By adding omp_get_initial_device and omp_get_num_devices builtins for
C, C++, and Fortran, the following can be achieved:
* By making them pure, multiple calls can be avoiding in some cases.
* Some comparisons can be optimized at compile time.
omp_get_initial_device will be converted to omp_get_num_devices for
consistency; note that OpenMP 6 also permits omp_initial_device (== -1)
as value.
If GCC has not been configure for offloading, either intrinsic will
leads to 0 - and on the offload side, -1 (= omp_initial_device) is
returned for omp_initial_device.
gcc/fortran/ChangeLog:
* f95-lang.cc (ATTR_PURE_NOTHROW_LIST): Define.
* trans-expr.cc (get_builtin_fn): Handle omp_get_num_devices
and omp_get_intrinsic_device.
* gfortran.h (gfc_option_t): Add disable_omp_... for them.
* options.cc (gfc_handle_option): Handle them with
-fno-builtin-.
* libgomp.texi (omp_get_num_devices, omp_get_intrinsic_device):
Document builtin handling.
gcc/testsuite/ChangeLog:
* c-c++-common/gomp/omp_get_num_devices_initial_device-2.c: New test.
* c-c++-common/gomp/omp_get_num_devices_initial_device.c: New test.
* gfortran.dg/gomp/omp_get_num_devices_initial_device-2.f90: New test.
* gfortran.dg/gomp/omp_get_num_devices_initial_device.f90: New test.
Tobias Burnus [Fri, 6 Jun 2025 13:57:50 +0000 (15:57 +0200)]
builtins.def: Enable OpenMP/OpenACC builtins also with -fno-nonansi-builtins
The flags -std=c.. and -std=c++.. imply -fno-nonansi-builtins, which disabled
the OpenMP/OpenACC intrinsics - but -fopenmp/-fopenacc builtin use should be
rather othogonal to the user's choice between -std=c... and -std=gnuc...
gcc/ChangeLog:
* builtins.def (DEF_GOACC_BUILTIN_COMPILER, DEF_GOMP_BUILTIN_COMPILER):
Set NONANSI_P = false to enable those also with -fno-nonansi-builtins.
Tobias Burnus [Thu, 5 Jun 2025 08:36:21 +0000 (10:36 +0200)]
gcn: Update --with-arch= for newer archs
Replace hard-coded list of supported devices by directly checking
config/gcn/gcn-devices.def.
gcc/ChangeLog:
* config.gcc (--with-{arch,tune}): Use .def file to validate gcn
processor names.
* doc/install.texi (amdgcn*-*-*): Update list of devices supported
by --with-arch/--with-tune.
Patrick Palka [Fri, 6 Jun 2025 13:34:17 +0000 (09:34 -0400)]
libstdc++: Fix flat_map::operator[] for const lvalue keys [PR120432]
The const lvalue operator[] overload wasn't properly forwarding the key
type to the generic overload, causing a hard error for const keys.
Rather than correcting the forwarded type this patch just makes the
non-template overloads call try_emplace directly instead. That way we
can remove the non-standard same_as constraint on the generic overload
and match the spec more closely.
PR libstdc++/120432
libstdc++-v3/ChangeLog:
* include/std/flat_map (flat_map::operator[]): Make the
non-template overloads call try_emplace directly. Remove
non-standard same_as constraint on the template overload.
* testsuite/23_containers/flat_map/1.cc (test08): New test.
Reviewed-by: Tomasz Kamiński <tkaminsk@redhat.com> Reviewed-by: Jonathan Wakely <jwakely@redhat.com>
(cherry picked from commit 91ed3248ce26aaaee4d7471aa4edbc07b3f1a90e)
Tomasz Kamiński [Wed, 4 Jun 2025 09:05:11 +0000 (11:05 +0200)]
libstdc++: Fix format call in formatting with empty specs for durations.
This patches fixes an obvious error, where the output iterator argument was
missing for call to format_to, when duration with custom representation types
are used.
libstdc++-v3/ChangeLog:
* include/bits/chrono_io.h (__formatter_chrono:_M_s): Add missing
__out argument to format_to call.
Reviewed-by: Jonathan Wakely <jwakely@redhat.com> Signed-off-by: Tomasz Kamiński <tkaminsk@redhat.com>
(cherry picked from commit ac0a04b7a254fb8e1d8d7088336bcb4375807b1e)
Richard Biener [Fri, 30 May 2025 12:11:47 +0000 (14:11 +0200)]
tree-optimization/120357 - ICE with early break vectorization
When doing early break vectorization of a loop with a conditional
reduction the epilog creation code is confused as to before which exit
to insert the conditional reduction induction IV update. The
following make sure this is done before the main IV exit.
PR tree-optimization/120357
* tree-vect-loop.cc (vect_create_epilog_for_reduction): Create
the conditional reduction induction IV increment before the
main IV exit.
* gcc.dg/vect/vect-early-break_136-pr120357.c: New testcase.
Richard Biener [Fri, 9 May 2025 06:38:45 +0000 (08:38 +0200)]
rtl-optimization/120182 - wrong-code with RTL DSE and constant addresses
RTL DSE forms store groups from unique invariant bases but that is
confused when presented with constant addresses where it assigns
one store group per unique address. That causes it to not consider
0x101:QI to alias 0x100:SI. Constant accesses can really alias
to every object, in practice they appear for I/O and for access
to objects fixed via linker scripts for example. So simply avoid
registering a store group for them.
PR rtl-optimization/120182
* dse.cc (canon_address): Constant addresses have no
separate store group.
Richard Biener [Wed, 30 Apr 2025 09:52:17 +0000 (11:52 +0200)]
tree-optimization/120003 - missed jump threading
The following allows the entry and exit block of a jump thread path
to be equal, which can easily happen when there isn't a forwarder
on the interesting edge for an FSM thread conditional. We just
don't want to enlarge the path from such a block.
PR tree-optimization/120003
* tree-ssa-threadbackward.cc (back_threader::find_paths_to_names):
Allow block re-use but do not enlarge the path beyond such a
re-use.
* gcc.dg/tree-ssa/ssa-thread-23.c: New testcase.
* gcc.dg/tree-ssa/ssa-dom-thread-7.c: Adjust.
The following addresses a too conservative sanity check of SLP nodes
we want to promote external. The issue lies in code generation
for such external which relies on get_later_stmt to figure an
insert location. But get_later_stmt relies on the ability to
totally order stmts, specifically implementation-wise that they
are all from the same BB, which is what is verified at the moment.
The patch changes this to require stmts to be orderable by
dominance queries. For simplicity and seemingly enough for the
testcase in PR119960, this handles the case of two distinct BBs.
PR tree-optimization/119960
* tree-vect-slp.cc (vect_slp_can_convert_to_external):
Handle cases where defs from multiple BBs are ordered
by their dominance relation.
Richard Biener [Thu, 8 May 2025 08:56:16 +0000 (10:56 +0200)]
tree-optimization/116352 - amend previous fix
The previous fix restricted external vector builds to defs from
the same basic-block. That turns out too restrictive so we have
to mitigate the original issue in a different way which is
restricting it to the original case where all defs are in the
same basic-block.
PR tree-optimization/116352
* tree-vect-slp.cc (vect_build_slp_tree_2): When compressing
operands from a two-operator node make sure the resulting
operation does not mix defs from different basic-blocks.
Richard Biener [Tue, 29 Apr 2025 13:08:52 +0000 (15:08 +0200)]
tree-optimization/119960 - add validity checking to SLP scheduling
The following adds checks that when we search for a vector stmt
insert location we arrive at one where all required operand defs
are dominating the insert location. At the moment any such
failure only blows up during SSA verification.
There's the long-standing issue that we do not verify there
exists a valid schedule of the SLP graph from BB vectorization
into the existing CFG. We do not have the ability to insert
vector stmts on the dominance frontier "end", nor to insert
LC PHIs that would be eventually required.
This should be done all differently, computing the schedule
during analysis and failing if we can't schedule.
Richard Biener [Wed, 14 May 2025 14:36:29 +0000 (16:36 +0200)]
Fix regression from x86 multi-epilogue tuning
With the avx512_two_epilogues tuning enabled for zen4 and zen5
the gcc.target/i386/vect-epilogues-5.c testcase below regresses
and ends up using AVX2 sized vectors for the masked epilogue
rather than AVX512 sized vectors. The following patch rectifies
this and adds coverage for the intended behavior.
* config/i386/i386.cc (ix86_vector_costs::finish_cost):
Do not suggest a first epilogue mode for AVX512 sized
main loops with X86_TUNE_AVX512_TWO_EPILOGUES as that
interferes with using a masked epilogue.
OpenMP: Fix regressions in metadirective-target-device-2.c [PR120518]
My previous patch that added a CLEANUP_POINT_EXPR around the device_num
selector expression in the C++ front end broke the testcase
c-c++-common/gomp/metadirective-target-device-2.c on offload targets.
It confused the code in omp_device_num_check that tries to bypass error
checking and do early resolution when the expression is a call to one
of the OpenMP library functions. The solution is to make that code smart
enough to look inside a CLEANUP_POINT_EXPR.
gcc/ChangeLog
PR c++/120518
* omp-general.cc (omp_device_num_check): Look inside a
CLEANUP_POINT_EXPR when trying to optimize special cases.
OpenMP: Handle more cases in user/condition selector
Tobias had noted that the C front end was not treating C23 constexprs
as constant in the user/condition selector property, which led to
missed opportunities to resolve metadirectives at parse time.
Additionally neither C nor C++ was permitting the expression to have
pointer or floating-point type -- the former being a common idiom in
other C/C++ conditional expressions. By using the existing front-end
hooks for the implicit conversion to bool in conditional expressions,
we also get free support for using a C++ class object that has a bool
conversion operator in the user/condition selector.
gcc/c/ChangeLog
* c-parser.cc (c_parser_omp_context_selector): Call
convert_lvalue_to_rvalue and c_objc_common_truthvalue_conversion
on the expression for OMP_TRAIT_PROPERTY_BOOL_EXPR.
gcc/cp/ChangeLog
* cp-tree.h (maybe_convert_cond): Declare.
* parser.cc (cp_parser_omp_context_selector): Call
maybe_convert_cond and fold_build_cleanup_point_expr on the
expression for OMP_TRAIT_PROPERTY_BOOL_EXPR.
* pt.cc (tsubst_omp_context_selector): Likewise.
* semantics.cc (maybe_convert_cond): Remove static declaration.
Javier Miranda [Thu, 6 Feb 2025 09:40:57 +0000 (09:40 +0000)]
ada: Spurious compilation error with repeated loop index
When multiple for-loop statements in the same scope use the
same index name to iterate through container elements, the
compiler reports a spurious error indicating a conflict
between index names.
gcc/ada/ChangeLog:
* exp_ch7.adb (Process_Object_Declaration): Avoid generating
duplicate names for master nodes.
If the body of a loop includes a raise statement then the loop should not be
considered to be free of side-effects and therefore eligible for elimination
by the compiler.
gcc/ada/ChangeLog:
* sem_util.adb
(Side_Effect_Free_Statements): Return False if the statement list
includes an explicit (i.e. Comes_From_Source) raise statement.
The generation of the check mandated by Ada issue AI05-0073 was not done
handled properly for protected types when used through subtypes. This
patch fixes the issue.
gcc/ada/ChangeLog:
* exp_ch4.adb (Tagged_Membership): Fix for protected types.
Eric Botcazou [Fri, 24 Jan 2025 09:26:13 +0000 (10:26 +0100)]
ada: Implement built-in-place expansion of two-pass array aggregates
These are array aggregates containing only component associations that are
iterated with iterator specifications, as per RM 4.3.3(20.2/5-20.4/5).
It is implemented for the array aggregates that are used to initialize an
object, as specified by RM 7.6(17.2/3-17.3/3) for immutably limited types
and types that need finalization, but for all types like other aggregates.
gcc/ada/ChangeLog:
* exp_aggr.adb (Build_Two_Pass_Aggr_Code): New function containing
most of the code initially present in Two_Pass_Aggregate_Expansion.
(Two_Pass_Aggregate_Expansion): Remove redundant N parameter.
Implement built-in-place expansion for (static) object declarations
and allocators, using Build_Two_Pass_Aggr_Code for the main work.
(Expand_Array_Aggregate): Adjust Two_Pass_Aggregate_Expansion call.
Replace Etype (N) by Typ in a couple of places.
* exp_ch3.adb (Expand_Freeze_Array_Type): Remove special case for
two-pass array aggregates.
(Expand_N_Object_Declaration): Do not adjust the object when it is
initialized by a two-pass array aggregate.
* exp_ch4.adb (Expand_Allocator_Expression): Apply the processing
used for container aggregates to two-pass array aggregates.
* exp_ch6.adb (Validate_Subprogram_Calls): Skip calls present in
initialization expressions of N_Object_Declaration nodes that have
No_Initialization set.
* sem_ch3.adb (Analyze_Object_Declaration): Detect the cases of an
array originally initialized by an aggregate consistently.
Viljar Indus [Mon, 20 Jan 2025 13:10:22 +0000 (15:10 +0200)]
ada: Reject Valid_Value arguments originating from Standard
The constraint for Valid_Value not applying to types from Standard
should also apply to all types derived from those types.
gcc/ada/ChangeLog:
* doc/gnat_rm/implementation_defined_attributes.rst: Update the
documentation for Valid_Value.
* sem_attr.adb (Analyze_Attribute): Reject types where
the root type originates from Standard.
* gnat_rm.texi: Regenerate.
Gary Dismukes [Sat, 18 Jan 2025 01:11:12 +0000 (01:11 +0000)]
ada: Error about assignment to limited target on aggregate with "for of" iterator
The compiler reports a spurious error about an assignment to a limited
object on an aggregate of a array type with limited components that has
an association with a "for of" iterator. This is fixed by arranging to
have the Assignment_OK flag set on the indexed_names generated by the
expander for initializing the aggregate object.
gcc/ada/ChangeLog:
* exp_aggr.adb (Two_Pass_Aggregate_Expansion): Change call to Make_Assignment
for the indexed aggregate object to call Change_Make_OK_Assignment instead.
Eric Botcazou [Wed, 15 Jan 2025 19:37:48 +0000 (20:37 +0100)]
ada: Fix buffer overflow for function call returning discriminated limited record
This occurs when the discriminated limited record type is declared with
default values for its discriminants, is not controlled, and the context
of the call is anonymous, i.e. the result of the call is not assigned
to an object. In this case, a temporary is created to hold the result
of the call, with the default values of the discriminants, but the result
may have different values for the discriminants and, in particular, may
be larger than the temporary, which leads to a buffer overflow.
This problem does not occur when the context is an object declaration, so
the fix just makes sure that the expansion in an anonymous context always
uses the model of an object declaration. It requires a minor tweak to the
helper function Entity_Of of the Sem_Util package.
gcc/ada/ChangeLog:
* exp_ch6.adb (Expand_Actuals): Remove obsolete comment.
(Make_Build_In_Place_Call_In_Anonymous_Context): Always use a proper
object declaration initialized with the function call in the cases
where a temporary is needed, with Assignment_OK set on it.
* sem_util.adb (Entity_Of): Deal with rewritten function call first.
Steve Baird [Mon, 13 Jan 2025 22:18:26 +0000 (14:18 -0800)]
ada: Fix compile-time failure due to duplicated attribute subprograms.
For a given type, and for certain attributes (the 4 streaming attributes
and, for Ada2022, the Put_Image attribute), the compiler needs to keep track
of whether a subprogram has already been generated for the given
type/attribute pair. In some cases this was being done incorrectly;
the compiler ended up generating duplicate subprograms (with the same
name), resulting in compilation failures. This could occur if the prefix
of an attribute reference denoted a subtype (more precisely, a non-first
subtype). This includes the case of a subtype declaration that is implicitly
introduced by the compiler to capture the binding between a formal type
in a generic and the corresponding actual type in an instantiation.
gcc/ada/ChangeLog:
* exp_attr.adb (Expand_N_Attribute_Reference): When accessing the
maps declared in package Cached_Attribute_Ops, the key value
passed to Get or to Set should never be the entity node for a
subtype. Use the entity of the corresponding type declaration
instead.
Steve Baird [Fri, 10 Jan 2025 21:15:18 +0000 (13:15 -0800)]
ada: Avoid calling Resolve with Stand.Any_Fixed as the expected type
When we call Resolve for an expression, we pass in the expected type
for that expression. In the absence of semantic errors, that expected type
should never be any of the "Any_xxx" types declared in stand.ads (e.g.,
Any_Array, Any_Numeric, Any_Real). In particular, it should never be Any_Fixed.
Fix a case in which this rule was being violated.
gcc/ada/ChangeLog:
* sem_res.adb
(Set_Mixed_Mode_Operand): If we are about to call Resolve
passing in Any_Fixed as the expected type, then instead pass in
the fixed point type of the other operand (i.e., B_Typ).
Gary Dismukes [Fri, 10 Jan 2025 22:39:52 +0000 (22:39 +0000)]
ada: Compiler crash on array aggregate association iterating over function result
The compiler triggers a bug box when compiling an array aggregate with
an iterated_component_association that iterates over another array object,
failing when trying to retrieve a Choices field, which isn't an allowed
field for N_Iterated_Component_Association nodes. This occurs in procedure
Check_Function_Writable_Actuals, which wasn't accounting for the iterated
association forms.
gcc/ada/ChangeLog:
* sem_util.adb (Check_Function_Writable_Actuals): Add handling for
N_Iterated_Component_Association and N_Iterated_Element_Association.
Fix a typo in an RM reference (6.4.1(20/3) => 6.4.1(6.20/3)).
(Collect_Expression_Ids): New procedure factoring code for collecting
identifiers from expressions of aggregate associations.
(Handle_Association_Choices): New procedure factoring code for handling
id collection for expressions of aggregate associations with multiple
choices. Removed redundant test of Box_Present from original code.
What happens is that symtab_remove_unreachable_nodes leaves the last symbol
in kind of a limbo state: in .remove_symbols, we have:
opt7_pkg__enum_name_table/13 (Opt7_Pkg.Enum_Name_Table)
Type: variable
Body removed by symtab_remove_unreachable_nodes
Visibility: externally_visible semantic_interposition external public
References:
Referring: opt7_pkg__image/2 (read) opt7_pkg__image/2 (read)
Availability: not_available
Varpool flags: initialized read-only const-value-known
This means that the "body" (DECL_INITIAL) of the symbol has been disregarded
during reachability analysis, causing the first two symbols to be discarded:
but the DECL_INITIAL is explicitly preserved for later constant folding,
which makes it possible to retrofit the DECLs corresponding to the first
two symbols in the GIMPLE IR and ultimately leads to the crash.
gcc/
* tree-vect-data-refs.cc (vect_can_force_dr_alignment_p): Return
false if the variable has no symtab node.
gcc/testsuite/
* gnat.dg/specs/opt7.ads: New test.
* gnat.dg/specs/opt7_pkg.ads: New helper.
* gnat.dg/specs/opt7_pkg.adb: Likewise.
Thomas Schwinge [Mon, 26 May 2025 11:31:54 +0000 (13:31 +0200)]
Avoid SIGSEGV in nvptx 'mkoffload' for voluminous PTX code
In commit 50be486dff4ea2676ed022e9524ef190b92ae2b1
"nvptx: libgomp+mkoffload.cc: Prepare for reverse offload fn lookup", some
additional tracking of the PTX code was added, and this assumes that
potentially every single character of PTX code needs to be tracked as a new
chunk of PTX code. That's problematic if we're dealing with voluminous PTX
code (for example, non-trivial C++ code), and the 'file_idx' 'alloca'tion then
causes stack overflow. For example:
FAIL: libgomp.c++/target-std__valarray-1.C (test for excess errors)
UNRESOLVED: libgomp.c++/target-std__valarray-1.C compilation failed to produce executable
lto-wrapper: fatal error: [...]/build-gcc/gcc//accel/nvptx-none/mkoffload terminated with signal 11 [Segmentation fault], core dumped
gcc/
* config/nvptx/mkoffload.cc (process): Use an 'auto_vec' for
'file_idx'.
Harald Anlauf [Fri, 30 May 2025 17:25:15 +0000 (19:25 +0200)]
Fortran: parameter inquiries of constant complex arrays [PR102599,PR114022]
PR fortran/102599
PR fortran/114022
gcc/fortran/ChangeLog:
* expr.cc (simplify_complex_array_inquiry_ref): Helper function for
simplification of inquiry references (%re/%im) of constant complex
arrays.
(find_inquiry_ref): Use it for handling %re/%im inquiry references
of complex arrays.
(scalarize_intrinsic_call): Fix frontend memleak.
* primary.cc (gfc_match_varspec): When the reference is NULL, the
previous simplification has succeeded in evaluating inquiry
references also of arrays.
Harald Anlauf [Tue, 27 May 2025 17:23:16 +0000 (19:23 +0200)]
Fortran: fix parsing of type parameter inquiries of substrings [PR101735]
Handling of type parameter inquiries of substrings failed to due either
parsing issues or not following or handling reference chains properly.
PR fortran/101735
gcc/fortran/ChangeLog:
* expr.cc (find_inquiry_ref): If an inquiry reference applies to
a substring, use that, and calculate substring length if needed.
* primary.cc (extend_ref): Also handle attaching to end of
reference chain for appending.
(gfc_match_varspec): Discrimate between arrays of character and
substrings of them. If a substring is taken from a character
component of a derived type, get the proper typespec so that
inquiry references work correctly.
(gfc_match_rvalue): Handle corner case where we hit a seemingly
dangling '%' and missed an inquiry reference. Try another match.
Jakub Jelinek [Sat, 10 May 2025 19:20:09 +0000 (21:20 +0200)]
fortran: Fix debug info for unsigned(kind=1) and unsigned(kind=4) [PR120193]
As the following testcase shows, debug info for unsigned(kind=1)
and unsigned(kind=4) vars is wrong while unsigned(kind=2), unsigned(kind=8)
and unsigned(kind=16) look right.
Instead of objects having unsigned(kind=1) type they have character(kind=1)
and instead of unsigned(kind=4) they have character(kind=4).
This means in gdb e.g. unsigned(kind=1) :: a(2) variable initialized to
97 will print as 'aa' rather than (97, 97) etc.
While there can be just one unsigned_char_type_node and one
unsigned_type_node type, each can have arbitrary number of variants
(e.g. consider C
typedef unsigned char uc;
where uc is a variant type to unsigned char) or even distinct types
with different TYPE_MAIN_VARIANT.
The following patch uses a variant of the character(kind=4) type
for unsigned(kind=4) and a distinct type based on character(kind=1)
type for unsigned(kind=1). The reason for the latter is that
unsigned_char_type_node has TYPE_STRING_FLAG set on it, so it has
DW_AT_encoding DW_ATE_unsigned_char rather than DW_ATE_unsigned and
so the debugger then likes to print it as characters rather than numbers.
That is IMHO in Fortran desirable for character(kind=1) but not for
unsigned(kind=1). I've made sure TYPE_CANONICAL of the unsigned(kind=1)
type is still character(kind=1), so they are considered compatible by
the middle-end also e.g. for aliasing etc.
2025-05-10 Jakub Jelinek <jakub@redhat.com>
PR fortran/120193
* trans-types.cc (gfc_init_types): For flag_unsigned use
build_distinct_type_copy or build_variant_type_copy from
gfc_character_types[index_char] if index_char > -1 instead of
gfc_character_types[index_char] or
gfc_build_unsigned_type (&gfc_unsigned_kinds[index]).
Jerry DeLisle [Sat, 31 May 2025 15:57:22 +0000 (08:57 -0700)]
Fortran: Fix handling of parsed format strings.
Previously parsed strings with errors were being cached such
that subsequent use of the format string were not being
checked for errors.
PR libfortran/119856
libgfortran/ChangeLog:
* io/format.c (parse_format_list): Set the fmt->error
message for missing comma.
(parse_format): Do not cache the parsed format string
if a previous error ocurred.
Jerry DeLisle [Wed, 28 May 2025 14:56:12 +0000 (07:56 -0700)]
Fortran: Adjust handling of optional comma in FORMAT.
This change adjusts the error messages for optional commas
in format strings to give a warning at compile time unless
-std=legacy is used. This is more consistant with the
runtime library. A missing comma separator should not be
encouraged as it is non-standard fortran.
PR fortran/119856
gcc/fortran/ChangeLog:
* io.cc: Set missing comma error checks to STD_STD_LEGACY.
Patrick Palka [Thu, 29 May 2025 14:12:23 +0000 (10:12 -0400)]
libstdc++: Compare keys and values separately in flat_map::operator==
Instead of effectively doing a zipped comparison of the keys and values,
compare them separately to leverage the underlying containers' optimized
equality implementations.
libstdc++-v3/ChangeLog:
* include/std/flat_map (_Flat_map_impl::operator==): Compare
keys and values separately.
Patrick Palka [Thu, 29 May 2025 14:11:57 +0000 (10:11 -0400)]
libstdc++: Fix tuple/pair confusion with std::erase_if(flat_map) [PR120465]
std::erase_if for flat_map/multimap is implemented via ranges::erase_if
over a zip_view of the keys and values, the value_type of which is a
tuple, but the given predicate needs to be called with a pair (flat_map's
value_type). So use a projection to convert the tuple into a suitable
pair.
PR libstdc++/120465
libstdc++-v3/ChangeLog:
* include/std/flat_map (_Flat_map_impl::_M_erase_if): Use a
projection with ranges::remove_if to pass a pair instead of
a tuple to the predicate.
* testsuite/23_containers/flat_map/1.cc (test07): Strengthen
to expect the argument passed to the predicate is a pair.
* testsuite/23_containers/flat_multimap/1.cc (test07): Likewise.
Co-authored-by: Jonathan Wakely <jwakely@redhat.com> Reviewed-by: Tomasz Kamiński <tkaminsk@redhat.com> Reviewed-by: Jonathan Wakely <jwakely@redhat.com>
(cherry picked from commit 589b27ec5769410e036df57645ff1eb7c765f692)
Jerry DeLisle [Tue, 20 May 2025 02:41:16 +0000 (19:41 -0700)]
Fortran: Fix c_associated argument checks.
PR fortran/120049
gcc/fortran/ChangeLog:
* check.cc (gfc_check_c_associated): Use new helper functions.
Only call check_c_ptr_1 if optional c_ptr_2 tests succeed.
(check_c_ptr_1): Handle only c_ptr_1 checks.
(check_c_ptr_2): Expand checks for c_ptr_2 and handle cases
where there is no derived pointer in the gfc_expr and check
the inmod_sym_id only if it exists. Rephrase error message.
* misc.cc (gfc_typename): Handle the case for BT_VOID rather
than throw an internal error.
gcc/testsuite/ChangeLog:
* gfortran.dg/pr120049_a.f90: Update test directives.
* gfortran.dg/pr120049_b.f90: Update test directives
* gfortran.dg/pr120049_2.f90: New test.
* gfortran.dg/c_f_pointer_tests_6.f90: Adjust dg-error
directive.
Thomas Koenig [Fri, 30 May 2025 11:31:58 +0000 (13:31 +0200)]
Type mismatch for passed external function
This obvious and simple patch fixes a 15/16 regression where the
typespec of a global function was in the RESULT clause and not
in the symbol itself.
gcc/fortran/ChangeLog:
PR fortran/120355
* interface.cc (compare_parameter): If the global function has a
result clause, take typespec from there for the comparison against
the dummy argument.
gcc/testsuite/ChangeLog:
PR fortran/120355
* gfortran.dg/interface_62.f90: New test.
Sandra Loosemore [Mon, 26 May 2025 19:21:48 +0000 (19:21 +0000)]
OpenMP: Fix ICE and other issues in C/C++ metadirective error recovery.
The new testcase included in this patch used to ICE in gcc after
diagnosing the first error, and in g++ it only diagnosed the error in
the first metadirective, ignoring the second one. The solution is to
make error recovery in the C front end more like that in the C++ front
end, and remove the code in both front ends that previously tried to
skip all the way over the following statement (instead of just to the
end of the metadirective pragma) after an error.
gcc/c/ChangeLog
* c-parser.cc (c_parser_skip_to_closing_brace): New, copied from
the equivalent function in the C++ front end.
(c_parser_skip_to_end_of_block_or_statement): Pass false to
the error flag.
(c_parser_omp_context_selector): Immediately return error_mark_node
after giving an error that the integer trait property is invalid,
similarly to C++ front end.
(c_parser_omp_context_selector_specification): Likewise handle
error return from c_parser_omp_context_selector similarly to C++.
(c_parser_omp_metadirective): Do not call
c_parser_skip_to_end_of_block_or_statement after an error.
gcc/cp/ChangeLog
* parser.cc (cp_parser_omp_metadirective): Do not call
cp_parser_skip_to_end_of_block_or_statement after an error.
gcc/testsuite/ChangeLog
* c-c++-common/gomp/declare-variant-2.c: Adjust patterns now that
C and C++ now behave similarly.
* c-c++-common/gomp/metadirective-error-recovery.c: New.
Waffl3x [Mon, 26 May 2025 08:38:27 +0000 (02:38 -0600)]
Add 'libgomp.c++/target-flex-[...].C' test cases
libgomp/ChangeLog:
* testsuite/libgomp.c++/target-flex-10.C: New test.
* testsuite/libgomp.c++/target-flex-100.C: New test.
* testsuite/libgomp.c++/target-flex-101.C: New test.
* testsuite/libgomp.c++/target-flex-11.C: New test.
* testsuite/libgomp.c++/target-flex-12.C: New test.
* testsuite/libgomp.c++/target-flex-2000.C: New test.
* testsuite/libgomp.c++/target-flex-2001.C: New test.
* testsuite/libgomp.c++/target-flex-2002.C: New test.
* testsuite/libgomp.c++/target-flex-2003.C: New test.
* testsuite/libgomp.c++/target-flex-30.C: New test.
* testsuite/libgomp.c++/target-flex-300.C: New test.
* testsuite/libgomp.c++/target-flex-31.C: New test.
* testsuite/libgomp.c++/target-flex-32.C: New test.
* testsuite/libgomp.c++/target-flex-33.C: New test.
* testsuite/libgomp.c++/target-flex-41.C: New test.
* testsuite/libgomp.c++/target-flex-60.C: New test.
* testsuite/libgomp.c++/target-flex-61.C: New test.
* testsuite/libgomp.c++/target-flex-62.C: New test.
* testsuite/libgomp.c++/target-flex-70.C: New test.
* testsuite/libgomp.c++/target-flex-80.C: New test.
* testsuite/libgomp.c++/target-flex-81.C: New test.
* testsuite/libgomp.c++/target-flex-90.C: New test.
* testsuite/libgomp.c++/target-flex-common.h: New test.
Tobias Burnus [Wed, 28 May 2025 13:14:14 +0000 (15:14 +0200)]
libgomp.fortran/metadirective-1.f90: Expect 'error:' for nvptx compile [PR118694]
This should have been part of commit r16-838-gb3d07ec7ac2ccd or r16-883-g5d6ed6d604ff94 - all showing the same issue:
'!$omp target' followed by a metadirective with 'teams'; if
the metadirective cannot be early resolved, a diagnostic
error is shown about using directives between 'target' and
'teams'.
While the message is misleading, the problem is that the
host invokes 'target' differently when 'teams' is present;
in this case, host fallback + amdgcn offload require the
no-teams case, nvptx offload the teams case such that it
only can be resolved at runtime.
Mark the error as 'dg-bogus + xfail' to silence the FAIL,
when nvptx offloading is compiled for. (If not, the
metadirective can be resolved early during compilation.)
libgomp/ChangeLog:
PR middle-end/118694
* testsuite/libgomp.fortran/metadirective-1.f90: xfail when
compiling (also) for nvptx offloading as an error is then expected.
Eric Botcazou [Tue, 27 May 2025 17:42:17 +0000 (19:42 +0200)]
Fix IPA-SRA issue with reverse SSO on specific pattern
IPA-SRA generally works fine in the presence of reverse Scalar_Storage_Order
by propagating the relevant flag onto the newly generated MEM_REFs. However
we have been recently faced with a specific Ada pattern that it does not
handle correctly: the 'Valid attribute applied to a floating-point component
of an aggregate type with reverse Scalar_Storage_Order.
The attribute is implemented by a call to a specific routine of the runtime
that expects a pointer to the object so, in the case of a component with
reverse SSO, the compiler first loads it from the aggregate to get back the
native storage order, but it does the load using an array of bytes instead
of the floating-point type to prevent the FPU from fiddling with the value,
which yields in the .original dump file:
Of course that's a bit convoluted, but it does not seem that another method
would be simpler or even work, and using VIEW_CONVERT_EXPR to toggle the SSO
is supposed to be supported in any case (unlike aliasing or type punning).
The attached patch makes it work. While the call to storage_order_barrier_p
from IPA-SRA is quite natural (the regular SRA has it too), the tweak to the
predicate itself is needed to handle the scalar->aggregate conversion, which
is admittedly awkward but again without clear alternative.
gcc/
* ipa-sra.cc (scan_expr_access): Also disqualify storage order
barriers from splitting.
* tree.h (storage_order_barrier_p): Also return false if the
operand of the VIEW_CONVERT_EXPR has reverse storage order.
gcc/testsuite/
* gnat.dg/sso19.adb: New test.
* gnat.dg/sso19_pkg.ads, gnat.dg/sso19_pkg.adb: New helper.
Jonathan Wakely [Wed, 21 May 2025 14:29:02 +0000 (15:29 +0100)]
libstdc++: Fix vector(from_range_t, R&&) for exceptions [PR120367]
Because this constructor delegates to vector(a) the object has been
fully constructed and the destructor will run if an exception happens.
That means we need to set _M_finish == _M_start so that the destructor
doesn't try to destroy any elements.
libstdc++-v3/ChangeLog:
PR libstdc++/120367
* include/bits/stl_vector.h (_M_range_initialize): Initialize
_M_impl._M_finish.
* testsuite/23_containers/vector/cons/from_range.cc: Check with
a type that throws on construction.
exceptions during construction.
Tobias Burnus [Mon, 26 May 2025 17:50:40 +0000 (19:50 +0200)]
c-c++-common/gomp/{attrs-,}metadirective-3.c: Fix expected result [PR118694]
With compilation for nvptx enabled, two issues showed up:
(a) "error: 'target' construct with nested 'teams' construct contains
directives outside of the 'teams' construct"
See PR comment 9 why this is difficult to fix.
Solution: Add dg-bogus and accept/expect the error for 'target offload_nvptx'.
(b) The assumptions about the dump for 'target offload_nvptx' were wrong
as the metadirective was already expanded to a OMP_NEXT_VARIANT
construct such that no 'omp metadirective' was left in either case.
Solution: Check that no 'omp metadirective' is left; additionally, expect
either OMP_NEXT_VARIANT (when offload_nvptx is available) or no 'teams'
directive at all (if not).
gcc/testsuite/ChangeLog:
PR middle-end/118694
* c-c++-common/gomp/attrs-metadirective-3.c: Change to never
expect 'omp metadirective' in the dump. If !offload_nvptx, check
that no 'teams' shows up in the dump; for offload_nvptx, expect
OMP_NEXT_VARIANT and an error about directive between 'target'
and 'teams'.
* c-c++-common/gomp/metadirective-3.c: Likewise.
Tobias Burnus [Fri, 23 May 2025 09:30:48 +0000 (11:30 +0200)]
libgomp.c-c++-common/metadirective-1.c: Expect 'error:' for nvptx compile [PR118694]
OpenMP's 'target teams' is strictly coupled with 'teams'; if the latter
exists, the kernel is launched in directly with multiple teams. Thus,
the host has to know whether the teams construct exists or not. For
#pragma omp target
#pragma omp metadirective when (device={arch("nvptx")}: teams loop)
it is simple when 'nvptx' offloading is not supported, otherwise it depends
on the default device at runtime as the user code asks for a single team for
host fallback and gcn offload and multiple for nvptx offload.
In any case, this commit ensures that no FAIL is printed, whatever a
future solution might look like. Instead of a dg-bogus combined with an
'xfail offload_target_nvptx', one an also argue that a dg-error for
'target offload_target_nvptx' would be more appropriate.
libgomp/ChangeLog:
PR middle-end/118694
* testsuite/libgomp.c-c++-common/metadirective-1.c: xfail when
compiling (also) for nvptx offloading as an error is then expected.
Nathaniel Shead [Thu, 22 May 2025 12:16:22 +0000 (22:16 +1000)]
c++/modules: Fix merge of TLS init functions [PR120363]
The PR notes that we missed setting DECL_CONTEXT on the TLS init
function; we missed this initially because this function is not created
in header units, only named modules.
I also noticed that 'DECL_CONTEXT (fn) = DECL_CONTEXT (var)' was
incorrect: for class members, this ends up having the modules merging
machinery treat the decl as a member function, which breaks when
attempting to dedup against an existing completed class type. Instead
we can just use the global_namespace as the context, because the name of
the function is already mangled appropriately so that we'll match the
correct duplicates.
PR c++/120363
gcc/cp/ChangeLog:
* decl2.cc (get_tls_init_fn): Set context as global_namespace.
(get_tls_wrapper_fn): Likewise.
gcc/testsuite/ChangeLog:
* g++.dg/modules/pr113292_a.H: Move to...
* g++.dg/modules/tls-1_a.H: ...here.
* g++.dg/modules/pr113292_b.C: Move to...
* g++.dg/modules/tls-1_b.C: ...here.
* g++.dg/modules/pr113292_c.C: Move to...
* g++.dg/modules/tls-1_c.C: ...here.
* g++.dg/modules/tls-2_a.C: New test.
* g++.dg/modules/tls-2_b.C: New test.
* g++.dg/modules/tls-2_c.C: New test.
* g++.dg/modules/tls-3.h: New test.
* g++.dg/modules/tls-3_a.H: New test.
* g++.dg/modules/tls-3_b.C: New test.
Signed-off-by: Nathaniel Shead <nathanieloshead@gmail.com> Reviewed-by: Jason Merrill <jason@redhat.com>
(cherry picked from commit 66e9a4f3083356b064cc64651edad466a56f762b)
Nathaniel Shead [Fri, 23 May 2025 14:51:49 +0000 (00:51 +1000)]
c++/modules: Fix stream-in of member using-decls [PR120414]
When streaming in a reference to a data member, we have an oversight
where we did not consider USING_DECLs, despite otherwise handling them
here the same as fields. This patch corrects that mistake.
PR c++/120414
gcc/cp/ChangeLog:
* module.cc (trees_in::tree_node): Allow reading a USING_DECL
when streaming tt_data_member.
gcc/testsuite/ChangeLog:
* g++.dg/modules/using-31_a.C: New test.
* g++.dg/modules/using-31_b.C: New test.
Signed-off-by: Nathaniel Shead <nathanieloshead@gmail.com> Reviewed-by: Jason Merrill <jason@redhat.com>
(cherry picked from commit 43dddeef7a870ce4db7407f73660504b67a0a919)
Harald Anlauf [Thu, 15 May 2025 19:07:07 +0000 (21:07 +0200)]
Fortran: default-initialization and functions returning derived type [PR85750]
Functions with non-pointer, non-allocatable result and of derived type did
not always get initialized although the type had default-initialization,
and a derived type component had the allocatable or pointer attribute.
Rearrange the logic when to apply default-initialization.
PR fortran/85750
gcc/fortran/ChangeLog:
* resolve.cc (resolve_symbol): Reorder conditions when to apply
default-initializers.
/* Handle empty records as per the x86-64 psABI. */
TYPE_EMPTY_P (type) = targetm.calls.empty_record_p (type);
(Indeed x86_64 is still the only target to define 'TARGET_EMPTY_RECORD_P',
calling 'gcc/tree.cc-default_is_empty_record'.)
And so it happens that for an empty struct used in code offloaded from x86_64
host (but not powerpc64le host, for example), we get to see 'TYPE_EMPTY_P' in
offloading compilation (where the offload targets (currently?) don't use it
themselves, and therefore aren't prepared to handle it).
For nvptx offloading compilation, this causes wrong code generation:
'ptxas [...] error : Call has wrong number of parameters', as nvptx code
generation for function definition doesn't pay attention to this flag (say, in
'gcc/config/nvptx/nvptx.cc:pass_in_memory', or whereever else would be
appropriate to handle that), but the generic code 'gcc/calls.cc:expand_call'
via 'gcc/function.cc:aggregate_value_p' does pay attention to it, and we thus
get mismatching function definition vs. function call.
This issue apparently isn't a problem for GCN offloading, but I don't know if
that's by design or by accident.
Richard Biener:
> It looks like TYPE_EMPTY_P is only used during RTL expansion for ABI
> purposes, so computing it during layout_type is premature as shown here.
>
> I would suggest to simply re-compute it at offload stream-in time.
(For avoidance of doubt, the additions to 'gcc.target/nvptx/abi-struct-arg.c',
'gcc.target/nvptx/abi-struct-ret.c' are not dependent on the offload streaming
code changes, but are just to mirror the changes to
'libgomp.oacc-c-c++-common/abi-struct-1.c'.)
Tobias Burnus [Thu, 22 May 2025 15:06:38 +0000 (17:06 +0200)]
git_repository.py: Fix handling of --last-commit with merges
contrib/ChangeLog:
* gcc-changelog/git_update_version.py (update_current_branch): Fix
handling of merges.
* gcc-changelog/git_repository.py (parse_git_revisions): Skip over
merge commits for merges from exclude branch.
projects / thirdparty / gcc.git / log
