Jakub Jelinek [Tue, 18 Oct 2022 08:31:20 +0000 (10:31 +0200)]
middle-end IFN_ASSUME support [PR106654]
My earlier patches gimplify the simplest non-side-effects assumptions
into if (cond) ; else __builtin_unreachable (); and throw the rest
on the floor.
The following patch attempts to do something with the rest too.
For -O0, it throws the more complex assumptions on the floor,
we don't expect optimizations and the assumptions are there to allow
optimizations. Otherwise arranges for the assumptions to be
visible in the IL as
.ASSUME (_Z2f4i._assume.0, i_1(D));
call where there is an artificial function like:
bool _Z2f4i._assume.0 (int i)
{
bool _2;
<bb 2> [local count:
1073741824]:
_2 = i_1(D) == 43;
return _2;
}
with the semantics that there is UB unless the assumption function
would return true.
Aldy, could ranger handle this? If it sees .ASSUME call,
walk the body of such function from the edge(s) to exit with the
assumption that the function returns true, so above set _2 [true, true]
and from there derive that i_1(D) [43, 43] and then map the argument
in the assumption function to argument passed to IFN_ASSUME (note,
args there are shifted by 1)?
During gimplification it actually gimplifies it into
[[assume (D.2591)]]
{
{
i = i + 1;
D.2591 = i == 44;
}
}
which is a new GIMPLE_ASSUME statement wrapping a GIMPLE_BIND and
specifying a boolean_type_node variable which contains the result.
The GIMPLE_ASSUME then survives just a couple of passes and is lowered
during gimple lowering into an outlined separate function and
IFN_ASSUME call. Variables declared inside of the
condition (both static and automatic) just change context, automatic
variables from the caller are turned into parameters (note, as the code
is never executed, I handle this way even non-POD types, we don't need to
bother pretending there would be user copy constructors etc. involved).
The assume_function artificial functions are then optimized until the
new assumptions pass which doesn't do much right now but I'd like to see
there the backwards ranger walk and filling up of SSA_NAME_RANGE_INFO
for the parameters.
There are a few further changes I'd like to do, like ignoring the
.ASSUME calls in inlining size estimations (but haven't figured out where
it is done), or for LTO arrange for the assume functions to be emitted
in all partitions that reference those (usually there will be just one,
unless code with the assumption got inlined, versioned etc.).
2022-10-18 Jakub Jelinek <jakub@redhat.com>
PR c++/106654
gcc/
* gimple.def (GIMPLE_ASSUME): New statement kind.
* gimple.h (struct gimple_statement_assume): New type.
(is_a_helper <gimple_statement_assume *>::test,
is_a_helper <const gimple_statement_assume *>::test): New.
(gimple_build_assume): Declare.
(gimple_has_substatements): Return true for GIMPLE_ASSUME.
(gimple_assume_guard, gimple_assume_set_guard,
gimple_assume_guard_ptr, gimple_assume_body_ptr, gimple_assume_body):
New inline functions.
* gsstruct.def (GSS_ASSUME): New.
* gimple.cc (gimple_build_assume): New function.
(gimple_copy): Handle GIMPLE_ASSUME.
* gimple-pretty-print.cc (dump_gimple_assume): New function.
(pp_gimple_stmt_1): Handle GIMPLE_ASSUME.
* gimple-walk.cc (walk_gimple_op): Handle GIMPLE_ASSUME.
* omp-low.cc (WALK_SUBSTMTS): Likewise.
(lower_omp_1): Likewise.
* omp-oacc-kernels-decompose.cc (adjust_region_code_walk_stmt_fn):
Likewise.
* tree-cfg.cc (verify_gimple_stmt, verify_gimple_in_seq_2): Likewise.
* function.h (struct function): Add assume_function bitfield.
* gimplify.cc (gimplify_call_expr): If the assumption isn't
simple enough, expand it into GIMPLE_ASSUME wrapped block or
for -O0 drop it.
* gimple-low.cc: Include attribs.h.
(create_assumption_fn): New function.
(struct lower_assumption_data): New type.
(find_assumption_locals_r, assumption_copy_decl,
adjust_assumption_stmt_r, adjust_assumption_stmt_op,
lower_assumption): New functions.
(lower_stmt): Handle GIMPLE_ASSUME.
* tree-ssa-ccp.cc (pass_fold_builtins::execute): Remove
IFN_ASSUME calls.
* lto-streamer-out.cc (output_struct_function_base): Pack
assume_function bit.
* lto-streamer-in.cc (input_struct_function_base): And unpack it.
* cgraphunit.cc (cgraph_node::expand): Don't verify assume_function
has TREE_ASM_WRITTEN set and don't release its body.
(symbol_table::compile): Allow assume functions not to have released
body.
* internal-fn.cc (expand_ASSUME): Remove gcc_unreachable.
* passes.cc (execute_one_pass): For TODO_discard_function don't
release body of assume functions.
* cgraph.cc (cgraph_node::verify_node): Don't verify cgraph nodes
of PROP_assumptions_done functions.
* tree-pass.h (PROP_assumptions_done): Define.
(TODO_discard_function): Adjust comment.
(make_pass_assumptions): Declare.
* passes.def (pass_assumptions): Add.
* timevar.def (TV_TREE_ASSUMPTIONS): New.
* tree-inline.cc (remap_gimple_stmt): Handle GIMPLE_ASSUME.
* tree-vrp.cc (pass_data_assumptions): New variable.
(pass_assumptions): New class.
(make_pass_assumptions): New function.
gcc/cp/
* cp-tree.h (build_assume_call): Declare.
* parser.cc (cp_parser_omp_assumption_clauses): Use build_assume_call.
* cp-gimplify.cc (build_assume_call): New function.
(process_stmt_assume_attribute): Use build_assume_call.
* pt.cc (tsubst_copy_and_build): Likewise.
gcc/testsuite/
* g++.dg/cpp23/attr-assume5.C: New test.
* g++.dg/cpp23/attr-assume6.C: New test.
* g++.dg/cpp23/attr-assume7.C: New test.
Richard Biener [Tue, 18 Oct 2022 07:38:03 +0000 (09:38 +0200)]
tree-optimization/107301 - check if we can duplicate block before doing so
Path isolation failed to do that.
PR tree-optimization/107301
* gimple-ssa-isolate-paths.cc (handle_return_addr_local_phi_arg):
Check whether we can duplicate the block.
(find_implicit_erroneous_behavior): Likewise.
* gcc.dg/torture/pr107301.c: New testcase.
Liwei Xu [Mon, 17 Oct 2022 03:08:55 +0000 (11:08 +0800)]
Move scanning pass of forwprop-19.c to dse1 for r13-3212-gb88adba751da63
gcc/testsuite/ChangeLog:
PR testsuite/107220
* gcc.dg/tree-ssa/forwprop-19.c: Move scanning pass from
forwprop1 to dse1, This fixs the test case fail.
Andrew MacLeod [Mon, 17 Oct 2022 23:00:49 +0000 (19:00 -0400)]
Merge partial relation precisions properly
When merging 2 groups of PE's, one group was simply being set to the
other instead of properly merging them.
PR tree-optimization/107273
gcc/
* value-relation.cc (equiv_oracle::add_partial_equiv): Merge
instead of copying precison of each member.
gcc/testsuite/
* gcc.dg/tree-ssa/pr107273-1.c: New.
* gcc.dg/tree-ssa/pr107273-2.c: New.
GCC Administrator [Tue, 18 Oct 2022 00:17:40 +0000 (00:17 +0000)]
Daily bump.
Jeff Law [Mon, 17 Oct 2022 23:52:18 +0000 (19:52 -0400)]
Fix bogus RTL on the H8.
This patch actually fixes the bogus RTL seen in PR101697.
Basically we continue to use the insn condition to catch most of the problem
cases related to autoinc addressing modes. This patch adds constraints which
can guide reload (and hopefully LRA) away from doing blind replacements during
register elimination that would ultimately result in bogus RTL. The idea is
from Paul K. who has done something very similar on the pdp11. I guess it
shouldn't be a big surprise that the H8 and pdp11 need the same kind of
handling given some of the similarities in their architectures.
gcc/
PR target/101697
* config/h8300/combiner.md: Replace '<' preincment constraint with
ZA/Z1..ZH/Z7 combinations.
* config/h8300/movepush.md: Similarly
Jeff Law [Mon, 17 Oct 2022 23:42:27 +0000 (19:42 -0400)]
More infrastructure to avoid bogus RTL on H8.
Continuing the work to add constraints to avoid invalid RTL
with autoinc addressing modes. Specifically this patch adds
the memory constraints similar to the pdp11.
gcc/
* config/h8300/constraints.md (Za..Zh): New constraints for
autoinc addresses using a specific register.
* config/h8300/h8300.cc (pre_incdec_with_reg): New function.
* config/h8300/h8300-protos.h (pre_incdec_with_reg): Add prototype.
Jeff Law [Mon, 17 Oct 2022 23:33:52 +0000 (17:33 -0600)]
Remove accidential commits
gcc/
* config/i386/cet.c: Remove accidental commit.
* config/i386/driver-mingw32.c: Likewise.
* config/i386/i386-builtins.c: Likewise.
* config/i386/i386-d.c: Likewise.
* config/i386/i386-expand.c: Likewise.
* config/i386/i386-features.c: Likewise.
* config/i386/i386-options.c: Likewise.
* config/i386/t-cet: Likewise.
* config/i386/x86-tune-sched-atom.c: Likewise.
* config/i386/x86-tune-sched-bd.c: Likewise.
* config/i386/x86-tune-sched-core.c: Likewise.
* config/i386/x86-tune-sched.c: Likewise.
Jeff Law [Mon, 17 Oct 2022 23:28:00 +0000 (19:28 -0400)]
Enable REE for H8
I was looking at H8 assembly code recently and noticed we had unnecessary
extensions. As it turns out we never enabled redundant extension elimination
on the H8. This patch fixes that oversight (and was the trigger for the
failure fixed my the prior patch).
gcc/common
* common/config/h8300/h8300-common.cc (h8300_option_optimization_table):
Enable redundant extension elimination at -O2 and above.
Jeff Law [Mon, 17 Oct 2022 23:19:25 +0000 (19:19 -0400)]
Add missing splitter for H8
While testing a minor optimization on the H8 my builds failed due to
failure to split a zero-extended memory load. That particular pattern
is a bit special on the H8 in that it's split at assembly time primarily
to get the length computations correct. Arguably that alternative should
go away completely, but I haven't really looked into that.
Anyway, with the final-asm split we obviously need to match a define_split
somewhere. But none was ever written after adding CCZN optimizations. So
if we had a zero extend of a memory operand and it was used to eliminate
a compare, then we'd abort at final asm time.
Regression tested (in conjunction with various other in-progress patches) on
H8 without regressions.
gcc/
* config/h8300/extensions.md (CCZN setting zero extended load): Add
missing splitter.
H.J. Lu [Thu, 11 Aug 2022 23:21:23 +0000 (16:21 -0700)]
x86: Check corrupted return address when unwinding stack
If shadow stack is enabled, when unwinding stack, we count how many stack
frames we pop to reach the landing pad and adjust shadow stack by the same
amount. When counting the stack frame, we compare the return address on
normal stack against the return address on shadow stack. If they don't
match, return _URC_FATAL_PHASE2_ERROR for the corrupted return address on
normal stack. Don't check the return address for
1. Non-catchable exception where exception_class == 0. Process will be
terminated.
2. Zero return address which marks the outermost stack frame.
3. Signal stack frame since kernel puts a restore token on shadow stack.
* unwind-generic.h (_Unwind_Frames_Increment): Add the EXC
argument.
* unwind.inc (_Unwind_RaiseException_Phase2): Pass EXC to
_Unwind_Frames_Increment.
(_Unwind_ForcedUnwind_Phase2): Likewise.
* config/i386/shadow-stack-unwind.h (_Unwind_Frames_Increment):
Take the EXC argument. Return _URC_FATAL_PHASE2_ERROR if the
return address on normal stack doesn't match the return address
on shadow stack.
Steve Kargl [Mon, 17 Oct 2022 20:42:40 +0000 (22:42 +0200)]
Fortran: NULL pointer dereference in gfc_simplify_image_index [PR104330]
gcc/fortran/ChangeLog:
PR fortran/104330
* simplify.cc (gfc_simplify_image_index): Do not dereference NULL
pointer.
gcc/testsuite/ChangeLog:
PR fortran/104330
* gfortran.dg/pr104330.f90: New test.
Aldy Hernandez [Mon, 17 Oct 2022 16:56:24 +0000 (18:56 +0200)]
Make sure exported range for SSA post-dominates the DEF in set_global_ranges_from_unreachable_edges.
The problem here is that we're exporting a range for an SSA range that
happens on the other side of a __builtin_unreachable, but the SSA does
not post-dominate the definition point. This is causing ivcanon to
unroll things incorrectly.
This was a snafu when converting the code from evrp.
PR tree-optimization/107293
gcc/ChangeLog:
* tree-ssa-dom.cc
(dom_opt_dom_walker::set_global_ranges_from_unreachable_edges):
Check that condition post-dominates the definition point.
gcc/testsuite/ChangeLog:
* gcc.dg/tree-ssa/pr107293.c: New test.
Harald Anlauf [Sat, 15 Oct 2022 19:56:56 +0000 (21:56 +0200)]
Fortran: handle bad array ctors with typespec [PR93483, PR107216, PR107219]
gcc/fortran/ChangeLog:
PR fortran/93483
PR fortran/107216
PR fortran/107219
* arith.cc (reduce_unary): Handled expressions are EXP_CONSTANT and
EXPR_ARRAY. Do not attempt to reduce otherwise.
(reduce_binary_ac): Likewise.
(reduce_binary_ca): Likewise.
(reduce_binary_aa): Moved check for EXP_CONSTANT and EXPR_ARRAY
from here ...
(reduce_binary): ... to here.
(eval_intrinsic): Catch failed reductions.
* gfortran.h (GFC_INTRINSIC_OPS): New enum ARITH_NOT_REDUCED to keep
track of expressions that were not reduced by the arithmetic evaluation
code.
gcc/testsuite/ChangeLog:
PR fortran/93483
PR fortran/107216
PR fortran/107219
* gfortran.dg/array_constructor_56.f90: New test.
* gfortran.dg/array_constructor_57.f90: New test.
Co-authored-by: Mikael Morin <mikael@gcc.gnu.org>
Harald Anlauf [Sun, 16 Oct 2022 18:32:27 +0000 (20:32 +0200)]
Fortran: check type of operands of logical operations, comparisons [PR107272]
gcc/fortran/ChangeLog:
PR fortran/107272
* arith.cc (gfc_arith_not): Operand must be of type BT_LOGICAL.
(gfc_arith_and): Likewise.
(gfc_arith_or): Likewise.
(gfc_arith_eqv): Likewise.
(gfc_arith_neqv): Likewise.
(gfc_arith_eq): Compare consistency of types of operands.
(gfc_arith_ne): Likewise.
(gfc_arith_gt): Likewise.
(gfc_arith_ge): Likewise.
(gfc_arith_lt): Likewise.
(gfc_arith_le): Likewise.
gcc/testsuite/ChangeLog:
PR fortran/107272
* gfortran.dg/pr107272.f90: New test.
Tobias Burnus [Mon, 17 Oct 2022 16:15:16 +0000 (18:15 +0200)]
Fortran: Fixes for kind=4 characters strings [PR107266]
PR fortran/107266
gcc/fortran/
* trans-expr.cc (gfc_conv_string_parameter): Use passed
type to honor character kind.
* trans-types.cc (gfc_sym_type): Honor character kind.
* trans-decl.cc (gfc_conv_cfi_to_gfc): Fix handling kind=4
character strings.
gcc/testsuite/
* gfortran.dg/char4_decl.f90: New test.
* gfortran.dg/char4_decl-2.f90: New test.
Patrick Palka [Mon, 17 Oct 2022 15:14:36 +0000 (11:14 -0400)]
c++ modules: streaming constexpr_fundef [PR101449]
It looks like we currently avoid streaming the RESULT_DECL and PARM_DECLs
of a constexpr_fundef entry under the assumption that they're just copies
of the DECL_RESULT and DECL_ARGUMENTS of the FUNCTION_DECL. Thus we can
just make new copies of DECL_RESULT and DECL_ARGUMENTS on stream in rather
than separately streaming them.
But the FUNCTION_DECL's DECL_RESULT and DECL_ARGUMENTS eventually get
genericized, whereas the constexpr_fundef entry consists of a copy of the
FUNCTION_DECL's pre-GENERIC trees. And notably during genericization we
lower invisref parms (which entails changing their TREE_TYPE and setting
DECL_BY_REFERENCE), the lowered form of which the constexpr evaluator
doesn't expect to see, and so this copying approach causes us to ICE for
the below testcase.
This patch fixes this by faithfully streaming the RESULT_DECL and
PARM_DECLs of a constexpr_fundef entry, which seems to just work.
Nathan says[1]: Hm, the reason for the complexity was that I wanted to
recreate the tree graph where the fndecl came from one TU and the defn
came from another one -- we need the definition to refer to argument
decls from the already-read decl. However, it seems that for constexpr
fns here, that is not needed, resulting in a significant simplification.
[1]: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/603662.html
PR c++/101449
gcc/cp/ChangeLog:
* module.cc (trees_out::write_function_def): Stream the
parms and result of the constexpr_fundef entry.
(trees_in::read_function_def): Likewise.
gcc/testsuite/ChangeLog:
* g++.dg/modules/cexpr-3_a.C: New test.
* g++.dg/modules/cexpr-3_b.C: New test.
Aldy Hernandez [Mon, 17 Oct 2022 13:32:35 +0000 (15:32 +0200)]
[PR tree-optimization/105820] Add test.
PR tree-optimization/105820
gcc/testsuite/ChangeLog:
* g++.dg/tree-ssa/pr105820.c: New test.
Aldy Hernandez [Mon, 17 Oct 2022 13:26:05 +0000 (15:26 +0200)]
Do not test for -Inf when flag_finite_math_only.
PR tree-optimization/107286
gcc/ChangeLog:
* value-range.cc (range_tests_floats): Do not test for -Inf when
flag_finite_math_only.
Aldy Hernandez [Mon, 10 Oct 2022 09:01:48 +0000 (11:01 +0200)]
Add 3 floating NAN tests.
x UNORD x should set NAN on the TRUE side.
The false side of x == x should set NAN.
The true side of x != x should set NAN.
gcc/testsuite/
* gcc.dg/tree-ssa/vrp-float-3a.c: New.
* gcc.dg/tree-ssa/vrp-float-4a.c: New.
* gcc.dg/tree-ssa/vrp-float-5a.c: New.
Andrew MacLeod [Thu, 13 Oct 2022 22:03:58 +0000 (18:03 -0400)]
Add relation_trio class for range-ops.
There are 3 possible relations range-ops might care about, but only the one
most likely to be needed is supplied. This patch provides a new class
relation_trio which allows 3 relations to be passed in a single word.
fold_range (), op1_range (), and op2_range () are adjusted to take a
relation_trio class instead of a relation_kind, then the routine can
extract which relation it wants to work with.
* gimple-range-fold.cc (fold_using_range::range_of_range_op):
Provide relation_trio class.
* gimple-range-gori.cc (gori_compute::refine_using_relation):
Provide relation_trio class.
(gori_compute::refine_using_relation): Ditto.
(gori_compute::compute_operand1_range): Provide lhs_op2 and
op1_op2 relations via relation_trio class.
(gori_compute::compute_operand2_range): Ditto.
* gimple-range-op.cc (gimple_range_op_handler::calc_op1): Use
relation_trio instead of relation_kind.
(gimple_range_op_handler::calc_op2): Ditto.
(*::fold_range): Ditto.
* gimple-range-op.h (gimple_range_op::calc_op1): Adjust prototypes.
(gimple_range_op::calc_op2): Adjust prototypes.
* range-op-float.cc (*::fold_range): Use relation_trio instead of
relation_kind.
(*::op1_range): Ditto.
(*::op2_range): Ditto.
* range-op.cc (*::fold_range): Use relation_trio instead of
relation_kind.
(*::op1_range): Ditto.
(*::op2_range): Ditto.
* range-op.h (class range_operator): Adjust prototypes.
(class range_operator_float): Ditto.
(class range_op_handler): Adjust prototypes.
(relop_early_resolve): Pickup op1_op2 relation from relation_trio.
* value-relation.cc (VREL_LAST): Adjust use to be one past the end of
the enum.
(relation_oracle::validate_relation): Use relation_trio in call
to fold_range.
* value-relation.h (enum relation_kind_t): Add VREL_LAST as
final element.
(class relation_trio): New.
(TRIO_VARYING, TRIO_SHIFT, TRIO_MASK): New.
Andrew MacLeod [Fri, 14 Oct 2022 13:29:23 +0000 (09:29 -0400)]
Fix nan updating in range-ops.
Calling clean_nan on an undefined type traps, set_varying first. Other
tweaks for correctness.
* range-op-float.cc (foperator_not_equal::op1_range): Check for
VREL_EQ after singleton.
(foperator_unordered::op1_range): Set VARYING before calling
clear_nan().
(foperator_ordered::op1_range): Set rather than clear NAN if both
operands are the same.
Andrew MacLeod [Fri, 14 Oct 2022 18:31:02 +0000 (14:31 -0400)]
Don't set useless relations.
The oracle will not register nonssense/useless relations, class
value_relation shouldn't either.
* value-relation.cc (value_relation::dump): Change message.
* value-relation.h (value_relation::set_relation): If op1 is the
same as op2 do not create a relation.
Thomas Schwinge [Fri, 14 Oct 2022 22:10:29 +0000 (00:10 +0200)]
GCN: Restore build with GCC 4.8
For example, for "g++-4.8 (Ubuntu 4.8.4-2ubuntu1~14.04.4) 4.8.4", the recent
commit r13-3220-g45381d6f9f4e7b5c7b062f5ad8cc9788091c2d07
"amdgcn: add multiple vector sizes" broke the build:
In file included from [...]/source-gcc/gcc/coretypes.h:458:0,
from [...]/source-gcc/gcc/config/gcn/gcn.cc:24:
[...]/source-gcc/gcc/config/gcn/gcn.cc: In function ‘machine_mode VnMODE(int, machine_mode)’:
./insn-modes.h:42:71: error: temporary of non-literal type ‘scalar_int_mode’ in a constant expression
#define QImode (scalar_int_mode ((scalar_int_mode::from_int) E_QImode))
^
[...]/source-gcc/gcc/config/gcn/gcn.cc:405:10: note: in expansion of macro ‘QImode’
case QImode:
^
In file included from [...]/source-gcc/gcc/coretypes.h:478:0,
from [...]/source-gcc/gcc/config/gcn/gcn.cc:24:
[...]/source-gcc/gcc/machmode.h:410:7: note: ‘scalar_int_mode’ is not literal because:
class scalar_int_mode
^
[...]/source-gcc/gcc/machmode.h:410:7: note: ‘scalar_int_mode’ is not an aggregate, does not have a trivial default constructor, and has no constexpr constructor that is not a copy or move constructor
[...]
Addressing this like simiar issues have been addressed in the past.
gcc/
* config/gcn/gcn.cc (VnMODE): Use 'case E_QImode:' instead of
'case QImode:', etc.
Thomas Schwinge [Wed, 15 Dec 2021 21:00:53 +0000 (22:00 +0100)]
Tag 'gcc/gimple-expr.cc:mark_addressable_2' as 'static'
Added in 2015 r229696 (commit
1b223a9f3489296c625bdb7cc764196d04fd9231)
"defer mark_addressable calls during expand till the end of expand",
it has never been used 'extern'ally.
gcc/
* gimple-expr.cc (mark_addressable_2): Tag as 'static'.
Thomas Schwinge [Fri, 23 Sep 2022 09:29:50 +0000 (11:29 +0200)]
Fix nvptx-specific '-foffload-options' syntax in 'libgomp.c/reverse-offload-sm30.c'
That is, '-mptx=_' is only valid in '-foffload-options=nvptx-none', too.
Fix test case added in recent
commit r13-2625-g6b43f556f392a7165582aca36a19fe7389d995b2 "nvptx/mkoffload.cc:
Warn instead of error when reverse offload is not possible".
libgomp/
* testsuite/libgomp.c/reverse-offload-sm30.c: Fix nvptx-specific
'-foffload-options' syntax.
Richard Biener [Thu, 6 Oct 2022 11:56:09 +0000 (13:56 +0200)]
Vectorization of first-order recurrences
The following picks up the prototype by Ju-Zhe Zhong for vectorizing
first order recurrences. That solves two TSVC missed optimization PRs.
There's a new scalar cycle def kind, vect_first_order_recurrence
and it's handling of the backedge value vectorization is complicated
by the fact that the vectorized value isn't the PHI but instead
a (series of) permute(s) shifting in the recurring value from the
previous iteration. I've implemented this by creating both the
single vectorized PHI and the series of permutes when vectorizing
the scalar PHI but leave the backedge values in both unassigned.
The backedge values are (for the testcases) computed by a load
which is also the place after which the permutes are inserted.
That placement also restricts the cases we can handle (without
resorting to code motion).
I added both costing and SLP handling though SLP handling is
restricted to the case where a single vectorized PHI is enough.
Missing is epilogue handling - while prologue peeling would
be handled transparently by adjusting iv_phi_p the epilogue
case doesn't work with just inserting a scalar LC PHI since
that a) keeps the scalar load live and b) that loads is the
wrong one, it has to be the last, much like when we'd vectorize
the LC PHI as live operation. Unfortunately LIVE
compute/analysis happens too early before we decide on
peeling. When using fully masked loop vectorization the
vect-recurr-6.c works as expected though.
I have tested this on x86_64 for now, but since epilogue
handling is missing there's probably no practical cases.
My prototype WHILE_ULT AVX512 patch can handle vect-recurr-6.c
just fine but I didn't feel like running SPEC within SDE nor
is the WHILE_ULT patch complete enough.
PR tree-optimization/99409
PR tree-optimization/99394
* tree-vectorizer.h (vect_def_type::vect_first_order_recurrence): Add.
(stmt_vec_info_type::recurr_info_type): Likewise.
(vectorizable_recurr): New function.
* tree-vect-loop.cc (vect_phi_first_order_recurrence_p): New
function.
(vect_analyze_scalar_cycles_1): Look for first order
recurrences.
(vect_analyze_loop_operations): Handle them.
(vect_transform_loop): Likewise.
(vectorizable_recurr): New function.
(maybe_set_vectorized_backedge_value): Handle the backedge value
setting in the first order recurrence PHI and the permutes.
* tree-vect-stmts.cc (vect_analyze_stmt): Handle first order
recurrences.
(vect_transform_stmt): Likewise.
(vect_is_simple_use): Likewise.
(vect_is_simple_use): Likewise.
* tree-vect-slp.cc (vect_get_and_check_slp_defs): Likewise.
(vect_build_slp_tree_2): Likewise.
(vect_schedule_scc): Handle the backedge value setting in the
first order recurrence PHI and the permutes.
* gcc.dg/vect/vect-recurr-1.c: New testcase.
* gcc.dg/vect/vect-recurr-2.c: Likewise.
* gcc.dg/vect/vect-recurr-3.c: Likewise.
* gcc.dg/vect/vect-recurr-4.c: Likewise.
* gcc.dg/vect/vect-recurr-5.c: Likewise.
* gcc.dg/vect/vect-recurr-6.c: Likewise.
* gcc.dg/vect/tsvc/vect-tsvc-s252.c: Un-XFAIL.
* gcc.dg/vect/tsvc/vect-tsvc-s254.c: Likewise.
* gcc.dg/vect/tsvc/vect-tsvc-s291.c: Likewise.
Co-authored-by: Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
Florian Weimer [Mon, 17 Oct 2022 09:09:17 +0000 (11:09 +0200)]
libgcc: Move cfa_how into potential padding in struct frame_state_reg_info
On many architectures, there is a padding gap after the how array
member, and cfa_how can be moved there. This reduces the size of the
struct and the amount of memory that uw_frame_state_for has to clear.
There is no measurable performance benefit from this on x86-64 (even
though the memset goes from 120 to 112 bytes), but it seems to be a
good idea to do anyway.
libgcc/
* unwind-dw2.h (struct frame_state_reg_info): Move cfa_how member
and reduce its size.
Jonathan Wakely [Mon, 17 Oct 2022 08:38:02 +0000 (09:38 +0100)]
libstdc++: Fix value of __cpp_lib_constexpr_charconv
libstdc++-v3/ChangeLog:
* include/std/charconv (__cpp_lib_constexpr_charconv): Define to
correct value.
* include/std/version (__cpp_lib_constexpr_charconv): Likewise.
* testsuite/20_util/to_chars/constexpr.cc: Check correct value.
* testsuite/20_util/to_chars/version.cc: Likewise.
Ju-Zhe Zhong [Mon, 17 Oct 2022 07:30:47 +0000 (15:30 +0800)]
RISC-V: Fix format[NFC]
gcc/ChangeLog:
* config/riscv/t-riscv: Change Tab into 2 space.
Ju-Zhe Zhong [Fri, 14 Oct 2022 23:02:36 +0000 (07:02 +0800)]
RISC-V: Reorganize mangle_builtin_type.[NFC]
Hi, this patch fixed my mistake in the previous commit patch.
Since "mangle_builtin_type" is a global function will be called in riscv.cc.
It's reasonable move it down and put them together stay with other global functions.
gcc/ChangeLog:
* config/riscv/riscv-vector-builtins.cc (mangle_builtin_type): Move down the function.
Arsen Arsenovic [Mon, 17 Oct 2022 06:58:07 +0000 (08:58 +0200)]
elf: ELF toolchain --without-{headers, newlib} should provide stdint.h
stdint.h is considered a freestanding headers by C, and a valid stdint.h
is required for certain parts of libstdc++' configuration, so we should
simply provide one when we have no other way (i.e. newlib or
user-specified sysroot) of getting one.
* config.gcc: --target=*-elf --without-{newlib,headers} should
provide stdint.h.
Hu, Lin1 [Fri, 16 Sep 2022 03:25:13 +0000 (11:25 +0800)]
Initial Meteorlake Support
gcc/ChangeLog:
* common/config/i386/cpuinfo.h:
(get_intel_cpu): Handle Meteorlake.
* common/config/i386/i386-common.cc:
(processor_alias_table): Add Meteorlake.
Haochen Jiang [Fri, 16 Sep 2022 05:59:01 +0000 (13:59 +0800)]
Initial Raptorlake Support
gcc/ChangeLog:
* common/config/i386/cpuinfo.h:
(get_intel_cpu): Handle Raptorlake.
* common/config/i386/i386-common.cc:
(processor_alias_table): Add Raptorlake.
GCC Administrator [Mon, 17 Oct 2022 00:16:37 +0000 (00:16 +0000)]
Daily bump.
Jeff Law [Sun, 16 Oct 2022 16:43:25 +0000 (12:43 -0400)]
Add new constraints for upcoming autoinc fixes
GCC does not allow a the operand of an autoinc addressing mode to
overlap with another soure operand in the same insn. This is primarly
enforced with insn conditions. However, cases can slip through LRA
and reload. To address those scenarios we'll take an idea from the
pdp11 port for describing the restriction in constraints as well.
To implement that we need register classes and constraints which are
"all general purpose hardware registers except r0". And similarly for
r1..r7(sp).
This patch adds those register classes and constraints, but does not
yet use them.
gcc/
* config/h8300/constraints.md (Z0..Z7): New register
constraints.
* config/h8300/h8300.h (reg_class): Add new classes.
(REG_CLASS_NAMES): Similarly.
(REG_CLASS_CONTENTS): Similarly.
Jeff Law [Sun, 16 Oct 2022 14:58:52 +0000 (10:58 -0400)]
Rename "z" constraint to "Zz" on the H8/300
I want to use Z as a multi-letter constraint. So first we have to
adjust the existing use of Z. This does not affect code generation.
gcc/
* config/h8300/constraints.md (Zz constraint): Renamed
from "z".
* config/h8300/movepush.md (movqi_h8sx, movhi_h8sx): Adjust
constraint to use Zz instead of Z.
Jeff Law [Sun, 16 Oct 2022 03:38:20 +0000 (23:38 -0400)]
Fix bug in register move costing on H8/300
gcc/
* config/h8300/h8300.cc (h8300_register_move_cost): Fix typo.
GCC Administrator [Sun, 16 Oct 2022 00:16:25 +0000 (00:16 +0000)]
Daily bump.
Jonathan Wakely [Sat, 15 Oct 2022 20:15:51 +0000 (21:15 +0100)]
libstdc++: Fix -Wunused-function warning in src/c++11/debug.cc
The only remaining use of print_raw is conditionally compiled, so when
libstdc++ i built without debug backtrace support, there's an unused
warning function for it. Move it inside the conditional block.
libstdc++-v3/ChangeLog:
* src/c++11/debug.cc (print_raw): Move inside #if block.
Jonathan Wakely [Sat, 15 Oct 2022 20:20:47 +0000 (21:20 +0100)]
libstdc++: Implement constexpr std::to_chars for C++23 (P2291R3)
Some of the helper functions use static constexpr local variables, which
is not permitted in a core constant expression. Removing the 'static'
seems to have negligible performance effect for __to_chars and
__to_chars_16. For __from_chars_alnum_to_val removing the 'static'
causes a significant performance impact for base 36 conversions. Use a
consteval lambda instead.
libstdc++-v3/ChangeLog:
* include/bits/charconv.h (__to_chars_10_impl): Add constexpr
for C++23. Remove 'static' from array.
* include/std/charconv (__cpp_lib_constexpr_charconv): Define.
(__to_chars, __to_chars_16): Remove 'static' from array, add
constexpr.
(__to_chars_10, __to_chars_8, __to_chars_2, __to_chars_i)
(to_chars, __raise_and_add, __from_chars_pow2_base)
(__from_chars_alnum, from_chars): Add constexpr.
(__from_chars_alnum_to_val): Avoid local static during constant
evaluation. Add constexpr.
* include/std/version (__cpp_lib_constexpr_charconv): Define.
* testsuite/20_util/from_chars/constexpr.cc: New test.
* testsuite/20_util/to_chars/constexpr.cc: New test.
* testsuite/20_util/to_chars/version.cc: New test.
Jonathan Wakely [Fri, 14 Oct 2022 13:25:48 +0000 (14:25 +0100)]
libstdc++: Fix uses_allocator_construction args for cv pair (LWG 3677)
The _Std_pair concept uses in <bits/uses_allocator_args.h> handles const
qualified pairs, but not volatile qualified. That's because it just uses
__is_pair which is specialized for const pairs.
This removes the partial specialization __is_pair<const pair<T,U>>, so
that __is_pair is now only true for cv-unqualified pairs. Then _Std_pair
needs to explicitly use remove_cv_t for the argument to __is_pair.
The other use of __is_pair is in map::insert(Pair&&) which doesn't want
to handle volatile so should just use remove_const_t.
libstdc++-v3/ChangeLog:
* include/bits/stl_map.h (map::insert(Pair&&)): Use
remove_const_t on argument to __is_pair.
* include/bits/stl_pair.h (__is_pair<const pair<T,U>>): Remove
partial specialization.
* include/bits/uses_allocator_args.h (_Std_pair): Use
remove_cv_t as per LWG 3677.
* testsuite/20_util/uses_allocator/lwg3677.cc: New test.
GCC Administrator [Sat, 15 Oct 2022 00:17:38 +0000 (00:17 +0000)]
Daily bump.
Joseph Myers [Fri, 14 Oct 2022 23:07:50 +0000 (23:07 +0000)]
preprocessor: C2x identifier rules
C2x has, like C++, adopted rules for identifiers based directly on an
unversioned normative reference to Unicode. Make libcpp follow those
rules for c2x / gnu2x standards (this involves bringing back a flag
separate from the C++ one for whether to use these identifier rules,
but this time enabled for all C++ language versions since that was the
conclusion adopted for C++ identifier handling).
There is one change here that affects C++. I believe the new
normative requirement for NFC only applies to identifiers, not to the
use of identifier-continue characters in pp-numbers, where there is no
such requirement and so the diagnostic ought to be a warning not a
pedwarn in pp-numbers, and that this is the case for both C and C++.
Bootstrapped with no regressions for x86_64-pc-linux-gnu.
libcpp/
* charset.cc (ucn_valid_in_identifier): Check xid_identifiers not
cplusplus to determine whether to use CXX23 and NXX23 flags.
* include/cpplib.h (struct cpp_options): Add xid_identifiers.
* init.cc (struct lang_flags, lang_defaults): Add xid_identifiers.
(cpp_set_lang): Set xid_identifiers.
* lex.cc (warn_about_normalization): Add parameter identifier.
Only pedwarn about non-NFC for identifiers, not pp-numbers.
(_cpp_lex_direct): Update calls to warn_about_normalization.
gcc/testsuite/
* gcc.dg/cpp/c2x-ucnid-1-utf8.c, gcc.dg/cpp/c2x-ucnid-1.c: New
tests.
Harald Anlauf [Sun, 9 Oct 2022 18:43:32 +0000 (20:43 +0200)]
Fortran: fix check of polymorphic elements in data transfers [PR100971]
gcc/fortran/ChangeLog:
PR fortran/100971
* resolve.cc (resolve_transfer): Extend check for permissibility
of polymorphic elements in a data transfer to arrays.
gcc/testsuite/ChangeLog:
PR fortran/100971
* gfortran.dg/der_io_5.f90: New test.
Aldy Hernandez [Fri, 14 Oct 2022 14:49:33 +0000 (16:49 +0200)]
Implement distinction between HONOR_SIGNED_ZEROS and MODE_HAS_SIGNED_ZEROS.
gcc/ChangeLog:
* value-range.cc (frange::set): Implement distinction between
HONOR_SIGNED_ZEROS and MODE_HAS_SIGNED_ZEROS.
Aldy Hernandez [Thu, 13 Oct 2022 15:51:29 +0000 (17:51 +0200)]
Implement range-op entry for __builtin_copysign.
copysign(MAGNITUDE, SIGN) is implemented as the absolute of MAGNITUDE,
with SIGN applied. If the sign of "SIGN" cannot be determined, we
return a range of [-MAGNITUDE, +MAGNITUDE].
gcc/ChangeLog:
* gimple-range-op.cc (class cfn_copysign): New.
(gimple_range_op_handler::maybe_builtin_call): Add
CFN_BUILT_IN_COPYSIGN*.
Tobias Burnus [Fri, 14 Oct 2022 16:34:49 +0000 (18:34 +0200)]
gfortran.dg/c-interop/deferred-character-2.f90: Fix dg-do
gcc/testsuite/
* gfortran.dg/c-interop/deferred-character-2.f90: Use 'dg-do run'.
Aldy Hernandez [Fri, 14 Oct 2022 10:08:11 +0000 (12:08 +0200)]
Check rvc_normal in real_isdenormal.
[-Inf, -Inf] is being flushed to [-Inf, -0.0] because real_isdenormal
is being overly pessimistic. It is missing a check for rvc_normal.
This doesn't cause problems in real.cc because all uses of
real_isdenormal are already on the rvc_normal path. The uses in
value-range.cc however, are not.
This patch adds a check for rvc_normal.
gcc/ChangeLog:
* real.h (real_isdenormal): Check rvc_normal.
* value-range.cc (range_tests_floats): New test.
Jonathan Wakely [Wed, 12 Oct 2022 22:04:53 +0000 (23:04 +0100)]
libstdc++: Disable all emergency EH pool code if obj-count == 0
For a zero-sized static pool we can completely elide all code for the EH
pool.
We no longer need to adjust the static buffer size to ensure at least
one free_entry can be created in it, because we no longer use a static
buffer at all if obj_count == 0. If the buffer exists, obj_count >= 1
and the buffer will be much larger than sizeof(free_entry).
libstdc++-v3/ChangeLog:
* libsupc++/eh_alloc.cc [USE_POOL]: New macro.
[!USE_POOL] (__gnu_cxx::__freeres, pool): Do not define.
[_GLIBCXX_EH_POOL_STATIC] (pool::arena): Do not use std::max.
(__cxxabiv1::__cxa_allocate_exception) [!USE_POOL]: Do not use
pool.
(__cxxabiv1::__cxa_free_exception) [!USE_POOL]: Likewise.
(__cxxabiv1::__cxa_allocate_dependent_exception) [!USE_POOL]:
Likewise.
(__cxxabiv1::__cxa_free_dependent_exception) [!USE_POOL]:
Likewise.
Jonathan Wakely [Wed, 12 Oct 2022 10:59:33 +0000 (11:59 +0100)]
libstdc++: Simplify print_raw function for debug assertions
Replace two uses of print_raw where it's clearer to just use fprintf
directly. Then the only remaining use of print_raw is as the print_func
argument of pretty_print. When called by pretty_print the count is
either a positive integer or -1, so we can simplify print_raw itself.
Remove the default argument, because it's never used. Remove the check
for nbc == 0, which never happens (but would be harmless if it did).
Replace the conditional expression with a single call to fprintf, using
INT_MAX as the maximum length.
libstdc++-v3/ChangeLog:
* src/c++11/debug.cc (print_raw): Simplify.
(print_word): Print indentation by calling fprintf directly.
(_Error_formatter::_M_error): Print unindented string by calling
fprintf directly.
Aldy Hernandez [Fri, 14 Oct 2022 13:02:06 +0000 (15:02 +0200)]
Replace CFN_BUILTIN_SIGNBIT* cases with CASE_FLT_FN.
gcc/ChangeLog:
* gimple-range-op.cc
(gimple_range_op_handler::maybe_builtin_call): Replace
CFN_BUILTIN_SIGNBIT* cases with CASE_FLT_FN.
Aldy Hernandez [Fri, 14 Oct 2022 10:07:40 +0000 (12:07 +0200)]
Normalize ranges over the range for both bounds when -ffinite-math-only.
[-Inf, +Inf] was being chopped correctly for -ffinite-math-only, but
[-Inf, -Inf] was not. This was latent because a bug in
real_isdenormal is causing us to flush -Inf to zero.
gcc/ChangeLog:
* value-range.cc (frange::set): Normalize ranges for both bounds.
Aldy Hernandez [Fri, 14 Oct 2022 10:06:56 +0000 (12:06 +0200)]
Drop -0.0 in frange::set() for !HONOR_SIGNED_ZEROS.
Similar to what we do for NANs when !HONOR_NANS and Inf when
flag_finite_math_only, we can remove -0.0 from the range at creation
time.
We were kinda sorta doing this because there is a bug in
real_isdenormal that is causing flush_denormals_to_zero to saturate
[x, -0.0] to [x, +0.0] when !HONOR_SIGNED_ZEROS. Fixing this bug
(upcoming), causes us to leave -0.0 in places where we aren't
expecting it (the intersection code).
gcc/ChangeLog:
* value-range.cc (frange::set): Drop -0.0 for !HONOR_SIGNED_ZEROS.
Patrick Palka [Fri, 14 Oct 2022 13:07:01 +0000 (09:07 -0400)]
c++ modules: ICE with dynamic_cast [PR106304]
The FUNCTION_DECL we build for __dynamic_cast has an empty DECL_CONTEXT
but trees_out::tree_node expects FUNCTION_DECLs to have non-empty
DECL_CONTEXT, thus we crash when streaming out the dynamic_cast in the
below testcase.
This patch naively fixes this by setting DECL_CONTEXT for __dynamic_cast
appropriately. I suppose we should push it into the namespace too, like
we do for __cxa_atexit which is similarly lazily declared.
PR c++/106304
gcc/cp/ChangeLog:
* constexpr.cc (cxx_dynamic_cast_fn_p): Check for abi_node
instead of global_namespace.
* rtti.cc (build_dynamic_cast_1): Set DECL_CONTEXT and
DECL_SOURCE_LOCATION when building dynamic_cast_node. Push
it into the namespace.
gcc/testsuite/ChangeLog:
* g++.dg/modules/pr106304_a.C: New test.
* g++.dg/modules/pr106304_b.C: New test.
Aldy Hernandez [Thu, 13 Oct 2022 15:45:38 +0000 (17:45 +0200)]
Add cases for CFN_BUILT_IN_SIGNBIT[FL].
gcc/ChangeLog:
* gimple-range-op.cc
(gimple_range_op_handler::maybe_builtin_call): Add
CFN_BUILT_IN_SIGNBIT[FL]* entries.
Richard Biener [Fri, 14 Oct 2022 09:14:59 +0000 (11:14 +0200)]
tree-optimization/107254 - check and support live lanes from permutes
The following fixes an omission from adding SLP permute nodes which
is live lanes originating from those. We have to check that we
can extract the lane and have to actually code generate them.
PR tree-optimization/107254
* tree-vect-slp.cc (vect_slp_analyze_node_operations_1):
For permutes also analyze live lanes.
(vect_schedule_slp_node): For permutes also code generate
live lane extracts.
* gfortran.dg/vect/pr107254.f90: New testcase.
Eric Botcazou [Fri, 14 Oct 2022 09:52:04 +0000 (11:52 +0200)]
Fix PR target/107248
This is the infamous PR rtl-optimization/38644 rearing its ugly head for
leaf functions on SPARC more than a decade later... Richard E.'s generic
solution has never been implemented so let's do as other RISC back-ends did.
gcc/
PR target/107248
* config/sparc/sparc.cc (sparc_expand_prologue): Emit a frame
blockage for leaf functions.
(sparc_flat_expand_prologue): Emit frame instead of full blockage.
(sparc_expand_epilogue): Emit a frame blockage for leaf functions.
(sparc_flat_expand_epilogue): Emit frame instead of full blockage.
Jonathan Wakely [Wed, 12 Oct 2022 11:07:14 +0000 (12:07 +0100)]
libstdc++: Use markdown in Doxygen comment
This makes the comment easier to read in the source, without altering
the Doxygen output.
libstdc++-v3/ChangeLog:
* include/std/iostream: Use markdown in Doxygen comment.
Jørgen Kvalsvik [Fri, 7 Oct 2022 11:29:20 +0000 (13:29 +0200)]
gcov: test line count for label in then/else block
Add a test to catch regression in line counts for labels on top of
then/else blocks. Only the 'goto <label>' should contribute to the line
counter for the label, not the if.
gcc/testsuite/ChangeLog:
* gcc.misc-tests/gcov-4.c: New testcase.
Jørgen Kvalsvik [Tue, 4 Oct 2022 13:45:59 +0000 (15:45 +0200)]
gcov: test switch/break line counts
The coverage support will under some conditions decide to split edges to
accurately report coverage. By running the test suite with/without this
edge splitting a small diff shows up, addressed by this patch, which
should catch future regressions.
Removing the edge splitting:
$ diff --git a/gcc/profile.cc b/gcc/profile.cc
--- a/gcc/profile.cc
+++ b/gcc/profile.cc
@@ -1244,19 +1244,7 @@ branch_prob (bool thunk)
Don't do that when the locuses match, so
if (blah) goto something;
is not computed twice. */
- if (last
- && gimple_has_location (last)
- && !RESERVED_LOCATION_P (e->goto_locus)
- && !single_succ_p (bb)
- && (LOCATION_FILE (e->goto_locus)
- != LOCATION_FILE (gimple_location (last))
- || (LOCATION_LINE (e->goto_locus)
- != LOCATION_LINE (gimple_location (last)))))
- {
- basic_block new_bb = split_edge (e);
- edge ne = single_succ_edge (new_bb);
- ne->goto_locus = e->goto_locus;
- }
+
if ((e->flags & (EDGE_ABNORMAL | EDGE_ABNORMAL_CALL))
&& e->dest != EXIT_BLOCK_PTR_FOR_FN (cfun))
need_exit_edge = 1;
Assuming the .gcov files from make chec-gcc RUNTESTFLAGS=gcov.exp are
kept:
$ diff -r no-split-edge with-split-edge | grep -C 2 -E "^[<>]\s\s"
diff -r sans-split-edge/gcc/gcov-4.c.gcov with-split-edge/gcc/gcov-4.c.gcov
228c228
< -: 224: break;
---
> 1: 224: break;
231c231
< -: 227: break;
---
> #####: 227: break;
237c237
< -: 233: break;
---
> 2: 233: break;
gcc/testsuite/ChangeLog:
* g++.dg/gcov/gcov-1.C: Add line count check.
* gcc.misc-tests/gcov-4.c: Likewise.
Jakub Jelinek [Fri, 14 Oct 2022 07:37:01 +0000 (09:37 +0200)]
middle-end, c++, i386, libgcc: std::bfloat16_t and __bf16 arithmetic support
Here is a complete patch to add std::bfloat16_t support on
x86 (AArch64 and ARM left for later). Almost no BFmode optabs
are added by the patch, so for binops/unops it extends to SFmode
first and then truncates back to BFmode.
For {HF,SF,DF,XF,TF}mode -> BFmode conversions libgcc has implementations
of all those conversions so that we avoid double rounding, for
BFmode -> {DF,XF,TF}mode conversions to avoid growing libgcc too much
it emits BFmode -> SFmode conversion first and then converts to the even
wider mode, neither step should be imprecise.
For BFmode -> HFmode, it first emits a precise BFmode -> SFmode conversion
and then SFmode -> HFmode, because neither format is subset or superset
of the other, while SFmode is superset of both.
expr.cc then contains a -ffast-math optimization of the BF -> SF and
SF -> BF conversions if we don't optimize for space (and for the latter
if -frounding-math isn't enabled either).
For x86, perhaps truncsfbf2 optab could be defined for TARGET_AVX512BF16
but IMNSHO should FAIL if !flag_finite_math || flag_rounding_math
|| !flag_unsafe_math_optimizations, because I think the insn doesn't
raise on sNaNs, hardcodes round to nearest and flushes denormals to zero.
By default (unless x86 -fexcess-precision=16) we use float excess
precision for BFmode, so truncate only on explicit casts and assignments.
The patch introduces a single __bf16 builtin - __builtin_nansf16b,
because (__bf16) __builtin_nansf ("") will drop the sNaN into qNaN,
and uses f16b suffix instead of bf16 because there would be ambiguity on
log vs. logb - __builtin_logbf16 could be either log with bf16 suffix
or logb with f16 suffix. In other cases libstdc++ should mostly use
__builtin_*f for std::bfloat16_t overloads (we have a problem with
std::nextafter though but that one we have also for std::float16_t).
2022-10-14 Jakub Jelinek <jakub@redhat.com>
gcc/
* tree-core.h (enum tree_index): Add TI_BFLOAT16_TYPE.
* tree.h (bfloat16_type_node): Define.
* tree.cc (excess_precision_type): Promote bfloat16_type_mode
like float16_type_mode.
(build_common_tree_nodes): Initialize bfloat16_type_node if
BFmode is supported.
* expmed.h (maybe_expand_shift): Declare.
* expmed.cc (maybe_expand_shift): No longer static.
* expr.cc (convert_mode_scalar): Don't ICE on BF -> HF or HF -> BF
conversions. If there is no optab, handle BF -> {DF,XF,TF,HF}
conversions as separate BF -> SF -> {DF,XF,TF,HF} conversions, add
-ffast-math generic implementation for BF -> SF and SF -> BF
conversions.
* builtin-types.def (BT_BFLOAT16, BT_FN_BFLOAT16_CONST_STRING): New.
* builtins.def (BUILT_IN_NANSF16B): New builtin.
* fold-const-call.cc (fold_const_call): Handle CFN_BUILT_IN_NANSF16B.
* config/i386/i386.cc (classify_argument): Handle E_BCmode.
(ix86_libgcc_floating_mode_supported_p): Also return true for BFmode
for -msse2.
(ix86_mangle_type): Mangle BFmode as DF16b.
(ix86_invalid_conversion, ix86_invalid_unary_op,
ix86_invalid_binary_op): Remove.
(TARGET_INVALID_CONVERSION, TARGET_INVALID_UNARY_OP,
TARGET_INVALID_BINARY_OP): Don't redefine.
* config/i386/i386-builtins.cc (ix86_bf16_type_node): Remove.
(ix86_register_bf16_builtin_type): Use bfloat16_type_node rather than
ix86_bf16_type_node, only create it if still NULL.
* config/i386/i386-builtin-types.def (BFLOAT16): Likewise.
* config/i386/i386.md (cbranchbf4, cstorebf4): New expanders.
gcc/c-family/
* c-cppbuiltin.cc (c_cpp_builtins): If bfloat16_type_node,
predefine __BFLT16_*__ macros and for C++23 also
__STDCPP_BFLOAT16_T__. Predefine bfloat16_type_node related
macros for -fbuilding-libgcc.
* c-lex.cc (interpret_float): Handle CPP_N_BFLOAT16.
gcc/c/
* c-typeck.cc (convert_arguments): Don't promote __bf16 to
double.
gcc/cp/
* cp-tree.h (extended_float_type_p): Return true for
bfloat16_type_node.
* typeck.cc (cp_compare_floating_point_conversion_ranks): Set
extended{1,2} if mv{1,2} is bfloat16_type_node. Adjust comment.
gcc/testsuite/
* lib/target-supports.exp (check_effective_target_bfloat16,
check_effective_target_bfloat16_runtime, add_options_for_bfloat16):
New.
* gcc.dg/torture/bfloat16-basic.c: New test.
* gcc.dg/torture/bfloat16-builtin.c: New test.
* gcc.dg/torture/bfloat16-builtin-issignaling-1.c: New test.
* gcc.dg/torture/bfloat16-complex.c: New test.
* gcc.dg/torture/builtin-issignaling-1.c: Allow to be includable
from bfloat16-builtin-issignaling-1.c.
* gcc.dg/torture/floatn-basic.h: Allow to be includable from
bfloat16-basic.c.
* gcc.target/i386/vect-bfloat16-typecheck_2.c: Adjust expected
diagnostics.
* gcc.target/i386/sse2-bfloat16-scalar-typecheck.c: Likewise.
* gcc.target/i386/vect-bfloat16-typecheck_1.c: Likewise.
* g++.target/i386/bfloat_cpp_typecheck.C: Likewise.
libcpp/
* include/cpplib.h (CPP_N_BFLOAT16): Define.
* expr.cc (interpret_float_suffix): Handle bf16 and BF16 suffixes for
C++.
libgcc/
* config/i386/t-softfp (softfp_extensions): Add bfsf.
(softfp_truncations): Add tfbf xfbf dfbf sfbf hfbf.
(CFLAGS-extendbfsf2.c, CFLAGS-truncsfbf2.c, CFLAGS-truncdfbf2.c,
CFLAGS-truncxfbf2.c, CFLAGS-trunctfbf2.c, CFLAGS-trunchfbf2.c): Add
-msse2.
* config/i386/libgcc-glibc.ver (GCC_13.0.0): Export
__extendbfsf2 and __trunc{s,d,x,t,h}fbf2.
* config/i386/sfp-machine.h (_FP_NANSIGN_B): Define.
* config/i386/64/sfp-machine.h (_FP_NANFRAC_B): Define.
* config/i386/32/sfp-machine.h (_FP_NANFRAC_B): Define.
* soft-fp/brain.h: New file.
* soft-fp/truncsfbf2.c: New file.
* soft-fp/truncdfbf2.c: New file.
* soft-fp/truncxfbf2.c: New file.
* soft-fp/trunctfbf2.c: New file.
* soft-fp/trunchfbf2.c: New file.
* soft-fp/truncbfhf2.c: New file.
* soft-fp/extendbfsf2.c: New file.
libiberty/
* cp-demangle.h (D_BUILTIN_TYPE_COUNT): Increment.
* cp-demangle.c (cplus_demangle_builtin_types): Add std::bfloat16_t
entry.
(cplus_demangle_type): Demangle DF16b.
* testsuite/demangle-expected (_Z3xxxDF16b): New test.
Jakub Jelinek [Fri, 14 Oct 2022 07:33:23 +0000 (09:33 +0200)]
c++: Excess precision for ? int : float or int == float [PR107097, PR82071, PR87390]
The following incremental patch implements the C11 behavior (for all C++
versions) for
cond ? int : float
cond ? float : int
int cmp float
float cmp int
where int is any integral type, float any floating point type with
excess precision and cmp ==, !=, >, <, >=, <= and <=>.
2022-10-14 Jakub Jelinek <jakub@redhat.com>
PR c/82071
PR c/87390
PR c++/107097
gcc/cp/
* cp-tree.h (cp_ep_convert_and_check): Remove.
* cvt.cc (cp_ep_convert_and_check): Remove.
* call.cc (build_conditional_expr): Use excess precision for ?: with
one arm floating and another integral. Don't convert first to
semantic result type from integral types.
(convert_like_internal): Don't call cp_ep_convert_and_check, instead
just strip EXCESS_PRECISION_EXPR before calling cp_convert_and_check
or cp_convert.
* typeck.cc (cp_build_binary_op): Set may_need_excess_precision
for comparisons or SPACESHIP_EXPR with at least one operand integral.
Don't compute semantic_result_type if build_type is non-NULL. Call
cp_convert_and_check instead of cp_ep_convert_and_check.
gcc/testsuite/
* gcc.target/i386/excess-precision-8.c: For C++ wrap abort and
exit declarations into extern "C" block.
* gcc.target/i386/excess-precision-10.c: Likewise.
* g++.target/i386/excess-precision-7.C: Remove.
* g++.target/i386/excess-precision-8.C: New test.
* g++.target/i386/excess-precision-9.C: Remove.
* g++.target/i386/excess-precision-10.C: New test.
* g++.target/i386/excess-precision-12.C: New test.
Jakub Jelinek [Fri, 14 Oct 2022 07:28:57 +0000 (09:28 +0200)]
c++: Implement excess precision support for C++ [PR107097, PR323]
The following patch implements excess precision support for C++.
Like for C, it uses EXCESS_PRECISION_EXPR tree to say that its operand
is evaluated in excess precision and what the semantic type of the
expression is.
In most places I've followed what the C FE does in similar spots, so
e.g. for binary ops if one or both operands are already
EXCESS_PRECISION_EXPR, strip those away or for operations that might need
excess precision (+, -, *, /) check if the operands should use excess
precision and convert to that type and at the end wrap into
EXCESS_PRECISION_EXPR with the common semantic type.
This patch follows the C99 handling where it differs from C11 handling.
There are some cases which needed to be handled differently, the C FE can
just strip EXCESS_PRECISION_EXPR (replace it with its operand) when handling
explicit cast, but that IMHO isn't right for C++ - the discovery what exact
conversion should be used (e.g. if user conversion or standard or their
sequence) should be decided based on the semantic type (i.e. type of
EXCESS_PRECISION_EXPR), and that decision continues in convert_like* where
we pick the right user conversion, again, if say some class has ctor
from double and long double and we are on ia32 with standard excess
precision promoting float/double to long double, then we should pick the
ctor from double. Or when some other class has ctor from just double,
and EXCESS_PRECISION_EXPR semantic type is float, we should choose the
user ctor from double, but actually just convert the long double excess
precision to double and not to float first. We need to make sure
even identity conversion converts from excess precision to the semantic one
though, but if identity is chained with other conversions, we don't want
the identity next_conversion to drop to semantic precision only to widen
afterwards.
The existing testcases tweaks were for cases on i686-linux where excess
precision breaks those tests, e.g. if we have
double d = 4.2;
if (d == 4.2)
then it does the expected thing only with -fexcess-precision=fast,
because with -fexcess-precision=standard it is actually
double d = 4.2;
if ((long double) d == 4.2L)
where 4.2L is different from 4.2. I've added -fexcess-precision=fast
to some tests and changed other tests to use constants that are exactly
representable and don't suffer from these excess precision issues.
There is one exception, pr68180.C looks like a bug in the patch which is
also present in the C FE (so I'd like to get it resolved incrementally
in both). Reduced testcase:
typedef float __attribute__((vector_size (16))) float32x4_t;
float32x4_t foo(float32x4_t x, float y) { return x + y; }
with -m32 -std=c11 -Wno-psabi or -m32 -std=c++17 -Wno-psabi
it is rejected with:
pr68180.c:2:52: error: conversion of scalar ‘long double’ to vector ‘float32x4_t’ {aka ‘__vector(4) float’} involves truncation
but without excess precision (say just -std=c11 -Wno-psabi or -std=c++17 -Wno-psabi)
it is accepted. Perhaps we should pass down the semantic type to
scalar_to_vector and use the semantic type rather than excess precision type
in the diagnostics.
2022-10-14 Jakub Jelinek <jakub@redhat.com>
PR middle-end/323
PR c++/107097
gcc/
* doc/invoke.texi (-fexcess-precision=standard): Mention that the
option now also works in C++.
gcc/c-family/
* c-common.def (EXCESS_PRECISION_EXPR): Remove comment part about
the tree being specific to C/ObjC.
* c-opts.cc (c_common_post_options): Handle flag_excess_precision
in C++ the same as in C.
* c-lex.cc (interpret_float): Set const_type to excess_precision ()
even for C++.
gcc/cp/
* parser.cc (cp_parser_primary_expression): Handle
EXCESS_PRECISION_EXPR with REAL_CST operand the same as REAL_CST.
* cvt.cc (cp_ep_convert_and_check): New function.
* call.cc (build_conditional_expr): Add excess precision support.
When type_after_usual_arithmetic_conversions returns error_mark_node,
use gcc_checking_assert that it is because of uncomparable floating
point ranks instead of checking all those conditions and make it
work also with complex types.
(convert_like_internal): Likewise. Add NESTED_P argument, pass true
to recursive calls to convert_like.
(convert_like): Add NESTED_P argument, pass it through to
convert_like_internal. For other overload pass false to it.
(convert_like_with_context): Pass false to NESTED_P.
(convert_arg_to_ellipsis): Add excess precision support.
(magic_varargs_p): For __builtin_is{finite,inf,inf_sign,nan,normal}
and __builtin_fpclassify return 2 instead of 1, document what it
means.
(build_over_call): Don't handle former magic 2 which is no longer
used, instead for magic 1 remove EXCESS_PRECISION_EXPR.
(perform_direct_initialization_if_possible): Pass false to NESTED_P
convert_like argument.
* constexpr.cc (cxx_eval_constant_expression): Handle
EXCESS_PRECISION_EXPR.
(potential_constant_expression_1): Likewise.
* pt.cc (tsubst_copy, tsubst_copy_and_build): Likewise.
* cp-tree.h (cp_ep_convert_and_check): Declare.
* cp-gimplify.cc (cp_fold): Handle EXCESS_PRECISION_EXPR.
* typeck.cc (cp_common_type): For COMPLEX_TYPEs, return error_mark_node
if recursive call returned it.
(convert_arguments): For magic 1 remove EXCESS_PRECISION_EXPR.
(cp_build_binary_op): Add excess precision support. When
cp_common_type returns error_mark_node, use gcc_checking_assert that
it is because of uncomparable floating point ranks instead of checking
all those conditions and make it work also with complex types.
(cp_build_unary_op): Likewise.
(cp_build_compound_expr): Likewise.
(build_static_cast_1): Remove EXCESS_PRECISION_EXPR.
gcc/testsuite/
* gcc.target/i386/excess-precision-1.c: For C++ wrap abort and
exit declarations into extern "C" block.
* gcc.target/i386/excess-precision-2.c: Likewise.
* gcc.target/i386/excess-precision-3.c: Likewise. Remove
check_float_nonproto and check_double_nonproto tests for C++.
* gcc.target/i386/excess-precision-7.c: For C++ wrap abort and
exit declarations into extern "C" block.
* gcc.target/i386/excess-precision-9.c: Likewise.
* g++.target/i386/excess-precision-1.C: New test.
* g++.target/i386/excess-precision-2.C: New test.
* g++.target/i386/excess-precision-3.C: New test.
* g++.target/i386/excess-precision-4.C: New test.
* g++.target/i386/excess-precision-5.C: New test.
* g++.target/i386/excess-precision-6.C: New test.
* g++.target/i386/excess-precision-7.C: New test.
* g++.target/i386/excess-precision-9.C: New test.
* g++.target/i386/excess-precision-11.C: New test.
* c-c++-common/dfp/convert-bfp-10.c: Add -fexcess-precision=fast
as dg-additional-options.
* c-c++-common/dfp/compare-eq-const.c: Likewise.
* g++.dg/cpp1z/constexpr-96862.C: Likewise.
* g++.dg/cpp1z/decomp12.C (main): Use 2.25 instead of 2.3 to
avoid excess precision differences.
* g++.dg/other/thunk1.C: Add -fexcess-precision=fast
as dg-additional-options.
* g++.dg/vect/pr64410.cc: Likewise.
* g++.dg/cpp1y/pr68180.C: Likewise.
* g++.dg/vect/pr89653.cc: Likewise.
* g++.dg/cpp0x/variadic-tuple.C: Likewise.
* g++.dg/cpp0x/nsdmi-union1.C: Use 4.25 instead of 4.2 to
avoid excess precision differences.
* g++.old-deja/g++.brendan/copy9.C: Add -fexcess-precision=fast
as dg-additional-options.
* g++.old-deja/g++.brendan/overload7.C: Likewise.
Joseph Myers [Fri, 14 Oct 2022 02:18:45 +0000 (02:18 +0000)]
c: C2x storage class specifiers in compound literals
Implement the C2x feature of storage class specifiers in compound
literals. Such storage class specifiers (static, register or
thread_local; also constexpr, but we don't yet have C2x constexpr
support implemented) can be used before the type name (not mixed with
type specifiers, unlike in declarations) and have the same semantics
and constraints as for declarations of named objects. Also allow GNU
__thread to be used, given that thread_local can be.
Bootstrapped with no regressions for x86_64-pc-linux-gnu.
gcc/c/
* c-decl.cc (build_compound_literal): Add parameter scspecs.
Handle storage class specifiers.
* c-parser.cc (c_token_starts_compound_literal)
(c_parser_compound_literal_scspecs): New.
(c_parser_postfix_expression_after_paren_type): Add parameter
scspecs. Call pedwarn_c11 for use of storage class specifiers.
Update call to build_compound_literal.
(c_parser_cast_expression, c_parser_sizeof_expression)
(c_parser_alignof_expression): Handle storage class specifiers for
compound literals. Update calls to
c_parser_postfix_expression_after_paren_type.
(c_parser_postfix_expression): Update syntax comment.
* c-tree.h (build_compound_literal): Update prototype.
* c-typeck.cc (c_mark_addressable): Diagnose taking address of
register compound literal.
gcc/testsuite/
* gcc.dg/c11-complit-1.c, gcc.dg/c11-complit-2.c,
gcc.dg/c11-complit-3.c, gcc.dg/c2x-complit-2.c,
gcc.dg/c2x-complit-3.c, gcc.dg/c2x-complit-4.c,
gcc.dg/c2x-complit-5.c, gcc.dg/c2x-complit-6.c,
gcc.dg/c2x-complit-7.c, gcc.dg/c90-complit-2.c,
gcc.dg/gnu2x-complit-1.c, gcc.dg/gnu2x-complit-2.c: New tests.
GCC Administrator [Fri, 14 Oct 2022 00:16:35 +0000 (00:16 +0000)]
Daily bump.
Eric Botcazou [Thu, 13 Oct 2022 22:55:40 +0000 (00:55 +0200)]
Fix bogus -Wstringop-overflow warning
If you compile the testcase with -O2 -fno-inline -Wall, you get:
In function 'process_array3':
cc1: warning: 'process_array4' accessing 4 bytes in a region of size 3 [-
Wstringop-overflow=]
cc1: note: referencing argument 1 of type 'char[4]'
t.c:6:6: note: in a call to function 'process_array4'
6 | void process_array4 (char a[4], int n)
| ^~~~~~~~~~~~~~
cc1: warning: 'process_array4' accessing 4 bytes in a region of size 3 [-
Wstringop-overflow=]
cc1: note: referencing argument 1 of type 'char[4]'
t.c:6:6: note: in a call to function 'process_array4'
That's because the ICF IPA pass has identified the two functions and turned
process_array3 into a wrapper of process_array4.
gcc/
* gimple-ssa-warn-access.cc (pass_waccess::check_call): Return
early for calls made from thunks.
gcc/testsuite/
* gcc.dg/Wstringop-overflow-89.c: New test.
Jason Merrill [Tue, 29 Jun 2021 21:45:21 +0000 (17:45 -0400)]
c++: trivial formatting cleanups
Split out from the C++ contracts patch.
gcc/cp/ChangeLog:
* cp-tree.h: Fix whitespace.
* parser.h: Fix whitespace.
* decl.cc: Fix whitespace.
* parser.cc: Fix whitespace.
* pt.cc: Fix whitespace.
David Malcolm [Thu, 13 Oct 2022 20:05:35 +0000 (16:05 -0400)]
analyzer: fix ICE introduced in r13-3168 [PR107210]
gcc/analyzer/ChangeLog:
PR analyzer/107210
* svalue.cc (constant_svalue::maybe_fold_bits_within): Only
attempt to extract individual bits when tree_fits_uhwi_p.
gcc/testsuite/ChangeLog:
PR analyzer/107210
* gfortran.dg/analyzer/pr107210.f90: New test.
Signed-off-by: David Malcolm <dmalcolm@redhat.com>
Tobias Burnus [Thu, 13 Oct 2022 18:38:27 +0000 (20:38 +0200)]
libgomp: Add Fortran testcases for omp_in_explicit_task
Fortranized testcases of commits r13-3257-ga58a965eb73
and r13-3258-g0ec4e93fb9f.
libgomp/ChangeLog:
* testsuite/libgomp.fortran/task-7.f90: New test.
* testsuite/libgomp.fortran/task-8.f90: New test.
* testsuite/libgomp.fortran/task-in-explicit-1.f90: New test.
* testsuite/libgomp.fortran/task-in-explicit-2.f90: New test.
* testsuite/libgomp.fortran/task-in-explicit-3.f90: New test.
* testsuite/libgomp.fortran/task-reduction-17.f90: New test.
* testsuite/libgomp.fortran/task-reduction-18.f90: New test.
Eric Botcazou [Wed, 12 Oct 2022 07:27:19 +0000 (09:27 +0200)]
Fix emit_group_store regression on big-endian
The recent optimization implemented for complex modes contains an oversight
for big-endian platforms: it uses a lowpart SUBREG when the integer modes
have different sizes, but this does not match the semantics of the PARALLELs
which have a bundled byte offset; this offset is always zero in the code
path and the lowpart is not at offset zero on big-endian platforms.
gcc/
* expr.cc (emit_group_stote): Fix handling of modes of different
sizes for big-endian targets in latest change and add commentary.
Martin Liska [Thu, 13 Oct 2022 13:39:08 +0000 (15:39 +0200)]
use proper DECL_INITIAL for VTV
gcc/cp/ChangeLog:
* vtable-class-hierarchy.cc (vtv_generate_init_routine): Emit
an artificial variable that would be put into .preinit_array
section.
gcc/ChangeLog:
* output.h (assemble_vtv_preinit_initializer): Remove.
* varasm.cc (assemble_vtv_preinit_initializer): Remove.
Andrew MacLeod [Wed, 5 Oct 2022 14:42:07 +0000 (10:42 -0400)]
propagate partial equivs in the cache.
Adjust on-entry cache propagation to look for and propagate both full
and partial equivalences.
gcc/
PR tree-optimization/102540
PR tree-optimization/102872
* gimple-range-cache.cc (ranger_cache::fill_block_cache):
Handle partial equivs.
(ranger_cache::range_from_dom): Cleanup dump output.
gcc/testsuite/
* gcc.dg/pr102540.c: New.
* gcc.dg/pr102872.c: New.
Andrew MacLeod [Thu, 6 Oct 2022 19:01:24 +0000 (15:01 -0400)]
Add partial equivalence recognition to cast and bitwise and.
This provides the hooks that will register partial equivalencies for
casts and bitwise AND operations with the appropriate bit pattern.
* range-op.cc (operator_cast::lhs_op1_relation): New.
(operator_bitwise_and::lhs_op1_relation): New.
Andrew MacLeod [Fri, 7 Oct 2022 16:55:32 +0000 (12:55 -0400)]
Add equivalence iterator to relation oracle.
Instead of looping over an exposed equivalence bitmap, provide iterators
to loop over equivalences, partial equivalences, or both.
* gimple-range-cache.cc (ranger_cache::fill_block_cache): Use
iterator.
* value-relation.cc
(equiv_relation_iterator::equiv_relation_iterator): New.
(equiv_relation_iterator::next): New.
(equiv_relation_iterator::get_name): New.
* value-relation.h (class relation_oracle): Privatize some methods.
(class equiv_relation_iterator): New.
(FOR_EACH_EQUIVALENCE): New.
(FOR_EACH_PARTIAL_EQUIV): New.
(FOR_EACH_PARTIAL_AND_FULL_EQUIV): New.
Andrew MacLeod [Thu, 6 Oct 2022 19:00:52 +0000 (15:00 -0400)]
Add partial equivalence support to the relation oracle.
This provides enhancements to the equivalence oracle to also track
partial equivalences. They are tracked similar to equivalences, except
it tracks a 'slice' of another ssa name. 8, 16, 32 and 64 bit slices are
tracked. This will allow casts and mask of the same value to compare
equal.
* value-relation.cc (equiv_chain::dump): Don't print empty
equivalences.
(equiv_oracle::equiv_oracle): Allocate a partial equiv table.
(equiv_oracle::~equiv_oracle): Release the partial equiv table.
(equiv_oracle::add_partial_equiv): New.
(equiv_oracle::partial_equiv_set): New.
(equiv_oracle::partial_equiv): New.
(equiv_oracle::query_relation): Check for partial equivs too.
(equiv_oracle::dump): Also dump partial equivs.
(dom_oracle::register_relation): Handle partial equivs.
(dom_oracle::query_relation): Check for partial equivs.
* value-relation.h (enum relation_kind_t): Add partial equivs.
(relation_partial_equiv_p): New.
(relation_equiv_p): New.
(class pe_slice): New.
(class equiv_oracle): Add prototypes.
(pe_to_bits): New.
(bits_to_pe): New.
(pe_min): New.
Marek Polacek [Tue, 11 Oct 2022 18:16:54 +0000 (14:16 -0400)]
c++: ICE with VEC_INIT_EXPR and defarg [PR106925]
Since r12-8066, in cxx_eval_vec_init we perform expand_vec_init_expr
while processing the default argument in this test. At this point
start_preparsed_function hasn't yet set current_function_decl.
expand_vec_init_expr then leads to maybe_splice_retval_cleanup which
checks DECL_CONSTRUCTOR_P (current_function_decl) without checking that
c_f_d is non-null first. It seems correct that c_f_d is null here, so
it seems to me that maybe_splice_retval_cleanup should check c_f_d as
in the following patch.
PR c++/106925
gcc/cp/ChangeLog:
* except.cc (maybe_splice_retval_cleanup): Check current_function_decl.
Make the bool const.
gcc/testsuite/ChangeLog:
* g++.dg/cpp0x/initlist-defarg3.C: New test.
Richard Biener [Thu, 13 Oct 2022 12:56:01 +0000 (14:56 +0200)]
tree-optimization/107247 - reduce SLP reduction accumulator
The following makes sure to reduce a multi-vector SLP reduction
accumulator to a single vector using vector operations if
easily possible (if the number of lanes in the vector type is
a multiple of the number of scalar accumulators).
PR tree-optimization/107247
* tree-vect-loop.cc (vect_create_epilog_for_reduction):
Reduce multi vector SLP reduction accumulators. Check
the adjusted number of accumulator vectors against
one for the re-use in the epilogue.
Jakub Jelinek [Thu, 13 Oct 2022 14:22:21 +0000 (16:22 +0200)]
machmode: Introduce GET_MODE_NEXT_MODE with previous GET_MODE_WIDER_MODE meaning, add new GET_MODE_WIDER_MODE
On Wed, Oct 05, 2022 at 04:02:25PM -0400, Jason Merrill wrote:
> > > > @@ -5716,7 +5716,13 @@ emit_store_flag_1 (rtx target, enum rtx_
> > > > {
> > > > machine_mode optab_mode = mclass == MODE_CC ? CCmode : compare_mode;
> > > > icode = optab_handler (cstore_optab, optab_mode);
> > > > - if (icode != CODE_FOR_nothing)
> > > > + if (icode != CODE_FOR_nothing
> > > > + /* Don't consider [BH]Fmode as usable wider mode, as neither is
> > > > + a subset or superset of the other. */
> > > > + && (compare_mode == mode
> > > > + || !SCALAR_FLOAT_MODE_P (compare_mode)
> > > > + || maybe_ne (GET_MODE_PRECISION (compare_mode),
> > > > + GET_MODE_PRECISION (mode))))
> > >
> > > Why do you need to do this here (and in prepare_cmp_insn, and similarly in
> > > can_compare_p)? Shouldn't get_wider skip over modes that are not actually
> > > wider?
> >
> > I'm afraid too many places rely on all modes of a certain class to be
> > visible when walking from "narrowest" to "widest" mode, say
> > FOR_EACH_MODE_IN_CLASS/FOR_EACH_MODE/FOR_EACH_MODE_UNTIL/FOR_EACH_WIDER_MODE
> > etc. wouldn't work at all if GET_MODE_WIDER_MODE (BFmode) == SFmode
> > && GET_MODE_WIDER_MODE (HFmode) == SFmode.
>
> Yes, it seems they need to change now that their assumptions have been
> violated. I suppose FOR_EACH_MODE_IN_CLASS would need to change to not use
> get_wider, and users of FOR_EACH_MODE/FOR_EACH_MODE_UNTIL need to decide
> whether they want an iteration that uses get_wider (likely with a new name)
> or not.
Here is a patch which does that.
Though I admit I didn't go carefully through all 24 GET_MODE_WIDER_MODE
uses, 54 FOR_EACH_MODE_IN_CLASS uses, 3 FOR_EACH_MODE uses, 24
FOR_EACH_MODE_FROM, 6 FOR_EACH_MODE_UNTIL and 15 FOR_EACH_WIDER_MODE uses.
It is more important to go through the GET_MODE_WIDER_MODE and
FOR_EACH_WIDER_MODE uses because the patch changes behavior for those,
the rest keep their previous meaning and so can be changed incrementally
if the other meaning is desirable to them (I've of course changed the 3
spots I had to change in the previous BFmode patch and whatever triggered
during the bootstraps).
2022-10-13 Jakub Jelinek <jakub@redhat.com>
* genmodes.cc (emit_mode_wider): Emit previous content of
mode_wider array into mode_next array and for mode_wider
emit always VOIDmode for !CLASS_HAS_WIDER_MODES_P classes,
otherwise skip through modes with the same precision.
* machmode.h (mode_next): Declare.
(GET_MODE_NEXT_MODE): New inline function.
(mode_iterator::get_next, mode_iterator::get_known_next): New
function templates.
(FOR_EACH_MODE_IN_CLASS): Use get_next instead of get_wider.
(FOR_EACH_MODE): Use get_known_next instead of get_known_wider.
(FOR_EACH_MODE_FROM): Use get_next instead of get_wider.
(FOR_EACH_WIDER_MODE_FROM): Define.
(FOR_EACH_NEXT_MODE): Define.
* expmed.cc (emit_store_flag_1): Use FOR_EACH_WIDER_MODE_FROM
instead of FOR_EACH_MODE_FROM.
* optabs.cc (prepare_cmp_insn): Likewise. Remove redundant
!CLASS_HAS_WIDER_MODES_P check.
(prepare_float_lib_cmp): Use FOR_EACH_WIDER_MODE_FROM instead of
FOR_EACH_MODE_FROM.
* config/i386/i386-expand.cc (get_mode_wider_vector): Use
GET_MODE_NEXT_MODE instead of GET_MODE_WIDER_MODE.
Wilco Dijkstra [Thu, 13 Oct 2022 13:41:55 +0000 (14:41 +0100)]
[AArch64] Improve bit tests [PR105773]
Since AArch64 sets all flags on logical operations, comparisons with zero
can be combined into an AND even if the condition is LE or GT. Add a new
CC_NZV mode used by ANDS/BICS/TST instructions.
gcc/
PR target/105773
* config/aarch64/aarch64.cc (aarch64_select_cc_mode): Allow
GT/LE for merging compare with zero into AND.
(aarch64_get_condition_code_1): Add CC_NZVmode support.
* config/aarch64/aarch64-modes.def: Add CC_NZV.
* config/aarch64/aarch64.md: Use CC_NZV in cmp+and patterns.
gcc/testsuite/
PR target/105773
* gcc.target/aarch64/ands_2.c: Test for ANDS.
* gcc.target/aarch64/bics_2.c: Test for BICS.
* gcc.target/aarch64/tst_2.c: Test for TST.
* gcc.target/aarch64/tst_imm_split_1.c: Fix test.
Richard Biener [Thu, 13 Oct 2022 12:24:05 +0000 (14:24 +0200)]
tree-optimization/107160 - avoid reusing multiple accumulators
Epilogue vectorization is not set up to re-use a vectorized
accumulator consisting of more than one vector. For non-SLP
we always reduce to a single but for SLP that isn't happening.
In such case we currenlty miscompile the epilog so avoid this.
PR tree-optimization/107160
* tree-vect-loop.cc (vect_create_epilog_for_reduction):
Do not register accumulator if we failed to reduce it
to a single vector.
* gcc.dg/vect/pr107160.c: New testcase.
Aldy Hernandez [Thu, 13 Oct 2022 06:08:26 +0000 (08:08 +0200)]
Add op1_op2_relation for float operands.
op1_op2_relation can be called for relops (bool = a < b) as well as
regular binary operators (z = a + b). This patch adds the overloaded
method for floating point results.
gcc/ChangeLog:
* range-op-float.cc (range_operator_float::op1_op2_relation): New.
(class foperator_equal): Add using.
(class foperator_not_equal): Same.
(class foperator_lt): Same.
(class foperator_le): Same.
(class foperator_gt): Same.
(class foperator_ge): Same.
* range-op.cc (range_op_handler::op1_op2_relation): New.
* range-op.h (range_operator_float::op1_op2_relation): New.
Richard Biener [Thu, 13 Oct 2022 10:59:09 +0000 (12:59 +0200)]
diagnose return statement in match.pd (with { ... } expressions
The expression in (with { ... } is used like a statement expression
which means control flow that leaves it is not allowed. The following
explicitely diagnoses 'return' and fixes up the few cases that crept
into match.pd (oops). Any such return will prematurely end matching
the current expression.
* genmatch.cc (parser::parse_c_expr): Diagnose 'return'.
* match.pd: Replace 'return' statements in with expressions
with appropriate variants.
Andre Vieira [Thu, 13 Oct 2022 11:09:38 +0000 (12:09 +0100)]
ifcvt: Fix bitpos calculation in bitfield lowering [PR107229]
The bitposition calculation for the bitfield lowering in loop if conversion was
not taking DECL_FIELD_OFFSET into account, which meant that it would result in
wrong bitpositions for bitfields that did not end up having representations
starting at the beginning of the struct.
gcc/ChangeLog:
PR tree-optimization/107229
* tree-if-conv.cc (get_bitfield_rep): Fix bitposition calculation.
gcc/testsuite/ChangeLog:
* gcc.dg/vect/pr107229-1.c: New test.
* gcc.dg/vect/pr107229-2.c: New test.
* gcc.dg/vect/pr107229-3.c: New test.
Xi Ruoyao [Wed, 12 Oct 2022 14:06:07 +0000 (22:06 +0800)]
LoongArch: implement count_{leading,trailing}_zeros
LoongArch always support clz and ctz instructions, so we can always use
__builtin_{clz,ctz} for count_{leading,trailing}_zeros. This improves
the code of libgcc, and also benefits Glibc once we merge longlong.h
there.
Bootstrapped and regtested on loongarch64-linux-gnu.
include/ChangeLog:
* longlong.h [__loongarch__] (count_leading_zeros): Define.
[__loongarch__] (count_trailing_zeros): Likewise.
[__loongarch__] (COUNT_LEADING_ZEROS_0): Likewise.
Andre Vieira [Thu, 13 Oct 2022 09:34:27 +0000 (10:34 +0100)]
vect: Don't pattern match BITFIELD_REF's of non-integrals [PR107226]
The original patch supported matching the vect_recog_bitfield_ref_pattern for
BITFIELD_REF's where the first operand didn't have a INTEGRAL_TYPE_P type.
That means it would also match vectors, leading to regressions in targets that
supported vectorization of those.
gcc/ChangeLog:
PR tree-optimization/107226
* tree-vect-patterns.cc (vect_recog_bitfield_ref_pattern): Reject
BITFIELD_REF's with non integral typed first operands.
Lulu Cheng [Wed, 12 Oct 2022 03:02:11 +0000 (11:02 +0800)]
LoongArch: Fixed a bug in the loongarch architecture of libitm package.
Add a soft floating point condition to the register recovery part of the code.
libitm/ChangeLog:
* config/loongarch/sjlj.S: Add a soft floating point condition to the
register recovery part of the code.
Joseph Myers [Thu, 13 Oct 2022 01:13:36 +0000 (01:13 +0000)]
c: Do not use *_IS_IEC_60559 == 2
A late change for C2x (addressing comments from the second round of
editorial review before the CD ballot, postdating the most recent
public working draft) removed the value 2 for *_IS_IEC_60559 (a new
<float.h> macro added in C2x). Adjust the implementation accordingly
not to use this value.
Bootstrapped with no regressions for x86_64-pc-linux-gnu.
gcc/
* ginclude/float.h (FLT_IS_IEC_60559, DBL_IS_IEC_60559)
(LDBL_IS_IEC_60559): Update comment.
gcc/c-family/
* c-cppbuiltin.cc (builtin_define_float_constants): Do not use
value 2 for *_IS_IEC_60559.
gcc/testsuite/
* gcc.dg/c2x-float-10.c: Do not expect value 2 for *_IS_IEC_60559.
GCC Administrator [Thu, 13 Oct 2022 00:17:37 +0000 (00:17 +0000)]
Daily bump.
Lewis Hyatt [Thu, 6 Oct 2022 22:05:02 +0000 (18:05 -0400)]
preprocessor: Fix tracking of system header state [PR60014,PR60723]
The token_streamer class (which implements gcc mode -E and
-save-temps/-no-integrated-cpp) needs to keep track whether the last tokens
output were in a system header, so that it can generate line marker
annotations as necessary for a downstream consumer to reconstruct the
state. The logic for tracking it, which was added by r5-1863 to resolve
PR60723, has some edge case issues as revealed by the three new test
cases. The first, coming from the original PR60014, was incidentally fixed by
r9-1926 for unrelated reasons. The other two were still failing on master
prior to this commit. Such code paths were not realizable prior to r13-1544,
which made it possible for the token streamer to see CPP_PRAGMA tokens in more
contexts.
The two main issues being corrected here are:
1) print.prev_was_system_token needs to indicate whether the previous token
output was in a system location. However, it was not being set on every token,
only on those that triggered the main code path; specifically it was not
triggered on a CPP_PRAGMA token. Testcase 2 covers this case.
2) The token_streamer uses a variable "line_marker_emitted" to remember
whether a line marker has been emitted while processing a given token, so that
it wouldn't be done more than once in case multiple conditions requiring a
line marker are true. There was no reason for this to be a member variable
that retains its value from token to token, since it is just needed for
tracking the state locally while processing a single given token. The fact
that it could retain its value for a subsequent token is rather difficult to
observe, but testcase 3 demonstrates incorrect behavior resulting from
that. Moving this to a local variable also simplifies understanding the
control flow going forward.
gcc/c-family/ChangeLog:
PR preprocessor/60014
PR preprocessor/60723
* c-ppoutput.cc (class token_streamer): Remove member
line_marker_emitted to...
(token_streamer::stream): ...a local variable here. Set
print.prev_was_system_token on all code paths.
gcc/testsuite/ChangeLog:
PR preprocessor/60014
PR preprocessor/60723
* gcc.dg/cpp/pr60014-1.c: New test.
* gcc.dg/cpp/pr60014-1.h: New test.
* gcc.dg/cpp/pr60014-2.c: New test.
* gcc.dg/cpp/pr60014-2.h: New test.
* gcc.dg/cpp/pr60014-3.c: New test.
* gcc.dg/cpp/pr60014-3.h: New test.
François Dumont [Wed, 10 Aug 2022 17:35:41 +0000 (19:35 +0200)]
libstdc++: Add __gnu_debug::basic_string<>::compare overloads
Rather than adding those implementations we are adding a:
using _Base::compare;
so that any compare method not implemented at __gnu_debug::basic_string
level are injected from the base class.
Also review how __gnu_debug::basic_string is tested. Now require to define
_GLIBCXX_TEST_DEBUG_STRING when running 'make check-debug'.
libstdc++-v3/ChangeLog
* include/debug/string: Add using _Base::compare.
(__gnu_debug::basic_string<>::compare(const basic_string<>&)): Remove.
(__gnu_debug::basic_string<>::compare(size_type, size_type, const basic_string<>&)):
Remove.
(__gnu_debug::basic_string<>::compare(size_type, size_type, const basic_string<>&,
size_type, size_type)): Remove.
* testsuite/util/testsuite_string.h [_GLIBCXX_TEST_DEBUG_STRING]: Include <debug/string>.
* testsuite/21_strings/basic_string/operations/compare/char/1.cc: Include testsuite_string.h
and use __gnu_test::string.
* testsuite/21_strings/basic_string/operations/compare/char/13650.cc: Likewise.
* testsuite/21_strings/basic_string/operations/compare/char/2.cc: Likewise.
* testsuite/21_strings/basic_string/operations/rfind/char/1.cc: Likewise.
* testsuite/21_strings/basic_string/operations/rfind/char/2.cc: Likewise.
* testsuite/21_strings/basic_string/operations/rfind/char/3.cc: Likewise.
* testsuite/21_strings/basic_string/operations/compare/wchar_t/1.cc:
Include testsuite_string.h
and use __gnu_test::wstring.
* testsuite/21_strings/basic_string/operations/compare/wchar_t/13650.cc: Likewise.
* testsuite/21_strings/basic_string/operations/compare/wchar_t/2.cc: Likewise.
Marek Polacek [Wed, 28 Sep 2022 14:37:34 +0000 (10:37 -0400)]
c++: Remove maybe-rvalue OR in implicit move
This patch removes the two-stage overload resolution when performing
implicit move, whereby the compiler does two separate overload resolutions:
one treating the operand as an rvalue, and then (if that resolution fails)
another one treating the operand as an lvalue. In the standard this was
introduced via CWG 1579 and implemented in gcc in r251035. In r11-2412,
we disabled the fallback OR in C++20 (but not in C++17). Then C++23 P2266
removed the fallback overload resolution, and changed the implicit move rules
once again. So we wound up with three different behaviors.
The two overload resolutions approach was complicated and quirky, so
users should transition to the newer model. Removing the maybe-rvalue
OR also allows us to simplify our code, for instance, now we can get
rid of LOOKUP_PREFER_RVALUE altogether.
This change means that code that previously didn't compile in C++17 will
now compile, for example:
struct S1 { S1(S1 &&); };
struct S2 : S1 {};
S1
f (S2 s)
{
return s; // OK, derived-to-base, use S1::S1(S1&&)
}
And conversely, code that used to work in C++17 may not compile anymore:
struct W {
W();
};
struct F {
F(W&);
F(W&&) = delete;
};
F fn ()
{
W w;
return w; // use w as rvalue -> use of deleted function F::F(W&&)
}
I plan to add a note to porting_to.html.
gcc/cp/ChangeLog:
* call.cc (standard_conversion): Remove LOOKUP_PREFER_RVALUE code.
(reference_binding): Honor clk_implicit_rval even pre-C++20.
(implicit_conversion_1): Remove LOOKUP_PREFER_RVALUE code.
(build_user_type_conversion_1): Likewise.
(convert_like_internal): Likewise.
(build_over_call): Likewise.
* cp-tree.h (LOOKUP_PREFER_RVALUE): Remove.
(LOOKUP_NO_NARROWING): Adjust definition.
* except.cc (build_throw): Don't perform two overload resolutions.
* typeck.cc (maybe_warn_pessimizing_move): Don't use
LOOKUP_PREFER_RVALUE.
(check_return_expr): Don't perform two overload resolutions.
gcc/testsuite/ChangeLog:
* g++.dg/cpp0x/Wredundant-move10.C: Adjust dg-warning.
* g++.dg/cpp0x/Wredundant-move7.C: Likewise.
* g++.dg/cpp0x/move-return2.C: Remove dg-error.
* g++.dg/cpp0x/move-return4.C: Likewise.
* g++.dg/cpp0x/ref-qual20.C: Adjust expected return value.
* g++.dg/cpp0x/move-return5.C: New test.
Aldy Hernandez [Tue, 11 Oct 2022 15:23:42 +0000 (17:23 +0200)]
Add range-op entry for floating point NEGATE_EXPR.
gcc/ChangeLog:
* range-op-float.cc (class foperator_negate): New.
(floating_op_table::floating_op_table): Add NEGATE_EXPR
(range_op_float_tests): Add negate tests.
Harald Anlauf [Tue, 11 Oct 2022 20:08:48 +0000 (22:08 +0200)]
Fortran: check types of operands of arithmetic binary operations [PR107217]
gcc/fortran/ChangeLog:
PR fortran/107217
* arith.cc (gfc_arith_plus): Compare consistency of types of operands.
(gfc_arith_minus): Likewise.
(gfc_arith_times): Likewise.
(gfc_arith_divide): Likewise.
(arith_power): Check that both operands are of numeric type.
gcc/testsuite/ChangeLog:
PR fortran/107217
* gfortran.dg/pr107217.f90: New test.
Jakub Jelinek [Wed, 12 Oct 2022 16:43:58 +0000 (18:43 +0200)]
libgomp: Fix up OpenMP 5.2 feature bullet
The previous bullet correctly mentions 5.2 added for Fortran
allocators directive which is a replacement of allocate directive
associated with ALLOCATE statement to differentiate it at parse time
from allocate directive as declarative one not associated with ALLOCATE
statement, but the deprecation bullet talks about non-existing allocator
directive.
2022-10-12 Jakub Jelinek <jakub@redhat.com>
* libgomp.texi (OpenMP 5.2): Fix up allocator -> allocate directive
in deprecation bullet.
Jakub Jelinek [Wed, 12 Oct 2022 16:39:20 +0000 (18:39 +0200)]
libgomp: Add omp_in_explicit_task support
This is pretty straightforward, if gomp_thread ()->task is NULL,
it can't be explicit task, otherwise if
gomp_thread ()->task->kind == GOMP_TASK_IMPLICIT, it is an implicit
task, otherwise explicit task.
2022-10-12 Jakub Jelinek <jakub@redhat.com>
* omp.h.in (omp_in_explicit_task): Declare.
* omp_lib.h.in (omp_in_explicit_task): Likewise.
* omp_lib.f90.in (omp_in_explicit_task): New interface.
* libgomp.map (OMP_5.2): New symbol version, export
omp_in_explicit_task and omp_in_explicit_task_.
* task.c (omp_in_explicit_task): New function.
* fortran.c (omp_in_explicit_task): Add ialias_redirect.
(omp_in_explicit_task_): New function.
* libgomp.texi (OpenMP 5.2): Mark omp_in_explicit_task as implemented.
* testsuite/libgomp.c-c++-common/task-in-explicit-1.c: New test.
* testsuite/libgomp.c-c++-common/task-in-explicit-2.c: New test.
* testsuite/libgomp.c-c++-common/task-in-explicit-3.c: New test.
Jakub Jelinek [Wed, 12 Oct 2022 15:54:08 +0000 (17:54 +0200)]
libgomp: Fix up creation of artificial teams
When not in explicit parallel/target/teams construct, we in some cases create
an artificial parallel with a single thread (either to handle target nowait
or for task reduction purposes). In those cases, it handled again artificially
created implicit task (created by gomp_new_icv for cases where we needed to write
to some ICVs), but as the testcases show, didn't take into account possibility
of this being done from explicit task(s). The code would destroy/free the previous
task and replace it with the new implicit task. If task is an explicit task
(when teams is NULL, all explicit tasks behave like if (0)), it is a pointer to
a local stack variable, so freeing it doesn't work, and additionally we shouldn't
lose the explicit tasks - the new implicit task should instead replace the
ancestor task which is the first implicit one.
2022-10-12 Jakub Jelinek <jakub@redhat.com>
* task.c (gomp_create_artificial_team): Fix up handling of invocations
from within explicit task.
* target.c (GOMP_target_ext): Likewise.
* testsuite/libgomp.c/task-7.c: New test.
* testsuite/libgomp.c/task-8.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-17.c: New test.
* testsuite/libgomp.c-c++-common/task-reduction-18.c: New test.
Jason Merrill [Sat, 8 Oct 2022 00:34:53 +0000 (20:34 -0400)]
c++: defer all consteval in default args [DR2631]
The proposed resolution of CWG2631 extends our current handling of
source_location::current to all consteval functions: default arguments
are not evaluated until they're used in a call, the same should apply to
evaluation of immediate invocations. And similarly for default member
initializers.
Previously we folded source_location::current in cp_fold_r; now we fold all
consteval calls in default arguments/member initializers in bot_replace.
DR 2631
gcc/cp/ChangeLog:
* cp-tree.h (source_location_current_p): Remove.
* name-lookup.h (struct cp_binding_level): Remove
immediate_fn_ctx_p.
* call.cc (in_immediate_context): All default args
and DMI are potentially immediate context.
(immediate_invocation_p): Don't treat source_location specially.
(struct in_consteval_if_p_temp_override): Move to cp-tree.h.
* constexpr.cc (get_nth_callarg): Move to cp-tree.h.
* cp-gimplify.cc (cp_fold_r): Don't fold consteval.
* name-lookup.cc (begin_scope): Don't set immediate_fn_ctx_p.
* parser.cc (cp_parser_lambda_declarator_opt): Likewise.
(cp_parser_direct_declarator): Likewise.
* pt.cc (tsubst_default_argument): Open sk_function_parms level.
* tree.cc (source_location_current_p): Remove.
(bot_replace): Fold consteval here.
(break_out_target_exprs): Handle errors.
gcc/testsuite/ChangeLog:
* g++.dg/cpp2a/consteval-defarg3.C: New test.
Patrick Palka [Wed, 12 Oct 2022 15:14:11 +0000 (11:14 -0400)]
libstdc++: Implement ranges::repeat_view from P2474R2
libstdc++-v3/ChangeLog:
* include/std/ranges (repeat_view): Define.
(repeat_view::_Iterator): Define.
(views::__detail::__can_repeat_view): Define.
(views::__detail::__can_bounded_repeat_view): Define.
(views::_Repeat, views::repeat): Define.
* testsuite/std/ranges/repeat/1.cc: New test.
Ju-Zhe Zhong [Tue, 11 Oct 2022 06:23:33 +0000 (14:23 +0800)]
RISC-V: Remove TUPLE size macro define. [NFC]
gcc/ChangeLog:
* config/riscv/riscv-vector-builtins.h: Remove unused macro.