gcc.git
15 years agodecl.c (validate_alignment): For the case of an implicit array base type...
Arnaud Charlet [Thu, 23 Jul 2009 10:21:03 +0000 (12:21 +0200)]
decl.c (validate_alignment): For the case of an implicit array base type...

* gcc-interface/decl.c (validate_alignment): For the case of an
implicit array base type, look for alignment clause on first subtype.
Code clean up.

From-SVN: r149994

15 years ago[multiple changes]
Arnaud Charlet [Thu, 23 Jul 2009 10:20:25 +0000 (12:20 +0200)]
[multiple changes]

2009-07-23  Olivier Hainque  <hainque@adacore.com>

* g-ssinty.ads: New unit. GNAT.SSE.Internal_Types. Factorize
low level internal type definitions for distinct higher level
binding development activities (user type definitions and
operations).
* gnat_rm.texi: Document it.
* g-ssvety.ads: Use it.
* gcc-interface/Makefile.in: (x86 32/64 linux, cygwin32 sections): Add
g-ssinty.o to EXTRA_GNATRTL_NONTASKING_OBJS.
* gcc-interface/utils.c (gnat_internal_attribute_table): Add entry
for the "may_alias" attribute.

2009-07-23  Thomas Quinot  <quinot@adacore.com>

* scos.ads: Minor typo fix
* gcc-interface/decl.c (validate_alignment): For the case of an
implicit array base type, look for alignment clause on first subtype.
Code clean up.

2009-07-23  Ed Schonberg  <schonberg@adacore.com>

* sem.adb (Walk_Library_Units): Handle properly the case where a unit
in the context depends on the spec of the main unit, by delaying
processing of the main unit body until all other units have been
processed.

From-SVN: r149993

15 years ago[multiple changes]
Arnaud Charlet [Thu, 23 Jul 2009 10:03:21 +0000 (12:03 +0200)]
[multiple changes]

2009-07-23  Arnaud Charlet  <charlet@adacore.com>

* a-convec.adb: Add comments about suspicious/subtle code.

2009-07-23  Ed Schonberg  <schonberg@adacore.com>

* einfo.ads: Document use of Alias in private overriding

2009-07-23  Thomas Quinot  <quinot@adacore.com>

* sem_ch13.adb (Analyze_Attribute_Definition_Clause): For the case of
an array type, propagate alignment from first subtype to implicit base
type so that other subtypes (such as the itypes for aggregates of the
type) also receive the expected alignment.

* g-comlin.ads: Minor documentation clarification/rewording.
* scos.ads: Minor comments update.
* lib-writ.ads: Minor reformatting

From-SVN: r149991

15 years ago[multiple changes]
Arnaud Charlet [Thu, 23 Jul 2009 09:56:17 +0000 (11:56 +0200)]
[multiple changes]

2009-07-23  Gary Dismukes  <dismukes@adacore.com>

* exp_ch3.adb (Expand_N_Object_Declaration): For an initialized object
of a class-wide interface type that is a return object of a
build-in-place function, bypass the interface-related expansions into
renamings with displacement conversions, etc.
* exp_ch5.adb (Expand_N_Extended_Return_Statement): Add an assertion
for the case where a renaming occurs in a build-in-place context, to
assert that the bypassing of the build-in-place treatment only occurs
in the case of a renaming that is an expansion of a return expression
that is itself a build-in-place function call.

2009-07-23  Ed Schonberg  <schonberg@adacore.com>

* sem_ch4.adb (Try_Primitive_Operation): A primitive operation is a
valid candidate interpretation in a prefixed view if it is hidden, but
overrides an inherited operation declared in the visible part.

2009-07-23  Robert Dewar  <dewar@adacore.com>

* exp_ch4.adb (Expand_N_Type_Conversion): Don't promote integer
division operands to 64-bit at all in any circumstances.

From-SVN: r149990

15 years agoUpdate comments.
Arnaud Charlet [Thu, 23 Jul 2009 09:51:55 +0000 (11:51 +0200)]
Update comments.

From-SVN: r149989

15 years agore PR target/40832 (gfortran 4.4.0 generates invalid .s file on solaris-x86 using...
Uros Bizjak [Thu, 23 Jul 2009 09:51:33 +0000 (11:51 +0200)]
re PR target/40832 (gfortran 4.4.0 generates invalid .s file on solaris-x86 using  -march=k8 for 130.socorro)

PR target/40832
* config/i386/i386.c (output_387_ffreep): Rewrite to return
ASM_SHORT instead of .word.
* config/i386/i386.md (*tls_global_dynamic_64): Use ASM_SHORT
instead of .word in asm template.

From-SVN: r149988

15 years ago[multiple changes]
Arnaud Charlet [Thu, 23 Jul 2009 09:51:19 +0000 (11:51 +0200)]
[multiple changes]

2009-07-23  Robert Dewar  <dewar@adacore.com>

* exp_ch4.adb (Analyze_N_Op_Rem): Assume operands are valid when
checking ranges for mod/rem to see if conditional jump will be
generated.
(Analyze_N_Op_Rem): Don't try to check actual lower bounds for
generating special -1 test for rem, generate it whenever both
operands can be negative (match circuit in Sem_Res).
(Analyze_N_Op_Rem): Don't go to base type, no longer needed and
destroys memory of positive range.
* sem_res.adb (Resolve_Arithmetic_Op): Assume operands are valid when
checking ranges for mod/rem to see if conditional jump will be generated

2009-07-23  Ed Schonberg  <schonberg@adacore.com>

* exp_ch3.adb (Build_Equivalent_Record_Aggregate): If the type of a
scalar components has non-static bounds, the equivalent aggregate
cannot be built, even if the expression is static, because range checks
will be generated.

From-SVN: r149987

15 years agoexp_ch4.adb (Expand_N_Type_Conversion): Don't promote integer division operands to...
Robert Dewar [Thu, 23 Jul 2009 09:44:59 +0000 (09:44 +0000)]
exp_ch4.adb (Expand_N_Type_Conversion): Don't promote integer division operands to 64-bit inside a conversion if...

2009-07-23  Robert Dewar  <dewar@adacore.com>

* exp_ch4.adb (Expand_N_Type_Conversion): Don't promote integer
division operands to 64-bit inside a conversion if 64-bit division not
available.

From-SVN: r149985

15 years ago[multiple changes]
Arnaud Charlet [Thu, 23 Jul 2009 09:42:18 +0000 (11:42 +0200)]
[multiple changes]

2009-07-23  Sergey Rybin  <rybin@adacore.com>

* gnat_ugn.texi: Update doc on Misnamed_Identifiers rule.

2009-07-23  Javier Miranda  <miranda@adacore.com>

* sinfo.ads, sinfo.adb (SCIL_Entity/Set_SCIL_Entity): new subprograms
(Entity/Set_Entity): not available in N_Null_Statement nodes
(Is_Scil_Node): renamed as Is_SCIL_Node
(Scil_Nkind): renamed as SCIL_Nkind
(Scil_Related_Node): renamed as SCIL_Related_Node
(Scil_Target_Prim): renamed as SCIL_Target_Prim
(Set_Is_Scil_Node): Renamed as Set_Is_SCIL_Node
(Set_Scil_Related_Node): Renamed as Set_SCIL_Related_Node
(Set_Scil_Target_Prim): Renamed as Set_SCIL_Target_Prim
Update documentation
* exp_disp.ads (Scil_Node_Kind): Renamed as SCIL_Node_Kind
(Get_Scil_Node_Kind): Renamed as Get_SCIL_Node_Kind
(New_Scil_Node): Renamed as New_SCIL_Node
* exp_disp.adb Update all occurrences of New_Scil_Node to New_SCIL_Node.
(Get_Scil_Node_Kind): Renamed as Get_SCIL_Node_Kind
(New_Scil_Node): Renamed as New_SCIL_Node
* exp_ch3.adb Update occurrence of New_Scil_Node to New_SCIL_Node.

2009-07-23  Robert Dewar  <dewar@adacore.com>

* sem_prag.adb: No_Return is an Ada 2005 pragma, not a GNAT pragma
* snames.ads-tmpl: No_Return is an Ada 2005 pragma, not a GNAT pragma
* einfo.adb: Minor reformatting

From-SVN: r149984

15 years agochecks.adb (Apply_Arithmetic_Overflow_Check): Add comments cross-referencing the...
Robert Dewar [Thu, 23 Jul 2009 09:34:26 +0000 (09:34 +0000)]
checks.adb (Apply_Arithmetic_Overflow_Check): Add comments cross-referencing the new related code in...

2009-07-23  Robert Dewar  <dewar@adacore.com>

* checks.adb (Apply_Arithmetic_Overflow_Check): Add comments
cross-referencing the new related code in
Exp_Ch4.Expand_N_Type_Conversion.
* exp_ch4.adb (Expand_N_Type_Conversion): Avoid unnecessary overflows

* exp_disp.adb, exp_disp.ads, sinfo.ads: Minor reformatting.
Add comment.

From-SVN: r149983

15 years agosinfo.ads (Is_Scil_Node, [...]): Add missing documentation.
Javier Miranda [Thu, 23 Jul 2009 09:27:18 +0000 (09:27 +0000)]
sinfo.ads (Is_Scil_Node, [...]): Add missing documentation.

2009-07-23  Javier Miranda  <miranda@adacore.com>

* sinfo.ads (Is_Scil_Node, Scil_Nkind, Scil_Related_Node,
Scil_Target_Prim, N_Has_Entity): Add missing documentation.
* exp_disp.ads (Scil_Node_Kind): Ditto.
* exp_disp.adb (Make_DT, Make_Tags): Ditto.
* exp_ch3.adb (Build_Init_Procedure): Ditto.

From-SVN: r149982

15 years agoeinfo.adb (Component_Type): Add missing assertion.
Javier Miranda [Thu, 23 Jul 2009 09:21:31 +0000 (09:21 +0000)]
einfo.adb (Component_Type): Add missing assertion.

2009-07-23  Javier Miranda  <miranda@adacore.com>

* einfo.adb (Component_Type): Add missing assertion.
* sem_res.adb (Resolve_Call): Ensure proper kind of entity before
reading attribute Component_Size.
* exp_ch4.adb (Is_Safe_In_Place_Array_Op): Ensure proper kind of entity
before reading attributes Component_Size and Component_Type.
* exp_ch3.adb (Build_Initialization_Call): Ensure proper kind of entity
before reading attribute Component_Type.

From-SVN: r149981

15 years agoMinor reformatting.
Arnaud Charlet [Thu, 23 Jul 2009 09:18:46 +0000 (11:18 +0200)]
Minor reformatting.

From-SVN: r149980

15 years ago[multiple changes]
Arnaud Charlet [Thu, 23 Jul 2009 09:17:56 +0000 (11:17 +0200)]
[multiple changes]

2009-07-23  Olivier Hainque  <hainque@adacore.com>

* gnat_rm.texi: Document the GNAT.SSE units.

2009-07-23  Ed Schonberg  <schonberg@adacore.com>

* sem_ch6.adb (Analyze_Return_Type): Do not create itype reference for
not null return if it appears on a subunit.

From-SVN: r149979

15 years agoexp_ch5.adb (Expand_N_Assignment_Statement): Do left-side validity check right away...
Robert Dewar [Thu, 23 Jul 2009 09:10:58 +0000 (09:10 +0000)]
exp_ch5.adb (Expand_N_Assignment_Statement): Do left-side validity check right away so it does not get skipped for early...

2009-07-23  Robert Dewar  <dewar@adacore.com>

* exp_ch5.adb (Expand_N_Assignment_Statement): Do left-side validity
check right away so it does not get skipped for early returns, e.g.
array assignments.
(Expand_N_Assignment_Statement): Don't propagate Is_Known_Valid to
left-side unless we really know the value is valid.

* errout.adb, exp_ch3.adb, exp_disp.ads, sinfo.ads, exp_disp.adb: Minor
reformatting. Minor code reorganization. Add comments.

From-SVN: r149978

15 years agoget_scos.adb (Skip_EOL): Fix error of mishandling end of line after complex condition.
Robert Dewar [Thu, 23 Jul 2009 08:49:47 +0000 (08:49 +0000)]
get_scos.adb (Skip_EOL): Fix error of mishandling end of line after complex condition.

2009-07-23  Robert Dewar  <dewar@adacore.com>

* get_scos.adb (Skip_EOL): Fix error of mishandling end of line after
complex condition.

From-SVN: r149975

15 years ago[multiple changes]
Arnaud Charlet [Thu, 23 Jul 2009 08:48:01 +0000 (10:48 +0200)]
[multiple changes]

2009-07-23  Gary Dismukes  <dismukes@adacore.com>

* sem_ch6.adb (Check_Return_Subtype_Indication): Replace type equality
with test of coverage, to allow specific type objects in extended
returns of class-wide functions. Remove now-unnecessary special-case
tests that allowed this in certain cases of expanded extended returns.

2009-07-23  Javier Miranda  <miranda@adacore.com>

* sinfo.ads,sinfo.adb (Entity/Set_Entity): Attribute available in
N_Null_Statements (for SCIL nodes).
(Is_Scil_Node/Set_Is_Scil_Node): New attribute (for SCIL nodes).
(Scil_Nkind/Set_Scil_Nkind): New attribute (for SCIL nodes).
(Scil_Related_Node/Set_Scil_Related_Node): New attribute (for SCIL
nodes).
(Scil_Target_Prim/Set_Scil_Target_Prim): New attribute (for SCIL nodes).
* exp_disp.adb (Expand_Dispatching_Call): Add generation of SCIL node
associated with dispatching call.
(Get_Scil_Node_Kind): New function that returns the kind of SCIL node.
(Make_DT, Make_Tags): Add generation of SCIL nodes associated with
initialization of dispatch tables and initialization of tags.
(New_Scil_Node): New function that creates a new SCIL node.
(Build_Init_Procedure): Add generation of SCIL node associated with the
initialization of tags done in the IP subprogram.

2009-07-23  Ed Schonberg  <schonberg@adacore.com>

* errout.adb (Error_Msg_NEL): If the entity in the initial message has
Warnings_Off, do not emit continuation messages.

* sem_ch10.adb: Set Is_Compilation_Unit on generated child subprogram
spec.

2009-07-23  Emmanuel Briot  <briot@adacore.com>

* ali.adb: Minor comment update

2009-07-23  Vasiliy Fofanov  <fofanov@adacore.com>

* s-win32.ads (HANDLE): Define to be the same size as address type.
Fix copyright.

2009-07-23  Olivier Hainque  <hainque@adacore.com>

* g-sse.ads: New file. Root of the SSE facilities trees, with
general description and common declarations.
* g-ssvety.ads: New file. Expose user level SSE vector types.
* impunit.adb (Non_Imp_File_Names_95): Register new units.
* gcc-interface/Makefile.in (x86 32/64 linux, win32): Add
EXTRA_GNATRTL_NONTASKING_OBJS entries for SSE units.

2009-07-23  Ben Brosgol  <brosgol@adacore.com>

* gnat_ugn.texi: Wordsmithing.

From-SVN: r149974

15 years ago* prj-conf.ads, prj-conf.adb: Switch to GPLv3.
Arnaud Charlet [Thu, 23 Jul 2009 08:45:14 +0000 (08:45 +0000)]
* prj-conf.ads, prj-conf.adb: Switch to GPLv3.

From-SVN: r149973

15 years agore PR libfortran/32784 ([win32] Using 'CONOUT$', 'CONIN$', or 'CONERR$' as assigned...
Jerry DeLisle [Thu, 23 Jul 2009 00:58:46 +0000 (00:58 +0000)]
re PR libfortran/32784 ([win32] Using 'CONOUT$', 'CONIN$', or 'CONERR$' as assigned file generates Fortran runtime error: Bad file descriptor)

2009-07-22  Jerry DeLisle  <jvdelisle@gcc.gnu.org>

PR libfortran/32784
* unix.c (regular_file): Check for CONIN$ CONOUT$, and CONERR$ and open
the respective /dev/conin or /dev/conout devices.  This is Cygwin
specific.

From-SVN: r149970

15 years agoDaily bump.
GCC Administrator [Thu, 23 Jul 2009 00:17:08 +0000 (00:17 +0000)]
Daily bump.

From-SVN: r149969

15 years agomangle.c (mangle_type_string_for_rtti): Rename to be clearer.
Jason Merrill [Wed, 22 Jul 2009 23:03:22 +0000 (19:03 -0400)]
mangle.c (mangle_type_string_for_rtti): Rename to be clearer.

* mangle.c (mangle_type_string_for_rtti): Rename to be clearer.
(needs_fake_anon): New.
(write_name): Check it.
(write_nested_name): Add a fake anonymous namespace scope if true.
* name-lookup.c (get_anonymous_namespace_name): No longer static.
* rtti.c, cp-tree.h: Adjust.

* libsupc++/typeinfo (__GXX_MERGED_TYPEINFO_NAMES): Default to 0.

From-SVN: r149964

15 years agore PR target/37488 (register allocation spills floats needlessly)
Vladimir Makarov [Wed, 22 Jul 2009 22:00:17 +0000 (22:00 +0000)]
re PR target/37488 (register allocation spills floats needlessly)

2009-07-22  Vladimir Makarov  <vmakarov@redhat.com>

PR target/37488
* ira-lives.c (bb_has_abnormal_call_pred): New function.
(process_bb_node_lives): Use it.

* ira.c (setup_cover_and_important_classes): Don't setup
ira_important_class_nums.  Add cover classes to the end of
important classes.
(cover_class_order, comp_reg_classes_func,
reorder_important_classes): New.
(find_reg_class_closure): Use reorder_important_classes.

* config/i386/i386.h (IRA_COVER_CLASSES): Remove.

* config/i386/i386.c (i386_ira_cover_classes): New function.
(TARGET_IRA_COVER_CLASSES): Redefine.

* doc/tm.texi (TARGET_IRA_COVER_CLASSES): Add a comment about
importance of order of cover classes in the array.

From-SVN: r149962

15 years agotree-pass.h (TDF_EH): Define.
Diego Novillo [Wed, 22 Jul 2009 19:44:46 +0000 (15:44 -0400)]
tree-pass.h (TDF_EH): Define.

* tree-pass.h (TDF_EH): Define.
* gimple-pretty-print.c (dump_gimple_stmt): If FLAGS
contains TDF_EH, print the EH region number holding GS.
* tree-dump.c (dump_options): Add "eh".
* doc/invoke.texi: Document it.

From-SVN: r149958

15 years ago[multiple changes]
Paul Thomas [Wed, 22 Jul 2009 18:12:35 +0000 (18:12 +0000)]
[multiple changes]

2009-07-22  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/40796
* trans-decl.c (generate_local_decl): Unreferenced result
variables with allocatable components should be treated like
INTENT_OUT dummy variables.

2009-07-21  Richard Guenther  <rguenther@suse.de>

PR fortran/40726
Change attribution to Richi.

From-SVN: r149952

15 years ago2009-07-22 Doug Kwan <dougkwan@google.com>
Doug Kwan [Wed, 22 Jul 2009 18:06:08 +0000 (18:06 +0000)]
2009-07-22  Doug Kwan  <dougkwan@google.com>

ChangeLog:

* config/arm/arm.md (subdi3) Copy non-reg values to DImode registers.

testsuite/ChangeLog:

* gcc.c-torture/compile/20090721-1.c: New test.

From-SVN: r149951

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 15:56:47 +0000 (17:56 +0200)]
[multiple changes]

2009-07-22  Eric Botcazou  <ebotcazou@adacore.com>

* exp_aggr.adb (Gen_Loop): Do not qualify the bounds of the range if
they are already of the base type of the index.

2009-07-22  Brett Porter  <porter@adacore.com>

* sysdep.c, init.c: Fix typo: _SPE_ should have been __SPE__.

2009-07-22  Robert Dewar  <dewar@adacore.com>

* vms_data.ads: Add entry for SCO_OUTPUT (-gnateS)
* gnat_ugn.texi: Add documentation for -gnateS switch
* ug_words: Add entry for -gnateS /SCO_OUTPUT
* gcc-interface/Make-lang.in: Update dependenciest.3

* get_scos.adb, get_scos.ads, gnat1drv.adb, par_sco.adb,
par_sco.ads, put_scos.adb, put_scos.ads, scos.adb, scos.ads: Initial
complete information for SCO input/output.

From-SVN: r149945

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 15:35:52 +0000 (17:35 +0200)]
[multiple changes]

2009-07-22  Sergey Rybin  <rybin@adacore.com>

* gnat_ugn.texi: Update doc for some gnatcheck rules.

2009-07-22  Robert Dewar  <dewar@adacore.com>

* par_sco.adb, par_sco.ads (pscos): New debug routine to output
contents of SCO tables.
* put_scos.adb, put_scos.ads, get_scos.adb, get_scos.ads,
scos.adb, scos.ads: New files.
* gcc-interface/Make-lang.in: Update dependencies.

* lib-util.ads, gnatbind.ads, ali.ads, binderr.ads: Minor comment
fixes and reformatting.

From-SVN: r149943

15 years agore PR tree-optimization/35229 (Vectorizer doesn't support dependence created by...
Michael Matz [Wed, 22 Jul 2009 15:30:50 +0000 (15:30 +0000)]
re PR tree-optimization/35229 (Vectorizer doesn't  support dependence created by predictive commoning or PRE)

        PR tree-optimization/35229
        PR tree-optimization/39300

        * tree-ssa-pre.c (includes): Include tree-scalar-evolution.h.
        (inhibit_phi_insertion): New function.
        (insert_into_preds_of_block): Call it for REFERENCEs.
        (init_pre): Initialize and finalize scalar evolutions.
        * Makefile.in (tree-ssa-pre.o): Depend on tree-scalar-evolution.h .

testsuite/
        * gcc.dg/vect/vect-pre-interact.c: New test.

From-SVN: r149942

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 15:21:51 +0000 (17:21 +0200)]
[multiple changes]

2009-07-22  Robert Dewar  <dewar@adacore.com>

* g-socket.ads: Minor reformatting

2009-07-22  Gary Dismukes  <dismukes@adacore.com>

* sem_warn.adb (Warn_On_Unreferenced_Entity): Add warning messages that
include the entity kind for following cases of unreferenced entities:
E_Label, E_Discriminant, E_Package, E_Exception, and Formal_Object_Kind.

From-SVN: r149940

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 15:13:23 +0000 (17:13 +0200)]
[multiple changes]

2009-07-22  Ed Falis  <falis@adacore.com>

* s-vxwext-kernel.adb, s-vxwext-kernel.ads: Replace use of taskStop
with taskSuspend.

2009-07-22  Arnaud Charlet  <charlet@adacore.com>

* adadecode.c: Make this file compilable outside of GCC.

2009-07-22  Thomas Quinot  <quinot@adacore.com>

* g-socket.adb, g-socket.ads (Check_Selector): Make sure that
(partially) default-initialized socket sets are handled properly by
clearing their Set component.

2009-07-22  Bob Duff  <duff@adacore.com>

* gnat_ugn.texi: Clarify the -gnatVx (validity checking) switches.

From-SVN: r149939

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 15:04:18 +0000 (17:04 +0200)]
[multiple changes]

2009-07-22  Robert Dewar  <dewar@adacore.com>

* gnat_ugn.texi: Minor reformatting

2009-07-22  Ed Schonberg  <schonberg@adacore.com>

* errout.adb (Error_Msg): A style message within an instantiation
should not be labelled as an error.

From-SVN: r149938

15 years agousing21.C: Fix duplicate paste.
Richard Guenther [Wed, 22 Jul 2009 15:01:30 +0000 (15:01 +0000)]
using21.C: Fix duplicate paste.

2009-07-22  Richard Guenther  <rguenther@suse.de>

        * g++.dg/lookup/using21.C: Fix duplicate paste.

From-SVN: r149937

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 14:58:15 +0000 (16:58 +0200)]
[multiple changes]

2009-07-22  Ed Schonberg  <schonberg@adacore.com>

* freeze.adb (Freeze_Entity): Do not generate extra formal for function
in initialization expression if function does not have convention Ada.

2009-07-22  Sergey Rybin  <rybin@adacore.com>

* gnat_ugn.texi, vms_data.ads: Add qualifier for new gnatpp option
'--separate-label' to control label layout.

From-SVN: r149936

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 13:24:46 +0000 (15:24 +0200)]
[multiple changes]

2009-07-22  Robert Dewar  <dewar@adacore.com>

* exp_tss.ads, sem_eval.adb: Minor reformatting

2009-07-22  Bob Duff  <duff@adacore.com>

* exp_dist.adb, exp_dist.ads: Update comments.

From-SVN: r149933

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 13:16:44 +0000 (15:16 +0200)]
[multiple changes]

2009-07-22  Brett Porter  <porter@adacore.com>

* init.c (__gnat_init_float): For SPE, set bits in SPEFSCR instead of
FPSCR.
* sysdep.c (__gnat_get_task_options): Set task option enabling SPE.

2009-07-22  Gary Dismukes  <dismukes@adacore.com>

* exp_ch5.adb, sem_util.adb, sem_attr.adb, exp_dbug.ads, exp_ch2.adb,
exp_tss.ads, exp_ch4.adb, sem_ch4.adb: Correct spelling error.
Minor reformatting.
* sem_res.adb (Resolve_Explicit_Dereference): Reword one comment that
used poor terminology.

2009-07-22  Robert Dewar  <dewar@adacore.com>

* freeze.adb, sem_ch3.adb, sem_prag.adb: Minor reformatting
* sem_eval.adb, exp_tss.adb: Minor comment update.
* stylesw.adb: Code clean up.

From-SVN: r149932

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 13:09:40 +0000 (15:09 +0200)]
[multiple changes]

2009-07-22  Ed Schonberg  <schonberg@adacore.com>

* freeze.adb (Freeze_Entity): If Implicit_Packing is enabled, and the
component size is an exact number of bytes, an array type can have a
size clause that forces packing even though the array type itself is
not bit-packed.

2009-07-22  Thomas Quinot  <quinot@adacore.com>

* sem_ch3.adb (Analyze_Object_Declaration): For a constant declaration,
if there is a previous entity with the same name in the scope, ignore
it if it is the renaming declaration for a generic package introduced
in instances.

2009-07-22  Nicolas Roche  <roche@adacore.com>

* seh_init.c: use RtlAddFunctionTable to register our SEH exception
handler on x86_64 windows.

2009-07-22  Arnaud Charlet  <charlet@adacore.com>

* sem_prag.adb (Analyze_Pragma): Initialize/Normalize_Scalars create
false positives in CodePeer, so ignore this pragma in this mode.

From-SVN: r149931

15 years agopredicates.md (zero_extended_scalar_load_operand): Use CONST_VECTOR_NUNITS to determi...
Uros Bizjak [Wed, 22 Jul 2009 13:04:19 +0000 (15:04 +0200)]
predicates.md (zero_extended_scalar_load_operand): Use CONST_VECTOR_NUNITS to determine number of elements.

* config/i386/predicates.md (zero_extended_scalar_load_operand):
Use CONST_VECTOR_NUNITS to determine number of elements.

From-SVN: r149930

15 years agovalarray (valarray<>::operator=(const valarray<>&), [...]): Implement resolution...
Paolo Carlini [Wed, 22 Jul 2009 12:19:58 +0000 (12:19 +0000)]
valarray (valarray<>::operator=(const valarray<>&), [...]): Implement resolution of DR 630, [Ready] in Frankfurt.

2009-07-22  Paolo Carlini  <paolo.carlini@oracle.com>

* include/std/valarray (valarray<>::operator=(const valarray<>&),
valarray<>::operator=(initializer_list<>)): Implement resolution
of DR 630, [Ready] in Frankfurt.
* testsuite/26_numerics/valarray/dr630-1.cc: New.
* testsuite/26_numerics/valarray/dr630-2.cc: Likewise.
* doc/xml/manual/intro.xml: Add an entry for DR 630.

From-SVN: r149929

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 10:39:30 +0000 (12:39 +0200)]
[multiple changes]

2009-07-22  Thomas Quinot  <quinot@adacore.com>

* sem_util.adb, sem_ch10.adb: Minor reformatting

* g-socket.adb (Receive_Socket, recvfrom(2) variant): Apply required
special handling for the case of no data received and Item'First =
Stream_Element_Offset'First.
(Last_Index): New subprogram factoring the above special handling
over the various locations where it is required.

2009-07-22  Arnaud Charlet  <charlet@adacore.com>

* gnat1drv.adb (Gnat1drv): Also disable division by zero and alignment
checks in CodePeer_Mode.
* gcc-interface/Make-lang.in: Update dependencies.

2009-07-22  Ed Schonberg  <schonberg@adacore.com>

* sem_aggr.adb: Improve error message.

* sem_ch13.adb: If Ignore_Rep_Clauses is enabled, do a minimal analysis
of an address representation clause.
* freeze.adb (Freeze_Static_Object): An local imported object is legal
if it has an address clause.

From-SVN: r149926

15 years ago[multiple changes]
Arnaud Charlet [Wed, 22 Jul 2009 10:31:30 +0000 (12:31 +0200)]
[multiple changes]

2009-07-22  Thomas Quinot  <quinot@adacore.com>

* sem_elab.adb (Insert_Elab_Check): When relocating an overloaded
expression to insert an elab check using a conditional expression, be
sure to carry the original list of interpretations to the new location.

2009-07-22  Gary Dismukes  <dismukes@adacore.com>

* gnat1drv.adb: Fix spelling error.

2009-07-22  Javier Miranda  <miranda@adacore.com>

* sem_type.ads, sem_type.adb (In_Generic_Actual): Leave this subprogram
at the library level and fix a hidden bug in its implementation: its
functionality for renaming objects was broken because
N_Object_Renaming_Declarations nodes are not a subclass of
N_Declaration nodes (as documented in sinfo.ads).
* sem_util.adb (Check_Dynamically_Tagged_Expression): Include in this
check nodes that are actuals of generic instantiations.

2009-07-22  Ed Schonberg  <schonberg@adacore.com>

* sinfo.ads, sinfo.adb (Pending_Context): New flag to indicate that the
context of a compilation unit is being analyzed. Used to detect
circularities created by with_clauses that are not detected by the
loading machinery.
* sem_ch10.adb (Analyze_Compilation_Unit): Set Pending_Context before
analyzing the context of the current compilation unit, to detect
possible circularities created by with_clauses.

From-SVN: r149925

15 years agosem_type.adb (Get_First_Interp): Fix wrong loop exit condition.
Thomas Quinot [Wed, 22 Jul 2009 10:26:19 +0000 (10:26 +0000)]
sem_type.adb (Get_First_Interp): Fix wrong loop exit condition.

2009-07-22  Thomas Quinot  <quinot@adacore.com>

* sem_type.adb (Get_First_Interp): Fix wrong loop exit condition.

From-SVN: r149924

15 years agosem_res.adb (Check_No_Direct_Boolean_Operators): Add check for -gnatyB
Robert Dewar [Wed, 22 Jul 2009 10:25:57 +0000 (10:25 +0000)]
sem_res.adb (Check_No_Direct_Boolean_Operators): Add check for -gnatyB

2009-07-22  Robert Dewar  <dewar@adacore.com>

* sem_res.adb (Check_No_Direct_Boolean_Operators): Add check for -gnatyB
* style.ads, styleg.adb, styleg.ads (Check_Boolean_Operator): New
procedure.
* usage.adb, stylesw.ads, stylesw.adb: Add handling of -gnatyB switch
* gnat_ugn.texi: Add documentation of -gnatyB
* vms_data.ads: Add entry for -gnatyB (STYLE=BOOLEAN_OPERATORS)

From-SVN: r149923

15 years agos-stchop.adb, [...]: Make sure sources obey short-circuit style rule.
Robert Dewar [Wed, 22 Jul 2009 10:25:32 +0000 (10:25 +0000)]
s-stchop.adb, [...]: Make sure sources obey short-circuit style rule.

2009-07-22  Robert Dewar  <dewar@adacore.com>

* s-stchop.adb, a-direct.adb, a-ztexio.adb, gnatchop.adb, prj-proc.adb,
make.adb, s-regpat.adb, ali-util.adb, a-ngcefu.adb, prep.adb,
s-tassta.adb, a-tifiio.adb, a-textio.adb, prj.adb, uintp.adb,
s-valrea.adb, a-ngelfu.adb, prepcomp.adb, sinput-l.adb, vms_conv.adb,
errout.adb, g-alleve.adb, repinfo.adb, a-wtedit.adb, ali.adb,
a-witeio.adb, prj-dect.adb, prj-nmsc.adb, sinput-c.adb, binde.adb,
s-regexp.adb, s-imgrea.adb, a-teioed.adb, errutil.adb, prj-util.adb,
a-ztedit.adb, gnatls.adb, prj-conf.adb, bcheck.adb, s-scaval.adb,
erroutc.adb, osint.adb, a-strfix.adb, s-fileio.adb: Make sure sources
obey short-circuit style rule.

From-SVN: r149921

15 years agore PR c++/40799 (Revision 149750 failed 483.xalancbmk in SPEC CPU 2006)
Richard Guenther [Wed, 22 Jul 2009 08:43:19 +0000 (08:43 +0000)]
re PR c++/40799 (Revision 149750 failed 483.xalancbmk in SPEC CPU 2006)

2009-07-22  Richard Guenther  <rguenther@suse.de>

PR c++/40799
* cp-gimplify.c (cp_gimplify_expr): Move handling of using
related exprs to ...
(cp_genericize_r): ... genericization stage.
(cp_genericize): Adjust.

* g++.dg/lookup/using21.C: New testcase.

From-SVN: r149919

15 years agotrans.h (gfc_set_decl_assembler_name): New prototype.
Francois-Xavier Coudert [Wed, 22 Jul 2009 08:28:10 +0000 (08:28 +0000)]
trans.h (gfc_set_decl_assembler_name): New prototype.

* trans.h (gfc_set_decl_assembler_name): New prototype.
* trans-decl.c (gfc_set_decl_assembler_name): New function.
(gfc_get_symbol_decl, gfc_get_extern_function_decl,
build_function_decl): Use gfc_set_decl_assembler_name instead of
SET_DECL_ASSEMBLER_NAME.
* trans-common.c (build_common_decl): Use
gfc_set_decl_assembler_name instead of SET_DECL_ASSEMBLER_NAME.

From-SVN: r149918

15 years agoconstraints.md (ZQ, ZR, ZS, ZT): New constraints.
Andreas Krebbel [Wed, 22 Jul 2009 07:47:04 +0000 (07:47 +0000)]
constraints.md (ZQ, ZR, ZS, ZT): New constraints.

2009-07-22  Andreas Krebbel  <krebbel1@de.ibm.com>

* config/s390/constraints.md (ZQ, ZR, ZS, ZT): New constraints.
(U, W): Constraints are now deprecated and will be removed if we
run out of letters.
* config/s390/s390.md (U, W): Replaced with ZQZR, ZSZT throughout
the file.
("prefetch"): Add the stcmh instruction for prefetching.
* config/s390/s390.c (s390_symref_operand_p): Function moved. No
changes.
(s390_short_displacement): Return always true if compiling for
machines not providing the long displacement facility.
(s390_mem_constraint): Support the new constraint letter Z.
(s390_check_qrst_address): New function.

From-SVN: r149917

15 years agodecl.c (reshape_init_r): { T } is not an aggregate initializer for class T.
Jason Merrill [Wed, 22 Jul 2009 03:32:39 +0000 (23:32 -0400)]
decl.c (reshape_init_r): { T } is not an aggregate initializer for class T.

* decl.c (reshape_init_r): { T } is not an aggregate initializer
for class T.

From-SVN: r149874

15 years agoCore issue 934
Jason Merrill [Wed, 22 Jul 2009 03:32:30 +0000 (23:32 -0400)]
Core issue 934

Core issue 934
* call.c (reference_binding): Implement binding to { }.
(initialize_reference): Binding temporary to non-const && is fine.
* decl.c (grok_reference_init): Remove error for CONSTRUCTOR.

From-SVN: r149873

15 years agoDaily bump.
GCC Administrator [Wed, 22 Jul 2009 00:16:47 +0000 (00:16 +0000)]
Daily bump.

From-SVN: r149872

15 years agomep.c (mep_legitimize_arg): Leave control registers alone too.
DJ Delorie [Tue, 21 Jul 2009 23:49:41 +0000 (19:49 -0400)]
mep.c (mep_legitimize_arg): Leave control registers alone too.

* config/mep/mep.c (mep_legitimize_arg): Leave control registers
alone too.

From-SVN: r149869

15 years ago* c-common.c (max_tinst_depth): Increase default to 1024.
Jason Merrill [Tue, 21 Jul 2009 21:03:25 +0000 (17:03 -0400)]
* c-common.c (max_tinst_depth): Increase default to 1024.

From-SVN: r149867

15 years agoAdd ccoutant to previous dwarf4 entry.
Cary Coutant [Tue, 21 Jul 2009 19:05:14 +0000 (19:05 +0000)]
Add ccoutant to previous dwarf4 entry.

From-SVN: r149866

15 years agosse.md (vec_unpacku_float_hi_v4si): New expander.
Uros Bizjak [Tue, 21 Jul 2009 15:32:51 +0000 (17:32 +0200)]
sse.md (vec_unpacku_float_hi_v4si): New expander.

* config/i386/sse.md (vec_unpacku_float_hi_v4si): New expander.
(vec_unpacku_float_lo_v4si): Ditto.

testsuite/ChangeLog:

* gcc.target/i386/vectorize8.c: New test.

From-SVN: r149862

15 years agore PR target/40811 (unsigned int to float isn't vectorized)
Uros Bizjak [Tue, 21 Jul 2009 15:17:23 +0000 (17:17 +0200)]
re PR target/40811 (unsigned int to float isn't vectorized)

PR target/40811
* config/i386/sse.md (sse2_cvtudq2ps): New expander.
(enum ix86_builtins): Add IX86_BUILTIN_CVTUDQ2PS.
(builtin_description): Add __builtin_ia32_cvtudq2ps.
(ix86_vectorize_builtin_conversion): Handle IX86_BUILTIN_CVTUDQ2PS.

testsuite/ChangeLog:

PR target/40811
* lib/target-supports.exp (check_effective_target_vect_uintfloat_cvt):
Add i?86 and x86_64 targets.
* gcc.target/i386/vectorize7.c: New test.

PR target/40809
* gcc.target/i386/pr40809.c: New test.

From-SVN: r149861

15 years agore PR tree-optimization/40813 (ICE in gsi_insert_seq_nodes_after, at gimple-iterator...
Jakub Jelinek [Tue, 21 Jul 2009 14:51:13 +0000 (16:51 +0200)]
re PR tree-optimization/40813 (ICE in gsi_insert_seq_nodes_after, at gimple-iterator.c:222)

PR tree-optimization/40813
* tree-inline.c (copy_bb): Regimplify RHS after last stmt, not before
it.

* g++.dg/opt/inline15.C: New test.

From-SVN: r149857

15 years agochrono (duration<>::operator%=, operator%): Add, per DR 934.
Paolo Carlini [Tue, 21 Jul 2009 14:48:47 +0000 (14:48 +0000)]
chrono (duration<>::operator%=, operator%): Add, per DR 934.

2009-07-21  Paolo Carlini  <paolo.carlini@oracle.com>

* include/std/chrono (duration<>::operator%=,  operator%):
Add, per DR 934.
* testsuite/20_util/duration/arithmetic/dr934-1.cc: New.
* testsuite/20_util/duration/arithmetic/dr934-2.cc: Likewise.

* include/std/chrono (operator/): Simplify implementation.

From-SVN: r149856

15 years agosh.c (sh_gimplify_va_arg_expr): Wrap the result with a NOP_EXPR if needed.
Kaz Kojima [Tue, 21 Jul 2009 12:45:52 +0000 (12:45 +0000)]
sh.c (sh_gimplify_va_arg_expr): Wrap the result with a NOP_EXPR if needed.

* config/sh/sh.c (sh_gimplify_va_arg_expr): Wrap the result
with a NOP_EXPR if needed.

From-SVN: r149854

15 years agotree-vectorizer.c (increase_alignment): Handle nested arrays.
Paul Brook [Tue, 21 Jul 2009 12:35:38 +0000 (12:35 +0000)]
tree-vectorizer.c (increase_alignment): Handle nested arrays.

2009-07-21  Paul Brook <paul@codesourcery.com>

gcc/
* tree-vectorizer.c (increase_alignment): Handle nested arrays.
Terminate debug dump with newline.

gcc/testsuite/
* gcc.dg/vect/section-anchors-nest-1.c: New test.
* lib/target-supports.exp (check_effective_target_section_anchors):
Add arm*-*-*.

From-SVN: r149853

15 years agore PR fortran/40726 (miscompilation at -O1)
Paul Thomas [Tue, 21 Jul 2009 04:51:30 +0000 (04:51 +0000)]
re PR fortran/40726 (miscompilation at -O1)

2009-07-21  Paul Thomas  <pault@gcc.gnu.org>

PR fortran/40726
* trans-decl.c (gfc_get_extern_function_decl): Do not set
DECL_IS_MALLOC for pointer valued functions.
(build_function_decl): The same.

From-SVN: r149846

15 years agointro.xml: Escape '&', validate.
Benjamin Kosnik [Tue, 21 Jul 2009 04:28:41 +0000 (04:28 +0000)]
intro.xml: Escape '&', validate.

2009-07-20  Benjamin Kosnik  <bkoz@redhat.com>

* doc/xml/manual/intro.xml: Escape '&', validate.
* doc/xml/manual/using.xml: Validate, dead link check.
* doc/xml/manual/strings.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/iterators.xml: Same.
* doc/xml/manual/spine.xml: Same.

* doc/xml/faq.xml: Remove redundant xreflabel entities.
* doc/xml/gnu/gpl-3.0.xml: Same.
* doc/xml/manual/mt_allocator.xml: Same.
* doc/xml/manual/allocator.xml: Same.
* doc/xml/manual/ctype.xml: Same.
* doc/xml/manual/codecvt.xml: Same.
* doc/xml/manual/backwards_compatibility.xml: Same.
* doc/xml/manual/shared_ptr.xml: Same.
* doc/xml/manual/abi.xml: Same.
* doc/xml/manual/auto_ptr.xml: Same.
* doc/xml/manual/internals.xml: Same.
* doc/xml/manual/parallel_mode.xml: Same.
* doc/xml/manual/bitmap_allocator.xml: Same.
* doc/xml/manual/build_hacking.xml: Same.
* doc/xml/manual/evolution.xml: Same.
* doc/xml/manual/debug.xml: Same.
* doc/xml/manual/localization.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/locale.xml: Same.
* doc/xml/manual/messages.xml: Same.
* doc/xml/manual/spine.xml: Same.
* doc/xml/manual/test.xml: Same.
* doc/xml/book.txml: Same.
* doc/xml/spine.xml: Same.

* doc/xml/api.xml: Clean up ulink targets, convert to link if possible.
* doc/xml/manual/backwards_compatibility.xml: Same.
* doc/xml/manual/concurrency.xml: Same.
* doc/xml/manual/intro.xml: Same.
* doc/xml/manual/parallel_mode.xml: Same.
* doc/xml/manual/status_cxx1998.xml: Same.
* doc/xml/manual/containers.xml: Same.
* doc/xml/manual/io.xml: Same.
* doc/xml/manual/support.xml: Same.
* doc/xml/manual/strings.xml: Same.
* doc/xml/manual/debug_mode.xml: Same.
* doc/xml/manual/extensions.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/messages.xml: Same.
* doc/xml/manual/test.xml: Same.

* doc/xml/manual/abi.xml: Same.
* doc/xml/manual/auto_ptr.xml: Same.
* doc/xml/manual/parallel_mode.xml: Same.
* doc/xml/manual/build_hacking.xml: Same.
* doc/xml/manual/evolution.xml: Same.
* doc/xml/manual/using.xml: Same.
* doc/xml/manual/debug_mode.xml: Same.
* doc/xml/manual/extensions.xml: Same.
* doc/xml/manual/diagnostics.xml: Same.
* doc/xml/manual/algorithms.xml: Same.

* doc/html: Regenerate.

From-SVN: r149845

15 years agointro.xml: Escape '&', validate.
Benjamin Kosnik [Tue, 21 Jul 2009 02:47:00 +0000 (02:47 +0000)]
intro.xml: Escape '&', validate.

2009-07-20  Benjamin Kosnik  <bkoz@redhat.com>

* doc/xml/manual/intro.xml: Escape '&', validate.
* doc/xml/manual/using.xml: Validate, dead link check.
* doc/xml/manual/strings.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/iterators.xml: Same.
* doc/xml/manual/spine.xml: Same.

* doc/xml/faq.xml: Remove redundant xreflabel entities.
* doc/xml/gnu/gpl-3.0.xml: Same.
* doc/xml/manual/mt_allocator.xml: Same.
* doc/xml/manual/allocator.xml: Same.
* doc/xml/manual/ctype.xml: Same.
* doc/xml/manual/codecvt.xml: Same.
* doc/xml/manual/backwards_compatibility.xml: Same.
* doc/xml/manual/shared_ptr.xml: Same.
* doc/xml/manual/abi.xml: Same.
* doc/xml/manual/auto_ptr.xml: Same.
* doc/xml/manual/internals.xml: Same.
* doc/xml/manual/parallel_mode.xml: Same.
* doc/xml/manual/bitmap_allocator.xml: Same.
* doc/xml/manual/build_hacking.xml: Same.
* doc/xml/manual/evolution.xml: Same.
* doc/xml/manual/debug.xml: Same.
* doc/xml/manual/localization.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/locale.xml: Same.
* doc/xml/manual/messages.xml: Same.
* doc/xml/manual/spine.xml: Same.
* doc/xml/manual/test.xml: Same.
* doc/xml/book.txml: Same.
* doc/xml/spine.xml: Same.

* doc/xml/api.xml: Clean up ulink targets, convert to link if possible.
* doc/xml/manual/backwards_compatibility.xml: Same.
* doc/xml/manual/concurrency.xml: Same.
* doc/xml/manual/intro.xml: Same.
* doc/xml/manual/parallel_mode.xml: Same.
* doc/xml/manual/status_cxx1998.xml: Same.
* doc/xml/manual/containers.xml: Same.
* doc/xml/manual/io.xml: Same.
* doc/xml/manual/support.xml: Same.
* doc/xml/manual/strings.xml: Same.
* doc/xml/manual/debug_mode.xml: Same.
* doc/xml/manual/extensions.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/messages.xml: Same.
* doc/xml/manual/test.xml: Same.

* doc/html: Regenerate.

From-SVN: r149844

15 years agopa.c (compute_zdepwi_operands): Limit deposit length to 32 - lsb.
John David Anglin [Tue, 21 Jul 2009 00:43:42 +0000 (00:43 +0000)]
pa.c (compute_zdepwi_operands): Limit deposit length to 32 - lsb.

* pa.c (compute_zdepwi_operands): Limit deposit length to 32 - lsb.
Cast "1" to unsigned HOST_WIDE_INT.
(compute_zdepdi_operands): Limit maximum length to 64 bits.  Limit
deposit length to the maximum length - lsb.  Extend length if
HOST_BITS_PER_WIDE_INT is 32.

From-SVN: r149843

15 years agoDaily bump.
GCC Administrator [Tue, 21 Jul 2009 00:16:52 +0000 (00:16 +0000)]
Daily bump.

From-SVN: r149842

15 years agoAdd some dwarf4 values.
Cary Coutant [Mon, 20 Jul 2009 21:00:52 +0000 (21:00 +0000)]
Add some dwarf4 values.

* dwarf2.h (enum dwarf_tag): Add DW_TAG_type_unit.
(enum dwarf_form): Add DW_FORM_sec_offset, DW_FORM_exprloc,
DW_FORM_flag_present, DW_FORM_sig8.
(enum dwarf_attribute): Add DW_AT_signature.

From-SVN: r149836

15 years agointro.xml: Escape '&', validate.
Benjamin Kosnik [Mon, 20 Jul 2009 20:35:28 +0000 (20:35 +0000)]
intro.xml: Escape '&', validate.

2009-07-20  Benjamin Kosnik  <bkoz@redhat.com>

* doc/xml/manual/intro.xml: Escape '&', validate.
* doc/xml/manual/using.xml: Validate, dead link check.
* doc/xml/manual/strings.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/iterators.xml: Same.
* doc/xml/manual/spine.xml: Same.

* doc/xml/faq.xml: Remove redundant xreflabel entities.
* doc/xml/gnu/gpl-3.0.xml: Same.
* doc/xml/manual/mt_allocator.xml: Same.
* doc/xml/manual/allocator.xml: Same.
* doc/xml/manual/ctype.xml: Same.
* doc/xml/manual/codecvt.xml: Same.
* doc/xml/manual/backwards_compatibility.xml: Same.
* doc/xml/manual/shared_ptr.xml: Same.
* doc/xml/manual/abi.xml: Same.
* doc/xml/manual/auto_ptr.xml: Same.
* doc/xml/manual/internals.xml: Same.
* doc/xml/manual/parallel_mode.xml: Same.
* doc/xml/manual/bitmap_allocator.xml: Same.
* doc/xml/manual/build_hacking.xml: Same.
* doc/xml/manual/evolution.xml: Same.
* doc/xml/manual/debug.xml: Same.
* doc/xml/manual/localization.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/locale.xml: Same.
* doc/xml/manual/messages.xml: Same.
* doc/xml/manual/spine.xml: Same.
* doc/xml/manual/test.xml: Same.
* doc/xml/book.txml: Same.
* doc/xml/spine.xml: Same.

* doc/html: Regenerate.

From-SVN: r149835

15 years agocgraph.h (constant_pool_htab): New function.
Olatunji Ruwase [Mon, 20 Jul 2009 19:35:25 +0000 (19:35 +0000)]
cgraph.h (constant_pool_htab): New function.

* cgraph.h (constant_pool_htab): New function.
(constant_descriptor_tree): Move from varasm.c.
* varasm.c (constant_pool_htab): New function.
(constant_descriptor_tree): Move to cgraph.h.

From-SVN: r149834

15 years agotoplev.c: Invoke FINISH_UNIT callbacks before call to finalize().
Olatunji Ruwase [Mon, 20 Jul 2009 19:28:05 +0000 (19:28 +0000)]
toplev.c: Invoke FINISH_UNIT callbacks before call to finalize().

./: * toplev.c: Invoke FINISH_UNIT callbacks before call to
finalize().
testsuite/:
* gcc.dg/plugin/finish_unit_plugin.c: New test.
* gcc.dg/plugin/finish_unit-test-1.c: New test.
* gcc.dg/plugin/plugin.exp: Added finish_unit_plugin.c test.

From-SVN: r149833

15 years agoMakefile.in (TREE_INLINE_H, [...]): Remove $(VARRAY_H).
Shujing Zhao [Mon, 20 Jul 2009 19:20:33 +0000 (19:20 +0000)]
Makefile.in (TREE_INLINE_H, [...]): Remove $(VARRAY_H).

2009-07-20  Shujing Zhao  <pearly.zhao@oracle.com>

* Makefile.in (TREE_INLINE_H, tree-inline.o, cgraph.o): Remove
$(VARRAY_H).

From-SVN: r149832

15 years agointro.xml: Escape '&', validate.
Benjamin Kosnik [Mon, 20 Jul 2009 19:20:27 +0000 (19:20 +0000)]
intro.xml: Escape '&', validate.

2009-07-20  Benjamin Kosnik  <bkoz@redhat.com>

* doc/xml/manual/intro.xml: Escape '&', validate.
* doc/xml/manual/using.xml: Validate, dead link check.
* doc/xml/manual/strings.xml: Same.
* doc/xml/manual/appendix_contributing.xml: Same.
* doc/xml/manual/iterators.xml: Same.
* doc/xml/manual/spine.xml: Same.

* doc/html: Regenerate.

From-SVN: r149831

15 years agodbgcnt name matching bug fix
Xinliang David Li [Mon, 20 Jul 2009 19:08:15 +0000 (19:08 +0000)]
dbgcnt name matching bug fix

From-SVN: r149830

15 years agomips.md (move_type): Add arith.
Adam Nemet [Mon, 20 Jul 2009 19:01:45 +0000 (19:01 +0000)]
mips.md (move_type): Add arith.

* config/mips/mips.md (move_type): Add arith.
(type): Handle arith.
(zero_extendsidi2): Rename this into ...
(*zero_extendsidi2): ... this.  Don't match if ISA_HAS_EXT_INS.
(zero_extendsidi2): New expander.
(*zero_extendsidi2_dext): New pattern.

testsuite/
* gcc.target/mips/ext-3.c: New test.

From-SVN: r149829

15 years agoconfig.gcc (mips64-*-*): Add definition of tm_defines in order to set MIPS_ABI_DEFAULT.
Nick Clifton [Mon, 20 Jul 2009 18:24:19 +0000 (18:24 +0000)]
config.gcc (mips64-*-*): Add definition of tm_defines in order to set MIPS_ABI_DEFAULT.

        * config.gcc (mips64-*-*): Add definition of tm_defines in order
        to set MIPS_ABI_DEFAULT.
        * config/mips/vr.h (MIPS_ABI_DEFAULT): Remove definition.

From-SVN: r149827

15 years agore PR regression/40800 (libcpp breaks bootstrap)
Jerry Quinn [Mon, 20 Jul 2009 18:09:33 +0000 (18:09 +0000)]
re PR regression/40800 (libcpp breaks bootstrap)

2009-07-20  Jerry Quinn  <jlquinn@optonline.net>

    PR regression/40800
    * configure.ac: Use = instead of == for testing
    ENABLE_BUILD_WITH_CXX.
    * configure: Rebuild.

From-SVN: r149826

15 years agotree-object-size.c (addr_object_size): Handle unions with array in it as last field...
Jakub Jelinek [Mon, 20 Jul 2009 17:39:22 +0000 (19:39 +0200)]
tree-object-size.c (addr_object_size): Handle unions with array in it as last field of structs in __bos (...

* tree-object-size.c (addr_object_size): Handle unions with
array in it as last field of structs in __bos (, 1) as __bos (, 0).

* gcc.dg/builtin-object-size-6.c: Adjust expected values.
* gcc.dg/builtin-object-size-8.c: New test.
* g++.dg/ext/builtin-object-size1.C: Adjust expected values.
* g++.dg/ext/builtin-object-size2.C: Likewise.

From-SVN: r149825

15 years agore PR tree-optimization/40792 (internal error in com_mpi.c at -O3 for gcc 4.4.0)
Jakub Jelinek [Mon, 20 Jul 2009 14:38:46 +0000 (16:38 +0200)]
re PR tree-optimization/40792 (internal error in com_mpi.c at -O3 for gcc 4.4.0)

PR tree-optimization/40792
* tree.c (build_function_type_skip_args): Remove bogus assert.

* gcc.dg/pr40792.c: New test.

From-SVN: r149822

15 years ago[multiple changes]
Arnaud Charlet [Mon, 20 Jul 2009 13:56:52 +0000 (15:56 +0200)]
[multiple changes]

2009-07-20  Bob Duff  <duff@adacore.com>

* sem_ch13.adb (Analyze_Record_Representation_Clause): Use "and then"
instead of "and", because otherwise Parent_Last_Bit is read
uninitialized in the case where it's not a tagged type, or the tagged
parent does not have a complete rep clause.

2009-07-20  Robert Dewar  <dewar@adacore.com>

* stylesw.ads: Minor documentation change.

* types.ads: Minor reformatting

2009-07-20  Javier Miranda  <miranda@adacore.com>

* exp_disp.ads (Apply_Access_Checks): New subprogram that takes care of
generating the tag checks associated with dispatching calls.
* exp_disp.adb (Apply_Access_Checks): New subprogram.
(New_Value): This routine was previously local to expand dispatching
calls but it is now used also by Apply_Access_Checks.
(Expand_Dispatching_Calls): Cleanup code because the functionality of
tag checks is now provided by Apply_Access_Checks.
* exp_ch6.adb (Expand_Call): Incorporate generation of tag checks in
case of dispatching calls.

2009-07-20  Arnaud Charlet  <charlet@adacore.com>

* gnat1drv.adb (Gnat1drv): Also disable Elaboration_Check in
CodePeer_Mode.

From-SVN: r149821

15 years agoswitch-m.ads (Normalize_Compiler_Switches): Take into account switches -gnatw.?
Vincent Celier [Mon, 20 Jul 2009 13:51:39 +0000 (13:51 +0000)]
switch-m.ads (Normalize_Compiler_Switches): Take into account switches -gnatw.?

2009-07-20  Vincent Celier  <celier@adacore.com>

* switch-m.ads (Normalize_Compiler_Switches): Take into account
switches -gnatw.?

From-SVN: r149820

15 years ago[multiple changes]
Arnaud Charlet [Mon, 20 Jul 2009 13:48:01 +0000 (15:48 +0200)]
[multiple changes]

2009-07-20  Gary Dismukes  <dismukes@adacore.com>

* exp_prag.adb (Expand_Pragma_Import_Export_Exception): When compiling
for VMS, only rewrite the first component of the associated exception's
aggregate init (as 'V'), and eliminate the bogus rewrites of the second
and third components  that were being replaced with 'M' and 'S'.

2009-07-20  Arnaud Charlet  <charlet@adacore.com>

* gnat1drv.adb (Gnat1drv): Suppress access checks in CodePeer mode.
Also do not generate error when parsing a spec in CodePeer mode.

2009-07-20  Javier Miranda  <miranda@adacore.com>

* checks.adb (Apply_Access_Check): Avoid checks on availability of
runtime function Offset_To_Top_Ptr when compiling with no tagged
types expansion.
* exp_ch3.adb (Build_Init_Procedure): Leave open the possibility of
adding code to the init proc when compiling for VM backends.

2009-07-20  Vincent Celier  <celier@adacore.com>

* switch-m.adb (Normalize_Compiler_Switches): Take into account
switches -gnatw.?

2009-07-20  Thomas Quinot  <quinot@adacore.com>

* sem_dist.adb, exp_dist.adb: Minor reformatting

* Make-generated.in: New file.

* gcc-interface/Make-lang.in: Use Make-generated.in fragment.

From-SVN: r149819

15 years agosem_util.ads, [...] (Check_Dynamically_Tagged_Expression): New subprogram.
Javier Miranda [Mon, 20 Jul 2009 13:31:05 +0000 (13:31 +0000)]
sem_util.ads, [...] (Check_Dynamically_Tagged_Expression): New subprogram.

2009-07-20  Javier Miranda  <miranda@adacore.com>

* sem_util.ads, sem_util.adb (Check_Dynamically_Tagged_Expression): New
subprogram.
* sem_aggr.adb (Resolve_Array_Aggregate): Check incorrect use of
dynamically tagged expression.
* sem_ch3.adb (Analyze_Object_Declaration): Call new routine that
factorizes code.
* sem_ch6.adb (Analyze_Function_Return, Process_Formals): Ditto.
* sem_ch8.adb (Analyze_Object_Renaming): Ditto.

From-SVN: r149817

15 years ago[multiple changes]
Arnaud Charlet [Mon, 20 Jul 2009 13:27:46 +0000 (15:27 +0200)]
[multiple changes]

2009-07-20  Arnaud Charlet  <charlet@adacore.com>

* gnat1drv.adb (Gnat1drv): Set operating mode to Generate_Code when
CodePeer_Mode is set, to benefit from full front-end expansion
(e.g. generics).

2009-07-20  Ed Schonberg  <schonberg@adacore.com>

* sem_res.adb: Add guard.

* exp_disp.adb, sem_disp.adb (Make_DT): Check underlying view of type
for possible attribute definition of External_Tag, in case clause
appears in the private part of a package.

From-SVN: r149816

15 years ago* gcc-interface/Makefile.in: cleanup powerpc linux target pairs.
Jerome Guitton [Mon, 20 Jul 2009 13:23:20 +0000 (13:23 +0000)]
* gcc-interface/Makefile.in: cleanup powerpc linux target pairs.

From-SVN: r149813

15 years ago[multiple changes]
Arnaud Charlet [Mon, 20 Jul 2009 13:18:34 +0000 (15:18 +0200)]
[multiple changes]

2009-07-20  Vadim Godunko  <godunko@adacore.com>

* a-coorma.adb: Minor reformatting.

2009-07-20  Ed Schonberg  <schonberg@adacore.com>

* sem_ch3 (Build_Itype_Reference): Make public, for use on non-null
access return types.
* sem_ch6.adb (Analyze_Return_Type): If return is a not null subtype,
provide an itype reference to gigi to force elaboration of the subtype
at the proper point.

2009-07-20  Tristan Gingold  <gingold@adacore.com>

* g-expect.adb: Avoid closeing already closed handle.

2009-07-20  Robert Dewar  <dewar@adacore.com>

* sprint.adb (Write_Subprogram_Name): New procedure to output
subprogram name with possible preceding $ (replaces
Note_Implicit_Run_Time_Call).

From-SVN: r149812

15 years ago[multiple changes]
Arnaud Charlet [Mon, 20 Jul 2009 13:06:01 +0000 (15:06 +0200)]
[multiple changes]

2009-07-20  Robert Dewar  <dewar@adacore.com>

* vms_data.ads: Minor reformatting

* einfo.ads, einfo.adb (Parent_Subtype): Now allowed on record subtype,
applies to base type.
(Parent_Subtype): Now allowed on record subtype, applies to base type
* exp_ch5.adb (Expand_Assign_Record): Handle Componentwise_Assignment
for case of fully repped tagged type.
(Make_Tag_Ctrl_Assignment): Set Componentwise_Assignment and avoid
tag save/restore for fully repped tagged type case.
* exp_util.ads, exp_util.adb (Is_Fully_Repped_Tagged_Type): New function
* fe.h (Is_Fully_Repped_Tagged_Type): New function
* sem_ch13.adb (Analyze_Recorrd_Representation_Clause): Check for
overlap of tagged type components with parent type if parent type is
fully repped.
* sinfo.ads, sinfo.adb (Componentwise_Assignment): New flag

* sem_res.adb (Check_No_Direct_Boolean_Operators): Remove handling of
comparisons.
(Resolve_Comparison_Operators): Remove No_Direct_Boolean_Operators check
(Resolve_Equality_Op): Remove No_Direct_Boolean_Operators check

* gnat_rm.texi: Restriction No_Direct_Boolean_Operators includes only
logical operators (AND/OR/XOR), not comparison operators.

* sprint.ads: Minor reformatting

2009-07-20  Ed Schonberg  <schonberg@adacore.com>

* sem_intr.adb (Check_Intrinsic_Call): For Import_Value and related
intrinsics, check that argument is a string literal, rather than
checking for staticness.

From-SVN: r149811

15 years ago[multiple changes]
Arnaud Charlet [Mon, 20 Jul 2009 12:55:43 +0000 (14:55 +0200)]
[multiple changes]

2009-07-20  Robert Dewar  <dewar@adacore.com>

* sem_ch13.adb: Minor reformatting

* einfo.ads: Minor reformatting
Component_Bit_Offset is no longer considered obsolescent

2009-07-20  Nicolas Roche  <roche@adacore.com>

* a-calend.adb: Redefine time_t as signed integer with same size as
Address type.
* s-os_lib.ads: Redefine OS_Time as signed integer with same size as
Address type
* adaint.h: On Windows 64bits declare OS_Time as long long instead of
long

2009-07-20  Javier Miranda  <miranda@adacore.com>

* exp_tss.adb (Init_Proc): Add missing support for non-default C++
constructors that have anonymous access type formals.

* sem_res.adb (Resolve_Actuals): Disable checks associated with Ada
class-wide arguments in case of imported C++ subprograms.

* exp_ch3.adb (Build_Initialization_Call): Add assertion.

2009-07-20  Sergey Rybin  <rybin@adacore.com>

* vms_data.ads: Update qualifiers.

From-SVN: r149810

15 years agoeinfo.ads, [...]: Minor reformatting
Robert Dewar [Mon, 20 Jul 2009 12:47:50 +0000 (12:47 +0000)]
einfo.ads, [...]: Minor reformatting

2009-07-20  Robert Dewar  <dewar@adacore.com>

* einfo.ads, switch.adb, gnatls.adb, inline.adb, sem_ch13.adb: Minor
reformatting

From-SVN: r149809

15 years agocgraph.h (combined_args_to_skip): New field.
Jan Hubicka [Mon, 20 Jul 2009 12:15:02 +0000 (14:15 +0200)]
cgraph.h (combined_args_to_skip): New field.

2009-07-20  Jan Hubicka  <jh@suse.cz>
            Martin Jambor  <mjambor@suse.cz>

* cgraph.h (combined_args_to_skip): New field.
* cgraph.c (cgraph_create_virtual_clone): Properly handle
combined_args_to_skip and args_to_skip.
* tree-inline.c (update_clone_info): New function.
(tree_function_versioning): Call update_clone_info.
* cgraphunit.c: (cgraph_materialize_clone): Dump materialized functions.
(cgraph_materialize_all_clones): More extensive dumping, working
with combined_args_to_skip rather than args_to_skip.

Co-Authored-By: Martin Jambor <mjambor@suse.cz>
From-SVN: r149808

15 years ago* config.gcc (mips64vr-*-*): Use tm_defines to set MIPS_ABI_DEFAULT.
Nick Clifton [Mon, 20 Jul 2009 12:13:50 +0000 (12:13 +0000)]
* config.gcc (mips64vr-*-*): Use tm_defines to set MIPS_ABI_DEFAULT.

From-SVN: r149807

15 years agotree-vectorizer.h (vectorizable_condition): Add parameters.
Ira Rosen [Mon, 20 Jul 2009 11:59:10 +0000 (11:59 +0000)]
tree-vectorizer.h (vectorizable_condition): Add parameters.

* tree-vectorizer.h (vectorizable_condition): Add parameters.
* tree-vect-loop.c (vect_is_simple_reduction): Support COND_EXPR.
(get_initial_def_for_reduction): Likewise.
(vectorizable_reduction): Skip the check of first operand in case
of COND_EXPR. Add check that it is outer loop vectorization if
nested cycle was detected. Call vectorizable_condition() for
COND_EXPR. If reduction epilogue cannot be created do not fail for
nested cycles (if it is not double reduction). Assert that there
is only one type in the loop in case of COND_EXPR. Call
vectorizable_condition() to vectorize COND_EXPR.
* tree-vect-stmts.c (vectorizable_condition): Update comment.
Add parameters. Allow nested cycles if called from
vectorizable_reduction(). Use reduction vector variable if provided.
(vect_analyze_stmt): Call vectorizable_reduction() before
vectorizable_condition().
(vect_transform_stmt): Update call to vectorizable_condition().

From-SVN: r149806

15 years agoChangeLog: Remove conflict markers.
Uros Bizjak [Mon, 20 Jul 2009 10:38:04 +0000 (12:38 +0200)]
ChangeLog: Remove conflict markers.

* ChangeLog: Remove conflict markers.
* testsuite/ChangeLog: Ditto.

From-SVN: r149805

15 years agoSH: resurect -mfmovd
Christian Bruel [Mon, 20 Jul 2009 07:37:37 +0000 (09:37 +0200)]
SH: resurect -mfmovd

From-SVN: r149803

15 years agoDaily bump.
GCC Administrator [Mon, 20 Jul 2009 00:16:33 +0000 (00:16 +0000)]
Daily bump.

From-SVN: r149800

15 years agore PR fortran/40714 (Fortran runtime error: Invalid argument)
Janne Blomqvist [Sun, 19 Jul 2009 23:26:20 +0000 (02:26 +0300)]
re PR fortran/40714 (Fortran runtime error: Invalid argument)

2009-07-19  Janne Blomqvist  <jb@gcc.gnu.org>
    Jerry DeLisle  <jvdelisle@gcc.gnu.org>

PR libfortran/40714
* gfortran.dg/eof_3.f90: New test.

Co-Authored-By: Jerry DeLisle <jvdelisle@gcc.gnu.org>
From-SVN: r149797

15 years agotree-ssa-dce.c (remove_dead_phis): Only look for abnormal PHIs when handling SSA...
Jan Hubicka [Sun, 19 Jul 2009 22:57:36 +0000 (00:57 +0200)]
tree-ssa-dce.c (remove_dead_phis): Only look for abnormal PHIs when handling SSA name.

* tree-ssa-dce.c (remove_dead_phis): Only look for abnormal PHIs
when handling SSA name.

From-SVN: r149794

15 years agore PR fortran/40727 ([4.4] ICE gfc_simplify_dcmplx(): Bad type when passing BT_COMPLE...
Steven G. Kargl [Sun, 19 Jul 2009 15:37:50 +0000 (15:37 +0000)]
re PR fortran/40727 ([4.4] ICE gfc_simplify_dcmplx(): Bad type when passing BT_COMPLEX to cmplx)

2009-07-18  Steven G. Kargl  <kargl@gcc.gnu.org>

PR fortran/40727
* fortran/check.c (gfc_check_cmplx, gfc_check_dcmplx): Add check that
the optional second argument isn't of COMPLEX type.

2009-07-18  Steven G. Kargl  <kargl@gcc.gnu.org>

PR fortran/40727
* gfortran.dg/intrinsic_cmplx.f90: New test.

From-SVN: r149793

15 years ago[multiple changes]
Thomas Koenig [Sun, 19 Jul 2009 15:07:21 +0000 (15:07 +0000)]
[multiple changes]

2009-07-19  Thomas Koenig  <tkoenig@gcc.gnu.org>

PR libfortran/34670
PR libfortran/36874
* Makefile.am:  Add bounds.c
* libgfortran.h (bounds_equal_extents):  Add prototype.
(bounds_iforeach_return):  Likewise.
(bounds_ifunction_return):  Likewise.
(bounds_reduced_extents):  Likewise.
* runtime/bounds.c:  New file.
(bounds_iforeach_return):  New function; correct typo in
error message.
(bounds_ifunction_return):  New function.
(bounds_equal_extents):  New function.
(bounds_reduced_extents):  Likewise.
* intrinsics/cshift0.c (cshift0):  Use new functions
for bounds checking.
* intrinsics/eoshift0.c (eoshift0):  Likewise.
* intrinsics/eoshift2.c (eoshift2):  Likewise.
* m4/iforeach.m4:  Likewise.
* m4/eoshift1.m4:  Likewise.
* m4/eoshift3.m4:  Likewise.
* m4/cshift1.m4:  Likewise.
* m4/ifunction.m4:  Likewise.
* Makefile.in:  Regenerated.
* generated/cshift1_16.c: Regenerated.
* generated/cshift1_4.c: Regenerated.
* generated/cshift1_8.c: Regenerated.
* generated/eoshift1_16.c: Regenerated.
* generated/eoshift1_4.c: Regenerated.
* generated/eoshift1_8.c: Regenerated.
* generated/eoshift3_16.c: Regenerated.
* generated/eoshift3_4.c: Regenerated.
* generated/eoshift3_8.c: Regenerated.
* generated/maxloc0_16_i1.c: Regenerated.
* generated/maxloc0_16_i16.c: Regenerated.
* generated/maxloc0_16_i2.c: Regenerated.
* generated/maxloc0_16_i4.c: Regenerated.
* generated/maxloc0_16_i8.c: Regenerated.
* generated/maxloc0_16_r10.c: Regenerated.
* generated/maxloc0_16_r16.c: Regenerated.
* generated/maxloc0_16_r4.c: Regenerated.
* generated/maxloc0_16_r8.c: Regenerated.
* generated/maxloc0_4_i1.c: Regenerated.
* generated/maxloc0_4_i16.c: Regenerated.
* generated/maxloc0_4_i2.c: Regenerated.
* generated/maxloc0_4_i4.c: Regenerated.
* generated/maxloc0_4_i8.c: Regenerated.
* generated/maxloc0_4_r10.c: Regenerated.
* generated/maxloc0_4_r16.c: Regenerated.
* generated/maxloc0_4_r4.c: Regenerated.
* generated/maxloc0_4_r8.c: Regenerated.
* generated/maxloc0_8_i1.c: Regenerated.
* generated/maxloc0_8_i16.c: Regenerated.
* generated/maxloc0_8_i2.c: Regenerated.
* generated/maxloc0_8_i4.c: Regenerated.
* generated/maxloc0_8_i8.c: Regenerated.
* generated/maxloc0_8_r10.c: Regenerated.
* generated/maxloc0_8_r16.c: Regenerated.
* generated/maxloc0_8_r4.c: Regenerated.
* generated/maxloc0_8_r8.c: Regenerated.
* generated/maxloc1_16_i1.c: Regenerated.
* generated/maxloc1_16_i16.c: Regenerated.
* generated/maxloc1_16_i2.c: Regenerated.
* generated/maxloc1_16_i4.c: Regenerated.
* generated/maxloc1_16_i8.c: Regenerated.
* generated/maxloc1_16_r10.c: Regenerated.
* generated/maxloc1_16_r16.c: Regenerated.
* generated/maxloc1_16_r4.c: Regenerated.
* generated/maxloc1_16_r8.c: Regenerated.
* generated/maxloc1_4_i1.c: Regenerated.
* generated/maxloc1_4_i16.c: Regenerated.
* generated/maxloc1_4_i2.c: Regenerated.
* generated/maxloc1_4_i4.c: Regenerated.
* generated/maxloc1_4_i8.c: Regenerated.
* generated/maxloc1_4_r10.c: Regenerated.
* generated/maxloc1_4_r16.c: Regenerated.
* generated/maxloc1_4_r4.c: Regenerated.
* generated/maxloc1_4_r8.c: Regenerated.
* generated/maxloc1_8_i1.c: Regenerated.
* generated/maxloc1_8_i16.c: Regenerated.
* generated/maxloc1_8_i2.c: Regenerated.
* generated/maxloc1_8_i4.c: Regenerated.
* generated/maxloc1_8_i8.c: Regenerated.
* generated/maxloc1_8_r10.c: Regenerated.
* generated/maxloc1_8_r16.c: Regenerated.
* generated/maxloc1_8_r4.c: Regenerated.
* generated/maxloc1_8_r8.c: Regenerated.
* generated/maxval_i1.c: Regenerated.
* generated/maxval_i16.c: Regenerated.
* generated/maxval_i2.c: Regenerated.
* generated/maxval_i4.c: Regenerated.
* generated/maxval_i8.c: Regenerated.
* generated/maxval_r10.c: Regenerated.
* generated/maxval_r16.c: Regenerated.
* generated/maxval_r4.c: Regenerated.
* generated/maxval_r8.c: Regenerated.
* generated/minloc0_16_i1.c: Regenerated.
* generated/minloc0_16_i16.c: Regenerated.
* generated/minloc0_16_i2.c: Regenerated.
* generated/minloc0_16_i4.c: Regenerated.
* generated/minloc0_16_i8.c: Regenerated.
* generated/minloc0_16_r10.c: Regenerated.
* generated/minloc0_16_r16.c: Regenerated.
* generated/minloc0_16_r4.c: Regenerated.
* generated/minloc0_16_r8.c: Regenerated.
* generated/minloc0_4_i1.c: Regenerated.
* generated/minloc0_4_i16.c: Regenerated.
* generated/minloc0_4_i2.c: Regenerated.
* generated/minloc0_4_i4.c: Regenerated.
* generated/minloc0_4_i8.c: Regenerated.
* generated/minloc0_4_r10.c: Regenerated.
* generated/minloc0_4_r16.c: Regenerated.
* generated/minloc0_4_r4.c: Regenerated.
* generated/minloc0_4_r8.c: Regenerated.
* generated/minloc0_8_i1.c: Regenerated.
* generated/minloc0_8_i16.c: Regenerated.
* generated/minloc0_8_i2.c: Regenerated.
* generated/minloc0_8_i4.c: Regenerated.
* generated/minloc0_8_i8.c: Regenerated.
* generated/minloc0_8_r10.c: Regenerated.
* generated/minloc0_8_r16.c: Regenerated.
* generated/minloc0_8_r4.c: Regenerated.
* generated/minloc0_8_r8.c: Regenerated.
* generated/minloc1_16_i1.c: Regenerated.
* generated/minloc1_16_i16.c: Regenerated.
* generated/minloc1_16_i2.c: Regenerated.
* generated/minloc1_16_i4.c: Regenerated.
* generated/minloc1_16_i8.c: Regenerated.
* generated/minloc1_16_r10.c: Regenerated.
* generated/minloc1_16_r16.c: Regenerated.
* generated/minloc1_16_r4.c: Regenerated.
* generated/minloc1_16_r8.c: Regenerated.
* generated/minloc1_4_i1.c: Regenerated.
* generated/minloc1_4_i16.c: Regenerated.
* generated/minloc1_4_i2.c: Regenerated.
* generated/minloc1_4_i4.c: Regenerated.
* generated/minloc1_4_i8.c: Regenerated.
* generated/minloc1_4_r10.c: Regenerated.
* generated/minloc1_4_r16.c: Regenerated.
* generated/minloc1_4_r4.c: Regenerated.
* generated/minloc1_4_r8.c: Regenerated.
* generated/minloc1_8_i1.c: Regenerated.
* generated/minloc1_8_i16.c: Regenerated.
* generated/minloc1_8_i2.c: Regenerated.
* generated/minloc1_8_i4.c: Regenerated.
* generated/minloc1_8_i8.c: Regenerated.
* generated/minloc1_8_r10.c: Regenerated.
* generated/minloc1_8_r16.c: Regenerated.
* generated/minloc1_8_r4.c: Regenerated.
* generated/minloc1_8_r8.c: Regenerated.
* generated/minval_i1.c: Regenerated.
* generated/minval_i16.c: Regenerated.
* generated/minval_i2.c: Regenerated.
* generated/minval_i4.c: Regenerated.
* generated/minval_i8.c: Regenerated.
* generated/minval_r10.c: Regenerated.
* generated/minval_r16.c: Regenerated.
* generated/minval_r4.c: Regenerated.
* generated/minval_r8.c: Regenerated.
* generated/product_c10.c: Regenerated.
* generated/product_c16.c: Regenerated.
* generated/product_c4.c: Regenerated.
* generated/product_c8.c: Regenerated.
* generated/product_i1.c: Regenerated.
* generated/product_i16.c: Regenerated.
* generated/product_i2.c: Regenerated.
* generated/product_i4.c: Regenerated.
* generated/product_i8.c: Regenerated.
* generated/product_r10.c: Regenerated.
* generated/product_r16.c: Regenerated.
* generated/product_r4.c: Regenerated.
* generated/product_r8.c: Regenerated.
* generated/sum_c10.c: Regenerated.
* generated/sum_c16.c: Regenerated.
* generated/sum_c4.c: Regenerated.
* generated/sum_c8.c: Regenerated.
* generated/sum_i1.c: Regenerated.
* generated/sum_i16.c: Regenerated.
* generated/sum_i2.c: Regenerated.
* generated/sum_i4.c: Regenerated.
* generated/sum_i8.c: Regenerated.
* generated/sum_r10.c: Regenerated.
* generated/sum_r16.c: Regenerated.
* generated/sum_r4.c: Regenerated.
* generated/sum_r8.c: Regenerated.

2009-07-19   Thomas Koenig  <tkoenig@gcc.gnu.org>

PR libfortran/34670
PR libfortran/36874
* gfortran.dg/cshift_bounds_1.f90:  New test.
* gfortran.dg/cshift_bounds_2.f90:  New test.
* gfortran.dg/cshift_bounds_3.f90:  New test.
* gfortran.dg/cshift_bounds_4.f90:  New test.
* gfortran.dg/eoshift_bounds_1.f90:  New test.
* gfortran.dg/maxloc_bounds_4.f90:  Correct typo in error message.
* gfortran.dg/maxloc_bounds_5.f90:  Correct typo in error message.
* gfortran.dg/maxloc_bounds_7.f90:  Correct typo in error message.

From-SVN: r149792

15 years agore PR tree-optimization/40676 (internal compiler error: verify_ssa error: definition...
Jan Hubicka [Sun, 19 Jul 2009 10:27:07 +0000 (12:27 +0200)]
re PR tree-optimization/40676 (internal compiler error: verify_ssa error: definition in block 5 does not dominate use in block 7)

PR tree-optimization/40676
* tree-ssa-dce.c (eliminate_unnecessary_stmts): Do renaming on all
virtual PHIs in empty BBs.

* gcc.c-torture/compile/pr40676.c: New testcase.

From-SVN: r149789

15 years agoDaily bump.
GCC Administrator [Sun, 19 Jul 2009 00:16:50 +0000 (00:16 +0000)]
Daily bump.

From-SVN: r149785

15 years agore PR libstdc++/40712 (locale(const locale&, const char*, locale::category) can creat...
Paolo Carlini [Sat, 18 Jul 2009 22:58:10 +0000 (22:58 +0000)]
re PR libstdc++/40712 (locale(const locale&, const char*, locale::category) can create broken locale)

2009-07-18  Paolo Carlini  <paolo.carlini@oracle.com>

PR libstdc++/40712
* config/locale/gnu/numeric_members.cc (numpunct<>::
_M_initialize_numpunct): Dynamically allocate _M_data->_M_grouping
and copy the langinfo data into it.
(numpunct<>::~numpunct): Free the allocated memory.
* config/locale/gnu/monetary_members.cc (moneypunct<>::
_M_initialize_moneypunct): Dynamically allocate _M_data->_M_grouping,
_M_data->_M_positive_sign, _M_data->_M_negative_sign,
_M_data->_M_curr_symbol.
(moneypunct<>::~moneypunct): Free the allocated memory.
* testsuite/22_locale/moneypunct/40712.cc: New.

* include/bits/locale_facets.tcc (__numpunct_cache<>::_M_cache):
Do not leak memory if new throws.
* include/bits/locale_facets_nonio.tcc
(__moneypunct_cache<>::_M_cache): Likewise.

From-SVN: r149782

15 years agocombine.c (make_compound_operation): If force_to_mode re-expanded the compound use...
Adam Nemet [Sat, 18 Jul 2009 21:55:40 +0000 (21:55 +0000)]
combine.c (make_compound_operation): If force_to_mode re-expanded the compound use gen_lowpart instead to convert to...

* combine.c (make_compound_operation) <SUBREG>: If force_to_mode
re-expanded the compound use gen_lowpart instead to convert to the
desired mode.

testsuite/
* gcc.target/mips/ext-2.c: New test.

From-SVN: r149781

15 years agocombine.c (try_widen_shift_mode): Add COUNT, OUTER_CODE and OUTER_CONST arguments.
Adam Nemet [Sat, 18 Jul 2009 21:52:48 +0000 (21:52 +0000)]
combine.c (try_widen_shift_mode): Add COUNT, OUTER_CODE and OUTER_CONST arguments.

* combine.c (try_widen_shift_mode): Add COUNT, OUTER_CODE and
OUTER_CONST arguments.
<LSHIFTRT>: Use them to allow widening if the bits shifted in from
the new wider mode will be masked off.
(simplify_shift_const_1): Adjust calls to try_widen_shift_mode.

From-SVN: r149780