2015-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb (New_Overloaded_Entity): In GNATprove mode, a
function wrapper may be a homonym of another local declaration.
* sem_ch8.adb (Analyze_Subprogram_Renaming): In GNATprove mode,
build function and operator wrappers after the actual subprogram
has been resolved, and replace the standard renaming declaration
with the declaration of wrapper.
* sem_ch12.ads (Build_Function_Wrapper, Build_Operator_Wraooer):
make public for use elsewhere.
* sem_ch12.adb (Build_Function_Wrapper, Build_Operator_Wraooer):
rewrite, now that actual is fully resolved when wrapper is
constructed.
2015-01-06 Javier Miranda <miranda@adacore.com>
* exp_disp.adb: Revert previous change.
From-SVN: r219232
2015-01-06 Robert Dewar <dewar@adacore.com>
* exp_util.adb: Change name Name_Table_Boolean to
Name_Table_Boolean1.
* namet.adb: Change name Name_Table_Boolean to Name_Table_Boolean1
Introduce Name_Table_Boolean2/3.
* namet.ads: Change name Name_Table_Boolean to Name_Table_Boolean1
Introduce Name_Table_Boolean2/3.
* par-ch13.adb: Change name Name_Table_Boolean to
Name_Table_Boolean1.
2015-01-06 Bob Duff <duff@adacore.com>
* gnat_rm.texi: Improve documentation regarding No_Task_Termination.
2015-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_aggr.adb (Resolve_Record_Aggregte, Get_Value): For an
others choice that covers multiple components, analyze each
copy with the type of the component even in compile-only mode,
to detect potential accessibility errors.
2015-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* sem_res.adb (Is_Assignment_Or_Object_Expression): New routine.
(Resolve_Actuals): An effectively volatile out
parameter cannot act as an in or in out actual in a call.
(Resolve_Entity_Name): An effectively volatile out parameter
cannot be read.
2015-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb (Analyze_Subprogram_Body_Helper): If the body is
the expansion of an expression function it may be pre-analyzed
if a 'access attribute is applied to the function, in which case
last_entity may have been assigned already.
2015-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch4.adb (Analyze_One_Call): If formal has an incomplete
type and actual has the corresponding full view, there is no
error, but a case of use of incomplete type in a predicate or
invariant expression.
2015-01-06 Vincent Celier <celier@adacore.com>
* makeutl.adb (Insert_No_Roots): Make sure that the same source
in two different project tree is checked in both trees, if they
are sources of two different projects, extended or not.
2015-01-06 Arnaud Charlet <charlet@adacore.com>
* gnat1drv.adb: Minor code clean up.
(Adjust_Global_Switches): Ignore gnatprove_mode in codepeer_mode.
2015-01-06 Bob Duff <duff@adacore.com>
* osint.adb (Read_Source_File): Don't print out
file name unless T = Source.
2015-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_util.adb (Is_Variable, Is_OK_Variable_For_Out_Formal):
recognize improper uses of constant_reference types as actuals
for in-out parameters.
(Check_Function_Call): Do not collect identifiers if function
name is missing because of previous error.
From-SVN: r219231
2015-01-06 Vincent Celier <celier@adacore.com>
* a-strsup.adb (Times (Natural;String;Positive)): Raise
Length_Error, not Index_Error, when the result is too long.
2015-01-06 Thomas Quinot <quinot@adacore.com>
* a-direct.adb (Create_Path): Minor error handling and
performance improvement.
2015-01-06 Robert Dewar <dewar@adacore.com>
* checks.ads, sem_ch12.adb: Minor reformatting.
* exp_ch4.adb (Expand_N_Op_Divide): Generate explicit divide by
zero check for fixed-point case if Backend_Divide_Checks_On_Target
is False.
2015-01-06 Robert Dewar <dewar@adacore.com>
* sem_prag.adb (Analyze_Pragma, case No_Elaboration_Code_All):
Do not set restriction No_Elaboration_Code unless the pragma
appears in the main unit).
2015-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch10.adb (Is_Regular_With_Clause): Add guard to verify
that with clause has already been analyzed before checking kind
of with_clause.
2015-01-06 Robert Dewar <dewar@adacore.com>
* exp_strm.adb (Build_Elementary_Input_Call): Return base type
(as required by RM).
From-SVN: r219228
2015-01-06 Arnaud Charlet <charlet@adacore.com>
* a-reatim.adb ("/"): Add explicit pragma Unsuppress (Division_Check).
2015-01-06 Robert Dewar <dewar@adacore.com>
* sem_prag.adb (Process_Suppress_Unsuppress): Add extra warning
for ignoring pragma Suppress (Elaboration_Check) in SPARK mode.
2015-01-06 Javier Miranda <miranda@adacore.com>
* exp_disp.adb (Expand_Interface_Conversion): No displacement
of the pointer to the object needed when the type of the operand
is not an interface type and the interface is one of its parent
types (since they share the primary dispatch table).
From-SVN: r219227
2015-01-06 Vincent Celier <celier@adacore.com>
* prj-env.adb: Minor comment update.
2015-01-06 Javier Miranda <miranda@adacore.com>
* sem_res.adb (Valid_Conversion): Restrict the checks on anonymous
access types whose target type is an interface type to operands
that are access types; required to report an error when the
operand is not an access type.
From-SVN: r219226
2015-01-06 Bob Duff <duff@adacore.com>
* a-cfinve.adb (Copy): Set the discriminant to the Length when
Capacity = 0.
* a-cofove.ads (Capacity): Add a postcondition.
* a-cfinve.ads (Capacity): Add a postcondition.
(Reserve_Capacity): Correct the postcondition in the case where
Capacity = 0; that means "Capacity => Length (Container)".
* a-cofove.adb (Elems[c]): Add a comment
explaining the dangers and how to avoid them.
2015-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch12.adb: Code clean up.
From-SVN: r219225
2015-01-06 Robert Dewar <dewar@adacore.com>
* s-taskin.ads, s-traces.ads: Minor reformatting.
* exp_util.adb: Minor typo fix.
2015-01-06 Vincent Celier <celier@adacore.com>
* gnatls.adb (Search_RTS): Invoke Initialize_Default_Project_Path
with the runtime name.
* prj-env.adb (Initialize_Default_Project_Path): When both
Target_Name and Runtime_Name are not empty string, add to the
project path the two directories .../lib/gnat and .../share/gpr
related to the runtime.
* prj-env.ads (Initialize_Default_Project_Path): New String
parameter Runtime_Name, defaulted to the empty string.
2015-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* frontend.adb: Guard against the case where a configuration
pragma may be split into multiple pragmas and the original
rewritten as a null statement.
* sem_prag.adb (Analyze_Pragma): Insert a brand new Check_Policy
pragma using Insert_Before rather than Insert_Action. This
takes care of the configuration pragma case where Insert_Action
would fail.
2015-01-06 Bob Duff <duff@adacore.com>
* a-coboho.ads (Element_Access): Add "pragma
No_Strict_Aliasing (Element_Access);". This is needed because
we are unchecked-converting from Address to Element_Access.
* a-cofove.ads, a-cofove.adb (Elems,Elemsc): Fix bounds of the
result to be 1.
2015-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* sem_res.adb (Resolve_Actuals): Remove the
restriction which prohibits volatile actual parameters with
enabled external propery Async_Writers to act appear in procedure
calls where the corresponding formal is of mode OUT.
From-SVN: r219222
If named types with similar names are defined inside of methods
with similar names, it was possible that the mangled type names would be
the same, leading to multiple definitions of type descriptor and
garbage collection info variables.
This can only occur with similarly named methods across different
receivers, so we use the receiver name to distinguish.
Fixes issue 33.
From-SVN: r219214
c-family/
PR c++/31397
* c.opt (Wsuggest-override): New option.
cp/
PR c++/31397
* class.c (check_for_override): Warn when a virtual function is an
override not marked override.
gcc/
PR c++/31397
* doc/invoke.texi: Document -Wsuggest-override.
From-SVN: r219213
Currently building gcc for xtensa configuration with XCHAL_HAVE_LOOPS
set to 0 fails with the following error:
xtensa.c:3952: undefined reference to `reorg_loops(bool, hw_doloop_hooks*)'
Only compile target-specific zero-overhead loop optimization code when
TARGET_LOOPS allows it.
2014-12-30 Max Filippov <jcmvbkbc@gmail.com>
* config/xtensa/xtensa.c (hwloop_optimize, hwloop_fail,
hwloop_pattern_reg, xtensa_doloop_hooks, xtensa_reorg_loops):
put under #if TARGET_LOOPS guard.
From-SVN: r219206
PR sanitizer/64265
* gimplify.c (gimplify_function_tree): Add TSAN_FUNC_EXIT internal
call as cleanup of the whole body.
* internal-fn.def (TSAN_FUNC_EXIT): New internal call.
* tsan.c (replace_func_exit): New function.
(instrument_func_exit): Moved earlier.
(instrument_memory_accesses): Adjust TSAN_FUNC_EXIT internal calls.
Call instrument_func_exit if no TSAN_FUNC_EXIT internal calls have
been found.
(tsan_pass): Don't call instrument_func_exit.
* internal-fn.c (expand_TSAN_FUNC_EXIT): New function.
* tree-inline.c (copy_bb): Drop TSAN_FUNC_EXIT internal calls during
inlining.
From-SVN: r219202
PR sanitizer/64344
* ubsan.h (ubsan_instrument_float_cast): Add ARG argument.
* ubsan.c (ubsan_instrument_float_cast): Add ARG argument, pass
it to libubsan handler instead of EXPR. Fold comparisons earlier,
if the result is integer_zerop, return NULL_TREE.
* convert.c (convert_to_integer): Pass expr as ARG.
c/
* c-typeck.c (convert_for_assignment, c_finish_return): For
-fsanitize=float-cast-overflow casts from REAL_TYPE to integer/enum
types also set in_late_binary_op around convert call.
* c-convert.c (convert): For -fsanitize=float-cast-overflow REAL_TYPE
to integral type casts, if not in_late_binary_op, pass c_fully_fold
result on expr as last argument to ubsan_instrument_float_cast,
if in_late_binary_op, don't use c_save_expr but save_expr.
testsuite/
* c-c++-common/ubsan/pr64344-1.c: New test.
* c-c++-common/ubsan/pr64344-2.c: New test.
From-SVN: r219201
PR tree-optimization/64465
* tree-inline.c (redirect_all_calls): During inlining
clean up EH stmts and EH edges if redirect_call_stmt_to_callee
changed the stmt to a non-throwing call.
* gcc.dg/pr64465.c: New test.
From-SVN: r219200
With my last change, `sed' is used to cut out the target name from a listed
target. Since there may be additional OPTions encoded in the "target", I tried
to get only the first submatch before an `OPT'. However, `sed' uses longest
match, so I'm re-writing this using awk.
If anybody is like using `gawk' or anything different, please feel free to
drop another patch. Since this is usually called by hand or by robots under
review, I don't see much of a problem here.
2015-01-05 Jan-Benedict Glaw <jbglaw@lug-owl.de>
contrib/
* config-list.mk: Use shortest match for OPT to find the actual
target name.
From-SVN: r219196
2015-01-05 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/47674
* dependency.c: Update copyright years.
(gfc_discard_nops): Add prototype.
* dependency.c (discard_nops): Rename to gfc_discard_nops,
make non-static.
(gfc_discard_nops): Use gfc_discard_nops.
(gfc_dep_difference): Likewise.
* frontend-passes.c Update copyright years.
(realloc_strings): New function. Add prototype.
(gfc_run_passes): Call realloc_strings.
(realloc_string_callback): New function.
(create_var): Add prototype. Handle case of a
scalar character variable.
(optimize_trim): Do not handle allocatable variables.
2015-01-05 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/47674
* gfortran.dg/realloc_on_assign_25.f90: New test.
From-SVN: r219193
PR c/64423
c-family/
* c-common.c (warn_array_subscript_with_type_char): Add location_t
parameter. Use it.
* c-common.h (warn_array_subscript_with_type_char): Update
declaration.
c/
* c-typeck.c (build_array_ref): Pass loc down to
warn_array_subscript_with_type_char.
cp/
* typeck.c (cp_build_array_ref): Pass loc down to
warn_array_subscript_with_type_char.
testsuite/
* gcc.dg/pr64423.c: New test.
From-SVN: r219186
* gcc.dg/debug/debug-1.c: Pass -fno-if-conversion for
mmix-knuth-mmixware for the same reason as for MIPS.
* gcc.dg/debug/debug-2.c: Ditto.
From-SVN: r219181
