| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This one was forgotten. Also constants/externals are costed explicitly
with SLP.
* tree-vect-stmts.cc (vect_model_promotion_demotion_cost): Pass
in SLP node and drop unused dr argument. Use SLP node for
costing, drop costing of constant/external operands.
(vectorizable_conversion): Adjust.
|
|
It was applied on the stmt_info rather than the slp_node.
* tree-vect-stmts.cc (vectorizable_store): Apply SLP_TREE_VECTYPE
to slp_node rather than stmt_info.
|
|
|
|
COBOL often requires the conversion of binary integers to string of characters.
These changes replace a naive routine that peels decimal digits from a binary
value one digit at a time, with a divide-and-conquer algorithm that is twice as
fast even for a couple of digits, and is about eight times faster past ten
digits.
Included here are some minor fixes to the lexer and parser.
gcc/cobol/ChangeLog:
* cbldiag.h (location_dump): Source code formatting.
* parse.y: error_msg formatting.
* scan.l: Remove UTF-8 character from regex pattern.
* scan_ante.h (numstr_of): error_msg formatting.
* show_parse.h (class ANALYZE): Suppress cppcheck error.
* util.cc (cbl_field_t::report_invalid_initial_value):
error_msg formatting.
libgcobol/ChangeLog:
* Makefile.am: Include new stringbin.cc file.
* Makefile.in: Regenerated.
* libgcobol.cc (__gg__power_of_ten): Improve error message.
(__gg__binary_to_string): Deleted.
(__gg__binary_to_string_internal): Deleted.
(int128_to_field): Use new conversion routine.
(__gg__move): Use new conversion routine.
* stringbin.cc: New file. Implements new conversion routine.
* stringbin.h: New file. Likewise.
|
|
After r16-2519-gba5a6787374dea, we'll never express a C++20 rewritten
comparison operator as a built-in operator acting on an operator<=>
call, e.g. operator<=>(x, y) @ 0. This is because operator<=> always
returns a class type (std::foo_ordering), so the outer operator@ will
necessarily resolve to a non-built-in operator@ for that class type
(even in the non-dependent templated case, after that commit).
So the corresponding handling in extract_call_expr is basically dead
code, except for the TRUTH_NOT_EXPR case where we can plausibly still
have !(operator==(x, y)), but it doesn't make sense to recognize just
that one special case of operator rewriting. So let's just remove it
altogether; apparently it's no longer needed.
Also, the handling is imprecise: it recognizes expressions such as
0 < f() which never corresponded to a call in the first place. All
the more reason to remove it.
gcc/cp/ChangeLog:
* call.cc (extract_call_expr): Remove handling of C++20
rewritten comparison operators.
Reviewed-by: Jason Merrill <jason@redhat.com>
|
|
libstdc++-v3:
* doc/xml/manual/build_hacking.xml: Switch gcc.gnu.org installation
docs to https.
* doc/html/manual/appendix_porting.html: Regenerate.
|
|
[PR117783]
The following patch implements the C++26
P1061R10 - Structured Bindings can introduce a Pack
paper.
One thing unresolved in the patch is mangling, I've raised
https://github.com/itanium-cxx-abi/cxx-abi/issues/200
for that but no comments there yet. One question is if it is ok
not to mention the fact that there is a structured binding pack in
the mangling of the structured bindings but more important is in case
of std::tuple* we might need to mangle individual structured binding
pack elements separately (each might need an exported name for the
var itself and perhaps its guard variable as well). The patch just
uses the normal mangling for the whole structured bindings and emits
sorry if we need to mangle the structured binding pack elements.
The patch just marks the structured binding pack specially (considered
e.g. using some bit on it, but in the end I'm identifying it using
a made up type which causes DECL_PACK_P to be true; it is kind of
self-referential solution, because the type on the pack mentions the
DECL_DECOMPOSITION_P VAR_DECL on which the type is attached as its pack,
so it needs to be handled carefully during instantiation to avoid infinite
recursion, but it is the type that should be used if something else actually
needs to use the same type as the structured binding pack, e.g. a capture
proxy), and stores the pack elements when actually processed through
cp_finish_decomp with non-dependent initializer into a TREE_VEC used as
DECL_VALUE_EXPR of the pack; though because several spots use the
DECL_VALUE_EXPR and assume it is ARRAY_REF from which they can find out the
base variable and the index, it stores the base variable and index in the
first 2 TREE_VEC elts and has the structured binding elements only after
that.
https://eel.is/c++draft/temp.dep.expr#3.6 says the packs are type dependent
regardless of whether the initializer of the structured binding is type
dependent or not, so I hope having a dependent type on the structured
binding VAR_DECL is ok.
The paper also has an exception for sizeof... which is then not value
dependent when the structured bindings are initialized with non-dependent
initializer: https://eel.is/c++draft/temp.dep.constexpr#4
The patch special cases that in 3 spots (I've been wondering if e.g. during
parsing I couldn't just fold the sizeof... to the INTEGER_CST right away,
but guess I'd need to repeat that also during partial instantiation).
And one thing still unresolved is debug info, I've just added DECL_IGNORED_P
on the structured binding pack VAR_DECL because there were ICEs with -g
for now, hope it can be fixed incrementally but am not sure what exactly
we should emit in the debug info for that.
Speaking of which, I see
DW_TAG_GNU_template_parameter_pack
DW_TAG_GNU_formal_parameter_pack
etc. DIEs emitted regardless of DWARF version, shouldn't we try to upstream
those into DWARF 6 or check what other compilers emit for the packs?
And bet we'd need DW_TAG_GNU_structured_binding_pack as well.
2025-08-07 Jakub Jelinek <jakub@redhat.com>
PR c++/117783
gcc/c-family/
* c-cppbuiltin.cc (c_cpp_builtins): Change __cpp_structured_bindings
predefined value for C++26 from 202403L to 202411L.
gcc/cp/
* parser.cc: Implement C++26 P1061R10 - Structured Bindings can
introduce a Pack.
(cp_parser_range_for): Also handle TREE_VEC as DECL_VALUE_EXPR
instead of ARRAY_REF.
(cp_parser_decomposition_declaration): Use sb-identifier-list instead
of identifier-list in comments. Parse structured bindings with
structured binding pack. Don't emit pedwarn about structured
binding attributes in structured bindings inside of a condition.
(cp_convert_omp_range_for): Also handle TREE_VEC as DECL_VALUE_EXPR
instead of ARRAY_REF.
* decl.cc (get_tuple_element_type): Change i argument type from
unsigned to unsigned HOST_WIDE_INT.
(get_tuple_decomp_init): Likewise.
(set_sb_pack_name): New function.
(cp_finish_decomp): Handle structured binding packs.
* pt.cc (tsubst_pack_expansion): Handle structured binding packs
and capture proxies for them. Formatting fixes.
(tsubst_decl): For structured binding packs don't tsubst TREE_TYPE
first, instead recreate the type after r is created.
(tsubst_omp_for_iterator): Also handle TREE_VEC as DECL_VALUE_EXPR
instead of ARRAY_REF.
(tsubst_expr): Handle sizeof... on non-dependent structure binding
packs.
(value_dependent_expression_p): Return false for sizeof... on
non-dependent structure binding packs.
(instantiation_dependent_r): Don't recurse on sizeof... on
non-dependent structure binding packs.
* constexpr.cc (potential_constant_expression_1): Also handle
TREE_VEC on DECL_VALUE_EXPR of structure binding packs.
gcc/testsuite/
* g++.dg/cpp26/decomp13.C: New test.
* g++.dg/cpp26/decomp14.C: New test.
* g++.dg/cpp26/decomp15.C: New test.
* g++.dg/cpp26/decomp16.C: New test.
* g++.dg/cpp26/decomp17.C: New test.
* g++.dg/cpp26/decomp18.C: New test.
* g++.dg/cpp26/decomp19.C: New test.
* g++.dg/cpp26/decomp20.C: New test.
* g++.dg/cpp26/decomp21.C: New test.
* g++.dg/cpp26/feat-cxx26.C (__cpp_structured_bindings): Expect
202411 rather than 202403.
|
|
Unlike base PCS functions, __arm_streaming and __arm_streaming_compatible
functions allow/require PSTATE.SM to be 1 on entry, so they need to
be treated as STO_AARCH64_VARIANT_PCS.
Similarly, functions that share ZA or ZT0 with their callers require
ZA to be active on entry, whereas the base PCS requires ZA to be
dormant or off. These functions too need to be marked as having
a variant PCS.
gcc/
PR target/121414
* config/aarch64/aarch64.cc (aarch64_is_variant_pcs): New function,
split out from...
(aarch64_asm_output_variant_pcs): ...here. Handle various types
of SME function type.
gcc/testsuite/
PR target/121414
* gcc.target/aarch64/sme/pr121414_1.c: New test.
|
|
simplify_gen_subreg rejected subregs of literal constants if
MODE_COMPOSITE_P. This was added by the fix for PR96648 in
g:c0f772894b6b3cd8ed5c5dd09d0c7917f51cf70f. Jakub said:
As for the simplify_gen_subreg change, I think it would be desirable
to just avoid creating SUBREGs of constants on all targets and for all
constants, if simplify_immed_subreg simplified, fine, otherwise punt,
but as we are late in GCC11 development, the patch instead guards this
behavior on MODE_COMPOSITE_P (outermode) - i.e. only conversions to
powerpc{,64,64le} double double long double - and only for the cases where
simplify_immed_subreg was called.
I'm not sure about relaxing the codes further, since subregs might
be wanted for CONST, SYMBOL_REF and LABEL_REF. But removing the
MODE_COMPOSITE_P is needed to fix PR120718, where we get an ICE
from generating a subreg of a V2SI const_vector.
gcc/
PR rtl-optimization/120718
* simplify-rtx.cc (simplify_context::simplify_gen_subreg):
Remove MODE_COMPOSITE_P condition.
gcc/testsuite/
PR rtl-optimization/120718
* gcc.target/aarch64/sve/acle/general/pr120718.c: New test.
|
|
The following handles value-numbering of a BIT_FIELD_REF of
a register that's defined by a load by looking up a subset
load similar to how we handle bit-and masked loads. This
allows the testcase to be simplified by two FRE passes,
the first one will create the BIT_FIELD_REF.
PR tree-optimization/121405
* tree-ssa-sccvn.cc (visit_nary_op): Handle BIT_FIELD_REF
with reference def by looking up a combination of both.
* gcc.dg/tree-ssa/ssa-fre-107.c: New testcase.
* gcc.target/i386/pr90579.c: Adjust.
|
|
This patch extends the support for peeling and versioning for alignment
from VLS modes to VLA modes. The key change is allowing the DR target
alignment to be set to a non-constant poly_int. Since the value must be
a power-of-two, for variable VFs, the power-of-two check is deferred to
runtime through loop versioning. The vectorizable check for speculative
loads is also refactored in this patch to handle both constant and
variable target alignment values.
Additional changes for VLA modes include:
1) Peeling
In VLA modes, we use peeling with masking - using a partial vector in
the first iteration of the vectorized loop to ensure aligned DRs in
subsequent iterations. It was already enabled for VLS modes to avoid
scalar peeling. This patch reuses most of the existing logic and just
fixes a small issue of incorrect IV offset in VLA code path. This also
removes a power-of-two rounding when computing the number of iterations
to peel, as power-of-two VF has been guaranteed by a new runtime check.
2) Versioning
The type of the mask for runtime alignment check is updated to poly_int
to support variable VFs. After this change, both standalone versioning
and peeling with versioning are available in VLA modes. This patch also
introduces another runtime check for speculative read amount, to ensure
that all speculative loads remain within current valid memory page. We
plan to remove these runtime checks in the future by introducing capped
VF - using partial vectors to limit the actual VF value at runtime.
3) Speculative read flag
DRs whose scalar accesses are known to be in-bounds will be considered
unaligned unsupported with a variable target alignment. But in fact,
speculative reads can be naturally avoided for in-bounds DRs as long as
partial vectors are used. Therefore, this patch clears the speculative
flags and sets the "must use partial vectors" flag for these cases.
This patch is bootstrapped and regression-tested on x86_64-linux-gnu,
arm-linux-gnueabihf and aarch64-linux-gnu with bootstrap-O3.
gcc/ChangeLog:
* tree-vect-data-refs.cc (vect_compute_data_ref_alignment):
Allow DR target alignment to be a poly_int.
(vect_enhance_data_refs_alignment): Support peeling and
versioning for VLA modes.
* tree-vect-loop-manip.cc (get_misalign_in_elems): Remove
power-of-two rounding in peeling.
(vect_create_cond_for_align_checks): Update alignment check
logic for poly_int mask.
(vect_create_cond_for_vla_spec_read): New runtime checks.
(vect_loop_versioning): Support new runtime checks.
* tree-vect-loop.cc (_loop_vec_info::_loop_vec_info): Add a new
loop_vinfo field.
(vectorizable_induction): Fix wrong IV offset issue.
* tree-vect-stmts.cc (get_load_store_type): Refactor
vectorizable checks for speculative loads.
* tree-vectorizer.h (LOOP_VINFO_MAX_SPEC_READ_AMOUNT): New
macro for new runtime checks.
(LOOP_REQUIRES_VERSIONING_FOR_SPEC_READ): Likewise
(LOOP_REQUIRES_VERSIONING): Update macro for new runtime checks.
gcc/testsuite/ChangeLog:
* gcc.target/aarch64/sve/peel_ind_11.c: New test.
* gcc.target/aarch64/sve/peel_ind_11_run.c: New test.
* gcc.target/aarch64/sve/peel_ind_12.c: New test.
* gcc.target/aarch64/sve/peel_ind_12_run.c: New test.
* gcc.target/aarch64/sve/peel_ind_13.c: New test.
* gcc.target/aarch64/sve/peel_ind_13_run.c: New test.
|
|
ChangeLog:
* MAINTAINERS: Add myself.
|
|
Download newer versions of GMP, MPFR and MPC (the latest); besides the usual
bug fixes and smaller features, MPFR adds new functions for C23, some of
which are already used in GCC in the middle (fold-const-call.cc) and in
Fortran 2023 for the 'pi' trignonometric functions, if MPFR is new enough.
contrib/ChangeLog:
PR other/120237
* download_prerequisites: Update to download GMP 6.3.0 (before 6.2.1),
MPFR 4.2.2 (before 4.1.0), and MPC 1.3.1 (before 1.2.1).
* prerequisites.md5: Update hash.
* prerequisites.sha512: Likewise.
|
|
And here is the last part of the paper. Contrary what the paper claims
(clearly they haven't tried -pedantic nor -pedantic-errors), I think we
already diagnose everything we should.
2025-08-07 Jakub Jelinek <jakub@redhat.com>
PR preprocessor/120778
* g++.dg/DRs/dr2577-1.C: New test.
* g++.dg/DRs/dr2577-2.C: New test.
* g++.dg/DRs/dr2577-2.h: New file.
* g++.dg/DRs/dr2577-3.C: New test.
* g++.dg/DRs/dr2577-3.h: New file.
|
|
From the paper it isn't clear what caused the decision changes, not to drop
the "the token defined is generated as a result of this replacement process or"
part and make [cpp.cond]/10 violations IFNDR rather than ill-formed (the
latter maybe so that the extension to handle e.g. !A(A) below etc. can be
accepted).
Anyway, because that case hasn't been dropped and we pedwarn on it already,
and diagnose everything else the way it should, the following patch just
adds testcase for it.
2025-08-07 Jakub Jelinek <jakub@redhat.com>
PR preprocessor/120778
* g++.dg/DRs/dr2575.C: New test.
|
|
Another part of the C++26 P2843R3, this time in https://eel.is/c++draft/cpp.include#7
changing former (compile time) undefined behavior to IFNDR.
With IFNDR we wouldn't have to test anything I guess, but this still adds
the cases from the 3.4 section of the paper plus related tests.
The paper wonders about implementation divergence between most compilers and
gcc in the case of glueing tokens together with <> around from multiple
macros, GCC doesn't add CPP_PADDING macros in that case (and even if it
would, it would ignore them through get_token_no_padding).
But I think this behavior is allowed by
https://eel.is/c++draft/cpp.include#7.sentence-3
and changing it now after 25+ years of such behavior might break real-world
stuff, especially making it really hard to construct the <> includes from
some name and adding < and > from some other macro around that. We handle
#define A <cstd def>
#include A
like clang++ and try to open 'cstd def' file. But not adding spaces for
all the padding means handling even stuff where nothing else can help,
while for cstd def coming from different macros one can use token pasting
to combine them together, e.g. < and following identifier or identifier
followed by . or . followed by identifier or identifier followed by > can't
be pasted together.
2025-08-07 Jakub Jelinek <jakub@redhat.com>
PR preprocessor/120778
* g++.dg/DRs/dr2576.C: New test.
|
|
[PR120778]
The following patch introduces a -Wkeyword-macro warning that clang has
since 2014 to implement part of C++26 P2843R3 Preprocessing is never undefined
paper.
The relevant change in the paper is moving [macro.names]/2 paragraph to
https://eel.is/c++draft/cpp.replace.general#9 :
"A translation unit shall not #define or #undef names lexically identical to
keywords, to the identifiers listed in Table 4, or to the attribute-tokens
described in [dcl.attr], except that the names likely and unlikely may be
defined as function-like macros."
Now, my understanding of the paper is that in [macro.names] and surrounding
sections the word shall bears different meaning from [cpp.replace.general],
where only the latter location implies ill-formed, diagnostic required.
The warning in clang when introduced diagnosed all #define/#undef directives
on keywords, but shortly after introduction has been changed not to
diagnose #undef at all (with "#undef a keyword is generally harmless but used
often in configuration scripts" message) and later on even the #define
part tweaked - not warn about say
#define inline
(or const, extern, static), or
#define keyword keyword
or
#define keyword __keyword
or
#define keyword __keyword__
Later on the warning has been moved to be only pedantic diagnostic unless
requested by users. Clearly some code in the wild does e.g.
#define private public
and similar games, or e.g. Linux kernel (sure, C) does
#define inline __inline__ __attribute__((__always_inline__))
etc.
Now, I believe at least with the current C++26 wording such exceptions
aren't allowed (unless it is changed to IFNDR). But given that this is just
pedantic stuff, the following patch makes the warning off by default for
C and C++ before C++26 and even for C++26 it enables it by default only
if -pedantic/-pedantic-errors (in that case it pedwarns, otherwise it
warns). And it diagnoses both #define and #undef without exceptions.
From what I can see, all the current NODE_WARN cases are macros starting
with __ with one exception (_Pragma). As the NODE_* flags seem to be a
limited resource, I chose to just use NODE_WARN as well and differentiate
on the node names (if they don't start with __ or _P, they are considered
to be -Wkeyword-macro registered ones, otherwise old NODE_WARN cases,
typically builtin macros or __STDC* macros).
2025-08-07 Jakub Jelinek <jakub@redhat.com>
PR preprocessor/120778
gcc/
* doc/invoke.texi (Wkeyword-macro): Document.
gcc/c-family/
* c.opt (Wkeyword-macro): New option.
* c.opt.urls: Regenerate.
* c-common.h (cxx_dialect): Comment formatting fix.
* c-opts.cc (c_common_post_options): Default to
-Wkeyword-macro for C++26 if pedantic.
gcc/c/
* c-decl.cc (c_init_decl_processing): Mark cpp nodes corresponding
to keywords as NODE_WARN if warn_keyword_macro.
gcc/cp/
* lex.cc (cxx_init): Mark cpp nodes corresponding
to keywords, identifiers with special meaning and standard
attribute identifiers as NODE_WARN if warn_keyword_macro.
gcc/testsuite/
* gcc.dg/Wkeyword-macro-1.c: New test.
* gcc.dg/Wkeyword-macro-2.c: New test.
* gcc.dg/Wkeyword-macro-3.c: New test.
* gcc.dg/Wkeyword-macro-4.c: New test.
* gcc.dg/Wkeyword-macro-5.c: New test.
* gcc.dg/Wkeyword-macro-6.c: New test.
* gcc.dg/Wkeyword-macro-7.c: New test.
* gcc.dg/Wkeyword-macro-8.c: New test.
* gcc.dg/Wkeyword-macro-9.c: New test.
* g++.dg/warn/Wkeyword-macro-1.C: New test.
* g++.dg/warn/Wkeyword-macro-2.C: New test.
* g++.dg/warn/Wkeyword-macro-3.C: New test.
* g++.dg/warn/Wkeyword-macro-4.C: New test.
* g++.dg/warn/Wkeyword-macro-5.C: New test.
* g++.dg/warn/Wkeyword-macro-6.C: New test.
* g++.dg/warn/Wkeyword-macro-7.C: New test.
* g++.dg/warn/Wkeyword-macro-8.C: New test.
* g++.dg/warn/Wkeyword-macro-9.C: New test.
* g++.dg/warn/Wkeyword-macro-10.C: New test.
* g++.dg/opt/pr82577.C: Don't #define register to nothing for
C++17 and later. Instead define reg macro to nothing for C++17
and later or to register and use it instead of register.
* g++.dg/modules/atom-preamble-3.C: Add -Wno-keyword-macro to
dg-additional-options.
* g++.dg/template/sfinae17.C (static_assert): Rename macro to ...
(my_static_assert): ... this.
(main): Use my_static_assert instead of static_assert.
libcpp/
* include/cpplib.h (struct cpp_options): Add cpp_warn_keyword_macro.
(enum cpp_warning_reason): Add CPP_W_KEYWORD_MACRO enumerator.
(cpp_keyword_p): New inline function.
* directives.cc (do_undef): Support -Wkeyword-macro diagnostics.
* macro.cc (warn_of_redefinition): Ignore NODE_WARN flag on nodes
registered for -Wkeyword-macro.
(_cpp_create_definition): Support -Wkeyword-macro diagnostics.
Formatting fixes.
|
|
gcc/ChangeLog:
* config/s390/s390.cc (print_operand): Allow arbitrary wide_int
constants for _BitInt.
(s390_bitint_type_info): Implement target hook
TARGET_C_BITINT_TYPE_INFO.
libgcc/ChangeLog:
* config/s390/libgcc-glibc.ver: Export _BitInt support
functions.
* config/s390/t-softfp (softfp_extras): Add fixtfbitint
floatbitinttf.
gcc/testsuite/ChangeLog:
* gcc.target/s390/bitint-1.c: New test.
* gcc.target/s390/bitint-2.c: New test.
* gcc.target/s390/bitint-3.c: New test.
* gcc.target/s390/bitint-4.c: New test.
|
|
Enable soft-fp for -m64 only.
libgcc/ChangeLog:
* config.host: Include makefiles t-softfp for -m64.
* config/s390/sfp-exceptions.c: New file.
* config/s390/sfp-machine.h: New file.
* config/s390/t-softfp: New file.
|
|
|
|
This fails for me as:
+ERROR: g++.dg/cpp26/constexpr-new3.C -std=c++26: unknown dg option: 2 for "
dg-error 40 "accessing value of 'int [2]' object through a 'long int' glvalue in a constant expression" "
otherwise.
Escape '[' and ']'.
gcc/testsuite/ChangeLog:
* g++.dg/cpp26/constexpr-new3.C: Escape '[' and ']'.
|
|
The following splitter from the commit r11-5747:
(define_split
[(set (match_operand:SWI 0 "register_operand")
(any_rotate:SWI
(match_operand:SWI 1 "const_int_operand")
(subreg:QI
(and
(match_operand 2 "int248_register_operand")
(match_operand 3 "const_int_operand")) 0)))]
"(INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode) - 1))
== GET_MODE_BITSIZE (<MODE>mode) - 1"
[(set (match_dup 4) (match_dup 1))
(set (match_dup 0)
(any_rotate:SWI (match_dup 4)
(subreg:QI
(and:SI (match_dup 2) (match_dup 3)) 0)))]
"operands[4] = gen_reg_rtx (<MODE>mode);")
matches any mode of (and ...) on input, but hard-codes (and:SI ...)
in the output. This causes an ICE if the incoming (and ...) is DImode
rather than SImode.
Co-developed-by: Richard Sandiford <richard.sandiford@arm.com>
PR target/96226
gcc/ChangeLog:
* config/i386/predicates.md (and_operator): New operator.
* config/i386/i386.md (splitter after *<rotate_insn><mode>3_mask):
Use and_operator to match AND RTX and use its mode
in the split pattern.
|
|
hardbools didn't behave quite like bools when incremented,
decremented, or otherwise modified using their previous value, as in
+= et al. Fix that.
Also fix some checking errors that come up when using qualified base
types.
for gcc/c-family/ChangeLog
* c-attribs.cc (handle_hardbool_attribute): Create distinct
enumeration types, with structural equality. Handle
base type qualifiers.
for gcc/c/ChangeLog
* c-tree.h (C_BOOLEAN_TYPE_P): Cover hardbools as well.
* c-typeck.cc (convert_lvalue_to_rvalue): New overload and
wrapper.
(build_atomic_assign, build_modify_expr): Use it.
(build_asm_expr, handle_omp-array_sections_1): Simplify with
it.
(build_unary_op): Handle hardbools.
for gcc/testsuite/ChangeLog
* gcc.dg/torture/hardbool-ai.c: New.
* gcc.dg/torture/hardbool-vi.c: New.
* gcc.dg/torture/hardbool.c: Handle NO_BITFIELDS.
(add1, preinc, postinc, sub1, predec, postdec): New.
(main): Exercise them.
|
|
In addition to gzip format our snapshots and releases have been using
xz instead of bzip2 for a while.
gcc:
PR target/69374
* doc/install.texi (Prerequisites): Replace bzip2 by xz.
|
|
This patch adds the missing PTA_POPCNT and PTA_LZCNT with the PTA_ABM
bitmask definition for the bdver1, btver1, and lujiazui architectures
in the i386 architecture configuration file.
Although these two features were not present in the original definition,
their absence does not affect the functionality of these architectures
because the POPCNT and LZCNT bits are set when ABM is enabled in the
ix86_option_override_internal function. However, including them in these
definitions improves consistency and clarity. This issue was discovered
while writing a script to extract these bitmasks from the i386.h file
referenced in [1].
Additionally, the PTA_YONGFENG bitmask appears incorrect as it includes
PTA_LZCNT while already inheriting PTA_ABM from PTA_LUJIAZUI. This seems
to be a typo and should be corrected.
[1] https://github.com/cyyself/x86-pta
gcc/ChangeLog:
* config/i386/i386.h (PTA_BDVER1):
Add missing PTA_POPCNT and PTA_LZCNT with PTA_ABM.
(PTA_ZNVER1): Ditto.
(PTA_BTVER1): Ditto.
(PTA_LUJIAZUI): Ditto.
(PTA_YONGFENG): Do not include extra PTA_LZCNT.
Signed-off-by: Yangyu Chen <cyy@cyyself.name>
|
|
Since r15-4719-ga9ec1bc06bd3cc, we require GCC 5.4, so a workaround for
< GCC 4.3 is long-obsolete. Before that, we required GCC 4.8.3 too.
i.e. it shouldn't be possible to build GCC with such a compiler for
quite some time.
gcc/ChangeLog:
PR libstdc++/29286
* Makefile.in (ALIASING_FLAGS): Drop.
* configure: Regenerate.
* configure.ac: Drop -fno-strict-aliasing workaround for < GCC 4.3.
gcc/d/ChangeLog:
* Make-lang.in (ALL_DFLAGS): Don't use ALIASING_FLAGS.
|
|
operator [PR108931]
For size expressions in the branches of the conditional operator should
be evaluated only for the active branch. To achieve this, construct
also the size expressions for the composite type as conditional
expressions depending on the condition of the conditional operator.
The change avoids some undefined behavior removed by N3652, but the
new constraints for C2Y are not yet implemented.
PR c/108931
gcc/c/ChangeLog:
* c-typeck.cc (composite_type_cond): Renamed from
composite_type with argument for condition
(composite_type): New function.
(composite_type_internal): Implement new logic.
(build_conditional_expr): Pass condition.
(common_pointer_type): Adapt.
(pointer_diff): Adapt.
(build_binary_op): Adapt.
gcc/testsuite/ChangeLog:
* gcc.dg/vla-tert-1.c: New test.
|
|
When the doubly-linked list tests were introduced, the tests were
printing the content of the list forward and backward. However, this
printing is not needed outside of debugging, and confuses people because
the output is not only composed of PASS: lines.
This patch disables the printing of the list content by default. If
one wants to re-enable it for debugging, he can set the macro DUMP_LIST
to 1.
libiberty/ChangeLog:
* testsuite/test-doubly-linked-list.c: disable debug logging on
stdout.
|
|
The following simplifies vect_supportable_dr_alignment based on
the observation that we always dereference dr_info so it cannot
be NULL, likewise we always dereference dr so that cannot be NULL
either.
* tree-vect-data-refs.cc (vect_supportable_dr_alignment):
Prune dead code.
|
|
Here the results of A::make(0, 0, 1), (0, 1, 0) and (1, 0, 0) are each
represented as a single-element CONSTRUCTOR with CONSTRUCTOR_NO_CLEARING
cleared, and when used as as a class NTTP argument we end up mangling them
all as A{1}, i.e. eliding both the implicit initial and trailing zeros.
Mangling them all the same seems clearly wrong since they're logically
different values.
It turns out the mangling also omits intermediate zeros, e.g.
A::make(1, 0, 1) is mangled as A{1, 1}, the same as A::make(1, 1, 0).
It seems we can't omit both trailing and non-trailing implicit zeros
without introducing mangling ambiguities. This patch makes us include
non-trailing zeros in these manglings (while continuing to omit trailing
zeros).
This also manifests as a wrong-code bug where the specializations table
would conflate different specializations that are in terms of different
class NTTP arguments, since the identity of such arguments is tied to
their mangling.
PR c++/121231
PR c++/119688
PR c++/94511
gcc/ChangeLog:
* common.opt: Document additional ABI version 21 change.
* doc/invoke.texi: Likewise.
gcc/cp/ChangeLog:
* mangle.cc (write_expression): Write out implicit non-trailing
zeroes of a CONSTRUCTOR when the ABI version is at least 21.
gcc/testsuite/ChangeLog:
* g++.dg/abi/mangle82.C: New test.
* g++.dg/cpp2a/nontype-class73.C: New test.
Reviewed-by: Jason Merrill <jason@redhat.com>
|
|
This diagnostic failed to specify the actual type of the object being
accessed through a glvalue of an incompatible type, and could also use to
indicate where that object comes from.
gcc/cp/ChangeLog:
* constexpr.cc (cxx_eval_indirect_ref): Improve diagnostic.
gcc/testsuite/ChangeLog:
* g++.dg/cpp26/constexpr-new3.C: Tweak diagnostic.
|
|
The main gather/scatter discovery happens at SLP discovery time,
the base address and the offset scale are currently not explicitly
represented in the SLP tree. This requires re-discovery of them
during vectorizable_store/load. The following fixes this by
recording this info into the SLP tree. This allows the main
vect_check_gather_scatter call to be elided from get_load_store_type
and replaced with target support checks for IFN/decl or fallback
emulated mode.
There's vect_check_gather_scatter left in the path using gather/scatter
for strided load/store. I hope to deal with this later.
* tree-vectorizer.h (_slp_tree::gs_scale): New.
(_slp_tree::gs_base): Likewise.
(SLP_TREE_GS_SCALE): Likewise.
(SLP_TREE_GS_BASE): Likewise.
(vect_describe_gather_scatter_call): Declare.
* tree-vect-slp.cc (_slp_tree::_slp_tree): Initialize
new members.
(vect_build_slp_tree_2): Record gather/scatter base and scale.
(vect_get_and_check_slp_defs): For gather/scatter IFNs
describe the call to first_gs_info.
* tree-vect-data-refs.cc (vect_gather_scatter_fn_p): Add
mode of operation with fixed offset vector type.
(vect_describe_gather_scatter_call): Export.
* tree-vect-stmts.cc (get_load_store_type): Do not call
vect_check_gather_scatter to fill gs_info, instead populate
from the SLP tree. Check which of, IFN, decl or fallback
is supported and record that decision.
|
|
Both vectorizable_load and vectorizable_store compute an alias
type commoning differences between SLP lanes. The following
makes sure to use that for gather/scatter rather than the
random one from the representative created during scalar stmt
gather/scatter analysis.
* tree-vect-stmts.cc (vectorizable_store): Build proper
alias + align pointer value for gather/scatter and SLP
and use it.
(vectorizable_load): Likewise.
|
|
The following removes the redundant check on supported gather/scatter
IFN in check_load_store_for_partial_vectors which is already done
and adjusts those to check the recorded ifn, also allowing legacy
gather/scatter which all handle masking.
* tree-vect-stmts.cc (check_load_store_for_partial_vectors):
Remove redundant gather/scatter target support check, instead
check the recorded ifns. Also allow legacy gather/scatter
with loop masking.
* gcc.dg/vect/vect-gather-1.c: Adjust to hide N.
|
|
Sorry, my bootstrap failed last night because of this, I've fixed it
up and it bootstrapped/regtested fine overnight, but in the morning
forgot to adjust the patch before committing.
Without this there is
.../gimple-lower-bitint.cc:7678:36: error: comparison of integer expressions of different signedness: ‘unsigned int’ and ‘int’ [-Werror=sign-compare]
7678 | if (min_prec > limb_prec && abi_limb_prec > limb_prec)
| ~~~~~~~~~^~~~~~~~~~~
2025-08-06 Jakub Jelinek <jakub@redhat.com>
PR tree-optimization/121413
* gimple-lower-bitint.cc (gimple_lower_bitint): Fix up last
commit, cast limb_prec to unsigned before comparison.
|
|
The test was unsafe when tested on different vector lengths. This patch
fixes it to work on all lengths.
gcc/testsuite/ChangeLog
* gcc.target/aarch64/sve/acle/general/cops.c: Fix test.
|
|
A _BitInt value may rely on a conversion to become properly extended.
So a conversion to _BitInt is not trivially removable even if the
types of the result and the operand have the same precision and size.
This patch fixes gcc.dg/torture/bitint-64.c at -O2 on LoongArch,
which fails because extension of the result is dropped in a
compare-and-swap loop generated for incrementing an _Atomic _BitInt,
causing an ABI violation.
* match.pd: Preserve conversion to _BitInt before a VCE
if the _BitInt is extended.
* gcc.dg/torture/bitint-84.c: New test.
|
|
A widening cast from a signed _BitInt operand to an unsigned _BitInt
type involves filling the extra limb(s) with sign extension.
On a target that wants _BitInts extended in memory, if this unsigned
type has a partial limb, the unused part of it should be zeroed.
e.g. Assuming limb_mode == E_DImode, at the end of
void
test (unsigned _BitInt(519) *t, _BitInt(512) x)
{
*t = -x;
}
the most significant limb of *t should be masked with 0x7f.
This patch also fixes gcc.dg/torture/bitint-16.c, which aborts at -O2
when the extension on load is optimized away.
* gimple-lower-bitint.cc (bitint_large_huge::lower_mergeable_stmt):
Zero-extend the partial limb of any unsigned _BitInt LHS assigned
with a widening sign-extension.
* gcc.dg/torture/bitint-83.c: New test.
|
|
In BEXTC, whether a _BitInt object is properly extended is examined
by a value comparison against a copied object in a wider _BitInt
type that utilizes all of the partial limb.
Since the (implicit) conversion to the wider type may be optimized
away now and cause the result of the comparison to always be true,
we need to cast the copied object down to the original type to
force a extension, so that it can serve as our reference.
* gcc.dg/bitintext.h (BEXTC1): Define. Convert the copied
object back to the original type before comparison.
(BEXTC): Use BEXTC1 for both the signed and the unsigned case.
|
|
This patch also make casts of ABI-extended large/huge _BitInts
behave the same as the small/middle case, i.e. no-op for casts
to a higher precision _BitInt with the same number of limbs /
extension for casts that turns a full top limb into a partial limb.
This conveniently helps keep some code with implementation-specific
extension semantics (e.g. BEXTC from gcc.dg/bitintext.h) the same
for _BitInts of any precision.
* gimple-lower-bitint.cc (bitint_large_huge::limb_access):
Add a parameter abi_load_p. If set, load a limb directly
in its actual precision without casting from m_limb_type.
(struct bitint_large_huge): Same.
(bitint_large_huge::handle_load): Use.
|
|
aarch64 for some strange reason unconditionally enables -Werror for libgcc
building and this particular file for some strange reason contains
a useless static forward declaration of a function only defined in the
#if defined __sun__ && defined __svr4__
block and not otherwise (with __attribute__((constructor))).
And we warn (with -Werror error) in the non-__sun__/__svr4__ case because
it declares a static function that is never defined.
The forward declaration makes no sense to me, for static functions
forward declarations shouldn't be needed even for -Wstrict-prototypes,
and AFAIK we don't warn on static __attribute__((constructor)) void foo (void) {}
being unused. And the function isn't used before being defined.
So, the following patch just removes the forward declaration.
2025-08-06 Jakub Jelinek <jakub@redhat.com>
PR libgcc/121397
* enable-execute-stack-mprotect.c (check_enabling): Remove useless
forward declaration.
|
|
The following testcase is miscompiled on aarch64-linux.
The problem is in the optimization to shorten large constants
in PHI arguments.
In a couple of places during bitint lowering we compute
minimal precision of constant and if it is significantly
smaller than the precision of the type, store smaller constant
in memory and extend it at runtime (zero or all ones).
Now, in most places that works fine, we handle the stored number
of limbs by loading them from memory and then the rest is
extended. In the PHI INTEGER_CST argument handling we do
it differently, we don't form there any loops (because we
insert stmt sequences on the edges).
The problem is that we copy the whole _BitInt variable from
memory to the PHI VAR_DECL + initialize the rest to = {} or
memset to -1. It has
min_prec = CEIL (min_prec, limb_prec) * limb_prec;
precision, so e.g. on x86_64 there is no padding and it works
just fine. But on aarch64 which has abi_limb_mode TImode
and limb_mode DImode it doesn't in some cases.
In the testcase the constant has 408 bits min precision, rounded up
to limb_prec (64) is 448, i.e. 7 limbs. But aarch64 with TImode
abi_limb_mode will actually allocate 8 limbs and the most significant
limb is solely padding. As we want to extend the constant with all
ones, copying the padding (from memory, so 0s) will result in
64 0 bits where 1 bits were needed.
The following patch fixes it by detecting this case and setting
min_prec to a multiple of abi limb precision so that it has
no padding.
2025-08-06 Jakub Jelinek <jakub@redhat.com>
PR tree-optimization/121413
* gimple-lower-bitint.cc (abi_limb_prec): New variable
(bitint_precision_kind): Initialize it.
(gimple_lower_bitint): Clear it at the start. For
min_prec > limb_prec descreased precision vars for
INTEGER_CST PHI arguments ensure min_prec is either
prec or multiple of abi_limb_prec.
* gcc.dg/torture/bitint-85.c: New test.
|
|
handle_operand_addr (used for the cases where we use libgcc APIs, so
multiplication, division, modulo, casts of _BitInt to float/dfp) when it
sees default definition of an SSA_NAME which is not PARM_DECL (i.e.
uninitialized one) just allocates single uninitialized limb, there is no
need to waste more memory on it, it can just tell libgcc that it has
64-bit precision, not say 1024 bit etc.
Unfortunately, doing this runs into some asserts when we have a narrowing
cast of the uninitialized SSA_NAME (but still large/huge _BitInt).
The following patch fixes that by using a magic value in *prec_stored
for the uninitialized cases (0) and just don't do any *prec tweaks for
narrowing casts from that. precs still needs to be maintained as before,
that one is used for big endian adjustment.
2025-08-06 Jakub Jelinek <jakub@redhat.com>
PR tree-optimization/121127
* gimple-lower-bitint.cc (bitint_large_huge::handle_operand_addr): For
uninitialized SSA_NAME, set *prec_stored to 0 rather than *prec.
Handle that case in narrowing casts. If prec_stored is non-NULL,
set *prec_stored to prec_stored_val.
* gcc.dg/bitint-125.c: New test.
|
|
OpenBSD headers apparently instead of just
#define SIZE_MAX something
do
#ifndef SIZE_MAX
#define SIZE_MAX something
#endif
This causes problem with gengtype-lex.cc, where the flex generated
code has
#ifndef SIZE_MAX
#define SIZE_MAX (~(size_t)0)
#endif
and system.h is included only after that and since my changes for
host size_t *printf printing SIZE_MAX is used in preprocessor
expressions,
#if SIZE_MAX <= UINT_MAX
etc.
In the preprocessor, identifiers are replaced with 0 and so
it is (~(0)0) <= 0xffffffffU
or so and thus invalid.
Now, normally we want to include system.h early, ideally immediately
after config.h or bconfig.h, but in gengtype-lex.cc case we have
#ifdef HOST_GENERATOR_FILE
#include "config.h"
#define GENERATOR_FILE 1
#else
#include "bconfig.h"
#endif
// flex generated start of file, including
#include <stdio.h>
#include <string.h>
#include <errno.h>
#include <stdlib.h>
#include <inttypes.h>
#ifndef SIZE_MAX
#define SIZE_MAX (~(size_t)0)
#endif
// start of gengtype-lex.l %{} section
#ifdef HOST_GENERATOR_FILE
#include "config.h"
#define GENERATOR_FILE 1
#else
#include "bconfig.h"
#endif
#include "system.h"
#define malloc xmalloc
#define realloc xrealloc
#include "gengtype.h"
As I'm not sure what flex we require for building gcc (%top{} which COBOL FE
*.l uses is only in flex from 2003-04-01), the patch keeps using the %top{}
done by hand in Makefile.in, but includes system.h in the top part, with
FLEX_SCANNER temporarily defined (I'm undefining it afterwards because
flex generated code defines it again and I don't want to guarantee it is
defined to the same value) so that malloc/realloc poisoning doesn't happen
and #define malloc xmalloc and realloc xrealloc are done in system.h.
Note, system.h already includes all the 5 headers flex generated code
includes.
2025-08-06 Jakub Jelinek <jakub@redhat.com>
PR bootstrap/121386
* Makefile.in (gengtype-lex.cc): Append #define FLEX_SCANNER,
#include "system.h" and #undef FLEX_SCANNER to the prepended lines.
* gengtype-lex.l: Remove inclusion of config.h or bconfig.h, system.h
and definition of malloc/realloc from %{} section.
|
|
tests_2_driver and tests_4_driver are identical, and tests_4_driver is not used
at all. This patch clean this up, and format the driver with gcc style.
gcc/testsuite/ChangeLog:
* gfortran.dg/c_f_pointer_shape_tests_2.f03: Use the new driver.
* gfortran.dg/c_f_pointer_shape_tests_4.f03: Ditto.
* gfortran.dg/c_f_pointer_shape_tests_4_driver.c: Removed.
* gfortran.dg/c_f_pointer_shape_tests_2_driver.c: Renamed to ...
* gfortran.dg/c_f_pointer_shape_tests_driver.c: ... this; format
with gcc style.
|
|
P2843R3 dropped the
"If there is a form-feed or a vertical-tab character in such a comment, only
whitespace characters shall appear between it and the new-line that terminates
the comment; no diagnostic is required."
sentence from [lex.comment]. AFAIK we've never diagnosed nor checked for
that and C23 doesn't have anything like that, so the following testcase
merely tests that we don't diagnose anything on it.
2025-08-06 Jakub Jelinek <jakub@redhat.com>
PR preprocessor/120778
* c-c++-common/cpp/comment-ff-1.c: New test.
* c-c++-common/cpp/comment-vtab-1.c: New test.
|
|
Under some error condition we can end up with NULL_TREEs for
the type of bitfields, which can cause an ICE when testing for
type compatibility. Add the missing check.
PR c/121217
gcc/c/ChangeLog:
* c-typeck.cc (tagged_types_tu_compatible_p): Add check.
gcc/testsuite/ChangeLog:
* gcc.dg/pr121217.c: New test.
|
|
Previously, arch-canonicalize used hardcoded data to handle IMPLIED_EXT.
But this data often got out of sync with the actual C++ implementation.
Earlier, we introduced riscv-ext.def to keep track of all extension info
and generate docs. Now, arch-canonicalize also uses this same data to handle
extension implication rules directly.
One limitation is that conditional implication rules still need to be written
manually. Luckily, there aren't many of them for now, so it's still manageable.
I really wanted to avoid writing a C++ + Python binding or trying to parse C++
logic in Python...
This version also adds a `--selftest` option to run some unit tests.
gcc/ChangeLog:
* config/riscv/arch-canonicalize: Read extension data from
riscv-ext*.def and adding unittest.
|
|
This patch introduces a new `-march=unset` option for RISC-V GCC that
allows users to explicitly ignore previous `-march` options and derive
the architecture string from the `-mcpu` option instead.
This feature is particularly useful for build systems and toolchain
configurations where you want to ensure the architecture is always
derived from the CPU specification rather than relying on potentially
conflicting `-march` options.
gcc/ChangeLog:
* common/config/riscv/riscv-common.cc (riscv_expand_arch):
Ignore `unset`.
* config/riscv/riscv.h (OPTION_DEFAULT_SPECS): Handle
`-march=unset`.
(ARCH_UNSET_CLEANUP_SPECS): New.
(DRIVER_SELF_SPECS): Handle -march=unset.
* doc/invoke.texi (RISC-V Options): Update documentation for
`-march=unset`.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/arch-unset-1.c: New test.
* gcc.target/riscv/arch-unset-2.c: New test.
* gcc.target/riscv/arch-unset-3.c: New test.
* gcc.target/riscv/arch-unset-4.c: New test.
* gcc.target/riscv/arch-unset-5.c: New test.
|
|
This adds support for iterators in 'to' and 'from' clauses in the
'target update' OpenMP directive.
gcc/c/
* c-parser.cc (c_parser_omp_clause_from_to): Parse 'iterator' modifier.
* c-typeck.cc (c_finish_omp_clauses): Finish iterators for to/from
clauses.
gcc/cp/
* parser.cc (cp_parser_omp_clause_from_to): Parse 'iterator' modifier.
* semantics.cc (finish_omp_clauses): Finish iterators for to/from
clauses.
gcc/
* gimplify.cc (remove_unused_omp_iterator_vars): Display unused
variable warning for 'to' and 'from' clauses.
(gimplify_scan_omp_clauses): Add argument for iterator loop sequence.
Gimplify the clause decl and size into the iterator loop if iterators
are used.
(gimplify_omp_workshare): Add argument for iterator loops sequence
in call to gimplify_scan_omp_clauses.
(gimplify_omp_target_update): Call remove_unused_omp_iterator_vars and
build_omp_iterators_loops. Add loop sequence as argument when calling
gimplify_scan_omp_clauses, gimplify_adjust_omp_clauses and building
the Gimple statement.
* tree-pretty-print.cc (dump_omp_clause): Call dump_omp_iterators
for to/from clauses with iterators.
* tree.cc (omp_clause_num_ops): Add extra operand for OMP_CLAUSE_FROM
and OMP_CLAUSE_TO.
* tree.h (OMP_CLAUSE_HAS_ITERATORS): Add check for OMP_CLAUSE_TO and
OMP_CLAUSE_FROM.
(OMP_CLAUSE_ITERATORS): Likewise.
gcc/testsuite/
* c-c++-common/gomp/target-update-iterators-1.c: New.
* c-c++-common/gomp/target-update-iterators-2.c: New.
* c-c++-common/gomp/target-update-iterators-3.c: New.
libgomp/
* target.c (gomp_update): Call gomp_merge_iterator_maps. Free
allocated variables.
* testsuite/libgomp.c-c++-common/target-update-iterators-1.c: New.
* testsuite/libgomp.c-c++-common/target-update-iterators-2.c: New.
* testsuite/libgomp.c-c++-common/target-update-iterators-3.c: New.
|
