PR middle-end/54146
* ira.c (init_live_subregs): Take live_subregs_used as a bitmap.
(build_insn_chain): Make live_subregs_used a bitmap.
Use SBITMAP_SIZE to ignore the paradoxical bytes of subregs.
Use sbitmap_free to free the live_subreg sbitmaps.
From-SVN: r190223
PR middle-end/54146
* ifcvt.c: Include pointer-set.h.
(cond_move_process_if_block): Change type of then_regs and
else_regs from alloca'd array to pointer_sets.
(check_cond_move_block): Update for this change.
(cond_move_convert_if_block): Likewise.
* Makefile.in: Fix dependencies for ifcvt.o.
From-SVN: r190222
gcc/
* cfg.c (debug_bb): Do not set TDF_BLOCKS.
* cfghooks.c (dump_bb): Honor ~TDF_BLOCKS.
* pretty-print.c (pp_base_flush): Do not add a newline, and do not
clear pp_needs_newline.
* pretty-print.h (pp_newline_and_flush): New macro with the behavior
of pp_base_flush before the above change.
* langhooks.c (lhd_print_error_function): Use pp_newline_and_flush
instead of pp_flush.
* diagnostic.c (diagnostic_finish): Likewise.
(diagnostic_report_diagnostic): Likewise.
(verbatim): Likewise.
(error_recursion): Likewise.
* tree-pretty-print.c (print_generic_stmt): Likewise.
(print_generic_stmt_indented): LIkewise.
* gimple-pretty-print.c (print_gimple_stmt): Likewise.
(print_gimple_seq): Likewise.
(gimple_dump_bb_buff): Likewise.
c-family/
* c-pretty-print.c (pp_c_function_definition): Use pp_newline_and_flush
instead of separate pp_newline and pp_flush.
(print_c_tree): Likewise.
cp/
* error.c (print_instantiation_context): Pretty-print a newline before
diagnostic_flush_buffer.
* cxx-pretty-print.c (pp_cxx_function_definition): Use
pp_newline_and_flush instead of separate pp_newline and pp_flush.
testsuite/
* gcc.dg/tree-prof/update-loopch.c: Ask for dump with blocks info.
* gcc.dg/tree-ssa/attr-hotcold-2.c: Likewise.
* gcc.dg/tree-ssa/pr18133-1.c: Likewise.
From-SVN: r190215
2012-08-07 Richard Guenther <rguenther@suse.de>
* tree-dfa.c (set_ssa_default_def): Clear the SSA_NAME_DEFAULT_DEF
bit of the old name when we clear the slot.
* tree-ssa-live.c (remove_unused_locals): Release any default
def associated with an unused var.
* tree-ssa-copy.c (may_propagate_copy_into_asm): Always return true.
From-SVN: r190201
2012-08-07 Richard Guenther <rguenther@suse.de>
* tree-into-ssa.c (rewrite_stmt): Remove clobbers for variables
we rewrite into SSA form.
(rewrite_enter_block): Adjust.
* gimple-iterator.c (gsi_replace): Also allow replacement with
a stmt without a lhs.
* tree-ssa-live.c (remove_unused_locals): Remove code handling
clobbers of SSA names.
* tree-nested.c (convert_local_reference_stmt): Remove clobbers
for variables we access through the local chain.
* tree-cfg.c (verify_gimple_assign_single): Verify clobbers
clobber full decls only.
From-SVN: r190200
runtime: support NumCPU() on more platforms
Added support for Solaris, Irix, *BSD (including Darwin).
Still missing support for RTEMS.
Fixes issue 3698 in Go issue tracker.
From-SVN: r190197
When working on something else, I noticed that failing to provide the
second argument to the static_assert operator would lead to an ICE.
Fixed thus, and tested against trunk on x86_64-unknown-linux-gnu.
gcc/cp/
* semantics.c (finish_static_assert): Don't crash on erroneous
message or condition.
gcc/testsuite/
* g++.dg/cpp0x/static_assert8.C: New test.
From-SVN: r190182
* tree-emutls.c (new_emutls_decl): When a var_section is requested by
the target, attach the new decl to that, not to the template section.
From-SVN: r190179
/cp
2012-08-06 Marc Glisse <marc.glisse@inria.fr>
Paolo Carlini <paolo.carlini@oracle.com>
PR c++/54165
* typeck.c (build_static_cast_1): Move the conversion to void case
before the perform_direct_initialization_if_possible call.
/testsuite
2012-08-06 Marc Glisse <marc.glisse@inria.fr>
Paolo Carlini <paolo.carlini@oracle.com>
PR c++/54165
* g++.dg/conversion/void2.C: New.
Co-Authored-By: Paolo Carlini <paolo.carlini@oracle.com>
From-SVN: r190175
2012-08-06 Arnaud Charlet <charlet@adacore.com>
* prj-attr.adb (Register_New_Package): Add missing blank.
2012-08-06 Ed Schonberg <schonberg@adacore.com>
* exp_aggr.adb (Is_Two_Dim_Packed_Array): New predicate,
used when computing maximum size allowable to construct static
aggregate.
2012-08-06 Vincent Pucci <pucci@adacore.com>
* freeze.adb (Freeze_Entity): Inherit_Aspects_At_Freeze_Point
calls added for derived types and subtypes.
* sem_aux.adb, sem_aux.ads (Get_Rep_Item, Get_Rep_Pragma,
Has_Rep_Pragma): New routines.
* sem_ch13.ads (Inherit_Aspects_At_Freeze_Point): New routine.
* sem_ch13.adb (Analyze_Aspect_Specifications): Error message
for aspect Lock_Free fixed.
(Inherits_Aspects_At_Freeze_Point): New routine.
* sem_ch3.adb: Several flag settings removed since inheritance
of aspects must be performed at freeze point.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* s-oscons-tmplt.c: Fix s-oscons.ads formatting on VxWorks.
2012-08-06 Vincent Pucci <pucci@adacore.com>
* sem_dim.adb (Analyze_Dimension_Binary_Op): Issue an error message
for unknown exponent at compile-time.
2012-08-06 Gary Dismukes <dismukes@adacore.com>
* sem_eval.ads (Compile_Time_Known_Value_Or_Aggr): Enhance
comment to make it clear that the aggregate's evaluation might
still involve run-time checks even though the aggregate is
considered known at compile time.
* sinfo.ads (Compile_Time_Known_Aggregate): Correct comment to
refer to Exp_Aggr instead of Sem_Aggr.
From-SVN: r190172
2012-08-06 Robert Dewar <dewar@adacore.com>
* xoscons.adb: Minor code reorganization (remove unused variable
E at line 331).
* g-sercom.ads, exp_attr.adb: Minor reformatting.
* sinfo.adb, sinfo.ads: Minor cleanup, remove unused flag
Static_Processing_OK.
2012-08-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Build_Derived_Concurrent_Type): Copy discriminant
constraint when building a constrained subtype, to prevent
undesirable tree sharing betweeb geberated subtype and derived
type definition.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* g-sercom-mingw.adb, s-oscons-tmplt.c: Add missing constants
on Windows.
2012-08-06 Sergey Rybin <rybin@adacore.com frybin>
* tree_io.ads: Update ASIS_Version_Number because of the tree fix
for discriminant constraints for concurrent types.
From-SVN: r190171
2012-08-06 Thomas Quinot <quinot@adacore.com>
* sem_ch4.adb: Minor reformatting.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* s-oscons-tmplt.c, xoscons.adb: Per the Single UNIX Specification,
types cc_t, speed_t, and tcflag_t defined in <termios.h> all are
unsigned types. Add required special handling to have their correct
unsigned values in s-oscons.ads.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* par-ch13.adb: Minor reformatting.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* g-sercom.adb, g-sercom.ads, g-sercom-mingw.adb, g-sercom-linux.adb
(Set): Add Local and Flow_Control settings.
2012-08-06 Ed Schonberg <schonberg@adacore.com>
* exp_attr.adb: Suppress saving of 'Old if assertions are not
enabled.
From-SVN: r190170
2012-08-06 Tom de Vries <tom@codesourcery.com>
* tree-switch-conversion.c (emit_case_bit_tests): Generate MINUS_EXPR in
unsigned type.
From-SVN: r190168
2012-08-06 Yannick Moy <moy@adacore.com>
* sem_ch4.adb (Analyze_Selected_Component): Issue an error in
Alfa mode for component not present.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* exp_ch7.adb (Insert_Actions_In_Scope_Around): Do not
use a renaming of Scope_Stack.Table (Scope_Stack.Last), as
Process_Transient_Object may introduce new scopes and cause
Scope_Stack.Table to be reallocated.
From-SVN: r190167
2012-08-06 Geert Bosch <bosch@adacore.com>
* a-ngelfu.adb: Change obsolete comment that this is a non-strict
implementation.
2012-08-06 Steve Baird <baird@adacore.com>
* exp_ch7.adb (Build_Finalizer.Process_Object_Declaration): If
CodePeer_Mode = True then omit exception handlers for finalization calls
2012-08-06 Robert Dewar <dewar@adacore.com>
* exp_aggr.adb: Minor reformatting.
From-SVN: r190165
2012-08-06 Robert Dewar <dewar@adacore.com>
* sem_ch6.adb: Remove useless flag Body_Deleted.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* sinfo.ads (End_Span): Change default from No_Uint to Uint_0,
as this is what all usage occurrences of this attribute are
expecting.
* uintp.adb (UI_To_Int): Add assertion to guard against calling
with No_Uint.
From-SVN: r190164
2012-08-06 Hristian Kirtchev <kirtchev@adacore.com>
* sem_mech.adb (Set_Mechanisms): OUT and IN OUT parameters are
now unconditionally passed by reference. IN parameters subject
to convention C_Pass_By_Copy are passed by copy, otherwise they
are passed by reference.
2012-08-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Analyze_Object_Declaration): According to
AI95-0303, protected objects with interrupt handlers can be
declared in nested scopes. This is a binding interpretation,
and thus applies to all versions of the compiler.
2012-08-06 Robert Dewar <dewar@adacore.com>
* frontend.adb, exp_aggr.adb: Minor reformatting.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* par-endh.adb: Minor reformatting.
From-SVN: r190162
2012-08-06 Hristian Kirtchev <kirtchev@adacore.com>
* exp_ch7.adb (Process_Transient_Objects): Remove obsolete loop
processing related to array initialization. The expansion of
loops already contains a mechanism to detect controlled objects
generated by expansion and introduce a block around the loop
statements for finalization purposes.
2012-08-06 Vincent Pucci <pucci@adacore.com>
* sem_ch13.adb: Current scope must be within
or same as the scope of the entity while analysing aspect
specifications at freeze point.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* par_sco.adb: Add note about dubious SCO for TERMINATE
alternative.
* sem_ch8.adb, exp_ch11.adb: Minor reformatting.
2012-08-06 Ed Schonberg <schonberg@adacore.com>
* exp_aggr.adb (Two_Dim_Packed_Array_Handled): New procedure to
transform an aggregate for a packed two-dimensional array into
a one-dimensional array of constant values, in order to avoid
the generation of component-by-component assignments.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* frontend.adb: Do not attempt to process deferred configuration
pragmas if the main unit failed to load, to avoid cascaded
inconsistencies that can lead to a compiler crash.
From-SVN: r190161
2012-08-06 Vincent Pucci <pucci@adacore.com>
* s-atopri.adb: Minor reformatting.
2012-08-06 Arnaud Charlet <charlet@adacore.com>
* gnat-style.texi: Clarify that all subprograms should be
documented. Minor rewording.
From-SVN: r190160
2012-08-06 Robert Dewar <dewar@adacore.com>
* aspects.ads: Define Aspect_Id_Exclude_No_Aspect.
* par-ch13.adb, restrict.adb: Use Aspect_Id_Exclude_No_Aspect to
simplify code.
2012-08-06 Yannick Moy <moy@adacore.com>
* gnat-style.texi: Update style guide for declarations.
2012-08-06 Yannick Moy <moy@adacore.com>
* sem_attr.adb (Analyze_Attribute): In the case for 'Old,
skip a special expansion which is not needed in Alfa mode.
2012-08-06 Yannick Moy <moy@adacore.com>
* sem_ch5.adb (Analyze_Iterator_Specification): Do not perform
an expansion of the iterator in Alfa mode.
From-SVN: r190159
2012-08-06 Robert Dewar <dewar@adacore.com>
* s-oscons-tmplt.c, sem_ch9.adb, osint.adb: Minor reformatting.
2012-08-06 Hristian Kirtchev <kirtchev@adacore.com>
* sem_ch5.adb (Analyze_Loop_Parameter_Specification): Remove obsolete
Alfa-specific guard. The code is a leftover from an early
implementation of iterators which is no longer in use.
2012-08-06 Vincent Celier <celier@adacore.com>
* par-ch13.adb (Get_Aspect_Specifications): Do not consider
No_Aspect when checking for a mispelled aspect.
From-SVN: r190158
2012-08-06 Robert Dewar <dewar@adacore.com>
* s-htable.adb: Minor reformatting.
2012-08-06 Thomas Quinot <quinot@adacore.com>
* s-oscons-tmplt.c, g-socket.adb (System.OS_Constants.Target_OS):
Suppress warnings so that we don't have to do it at each usage.
From-SVN: r190157
2012-08-06 Vincent Pucci <pucci@adacore.com>
* exp_ch9.adb (Build_Lock_Free_Unprotected_Subprogram_Body):
Use of Known_Static_Esize instead of Known_Esize and
Known_Static_RM_Size instead of Known_RM_Size in order to
properly call UI_To_Int. Don't check the size of the component
type in case of generic.
* sem_ch9.adb (Allows_Lock_Free_Implementation):
Use of Known_Static_Esize instead of Known_Esize and
Known_Static_RM_Size instead of Known_RM_Size in order to properly
call UI_To_Int. Don't check the size of the component type in
case of generic.
2012-08-06 Hristian Kirtchev <kirtchev@adacore.com>
* checks.adb (Discrete_Range_Cond): Do not try to optimize on
the assumption that the type of an expression can always fit in
the target type of a conversion.
From-SVN: r190156
Nick Clifton
Steven Bosscher
Bill Schmidt
GCC Administrator
Benjamin Kosnik
Uros Bizjak
Uros Bizjak
Richard Henderson
Paul Brook
Richard Guenther
Nick Clifton
Ian Lance Taylor
Cary Coutant
Sandra Loosemore
Janus Weil
Paolo Carlini
Marc Glisse
Dodji Seketeli
Olivier Hainque
Jason Merrill
Jose Ruiz
Arnaud Charlet
Tom de Vries
Vincent Pucci
Venkataramanan Kumar
Catherine Moore