Vincent Celier [Wed, 3 Aug 2011 12:36:13 +0000 (12:36 +0000)]
make.adb (Gnatmake): Add to table Library_Projs only library projects that are not extended.
2011-08-03 Vincent Celier <celier@adacore.com>
* make.adb (Gnatmake): Add to table Library_Projs only library projects
that are not extended. Do not check if a library project that is
extended should be regenerated. Do not add to table Library_Paths the
libraries that are in extended projects. Link only with libraries that
are in libray projects that are not extended.
* prj-nmsc.adb (Check_Library_Attributes): Do not reset flag Library
for a library project that is extended.
From-SVN: r177269
Alan Modra [Wed, 3 Aug 2011 11:48:21 +0000 (21:18 +0930)]
re PR rtl-optimization/49941 (segmentation fault in redirect_jump_2)
PR rtl-optimization/49941
* jump.c (mark_jump_label): Comment.
(mark_jump_label_1): Set JUMP_LABEL for return jumps.
* emit-rtl.c (copy_rtx_if_shared_1, copy_insn_1): Leave RETURN shared.
(mark_used_flags): Don't mark RETURN.
From-SVN: r177268
Richard Guenther [Wed, 3 Aug 2011 11:09:35 +0000 (11:09 +0000)]
re PR middle-end/49938 (ICE in interpret_loop_phi, at tree-scalar-evolution.c:1645)
2011-08-03 Richard Guenther <rguenther@suse.de>
PR tree-optimization/49938
* tree-scalar-evolution.c (interpret_loop_phi): Gracefully
deal with a POLYNOMIAL_CHREC.
* g++.dg/torture/pr49938.C: New testcase.
From-SVN: r177267
Arnaud Charlet [Wed, 3 Aug 2011 10:50:14 +0000 (12:50 +0200)]
[multiple changes]
2011-08-03 Gary Dismukes <dismukes@adacore.com>
* sem_ch6.adb (Find_Corresponding_Spec): When in an instance, skip
conforming subprogram renamings that appear to be completions if they
are not fully conformant.
Such renamings are homographs but not completions.
* sem_type.adb (Disambiguate): Handle disambiguation of overloaded
names in a subprogram renaming that appears in an instance.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* exp_util.adb (Expand_Subtype_From_Expr): if the type is limited but
not immutably limited, build actual subtype from expression to provide
proper bounds to caller.
2011-08-03 Gary Dismukes <dismukes@adacore.com>
* sem_ch8.adb: Minor comment correction.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* exp_strm.adb (Build_Array_Input_Function): In Ada 2005 mode, when
returning a limited array, use an extended return statement.
2011-08-03 Vincent Celier <celier@adacore.com>
* make.adb (Initialize): If --subdirs= is used, but no project file is
specified, attempt to create the specify subdir if it does not already
exist and use it as the object directory as if -D had been specified.
2011-08-03 Arnaud Charlet <charlet@adacore.com>
* s-tpopsp-vms.adb: New file.
* s-taprop-vms.adb: Put back ATCB_Key, since needed by this file on VMS.
* gcc-interfaces/Makefile.in: Use s-taprop-vms.adb on VMS.
From-SVN: r177266
Uros Bizjak [Wed, 3 Aug 2011 10:46:04 +0000 (12:46 +0200)]
proc.h: New.
* config/linux/proc.h: New.
* config/linux/proc.c: Include "proc.h". Do not include <sched.h>.
(gomp_cpuset_popcount): Rename from cpuset_popcount. No more static.
(gomp_init_num_threads): Update call to cpuset_popcount.
(get_num_procs): Ditto.
* config/linux/affinity.c (gomp_init_affinity): Call
gomp_cpuset_popcount.
From-SVN: r177265
Arnaud Charlet [Wed, 3 Aug 2011 10:45:20 +0000 (12:45 +0200)]
[multiple changes]
2011-08-03 Emmanuel Briot <briot@adacore.com>
* make.adb, makeutl.adb, makeutl.ads, clean.adb (Makeutl.Queue): new
package.
2011-08-03 Yannick Moy <moy@adacore.com>
* cstand.adb (Create_Standard): select Universal_Integer as an ALFA type
* sem_ch3.adb (Array_Type_Declaration): detect array types in ALFA
* sem_util.adb, sem_util.ads (Has_Static_Array_Bounds): new function to
detect that an array has static bounds.
From-SVN: r177264
Arnaud Charlet [Wed, 3 Aug 2011 10:42:00 +0000 (12:42 +0200)]
[multiple changes]
2011-08-03 Thomas Quinot <quinot@adacore.com>
* exp_dist.adb: Minor reformatting.
2011-08-03 Arnaud Charlet <charlet@adacore.com>
* s-taprop-irix.adb, s-taprop-hpux-dce.adb, s-taprop-vms.adb
(ATCB_Key): Removed, not always used.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* gnatcmd.adb, make.adb, prj.adb, prj.ads, makeutl.adb, makeutl.ads,
clean.adb, prj-conf.adb, prj-env.adb, prj-env.ads (Makeutl): remove
most remaining global variables.
From-SVN: r177263
Arnaud Charlet [Wed, 3 Aug 2011 10:38:26 +0000 (12:38 +0200)]
[multiple changes]
2011-08-03 Robert Dewar <dewar@adacore.com>
* gnatcmd.adb, prj-proc.adb, mlib-prj.adb, prj.adb, makeutl.ads,
prj-util.adb, prj-util.ads, prj-conf.adb, prj-env.adb: Minor
reformatting.
2011-08-03 Javier Miranda <miranda@adacore.com>
* exp_util.adb (Is_VM_By_Copy_Actual): Include N_Slide nodes as actuals
that must be passed by copy in VM targets.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj.ads, prj-nmsc.adb (Files_Htable): removed this htable, which
duplicates a similar htable now in the project tree.
2011-08-03 Claire Dross <dross@adacore.com>
* a-cfdlli.adb, a-cfhama.adb, a-cfhase.adb, a-cforma.adb, a-cforse.adb,
a-cofove.adb ("=", Length, Is_Empty, Clear, Assign, Copy, Element,
Replace_Element, Query_Element, Update_Element, Move, Insert, Prepend,
Append, Delete, Delete_First, Delete_Last, Reverse_Element, Swap,
Splice, First, First_Element, Last, Last_Element, Next, Previous, Find,
Reverse_Find, Contains, Has_Element, Iterate, Reverse_Iterate, Capacity,
Reserve_Length, Length, Strict_Equal, Left, Right): Data-structure
update.
2011-08-03 Arnaud Charlet <charlet@adacore.com>
* s-taprop-posix.adb, s-taprop-linux.adb, s-taprop-tru64.adb
(ATCB_Key): Removed, not always used.
* s-tpopsp-posix.adb, s-tpopsp-posix-foreign.adb (ATCB_Key): Moved from
s-taprop-posix.adb.
* s-tpopsp-tls.adb: New file.
* gcc-interface/Makefile.in: Use TLS implementation of s-tpopsp.adb on
x86/x64/ia64/powerpc/sparc Linux.
2011-08-03 Arnaud Charlet <charlet@adacore.com>
* system-aix.ads, system-aix64.ads: Set ZCX_By_Default to True.
* gcc-interface/Makefile.in: Switch to ZCX by default on AIX ports.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* rtsfind.ads, exp_dist.adb, exp_dist.ads
(Build_From_Any_Call, Build_To_Any_Call, Build_TypeCode_Call):
Fix type selection for mapping integer types to PolyORB types.
2011-08-03 Bob Duff <duff@adacore.com>
* sem_ch7.adb: Minor comment clarification.
2011-08-03 Bob Duff <duff@adacore.com>
* sem_ch13.adb (Analyze_Enumeration_Representation_Clause): If we get
an error analyzing a choice, skip further processing. Further
processing could cause a crash or cascade errors.
From-SVN: r177262
Emmanuel Briot [Wed, 3 Aug 2011 10:19:32 +0000 (10:19 +0000)]
gnatcmd.adb, [...] (Shared_Project_Tree_Data): new type An aggregate project and its aggregated trees need to share the common...
2011-08-03 Emmanuel Briot <briot@adacore.com>
* gnatcmd.adb, prj-proc.adb, prj-proc.ads, make.adb, mlib-prj.adb,
prj.adb, prj.ads, makeutl.adb, makeutl.ads, clean.adb, prj-nmsc.adb,
prj-util.adb, prj-util.ads, prj-conf.adb, prj-conf.ads, prj-env.adb,
prj-env.ads (Shared_Project_Tree_Data): new type
An aggregate project and its aggregated trees need to share the common
data structures used for lists of strings, packages,... This makes the
code simpler since otherwise we have to pass the root tree (also used
for the configuration file data) in addition to the current project
tree. This also avoids ambiguities as to which tree should be used.
And finally this saves a bit of memory.
(For_Every_Project_Imported): new parameter Tree.
Since aggregated projects are using a different tree, we need to let
the caller know which tree to use to manipulate the returned project.
From-SVN: r177261
Robert Dewar [Wed, 3 Aug 2011 10:17:45 +0000 (10:17 +0000)]
prj-proc.adb, [...]: Minor reformatting.
2011-08-03 Robert Dewar <dewar@adacore.com>
* prj-proc.adb, exp_util.ads, exp_ch9.adb, make.adb, prj-ext.adb,
prj-ext.ads, alfa.ads, sem_ch4.adb, makeutl.adb, makeutl.ads,
lib-xref-alfa.adb, sem_cat.adb, exp_dist.adb, get_alfa.adb,
prj-env.adb, prj-env.ads, prj-tree.adb, alfa.ads: Minor reformatting.
From-SVN: r177260
Arnaud Charlet [Wed, 3 Aug 2011 10:10:40 +0000 (12:10 +0200)]
[multiple changes]
2011-08-03 Javier Miranda <miranda@adacore.com>
* exp_util.adb, sem_aux.adb, exp_util.ads, sem_aux.ads:
Move routine Is_VM_By_Copy_Actual from sem_aux to exp_util.
2011-08-03 Bob Duff <duff@adacore.com>
* sem_ch8.adb (Is_Primitive_Operator_In_Use): Always check In_Use on
the Base_Type.
2011-08-03 Joel Brobecker <brobecker@adacore.com brobecker>
* g-sha256.ads, g-sha512.ads, g-sha1.ads, g-sha224.ads, g-sha384.ads,
g-sehash.ads: Fix typo. Update header.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* exp_ch6.adb (Expand_Actuals): Remove Ada_2005 guard on build-in-place
expansion.
2011-08-03 Gary Dismukes <dismukes@adacore.com>
* sem_ch4.adb (Analyze_Overloaded_Selected_Component): Consider
prefixes of private types along with records, since the selector may be
a discriminant.
* sem_res.adb (Resolve_Selected_Component): Consider prefixes of
private types along with records, since the selector may be a
discriminant.
From-SVN: r177259
Arnaud Charlet [Wed, 3 Aug 2011 10:04:04 +0000 (12:04 +0200)]
[multiple changes]
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-pp.adb (Pretty_Print): remove trailing blank line in the file
2011-08-03 Javier Miranda <miranda@adacore.com>
* sem_aux.adb (Is_VM_By_Copy_Actual): Fix wrong use of
Tagged_Type_Expansion.
2011-08-03 Yannick Moy <moy@adacore.com>
* sem_res.adb
(Resolve_Op_Concat_Arg): do not issue a SPARK violation when argument of
concatenation is the name of a constant
From-SVN: r177258
Emmanuel Briot [Wed, 3 Aug 2011 10:01:51 +0000 (10:01 +0000)]
[multiple changes]
2011-08-03 Yannick Moy <moy@adacore.com>
* alfa.ads Update format of ALFA section in ALI file in order to add a
mapping from bodies to specs when both are present
(ALFA_Scope_Record): add components for spec file/scope
* get_alfa.adb (Get_ALFA): read the new file/scope for spec when present
* lib-xref-alfa.adb
(Collect_ALFA): after all scopes have been collected, fill in the spec
information when relevant
* put_alfa.adb (Put_ALFA): write the new file/scope for spec when
present.
2011-08-03 Eric Botcazou <ebotcazou@adacore.com>
* inline.adb (Add_Inlined_Subprogram): Do not consider the enclosing
code unit to decide whether to add internally generated subprograms.
2011-08-03 Javier Miranda <miranda@adacore.com>
* sem_aux.ads, sem_aux.adb (Is_VM_By_Copy_Actual): New subprogram.
* exp_ch9.adb
(Build_Simple_Entry_Call): Handle actuals that must be handled by copy
in VM targets.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* make.adb, makeutl.adb, makeutl.ads (Make.Switches_Of): now shares
code with Makeutl.Get_Switches.
* prj-tree.adb: Update comment.
From-SVN: r177257
Arnaud Charlet [Wed, 3 Aug 2011 09:59:55 +0000 (11:59 +0200)]
[multiple changes]
2011-08-03 Yannick Moy <moy@adacore.com>
* alfa.ads Update format of ALFA section in ALI file in order to add a
mapping from bodies to specs when both are present
(ALFA_Scope_Record): add components for spec file/scope
* get_alfa.adb (Get_ALFA): read the new file/scope for spec when present
* lib-xref-alfa.adb
(Collect_ALFA): after all scopes have been collected, fill in the spec
information when relevant
* put_alfa.adb (Put_ALFA): write the new file/scope for spec when
present.
2011-08-03 Eric Botcazou <ebotcazou@adacore.com>
* inline.adb (Add_Inlined_Subprogram): Do not consider the enclosing
code unit to decide whether to add internally generated subprograms.
2011-08-03 Javier Miranda <miranda@adacore.com>
* sem_aux.ads, sem_aux.adb (Is_VM_By_Copy_Actual): New subprogram.
* exp_ch9.adb
(Build_Simple_Entry_Call): Handle actuals that must be handled by copy
in VM targets.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* make.adb, makeutl.adb, makeutl.ads (Make.Switches_Of): now shares
code with Makeutl.Get_Switches.
* prj-tree.adb: Update comment.
From-SVN: r177256
Arnaud Charlet [Wed, 3 Aug 2011 09:56:17 +0000 (11:56 +0200)]
[multiple changes]
2011-08-03 Eric Botcazou <ebotcazou@adacore.com>
* inline.adb: Revert previous change.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* sem_cat.adb (Validate_RCI_Subprogram_Declaration): Reject a remote
subprogram with a limited formal that does not support external
streaming.
2011-08-03 Yannick Moy <moy@adacore.com>
* get_alfa.adb (Get_ALFA): add missing Skip_Spaces at start of
continuation line
* lib-xref-alfa.adb (Add_ALFA_File): split removal of scopes that are
not from current unit in two phases, because it is not possible to
change the table while iterating over its content.
* put_alfa.adb (Put_ALFA): reset current file/scope at each new entity
2011-08-03 Sergey Rybin <rybin@adacore.com>
* vms_data.ads: Add qualifier for gnatmetric --no-static-loop option
* gnat_ugn.texi: Update description of complexity metrics (gnatmetric)
From-SVN: r177255
Ed Schonberg [Wed, 3 Aug 2011 09:53:11 +0000 (09:53 +0000)]
inline.adb: proper handling of init_procs.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* inline.adb: proper handling of init_procs.
* sem_res.adb (Resolve_Op_Concat_Arg): if the argument is an aggregate
and the component type is composite, this is ambiguous for predefined
concatenation, but if the node is not overloaded and the entity is a use
-defined function its profile can be used to resolve that aggregate.
From-SVN: r177254
Arnaud Charlet [Wed, 3 Aug 2011 09:51:09 +0000 (11:51 +0200)]
[multiple changes]
2011-08-03 Robert Dewar <dewar@adacore.com>
* exp_ch4.adb: Minor code cleanup.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* exp_dist.adb: Do not generate distribution stubs if serious errors
have been diagnosed.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_ch4.adb (Try_Primitive_Operation): when checking the validity of
the first formal of a primitive operation, go to the base type to
handle properly an access to a derived type with constrained
discriminants.
* sem_type.adb (Disambiguate): an immediately visible operator hides a
user-defined function that is only use-visible.
* inline.adb: init procs are inlineable.
From-SVN: r177253
Robert Dewar [Wed, 3 Aug 2011 09:47:07 +0000 (09:47 +0000)]
sem_aggr.adb, [...]: Minor reformatting
2011-08-03 Robert Dewar <dewar@adacore.com>
* sem_aggr.adb, sem_ch3.adb, lib.ads, gnatcmd.adb, prj-proc.adb,
make.adb, lib-writ.adb, prj-part.adb, prj-part.ads, prj-ext.adb,
fname-uf.adb, prj-ext.ads, prj.adb, prj.ads, sem_attr.adb, alfa.adb,
prj-makr.adb, errout.adb, makeutl.adb, makeutl.ads, restrict.ads,
sem_ch6.adb, g-pehage.adb, clean.adb, put_alfa.adb, lib-xref-alfa.adb,
prj-nmsc.adb, prj-nmsc.ads, sem_ch8.adb, prj-pars.ads, exp_aggr.adb,
prj-attr.ads, sem_ch13.adb, get_alfa.adb, prj-env.adb, prj-env.ads,
alfa_test.adb, prj-tree.adb, prj-tree.ads, einfo.ads: Minor reformatting
2011-08-03 Robert Dewar <dewar@adacore.com>
* repinfo.adb (List_Mechanism): Add handling of
Convention_Ada_Pass_By_XXX.
* sem_mech.adb (Set_Mechanism): Ditto.
* sem_prag.adb (Process_Convention): Add entries for
Convention_Ada_Pass_By_XXX.
* snames.adb-tmpl, snames.ads-tmpl: Ditto.
From-SVN: r177252
Arnaud Charlet [Wed, 3 Aug 2011 09:43:16 +0000 (11:43 +0200)]
[multiple changes]
2011-08-03 Pascal Obry <obry@adacore.com>
* makeutl.adb: Minor reformatting.
2011-08-03 Vincent Celier <celier@adacore.com>
* fname-uf.adb
(Get_File_Name) If the file cannot be found when there are no config
pragmas file and May_Fail is True, return No_File instead of the file
name, to be consistent.
2011-08-03 Pascal Obry <obry@adacore.com>
* adaint.c (__gnat_is_executable_file_attr): Avoid access
to null pointer.
2011-08-03 Javier Miranda <miranda@adacore.com>
* sem_ch13.adb
(Same_Representation): In VM targets the representation of arrays with
aliased components differs from arrays with non-aliased components.
From-SVN: r177251
Emmanuel Briot [Wed, 3 Aug 2011 09:38:56 +0000 (09:38 +0000)]
prj-proc.adb, [...] (Get_Attribute_Index): do not systematically lower case attribute indexes that contain no "." Fix...
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-proc.adb, prj-attr.adb, prj-attr.ads (Get_Attribute_Index): do
not systematically lower case attribute indexes that contain no "."
Fix definition of several Naming attributes, which take
a unit name as index and therefore should be case insensitive.
Minor refactoring (reduce length of variable names).
2011-08-03 Emmanuel Briot <briot@adacore.com>
* makeutl.adb, makeutl.ads (Get_Switches): new subprogram.
From-SVN: r177250
Emmanuel Briot [Wed, 3 Aug 2011 09:36:24 +0000 (09:36 +0000)]
gnatcmd.adb, [...] (Prj.Tree.Environment): new type.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* gnatcmd.adb, prj-proc.adb, prj-proc.ads, make.adb, prj-part.adb,
prj-part.ads, switch-m.adb, switch-m.ads, prj-makr.adb, clean.adb,
prj-pars.adb, prj-pars.ads, prj-conf.adb, prj-conf.ads, prj-tree.adb,
prj-tree.ads (Prj.Tree.Environment): new type.
From-SVN: r177248
Emmanuel Briot [Wed, 3 Aug 2011 09:30:45 +0000 (09:30 +0000)]
prj-proc.adb, [...] (Environment): new type.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-proc.adb, prj.ads, makeutl.adb, makeutl.ads, prj-conf.adb,
prj-tree.adb, prj-tree.ads (Environment): new type.
From-SVN: r177247
Arnaud Charlet [Wed, 3 Aug 2011 09:27:50 +0000 (11:27 +0200)]
[multiple changes]
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-tree.ads: Remove unused variable.
2011-08-03 Yannick Moy <moy@adacore.com>
* restrict.adb (Restriction_Msg): set "SPARK" all in upper case for
issuing messages related to SPARK restriction violation.
From-SVN: r177246
Arnaud Charlet [Wed, 3 Aug 2011 09:26:00 +0000 (11:26 +0200)]
[multiple changes]
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb: guard against error nodes in return statements.
2011-08-03 Arnaud Charlet <charlet@adacore.com>
* errout.adb (Error_Msg_Internal): the main unit has not been read yet,
a warning can only appear on a configuration file, so emit warning
without further checks.
2011-08-03 Arnaud Charlet <charlet@adacore.com>
* s-interr.ads: add overriding keyword.
2011-08-03 Geert Bosch <bosch@adacore.com>
* exp_attr.adb: Fix minor typo.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* par-ch4.adb: improve error recovery.
From-SVN: r177245
Arnaud Charlet [Wed, 3 Aug 2011 09:21:55 +0000 (11:21 +0200)]
[multiple changes]
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-part.adb, prj-part.ads, prj-makr.adb, prj-pars.adb, prj-conf.adb,
prj-env.adb (Prj.Part.Parse): change parameter Always_Errout_Finalize
to Errout_Handling.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-dect.adb (Parse_Attribute_Declaration): make sure we can use
"external" as an attribute name in aggregate projects.
2011-08-03 Jose Ruiz <ruiz@adacore.com>
* s-taprop-vxworks.adb: (Create_Task, Initialize): Ada 2012 pragma CPU
uses CPU numbers starting 1, while VxWorks uses CPU numbers starting
from 0, so we need to adjust.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-proc.adb, prj-ext.adb, prj-ext.ads, makeutl.adb, prj-tree.adb,
prj-tree.ads, gnatcmd.adb, clean.adb (External_References): new type.
From-SVN: r177244
Arnaud Charlet [Wed, 3 Aug 2011 09:17:28 +0000 (11:17 +0200)]
[multiple changes]
2011-08-03 Yannick Moy <moy@adacore.com>
* sem_ch6.adb (New_Overloaded_Entity): only issue error for SPARK
restriction on overloaded entity if the entity is not an operator.
2011-08-03 Yannick Moy <moy@adacore.com>
* sem_ch7.adb, sem_res.adb, sem_attr.adb, restrict.adb,
restrict.ads: Rename remaining Check_Formal_Restriction* into
Check_SPARK_Restriction*.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-proc.adb, prj-part.adb, prj.adb, prj.ads, prj-nmsc.adb,
prj-nmsc.ads, prj-err.adb (Project_Data): now discriminated on its
qualifier.
(Project_Empty): removed
(Empty_Project): new parameter Qualifier
This is used to have fields specific to aggregate projects, cleaner
New field to store the list of aggregated projects.
(Check_Aggregate_Project): removed
(Process_Aggregated_Projects, Free): new subprograms.
From-SVN: r177243
Arnaud Charlet [Wed, 3 Aug 2011 08:32:57 +0000 (10:32 +0200)]
[multiple changes]
2011-08-03 Olivier Hainque <hainque@adacore.com>
* tracebak.c (STOP_FRAME ppc AIX): Stop at null return address as well.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_ch8.adb (Analyze_Object_Renaming): if the renamed object is an
explicit dereference of an unconstrained type, create a constrained
subtype for it, as is done for function calls that return an
unconstrained type.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* g-pehage.adb (Finalize): Avoid possible double-free.
2011-08-03 Steve Baird <baird@adacore.com>
* exp_attr.adb (Expand_N_Attribute_Reference): Don't expand
Elab_Spec/Body attrs in CodePeer_Mode.
2011-08-03 Javier Miranda <miranda@adacore.com>
* exp_aggr.adb (Flatten): Convert to positional form aggregates whose
low bound is not known at compile time but they have no others choice.
Done because in this case the bounds can be obtained directly from the
aggregate.
2011-08-03 Ed Falis <falis@adacore.com>
* s-tasinf-vxworks.ads: Update comment to reflect 0 indexing of CPUs
on VxWorks SMP. Remove unusable constant ANY_CPU.
From-SVN: r177242
Emmanuel Briot [Wed, 3 Aug 2011 08:28:47 +0000 (08:28 +0000)]
gnatcmd.adb, [...] (Prj.Env.Initialize_Default_Project_Path, [...]): new subprograms
2011-08-03 Emmanuel Briot <briot@adacore.com>
* gnatcmd.adb, make.adb, prj-part.adb, prj-part.ads, prj-makr.adb,
clean.adb, prj-nmsc.adb, prj-pars.adb, prj-conf.adb, prj-env.adb,
prj-env.ads (Prj.Env.Initialize_Default_Project_Path,
Prj.Env.Initialize_Empty): new subprograms
(Get_Env, Find_Project): remove parameter Target_Name.
From-SVN: r177241
Arnaud Charlet [Wed, 3 Aug 2011 08:26:17 +0000 (10:26 +0200)]
[multiple changes]
2011-08-03 Gary Dismukes <dismukes@adacore.com>
* sem_ch3.adb (Build_Derived_Record_Type): Test the Derive_Subps formal
as a condition for the delayed call to Derived_Subprograms done for the
case of the rewriting of a derived type that constrains the
discriminants of its parent type.
Avoids redundant subprogram derivations for private subtype derivations.
2011-08-03 Javier Miranda <miranda@adacore.com>
* exp_aggr.adb (Init_Hidden_Discriminants): New subprogram of
Build_Record_Aggr_Code.
(Build_Record_Aggr_Code): Add missing support to initialize hidden
discriminants in extension aggregates.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-pp.adb (Print): also output project qualifiers, since in
particular "aggregate" is mandatory in an aggregate project.
2011-08-03 Emmanuel Briot <briot@adacore.com>
* prj-part.adb, prj.adb, prj.ads, prj-nmsc.adb, prj-env.adb:
(Debug_Output): new function.
From-SVN: r177240
Arnaud Charlet [Wed, 3 Aug 2011 08:22:52 +0000 (10:22 +0200)]
[multiple changes]
2011-08-03 Eric Botcazou <ebotcazou@adacore.com>
* gnat_ugn.texi: Document -Wstack-usage.
* gcc-interface/misc.c (enumerate_modes): Add guard for ghost FP modes.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* sem_prag.adb Issue an error (not a warning) when a C++ type does not
have keyword LIMITED.
2011-08-03 Yannick Moy <moy@adacore.com>
* alfa.adb, alfa.ads, alfa_test.adb: New files.
* ali.adb (Known_ALI_Lines): add 'C' lines (SCO) and 'F' lines (ALFA)
(Scan_ALI): do not issue a fatal error if parsing known lines after Xref
section (does not happen in compiler, only if code directly calls
Scan_ALI).
* get_alfa.adb, get_alfa.ads: New files.
* lib-writ.adb, lib-writ.ads (Write_ALI): output ALFA information if
needed.
* lib-xref-alfa.adb: New file.
* lib-xref.adb, lib-xref.ads
(Xref_Entry): redefine information needed in cross-references for ALFA.
Push ALFA treatments in separated local package.
(Enclosing_Subpragram_Or_Package): treat specially subprogram
identifiers. Return entity of package body instead of spec. Return
Empty for a scope with no location.
(Generate_Reference): adapt to new components for ALFA information.
Remove the need for D references on definitions.
(Is_Local_Reference): moved to ALFA local package
(Output_References): extract subfunction as Extract_Source_Name
(Output_Local_References): remove procedure, replaced by filtering of
cross-references in package ALFA and printing in Put_ALFA.
(Write_Entity_Name): remove procedure
* lib.adb, lib.ads (Extract_Source_Name): extract here function to
print exact name of entity as it appears in source file
(Unit_Ref_Table): make type public for use in Lib.Xref.ALFA
* put_alfa.adb, put_alfa.ads: New files.
* xref_lib.adb (Search_Xref): protect read of cross-references against
reading other sections of the ALI file, in gnatxref
(Search): protect read of cross-references against reading other
sections of the ALI file, in gnatfind.
* gcc-interface/Make-lang.in: Update dependencies.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* sem_ch3.adb: Minor reformatting.
2011-08-03 Jose Ruiz <ruiz@adacore.com>
* s-inmaop-vxworks.adb (Setup_Interrupt_Mask): Do nothing instead of
raising an exception.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_res.adb (Set_String_Literal_Subtype): if index type is an integer
type, always use 1 as the lower bound or string, even if lower bound of
context is not static, to handle properly null strings in a non-static
context.
2011-08-03 Bob Duff <duff@adacore.com>
* sem_prag.adb (Resolve_Aggregate): An array aggregate with 'others' is
always legal on the right-hand side of an assignment statement; there
is always an applicable index constraint in this case. Therefore, the
check for Pkind = N_Assignment_Statement is now unconditional -- it
doesn't depend on whether Is_Constrained (Typ).
From-SVN: r177239
Arnaud Charlet [Wed, 3 Aug 2011 08:11:09 +0000 (10:11 +0200)]
[multiple changes]
2011-08-03 Robert Dewar <dewar@adacore.com>
* exp_ch5.adb, sem_prag.adb, exp_ch4.adb: Minor reformatting.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Build_Itype_Reference): do not create an itype reference
for an itype created within a generic unit.
2011-08-03 Eric Botcazou <ebotcazou@adacore.com>
* gnat_ugn.texi (Switches for gcc): Make it clearer that -fno-inline
suppresses all inlining.
From-SVN: r177238
Arnaud Charlet [Wed, 3 Aug 2011 08:08:31 +0000 (10:08 +0200)]
[multiple changes]
2011-08-03 Robert Dewar <dewar@adacore.com>
* sem_ch3.adb, sem_res.adb, exp_ch13.adb, exp_disp.adb,
exp_aggr.adb: Minor reformatting.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* exp_ch5.adb (Expand_N_Assignment_Statement): Do not force inlining of
tagged assignment when discriminant checks are suppressed. This is
useless and extremely costly in terms of static stack usage.
2011-08-03 Bob Duff <duff@adacore.com>
* sem_prag.adb (Get_Base_Subprogram): Do not follow Alias for instances
of generics, because this leads to the wrong entity in the wrong scope,
causing (e.g.) pragma Export_Procedure to get an error if the entity is
an instance.
(Process_Interface_Name): Follow Alias for instances of generics, to
correct for the above change.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* exp_ch4.adb (Expand_N_Selected_Component): If the discriminant value
is an integer literal it is always safe to replace the reference. In
addition, if the reference appears in the generated code for an object
declaration it is necessary to copy because otherwise the reference
might be to the uninitilized value of the discriminant of the object
itself.
2011-08-03 Pascal Obry <obry@adacore.com>
* adaint.c (__gnat_is_executable_file_attr): Fix Win32 circuitry when no
ACL used, in this case we want to check for ending .exe, not .exe
anywhere in the path.
2011-08-03 Sergey Rybin <rybin@adacore.com>
* tree_io.ads (ASIS_Version_Number): Update because of the changes in
the tree structure (semantic decoration of references to record
discriminants).
From-SVN: r177237
Arnaud Charlet [Wed, 3 Aug 2011 08:02:56 +0000 (10:02 +0200)]
[multiple changes]
2011-08-03 Gary Dismukes <dismukes@adacore.com>
* sem_aggr.adb (Analyze_Array_Aggregate): When checking the discrete
choices of a named array aggregate, bail out when any choices are
marked as Errors_Posted.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* exp_ch13.adb (Expand_N_Freeze_Entity): cleanup determination of scope
in which entity is frozen, to handle properly loop variables in
iterators.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_res.adb (Set_String_Literal_Subtype): if the lower bound of the
subtype is not static, compute the upper bound using attributes, to
handle properly index types that are not integer types.
2011-08-03 Bob Duff <duff@adacore.com>
* gnat_rm.texi, gnat_ugn.texi: Fix some dangling URLs.
Update copyright notice.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Build_Discriminant_Constraints): Only use
Original_Discriminant if within an instance.
* sem_ch4.adb (Analyze_Selected_Component): Ditto.
2011-08-03 Thomas Quinot <quinot@adacore.com>
* einfo.ads: Minor reformatting.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* exp_disp.adb (Check_Premature_Freezing): diagnose the presence of a
composite type with an unfrozen subcomponent, in the profile of a
primitive operation.
From-SVN: r177236
Revital Eres [Wed, 3 Aug 2011 07:59:23 +0000 (07:59 +0000)]
Optimize stage count
From-SVN: r177235
Arnaud Charlet [Wed, 3 Aug 2011 07:52:00 +0000 (09:52 +0200)]
[multiple changes]
2011-08-03 Robert Dewar <dewar@adacore.com>
* sem_util.ads, exp_aggr.adb, exp_ch3.adb: Minor reformatting.
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Analyze_Object_Declaration): if a constant object of an
unconstrained type with discriminants is initialized with an aggregate,
use the constrained subtype of the aggregate as the type of the object,
because it is immutable, and this allows the back-end to generate no
code for the object declaration.
From-SVN: r177234
Arnaud Charlet [Wed, 3 Aug 2011 07:45:36 +0000 (09:45 +0200)]
[multiple changes]
2011-08-03 Robert Dewar <dewar@adacore.com>
* exp_ch4.adb (Optimize_Length_Check): Fix bad handling of case where
comparison operand is variable, and turns out to be zero or negative.
2011-08-03 Javier Miranda <miranda@adacore.com>
* exp_intr.adb
(Expand_Dispatching_Constructor_Call): Disable expansion of
code required for native targets. Done to avoid generating
references to unavailable runtime entities in VM targets.
* exp_ch3.adb
(Expand_N_Object_Declaration): Add missing support to handle
the explicit initialization of class-wide interface objects.
Fix documentation.
2011-08-03 Matthew Heaney <heaney@adacore.com>
* a-cobove.adb (Merge): Move source onto target, instead of using Assign
2011-08-03 Matthew Heaney <heaney@adacore.com>
* a-cbdlli.adb (Splice): move source items from first to last
2011-08-03 Yannick Moy <moy@adacore.com>
* sem_util.ads: comment added.
2011-08-03 Javier Miranda <miranda@adacore.com>
* exp_aggr.adb
(Expand_Record_Aggregate): In VM targets disable the expansion into
assignments of aggregates whose type is not known at compile time.
From-SVN: r177233
Ed Schonberg [Wed, 3 Aug 2011 07:37:15 +0000 (07:37 +0000)]
exp_ch9.adb (Build_Renamed_Formal_Declaration): common procedure for protected entries and task entries...
2011-08-03 Ed Schonberg <schonberg@adacore.com>
* exp_ch9.adb (Build_Renamed_Formal_Declaration): common procedure for
protected entries and task entries, to build the proper renaming
declaration for entry formals, used in debugging.
* exp_ch2.adb (Expand_Entry_Parameter): handle task and entry
parameters in the same way.
From-SVN: r177232
H.J. Lu [Wed, 3 Aug 2011 04:59:41 +0000 (04:59 +0000)]
Update x32 __NR_rt_sigreturn system call number.
2011-08-02 H.J. Lu <hongjiu.lu@intel.com>
* config/i386/linux-unwind.h (RT_SIGRETURN_SYSCALL): New.
(x86_64_fallback_frame_state): Use RT_SIGRETURN_SYSCALL and
long long to check rt_sigreturn syscall.
From-SVN: r177229
Roberto Lublinerman [Wed, 3 Aug 2011 00:37:26 +0000 (00:37 +0000)]
gccgo: Added code to dump the AST tree.
gccgo: Added code to dump the AST tree. The AST dump is
activated with -fgo-dump-ast.
Initial version, it only dumps (most) constructs that
are expected after the lowering transformation.
* Make-lang.in (GO_OBJS): Add go/ast-dump.o.
(go/ast-dump.o): New target.
(go/expressions.o): Depend on go/gofrontend/ast-dump.h.
(go/statements.o): Likewise.
From-SVN: r177225
GCC Administrator [Wed, 3 Aug 2011 00:18:01 +0000 (00:18 +0000)]
Daily bump.
From-SVN: r177224
Richard Henderson [Tue, 2 Aug 2011 22:20:17 +0000 (15:20 -0700)]
Add another PR tag.
From-SVN: r177219
Richard Henderson [Tue, 2 Aug 2011 22:18:35 +0000 (15:18 -0700)]
re PR debug/49864 (ICE: in maybe_record_trace_start, at dwarf2cfi.c:2439)
PR target/49864
* reg-notes.def (REG_ARGS_SIZE): New.
* calls.c (emit_call_1): Emit REG_ARGS_SIZE for call_pop.
(expand_call): Add REG_ARGS_SIZE to emit_stack_restore.
* cfgcleanup.c (old_insns_match_p): Don't allow cross-jumping to
different stack levels.
* combine-stack-adj.c (adjust_frame_related_expr): Remove.
(maybe_move_args_size_note): New.
(combine_stack_adjustments_for_block): Use it.
* combine.c (distribute_notes): Place REG_ARGS_SIZE.
* dwarf2cfi.c (dw_cfi_row_struct): Remove args_size member.
(dw_trace_info): Add beg_true_args_size, end_true_args_size,
beg_delay_args_size, end_delay_args_size, eh_head, args_size_undefined.
(cur_cfa): New.
(queued_args_size): Remove.
(add_cfi_args_size): Assert size is non-negative.
(stack_adjust_offset, dwarf2out_args_size): Remove.
(dwarf2out_stack_adjust, dwarf2out_notice_stack_adjust): Remove.
(notice_args_size, notice_eh_throw): New.
(dwarf2out_frame_debug_def_cfa): Use cur_cfa.
(dwarf2out_frame_debug_adjust_cfa): Likewise.
(dwarf2out_frame_debug_cfa_offset): Likewise.
(dwarf2out_frame_debug_expr): Likewise. Don't stack_adjust_offset.
(dwarf2out_frame_debug): Don't handle non-frame-related-p insns.
(change_cfi_row): Don't emit args_size.
(maybe_record_trace_start_abnormal): Split out from ...
(maybe_record_trace_start): Here. Set args_size_undefined.
(create_trace_edges): Update to match.
(scan_trace): Handle REG_ARGS_SIZE.
(connect_traces): Connect args_size between EH insns.
* emit-rtl.c (try_split): Handle REG_ARGS_SIZE.
* explow.c (suppress_reg_args_size): New.
(adjust_stack_1): Split out from ...
(adjust_stack): ... here.
(anti_adjust_stack): Use it.
(allocate_dynamic_stack_space): Suppress REG_ARGS_SIZE.
* expr.c (mem_autoinc_base): New.
(fixup_args_size_notes): New.
(emit_single_push_insn_1): Rename from emit_single_push_insn.
(emit_single_push_insn): New. Generate REG_ARGS_SIZE.
* recog.c (peep2_attempt): Handle REG_ARGS_SIZE.
* reload1.c (reload_as_needed): Likewise.
* rtl.h (fixup_args_size_notes): Declare.
From-SVN: r177218
Richard Henderson [Tue, 2 Aug 2011 21:23:58 +0000 (14:23 -0700)]
Fix my last changelog entry.
From-SVN: r177217
Jason Merrill [Tue, 2 Aug 2011 21:09:26 +0000 (17:09 -0400)]
re PR c++/43886 ([C++0x] name lookup failure on un-used local variable in lambda compound-statement)
PR c++/43886
* parser.c (cp_parser_lambda_body): Clear local_variables_forbidden_p.
From-SVN: r177216
Jason Merrill [Tue, 2 Aug 2011 21:09:17 +0000 (17:09 -0400)]
re PR c++/49577 ([C++0x] narrowing conversion not rejected)
PR c++/49577
* typeck2.c (check_narrowing): Check unsigned mismatch.
* semantics.c (finish_compound_literal): check_narrowing.
From-SVN: r177215
Jason Merrill [Tue, 2 Aug 2011 21:09:08 +0000 (17:09 -0400)]
re PR c++/49593 ([C++0x] cannot use T{t}... as pack expansion)
PR c++/49593
* pt.c (find_parameter_packs_r): Handle CONSTRUCTOR.
From-SVN: r177214
Jason Merrill [Tue, 2 Aug 2011 21:08:57 +0000 (17:08 -0400)]
re PR c++/49803 ([C++0x] erroneous variant-member initialization in a union containing an anonymous struct)
PR c++/49803
* init.c (sort_mem_initializers): Initialize uses_unions_p here.
(build_field_list): Not here.
From-SVN: r177213
Paolo Carlini [Tue, 2 Aug 2011 20:26:57 +0000 (20:26 +0000)]
re PR bootstrap/49914 (call to abs(long long) in gcc/fold-const.c)
2011-08-02 Paolo Carlini <paolo.carlini@oracle.com>
PR bootstrap/49914
* fold-const.c (fold_plusminus_mult_expr): Use abs_hwi instead
of abs.
* tree-ssa-math-opts.c (gimple_expand_builtin_pow): Likewise.
* tree-ssa-loop-prefetch.c (prune_ref_by_group_reuse): Likewise.
From-SVN: r177212
Daniel Kraft [Tue, 2 Aug 2011 20:10:13 +0000 (22:10 +0200)]
re PR fortran/49885 (Segmentation fault when writing to an automatic character array)
2011-08-02 Daniel Kraft <d@domob.eu>
PR fortran/49885
* trans-array.c (gfc_trans_auto_array_allocation): Change
gfc_start_block to gfc_init_block to avoid spurious extra-scope.
2011-08-02 Daniel Kraft <d@domob.eu>
PR fortran/49885
* gfortran.dg/auto_char_dummy_array_3.f90: New test.
From-SVN: r177211
Richard Henderson [Tue, 2 Aug 2011 19:56:29 +0000 (12:56 -0700)]
h8300: Generate correct unwind info around swap_into/out_of_er6.
Minimal bug fix is to unset RTX_FRAME_RELATED_P on the PUSH
insn generated in h8300_swap_into_er6. But with a tiny bit
of effort we can generate real unwind info around the sequence.
From-SVN: r177209
Kai Tietz [Tue, 2 Aug 2011 19:31:30 +0000 (21:31 +0200)]
re PR middle-end/49947 (FAIL: gcc.dg/tree-ssa/forwprop-15.c)
2011-08-02 Kai Tietz <ktietz@redhat.com>
PR middle-end/49947
* gcc.dg/tree-ssa/forwprop-15.c
From-SVN: r177205
H.J. Lu [Tue, 2 Aug 2011 19:23:10 +0000 (19:23 +0000)]
Fix a typo in host_detect_local_cpu.
2011-08-02 H.J. Lu <hongjiu.lu@intel.com>
* config/i386/driver-i386.c (host_detect_local_cpu): Fix a typo.
From-SVN: r177204
Richard Henderson [Tue, 2 Aug 2011 18:48:00 +0000 (11:48 -0700)]
re PR target/49878 ([h8300] ICE in based_loc_descr, at dwarf2out.c:10478)
PR target/49878
h8300: Don't allow eliminable registers to be pushed.
From-SVN: r177200
Jason Merrill [Tue, 2 Aug 2011 18:08:58 +0000 (14:08 -0400)]
re PR c++/49834 ([C++0x] type deduction in for-each loop in template function)
PR c++/49834
* parser.c (build_range_temp): Split out from...
(cp_convert_range_for): ...here.
(do_range_for_auto_deduction): New.
(cp_parser_range_for): Use it.
From-SVN: r177199
Tobias Burnus [Tue, 2 Aug 2011 18:07:52 +0000 (20:07 +0200)]
trans-array.c (gfc_array_allocate): Pass token to gfc_allocate_allocatable for -fcoarray=lib.
2011-08-02 Tobias Burnus <burnus@net-b.de>
* trans-array.c (gfc_array_allocate): Pass token to
gfc_allocate_allocatable for -fcoarray=lib.
* trans-stmt.c (gfc_trans_allocate): Update
gfc_allocate_allocatable call.
* trans.h (gfc_allocate_allocatable): Update prototype.
(gfc_allocate_using_lib): Remove.
* trans.c (gfc_allocate_using_lib): Make static, handle
token.
(gfc_allocate_allocatable): Ditto.
2011-08-02 Tobias Burnus <burnus@net-b.de>
* gfortran.dg/coarray_lib_token_3.f90: New.
From-SVN: r177198
Richard Henderson [Tue, 2 Aug 2011 17:38:16 +0000 (10:38 -0700)]
re PR target/49881 ([AVR] Inefficient stack manipulation around calls)
PR target/49881
* config/avr/avr.md (push<MPUSH>1): Don't constrain the operand.
From-SVN: r177196
Jakub Jelinek [Tue, 2 Aug 2011 16:13:29 +0000 (18:13 +0200)]
backport: re PR fortran/46752 (OpenMP - Seg fault for unallocated allocatable array in firstprivate clause)
Merge from gomp-3_1-branch branch:
2011-08-02 Jakub Jelinek <jakub@redhat.com>
gcc/
* c-parser.c (enum c_parser_prec): New enum, moved from within
c_parser_binary_expression.
(c_parser_binary_expression): Add PREC argument. Stop parsing
if operator has lower or equal precedence than PREC.
(c_parser_conditional_expression, c_parser_omp_for_loop): Adjust
callers.
(c_parser_omp_atomic): Handle parsing OpenMP 3.1 atomics.
Adjust c_finish_omp_atomic caller.
(c_parser_omp_taskyield): New function.
(c_parser_pragma): Handle PRAGMA_OMP_TASKYIELD.
(c_parser_omp_clause_name): Handle final and mergeable clauses.
(c_parser_omp_clause_final, c_parser_omp_clause_mergeable): New
functions.
(c_parser_omp_all_clauses): Handle PRAGMA_OMP_CLAUSE_FINAL
and PRAGMA_OMP_CLAUSE_MERGEABLE.
(OMP_TASK_CLAUSE_MASK): Allow final and mergeable clauses.
(c_parser_omp_clause_reduction): Handle min and max.
* c-typeck.c (c_finish_omp_clauses): Don't complain about
const qualified predetermined vars in firstprivate clause.
andle OMP_CLAUSE_FINAL and OMP_CLAUSE_MERGEABLE.
Handle MIN_EXPR and MAX_EXPR.
* tree-pretty-print.c (dump_omp_clause): Handle OMP_CLAUSE_FINAL
and OMP_CLAUSE_MERGEABLE.
(dump_generic_node): Handle OMP_ATOMIC_READ, OMP_ATOMIC_CAPTURE_OLD
and OMP_ATOMIC_CAPTURE_NEW.
* tree.c (omp_clause_num_ops): Add OMP_CLAUSE_FINAL and
OMP_CLAUSE_MERGEABLE.
(omp_clause_code_name): Likewise.
(walk_tree_1): Handle OMP_CLAUSE_FINAL and OMP_CLAUSE_MERGEABLE.
* tree.h (enum omp_clause_code): Add OMP_CLAUSE_FINAL
and OMP_CLAUSE_MERGEABLE.
(OMP_CLAUSE_FINAL_EXPR): Define.
* omp-low.c (scan_sharing_clauses): Handle OMP_CLAUSE_FINAL and
OMP_CLAUSE_MERGEABLE.
(expand_task_call): Likewise.
(expand_omp_atomic_load, expand_omp_atomic_store): New functions.
(expand_omp_atomic_fetch_op): Handle cases where old or new
value is needed afterwards.
(expand_omp_atomic): Call expand_omp_atomic_load resp.
expand_omp_atomic_store.
* gimplify.c (gimplify_omp_atomic, gimplify_expr): Handle
OMP_ATOMIC_READ, OMP_ATOMIC_CAPTURE_OLD and OMP_ATOMIC_CAPTURE_NEW.
(gimplify_scan_omp_clauses, gimplify_adjust_omp_clauses): Handle
OMP_CLAUSE_FINAL and OMP_CLAUSE_MERGEABLE.
* tree-nested.c (convert_nonlocal_omp_clauses,
convert_local_omp_clauses): Likewise.
* tree.def (OMP_ATOMIC_READ, OMP_ATOMIC_CAPTURE_OLD,
OMP_ATOMIC_CAPTURE_NEW): New.
* gimple.h (GF_OMP_ATOMIC_NEED_VALUE): New.
(gimple_omp_atomic_need_value_p, gimple_omp_atomic_set_need_value):
New inlines.
* omp-builtins.def (BUILT_IN_GOMP_TASKYIELD): New builtin.
* doc/generic.texi: Mention OMP_CLAUSE_COLLAPSE,
OMP_CLAUSE_UNTIED, OMP_CLAUSE_FINAL and OMP_CLAUSE_MERGEABLE.
gcc/c-family/
* c-common.h (c_finish_omp_atomic): Adjust prototype.
(c_finish_omp_taskyield): New prototype.
* c-omp.c (c_finish_omp_atomic): Add OPCODE, V, LHS1 and RHS1
arguments. Handle OMP_ATOMIC_READ, OMP_ATOMIC_CAPTURE_OLD and
OMP_ATOMIC_CAPTURE_NEW in addition to OMP_ATOMIC. If LHS1
or RHS1 have side-effects, evaluate those too in the right spot,
if it is a decl and LHS is also a decl, error out if they
aren't the same.
(c_finish_omp_taskyield): New function.
* c-cppbuiltin.c (c_cpp_builtins): Change _OPENMP to 201107.
* c-pragma.c (omp_pragmas): Add taskyield.
* c-pragma.h (enum pragma_kind): Add PRAGMA_OMP_TASKYIELD.
(enum pragma_omp_clause): Add PRAGMA_OMP_CLAUSE_FINAL and
PRAGMA_OMP_CLAUSE_MERGEABLE.
gcc/cp/
* cp-tree.h (finish_omp_atomic): Adjust prototype.
(cxx_omp_const_qual_no_mutable): New prototype.
(finish_omp_taskyield): New prototype.
* parser.c (cp_parser_omp_atomic): (cp_parser_omp_atomic): Handle
parsing OpenMP 3.1 atomics. Adjust finish_omp_atomic caller.
(cp_parser_omp_clause_name): Handle final and mergeable clauses.
(cp_parser_omp_clause_final, cp_parser_omp_clause_mergeable): New
functions.
(cp_parser_omp_all_clauses): Handle PRAGMA_OMP_CLAUSE_FINAL
and PRAGMA_OMP_CLAUSE_MERGEABLE.
(OMP_TASK_CLAUSE_MASK): Allow final and mergeable clauses.
(cp_parser_omp_taskyield): New function.
(cp_parser_pragma): Handle PRAGMA_OMP_TASKYIELD.
(cp_parser_omp_clause_reduction): Handle min and max.
* pt.c (tsubst_expr) <case OMP_ATOMIC>: Handle OpenMP 3.1 atomics.
(tsubst_omp_clauses): Handle OMP_CLAUSE_FINAL and
OMP_CLAUSE_MERGEABLE.
* semantics.c (finish_omp_atomic): Add OPCODE, V, LHS1 and RHS1
arguments. Handle OpenMP 3.1 atomics. Adjust c_finish_omp_atomic
caller.
(finish_omp_clauses): Don't complain about const qualified
predetermined vars and static data members in firstprivate clause.
Handle OMP_CLAUSE_FINAL and OMP_CLAUSE_MERGEABLE. Handle MIN_EXPR
and MAX_EXPR.
(finish_omp_taskyield): New function.
* cp-gimplify.c (cxx_omp_const_qual_no_mutable): New function.
(cxx_omp_predetermined_sharing): Use it.
gcc/fortran/
PR fortran/46752
* cpp.c (cpp_define_builtins): Change _OPENMP to 201107.
* openmp.c (gfc_free_omp_clauses): Free also final_expr.
(OMP_CLAUSE_FINAL, OMP_CLAUSE_MERGEABLE): Define.
(gfc_match_omp_clauses): Handle parsing final and mergeable
clauses.
(OMP_TASK_CLAUSES): Allow final and mergeable clauses.
(gfc_match_omp_taskyield): New function.
(resolve_omp_clauses): Resolve final clause. Allow POINTERs and
Cray pointers in clauses other than REDUCTION.
(gfc_match_omp_atomic): Match optional
read/write/update/capture keywords after !$omp atomic.
(resolve_omp_atomic): Handle all OpenMP 3.1 atomic forms.
* dump-parse-tree.c (show_omp_node): Handle EXEC_OMP_TASKYIELD,
print final and mergeable clauses.
(show_code_node): Handle EXEC_OMP_TASKYIELD.
* trans-openmp.c (gfc_trans_omp_clauses): Handle final and
mergeable clauses.
(gfc_trans_omp_taskyield): New function.
(gfc_trans_omp_directive): Handle EXEC_OMP_TASKYIELD.
(gfc_trans_omp_atomic): Handle all OpenMP 3.1 atomic forms.
(gfc_omp_clause_copy_ctor): Handle non-allocated allocatable.
(gfc_omp_predetermined_sharing): Adjust comment.
* gfortran.h (gfc_statement): Add ST_OMP_TASKYIELD and
ST_OMP_END_ATOMIC.
(gfc_omp_clauses): Add final_expr and mergeable fields.
(gfc_exec_op): Add EXEC_OMP_TASKYIELD.
(gfc_omp_atomic_op): New enum typedef.
(struct gfc_code): Add ext.omp_atomic.
* trans.c (trans_code): Handle EXEC_OMP_TASKYIELD.
* frontend-passes.c (gfc_code_walker): Also walk final_expr.
* resolve.c (gfc_resolve_blocks, resolve_code): Handle
EXEC_OMP_TASKYIELD.
* st.c (gfc_free_statement): Likewise.
* match.h (gfc_match_omp_taskyield): New prototype.
* parse.c (decode_omp_directive): Handle taskyield directive.
Handle !$omp end atomic.
(case_executable): Add ST_OMP_TASKYIELD case.
(gfc_ascii_statement): Handle ST_OMP_TASKYIELD.
(parse_omp_atomic): Return gfc_statement instead of void.
For !$omp atomic capture parse two assignments instead of
just one and require !$omp end atomic afterwards, for
other !$omp atomic forms just allow !$omp end atomic at the
end.
(parse_omp_structured_block, parse_executable): Adjust
parse_omp_atomic callers.
2011-08-02 Tobias Burnus <burnus@net-b.de>
* intrinsic.c (OMP_LIB): Updated openmp_version's
value to 201107.
* gfortran.texi (OpenMP): Update ref to OpenMP 3.1.
* intrinsic.texi (OpenMP Modules): Update ref to OpenMP 3.1;
remove deleted omp_integer_kind and omp_logical_kind constants.
gcc/testsuite/
PR fortran/46752
* gcc.dg/gomp/atomic-5.c: Adjust expected diagnostics.
* gcc.dg/gomp/atomic-15.c: New test.
* g++.dg/gomp/atomic-5.C: Adjust expected diagnostics.
* g++.dg/gomp/atomic-15.C: New test.
* g++.dg/gomp/private-1.C: New test.
* g++.dg/gomp/sharing-2.C: New test.
* gfortran.dg/gomp/crayptr1.f90: Don't expect error
about Cray pointer in FIRSTPRIVATE/LASTPRIVATE.
* gfortran.dg/gomp/omp_atomic2.f90: New test.
libgomp/
PR fortran/42041
PR fortran/46752
* omp.h.in (omp_in_final): New prototype.
* omp_lib.f90.in (omp_in_final): New interface.
(omp_integer_kind, omp_logical_kind): Remove
and replace all its uses in the module with 4.
(openmp_version): Change to 201107.
* omp_lib.h.in (omp_sched_static, omp_sched_dynamic,
omp_sched_guided, omp_sched_auto): Use omp_sched_kind
kind for the parameters.
(omp_in_final): New external.
(openmp_version): Change to 201107.
* task.c (omp_in_final): New function.
(gomp_init_task): Initialize final_task.
(GOMP_task): Remove unused attribute from flags. Handle final
tasks.
(GOMP_taskyield): New function.
(omp_in_final): Return true if if (false) or final (true) task
or descendant of final (true).
* fortran.c (omp_in_final_): New function.
* libgomp.map (OMP_3.1): Export omp_in_final and omp_in_final_.
(GOMP_3.0): Export GOMP_taskyield.
* env.c (gomp_nthreads_var_list, gomp_nthreads_var_list_len): New
variables.
(parse_unsigned_long_list): New function.
(initialize_env): Use it for OMP_NUM_THREADS. Call parse_boolean
with "OMP_PROC_BIND". If OMP_PROC_BIND=true, call gomp_init_affinity
even if parse_affinity returned false.
* config/linux/affinity.c (gomp_init_affinity): Handle
gomp_cpu_affinity_len == 0.
* libgomp_g.h (GOMP_taskyield): New prototype.
* libgomp.h (struct gomp_task): Add final_task field.
(gomp_nthreads_var_list, gomp_nthreads_var_list_len): New externs.
* team.c (gomp_team_start): Override new task's nthreads_var icv
if list form OMP_NUM_THREADS has been used and it has value for
the new nesting level.
* testsuite/libgomp.c/atomic-11.c: New test.
* testsuite/libgomp.c/atomic-12.c: New test.
* testsuite/libgomp.c/atomic-13.c: New test.
* testsuite/libgomp.c/atomic-14.c: New test.
* testsuite/libgomp.c/reduction-6.c: New test.
* testsuite/libgomp.c/task-5.c: New test.
* testsuite/libgomp.c++/atomic-2.C: New test.
* testsuite/libgomp.c++/atomic-3.C: New test.
* testsuite/libgomp.c++/atomic-4.C: New test.
* testsuite/libgomp.c++/atomic-5.C: New test.
* testsuite/libgomp.c++/atomic-6.C: New test.
* testsuite/libgomp.c++/atomic-7.C: New test.
* testsuite/libgomp.c++/atomic-8.C: New test.
* testsuite/libgomp.c++/atomic-9.C: New test.
* testsuite/libgomp.c++/task-8.C: New test.
* testsuite/libgomp.c++/reduction-4.C: New test.
* testsuite/libgomp.fortran/allocatable7.f90: New test.
* testsuite/libgomp.fortran/allocatable8.f90: New test.
* testsuite/libgomp.fortran/crayptr3.f90: New test.
* testsuite/libgomp.fortran/omp_atomic3.f90: New test.
* testsuite/libgomp.fortran/omp_atomic4.f90: New test.
* testsuite/libgomp.fortran/pointer1.f90: New test.
* testsuite/libgomp.fortran/pointer2.f90: New test.
* testsuite/libgomp.fortran/task4.f90: New test.
2011-08-02 Tobias Burnus <burnus@net-b.de>
* libgomp.texi: Update OpenMP spec references to 3.1.
(omp_in_final,OMP_PROC_BIND): New sections.
(OMP_NUM_THREADS): Document that the value can be now a list.
(GOMP_STACKSIZE,GOMP_CPU_AFFINITY): Update @ref.
From-SVN: r177194
Arnaud Charlet [Tue, 2 Aug 2011 15:53:12 +0000 (17:53 +0200)]
New file.
From-SVN: r177193
Arnaud Charlet [Tue, 2 Aug 2011 15:52:59 +0000 (17:52 +0200)]
Minor reformatting.
From-SVN: r177192
Arnaud Charlet [Tue, 2 Aug 2011 15:49:38 +0000 (17:49 +0200)]
Update dependencies
From-SVN: r177191
Arnaud Charlet [Tue, 2 Aug 2011 15:47:39 +0000 (17:47 +0200)]
[multiple changes]
2011-08-02 Robert Dewar <dewar@adacore.com>
* a-direct.adb, sinfo.ads, exp_ch9.adb, scng.adb, sem_util.adb,
sem_util.ads, restrict.ads, par-prag.adb: Minor reformatting and/or
code reorganization.
2011-08-02 Robert Dewar <dewar@adacore.com>
* debug.adb: Debug flag d.P to suppress length comparison optimization
* exp_ch4.adb (Optimize_Length_Comparison): New routine to optimize
comparison of Length by comparing First/Last instead.
2011-08-02 Matthew Heaney <heaney@adacore.com>
* a-cobove.ads: Code clean up.
From-SVN: r177190
Arnaud Charlet [Tue, 2 Aug 2011 15:36:49 +0000 (17:36 +0200)]
[multiple changes]
2011-08-02 Vincent Celier <celier@adacore.com>
* adaint.c (file_names_case_sensitive_cache): New static int.
(__gnat_get_file_names_case_sensitive): Cache the return value in
file_names_case_sensitive_cache at the first invocation, to avoid
multiple calls to getenv.
2011-08-02 Bob Duff <duff@adacore.com>
* sem_ch12.adb (Validate_Derived_Type_Instance): Implement AI05-0218-1.
From-SVN: r177189
Arnaud Charlet [Tue, 2 Aug 2011 15:35:25 +0000 (17:35 +0200)]
[multiple changes]
2011-08-02 Yannick Moy <moy@adacore.com>
* sem_ch3.adb, sem_ch5.adb, sem_ch9.adb, sem_prag.adb, sem.ads,
sem_util.adb, sem_util.ads, sem_res.adb, sem_ch2.adb, sem_ch4.adb,
sem_ch6.adb, sem_ch11.adb: Add semantic flag In_Pre_Post_Expression to
indicate that we are in a precondition or postcondition. This is used in
Mark_Non_ALFA_Subprogram (renaming of Mark_Non_ALFA_Subprogram_Body) to
decide whether to flag the spec or body of the current subprogram as
not in ALFA.
2011-08-02 Fabien Chouteau <chouteau@adacore.com>
* impunit.adb: Add Ada.Execution_Time.Interrupts in the
Ada2012 package
list.
* a-extiin.ads: New file.
2011-08-02 Bob Duff <duff@adacore.com>
* a-direct.adb (Rename): Implement AI05-0231-1. In particular, Rename
now raises Name_Error instead of Use_Error in certain cases. The other
parts of this AI were already implemented properly.
From-SVN: r177187
Jason Merrill [Tue, 2 Aug 2011 15:35:09 +0000 (11:35 -0400)]
* call.c (build_call_a): Also check at_function_scope_p.
From-SVN: r177185
Tobias Burnus [Tue, 2 Aug 2011 15:32:31 +0000 (17:32 +0200)]
trim_optimize_5.f90: Remove spurious "use foo".
2011-08-02 Tobias Burnus <burnus@net-b.de>
* trim_optimize_5.f90: Remove spurious "use foo".
* actual_array_vect_1.f90: Fix or add dg-final cleanup-module
* alloc_comp_assign_7.f90: Ditto.
* allocatable_function_5.f90: Ditto.
* allocate_stat.f90: Ditto.
* array_constructor_20.f90: Ditto.
* array_constructor_21.f90: Ditto.
* array_constructor_22.f90: Ditto.
* array_constructor_26.f03: Ditto.
* array_function_4.f90: Ditto.
* assumed_charlen_function_1.f90: Ditto.
* assumed_size_dt_dummy.f90: Ditto.
* bind_c_usage_15.f90: Ditto.
* bind_c_usage_16.f03: Ditto.
* bind_c_usage_3.f03: Ditto.
* binding_label_tests_7.f03: Ditto.
* class_15.f03: Ditto.
* class_27.f03: Ditto.
* class_33.f90: Ditto.
* class_37.f03: Ditto.
* class_40.f03: Ditto.
* class_42.f03: Ditto.
* class_defined_operator_1.f03: Ditto.
* coarray/registering_1.f90: Ditto.
* convert_1.f90: Ditto.
* default_initialization_3.f90: Ditto.
* dependency_25.f90: Ditto.
* dependency_26.f90: Ditto.
* dependency_36.f90: Ditto.
* dependency_37.f90: Ditto.
* derived_array_intrinisics_1.f90: Ditto.
* dynamic_dispatch_10.f03: Ditto.
* elemental_non_intrinsic_dummy_1.f90 |: Ditto.
* elemental_pointer_1.f90: Ditto.
* elemental_result_1.f90: Ditto.
* empty_derived_type.f90: Ditto.
* entry_7.f90: Ditto.
* entry_dummy_ref_2.f90: Ditto.
* equiv_constraint_3.f90: Ditto.
* error_recovery_3.f90: Ditto.
* extends_4.f03: Ditto.
* func_assign.f90: Ditto.
* func_assign_3.f90: Ditto.
* generic_1.f90: Ditto.
* generic_18.f90: Ditto.
* generic_22.f03: Ditto.
* generic_4.f90: Ditto.
* generic_6.f90: Ditto.
* generic_actual_arg.f90: Ditto.
* graphite/id-21.f: Ditto.
* graphite/pr45758.f90: Ditto.
* host_assoc_call_2.f90: Ditto.
* host_assoc_call_3.f90: Ditto.
* host_assoc_call_4.f90: Ditto.
* host_assoc_function_4.f90: Ditto.
* host_used_types_1.f90: Ditto.
* implicit_1.f90: Ditto.
* implicit_11.f90: Ditto.
* implicit_derived_type_1.f90: Ditto.
* impure_actual_1.f90: Ditto.
* impure_assignment_1.f90: Ditto.
* impure_constructor_1.f90: Ditto.
* initialization_10.f90: Ditto.
* initialization_12.f90: Ditto.
* interface_14.f90: Ditto.
* interface_15.f90: Ditto.
* interface_2.f90: Ditto.
* interface_25.f90: Ditto.
* interface_26.f90: Ditto.
* interface_29.f90: Ditto.
* interface_assignment_1.f90: Ditto.
* internal_pack_6.f90: Ditto.
* internal_pack_7.f90: Ditto.
* internal_pack_8.f90: Ditto.
* lto/pr45586_0.f90: Ditto.
* lto/pr47839_0.f90: Ditto.
* module_commons_3.f90: Ditto.
* module_equivalence_4.f90: Ditto.
* module_equivalence_6.f90: Ditto.
* module_function_type_1.f90: Ditto.
* module_naming_1.f90: Ditto.
* namelist_4.f90: Ditto.
* operator_c1202.f90: Ditto.
* parens_7.f90: Ditto.
* pr32921.f: Ditto.
* pr33646.f90: Ditto.
* pr41928.f90: Ditto.
* pr42119.f90: Ditto.
* pr43984.f90: Ditto.
* present_1.f90: Ditto.
* private_type_2.f90: Ditto.
* proc_decl_2.f90: Ditto.
* proc_ptr_10.f90: Ditto.
* proc_ptr_22.f90: Ditto.
* proc_ptr_8.f90: Ditto.
* proc_ptr_comp_12.f90: Ditto.
* proc_ptr_comp_pass_6.f90: Ditto.
* pure_formal_proc_2.f90: Ditto.
* realloc_on_assign_4.f03: Ditto.
* substring_equivalence.f90: Ditto.
* transfer_assumed_size_1.f90: Ditto.
* transpose_optimization_1.f90: Ditto.
* trim_optimize_6.f90: Ditto.
* typebound_call_12.f03: Ditto.
* typebound_operator_5.f03: Ditto.
* typebound_operator_6.f03: Ditto.
* typebound_proc_18.f03: Ditto.
* unreferenced_use_assoc_1.f90: Ditto.
* use_1.f90: Ditto.
* use_10.f90: Ditto.
* use_11.f90: Ditto.
* use_14.f90: Ditto.
* use_only_2.f90: Ditto.
* use_rename_4.f90: Ditto.
* use_rename_5.f90: Ditto.
* used_dummy_types_3.f90: Ditto.
* used_interface_ref.f90: Ditto.
* used_types_11.f90: Ditto.
* used_types_3.f90: Ditto.
* used_types_4.f90: Ditto.
* userdef_operator_2.f90: Ditto.
* vect/fast-math-vect-8.f90: Ditto.
* vect/pr46213.f90: Ditto.
* whole_file_21.f90: Ditto.
* whole_file_22.f90: Ditto.
* coarray_lib_token_1.f90: Ditto. Add space before "}".
* coarray_lib_token_2.f90: Ditto.
From-SVN: r177184
Arnaud Charlet [Tue, 2 Aug 2011 15:30:55 +0000 (17:30 +0200)]
[multiple changes]
2011-08-02 Vincent Celier <celier@adacore.com>
* link.c: Only import "auto-host.h" when building the gnattools.
2011-08-02 Yannick Moy <moy@adacore.com>
* sem_util.adb: Inter-unit inlining does not work for a subprogram
which calls a local subprogram, so extract subprogram
from Mark_Non_ALFA_Subprogram_Body.
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_ch9.adb
(Extract_Dispatching_Call): If the type of the dispatching object is an
access type then return an explicit dereference in the Object out-mode
parameter.
2011-08-02 Gary Dismukes <dismukes@adacore.com>
* sem_ch3.adb (Analyze_Subtype_Declaration): Generate range
compatibility checks for all indexes of an array subtype, not just the
first. Reset Has_Dynamic_Range_Check on the subtype before each
potential check to ensure that Insert_Range_Checks will not elide any
of the dynamic checks.
2011-08-02 Yannick Moy <moy@adacore.com>
* par-prag.ad (Process_Restrictions_Or_Restriction_Warnings): recognize
SPARK restriction at parsing time.
* scng.adb (Scan): Generate a token Tok_SPARK_Hide for a SPARK HIDE
directive only if the SPARK restriction is set for this unit.
From-SVN: r177183
Yannick Moy [Tue, 2 Aug 2011 15:28:12 +0000 (15:28 +0000)]
sem_ch3.adb, [...]: Protect call to Current_Subprogram which might be costly when repeated.
2011-08-02 Yannick Moy <moy@adacore.com>
* sem_ch3.adb, sem_ch5.adb, sem_ch9.adb, sem_util.adb, sem_util.ads,
sem_res.adb, sem_ch2.adb, sem_ch4.adb, sem_ch6.adb,
sem_ch11.adb: Protect call to Current_Subprogram which might be costly
when repeated. Rename Current_Subprogram_Is_Not_In_ALFA into
Mark_Non_ALFA_Subprogram_Body.
Split body of Mark_Non_ALFA_Subprogram_Body to get body small and
inlined.
From-SVN: r177182
Yannick Moy [Tue, 2 Aug 2011 15:26:43 +0000 (15:26 +0000)]
sem_res.adb: Protect calls to Matching_Static_Array_Bounds which might be costly.
2011-08-02 Yannick Moy <moy@adacore.com>
* sem_res.adb: Protect calls to Matching_Static_Array_Bounds which
might be costly.
From-SVN: r177181
Arnaud Charlet [Tue, 2 Aug 2011 15:25:25 +0000 (17:25 +0200)]
[multiple changes]
2011-08-02 Robert Dewar <dewar@adacore.com>
* exp_ch9.adb, exp_sel.adb, restrict.ads, exp_disp.adb, erroutc.ads,
exp_ch3.adb: Minor reformatting.
2011-08-02 Emmanuel Briot <briot@adacore.com>
* adaint.c (__gnat_locate_exec_on_path): only returns executable
files, not any regular file.
(__gnat_locate_file_with_predicate): new subprogram.
2011-08-02 Yannick Moy <moy@adacore.com>
* sinfo.adb, sinfo.ads: Restrict the use of flags
Has_Dynamic_Length_Check and Has_Dynamic_Range_Check to expression
nodes, plus N_Subtype_Declaration for the 2nd one.
From-SVN: r177180
Arnaud Charlet [Tue, 2 Aug 2011 15:21:19 +0000 (17:21 +0200)]
[multiple changes]
2011-08-02 Sergey Rybin <rybin@adacore.com>
* gnat_rm.texi: Ramification of pragma Eliminate documentation
- fix bugs in the description of Source_Trace;
- get rid of UNIT_NAME;
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_ch9.adb
(Build_Dispatching_Requeue): Adding support for VM targets
since we cannot directly reference the Tag entity.
* exp_sel.adb (Build_K): Adding support for VM targets.
(Build_S_Assignment): Adding support for VM targets.
* exp_disp.adb
(Default_Prim_Op_Position): In VM targets do not restrict availability
of predefined interface primitives to compiling in Ada 2005 mode.
(Is_Predefined_Interface_Primitive): In VM targets this service is not
restricted to compiling in Ada 2005 mode.
(Make_VM_TSD): Generate code that declares and initializes the OSD
record. Needed to support dispatching calls through synchronized
interfaces.
* exp_ch3.adb
(Make_Predefined_Primitive_Specs): Enable generation of predefined
primitives associated with synchronized interfaces.
(Make_Predefined_Primitive_Bodies): Enable generation of predefined
primitives associated with synchronized interfaces.
2011-08-02 Yannick Moy <moy@adacore.com>
* par-ch11.adb (P_Handled_Sequence_Of_Statements): mark a sequence of
statements hidden in SPARK if preceded by the HIDE directive
(Parse_Exception_Handlers): mark each exception handler in a sequence of
exception handlers as hidden in SPARK if preceded by the HIDE directive
* par-ch6.adb (P_Subprogram): mark a subprogram body hidden in SPARK
if starting with the HIDE directive
* par-ch7.adb (P_Package): mark a package body hidden in SPARK if
starting with the HIDE directive; mark the declarations in a private
part as hidden in SPARK if the private part starts with the HIDE
directive
* restrict.adb, restrict.ads
(Set_Hidden_Part_In_SPARK): record a range of slocs as hidden in SPARK
(Is_In_Hidden_Part_In_SPARK): new function which returns whether its
argument node belongs to a part which is hidden in SPARK
(Check_SPARK_Restriction): do not issue violations on nodes in hidden
parts in SPARK; protect the possibly costly call to
Is_In_Hidden_Part_In_SPARK by a check that the SPARK restriction is on
* scans.ads (Token_Type): new value Tok_SPARK_Hide in enumeration
* scng.adb (Accumulate_Token_Checksum_GNAT_6_3,
Accumulate_Token_Checksum_GNAT_5_03): add case for new token
Tok_SPARK_Hide.
(Scan): recognize special comment starting with '#' and followed by
SPARK keyword "hide" as a HIDE directive.
2011-08-02 Yannick Moy <moy@adacore.com>
* types.ads, erroutc.ads: Minor reformatting.
2011-08-02 Vincent Celier <celier@adacore.com>
* link.c: Add response file support for cross platforms.
From-SVN: r177179
Ed Schonberg [Tue, 2 Aug 2011 15:17:00 +0000 (15:17 +0000)]
sem_aggr.adb (Resolve_Array_Aggregate): when copying the expression in an association...
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_aggr.adb (Resolve_Array_Aggregate): when copying the expression
in an association, set parent field of copy before partial analysis.
* sem_res.adb (Resolve_Slice): create reference to itype only when
expansion is enabled.
From-SVN: r177178
Yannick Moy [Tue, 2 Aug 2011 15:15:07 +0000 (15:15 +0000)]
einfo.adb, einfo.ads (Body_Is_In_ALFA, [...]): get/set for new flag denoting which subprogram bodies are in ALFA
2011-08-02 Yannick Moy <moy@adacore.com>
* einfo.adb, einfo.ads (Body_Is_In_ALFA, Set_Body_Is_In_ALFA): get/set
for new flag denoting which subprogram bodies are in ALFA
* restrict.adb, sem_ch7.adb: Update comment
* sem_ch11.adb, sem_ch2.adb, sem_ch3.adb, sem_ch4.adb, sem_ch5.adb,
sem_ch9.adb, sem_res.adb: Add calls to
Current_Subprogram_Body_Is_Not_In_ALFA on unsupported constructs.
* sem_ch6.adb (Analyze_Function_Return): add calls to
Current_Subprogram_Body_Is_Not_In_ALFA on return statement in the
middle of the body, and extended return.
(Check_Missing_Return): add calls to Set_Body_Is_In_ALFA with argument
False when missing return.
(Analyze_Subprogram_Body_Helper): initialize the flag Body_Is_In_ALFA
to True for subprograms whose spec is in ALFA. Remove later on the flag
on the entity used for a subprogram body when there exists a separate
declaration.
* sem_util.adb, sem_util.ads (Current_Subprogram_Body_Is_Not_In_ALFA):
if Current_Subprogram is not Empty, set its flag Body_Is_In_ALFA to
False, otherwise do nothing.
From-SVN: r177177
Arnaud Charlet [Tue, 2 Aug 2011 15:12:44 +0000 (17:12 +0200)]
[multiple changes]
2011-08-02 Robert Dewar <dewar@adacore.com>
* inline.adb, stand.ads, sem_ch6.adb, sem_ch8.adb: Minor reformatting.
2011-08-02 Yannick Moy <moy@adacore.com>
* sem_ch4.ads: minor formatting.
From-SVN: r177176
Yannick Moy [Tue, 2 Aug 2011 15:10:17 +0000 (15:10 +0000)]
sem_aggr.adb, [...]: cleanup of SPARK mode
2011-08-02 Yannick Moy <moy@adacore.com>
* sem_aggr.adb, err_vars.ads, sem_ch3.adb, sem_ch5.adb, sem_ch9.adb,
debug.adb, sem_util.adb, sem_res.adb, sem_attr.adb, gnat1drv.adb,
errout.adb, errout.ads, exp_ch6.adb, sem_ch4.adb, restrict.adb,
restrict.ads, sem_ch6.adb, sem_ch8.adb, sem_ch11.adb,
opt.ads: cleanup of SPARK mode
From-SVN: r177175
Yannick Moy [Tue, 2 Aug 2011 15:07:10 +0000 (15:07 +0000)]
cstand.adb (Create_Standard): sets Is_In_ALFA component of standard types.
2011-08-02 Yannick Moy <moy@adacore.com>
* cstand.adb (Create_Standard): sets Is_In_ALFA component of standard
types.
* einfo.adb, einfo.ads (Is_In_ALFA): add flag for all entities
(Is_In_ALFA, Set_Is_In_ALFA): new subprograms to access flag Is_In_ALFA
* sem_ch3.adb
(Analyze_Object_Declaration): set Is_In_ALFA flag for objects
(Constrain_Enumeration): set Is_In_ALFA flag for enumeration subtypes
(Constrain_Integer): set Is_In_ALFA flag for integer subtypes
(Enumeration_Type_Declaration): set Is_In_ALFA flag for enumeration
types.
(Set_Scalar_Range_For_Subtype): unset Is_In_ALFA flag for subtypes with
non-static range.
* sem_ch6.adb (Analyze_Return_Type): unset Is_In_ALFA flag for
functions whose return type is not in ALFA.
(Analyze_Subprogram_Specification): set Is_In_ALFA flag for subprogram
specifications.
(Process_Formals): unset Is_In_ALFA flag for subprograms if a
parameter's type is not in ALFA.
* stand.ads (Standard_Type_Is_In_ALFA): array defines which standard
types are in ALFA.
From-SVN: r177174
Ed Schonberg [Tue, 2 Aug 2011 15:04:46 +0000 (15:04 +0000)]
sem_ch6 (Analyze_Expression_Function): treat the function as Inline_Always...
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_ch6 (Analyze_Expression_Function): treat the function as
Inline_Always, and introduce a subprogram declaration for it when it is
not a completion.
* inline.adb (Add_Inlined_Body): recognize bodies that come from
expression functions, so that the back-end can determine whether they
can in fact be inlined.
* sem_util.adb (Is_Expression_Function): predicate to determine whether
a function body comes from an expression function.
From-SVN: r177173
Arnaud Charlet [Tue, 2 Aug 2011 15:00:07 +0000 (17:00 +0200)]
[multiple changes]
2011-08-02 Gary Dismukes <dismukes@adacore.com>
* sem_ch6.adb (Check_Conformance): Revise the check for nonconforming
null exclusions to test Can_Never_Be_Null on the anonymous access types
of the formals rather than testing the formals themselves. Exclude this
check in cases where the Old_Formal is marked as a controlling formal,
to avoid issuing spurious errors for bodies completing dispatching
operations (due to the flag not getting set on controlling access
formals in body specs).
(Find_Corresponding_Spec): When checking full and subtype conformance of
subprogram bodies in instances, pass Designated and E in that order, for
consistency with the expected order of the formals (New_Id followed by
Old_Id).
2011-08-02 Robert Dewar <dewar@adacore.com>
* sem_ch8.adb: Minor reformatting.
From-SVN: r177172
Arnaud Charlet [Tue, 2 Aug 2011 14:56:42 +0000 (16:56 +0200)]
[multiple changes]
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_ch8.adb (Analyze_Subprogram_Renaming): new procedure
Check_Class_Wide_Actual, to implement AI05-0071, on defaulted
primitive operations of class-wide actuals.
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_atag.ads, exp_atag.adb
(Build_Common_Dispatching_Select_Statements): Remove argument Loc
since its value is implicitly passed in argument Typ.
* exp_disp.adb (Make_Disp_Conditional_Select_Body,
Make_Disp_Timed_Select_Body): Remove Loc in calls to routine
Build_Common_Dispatching_Select_Statements.
From-SVN: r177171
Kai Tietz [Tue, 2 Aug 2011 14:55:47 +0000 (16:55 +0200)]
gimple.c (canonicalize_cond_expr_cond): Handle cast from boolean-type.
* gimple.c (canonicalize_cond_expr_cond): Handle cast from
boolean-type.
(ssa_forward_propagate_and_combine): Interprete result of
forward_propagate_comparison.
* gcc/gimple-fold.c (fold_gimple_assign): Add canonicalization for
boolean-typed operands for comparisons.
* gcc.dg/tree-ssa/forwprop-15.c: New testcase.
From-SVN: r177170
Robert Dewar [Tue, 2 Aug 2011 14:54:14 +0000 (14:54 +0000)]
sem_ch3.adb, [...]: Update comments.
2011-08-02 Robert Dewar <dewar@adacore.com>
* sem_ch3.adb, exp_atag.ads, get_scos.adb, get_scos.ads,
exp_disp.adb, lib-xref.adb, lib-xref.ads: Update comments.
Minor reformatting.
From-SVN: r177169
Arnaud Charlet [Tue, 2 Aug 2011 14:50:56 +0000 (16:50 +0200)]
[multiple changes]
2011-08-02 Robert Dewar <dewar@adacore.com>
* sem_res.adb: Minor reformatting.
* sem_prag.adb: Minor reformatting.
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_atag.adb, exp_atags.ads
(Build_Common_Dispatching_Select_Statement): Replace argument DT_Ptr
by the tagged type Entity. Required to use this routine in the VM
targets since we do not have available the Tag entity in the VM
platforms.
* exp_ch6.adb
(Expand_N_Subprogram_Body): Do not invoke Build_VM_TSDs if package
Ada.Tags has not been previously loaded.
* exp_ch7.adb
(Expand_N_Package_Declaration, Expand_N_Package_Body): Do not invoke
Build_VM_TSDs if package Ada.Tags has not been previously loaded.
* sem_aux.adb
(Enclosing_Dynamic_Scope): Add missing support to handle the full
view of enclosing scopes. Required to handle enclosing scopes that
are synchronized types whose full view is a task type.
* exp_disp.adb
(Build_VM_TSDs): Minor code improvement to avoid generating and
analyzing lists with empty nodes.
(Make_Disp_Asynchronous_Select_Body): Add support for VM targets.
(Make_Disp_Conditional_Select_Body): Add support for VM targets.
(Make_Disp_Get_Prim_Op_Kind): Add support for VM targets.
(Make_Disp_Timed_Select_Body): Add support for VM targets.
(Make_Select_Specific_Data_Table): Add support for VM targets.
(Make_VM_TSD): Generate code to initialize the SSD structure of
the TSD.
2011-08-02 Yannick Moy <moy@adacore.com>
* lib-writ.adb (Write_ALI): when ALFA mode is set, write local
cross-references section in ALI.
* lib-xref.adb, lib-xref.ads (Xref_Entry): add components Sub
(enclosing subprogram), Slc (location of Sub) and Sun (unit number of
Sub).
(Enclosing_Subprogram_Or_Package): new function to return the enclosing
subprogram or package entity of a node
(Is_Local_Reference_Type): new function returns True for references
selected in local cross-references.
(Lt): function extracted from Lt in Output_References
(Write_Entity_Name): function extracted from Output_References
(Generate_Definition): generate reference with type 'D' for definition
of objects (object declaration and parameter specification), with
appropriate locations and units, for use in local cross-references.
(Generate_Reference): update fields Sub, Slc and Sun. Keep newly created
references of type 'I' for initialization in object definition.
(Output_References): move part of function Lt and procedure
Write_Entity_Name outside of the body. Ignore references of types 'D'
and 'I' introduced for local cross-references.
(Output_Local_References): new procedure to output the local
cross-references sections.
(Lref_Entity_Status): new array defining whether an entity is a local
* sem_ch3.adb (Analyze_Object_Declaration): call Generate_Reference
with 'I' type when initialization expression is present.
* get_scos.adb, get_scos.ads: Correct comments and typos
From-SVN: r177168
Arnaud Charlet [Tue, 2 Aug 2011 14:46:28 +0000 (16:46 +0200)]
[multiple changes]
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_ch6.adb (Expand_N_Subprogram_Body): Enable generation of TSDs in
the JVM target.
* exp_ch7.adb (Expand_N_Package_Body): Enable generation of TSDs in
the JVM target.
* exp_disp.adb (Build_VM_TSDs): No action needed if the runtime has no
TSD support.
2011-08-02 Vincent Celier <celier@adacore.com>
* prj-nmsc.adb (File_Found): New components Excl_File and Excl_Line
(No_Space_Img): New function
(Find_Excluded_Sources): When reading from a file, record the file name
and the line number for each excluded source.
(Mark_Excluded_Sources): When reporting an error, if the excluded
sources were read from a file, include file name and line number in
the error message.
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_res.adb (Resolve_Call): implement rule in RM 12.5.1 (23.3/2).
From-SVN: r177167
H.J. Lu [Tue, 2 Aug 2011 14:44:06 +0000 (14:44 +0000)]
Check __x86_64__ instead of __LP64__ for x86 futex.
2011-08-02 H.J. Lu <hongjiu.lu@intel.com>
* config/linux/x86/futex.h: Check __x86_64__ instead of
__LP64__.
From-SVN: r177166
Robert Dewar [Tue, 2 Aug 2011 14:43:43 +0000 (14:43 +0000)]
exp_ch7.adb [...]: Minor reformatting
2011-08-02 Robert Dewar <dewar@adacore.com>
* exp_ch7.adb exp_ch6.adb, exp_disp.adb: Minor reformatting
From-SVN: r177165
Javier Miranda [Tue, 2 Aug 2011 14:42:28 +0000 (14:42 +0000)]
exp_ch6.adb (Expand_N_Subprogram_Body): Temporarily restrict the generation of TSDs to the DOTNET compiler.
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_ch6.adb (Expand_N_Subprogram_Body): Temporarily restrict the
generation of TSDs to the DOTNET compiler.
* exp_ch7.adb (Expand_N_Package_Body): Temporarily restrict the
generation of TSDs to the DOTNET compiler.
From-SVN: r177164
Javier Miranda [Tue, 2 Aug 2011 14:41:13 +0000 (14:41 +0000)]
exp_disp.ads (Build_VM_TSDs): Build the runtime Type Specific Data record of all the tagged types declared...
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_disp.ads (Build_VM_TSDs): Build the runtime Type Specific Data
record of all the tagged types declared inside library level package
declarations, library level package bodies or library level subprograms.
* exp_disp.adb (Make_VM_TSD): New subprogram that builds the TSD
associated with a given tagged type.
(Build_VM_TSDs): New subprogram.
* exp_ch6.adb (Expand_N_Subprogram_Body): Generate TSDs records of main
compilation units that are subprograms.
* exp_ch7.adb (Expand_N_Package_Body): Generate TSDs of main
compilation units that are package bodies.
(Expand_N_Package_Declaration): Generate TSDs of the main compilation
units that are a package declaration or a package instantiation.
* exp_intr.adb (Expand_Dispatching_Constructor_Call): Minor code
reorganization to improve the error generated by the frontend when the
function Ada.Tags.Secondary_Tag is not available.
* rtsfind.ads (RE_Register_TSD): New runtime entity.
* exp_ch4.adb (Expand_N_Type_Conversion): Minor code cleanup.
From-SVN: r177163
Arnaud Charlet [Tue, 2 Aug 2011 14:39:06 +0000 (16:39 +0200)]
[multiple changes]
2011-08-02 Javier Miranda <miranda@adacore.com>
* exp_disp.adb (Make_DT): Generate call to Check_TSD in Ada 2005 mode.
2011-08-02 Robert Dewar <dewar@adacore.com>
* s-imenne.ads: Minor reformatting.
2011-08-02 Robert Dewar <dewar@adacore.com>
* a-stunau.ads: Add pragma Suppress_Initialization for Big_String
* freeze.adb (Warn_Overlay): Don't warn if initialization suppressed
* s-stalib.ads: Add pragma Suppress_Initialization for Big_String
2011-08-02 Robert Dewar <dewar@adacore.com>
* einfo.ads (Materialize_Entity): Document this is only for renamings
* exp_ch3.adb (Expand_N_Object_Declaration): Make sure we generate
required debug information in the case where we transform the object
declaration into a renaming declaration.
* exp_ch4.adb (Expand_Concatenate): Generate debug info for result
object
* exp_dbug.ads (Debug_Renaming_Declaration): Document setting of
Materialize_Entity.
From-SVN: r177162
Robert Dewar [Tue, 2 Aug 2011 14:35:51 +0000 (14:35 +0000)]
einfo.ads, einfo.adb (Suppress_Initialization): Replaces Suppress_Init_Procs.
2011-08-02 Robert Dewar <dewar@adacore.com>
* einfo.ads, einfo.adb (Suppress_Initialization): Replaces
Suppress_Init_Procs.
* exp_ch3.adb, exp_disp.adb, freeze.adb: Use
Suppress_Initialization/Initialization_Suppressed.
* gnat_rm.texi: New documentation for pragma Suppress_Initialization
* sem_aux.ads, sem_aux.adb (Initialization_Suppressed): New function
* sem_dist.adb: Use Suppress_Initialization/Initialization_Suppressed
* sem_prag.adb: New processing for pragma Suppress_Initialization.
From-SVN: r177161
Robert Dewar [Tue, 2 Aug 2011 14:30:35 +0000 (14:30 +0000)]
gnat_rm.texi, [...]: Minor reformatting.
2011-08-02 Robert Dewar <dewar@adacore.com>
* gnat_rm.texi, a-tags.ads, sem_prag.adb, sem_ch12.adb, exp_disp.adb:
Minor reformatting.
From-SVN: r177160
Arnaud Charlet [Tue, 2 Aug 2011 14:28:32 +0000 (16:28 +0200)]
[multiple changes]
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_prag.adb (Chain_PPC): Implement AI04-0230: null procedures can
only have inheritable classwide pre/postconditions.
2011-08-02 Javier Miranda <miranda@adacore.com>
* a-tags.ads, a-tags.adb (Check_TSD): New subprogram.
* rtsfind.ads (RE_Check_TSD): New runtime entity.
* exp_disp.adb (Make_DT): Generate call to the new runtime routine that
checks if the external tag of a type is the same as the external tag
of some other declaration.
From-SVN: r177159
Arnaud Charlet [Tue, 2 Aug 2011 14:02:45 +0000 (16:02 +0200)]
[multiple changes]
2011-08-02 Thomas Quinot <quinot@adacore.com>
* s-taskin.ads: Minor reformatting.
2011-08-02 Emmanuel Briot <briot@adacore.com>
* g-comlin.adb (Display_Help): swap the order in which it prints the
short help and the general usage.
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_ch12.adb (Analyze_Generic_Subprogram_Declaration): copy properly
the aspect declarations and attach them to the generic copy for
subsequent analysis.
(Analyze_Subprogram_Instantiation): copy explicitly the aspect
declarations of the generic tree to the new subprogram declarations.
* sem_attr.adb (Check_Precondition_Postcondition): recognize
conditions that apply to a subprogram instance.
2011-08-02 Robert Dewar <dewar@adacore.com>
* gnat_rm.texi: Clarify doc on pragma Source_File_Name[_Project].
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_ch3.adb (Derived_Type_Declaration): When checking that a untagged
private type with a tagged full view is not derived in the immediate
scope of the partial view, (RM 7.3 (7)) use the scope of the base type.
From-SVN: r177157
Arnaud Charlet [Tue, 2 Aug 2011 13:51:43 +0000 (15:51 +0200)]
[multiple changes]
2011-08-02 Robert Dewar <dewar@adacore.com>
* exp_ch4.adb: Minor reformatting.
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_ch5.adb (Analyze_Loop_Statement): If the iteration scheme is an
Ada2012 iterator, the loop will be rewritten during expansion into a
while loop with a cursor and an element declaration. Do not analyze the
body in this case, because if the container is for indefinite types the
actual subtype of the elements will only be determined when the cursor
declaration is analyzed.
2011-08-02 Arnaud Charlet <charlet@adacore.com>
* sem_ch13.adb (Analyze_Attribute_Definition_Clause): Ignore
size/alignment related attributes in CodePeer_Mode.
2011-08-02 Gary Dismukes <dismukes@adacore.com>
* sem_ch3.adb (Check_Ops_From_Incomplete_Type): Remove call to
Prepend_Element, since this can result in the operation getting the
wrong slot in the full type's dispatch table if the full type has
inherited operations. The incomplete type's operation will get added
to the proper position in the full type's primitives
list later in Sem_Disp.Check_Operation_From_Incomplete_Type.
(Process_Incomplete_Dependents): Add Is_Primitive test when checking for
dispatching operations, since there are cases where nonprimitive
subprograms can get added to the list of incomplete dependents (such
as subprograms in nested packages).
* sem_ch6.adb (Process_Formals): First, remove test for being in a
private part when determining whether to add a primitive with a
parameter of a tagged incomplete type to the Private_Dependents list.
Such primitives can also occur in the visible part, and should not have
been excluded from being private dependents.
* sem_ch7.adb (Uninstall_Declarations): When checking the rule of
RM05-3.10.1(9.3/2), test that a subprogram in the Private_Dependents
list of a Taft-amendment incomplete type is a primitive before issuing
an error that the full type must appear in the same unit. There are
cases where nonprimitives can be in the list (such as subprograms in
nested packages).
* sem_disp.adb (Derives_From): Use correct condition for checking that
a formal's type is derived from the type of the corresponding formal in
the parent subprogram (the condition was completely wrong). Add
checking that was missing for controlling result types being derived
from the result type of the parent operation.
From-SVN: r177156
Yannick Moy [Tue, 2 Aug 2011 13:46:38 +0000 (13:46 +0000)]
errout.adb (First_Node): minor renaming
2011-08-02 Yannick Moy <moy@adacore.com>
* errout.adb (First_Node): minor renaming
* restrict.adb (Check_Formal_Restriction): put restriction warning on
first node.
From-SVN: r177155
Arnaud Charlet [Tue, 2 Aug 2011 13:43:04 +0000 (15:43 +0200)]
[multiple changes]
2011-08-02 Yannick Moy <moy@adacore.com>
* sem_res.adb (Resolve_Logical_Op): ensure N is a binary operator
before accessing operands.
* sem_util.adb (Is_SPARK_Initialization_Expr): follow original nodes to
decide whether an initialization expression respects SPARK rules, as
the plain node is the expanded one. This allows for more valid warnings
to be issued.
* gnat_rm.texi: Minor update.
2011-08-02 Arnaud Charlet <charlet@adacore.com>
* sem_ch13.adb (Analyze_Enumeration_Representation_Clause): Revert
previous change.
2011-08-02 Robert Dewar <dewar@adacore.com>
* sem_ch3.adb, sem_ch4.adb: Minor reformatting.
2011-08-02 Hristian Kirtchev <kirtchev@adacore.com>
* exp_ch5.adb (Expand_Iterator_Loop): Reformatting. Wrap the original
loop statements and the element renaming declaration with a block when
the element type is controlled.
2011-08-02 Yannick Moy <moy@adacore.com>
* sinfo.ads: Minor formatting.
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_aggr.adb (Add_Association): if the association has a box and no
expression, use the Sloc of the aggregate itself for the new
association.
* errout.adb (First_Node): Exclude nodes with no Sloc, and always use
the Original_Node.
From-SVN: r177153
Arnaud Charlet [Tue, 2 Aug 2011 13:39:08 +0000 (15:39 +0200)]
[multiple changes]
2011-08-02 Hristian Kirtchev <kirtchev@adacore.com>
* exp_ch5.adb (Expand_Iterator_Loop): Code cleanup and refactoring.
When a container is provided via a function call, generate a renaming
of the function result. This avoids the creation of a transient scope
and the premature finalization of the container.
* exp_ch7.adb (Is_Container_Cursor): Removed.
(Wrap_Transient_Declaration): Remove the supression of the finalization
of the list controller when the declaration denotes a container cursor,
it is not needed.
2011-08-02 Yannick Moy <moy@adacore.com>
* restrict.adb (Check_Formal_Restriction): only issue a warning if the
node is from source, instead of the original node being from source.
* sem_aggr.adb
(Resolve_Array_Aggregate): refine the check for a static expression, to
recognize also static ranges
* sem_ch3.adb, sem_ch3.ads (Analyze_Component_Declaration,
Array_Type_Declaration): postpone the test for the type being a subtype
mark after the type has been resolved, so that component-selection and
expanded-name are discriminated.
(Make_Index, Process_Range_Expr_In_Decl): add a parameter In_Iter_Schm
to distinguish the case of an iteration scheme, so that an error is
issed on a non-static range in SPARK except in an iteration scheme.
* sem_ch5.adb (Analyze_Iteration_Scheme): call Make_Index with
In_Iter_Schm = True.
* sem_ch6.adb (Analyze_Subprogram_Specification): refine the check for
user-defined operators so that they are allowed in renaming
* sem_ch8.adb
(Find_Selected_Component): refine the check for prefixing of operators
so that they are allowed in renaming. Move the checks for restrictions
on selector name after analysis discriminated between
component-selection and expanded-name.
* sem_res.adb (Resolve_Op_Concat_Arg): do not issue a warning on
concatenation argument of string type if it is static.
* sem_util.adb, sem_util.ads
(Check_Later_Vs_Basic_Declarations): add a new function
Is_Later_Declarative_Item to decice which declarations are allowed as
later items, in the two different modes Ada 83 and SPARK. In the SPARK
mode, add that renamings are considered as later items.
(Enclosing_Package): new function to return the enclosing package
(Enter_Name): correct the rule for homonyms in SPARK
(Is_SPARK_Initialization_Expr): default to returning True on nodes not
from source (result of expansion) to avoid issuing wrong warnings.
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* errout.adb: On anything but an expression First_Node returns its
argument.
From-SVN: r177152
Arnaud Charlet [Tue, 2 Aug 2011 13:34:00 +0000 (15:34 +0200)]
[multiple changes]
2011-08-02 Pascal Obry <obry@adacore.com>
* prj-proc.adb, make.adb, makeutl.adb: Minor reformatting.
2011-08-02 Hristian Kirtchev <kirtchev@adacore.com>
* exp_ch5.adb (Expand_Iterator_Loop): Code cleanup and reorganization.
Set the associated loop as the related expression of internally
generated cursors.
* exp_ch7.adb (Is_Container_Cursor): New routine.
(Wrap_Transient_Declaration): Supress the finalization of the list
controller when the declaration denotes a container cursor.
2011-08-02 Yannick Moy <moy@adacore.com>
* opt.ads (SPARK_Mode): update comment, SPARK_Mode only set through
command line now.
* par-ch3.adb (P_Delta_Constraint): remove check in SPARK mode that the
expression is a simple expression. This check cannot be performed in
the semantics, so just drop it.
(P_Index_Or_Discriminant_Constraint): move check that the index or
discriminant is a subtype mark to Analyze_Subtype_Declaration in the
semantics. Other cases were previously checked in the semantics.
* par-ch4.adb (P_Name): move checks that a selector name is not
character literal or an operator symbol to Find_Selected_Component in
the semantics
* par-ch5.adb (Parse_Decls_Begin_End): move check that basic
declarations are not placed after later declarations in a separate
procedure in Sem_Util (possibly not the best choice?), to be used both
during parsing, for Ada 83 mode, and during semantic analysis, for
SPARK mode.
* par-endh.adb (Check_End): move check that end label is not missing
to Process_End_Label in the semantics
* par-prag.adb (Process_Restrictions_Or_Restriction_Warnings): remove
the special case for SPARK restriction
* par.adb: use and with Sem_Util, for use in Parse_Decls_Begin_End
* restrict.adb, restrict.ads (Check_Formal_Restriction): add a
parameter Force to issue the error message even on internal node (used
for generated end label). Call Check_Restriction to check when an error
must be issued. In SPARK mode, issue an error message even if the
restriction is not set.
(Check_Restriction): new procedure with an additional out parameter to
inform the caller that a message has been issued
* sem_aggr.adb: Minor modification of message
* sem_attr.adb (Analyze_Attribute): call Check_Formal_Restriction
instead of issuing an error message directly
* sem_ch3.adb (Analyze_Declarations): move here the check that basic
declarations are not placed after later declarations, by calling
Check_Later_Vs_Basic_Declarations
(Analyze_Subtype_Declaration): move here the check that an index or
discriminant constraint must be a subtype mark. Change the check that
a subtype of String must start at one so that it works on subtype marks.
* sem_ch4.adb (Analyze_Call): move here the check that a named
association cannot follow a positional one in a call
* sem_ch5.adb (Check_Unreachable_Code): call Check_Formal_Restriction
instead of issuing an error message directly
* sem_ch8.adb (Find_Selected_Component): move here the check that a
selector name is not a character literal or an operator symbol. Move
here the check that the prefix of an expanded name cannot be a
subprogram or a loop statement.
* sem_util.adb, sem_util.ads (Check_Later_Vs_Basic_Declarations): new
procedure called from parsing and semantics to check that basic
declarations are not placed after later declarations
(Process_End_Label): move here the check that end label is not missing
2011-08-02 Arnaud Charlet <charlet@adacore.com>
* sem_ch13.adb (Analyze_Enumeration_Representation_Clause): Ignore enum
representation clause in codepeer mode, since it confuses CodePeer and
does not bring useful info.
2011-08-02 Ed Falis <falis@adacore.com>
* init.c: initialize fp hw on MILS.
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* errout.adb (First_Node): for bodies, return the node itself (small
optimization). For other nodes, do not check source_unit if the node
comes from Standard.
From-SVN: r177151
Arnaud Charlet [Tue, 2 Aug 2011 13:29:08 +0000 (15:29 +0200)]
[multiple changes]
2011-08-02 Robert Dewar <dewar@adacore.com>
* exp_ch3.adb: Minor comment additions.
* sem_ch13.adb: Minor reformatting.
2011-08-02 Arnaud Charlet <charlet@adacore.com>
* exp_pakd.adb (Expand_Packed_Element_Reference): Revert previous change
2011-08-02 Pascal Obry <obry@adacore.com>
* make.adb, makeutl.adb: Removes some superfluous directory separator.
From-SVN: r177150
Arnaud Charlet [Tue, 2 Aug 2011 13:26:46 +0000 (15:26 +0200)]
Refine previous change.
From-SVN: r177149
Arnaud Charlet [Tue, 2 Aug 2011 13:24:06 +0000 (15:24 +0200)]
Fix bootstrap with g++
From-SVN: r177148
Arnaud Charlet [Tue, 2 Aug 2011 13:16:09 +0000 (15:16 +0200)]
[multiple changes]
2011-08-02 Robert Dewar <dewar@adacore.com>
* sem_attr.adb: Minor reformatting.
2011-08-02 Ed Falis <falis@adacore.com>
* init.c: Revert previous change.
2011-08-02 Robert Dewar <dewar@adacore.com>
* einfo.adb (Has_Default_Aspect): Replaces Has_Default_Value
(Has_Default_Component_Value): Removed
* einfo.ads Comment updates
(Has_Default_Aspect): Replaces Has_Default_Value
(Has_Default_Component_Value): Removed
* exp_ch13.adb
(Expand_N_Freeze_Entity): Handle Default[_Component]_Value aspects
* exp_ch3.adb
(Build_Array_Init_Proc): Handle Default_[Component_]Value aspects
(Get_Simple_Init_Val): Handle Default_Value aspect
(Needs_Simple_Initialization): Handle Default_Value aspect
* exp_ch3.ads: Needs_Simple_Initialization
* freeze.adb (Freeze_Entity): Handle Default_[Component_]Value aspect
* par-prag.adb (Pragma_Default[_Component]Value) Removed
* sem_ch13.adb
(Analyze_Aspect_Specifications): Fix Default[_Component]_Value aspects
* sem_prag.adb (Pragma_Default[_Component]Value) Removed
* snames.ads-tmpl (Pragma_Default[_Component]Value) Removed
2011-08-02 Ed Schonberg <schonberg@adacore.com>
* sem_ch5.adb (Analyze_Iterator_Specification): use base type to locate
package containing iteration primitives.
exp_ch5.adb (Expand_Iterator_Loop): ditto.
From-SVN: r177147
projects / gcc.git / log