The original code only propagated into PHI arguments if the value was
a constant. This behavior was lost in the conversion, allowing
any value (SSAs for instance) to be propagated into PHIs.
gcc/ChangeLog:
PR tree-optimization/95649
* tree-ssa-propagate.c (propagate_into_phi_args): Do not propagate unless
value is a constant.
gcc/testsuite/ChangeLog:
* g++.dg/tree-ssa/pr95649.C: New test.
* gcc.dg/tree-ssa/pr95649.c: New test.
gcc/fortran/ChangeLog
* parse.c (decode_oacc_directive): Permit 'acc routine' also
inside pure procedures.
* openmp.c (gfc_match_oacc_routine): Inside pure procedures
do not permit gang, worker or vector clauses.
libgomp/ChangeLog:
* testsuite/libgomp.oacc-fortran/routine-10.f90: New test.
gcc/testsuite/ChangeLog:
* gfortran.dg/goacc/pure-elemental-procedures-2.f90: New test.
Reviewed-by: Thomas Schwinge <thomas@codesourcery.com>
OpenMP since 4.5 permits IMPURE ELEMENTAL in directives and
the code already only checked for PURE. – Followup for
-fopenmp-simd.
gcc/fortran/ChangeLog:
* parse.c (decode_omp_directive): Remove "or ELEMENTAL"
from "in PURE" error message also for -fopenmp-simd.
gcc/testsuite/ChangeLog:
* gfortran.dg/gomp/pr79154-simd.f90: New test.
Another part of 95369 is that we accept designated initializers with
non-aggregate types. That seems to be wrong since they're part of
aggregate initialization. clang/icc also reject it.
There are multiple contexts where we can use designated initializers:
function-like casts, member list initializers, NTTP, etc. I've adjusted
add_list_candidates and implicit_conversion_error in order to to detect
this case.
gcc/cp/ChangeLog:
PR c++/95369
* call.c (add_list_candidates): Return if a designated initializer
is used with a non-aggregate.
(implicit_conversion_error): Give an error for the case above.
gcc/testsuite/ChangeLog:
PR c++/95369
* g++.dg/cpp2a/desig11.C: Adjust dg-error.
* g++.dg/cpp2a/desig16.C: New test.
Another indication that perhaps this warning is emitted too early. We
crash because same_type_p gets a null type: we have an enumerator
without a fixed underlying type and finish_enum_value_list hasn't yet
run. So check if the type is null before calling same_type_p.
PR c++/95560
* name-lookup.c (check_local_shadow): Check if types are
non-null before calling same_type_p.
* g++.dg/warn/Wshadow-local-3.C: New test.
These tests are UNRESOLVED because -fdump-tree-optimized can't be scanned
with slim LTO. Other *san/ tests deal with this by adding -ffat-lto-objects.
2020-06-16 Jakub Jelinek <jakub@redhat.com>
* c-c++-common/asan/inline.c: Add -ffat-lto-objects to dg-options.
* c-c++-common/asan/inline-kernel.c: Likewise.
* c-c++-common/ubsan/inline.c: Likewise.
Since 87cb9423ad vector alignment hints are emitted for target z13,
too. This patch changes this behaviour in the sense that alignment
hints are only emitted for target z13 if the assembler accepts them.
gcc/ChangeLog:
* config.in: Regenerate.
* config/s390/s390.c (print_operand): Emit vector alignment hints
for target z13, if AS accepts them. For other targets the logic
stays the same.
* config/s390/s390.h (TARGET_VECTOR_LOADSTORE_ALIGNMENT_HINTS): Define
macro.
* configure: Regenerate.
* configure.ac: Check HAVE_AS_VECTOR_LOADSTORE_ALIGNMENT_HINTS_ON_Z13.
This patch modifies the MVE scalar shift RTL patterns. The current patterns
have wrong constraints and predicates due to which the values returned from
MVE scalar shift instructions are overwritten in the code-gen.
example:
$ cat x.c
int32_t foo(int64_t acc, int shift)
{
return sqrshrl_sat48 (acc, shift);
}
Code-gen before applying this patch:
$ arm-none-eabi-gcc -march=armv8.1-m.main+mve -mfloat-abi=hard -O2 -S
$ cat x.s
foo:
push {r4, r5}
sqrshrl r0, r1, #48, r2 ----> (a)
mov r0, r4 ----> (b)
pop {r4, r5}
bx lr
Code-gen after applying this patch:
foo:
sqrshrl r0, r1, #48, r2
bx lr
In the current compiler the return value (r0) from sqrshrl (a) is getting
overwritten by the mov statement (b).
This patch fixes above issue.
2020-06-12 Srinath Parvathaneni <srinath.parvathaneni@arm.com>
gcc/
* config/arm/mve.md (mve_uqrshll_sat<supf>_di): Correct the predicate
and constraint of all the operands.
(mve_sqrshrl_sat<supf>_di): Likewise.
(mve_uqrshl_si): Likewise.
(mve_sqrshr_si): Likewise.
(mve_uqshll_di): Likewise.
(mve_urshrl_di): Likewise.
(mve_uqshl_si): Likewise.
(mve_urshr_si): Likewise.
(mve_sqshl_si): Likewise.
(mve_srshr_si): Likewise.
(mve_srshrl_di): Likewise.
(mve_sqshll_di): Likewise.
* config/arm/predicates.md (arm_low_register_operand): Define.
gcc/testsuite/
* gcc.target/arm/mve/intrinsics/mve_scalar_shifts1.c: New test.
* gcc.target/arm/mve/intrinsics/mve_scalar_shifts2.c: Likewise.
* gcc.target/arm/mve/intrinsics/mve_scalar_shifts3.c: Likewise.
* gcc.target/arm/mve/intrinsics/mve_scalar_shifts4.c: Likewise.
OpenMP 5.0 adds support for non-rectangular loop collapses, e.g.
triangular and more complex.
This patch deals just with the diagnostics so that they aren't rejected
immediately as before. As the spec generally requires as before that the
iteration variable initializer and bound in the comparison as invariant
vs. the outermost loop, and just add some exceptional forms that can violate
that, we need to avoid folding the expressions until we can detect them and
in order to avoid folding it later on, I chose to use a TREE_VEC in those
expressions to hold the var_outer * expr1 + expr2 triplet, the patch adds
pretty-printing of that, gimplification etc. and just sorry_at during
omp expansion for now.
The next step will be to implement the different cases of that one by one.
2020-06-16 Jakub Jelinek <jakub@redhat.com>
gcc/
* tree.h (OMP_FOR_NON_RECTANGULAR): Define.
* gimplify.c (gimplify_omp_for): Diagnose schedule, ordered
or dist_schedule clause on non-rectangular loops. Handle
gimplification of non-rectangular lb/b expressions. When changing
iteration variable, adjust also non-rectangular lb/b expressions
referencing that.
* omp-general.h (struct omp_for_data_loop): Add m1, m2 and outer
members.
(struct omp_for_data): Add non_rect member.
* omp-general.c (omp_extract_for_data): Handle non-rectangular
loops. Fill in non_rect, m1, m2 and outer.
* omp-low.c (lower_omp_for): Handle non-rectangular lb/b expressions.
* omp-expand.c (expand_omp_for): Emit sorry_at for unsupported
non-rectangular loop cases and assert for cases that can't be
non-rectangular.
* tree-pretty-print.c (dump_mem_ref): Formatting fix.
(dump_omp_loop_non_rect_expr): New function.
(dump_generic_node): Handle non-rectangular OpenMP loops.
* tree-pretty-print.h (dump_omp_loop_non_rect_expr): Declare.
* gimple-pretty-print.c (dump_gimple_omp_for): Handle non-rectangular
OpenMP loops.
gcc/c-family/
* c-common.h (c_omp_check_loop_iv_exprs): Add an int argument.
* c-omp.c (struct c_omp_check_loop_iv_data): Add maybe_nonrect and
idx members.
(c_omp_is_loop_iterator): New function.
(c_omp_check_loop_iv_r): Use it. Add support for silent scanning
if outer loop iterator is present. Perform duplicate checking through
hash_set in the function rather than expecting caller to do that.
Pass NULL instead of d->ppset to walk_tree_1.
(c_omp_check_nonrect_loop_iv): New function.
(c_omp_check_loop_iv): Use it. Fill in new members, allow
non-rectangular loop forms, diagnose multiple associated loops with
the same iterator. Pass NULL instead of &pset to walk_tree_1.
(c_omp_check_loop_iv_exprs): Likewise.
gcc/c/
* c-parser.c (c_parser_expr_no_commas): Save, clear and restore
c_in_omp_for.
(c_parser_omp_for_loop): Set c_in_omp_for around some calls to avoid
premature c_fully_fold. Defer explicit c_fully_fold calls to after
c_finish_omp_for.
* c-tree.h (c_in_omp_for): Declare.
* c-typeck.c (c_in_omp_for): Define.
(build_modify_expr): Avoid c_fully_fold if c_in_omp_for.
(digest_init): Likewise.
(build_binary_op): Likewise.
gcc/cp/
* semantics.c (handle_omp_for_class_iterator): Adjust
c_omp_check_loop_iv_exprs caller.
(finish_omp_for): Likewise. Don't call fold_build_cleanup_point_expr
before calling c_finish_omp_for and c_omp_check_loop_iv, move it
after those calls.
* pt.c (tsubst_omp_for_iterator): Handle non-rectangular loops.
gcc/testsuite/
* c-c++-common/gomp/loop-6.c: New test.
* gcc.dg/gomp/loop-1.c: Don't expect diagnostics on valid
non-rectangular loops.
* gcc.dg/gomp/loop-2.c: New test.
* g++.dg/gomp/loop-1.C: Don't expect diagnostics on valid
non-rectangular loops.
* g++.dg/gomp/loop-2.C: Likewise.
* g++.dg/gomp/loop-5.C: New test.
* g++.dg/gomp/loop-6.C: New test.
2020-06-16 Jakub Jelinek <jakub@redhat.com>
gcc/c/
* c-parser.c (c_parser_omp_clause_schedule): Reject modifier separated
from kind by comma rather than colon.
gcc/cp/
* parser.c (cp_parser_omp_clause_schedule): Reject modifier separated
from kind by comma rather than colon.
gcc/testsuite/
* c-c++-common/gomp/schedule-modifiers-2.c: New test.
OpenMP since 4.5 permits IMPURE ELEMENTAL in directives and
the code already only checked for PURE.
gcc/fortran/ChangeLog:
* parse.c (decode_omp_directive): Remove "or ELEMENTAL"
from "in PURE" error message.
gcc/testsuite/ChangeLog:
* gfortran.dg/gomp/pr79154-1.f90: Update dg-*;
add an impure elemental example.
* gfortran.dg/gomp/pr79154-2.f90: Likewise.
2020-06-16 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* sinfo.ads (Conversion_OK): Document use for 'Pos and 'Val.
* exp_attr.adb (Get_Integer_Type): New function returning a
small integer type appropriate for an enumeration type.
(Expand_N_Attribute_Reference) <Attribute_Enum_Rep>: Call it.
<Attribute_Pos>: For an enumeration type with a standard
representation, expand to a conversion with Conversion_OK.
<Attribute_Val>: Likewise.
* exp_ch4.adb (Expand_N_Type_Conversion): Do not expand when
the target is an enumeration type and Conversion_OK is set.
2020-06-16 Piotr Trojanek <trojanek@adacore.com>
gcc/ada/
* exp_util.adb (Evaluate_Name): Force evaluation of aggregates;
recursively evaluate expression of a qualified expression; fix
location of the comment for an attribute referenced and an
indexed component.
2020-06-16 Steve Baird <baird@adacore.com>
gcc/ada/
* libgnat/a-nbnbin.ads, libgnat/a-nbnbre.ads: Uncomment the
commented-out Integer_Literal aspect specification for type
Big_Integer.
2020-06-16 Javier Miranda <miranda@adacore.com>
gcc/ada/
* exp_ch6.adb (BIP_Suffix_Kind, Check_BIP_Actuals,
Is_Build_In_Place_Entity): New subprograms.
(Make_Build_In_Place_Call_In_Allocator,
Make_Build_In_Place_Call_In_Anonymous_Context,
Make_Build_In_Place_Call_In_Assignment,
Make_Build_In_Place_Call_In_Object_Declaration): Add assertions.
(Needs_BIP_Task_Actuals): Add missing support for thunks.
(Expand_Actuals): Ensure that the BIP call has available an
activation chain and the _master variable.
* exp_ch9.adb (Find_Enclosing_Context): Initialize the list of
declarations of empty blocks when the _master variable must be
declared and the list was not available.
2020-06-16 Bob Duff <duff@adacore.com>
gcc/ada/
* par-ch4.adb (P_Case_Expression): Move to be local.
(P_Declare_Expression): New parsing routine.
(P_Unparen_Cond_Expr_Etc): New name for
P_Unparen_Cond_Case_Quant_Expression which was missing one case
in its name (iterated component association), and we're adding a
new case (declare expression), so lets use "Etc" instead of
trying to pack all those things into the name. Add call to
P_Declare_Expression, and check for missing parens.
(P_Expression_If_OK, P_Expression_Or_Range_Attribute_If_OK): Add
Tok_Declare.
* par.adb (P_Basic_Declarative_Items): Add parameter
Declare_Expression so we can tailor the error message about
incorrect bodies.
(P_Case_Expression): Move to body.
* par-ch3.adb (P_Basic_Declarative_Items): Tailor the error
message about incorrect bodies.
* par-ch7.adb (P_Package): Pass Declare_Expression => False to
P_Basic_Declarative_Items.
* sem.ads (In_Declare_Expr): Counter used to determine whether
we are analyzing a declare_expression. Needed to give errors
about things that are not allowed in declare_expression, such as
the 'Access attribute.
* sem.adb (Do_Analyze): Save/restore In_Declare_Expr.
* sem_ch4.adb (Analyze_Expression_With_Actions): Give this node
its own scope. That seems better in general, but it is
necessary for declare_expressions. For example, an identifier
declared in a declare_expression should not clash with the same
identifier in an outer scope. If this is a declare_expression,
indicated by Comes_From_Source, then check legality rules, and
incr/decr In_Declare_Expr.
* sem_aggr.adb (Resolve_Aggregate): Allow an applicable index
constraint for a declare_expression, so if its expression is an
array aggregate, it can have "others => ...".
* sem_attr.adb (Analyze_Access_Attribute): Disallow these
attributes in declare_expressions. Add comment to make it clear
that Unrestricted_Access is included.
* sinfo.ads, sinfo.adb, atree.ads, atree.adb: Remove the
now-incorrect comment in sinfo.ads that says
N_Expression_With_Actions has no proper scope. Add 17-parameter
versions of Nkind_In. Remove the 16-parameter versions of
Nkind_In.
2020-06-16 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_warn.adb (Warn_On_Overlapping_Actuals): Simplify code and
implement AI12-0216 which clarifies the conditions under which
overlapping actuals in a call are illegal. If proper warnings
are enabled, GNAT also emits warnings in legal cases of
overlopping actuals.
2020-06-16 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* freeze.adb (Freeze_Expression): Stop climbing the parent chain
at a N_{Case,If}_Expression node for a type or an entity that
does not come from source.
2020-06-16 Steve Baird <baird@adacore.com>
gcc/ada/
* snames.ads-tmpl: Define names of the three new aspects.
* aspects.ads: Define the three new aspects.
* sem_util.ads, sem_util.adb, sem_dim.adb: Move the function
String_From_Numeric_Literal from being declared in the body of
package Sem_Dim to being declared in the visible part of package
Sem_Util.
* sem_ch13.ads, sem_ch13.adb: Declare new visible procedure
Validate_Literal_Aspect. This is where most of the legality
checking occurs for an aspect specification for one of the three
new aspects, as well as resolution of the subprogram named in
the aspect specification. Follow example of other aspects (e.g.,
Validate_Literal_Aspect is called in much the same way as
Validate_Iterable_Aspect in Analyze_Aspects_At_Freeze_Point; a
small amount of legality checking is performed in
Analyze_One_Aspect in much the same way as for Default_Value or
Default_Component_Value aspects). Most of the work is done in
Validate_Literal_Aspect.
* contracts.adb (Add_Contract_Item): Call
Validate_Literal_Aspect in much the same way that
Validate_Iterable_Aspect was already being called.
* sem_res.adb (Resolve): Rewrite a literal as a call if it is a
user-defined literal. This is where the dynamic semantics of
the 3 new aspects are implemented.
* sem_ch6.adb (Fully_Conformant_Expressions): Two numeric
literals that have different text but the same value (e.g.,
12345 and 12_345) do not conform if they are user-defined
literals. Introduce a new function
User_Defined_Numeric_Literal_Mismatch to avoid duplication in
making this check.
* sem_type.adb (Has_Compatible_Type): A numeric literal can be
compatible with a non-numeric type (and a string literal can be
compatible with a non-string type) if it can be interpreted as a
user-defined literal.
2020-06-16 Arnaud Charlet <charlet@adacore.com>
gcc/ada/
* sem_ch4.adb (Analyze_Membership_Op): Reset entity of equality
nodes for membership tests with singletons.
(Analyze_User_Defined_Binary_Op): Always perform the analysis
since nodes coming from the expander also may refer to non
standard operators as part of membership expansion.
* exp_ch4.adb (Expand_Set_Membership.Make_Cond): Reset entity of
equality node.
* sem_type.ads: Fix typo in comment.
2020-06-16 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* freeze.adb (In_Expanded_Body): Remove unreachable code.
(Freeze_Expression): Rename a couple of local variables.
In the case of an expanded body, also freeze locally the
entities declared in a nested block.
2020-06-16 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* sem_ch4.adb (Transform_Object_Operation): Document that it
may be partially destructive for the parent of the node.
(Try_Object_Operation): Undo the changes made above on failure.
2020-06-16 Javier Miranda <miranda@adacore.com>
gcc/ada/
* restrict.adb (Global_No_Tasking): Adding
Targparm.Restrictions_On_Target Fixes regressions with zfp.
2020-06-16 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* freeze.adb: (Freeze_Expression, In_Expanded_Body): Treat the
generated body of an expression function like other bodies
generated during expansion (e.g. stream subprograms) so that
those bodies are not treated as freezing points. Handle properly
other global references in such completions.
2020-06-16 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* sem_ch12.adb (Validate_Derived_Type_Instance): Reword error
message for 12.5.1(8) subclause and add secondary message if
the incompatibility comes from the predicates.
* sem_ch3.adb (Check_Constraining_Discriminant): New procedure
to give the error required by the 3.7(15) subclause. Mention
"statically" in the error message and add secondary message
if the incompatibility comes from the predicates.
(Build_Derived_Concurrent_Type): Call it when a new discriminant
constrains an old one.
(Build_Derived_Record_Type): Likewise.
* sem_eval.ads (Predicates_Compatible): Declare.
* sem_eval.adb (Predicates_Compatible): New function to implement
the compatibility of predicates specified by the 4.9.1 clause.
(Subtypes_Statically_Compatible): Call it.
This adds an assert to enforce_access to verify that we don't defer
access checks of dependent decls -- we should instead be rechecking the
access of such a decl after tsubst'ing into the user of the decl.
gcc/cp/ChangeLog:
* pt.c (perform_instantiation_time_access_checks): No need to
tsubst into decl.
* semantics.c (enforce_access): Verify that decl is not
dependent.
The previous patch mostly avoided making any changes that had no
functional impact, such as adjusting now-outdated comments and
performing renamings. Such changes have been consolidated to this
followup patch for easier review.
The main change here is that we now reuse struct deferred_access_check
as the element type of the vector TI_TYPEDEFS_NEEDING_ACCESS_CHECKING
(now renamed to TI_DEFERRED_ACCESS_CHECKS, since it may contain any kind
of access check).
gcc/cp/ChangeLog:
PR c++/41437
PR c++/47346
* cp-tree.h (qualified_typedef_usage_s): Delete.
(qualified_typedef_usage_t): Delete.
(deferred_access_check): Move up in file.
(tree_template_info::typedefs_needing_access_checking): Delete.
(tree_template_info::deferred_access_checks): New field.
(TI_TYPEDEFS_NEEDING_ACCESS_CHECKING): Rename to ...
(TI_DEFERRED_ACCESS_CHECKS): ... this, and adjust accordingly.
* pt.c (perform_typedefs_access_check): Rename to ...
(perform_instantiation_time_access_checks): ... this, and adjust
accordingly. Remove unnecessary tree tests.
(instantiate_class_template_1): Adjust accordingly.
(instantiate_decl): Likewise.
* semantics.c (enforce_access): Likewise.
This patch generalizes our existing functionality for deferring access
checking of typedefs when parsing a function or class template to now
defer all kinds of access checks until template instantiation time,
including member function and member object accesses.
Since all access checks eventually go through enforce_access, the main
component of this patch is new handling inside enforce_access to defer
the current access check if we're inside a template. The bulk of the
rest of the patch consists of removing now-unneeded code pertaining to
suppressing access checks inside templates or pertaining to
typedef-specific access handling. Renamings and other changes with no
functional impact have been split off into the followup patch.
gcc/cp/ChangeLog:
PR c++/41437
PR c++/47346
* call.c (enforce_access): Move to semantics.c.
* cp-tree.h (enforce_access): Delete.
(get_types_needing_access_check): Delete.
(add_typedef_to_current_template_for_access_check): Delete.
* decl.c (make_typename_type): Adjust accordingly. Use
check_accessibility_of_qualified_id instead of directly using
perform_or_defer_access_check.
* parser.c (cp_parser_template_declaration_after_parameters):
Don't push a dk_no_check access state when parsing a template.
* pt.c (get_types_needing_access_check): Delete.
(append_type_to_template_for_access_check_1): Delete.
(perform_typedefs_access_check): Adjust. If type_decl is a
FIELD_DECL, also check its DECL_CONTEXT for dependence. Use
tsubst_copy instead of tsubst to substitute into type_decl so
that we substitute into the DECL_CONTEXT of a FIELD_DECL.
(append_type_to_template_for_access_check): Delete.
* search.c (accessible_p): Remove the processing_template_decl
early exit.
* semantics.c (enforce_access): Moved from call.c. If we're
parsing a template and the access check failed, add the check to
TI_TYPEDEFS_NEEDING_ACCESS_CHECKING.
(perform_or_defer_access_check): Adjust comment.
(add_typedef_to_current_template_for_access_check): Delete.
(check_accessibility_of_qualified_id): Adjust accordingly.
Exit early if the scope is dependent.
gcc/testsuite/ChangeLog:
PR c++/41437
PR c++/47346
* g++.dg/cpp2a/concepts-using2.C: Adjust.
* g++.dg/lto/20081219_1.C: Adjust.
* g++.dg/lto/20091002-1_0.C: Adjust.
* g++.dg/lto/pr65475c_0.C: Adjust.
* g++.dg/opt/dump1.C: Adjust.
* g++.dg/other/pr53574.C: Adjust.
* g++.dg/template/access30.C: New test.
* g++.dg/template/access31.C: New test.
* g++.dg/wrappers/wrapper-around-type-pack-expansion.C: Adjust.
libstdc++-v3/ChangeLog:
PR libstdc++/94003
* testsuite/20_util/is_constructible/94003.cc: New test.
The following avoids calling set_mem_attributes on the
DECL_INITIAL of a CONST_DECL which seems pointless since there
cannot be a sensible MEM_EXPR derived from that. We're overwriting
both other possibly useful info, alias-set and alignment immediately
so the following patch simply removes the call instead of making
the function deal with even more (unexpected) trees that are not
memory accesses.
2020-06-16 Richard Biener <rguenther@suse.de>
PR middle-end/95690
* varasm.c (build_constant_desc): Remove set_mem_attributes call.
* gfortran.dg/pr95690.f90: New testcase.
