id for Part_Of.
From-SVN: r207251
2014-01-29 Ed Schonberg <schonberg@adacore.com>
* checks.adb: minor clarification.
* sem_ch7.adb (Declare_Inherited_Private_Subprograms): Limit
search for primitive operations to the entities that immediately
follow the type declaration.
From-SVN: r207248
2014-01-29 Tristan Gingold <gingold@adacore.com>
* exp_ch9.adb (Build_Protected_Entry): Do not call
Complete_Entry_Body anymore.
* rtsfind.ads (RE_Complete_Single_Entry_Body): Remove.
* s-tposen.ads, s-tposen.adb (Complete_Single_Entry_Body): Remove.
2014-01-29 Pierre-Marie Derodat <derodat@adacore.com>
* s-os_lib.adb, s-os_lib.ads (Normalize_Pathname): Return an empty
string when the Name input bigger than allowed. Adapt the function
specification.
2014-01-29 Ed Schonberg <schonberg@adacore.com>
* checks.adb (Install_Null_Excluding_Check): Do not emit warning
if expression is within a case_expression of if_expression.
From-SVN: r207247
2014-01-29 Tristan Gingold <gingold@adacore.com>
* exp_ch9.adb (Is_Exception_Safe): Return true if no exceptions.
2014-01-29 Yannick Moy <moy@adacore.com>
* inline.ads (Pending_Body_Info): Add SPARK_Mode and
SPARK_Mode_Pragma components to be able to analyze generic
instance.
* sem_ch12.adb (Analyze_Package_Instantiation,
Inline_Instance_Body, Need_Subprogram_Instance_Body,
Load_Parent_Of_Generic): Pass in SPARK_Mode from instantiation
for future analysis of the instance.
(Instantiate_Package_Body,
Instantiate_Subprogram_Body, Set_Instance_Inv): Set SPARK_Mode
from instantiation to analyze the instance.
From-SVN: r207244
2014-01-29 Thomas Quinot <quinot@adacore.com>
* sem_ch4.adb (Find_Component_In_Instance): Update comment.
2014-01-29 Ed Schonberg <schonberg@adacore.com>
* exp_util.adb (Build_Task_Image_Prefix): Indicate that the
resulting string is an internal entity. and thus requires no
initialization. This is relevant when Initialize_ Scalars is
enabled, because the resultant spurious initialization may lead to
secondary stack anomalies that produce a mangled name for a task.
2014-01-29 Yannick Moy <moy@adacore.com>
* sem_ch6.adb (Analyze_Subprogram_Body_Helper): SPARK_Mode
not inherited from spec anymore. Check consistency
rules after processing of declarations.
* sem_ch7.adb (Analyze_Package_Body_Helper): SPARK_Mode not inherited
from spec anymore. Check consistency rules after processing of
declarations.
(Analyze_Package_Declaration): Set SPARK_Mode only for non-generic
packages.
* sem_prag.adb (Analyze_Pragma/Pragma_SPARK_Mode): Implement new
consistency rules.
From-SVN: r207242
PR middle-end/59917
PR tree-optimization/59920
* tree.c (build_common_builtin_nodes): Remove
__builtin_setjmp_dispatcher initialization.
* omp-low.h (make_gimple_omp_edges): Add a new int * argument.
* profile.c (branch_prob): Use gsi_start_nondebug_after_labels_bb
instead of gsi_after_labels + manually skipping debug stmts.
Don't ignore bbs with BUILT_IN_SETJMP_DISPATCHER, instead
ignore bbs with IFN_ABNORMAL_DISPATCHER.
* tree-inline.c (copy_edges_for_bb): Remove
can_make_abnormal_goto argument, instead add abnormal_goto_dest
argument. Ignore computed_goto_p stmts. Don't call
make_abnormal_goto_edges. If a call might need abnormal edges
for non-local gotos, see if it already has an edge to
IFN_ABNORMAL_DISPATCHER or if it is IFN_ABNORMAL_DISPATCHER
with true argument, don't do anything then, otherwise add
EDGE_ABNORMAL from the call's bb to abnormal_goto_dest.
(copy_cfg_body): Compute abnormal_goto_dest, adjust copy_edges_for_bb
caller.
* gimple-low.c (struct lower_data): Remove calls_builtin_setjmp.
(lower_function_body): Don't emit __builtin_setjmp_dispatcher.
(lower_stmt): Don't set data->calls_builtin_setjmp.
(lower_builtin_setjmp): Adjust comment.
* builtins.def (BUILT_IN_SETJMP_DISPATCHER): Remove.
* tree-cfg.c (found_computed_goto): Remove.
(factor_computed_gotos): Remove.
(make_goto_expr_edges): Return bool, true for computed gotos.
Don't call make_abnormal_goto_edges.
(build_gimple_cfg): Don't set found_computed_goto, don't call
factor_computed_gotos.
(computed_goto_p): No longer static.
(make_blocks): Don't set found_computed_goto.
(get_abnormal_succ_dispatcher, handle_abnormal_edges): New functions.
(make_edges): If make_goto_expr_edges returns true, push bb
into ab_edge_goto vector, for stmt_can_make_abnormal_goto calls
instead of calling make_abnormal_goto_edges push bb into ab_edge_call
vector. Record mapping between bbs and OpenMP regions if there
are any, adjust make_gimple_omp_edges caller. Call
handle_abnormal_edges.
(make_abnormal_goto_edges): Remove.
* tree-cfg.h (make_abnormal_goto_edges): Remove.
(computed_goto_p, get_abnormal_succ_dispatcher): New prototypes.
* internal-fn.c (expand_ABNORMAL_DISPATCHER): New function.
* builtins.c (expand_builtin): Don't handle
BUILT_IN_SETJMP_DISPATCHER.
* internal-fn.def (ABNORMAL_DISPATCHER): New.
* omp-low.c (make_gimple_omp_edges): Add region_idx argument, when
filling *region also set *region_idx to (*region)->entry->index.
* gcc.dg/pr59920-1.c: New test.
* gcc.dg/pr59920-2.c: New test.
* gcc.dg/pr59920-3.c: New test.
* c-c++-common/gomp/pr59917-1.c: New test.
* c-c++-common/gomp/pr59917-2.c: New test.
From-SVN: r207231
2014-01-29 Bingfeng Mei <bmei@broadcom.com>
* doc/md.texi: Mention that a target shouldn't implement
vec_widen_(s|u)mul_even/odd pair if it is less efficient
than hi/lo pair.
From-SVN: r207228
PR tree-optimization/59594
* tree-vect-data-refs.c (vect_analyze_data_ref_accesses): Sort
a copy of the datarefs vector rather than the vector itself.
* gcc.dg/vect/no-vfa-vect-depend-2.c: New test.
* gcc.dg/vect/no-vfa-vect-depend-3.c: New test.
* gcc.dg/vect/pr59594.c: New test.
From-SVN: r207225
2014-01-28 Paul Thomas <pault@gcc.gnu.org>
PR fortran/59414
* trans-stmt.c (gfc_trans_allocate): Before the pointer
assignment to transfer the source _vptr to a class allocate
expression, the final class reference should be exposed. The
tail that includes the _data and array references is stored.
This reduced expression is transferred to 'lhs' and the _vptr
added. Then the tail is restored to the allocate expression.
2014-01-28 Paul Thomas <pault@gcc.gnu.org>
PR fortran/59414
* gfortran.dg/allocate_class_3.f90 : New test
From-SVN: r207204
PR c++/53756
gcc/
* dwarf2out.c (auto_die): New static.
(gen_type_die_with_usage): Handle C++1y 'auto'.
(gen_subprogram_die): If in-class DIE had 'auto', emit type again
on definition.
gcc/cp/
* mangle.c (write_unqualified_name): Handle operator auto.
From-SVN: r207197
The .code16gcc directive was added to binutils back in 1999:
---
'.code16gcc' provides experimental support for generating 16-bit code
from gcc, and differs from '.code16' in that 'call', 'ret', 'enter',
'leave', 'push', 'pop', 'pusha', 'popa', 'pushf', and 'popf'
instructions default to 32-bit size. This is so that the stack pointer
is manipulated in the same way over function calls, allowing access to
function parameters at the same stack offsets as in 32-bit mode.
'.code16gcc' also automatically adds address size prefixes where
necessary to use the 32-bit addressing modes that gcc generates.
---
It encodes 32-bit assembly instructions generated by GCC in 16-bit format
so that GCC can be used to generate 16-bit instructions. To do that, the
.code16gcc directive must be placed at the very beginning of the assembly
code. This patch adds -m16 to x86 backend by:
1. Add -m16 and make it mutually exclusive with -m32, -m64 and -mx32.
2. Treat -m16 like -m32 so that --32 is passed to assembler.
3. Output .code16gcc at the very beginning of the assembly code.
4. Turn off 64-bit ISA when -m16 is used.
PR target/59672
* config/i386/gnu-user64.h (SPEC_32): Add "m16|" to "m32".
(SPEC_X32): Likewise.
(SPEC_64): Likewise.
* config/i386/i386.c (ix86_option_override_internal): Turn off
OPTION_MASK_ISA_64BIT, OPTION_MASK_ABI_X32 and OPTION_MASK_ABI_64
for TARGET_16BIT.
(x86_file_start): Output .code16gcc for TARGET_16BIT.
* config/i386/i386.h (TARGET_16BIT): New macro.
(TARGET_16BIT_P): Likewise.
* config/i386/i386.opt: Add m16.
* doc/invoke.texi: Document -m16.
From-SVN: r207196
gcc/ChangeLog
* input.c (location_get_source_line): Bail out on when line number
is zero, and test the return value of
lookup_or_add_file_to_cache_tab.
gcc/testsuite/ChangeLog
* c-c++-common/cpp/warning-zero-location.c: New test.
* c-c++-common/cpp/warning-zero-location-2.c: Likewise.
Signed-off-by: Dodji Seketeli <dodji@seketeli.org>
From-SVN: r207195
