Balaji V. Iyer [Mon, 10 Feb 2014 16:56:54 +0000 (16:56 +0000)]
re PR target/59691 (cilk-plus run failures on non-sse processors)
Fix for PR target/59691.
+2014-02-10 Balaji V. Iyer <balaji.v.iyer@intel.com>
+
+ PR target/59691
+ * runtime/config/x86/os-unix-sysdep.c (__builtin_cpu_supports): New
+ function.
+ (restore_x86_fp_state): Added a check if the cpu supports the
+ instruction before emitting it.
+ (sysdep_save_fp_ctrl_state): Likewise.
+
Jeff Law [Mon, 10 Feb 2014 16:25:44 +0000 (09:25 -0700)]
re PR middle-end/52306 (ICE in cselib_record_set, at cselib.c:2158)
PR middle-end/52306
* reload1.c (emit_input_reload_insns): Do not create invalid RTL
when changing the SET_DEST of a prior insn to avoid an input
reload.
PR middle-end-52306
* gcc.c-torture/compile/pr52306.c: New test.
Richard Biener [Mon, 10 Feb 2014 13:51:23 +0000 (13:51 +0000)]
re PR tree-optimization/60115 (wrong code at -O3 on x86_64-linux-gnu)
2014-02-10 Richard Biener <rguenther@suse.de>
PR tree-optimization/60115
* tree-eh.c (tree_could_trap_p): Unify TARGET_MEM_REF and
MEM_REF handling. Properly verify that the accesses are not
out of the objects bound.
Paul Thomas [Sun, 9 Feb 2014 20:50:21 +0000 (20:50 +0000)]
re PR fortran/57522 ([F03] ASSOCIATE construct creates array descriptor with incorrect stride for derived type array component)
2014-02-09 Paul Thomas <pault@gcc.gnu.org>
PR fortran/57522
* resolve.c (resolve_assoc_var): Set the subref_array_pointer
attribute for the 'associate-name' if necessary.
* trans-stmt.c (trans_associate_var): If the 'associate-name'
is a subref_array_pointer, assign the element size of the
associate variable to 'span'.
2014-02-09 Paul Thomas <pault@gcc.gnu.org>
PR fortran/57522
* gfortran.dg/associated_target_5.f03 : New test
Paul Thomas [Sun, 9 Feb 2014 19:45:06 +0000 (19:45 +0000)]
re PR fortran/59026 (ELEMENTAL procedure with VALUE arguments emits wrong code)
2014-02-09 Paul Thomas <pault@gcc.gnu.org>
PR fortran/59026
* trans-expr.c (gfc_conv_procedure_call): Pass the value of the
actual argument to a formal argument with the value attribute
in an elemental procedure.
2014-02-09 Paul Thomas <pault@gcc.gnu.org>
PR fortran/59026
* gfortran.dg/elemental_by_value_1.f90 : New test
Janus Weil [Sat, 8 Feb 2014 21:32:25 +0000 (22:32 +0100)]
re PR fortran/58470 ([OOP] ICE on invalid with FINAL procedure and type extension)
2014-02-08 Janus Weil <janus@gcc.gnu.org>
Mikael Morin <mikael.morin@gcc.gnu.org>
PR fortran/58470
* class.c (generate_finalization_wrapper): Assert that proc_tree has
been set in gfc_resolve_finalizers.
* resolve.c (resolve_fl_derived0): Remove unnecessary call to
gfc_is_finalizable.
Jakub Jelinek [Sat, 8 Feb 2014 14:52:39 +0000 (15:52 +0100)]
tree-vect-data-refs.c (vect_analyze_data_refs): For clobbers not at the end of datarefs vector use ordered_remove to avoid...
* tree-vect-data-refs.c (vect_analyze_data_refs): For clobbers
not at the end of datarefs vector use ordered_remove to avoid
reordering datarefs vector.
Jakub Jelinek [Sat, 8 Feb 2014 09:10:14 +0000 (10:10 +0100)]
re PR c/59984 (OpenMP pragma makes loop incorrect)
PR c/59984
* gimplify.c (gimplify_bind_expr): In ORT_SIMD region
mark local addressable non-static vars as GOVD_PRIVATE
instead of GOVD_LOCAL.
* omp-low.c (lower_omp_for): Move gimple_bind_vars
and BLOCK_VARS of gimple_bind_block to new_stmt rather
than copying them.
Jakub Jelinek [Sat, 8 Feb 2014 09:09:01 +0000 (10:09 +0100)]
re PR middle-end/60092 (posix_memalign not recognized to derive alias and alignment info)
PR middle-end/60092
* tree-ssa-ccp.c (surely_varying_stmt_p): Don't return true
if TYPE_ATTRIBUTES (gimple_call_fntype ()) contain
assume_aligned or alloc_align attributes.
(bit_value_assume_aligned): Add ATTR, PTRVAL and ALLOC_ALIGN
arguments. Handle also assume_aligned and alloc_align attributes.
(evaluate_stmt): Adjust bit_value_assume_aligned caller.
Handle calls to functions with assume_aligned or alloc_align
attributes.
* doc/extend.texi: Document assume_aligned and alloc_align
attributes.
c-family/
* c-common.c (handle_alloc_size_attribute): Use tree_fits_uhwi_p
and tree_to_uhwi.
(handle_alloc_align_attribute, handle_assume_aligned_attribute): New
functions.
(c_common_attribute_table): Add alloc_align and assume_aligned
attributes.
testsuite/
* gcc.dg/attr-alloc_align-1.c: New test.
* gcc.dg/attr-alloc_align-2.c: New test.
* gcc.dg/attr-alloc_align-3.c: New test.
* gcc.dg/attr-assume_aligned-1.c: New test.
* gcc.dg/attr-assume_aligned-2.c: New test.
* gcc.dg/attr-assume_aligned-3.c: New test.
Jan Hubicka [Fri, 7 Feb 2014 23:49:18 +0000 (00:49 +0100)]
varpool.c: Include pointer-set.h.
* varpool.c: Include pointer-set.h.
(varpool_remove_unreferenced_decls): Variables in other partitions
will not be output; be however careful to not lose information
about partitioning.
s390.c (s390_restore_gprs_from_fprs): Add REG_CFA_RESTORE notes to each restore.
gcc/
* config/s390/s390.c (s390_restore_gprs_from_fprs): Add REG_CFA_RESTORE
notes to each restore. Also add REG_CFA_DEF_CFA when restoring %r15.
(s390_optimize_prologue): Don't clear RTX_FRAME_RELATED_P. Update the
REG_CFA_RESTORE list when deciding not to restore a register.
gcc/
* config/s390/s390.c: Include tree-pass.h and context.h.
(s390_early_mach): New function, split out from...
(s390_emit_prologue): ...here.
(pass_data_s390_early_mach): New pass structure.
(pass_s390_early_mach): New class.
(s390_option_override): Create and register early_mach pass.
Move to end of file.
Richard Biener [Fri, 7 Feb 2014 13:41:10 +0000 (13:41 +0000)]
re PR middle-end/60092 (posix_memalign not recognized to derive alias and alignment info)
2014-02-07 Richard Biener <rguenther@suse.de>
PR middle-end/60092
* gimple-low.c (lower_builtin_posix_memalign): New function.
(lower_stmt): Call it to lower posix_memalign in a way
to make alignment info accessible.
Jakub Jelinek [Thu, 6 Feb 2014 21:54:21 +0000 (22:54 +0100)]
re PR rtl-optimization/60030 (ICE in simplify_subreg, at simplify-rtx.c:5903)
PR rtl-optimization/60030
* internal-fn.c (ubsan_expand_si_overflow_mul_check): Surround
lopart with paradoxical subreg before shifting it up by hprec.
Kyrylo Tkachov [Thu, 6 Feb 2014 16:51:03 +0000 (16:51 +0000)]
[ARM] Cortex-A57 rtx costs
* config/arm/aarch-cost-tables.h (cortexa57_extra_costs): New table.
Remove extra newline at end of file.
* config/arm/arm.c (arm_cortex_a57_tune): New tuning struct.
(arm_issue_rate): Handle cortexa57.
* config/arm/arm-cores.def (cortex-a57): Use cortex_a57 tuning.
(cortex-a57.cortex-a53): Likewise.
Jakub Jelinek [Thu, 6 Feb 2014 15:52:36 +0000 (16:52 +0100)]
re PR debug/59575 (ICE in maybe_record_trace_start, at dwarf2cfi.c:2239)
PR target/59575
* config/arm/arm.c (emit_multi_reg_push): Add dwarf_regs_mask argument,
don't record in REG_FRAME_RELATED_EXPR registers not set in that
bitmask.
(arm_expand_prologue): Adjust all callers.
(arm_unwind_emit_sequence): Allow saved, but not important for unwind
info, registers also at the lowest numbered registers side. Use
gcc_assert instead of abort, and SET_SRC/SET_DEST macros instead of
XEXP.
Jakub Jelinek [Thu, 6 Feb 2014 15:52:17 +0000 (16:52 +0100)]
re PR debug/59575 (ICE in maybe_record_trace_start, at dwarf2cfi.c:2239)
PR target/59575
* config/arm/arm.c (emit_multi_reg_push): Add dwarf_regs_mask argument,
don't record in REG_FRAME_RELATED_EXPR registers not set in that
bitmask.
(arm_expand_prologue): Adjust all callers.
(arm_unwind_emit_sequence): Allow saved, but not important for unwind
info, registers also at the lowest numbered registers side. Use
gcc_assert instead of abort, and SET_SRC/SET_DEST macros instead of
XEXP.
* sem_prag.adb (Analyze_Refined_Pragma): Remove
local variable Pack_Spec. Refinement pragmas may now apply to
bodies of both visible and private subprograms.
2014-02-06 Robert Dewar <dewar@adacore.com>
* exp_attr.adb (Expand_Loop_Entry_Attribute):
Minor change (Attr => N) (Expand_Pred_Succ): New name
Expand_Pred_Succ_Attribute (Expand_N_Attribute_Reference, case
Max): Expand into if expression if Modify_Tree_For_C mode.
(Expand_N_Attribute_Reference, case Min): ditto
* sinfo.ads: Modify_Tree_For_C takes care of expanding Min and
Max attributes.
2014-02-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Analyze_Object_Declaration): Do not generate
predicate check if this is an internal declaration with
No_Initialization set, as for an expanded aggregate component.
2014-02-06 Doug Rupp <rupp@adacore.com>
* init.c (__gnat_default_resignal_p) [VMS]: Test for and resignal
conditions with severity of "SUCCESS" or "INFORMATIONAL".
* sem_prag.adb (Analyze_Abstract_State): Update
all calls to Create_Abstract_State to pass the proper state
"declaration".
(Create_Abstract_State): Add new formal parameter
State_Decl along with comment on usage. Establish a link between
the abstract state entity and the state declaration.
Robert Dewar [Thu, 6 Feb 2014 14:09:36 +0000 (14:09 +0000)]
sem_attr.adb (Analyze_Attribute, case Max): Check for improper comparison of unordered enumeration type.
2014-02-06 Robert Dewar <dewar@adacore.com>
* sem_attr.adb (Analyze_Attribute, case Max): Check for improper
comparison of unordered enumeration type.
(Analyze_Attribute, case Max): Check for improper comparison of
unordered enumeration type.
* sem_res.adb (Bad_Unordered_Enumeration_Reference): Moved to
sem_util.adb.
* sem_util.ads, sem_util.adb (Bad_Unordered_Enumeration_Reference):
Moved here from Sem_Res.
Jakub Jelinek [Thu, 6 Feb 2014 10:59:30 +0000 (11:59 +0100)]
re PR middle-end/59150 (ICE: in expand_one_var, at cfgexpand.c:1242 with -fopenmp)
PR middle-end/59150
* tree-vect-data-refs.c (vect_analyze_data_refs): For clobbers, call
free_data_ref on the dr first, and before goto again also set dr
to the next dr. For simd_lane_access, free old datarefs[i] before
overwriting it. For get_vectype_for_scalar_type failure, don't
free_data_ref if simd_lane_access.
* sem_ch3.adb (Handle_Late_Controlled_Primitive): Add local
variable Spec. Do not inherit the null indicator from the
subprogram body when generating the spec.
Arnaud Charlet [Thu, 6 Feb 2014 10:23:17 +0000 (11:23 +0100)]
[multiple changes]
2014-02-06 Robert Dewar <dewar@adacore.com>
* casing.adb (Determine_Casing): Consider SPARK_Mode to be
mixed case.
2014-02-06 Ed Schonberg <schonberg@adacore.com>
* exp_ch6.adb (Is_Build_In_Place_Function): Predicate is false
when the function has a foreign convention, but not if only the
limited return type has such a convention.
* sem_ch3.adb (Handle_Late_Controlled_Primitive): Remove local
variable Spec. Comment reformatting. Use Copy_Separate_Tree
rather than New_Copy_Tree when building the corresponding
subprogram declaration.
* gnat_ugn.texi, vms_data.ads: Add documentation of -j option for
gnatmetric.
2014-02-06 Robert Dewar <dewar@adacore.com>
* exp_ch4.adb (Expand_N_Shift_Left): Handle shift counts greater
than the word size when operating in Modify_Tree_For_C mode.
* sinfo.ads: Add documentation section on Modify_Tree_For_C mode.
2014-02-06 Robert Dewar <dewar@adacore.com>
* erroutc.adb (Warning_Specifically_Suppressed.Matches):
compare is case insensitive.
* gnat_rm.texi: Document that string compare for Warnings Off
is now case insensitive.
* sem_ch10.adb (Analyze_Package_Body_Stub): Maintain
the configuration options of the enclosing context in a
stack-like fasion.
(Analyze_Subprogram_Body_Stub): Maintain the
configuration options of the enclosing context in a stack-like
fashion.
2014-02-06 Robert Dewar <dewar@adacore.com>
* debug.adb: -gnatd.u sets Modify_Tree_For C
* exp_ch4.adb (Expand_N_Op_Rotate_Left): Expand out
if Modify_Tree_For_C (Expand_N_Op_Rotate_Right): ditto.
(Expand_N_Op_Arithmetic_Right_Shift): ditto.
* exp_intr.adb (Expand_Shift): Call expander so we do
Modify_Tree_For_C expansions.
* gnat1drv.adb (Adjust_Global_Switches): Set Modify_Tree_For_C
if -gnatd.u set.
2014-02-06 Fedor Rybin <frybin@adacore.com>
* prj-proc.ads (Tree_Loaded_Callback): new type Callback used
after the phase 1 of the processing of each aggregated project
to get access to project trees of aggregated projects.
(Process_Project_Tree_Phase_1): new parameter On_New_Tree_Loaded
If specified, On_New_Tree_Loaded is called after each aggregated
project has been processed succesfully.
(Process): new parameter On_New_Tree_Loaded.
* prj-proc.adb (Process_Aggregated_Projects): On_New_Tree_Loaded
callback added after processing of each aggregated project.
(Recursive_Process): new parameter On_New_Tree_Loaded.
(Process): new parameter On_New_Tree_Loaded.
(Process_Project_Tree_Phase_1): new parameter On_New_Tree_Loaded.
* prj-conf.ads (Parse_Project_And_Apply_Config): new parameter
On_New_Tree_Loaded.
* prj-conf.adb (Parse_Project_And_Apply_Config): new parameter
On_New_Tree_Loaded.
2014-02-06 Bob Duff <duff@adacore.com>
* gnat_ugn.texi: Implement --insert-blank-lines and
--preserve-blank-lines switches.
The symbol classification logic in correctly classifies all
SYMBOL_REF's as SYMBOL_FORCE_TO_MEM, the intention of the code is to
catch only the large memory model case.
* einfo.adb Node29 is now used as BIP_Initialization_Call.
(BIP_Initialization_Call): New routine.
(Set_BIP_Initialization_Call): New routine.
(Write_Field29_Name): Add an entry for constants and variables.
* einfo.ads Add new attribute BIP_Initialization_Call and update
its usage in nodes.
(BIP_Initialization_Call): New routine along with pragma Inline.
(Set_BIP_Initialization_Call): New routine along with pragma Inline.
* exp_ch6.adb (Make_Build_In_Place_Call_In_Object_Declaration):
Add local declaration Res_Decl. Capture the build-in-place
initialization call when the related object declaration has
created a transient block.
* exp_ch7.adb (Process_Transient_Objects): Add local variable
Fin_Insrt. Recognize a scenario where an object declaration acts
as a transient context and is initialized by a build-in-place
function call.
Arnaud Charlet [Thu, 6 Feb 2014 10:11:07 +0000 (11:11 +0100)]
[multiple changes]
2014-02-06 Pascal Obry <obry@adacore.com>
* prj-util.adb (For_Interface_Sources): Fix handling of required
bodies for aggregate libs.
2014-02-06 Robert Dewar <dewar@adacore.com>
* nlists.ads: Minor comment clarifications.
2014-02-06 Robert Dewar <dewar@adacore.com>
* gnat1drv.adb (Adjust_Global_Switches): Set Modify_Tree_For_C
if gnatd.V set.
* opt.ads (Modify_Tree_For_C): New flag.
* output.ads, output.adb (Last_Char): New function.
2014-02-06 Pascal Obry <obry@adacore.com>
* projects.texi, prj-env.adb (Initialize_Default_Project_Path): Add
share/gpr for cross-builds.
* sem_prag.adb (Analyze_Depends_In_Decl_Part): Add
local variable Expr. Flag clauses with extra parenthesis as this
is not allowed by the syntax of the pragma. Code reformatting.
* exp_attr.adb (Expand_N_Attribute_Reference): Alphabetize
variables. Rename variabme Tnn to Temp. Do not create a temporary
if assertions are disabled. Find enclosing routine _Postconditions
and insert the temporary that captures the value of the prefix
before the routine.
* exp_ch6.adb (Build_Postconditions_Procedure):
Insert the generated _Postconditions routine
before the first source declaration of the related
subprogram.
(Insert_After_Last_Declaration): Removed.
(Insert_Before_First_Source_Declaration): New routine.
2014-02-06 Ed Schonberg <schonberg@adacore.com>
* exp_util.adb, exp_util.ads (Within_Internal_Subprogram):
Utility to determine whether current expansion is for the body
of a predefined primitive operation.
(Make_Predicate_Check): Use Within_Internal_Subpgram
* checks.adb (Apply_Predicate_Check): Use
Within_Internal_Subprogram
* sem_ch13.adb (Freeze_Entity_Checks): Ditto.
Arnaud Charlet [Thu, 6 Feb 2014 09:58:37 +0000 (10:58 +0100)]
[multiple changes]
2014-02-06 Ed Schonberg <schonberg@adacore.com>
* exp_ch6.adb (Expand_Subprogram_Contract, Append_Enabled_Item):
Take into account the Split_PPC flag to ensure that conjuncts
in a composite postcondition aspect are tested in source order.
* sem_ch6.adb (Analyze_Generic_Subprogram_Body): Flag illegal
use of SPARK_Mode.
* sem_ch12.adb (Analyze_Generic_Subprogram_Declaration): Flag
illegal use of SPARK_Mode.
(Instantiate_Subprogram_Body): Flag illegal use of SPARK_Mode.
* sem_prag.adb (Analyze_Pragma): Code reformatting.
* sem_util.adb Add with and use clause for Aspects.
(Check_SPARK_Mode_In_Generic): New routine.
* sem_util.ads (Check_SPARK_Mode_In_Generic): New routine.
2014-02-06 Thomas Quinot <quinot@adacore.com>
* a-calend.adb (Formatting_Operations.Split): Ensure that
Time_Error is raised for invalid time values.
projects / thirdparty / gcc.git / log
