platform/upstream/gcc.git
2 years agolto-wrapper.cc: Delete offload_names temp files in case of error [PR106686]
Tobias Burnus [Mon, 22 Aug 2022 07:49:52 +0000 (09:49 +0200)]
lto-wrapper.cc: Delete offload_names temp files in case of error [PR106686]

Usually, the caller takes care of the .o files for the offload compilers
(suffix: ".target.o"). However, if an error occurs during processing
(e.g. fatal error by lto1), they were not deleted.

gcc/ChangeLog:

PR lto/106686
* lto-wrapper.cc (free_array_of_ptrs): Move before tool_cleanup.
(tool_cleanup): Unlink offload_names.
(compile_offload_image): Take filename argument to set it early.
(compile_images_for_offload_targets): Update call; set
offload_names to NULL after freeing the array.

2 years agotree-optimization/105937 - avoid uninit diagnostics crossing iterations
Richard Biener [Fri, 19 Aug 2022 12:12:52 +0000 (14:12 +0200)]
tree-optimization/105937 - avoid uninit diagnostics crossing iterations

The following avoids adding PHIs to the worklist for uninit processing
if we reach them following backedges.  That confuses predicate analysis
because it assumes the use is happening in the same iteration as the the
definition.  For the testcase in the PR the situation is like

void foo (int val)
{
  int uninit;
  # val = PHI <..> (B)
  for (..)
    {
      if (..)
        {
          .. = val; (C)
          val = uninit;
        }
      # val = PHI <..> (A)
    }
}

and starting from (A) with 'uninit' as argument we arrive at (B)
and from there at (C).  Predicate analysis then tries to prove
the predicate of (B) (not the backedge) can prove that the
path from (B) to (C) is unreachable which isn't really what it
necessary - that's what we'd need to do when the preheader
edge of the loop were the edge with the uninitialized def.

So the following makes those cases intentionally false negatives.

PR tree-optimization/105937
* tree-ssa-uninit.cc (find_uninit_use): Do not queue PHIs
on backedges.
(execute_late_warn_uninitialized): Mark backedges.

* g++.dg/uninit-pr105937.C: New testcase.

2 years agoImprove uninit analysis
Richard Biener [Thu, 18 Aug 2022 13:49:58 +0000 (15:49 +0200)]
Improve uninit analysis

The following reduces the number of false positives in uninit analysis
by providing fallback for situations the current analysis gives up
and thus warns because it cannot prove initialization.

The first situation is when compute_control_dep_chain gives up walking
because it runs into either param_uninit_control_dep_attempts or
MAX_CHAIN_LEN.  If in the process it did not collect a single path
from function entry to the interesting PHI edge then we'll give up
and diagnose.  The following patch insteads provides a sparse path
including only those predicates that always hold when the PHI edge
is reached in that case.  That's cheap to produce but may in some
odd cases prove less precise than what the code tries now (enumerating
all possible paths from function entry to the PHI edge, but only
use the first N of those and only require unreachability of those N).

The second situation is when the set of predicates computed to hold
on the use stmt was formed from multiple paths (there's a similar
enumeration of all paths and their predicates from the PHI def to the
use).  In that case use_preds.use_cannot_happen gives up because
it doesn't know which of the predicates from which path from PHI to
the use it can use to prove unreachability of the PHI edge that has
the uninitialized def.  The patch for this case simply computes
the intersection of the predicates and uses that for further analysis,
but in a crude way since the predicate vectors are not sorted.
Fortunately the total size is limited - we have max MAX_NUM_CHAINS
number of predicates each of length MAX_CHAIN_LEN so the brute
force intersection code should behave quite reasonable in practice.

* gimple-predicate-analysis.cc (predicate::use_cannot_happen):
If the use is guarded with multiple predicate paths compute
the predicates intersection before going forward.  When
compute_control_dep_chain wasn't able to come up with at
least one path from function entry to the PHI edge compute
a conservative sparse path instead.

2 years agoanalyzer: add missing final keyword
Martin Liska [Mon, 22 Aug 2022 03:13:18 +0000 (05:13 +0200)]
analyzer: add missing final keyword

Fixes the following clang warning:
gcc/analyzer/region-model.cc:5096:8: warning: 'subclass_equal_p' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]

gcc/analyzer/ChangeLog:

* region-model.cc: Add missing final keyword.

2 years agoDaily bump.
GCC Administrator [Mon, 22 Aug 2022 00:16:22 +0000 (00:16 +0000)]
Daily bump.

2 years agoDaily bump.
GCC Administrator [Sun, 21 Aug 2022 00:16:33 +0000 (00:16 +0000)]
Daily bump.

2 years agofortran: Drop -static-lib{gfortran,quadmath} from f951 [PR46539]
Jakub Jelinek [Sat, 20 Aug 2022 19:20:04 +0000 (21:20 +0200)]
fortran: Drop -static-lib{gfortran,quadmath} from f951 [PR46539]

As discussed earlier, all other -static-lib* options are Driver only,
these 2 are Driver in common.opt and Fortran in lang.opt.

The spec files never pass the -static-lib* options down to any compiler
(f951 etc.), so the 2 errors below are reported only when one
runs ./f951 -static-libgfortran by hand.

The following patch just removes f951 support of these options, the
gfortran driver behavior remains as before.  For other -static-lib*
option (and even these because it is never passed to f951) we never
error if we can't support those options, and e.g. Darwin is actually
able to handle those options through other means.

2022-08-20  Jakub Jelinek  <jakub@redhat.com>

PR fortran/46539
* lang.opt (static-libgfortran, static-libquadmath): Change Fortran
to Driver.
* options.cc (gfc_handle_option): Don't handle OPT_static_libgfortran
nor OPT_static_libquadmath here.

2 years agoLoongArch: Add support code model extreme.
Lulu Cheng [Thu, 18 Aug 2022 09:26:13 +0000 (17:26 +0800)]
LoongArch: Add support code model extreme.

Use five instructions to calculate a signed 64-bit offset relative to the pc.

gcc/ChangeLog:

* config/loongarch/loongarch-opts.cc: Allow cmodel to be extreme.
* config/loongarch/loongarch.cc (loongarch_call_tls_get_addr):
Add extreme support for TLS GD and LD types.
(loongarch_legitimize_tls_address): Add extreme support for TLS LE
and IE.
(loongarch_split_symbol): When compiling with -mcmodel=extreme,
the symbol address will be obtained through five instructions.
(loongarch_print_operand_reloc): Add support.
(loongarch_print_operand): Add support.
(loongarch_print_operand_address): Add support.
(loongarch_option_override_internal): Set '-mcmodel=extreme' option
incompatible with '-mno-explicit-relocs'.
* config/loongarch/loongarch.md (@lui_l_hi20<mode>):
Loads bits 12-31 of data into registers.
(lui_h_lo20): Load bits 32-51 of the data and spell bits 0-31 of
the source register.
(lui_h_hi12): Load bits 52-63 of the data and spell bits 0-51 of
the source register.
* config/loongarch/predicates.md: Symbols need to be decomposed
when defining the macro TARGET_CMODEL_EXTREME
* doc/invoke.texi: Modify the description information of cmodel in the document.
Document -W[no-]extreme-plt.

gcc/testsuite/ChangeLog:

* gcc.target/loongarch/func-call-1.c: Add option '-mcmodel=normal'.
* gcc.target/loongarch/func-call-2.c: Likewise.
* gcc.target/loongarch/func-call-3.c: Likewise.
* gcc.target/loongarch/func-call-4.c: Likewise.
* gcc.target/loongarch/func-call-5.c: Likewise.
* gcc.target/loongarch/func-call-6.c: Likewise.
* gcc.target/loongarch/func-call-7.c: Likewise.
* gcc.target/loongarch/func-call-8.c: Likewise.
* gcc.target/loongarch/relocs-symbol-noaddend.c: Likewise.
* gcc.target/loongarch/func-call-extreme-1.c: New test.
* gcc.target/loongarch/func-call-extreme-2.c: New test.

2 years agolibcpp: Implement C++23 P2290R3 - Delimited escape sequences [PR106645]
Jakub Jelinek [Sat, 20 Aug 2022 08:26:55 +0000 (10:26 +0200)]
libcpp: Implement C++23 P2290R3 - Delimited escape sequences [PR106645]

The following patch implements the C++23 P2290R3 paper.

2022-08-20  Jakub Jelinek  <jakub@redhat.com>

PR c++/106645
libcpp/
* include/cpplib.h (struct cpp_options): Implement
P2290R3 - Delimited escape sequences.  Add delimite_escape_seqs
member.
* init.cc (struct lang_flags): Likewise.
(lang_defaults): Add delim column.
(cpp_set_lang): Copy over delimite_escape_seqs.
* charset.cc (extend_char_range): New function.
(_cpp_valid_ucn): Use it.  Handle delimited escape sequences.
(convert_hex): Likewise.
(convert_oct): Likewise.
(convert_ucn): Use extend_char_range.
(convert_escape): Call convert_oct even for \o.
(_cpp_interpret_identifier): Handle delimited escape sequences.
* lex.cc (get_bidi_ucn_1): Likewise.  Add end argument, fill it in.
(get_bidi_ucn): Adjust get_bidi_ucn_1 caller.  Use end argument to
compute num_bytes.
gcc/testsuite/
* c-c++-common/cpp/delimited-escape-seq-1.c: New test.
* c-c++-common/cpp/delimited-escape-seq-2.c: New test.
* c-c++-common/cpp/delimited-escape-seq-3.c: New test.
* c-c++-common/Wbidi-chars-24.c: New test.
* gcc.dg/cpp/delimited-escape-seq-1.c: New test.
* gcc.dg/cpp/delimited-escape-seq-2.c: New test.
* g++.dg/cpp/delimited-escape-seq-1.C: New test.
* g++.dg/cpp/delimited-escape-seq-2.C: New test.

2 years agoDaily bump.
GCC Administrator [Sat, 20 Aug 2022 00:16:26 +0000 (00:16 +0000)]
Daily bump.

2 years agomkoffload: Cleanup temporary omp_requires_file
Tobias Burnus [Fri, 19 Aug 2022 14:09:28 +0000 (16:09 +0200)]
mkoffload: Cleanup temporary omp_requires_file

The file (suffix ".mkoffload.omp_requires") used to save the 'omp requires'
data has to be passed to maybe_unlink for cleanup or -v -save-temps stderr
diagnostic. That was missed before. - For GCN, the same has to be done for
the files with suffix ".mkoffload.dbg.o".

gcc/ChangeLog:

* config/gcn/mkoffload.cc (main): Add omp_requires_file and dbgobj to
files_to_cleanup.
* config/i386/intelmic-mkoffload.cc (prepare_target_image): Add
omp_requires_file to temp_files.
* config/nvptx/mkoffload.cc (omp_requires_file): New global static var.
(main): Remove local omp_requires_file var.
(tool_cleanup): Handle omp_requires_file.

2 years agoRemove path_range_query constructor that takes an edge.
Aldy Hernandez [Thu, 18 Aug 2022 15:55:19 +0000 (17:55 +0200)]
Remove path_range_query constructor that takes an edge.

The path_range_query constructor that takes an edge is really a
convenience function for the loop-ch pass.  It feels wrong to pollute
the API with such a specialized function that could be done with
a small inline function closer to its user.

As an added benefit, we remove one use of reset_path.  The last
remaining one is the forward threader one.

Tested, thread-counted, and benchmarked on x86-64 Linux.

gcc/ChangeLog:

* gimple-range-path.cc (path_range_query::path_range_query):
Remove constructor that takes edge.
* gimple-range-path.h (class path_range_query): Same.
* tree-ssa-loop-ch.cc (edge_range_query): New.
(entry_loop_condition_is_static): Call edge_range_query.

2 years agoAdd further FOR_EACH_ macros
Stefan Schulze Frielinghaus [Fri, 19 Aug 2022 07:33:13 +0000 (09:33 +0200)]
Add further FOR_EACH_ macros

contrib/ChangeLog:

* clang-format: Add further FOR_EACH_ macros.

2 years agoi386: Add ABI test for __bf16 type
Haochen Jiang [Thu, 23 Jun 2022 06:16:43 +0000 (14:16 +0800)]
i386: Add ABI test for __bf16 type

gcc/testsuite/ChangeLog:

* gcc.target/x86_64/abi/bf16/abi-bf16.exp: New test.
* gcc.target/x86_64/abi/bf16/args.h: Ditto.
* gcc.target/x86_64/abi/bf16/asm-support.S: Ditto.
* gcc.target/x86_64/abi/bf16/bf16-check.h: Ditto.
* gcc.target/x86_64/abi/bf16/bf16-helper.h: Ditto.
* gcc.target/x86_64/abi/bf16/defines.h: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/abi-bf16-ymm.exp: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/args.h: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/bf16-ymm-check.h: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/test_m256_returning.c: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/test_passing_m256.c: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/test_passing_structs.c: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/test_passing_unions.c: Ditto.
* gcc.target/x86_64/abi/bf16/m256bf16/test_varargs-m256.c: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/abi-bf16-zmm.exp: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/args.h: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/bf16-zmm-check.h: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/test_m512_returning.c: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/test_passing_m512.c: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/test_passing_structs.c: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/test_passing_unions.c: Ditto.
* gcc.target/x86_64/abi/bf16/m512bf16/test_varargs-m512.c: Ditto.
* gcc.target/x86_64/abi/bf16/macros.h: Ditto.
* gcc.target/x86_64/abi/bf16/test_3_element_struct_and_unions.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_basic_alignment.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_basic_array_size_and_align.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_basic_returning.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_basic_sizes.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_basic_struct_size_and_align.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_basic_union_size_and_align.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_m128_returning.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_passing_floats.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_passing_m128.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_passing_structs.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_passing_unions.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_struct_returning.c: Ditto.
* gcc.target/x86_64/abi/bf16/test_varargs-m128.c: Ditto.

2 years agoDaily bump.
GCC Administrator [Fri, 19 Aug 2022 00:16:27 +0000 (00:16 +0000)]
Daily bump.

2 years agopreprocessor: Support #warning for standard C2x
Joseph Myers [Thu, 18 Aug 2022 23:01:01 +0000 (23:01 +0000)]
preprocessor: Support #warning for standard C2x

ISO C2x standardizes the existing #warning extension.  Arrange
accordingly for it not to be diagnosed with -std=c2x -pedantic, but to
be diagnosed with -Wc11-c2x-compat.

Bootstrapped with no regressions for x86_64-pc-linux-gnu.

gcc/testsuite/
* gcc.dg/cpp/c11-warning-1.c, gcc.dg/cpp/c11-warning-2.c,
gcc.dg/cpp/c11-warning-3.c, gcc.dg/cpp/c11-warning-4.c,
gcc.dg/cpp/c2x-warning-1.c, gcc.dg/cpp/c2x-warning-2.c,
gcc.dg/cpp/gnu11-warning-1.c, gcc.dg/cpp/gnu11-warning-2.c,
gcc.dg/cpp/gnu11-warning-3.c, gcc.dg/cpp/gnu11-warning-4.c,
gcc.dg/cpp/gnu2x-warning-1.c, gcc.dg/cpp/gnu2x-warning-2.c: New
tests.

libcpp/
* include/cpplib.h (struct cpp_options): Add warning_directive.
* init.cc (struct lang_flags, lang_defaults): Add
warning_directive.
* directives.cc (DIRECTIVE_TABLE): Mark #warning as STDC2X not
EXTENSION.
(directive_diagnostics): Diagnose #warning with -Wc11-c2x-compat,
or with -pedantic for a standard not supporting #warning.

2 years agoxtensa: Improve indirect sibling call handling
Takayuki 'January June' Suwa [Thu, 18 Aug 2022 10:03:51 +0000 (19:03 +0900)]
xtensa: Improve indirect sibling call handling

No longer needs the dedicated hard register (A11) for the address of the
call and the split patterns for fixups, due to the introduction of appropriate
register class and constraint.

(Note: "ISC_REGS" contains a hard register A8 used as a "static chain"
 pointer for nested functions, but no problem;  Pointer to nested function
 actually points to "trampoline", and trampoline itself doesn't receive
 "static chain" pointer to its parent's stack frame from the caller.)

gcc/ChangeLog:

* config/xtensa/xtensa.h
(enum reg_class, REG_CLASS_NAMES, REG_CLASS_CONTENTS):
Add new register class "ISC_REGS".
* config/xtensa/constraints.md (c): Add new register constraint.
* config/xtensa/xtensa.md (define_constants): Remove "A11_REG".
(sibcall_internal, sibcall_value_internal):
Change to use the new register constraint, and remove two split
patterns for fixups that are no longer needed.

gcc/testsuite/ChangeLog:

* gcc.target/xtensa/sibcalls.c: Add a new test function to ensure
that registers for arguments (occupy from A2 to A7) and for indirect
sibcall (should be assigned to A8) neither conflict nor spill out.

2 years agoRevert "Fortran: fix invalid rank error in ASSOCIATED when rank is remapped [PR77652]"
Harald Anlauf [Thu, 18 Aug 2022 19:24:29 +0000 (21:24 +0200)]
Revert "Fortran: fix invalid rank error in ASSOCIATED when rank is remapped [PR77652]"

This reverts commit 0110cfd5449bae3a772f45ea2e4c5dab5b7a8ccd.

2 years agoRISC-V: Standardize formatting of SFB ALU conditional move
Maciej W. Rozycki [Thu, 18 Aug 2022 15:50:26 +0000 (16:50 +0100)]
RISC-V: Standardize formatting of SFB ALU conditional move

Standardize the formatting of SFB ALU conditional move operations from:

beq a2,zero,1f; mv a0,zero; 1: # movcc

to:

beq a2,zero,1f # movcc
mv a0,zero
1:

for consistency with other assembly code produced.  No functional change.

gcc/
* config/riscv/riscv.md (*mov<GPR:mode><X:mode>cc): Fix output
pattern formatting.

2 years agocontrib: Fix a typo in contrib/git-fetch-vendor.sh
Andrea Corallo [Thu, 18 Aug 2022 15:20:13 +0000 (17:20 +0200)]
contrib: Fix a typo in contrib/git-fetch-vendor.sh

2022-08-18  Andrea Corallo  <andrea.corallo@arm.com>

* git-fetch-vendor.sh : Fix typo.

2 years agoanalyzer: warn on the use of floating-points operands in the size argument [PR106181]
Tim Lange [Thu, 18 Aug 2022 09:36:08 +0000 (11:36 +0200)]
analyzer: warn on the use of floating-points operands in the size argument [PR106181]

This patch fixes the ICE reported in PR106181 and adds a new warning to
the analyzer complaining about the use of floating-point operands.

Regrtested on Linux x86_64.

2022-08-17  Tim Lange  <mail@tim-lange.me>

gcc/analyzer/ChangeLog:

PR analyzer/106181
* analyzer.opt: Add Wanalyzer-imprecise-floating-point-arithmetic.
* region-model.cc (is_any_cast_p): Formatting.
(region_model::check_region_size): Ensure precondition.
(class imprecise_floating_point_arithmetic): New abstract
diagnostic class for all floating-point related warnings.
(class float_as_size_arg): Concrete diagnostic class to complain
about floating-point operands inside the size argument.
(class contains_floating_point_visitor):
New visitor to find floating-point operands inside svalues.
(region_model::check_dynamic_size_for_floats): New function.
(region_model::set_dynamic_extents):
Call to check_dynamic_size_for_floats.
* region-model.h (class region_model):
Add region_model::check_dynamic_size_for_floats.

gcc/ChangeLog:

PR analyzer/106181
* doc/invoke.texi: Add Wanalyzer-imprecise-fp-arithmetic.

gcc/testsuite/ChangeLog:

PR analyzer/106181
* gcc.dg/analyzer/allocation-size-1.c: New test.
* gcc.dg/analyzer/imprecise-floating-point-1.c: New test.
* gcc.dg/analyzer/pr106181.c: New test.

2 years agoMake path_range_query standalone and add reset_path.
Aldy Hernandez [Wed, 17 Aug 2022 11:18:01 +0000 (13:18 +0200)]
Make path_range_query standalone and add reset_path.

These are a bunch of cleanups inspired by Richi's suggestion of making
path_range_query standalone, instead of having to call
compute_ranges() for each path.

I've made the ranger need explicit, and moved the responsibility for
its creation to the caller.

I've also investigated and documented why the forward threader needs its
own compute exit dependencies variant.  I can't wait for it to go away
:-/.

I've also added constructors that take a path and dependencies, and
made compute_ranges() private.  Unfortunately, reinstantiating
path_range_query in the forward threader caused a 14% performance
regression in DOM, because the old threader calls it over and over on
the same path to simplify statements (some of which not even in the
IL, but that's old news).

In the meantime, I've left the ability to reset a path, but this time
appropriately called reset_path().

Tested, benchmarked, and thread counted on x86-64 Linux.

gcc/ChangeLog:

* gimple-range-path.cc (path_range_query::path_range_query): Add
various constructors to take a path.
(path_range_query::~path_range_query): Remove m_alloced_ranger.
(path_range_query::range_on_path_entry): Adjust for m_ranger being
a reference.
(path_range_query::set_path): Rename to...
(path_range_query::reset_path): ...this and call compute_ranges.
(path_range_query::ssa_range_in_phi): Adjust for m_ranger
reference.
(path_range_query::range_defined_in_block): Same.
(path_range_query::compute_ranges_in_block): Same.
(path_range_query::adjust_for_non_null_uses): Same.
(path_range_query::compute_exit_dependencies): Use m_path instead
of argument.
(path_range_query::compute_ranges): Remove path argument.
(path_range_query::range_of_stmt): Adjust for m_ranger reference.
(path_range_query::compute_outgoing_relations): Same.
* gimple-range-path.h (class path_range_query): Add various
constructors.
Make compute_ranges and compute_exit_dependencies private.
Rename set_path to reset_path.
Make m_ranger a reference.
Remove m_alloced_ranger.
* tree-ssa-dom.cc (pass_dominator::execute): Adjust constructor to
path_range_query.
* tree-ssa-loop-ch.cc (entry_loop_condition_is_static): Take a
ranger and instantiate a new path_range_query every time.
(ch_base::copy_headers): Pass ranger instead of path_range_query.
* tree-ssa-threadbackward.cc (class back_threader): Remove m_solver.
(back_threader::~back_threader): Remove m_solver.
(back_threader::find_taken_edge_switch): Adjust for m_ranger
reference.
(back_threader::find_taken_edge_cond): Same.
(back_threader::dump): Remove m_solver.
(back_threader::back_threader): Move verify_marked_backedges
here from the path_range_query constructor.
* tree-ssa-threadedge.cc (hybrid_jt_simplifier::simplify): Move
some code from compute_ranges_from_state here.
(hybrid_jt_simplifier::compute_ranges_from_state): Rename...
(hybrid_jt_simplifier::compute_exit_dependencies): ...to this.
* tree-ssa-threadedge.h (class hybrid_jt_simplifier): Rename
compute_ranges_from_state to compute_exit_dependencies.
Remove m_path.

2 years agomiddle-end/106617 - fix fold_binary_op_with_conditional_arg pattern issue
Richard Biener [Thu, 18 Aug 2022 09:10:30 +0000 (11:10 +0200)]
middle-end/106617 - fix fold_binary_op_with_conditional_arg pattern issue

Now that we have parts of fold_binary_op_with_conditional_arg duplicated
in match.pd and are using ! to take or throw away the result we have to
be careful to not have both implementations play games which each other,
causing quadratic behavior.  In particular the match.pd implementation
requires both arms to simplify while the fold-const.cc is happy with
just one arm simplifying (something we cannot express in match.pd).

The fix is to simply not enable the match.pd pattern for GENERIC.

PR middle-end/106617
* match.pd ((a ? b : c) > d -> a ? (b > d) : (c > d)): Fix
guard, disable on GENERIC to not cause quadratic behavior
with the fold-const.cc implementation and the use of !

* gcc.dg/pr106617.c: New testcase.

2 years agogcov-dump: properly use INCLUDE_VECTOR
Andrew Pinski [Thu, 18 Aug 2022 12:36:28 +0000 (14:36 +0200)]
gcov-dump: properly use INCLUDE_VECTOR

PR gcov-profile/106659

gcc/ChangeLog:

* gcov-dump.cc (INCLUDE_VECTOR): Include vector.h with
  INCLUDE_VECTOR.

2 years agox86: Support vector __bf16 type
konglin1 [Thu, 18 Aug 2022 07:28:10 +0000 (15:28 +0800)]
x86: Support vector __bf16 type

gcc/ChangeLog:

* config/i386/i386-expand.cc (ix86_expand_sse_movcc): Handle vector
BFmode.
(ix86_expand_vector_init_duplicate): Support vector BFmode.
(ix86_expand_vector_init_one_nonzero): Ditto.
(ix86_expand_vector_init_one_var): Ditto.
(ix86_expand_vector_init_concat): Ditto.
(ix86_expand_vector_init_interleave): Ditto.
(ix86_expand_vector_init_general): Ditto.
(ix86_expand_vector_init): Ditto.
(ix86_expand_vector_set_var): Ditto.
(ix86_expand_vector_set): Ditto.
(ix86_expand_vector_extract): Ditto.
* config/i386/i386.cc (classify_argument): Add BF vector modes.
(function_arg_64): Ditto.
(ix86_gimplify_va_arg): Ditto.
(ix86_get_ssemov): Ditto.
* config/i386/i386.h (VALID_AVX256_REG_MODE): Add BF vector modes.
(VALID_AVX512F_REG_MODE): Ditto.
(host_detect_local_cpu): Ditto.
(VALID_SSE2_REG_MODE): Ditto.
* config/i386/i386.md: Add BF vector modes.
(MODE_SIZE): Ditto.
(ssemodesuffix): Add bf suffix for BF vector modes.
(ssevecmode): Ditto.
* config/i386/sse.md (VMOVE): Adjust for BF vector modes.
(VI12HFBF_AVX512VL): Ditto.
(V_256_512): Ditto.
(VF_AVX512HFBF16): Ditto.
(VF_AVX512BWHFBF16): Ditto.
(VIHFBF): Ditto.
(avx512): Ditto.
(VIHFBF_256): Ditto.
(VIHFBF_AVX512BW): Ditto.
(VI2F_256_512):Ditto.
(V8_128):Ditto.
(V16_256): Ditto.
(V32_512): Ditto.
(sseinsnmode): Ditto.
(sseconstm1): Ditto.
(sseintmodesuffix): New mode_attr.
(avx512fmaskmode): Ditto.
(avx512fmaskmodelower): Ditto.
(ssedoublevecmode): Ditto.
(ssehalfvecmode): Ditto.
(ssehalfvecmodelower): Ditto.
(ssescalarmode): Add vector BFmode mapping.
(ssescalarmodelower): Ditto.
(ssexmmmode): Ditto.
(ternlogsuffix): Ditto.
(ssescalarsize): Ditto.
(sseintprefix): Ditto.
(i128): Ditto.
(xtg_mode): Ditto.
(bcstscalarsuff): Ditto.
(<avx512>_blendm<mode>): New define_insn for BFmode.
(<avx512>_store<mode>_mask): Ditto.
(vcond_mask_<mode><avx512fmaskmodelower>): Ditto.
(vec_set<mode>_0): New define_insn for BF vector set.
(V8BFH_128): New mode_iterator for BFmode.
(avx512fp16_mov<mode>): Ditto.
(vec_set<mode>): New define_insn for BF vector set.
(@vec_extract_hi_<mode>): Ditto.
(@vec_extract_lo_<mode>): Ditto.
(vec_set_hi_<mode>): Ditto.
(vec_set_lo_<mode>): Ditto.
(*vec_extract<mode>_0): New define_insn_and_split for BF
vector extract.
(*vec_extract<mode>): New define_insn.
(VEC_EXTRACT_MODE): Add BF vector modes.
(PINSR_MODE): Add V8BF.
(sse2p4_1): Ditto.
(pinsr_evex_isa): Ditto.
(<sse2p4_1>_pinsr<ssemodesuffix>): Adjust to support
insert for V8BFmode.
(pbroadcast_evex_isa): Add BF vector modes.
(AVX2_VEC_DUP_MODE): Ditto.
(VEC_INIT_MODE): Ditto.
(VEC_INIT_HALF_MODE): Ditto.
(avx2_pbroadcast<mode>): Adjust to support BF vector mode
broadcast.
(avx2_pbroadcast<mode>_1): Ditto.
(<avx512>_vec_dup<mode>_1): Ditto.
(<mask_codefor><avx512>_vec_dup_gpr<mode><mask_name>):
Ditto.

gcc/testsuite/ChangeLog:

* g++.target/i386/vect-bfloat16-1.C: New test.
* gcc.target/i386/vect-bfloat16-1.c: New test.
* gcc.target/i386/vect-bfloat16-2a.c: New test.
* gcc.target/i386/vect-bfloat16-2b.c: New test.
* gcc.target/i386/vect-bfloat16-typecheck_1.c: New test.
* gcc.target/i386/vect-bfloat16-typecheck_2.c: New test.

2 years agobuild: regenerate gcc/configure
Martin Liska [Thu, 18 Aug 2022 10:38:59 +0000 (12:38 +0200)]
build: regenerate gcc/configure

After the change 71f068a9b3332a2179dfc807cf9138f691d77461, gcc/configure
needs to re-generated.

gcc/ChangeLog:

* configure: Regenerate.

2 years agoMakefile.def: drop remnants of unused libelf
Sergei Trofimovich [Thu, 18 Aug 2022 06:46:03 +0000 (07:46 +0100)]
Makefile.def: drop remnants of unused libelf

Use of libelf was removed from gcc in r0-104274-g48215350c24d52 ("re PR
lto/46273 (Failed to bootstrap)") around 2010, before gcc-4.6.0.

This change removes unused references to libelf from top-level configure
and Makefile.

/
* Makefile.def: Drop libelf module and gcc-configure dependency
on it.
* Makefile.in: Regenerate with 'autogen Makefile.def'.
* Makefile.tpl (HOST_EXPORTS): Drop unused LIBELFLIBS and
LIBELFINC.
* configure: Regenrate.
* configure.ac (host_libs): Drop unused libelf.

2 years agoAdd libgo dependency on libbacktrace.
Sergei Trofimovich [Thu, 18 Aug 2022 06:55:15 +0000 (07:55 +0100)]
Add libgo dependency on libbacktrace.

Noticed missing dependency when regenerated Makefile.in for unrelated
change with 'autogen Makefile.def'.

The change was lost in r12-6861-gaeac414923aa1e ("Revert "Fix PR 67102:
Add libstdc++ dependancy to libffi" [PR67102]").

/
* Makefile.in: Regenerate.

2 years agors6000: Add expand pattern for multiply-add (PR103109)
Haochen Gui [Thu, 18 Aug 2022 08:23:11 +0000 (16:23 +0800)]
rs6000: Add expand pattern for multiply-add (PR103109)

gcc/
PR target/103109
* config/rs6000/rs6000.md (<u>maddditi4): New pattern for multiply-add.
(<u>madddi4_highpart): New.
(<u>madddi4_highpart_le): New.

gcc/testsuite/
PR target/103109
* gcc.target/powerpc/pr103109.h: New.
* gcc.target/powerpc/pr103109-1.c: New.
* gcc.target/powerpc/pr103109-2.c: New.

2 years agoUse gimple_range_ssa_names in path_range_query.
Aldy Hernandez [Wed, 17 Aug 2022 23:05:08 +0000 (01:05 +0200)]
Use gimple_range_ssa_names in path_range_query.

gcc/ChangeLog:

* gimple-range-path.cc
(path_range_query::compute_exit_dependencies): Use
gimple_range_ssa_names.

2 years agoRISC-V: Add runtime invariant support
zhongjuzhe [Wed, 17 Aug 2022 07:19:50 +0000 (15:19 +0800)]
RISC-V: Add runtime invariant support

RISC-V 'V' Extension support scalable vector like ARM SVE.
To support RVV, we need to introduce runtime invariant.

- For zve32*, the runtime invariant uses 32-bit chunk.
- For zve64*, the runtime invariant uses 64-bit chunk.

[1] https://github.com/riscv/riscv-v-spec/blob/master/v-spec.adoc#sec-vector-extensions

This patch is preparing patch for RVV support.
Because we didn't introduce vector machine_mode yet, it safe to just change HOST_WIDE_INT into poly_int.
Also it safe to use "to_constant()" function for scalar operation.
This patch has been tested by full dejagnu regression.

gcc/ChangeLog:

* config/riscv/predicates.md: Adjust runtime invariant.
* config/riscv/riscv-modes.def (MAX_BITSIZE_MODE_ANY_MODE): New.
(NUM_POLY_INT_COEFFS): New.
* config/riscv/riscv-protos.h (riscv_initial_elimination_offset):Adjust
runtime invariant.
* config/riscv/riscv-sr.cc (riscv_remove_unneeded_save_restore_calls):
Adjust runtime invariant.
* config/riscv/riscv.cc (struct riscv_frame_info): Adjust runtime
invariant.
(enum riscv_microarchitecture_type): Ditto.
(riscv_valid_offset_p): Ditto.
(riscv_valid_lo_sum_p): Ditto.
(riscv_address_insns): Ditto.
(riscv_load_store_insns): Ditto.
(riscv_legitimize_move): Ditto.
(riscv_binary_cost): Ditto.
(riscv_rtx_costs): Ditto.
(riscv_output_move): Ditto.
(riscv_extend_comparands): Ditto.
(riscv_flatten_aggregate_field): Ditto.
(riscv_get_arg_info): Ditto.
(riscv_pass_by_reference): Ditto.
(riscv_elf_select_rtx_section): Ditto.
(riscv_stack_align): Ditto.
(riscv_compute_frame_info): Ditto.
(riscv_initial_elimination_offset): Ditto.
(riscv_set_return_address): Ditto.
(riscv_for_each_saved_reg): Ditto.
(riscv_first_stack_step): Ditto.
(riscv_expand_prologue): Ditto.
(riscv_expand_epilogue): Ditto.
(riscv_can_use_return_insn): Ditto.
(riscv_secondary_memory_needed): Ditto.
(riscv_hard_regno_nregs): Ditto.
(riscv_convert_vector_bits): New.
(riscv_option_override): Adjust runtime invariant.
(riscv_promote_function_mode): Ditto.
* config/riscv/riscv.h (POLY_SMALL_OPERAND_P): New.
(BITS_PER_RISCV_VECTOR): New.
(BYTES_PER_RISCV_VECTOR): New.
* config/riscv/riscv.md: Adjust runtime invariant.

2 years agoLoongArch: Get __tls_get_addr address through got table when disable plt.
Lulu Cheng [Thu, 18 Aug 2022 01:57:14 +0000 (09:57 +0800)]
LoongArch: Get __tls_get_addr address through got table when disable plt.

Fix bug, ICE with tls gd/ld var with -fno-plt.

gcc/ChangeLog:

* config/loongarch/loongarch.cc (loongarch_call_tls_get_addr):
Get __tls_get_addr address through got table when disable plt.

gcc/testsuite/ChangeLog:

* gcc.target/loongarch/tls-gd-noplt.c: New test.

2 years agoxtensa: Optimize stack pointer updates in function pro/epilogue under certain conditions
Takayuki 'January June' Suwa [Wed, 17 Aug 2022 19:31:16 +0000 (04:31 +0900)]
xtensa: Optimize stack pointer updates in function pro/epilogue under certain conditions

This patch enforces the use of "addmi" machine instruction instead of
addition/subtraction with two source registers for adjusting the stack
pointer, if the adjustment fits into a signed 16-bit and is also a multiple
of 256.

    /* example */
    void test(void) {
      char buffer[4096];
      __asm__(""::"m"(buffer));
    }

    ;; before
    test:
movi.n a9, 1
slli a9, a9, 12
sub sp, sp, a9
movi.n a9, 1
slli a9, a9, 12
add.n sp, sp, a9
addi sp, sp, 0
ret.n

    ;; after
    test:
addmi sp, sp, -0x1000
addmi sp, sp, 0x1000
ret.n

gcc/ChangeLog:

* config/xtensa/xtensa.cc (xtensa_expand_prologue):
Use an "addmi" machine instruction for updating the stack pointer
rather than addition/subtraction via hard register A9, if the amount
of change satisfies the literal value conditions of that instruction
when the CALL0 ABI is used.
(xtensa_expand_epilogue): Ditto.
And also inhibit the stack pointer addition of constant zero.

2 years agoDaily bump.
GCC Administrator [Thu, 18 Aug 2022 00:16:43 +0000 (00:16 +0000)]
Daily bump.

2 years agoRISC-V/testsuite: Restrict remaining `fmin'/`fmax' tests to hard float
Maciej W. Rozycki [Wed, 17 Aug 2022 22:22:18 +0000 (23:22 +0100)]
RISC-V/testsuite: Restrict remaining `fmin'/`fmax' tests to hard float

Complement commit 7915f6551343 ("RISC-V/testsuite: constraint some of
tests to hard_float") and also restrict the remaining `fmin'/`fmax'
tests to hard-float test configurations.

gcc/testsuite/
* gcc.target/riscv/fmax-snan.c: Add `dg-require-effective-target
hard_float'.
* gcc.target/riscv/fmaxf-snan.c: Likewise.
* gcc.target/riscv/fmin-snan.c: Likewise.
* gcc.target/riscv/fminf-snan.c: Likewise.

2 years ago[Committed] PR target/106640: Fix use of XINT in TImode compute_convert_gain.
Roger Sayle [Wed, 17 Aug 2022 21:23:33 +0000 (22:23 +0100)]
[Committed] PR target/106640: Fix use of XINT in TImode compute_convert_gain.

Thanks to Zdenek Sojka for reporting PR target/106640 where an RTL checking
build reveals a thinko in my recent patch to support TImode shifts/rotates
in STV.  My "senior moment" was to inappropriately use XINT where I should
be using INTVAL of XEXP.

2022-08-17  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
PR target/106640
* config/i386/i386-features.cc
(timde_scalar_chain::compute_convert_gain): Replace incorrect use
of XINT with INTVAL (XEXP (src, 1)).

2 years agoc++: Add new std::move test [PR67906]
Marek Polacek [Wed, 17 Aug 2022 19:43:45 +0000 (15:43 -0400)]
c++: Add new std::move test [PR67906]

As discussed in 67906, let's make sure we don't warn about a std::move
when initializing when there's a T(const T&&) ctor.

PR c++/67906

gcc/testsuite/ChangeLog:

* g++.dg/cpp0x/Wredundant-move11.C: New test.

2 years agoReset root oracle from path_oracle::reset_path.
Aldy Hernandez [Wed, 17 Aug 2022 15:47:21 +0000 (17:47 +0200)]
Reset root oracle from path_oracle::reset_path.

When we cross a backedge in the path solver, we reset the path
relations and nuke the root oracle.  However, we forget to reset it
for the next path.  This is causing us to miss threads because
subsequent paths will have no root oracle to use.

With this patch we get 201 more threads in the threadfull passes in my
.ii files and 118 more overall (DOM gets less because threadfull runs
before).

Normally, I'd recommend this for the GCC 12 branch, but considering
how sensitive other passes are to jump threading, and that there is no
PR associated with this, perhaps we should leave this out.  Up to the
release maintainers of course.

gcc/ChangeLog:

* gimple-range-path.cc
(path_range_query::compute_ranges_in_block): Remove
set_root_oracle call.
(path_range_query::compute_ranges): Pass ranger oracle to
reset_path.
* value-relation.cc (path_oracle::reset_path): Set root oracle.
* value-relation.h (path_oracle::reset_path): Add root oracle
argument.

2 years agoc++: Extend -Wredundant-move for const-qual objects [PR90428]
Marek Polacek [Wed, 17 Aug 2022 17:36:52 +0000 (13:36 -0400)]
c++: Extend -Wredundant-move for const-qual objects [PR90428]

In this PR, Jon suggested extending the -Wredundant-move warning
to warn when the user is moving a const object as in:

  struct T { };

  T f(const T& t)
  {
    return std::move(t);
  }

where the std::move is redundant, because T does not have
a T(const T&&) constructor (which is very unlikely).  Even with
the std::move, T(T&&) would not be used because it would mean
losing the const.  Instead, T(const T&) will be called.

I had to restructure the function a bit, but it's better now.  This patch
depends on my other recent patches to maybe_warn_pessimizing_move.

PR c++/90428

gcc/cp/ChangeLog:

* typeck.cc (can_do_rvo_p): Rename to ...
(can_elide_copy_prvalue_p): ... this.
(maybe_warn_pessimizing_move): Extend the
-Wredundant-move warning to warn about std::move on a
const-qualified object.

gcc/testsuite/ChangeLog:

* g++.dg/cpp0x/Wredundant-move1.C: Adjust dg-warning.
* g++.dg/cpp0x/Wredundant-move9.C: Likewise.
* g++.dg/cpp0x/Wredundant-move10.C: New test.

2 years agoc++: Tweak for -Wpessimizing-move in templates [PR89780]
Marek Polacek [Wed, 3 Aug 2022 23:47:30 +0000 (19:47 -0400)]
c++: Tweak for -Wpessimizing-move in templates [PR89780]

In my previous patches I've been extending our std::move warnings,
but this tweak actually dials it down a little bit.  As reported in
bug 89780, it's questionable to warn about expressions in templates
that were type-dependent, but aren't anymore because we're instantiating
the template.  As in,

  template <typename T>
  Dest withMove() {
    T x;
    return std::move(x);
  }

  template Dest withMove<Dest>(); // #1
  template Dest withMove<Source>(); // #2

Saying that the std::move is pessimizing for #1 is not incorrect, but
it's not useful, because removing the std::move would then pessimize #2.
So the user can't really win.  At the same time, disabling the warning
just because we're in a template would be going too far, I still want to
warn for

  template <typename>
  Dest withMove() {
    Dest x;
    return std::move(x);
  }

because the std::move therein will be pessimizing for any instantiation.

So I'm using the suppress_warning machinery to that effect.
Problem: I had to add a new group to nowarn_spec_t, otherwise
suppressing the -Wpessimizing-move warning would disable a whole bunch
of other warnings, which we really don't want.

PR c++/89780

gcc/cp/ChangeLog:

* pt.cc (tsubst_copy_and_build) <case CALL_EXPR>: Maybe suppress
-Wpessimizing-move.
* typeck.cc (maybe_warn_pessimizing_move): Don't issue warnings
if they are suppressed.
(check_return_expr): Disable -Wpessimizing-move when returning
a dependent expression.

gcc/ChangeLog:

* diagnostic-spec.cc (nowarn_spec_t::nowarn_spec_t): Handle
OPT_Wpessimizing_move and OPT_Wredundant_move.
* diagnostic-spec.h (nowarn_spec_t): Add NW_REDUNDANT enumerator.

gcc/testsuite/ChangeLog:

* g++.dg/cpp0x/Wpessimizing-move3.C: Remove dg-warning.
* g++.dg/cpp0x/Wredundant-move2.C: Likewise.

2 years agoc++: Extend -Wpessimizing-move to other contexts
Marek Polacek [Mon, 1 Aug 2022 21:02:23 +0000 (17:02 -0400)]
c++: Extend -Wpessimizing-move to other contexts

In my recent patch which enhanced -Wpessimizing-move so that it warns
about class prvalues too I said that I'd like to extend it so that it
warns in more contexts where a std::move can prevent copy elision, such
as:

  T t = std::move(T());
  T t(std::move(T()));
  T t{std::move(T())};
  T t = {std::move(T())};
  void foo (T);
  foo (std::move(T()));

This patch does that by adding two maybe_warn_pessimizing_move calls.
These must happen before we've converted the initializers otherwise the
std::move will be buried in a TARGET_EXPR.

PR c++/106276

gcc/cp/ChangeLog:

* call.cc (build_over_call): Call maybe_warn_pessimizing_move.
* cp-tree.h (maybe_warn_pessimizing_move): Declare.
* decl.cc (build_aggr_init_full_exprs): Call
maybe_warn_pessimizing_move.
* typeck.cc (maybe_warn_pessimizing_move): Handle TREE_LIST and
CONSTRUCTOR.  Add a bool parameter and use it.  Adjust a diagnostic
message.
(check_return_expr): Adjust the call to maybe_warn_pessimizing_move.

gcc/testsuite/ChangeLog:

* g++.dg/cpp0x/Wpessimizing-move7.C: Add dg-warning.
* g++.dg/cpp0x/Wpessimizing-move8.C: New test.

2 years agofortran: Add -static-libquadmath support [PR46539]
Jakub Jelinek [Wed, 17 Aug 2022 15:00:33 +0000 (17:00 +0200)]
fortran: Add -static-libquadmath support [PR46539]

The following patch is a revival of the
https://gcc.gnu.org/legacy-ml/gcc-patches/2014-10/msg00771.html
patch.  While trunk configured against recent glibc and with linker
--as-needed support doesn't really need to link against -lquadmath
anymore, there are still other targets where libquadmath is still in
use.
As has been discussed, making -static-libgfortran imply statically
linking both libgfortran and libquadmath is undesirable because of
the significant licensing differences between the 2 libraries.
Compared to the 2014 patch, this one doesn't handle -lquadmath
addition in the driver, which to me looks incorrect, libgfortran
configure determines where in libgfortran.spec -lquadmath should
be present if at all and with what it should be wrapped, but
analyzes gfortran -### -static-libgfortran stderr and based on
that figures out what gcc/configure.ac determined.

2022-08-17  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
    Jakub Jelinek  <jakub@redhat.com>

PR fortran/46539
gcc/
* common.opt (static-libquadmath): New option.
* gcc.cc (driver_handle_option): Always accept -static-libquadmath.
* config/darwin.h (LINK_SPEC): Handle -static-libquadmath.
gcc/fortran/
* lang.opt (static-libquadmath): New option.
* invoke.texi (-static-libquadmath): Document it.
* options.cc (gfc_handle_option): Error out if -static-libquadmath
is passed but we do not support it.
libgfortran/
* acinclude.m4 (LIBQUADSPEC): From $FC -static-libgfortran -###
output determine -Bstatic/-Bdynamic, -bstatic/-bdynamic,
-aarchive_shared/-adefault linker support or Darwin remapping
of -lgfortran to libgfortran.a%s and use that around or instead
of -lquadmath in LIBQUADSPEC.
* configure: Regenerated.

Co-Authored-By: Francois-Xavier Coudert <fxcoudert@gcc.gnu.org>
2 years agoFortran: OpenMP fix declare simd inside modules and absent linear step [PR106566]
Tobias Burnus [Wed, 17 Aug 2022 14:19:58 +0000 (16:19 +0200)]
Fortran: OpenMP fix declare simd inside modules and absent linear step [PR106566]

gcc/fortran/ChangeLog:

PR fortran/106566
* openmp.cc (gfc_match_omp_clauses): Fix setting linear-step value
to 1 when not specified.
(gfc_match_omp_declare_simd): Accept module procedures.

gcc/testsuite/ChangeLog:

PR fortran/106566
* gfortran.dg/gomp/declare-simd-4.f90: New test.
* gfortran.dg/gomp/declare-simd-5.f90: New test.
* gfortran.dg/gomp/declare-simd-6.f90: New test.

2 years agoOpenMP requires: Fix diagnostic filename corner case
Tobias Burnus [Wed, 17 Aug 2022 13:52:23 +0000 (15:52 +0200)]
OpenMP requires: Fix diagnostic filename corner case

The issue occurs when there is, e.g., main._omp_fn.0 in two files with
different OpenMP requires clauses.  The function entries in the offload
table ends up having the same decl tree and, hence, the diagnostic showed
the same filename for both.  Solution: Use the .o filename in this case.

Note that the issue does not occur with same-named 'static' functions and
without the fatal error from the requires diagnostic, there would be
later a linker error due to having two 'main'.

gcc/
* lto-cgraph.cc (input_offload_tables): Improve requires diagnostic
when filenames come out identically.

2 years agoOpenMP: Fix var replacement with 'simd' and linear-step vars [PR106548]
Tobias Burnus [Wed, 17 Aug 2022 13:45:56 +0000 (15:45 +0200)]
OpenMP: Fix var replacement with 'simd' and linear-step vars [PR106548]

gcc/ChangeLog:

PR middle-end/106548
* omp-low.cc (lower_rec_input_clauses): Use build_outer_var_ref
for 'simd' linear-step values that are variable.

libgomp/ChangeLog:

PR middle-end/106548
* testsuite/libgomp.c/linear-2.c: New test.

2 years agolibgomp/splay-tree.h: Fix splay_tree_prefix handling
Tobias Burnus [Wed, 17 Aug 2022 13:36:41 +0000 (15:36 +0200)]
libgomp/splay-tree.h: Fix splay_tree_prefix handling

When splay_tree_prefix is defined, the .h file
defines splay_* macros to add the prefix. However,
before those were only unset when additionally
splay_tree_c was defined.
Additionally, for consistency undefine splay_tree_c
also when no splay_tree_prefix is defined - there
is no interdependence either.

libgomp/ChangeLog:

* splay-tree.h: Fix splay_* macro unsetting if
splay_tree_prefix is defined.

2 years agoOpenMP/C++: Allow classes with static members to be mappable [PR104493]
Tobias Burnus [Wed, 17 Aug 2022 12:17:08 +0000 (14:17 +0200)]
OpenMP/C++: Allow classes with static members to be mappable [PR104493]

As this is the last lang-specific user of the omp_mappable_type hook,
the hook is removed, keeping only a generic omp_mappable_type for
incomplete types (or error_node).

PR c++/104493

gcc/c/ChangeLog:

* c-decl.cc (c_decl_attributes, finish_decl): Call omp_mappable_type
instead of removed langhook.
* c-typeck.cc (c_finish_omp_clauses): Likewise.

gcc/cp/ChangeLog:

* cp-objcp-common.h (LANG_HOOKS_OMP_MAPPABLE_TYPE): Remove.
* cp-tree.h (cp_omp_mappable_type, cp_omp_emit_unmappable_type_notes):
Remove.
* decl2.cc (cp_omp_mappable_type_1, cp_omp_mappable_type,
cp_omp_emit_unmappable_type_notes): Remove.
(cplus_decl_attributes): Call omp_mappable_type instead of
removed langhook.
* decl.cc (cp_finish_decl): Likewise; call cxx_incomplete_type_inform
in lieu of cp_omp_emit_unmappable_type_notes.
* semantics.cc (finish_omp_clauses): Likewise.

gcc/ChangeLog:

* gimplify.cc (omp_notice_variable): Call omp_mappable_type
instead of removed langhook.
* omp-general.h (omp_mappable_type): New prototype.
* omp-general.cc (omp_mappable_type):  New; moved from ...
* langhooks.cc (lhd_omp_mappable_type): ... here.
* langhooks-def.h (lhd_omp_mappable_type,
LANG_HOOKS_OMP_MAPPABLE_TYPE): Remove.
(LANG_HOOKS_FOR_TYPES_INITIALIZER): Remote the latter.
* langhooks.h (struct lang_hooks_for_types): Remove
omp_mappable_type.

gcc/testsuite/ChangeLog:

* g++.dg/gomp/unmappable-1.C: Remove dg-error; remove dg-note no
longer shown as TYPE_MAIN_DECL is NULL.
* c-c++-common/gomp/map-incomplete-type.c: New test.

Co-authored-by: Chung-Lin Tang <cltang@codesourcery.com>
2 years agoarm: Define with_float to hard when target name ends with hf
Christophe Lyon [Wed, 17 Aug 2022 08:06:24 +0000 (10:06 +0200)]
arm: Define with_float to hard when target name ends with hf

On arm, the --with-float= configure option is used to define include
files search path (among other things).  However, when targeting
arm-linux-gnueabihf, one would expect to automatically default to the
hard-float ABI, but this is not the case. As a consequence, GCC
bootstrap fails on an arm-linux-gnueabihf target if --with-float=hard
is not used.

This patch checks if the target name ends with 'hf' and defines
with_float to hard if not already defined.  This is achieved in
gcc/config.gcc, just before selecting the default CPU depending on the
$with_float value.

2022-08-17  Christophe Lyon  <christophe.lyon@arm.com>

gcc/
* config.gcc (arm): Define with_float to hard if target name ends
with 'hf'.

2 years agoRefactor back_threader_profitability
Richard Biener [Mon, 15 Aug 2022 13:38:09 +0000 (15:38 +0200)]
Refactor back_threader_profitability

The following refactors profitable_path_p in the backward threader,
splitting out parts that can be computed once the exit block is known,
parts that contiguously update and that can be checked allowing
for the path to be later identified as FSM with larger limits,
possibly_profitable_path_p, and final checks done when the whole
path is known, profitable_path_p.

I've removed the back_threader_profitability instance from the
back_threader class and instead instantiate it once per path
discovery.  I've kept the size compute non-incremental to simplify
the patch and not worry about unwinding.

There's key changes to previous behavior - namely we apply
the param_max_jump_thread_duplication_stmts early only when
we know the path cannot become an FSM one (multiway + thread through
latch) but make sure to elide the path query when we we didn't
yet discover that but are over this limit.  Similarly the
speed limit is now used even when we did not yet discover a
hot BB on the path.  Basically the idea is to only stop path
discovery when we know the path will never become profitable
but avoid the expensive path range query when we know it's
currently not.

I've done a few cleanups, merging functions, on the way.

* tree-ssa-threadbackward.cc
(back_threader_profitability): Split profitable_path_p
into possibly_profitable_path_p and itself, keep state
as new members.
(back_threader::m_profit): Remove.
(back_threader::find_paths): Likewise.
(back_threader::maybe_register_path): Take profitability
instance as parameter.
(back_threader::find_paths_to_names): Likewise.  Use
possibly_profitable_path_p and avoid the path range query
when the path is currently too large.
(back_threader::find_paths): Fold into ...
(back_threader::maybe_thread_block): ... this.
(get_gimple_control_stmt): Remove.
(back_threader_profitability::possibly_profitable_path_p):
Split out from profitable_path_p, do early profitability
checks.
(back_threader_profitability::profitable_path_p): Do final
profitability path after the taken edge has been determined.

2 years agoFix bug in emergency cxa pool free
Keef Aragon [Wed, 17 Aug 2022 06:45:15 +0000 (08:45 +0200)]
Fix bug in emergency cxa pool free

This probably has never actually affected anyone in practice. The normal
ABI implementation just uses malloc and only falls back to the pool on
malloc failure. But if that happens a bunch of times the freelist gets out
of order which violates some of the invariants of the freelist (as well as
the comments that follow the bug). The bug is just a comparison reversal
when traversing the freelist in the case where the pointer being returned
to the pool is after the existing freelist.

libstdc++-v3/
* libsupc++/eh_alloc.cc (pool::free): Inverse comparison.

2 years agoLoongArch: Provide fmin/fmax RTL pattern
Xi Ruoyao [Tue, 16 Aug 2022 07:34:36 +0000 (15:34 +0800)]
LoongArch: Provide fmin/fmax RTL pattern

We already had smin/smax RTL pattern using fmin/fmax instruction.  But
for smin/smax, it's unspecified what will happen if either operand is
NaN.  So we would generate calls to libc fmin/fmax functions with
-fno-finite-math-only (the default for all optimization levels expect
-Ofast).

But, LoongArch fmin/fmax instruction is IEEE-754-2008 conformant so we
can also use the instruction for fmin/fmax pattern and avoid the library
function call.

gcc/ChangeLog:

* config/loongarch/loongarch.md (fmax<mode>3): New RTL pattern.
(fmin<mode>3): Likewise.

gcc/testsuite/ChangeLog:

* gcc.target/loongarch/fmax-fmin.c: New test.

2 years agoAbstract interesting ssa-names from GORI.
Andrew MacLeod [Tue, 16 Aug 2022 17:18:37 +0000 (13:18 -0400)]
Abstract interesting ssa-names from GORI.

Provide a routine to pick out the ssa-names from interesting statements.

* gimple-range-fold.cc (gimple_range_ssa_names): New.
* gimple-range-fold.h (gimple_range_ssa_names): New prototype.
* gimple-range-gori.cc (range_def_chain::get_def_chain): Move
  code to new routine.

2 years agoDaily bump.
GCC Administrator [Wed, 17 Aug 2022 00:17:17 +0000 (00:17 +0000)]
Daily bump.

2 years agoc++: remove some xfails
Jason Merrill [Tue, 16 Aug 2022 19:19:46 +0000 (15:19 -0400)]
c++: remove some xfails

These tests are now passing.

gcc/testsuite/ChangeLog:

* g++.dg/warn/Wstringop-overflow-4.C: Only xfail for C++98.
* g++.target/i386/bfloat_cpp_typecheck.C: Remove xfail.

2 years agoc++: Fix pragma suppression of -Wc++20-compat diagnostics [PR106423]
Tom Honermann [Mon, 1 Aug 2022 18:49:00 +0000 (14:49 -0400)]
c++: Fix pragma suppression of -Wc++20-compat diagnostics [PR106423]

Gcc's '#pragma GCC diagnostic' directives are processed in "early mode"
(see handle_pragma_diagnostic_early) for the C++ frontend and, as such,
require that the target diagnostic option be enabled for the preprocessor
(see c_option_is_from_cpp_diagnostics).  This change modifies the
-Wc++20-compat option definition to register it as a preprocessor option
so that its associated diagnostics can be suppressed.  The changes also
implicitly disable the option in C++20 and later modes.  These changes
are consistent with the definition of the -Wc++11-compat option.

This support is motivated by the need to suppress the following diagnostic
otherwise issued in C++17 and earlier modes due to the char8_t typedef
present in the uchar.h header file in glibc 2.36.
  warning: identifier ‘char8_t’ is a keyword in C++20 [-Wc++20-compat]

Tests are added to validate suppression of both -Wc++11-compat and
-Wc++20-compat related diagnostics (fixes were only needed for the C++20
case).

PR c++/106423

gcc/c-family/ChangeLog:
* c-opts.cc (c_common_post_options): Disable -Wc++20-compat
diagnostics in C++20 and later.
* c.opt (Wc++20-compat): Enable hooks for the preprocessor.

gcc/cp/ChangeLog:
* parser.cc (cp_lexer_saving_tokens): Add comment regarding
diagnostic requirements.

gcc/testsuite/ChangeLog:
* g++.dg/cpp0x/keywords2.C: New test.
* g++.dg/cpp2a/keywords2.C: New test.

libcpp/ChangeLog:
* include/cpplib.h (cpp_warning_reason): Add CPP_W_CXX20_COMPAT.
* init.cc (cpp_create_reader): Add cpp_warn_cxx20_compat.

2 years agodocs: remove link to www.bullfreeware.com from install
Martin Liska [Tue, 16 Aug 2022 16:15:55 +0000 (18:15 +0200)]
docs: remove link to bullfreeware.com from install

As mentioned at https://gcc.gnu.org/PR106637#c2, the discontinued
providing binaries.

PR target/106637

gcc/ChangeLog:

* doc/install.texi: Remove link to www.bullfreeware.com

2 years agoRISC-V: Support zfh and zfhmin extension
Kito Cheng [Mon, 4 Jul 2022 15:09:57 +0000 (23:09 +0800)]
RISC-V: Support zfh and zfhmin extension

Zfh and Zfhmin are extensions for IEEE half precision, both are ratified
in Jan. 2022[1]:

- Zfh has full set of operation like F or D for single or double precision.
- Zfhmin has only provide minimal support for half precision operation,
  like conversion, load, store and move instructions.

[1] https://github.com/riscv/riscv-isa-manual/commit/b35a54079e0da11740ce5b1e6db999d1d5172768

gcc/ChangeLog:

* common/config/riscv/riscv-common.cc (riscv_implied_info): Add
zfh and zfhmin.
(riscv_ext_version_table): Ditto.
(riscv_ext_flag_table): Ditto.
* config/riscv/riscv-opts.h (MASK_ZFHMIN): New.
(MASK_ZFH): Ditto.
(TARGET_ZFHMIN): Ditto.
(TARGET_ZFH): Ditto.
* config/riscv/riscv.cc (riscv_output_move): Handle HFmode move
for zfh and zfhmin.
(riscv_emit_float_compare): Handle HFmode.
* config/riscv/riscv.md (ANYF): Add HF.
(SOFTF): Add HF.
(load): Ditto.
(store): Ditto.
(truncsfhf2): New.
(truncdfhf2): Ditto.
(extendhfsf2): Ditto.
(extendhfdf2): Ditto.
(*movhf_hardfloat): Ditto.
(*movhf_softfloat): Make sure not ZFHMIN.
* config/riscv/riscv.opt (riscv_zf_subext): New.

gcc/testsuite/ChangeLog:

* gcc.target/riscv/_Float16-zfh-1.c: New.
* gcc.target/riscv/_Float16-zfh-2.c: Ditto.
* gcc.target/riscv/_Float16-zfh-3.c: Ditto.
* gcc.target/riscv/_Float16-zfhmin-1.c: Ditto.
* gcc.target/riscv/_Float16-zfhmin-2.c: Ditto.
* gcc.target/riscv/_Float16-zfhmin-3.c: Ditto.
* gcc.target/riscv/arch-16.c: Ditto.
* gcc.target/riscv/arch-17.c: Ditto.
* gcc.target/riscv/predef-21.c: Ditto.
* gcc.target/riscv/predef-22.c: Ditto.

2 years agoRISC-V: Support _Float16 type.
Kito Cheng [Mon, 4 Jul 2022 12:17:47 +0000 (20:17 +0800)]
RISC-V: Support _Float16 type.

RISC-V decide use _Float16 as primary IEEE half precision type, and this
already become part of psABI, this patch has added folloing support for
_Float16:

- Soft-float support for _Float16.
- Make sure _Float16 available on C++ mode.
- Name mangling for _Float16 on C++ mode.

gcc/ChangeLog

* config/riscv/riscv-builtins.cc: include stringpool.h
(riscv_float16_type_node): New.
(riscv_init_builtin_types): Ditto.
(riscv_init_builtins): Call riscv_init_builtin_types.
* config/riscv/riscv-modes.def (HF): New.
* config/riscv/riscv.cc (riscv_output_move): Handle HFmode.
(riscv_mangle_type): New.
(riscv_scalar_mode_supported_p): Ditto.
(riscv_libgcc_floating_mode_supported_p): Ditto.
(riscv_excess_precision): Ditto.
(riscv_floatn_mode): Ditto.
(riscv_init_libfuncs): Ditto.
(TARGET_MANGLE_TYPE): Ditto.
(TARGET_SCALAR_MODE_SUPPORTED_P): Ditto.
(TARGET_LIBGCC_FLOATING_MODE_SUPPORTED_P): Ditto.
(TARGET_INIT_LIBFUNCS): Ditto.
(TARGET_C_EXCESS_PRECISION): Ditto.
(TARGET_FLOATN_MODE): Ditto.
* config/riscv/riscv.md (mode): Add HF.
(softload): Add HF.
(softstore): Ditto.
(fmt): Ditto.
(UNITMODE): Ditto.
(movhf): New.
(*movhf_softfloat): New.

libgcc/ChangeLog:

* config/riscv/sfp-machine.h (_FP_NANFRAC_H): New.
(_FP_NANFRAC_H): Ditto.
(_FP_NANSIGN_H): Ditto.
* config/riscv/t-softfp32 (softfp_extensions): Add HF related
routines.
(softfp_truncations): Ditto.
(softfp_extras): Ditto.
* config/riscv/t-softfp64 (softfp_extras): Add HF related routines.

gcc/testsuite/ChangeLog:

* g++.target/riscv/_Float16.C: New.
* gcc.target/riscv/_Float16-soft-1.c: Ditto.
* gcc.target/riscv/_Float16-soft-2.c: Ditto.
* gcc.target/riscv/_Float16-soft-3.c: Ditto.
* gcc.target/riscv/_Float16-soft-4.c: Ditto.
* gcc.target/riscv/_Float16.c: Ditto.

2 years agosoft-fp: Update soft-fp from glibc
Kito Cheng [Wed, 10 Aug 2022 13:54:24 +0000 (21:54 +0800)]
soft-fp: Update soft-fp from glibc

This patch is updating all soft-fp from glibc, most changes are
copyright years update, removing "Contributed by" lines and update URL for
license, and changes other than those update are adding conversion
function between IEEE half and 32-bit/64-bit integer, those functions are
required by RISC-V _Float16 support.

libgcc/ChangeLog:

* soft-fp/fixhfdi.c: New.
* soft-fp/fixhfsi.c: Likewise.
* soft-fp/fixunshfdi.c: Likewise.
* soft-fp/fixunshfsi.c: Likewise.
* soft-fp/floatdihf.c: Likewise.
* soft-fp/floatsihf.c: Likewise.
* soft-fp/floatundihf.c: Likewise.
* soft-fp/floatunsihf.c: Likewise.
* soft-fp/adddf3.c: Updating copyright years, removing "Contributed by"
lines and update URL for license.
* soft-fp/addsf3.c: Likewise.
* soft-fp/addtf3.c: Likewise.
* soft-fp/divdf3.c: Likewise.
* soft-fp/divsf3.c: Likewise.
* soft-fp/divtf3.c: Likewise.
* soft-fp/double.h: Likewise.
* soft-fp/eqdf2.c: Likewise.
* soft-fp/eqhf2.c: Likewise.
* soft-fp/eqsf2.c: Likewise.
* soft-fp/eqtf2.c: Likewise.
* soft-fp/extenddftf2.c: Likewise.
* soft-fp/extended.h: Likewise.
* soft-fp/extendhfdf2.c: Likewise.
* soft-fp/extendhfsf2.c: Likewise.
* soft-fp/extendhftf2.c: Likewise.
* soft-fp/extendhfxf2.c: Likewise.
* soft-fp/extendsfdf2.c: Likewise.
* soft-fp/extendsftf2.c: Likewise.
* soft-fp/extendxftf2.c: Likewise.
* soft-fp/fixdfdi.c: Likewise.
* soft-fp/fixdfsi.c: Likewise.
* soft-fp/fixdfti.c: Likewise.
* soft-fp/fixhfti.c: Likewise.
* soft-fp/fixsfdi.c: Likewise.
* soft-fp/fixsfsi.c: Likewise.
* soft-fp/fixsfti.c: Likewise.
* soft-fp/fixtfdi.c: Likewise.
* soft-fp/fixtfsi.c: Likewise.
* soft-fp/fixtfti.c: Likewise.
* soft-fp/fixunsdfdi.c: Likewise.
* soft-fp/fixunsdfsi.c: Likewise.
* soft-fp/fixunsdfti.c: Likewise.
* soft-fp/fixunshfti.c: Likewise.
* soft-fp/fixunssfdi.c: Likewise.
* soft-fp/fixunssfsi.c: Likewise.
* soft-fp/fixunssfti.c: Likewise.
* soft-fp/fixunstfdi.c: Likewise.
* soft-fp/fixunstfsi.c: Likewise.
* soft-fp/fixunstfti.c: Likewise.
* soft-fp/floatdidf.c: Likewise.
* soft-fp/floatdisf.c: Likewise.
* soft-fp/floatditf.c: Likewise.
* soft-fp/floatsidf.c: Likewise.
* soft-fp/floatsisf.c: Likewise.
* soft-fp/floatsitf.c: Likewise.
* soft-fp/floattidf.c: Likewise.
* soft-fp/floattihf.c: Likewise.
* soft-fp/floattisf.c: Likewise.
* soft-fp/floattitf.c: Likewise.
* soft-fp/floatundidf.c: Likewise.
* soft-fp/floatundisf.c: Likewise.
* soft-fp/floatunditf.c: Likewise.
* soft-fp/floatunsidf.c: Likewise.
* soft-fp/floatunsisf.c: Likewise.
* soft-fp/floatunsitf.c: Likewise.
* soft-fp/floatuntidf.c: Likewise.
* soft-fp/floatuntihf.c: Likewise.
* soft-fp/floatuntisf.c: Likewise.
* soft-fp/floatuntitf.c: Likewise.
* soft-fp/gedf2.c: Likewise.
* soft-fp/gesf2.c: Likewise.
* soft-fp/getf2.c: Likewise.
* soft-fp/half.h: Likewise.
* soft-fp/ledf2.c: Likewise.
* soft-fp/lesf2.c: Likewise.
* soft-fp/letf2.c: Likewise.
* soft-fp/muldf3.c: Likewise.
* soft-fp/mulsf3.c: Likewise.
* soft-fp/multf3.c: Likewise.
* soft-fp/negdf2.c: Likewise.
* soft-fp/negsf2.c: Likewise.
* soft-fp/negtf2.c: Likewise.
* soft-fp/op-1.h: Likewise.
* soft-fp/op-2.h: Likewise.
* soft-fp/op-4.h: Likewise.
* soft-fp/op-8.h: Likewise.
* soft-fp/op-common.h: Likewise.
* soft-fp/quad.h: Likewise.
* soft-fp/single.h: Likewise.
* soft-fp/soft-fp.h: Likewise.
* soft-fp/subdf3.c: Likewise.
* soft-fp/subsf3.c: Likewise.
* soft-fp/subtf3.c: Likewise.
* soft-fp/truncdfhf2.c: Likewise.
* soft-fp/truncdfsf2.c: Likewise.
* soft-fp/truncsfhf2.c: Likewise.
* soft-fp/trunctfdf2.c: Likewise.
* soft-fp/trunctfhf2.c: Likewise.
* soft-fp/trunctfsf2.c: Likewise.
* soft-fp/trunctfxf2.c: Likewise.
* soft-fp/truncxfhf2.c: Likewise.
* soft-fp/unorddf2.c: Likewise.
* soft-fp/unordsf2.c: Likewise.
* soft-fp/unordtf2.c: Likewise.

2 years agoStop backwards thread discovery when leaving a loop
Richard Biener [Tue, 16 Aug 2022 12:14:55 +0000 (14:14 +0200)]
Stop backwards thread discovery when leaving a loop

The backward threader copier cannot deal with the situation of
copying blocks belonging to different loops and will reject those
paths late.  The following uses this to prune path discovery,
saving on compile-time.  Note the off-loop block is still considered
as entry edge origin.

* tree-ssa-threadbackward.cc (back_threader::find_paths_to_names):
Do not walk further if we are leaving the current loop.

2 years agodriver: fix environ corruption after putenv() [PR106624]
Sergei Trofimovich [Tue, 16 Aug 2022 11:35:07 +0000 (12:35 +0100)]
driver: fix environ corruption after putenv() [PR106624]

The bug appeared afte r13-2010-g1270ccda70ca09 "Factor out
jobserver_active_p" slightly changed `putenv()` use from allocating
to non-allocating:

    -xputenv (concat ("MAKEFLAGS=", dup, NULL));
    +xputenv (jinfo.skipped_makeflags.c_str ());

`xputenv()` (and `putenv()`) don't copy strings and only store the
pointer in the `environ` global table. As a result `environ` got
corrupted as soon as `jinfo.skipped_makeflags` store got deallocated.

This started causing bootstrap crashes in `execv()` calls:

    xgcc: fatal error: cannot execute '/build/build/./prev-gcc/collect2': execv: Bad address

The change restores memory allocation for `xputenv()` argument.

gcc/

PR driver/106624
* gcc.cc (driver::detect_jobserver): Allocate storage xputenv()
argument using xstrdup().

2 years agoc++: Implement P2327R1 - De-deprecating volatile compound operations
Jakub Jelinek [Tue, 16 Aug 2022 11:15:32 +0000 (13:15 +0200)]
c++: Implement P2327R1 - De-deprecating volatile compound operations

From what I can see, this has been voted in as a DR and as it means
we warn less often than before in -std={gnu,c}++2{0,3} modes or with
-Wvolatile, I wonder if it shouldn't be backported to affected release
branches as well.

2022-08-16  Jakub Jelinek  <jakub@redhat.com>

* typeck.cc (cp_build_modify_expr): Implement
P2327R1 - De-deprecating volatile compound operations.  Don't warn
for |=, &= or ^= with volatile lhs.
* expr.cc (mark_use) <case MODIFY_EXPR>: Adjust warning wording,
leave out simple.

* g++.dg/cpp2a/volatile1.C: Adjust for de-deprecation of volatile
compound |=, &= and ^= operations.
* g++.dg/cpp2a/volatile3.C: Likewise.
* g++.dg/cpp2a/volatile5.C: Likewise.

2 years agod: Update DIP links in gdc documentation to point at upstream repository
Iain Buclaw [Tue, 16 Aug 2022 10:22:10 +0000 (12:22 +0200)]
d: Update DIP links in gdc documentation to point at upstream repository

The wiki links probably worked at some point in the distant past, but
now the official location of tracking all D Improvement Proposals is on
the upstream dlang/DIPs GitHub repository.

PR d/106638

gcc/d/ChangeLog:

* gdc.texi: Update DIP links to point at upstream dlang/DIPs
repository.

2 years agoRename imports nomenclature in path_range_query to exit_dependencies.
Aldy Hernandez [Tue, 16 Aug 2022 08:52:37 +0000 (10:52 +0200)]
Rename imports nomenclature in path_range_query to exit_dependencies.

The purpose of this change is to disambiguate the imports name with
its use in GORI.

gcc/ChangeLog:

* gimple-range-path.cc (path_range_query::import_p): Rename to...
(path_range_query::exit_dependency_p): ...this.
(path_range_query::dump): Rename imports to exit dependencies.
(path_range_query::compute_ranges_in_phis): Same.
(path_range_query::compute_ranges_in_block): Same.
(path_range_query::adjust_for_non_null_uses): Same.
(path_range_query::compute_ranges): Same.
(path_range_query::compute_phi_relations): Same.
(path_range_query::add_to_imports): Rename to...
(path_range_query::add_to_exit_dependencies): ...this.
(path_range_query::compute_imports): Rename to...
(path_range_query::compute_exit_dependencies): ...this.
* gimple-range-path.h (class path_range_query): Rename imports to
exit dependencies.

2 years agoVR: mitigate -Wfinal-dtor-non-final-class clang warnings
Martin Liska [Tue, 16 Aug 2022 09:23:43 +0000 (11:23 +0200)]
VR: mitigate -Wfinal-dtor-non-final-class clang warnings

Fixes:

gcc/value-range-storage.h:129:40: warning: class with destructor marked 'final' cannot be inherited from [-Wfinal-dtor-non-final-class]
gcc/value-range-storage.h:146:36: warning: class with destructor marked 'final' cannot be inherited from [-Wfinal-dtor-non-final-class]

gcc/ChangeLog:

* value-range-storage.h (class obstack_vrange_allocator): Mark
  the class as final.
(class ggc_vrange_allocator): Likewise.

2 years agoVR: add more virtual dtors
Martin Liska [Tue, 16 Aug 2022 09:15:30 +0000 (11:15 +0200)]
VR: add more virtual dtors

Add 2 virtual destructors in order to address:

gcc/alloc-pool.h:522:5: warning: destructor called on non-final 'value_range_equiv' that has virtual functions but non-virtual destructor [-Wdelete-non-abstract-non-virtual-dtor]
gcc/ggc.h:166:3: warning: destructor called on non-final 'int_range<1>' that has virtual functions but non-virtual destructor [-Wdelete-non-abstract-non-virtual-dtor]

gcc/ChangeLog:

* value-range-equiv.h (class value_range_equiv): Add virtual
  destructor.
* value-range.h: Likewise.

2 years agomiddle-end/106630 - avoid ping-pong between extract_muldiv and match.pd
Richard Biener [Tue, 16 Aug 2022 07:43:24 +0000 (09:43 +0200)]
middle-end/106630 - avoid ping-pong between extract_muldiv and match.pd

The following avoids ping-pong between the match.pd pattern changing
(sizetype) ((a_9 + 1) * 48) to (sizetype)(a_9 + 1) * 48 and
extract_muldiv performing the reverse transform by restricting the
match.pd pattern to narrowing conversions as the comment indicates.

PR middle-end/106630
* match.pd ((T)(x * CST) -> (T)x * CST): Restrict to
narrowing conversions.

* gcc.dg/torture/pr106630.c: New testcase.

2 years agoVR: add missing override keyworks
Martin Liska [Tue, 16 Aug 2022 09:06:04 +0000 (11:06 +0200)]
VR: add missing override keyworks

Address:

gcc/value-range-equiv.h:57:8: warning: 'set_undefined' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]
gcc/value-range-equiv.h:58:8: warning: 'set_varying' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]

gcc/ChangeLog:

* value-range-equiv.h (class value_range_equiv):

2 years agoanalyzer: add more final override keywords
Martin Liska [Tue, 16 Aug 2022 09:00:14 +0000 (11:00 +0200)]
analyzer: add more final override keywords

gcc/analyzer/ChangeLog:

* region-model.cc: Fix -Winconsistent-missing-override clang
  warning.
* region.h: Likewise.

2 years agoi386: add 'final' and 'override' to scalar_chain
Martin Liska [Tue, 16 Aug 2022 08:53:00 +0000 (10:53 +0200)]
i386: add 'final' and 'override' to scalar_chain

In c3ed9e0d6e96d8697e4bab994f8acbc5506240ee, David added some
"final override" and since that there are 2 new warnings that
need the same treatment:

gcc/config/i386/i386-features.h:186:8: warning: 'convert_op' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]
gcc/config/i386/i386-features.h:186:8: warning: 'convert_op' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]
gcc/config/i386/i386-features.h:199:8: warning: 'convert_op' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]
gcc/config/i386/i386-features.h:199:8: warning: 'convert_op' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]

gcc/ChangeLog:

* config/i386/i386-features.h (class general_scalar_chain): Add
  final override for a method.
(class timode_scalar_chain): Likewise.

2 years agodocs: fix link destination
Martin Liska [Tue, 16 Aug 2022 08:04:36 +0000 (10:04 +0200)]
docs: fix link destination

gcc/fortran/ChangeLog:

* gfortran.texi: Fix link destination to a valid URL.

2 years agoAdjust max-jump-thread-paths docs
Richard Biener [Tue, 16 Aug 2022 07:51:16 +0000 (09:51 +0200)]
Adjust max-jump-thread-paths docs

The following fixes spelling and changes edge degree for number of
incoming edges.

* doc/invoke.texi (max-jump-thread-paths): Adjust.

2 years agojobserver: fix fifo mode by opening pipe in proper mode
Martin Liska [Tue, 16 Aug 2022 06:22:29 +0000 (08:22 +0200)]
jobserver: fix fifo mode by opening pipe in proper mode

The current jobserver_info relies on non-blocking FDs,
thus one the pipe in such mode.

gcc/ChangeLog:

* opts-common.cc (jobserver_info::connect): Open fifo
in non-blocking mode.

2 years agors6000: Adjust mov optabs for opaque modes [PR103353]
Kewen.Lin [Tue, 16 Aug 2022 05:24:07 +0000 (00:24 -0500)]
rs6000: Adjust mov optabs for opaque modes [PR103353]

As PR103353 shows, we may want to continue to expand built-in
function __builtin_vsx_lxvp, even if we have already emitted
error messages about some missing required conditions.  As
shown in that PR, without one explicit mov optab on OOmode
provided, it would call emit_move_insn recursively.

So this patch is to allow the mov pattern to be generated during
expanding phase if compiler has already seen errors.

PR target/103353

gcc/ChangeLog:

* config/rs6000/mma.md (define_expand movoo): Move TARGET_MMA condition
check to preparation statements and add handlings for !TARGET_MMA.
(define_expand movxo): Likewise.

gcc/testsuite/ChangeLog:

* gcc.target/powerpc/pr103353.c: New test.

2 years agovect: Don't allow vect_emulated_vector_p type in vectorizable_call [PR106322]
Kewen Lin [Tue, 16 Aug 2022 05:18:51 +0000 (00:18 -0500)]
vect: Don't allow vect_emulated_vector_p type in vectorizable_call [PR106322]

As PR106322 shows, in some cases for some vector type whose
TYPE_MODE is a scalar integral mode instead of a vector mode,
it's possible to obtain wrong target support information when
querying with the scalar integral mode.  For example, for the
test case in PR106322, on ppc64 32bit vectorizer gets vector
type "vector(2) short unsigned int" for scalar type "short
unsigned int", its mode is SImode instead of V2HImode.  The
target support querying checks umul_highpart optab with SImode
and considers it's supported, then vectorizer further generates
.MULH IFN call for that vector type.  Unfortunately it's wrong
to use SImode support for that vector type multiply highpart
here.

This patch is to teach vectorizable_call analysis not to allow
vect_emulated_vector_p type for both vectype_in and vectype_out
as Richi suggested.

PR tree-optimization/106322

gcc/ChangeLog:

* tree-vect-stmts.cc (vectorizable_call): Don't allow
vect_emulated_vector_p type for both vectype_in and vectype_out.

gcc/testsuite/ChangeLog:

* gcc.target/i386/pr106322.c: New test.
* gcc.target/powerpc/pr106322.c: New test.

2 years agoxtensa: Turn on -fsplit-wide-types-early by default
Takayuki 'January June' Suwa [Sun, 14 Aug 2022 09:29:33 +0000 (18:29 +0900)]
xtensa: Turn on -fsplit-wide-types-early by default

Since GCC10, the "subreg2" optimization pass was no longer tied to enabling
"subreg1" unless -fsplit-wide-types-early was turned on (PR88233).  However
on the Xtensa port, the lack of "subreg2" can degrade the quality of the
output code, especially for those that produce many D[FC]mode pseudos.

This patch turns on -fsplit-wide-types-early by default in order to restore
the previous behavior.

gcc/ChangeLog:

* common/config/xtensa/xtensa-common.cc
(xtensa_option_optimization_table): Add OPT_fsplit_wide_types_early
for OPT_LEVELS_ALL in order to restore pre-GCC10 behavior.

2 years agoDaily bump.
GCC Administrator [Tue, 16 Aug 2022 00:17:23 +0000 (00:17 +0000)]
Daily bump.

2 years agod: Defer compiling inline definitions until after the module has finished.
Iain Buclaw [Mon, 15 Aug 2022 17:00:43 +0000 (19:00 +0200)]
d: Defer compiling inline definitions until after the module has finished.

This is to prevent the case of when generating the methods of a struct
type, we don't accidentally emit an inline function that references it,
as the outer struct itself would still be incomplete.

gcc/d/ChangeLog:

* d-tree.h (d_defer_declaration): Declare.
* decl.cc (function_needs_inline_definition_p): Defer checking
DECL_UNINLINABLE and DECL_DECLARED_INLINE_P.
(maybe_build_decl_tree): Call d_defer_declaration instead of
build_decl_tree.
* modules.cc (deferred_inline_declarations): New variable.
(build_module_tree): Set deferred_inline_declarations and a handle
declarations pushed to it.
(d_defer_declaration): New function.

2 years agod: Fix internal compiler error: Segmentation fault at gimple-expr.cc:88
Iain Buclaw [Mon, 15 Aug 2022 15:51:03 +0000 (17:51 +0200)]
d: Fix internal compiler error: Segmentation fault at gimple-expr.cc:88

Because complex types are deprecated in the language, the new way to
expose native complex types is by defining an enum with a basetype of a
library-defined struct that is implicitly treated as-if it is native.
As casts are not implicitly added by the front-end when downcasting from
enum to its underlying type, we must insert an explicit cast during the
code generation pass.

PR d/106623

gcc/d/ChangeLog:

* d-codegen.cc (underlying_complex_expr): New function.
(d_build_call): Handle passing native complex objects as the
library-defined equivalent.
* d-tree.h (underlying_complex_expr): Declare.
* expr.cc (ExprVisitor::visit (DotVarExp *)): Call
underlying_complex_expr instead of build_vconvert.

gcc/testsuite/ChangeLog:

* gdc.dg/torture/pr106623.d: New test.

2 years agod: Build internal TypeInfo types when module name is "object"
Iain Buclaw [Mon, 15 Aug 2022 11:43:29 +0000 (13:43 +0200)]
d: Build internal TypeInfo types when module name is "object"

If for whatever reason the module declaration doesn't exist in the
object file, ensure that the internal definitions for TypeInfo and
TypeInfo_Class are still created, otherwise an ICE could occur later if
they are required for a run-time helper call.

gcc/d/ChangeLog:

* d-compiler.cc (Compiler::onParseModule): Call create_tinfo_types
when module name is object.
* typeinfo.cc (create_tinfo_types): Add guard for multiple
invocations.

2 years agod: Field names of anonymous delegates should be same as regular delegate types.
Iain Buclaw [Mon, 15 Aug 2022 07:54:15 +0000 (09:54 +0200)]
d: Field names of anonymous delegates should be same as regular delegate types.

Doesn't change anything in the code generation or ABI, but makes it
consistent with regular delegates as names would match up when
inspecting tree dumps.

gcc/d/ChangeLog:

* d-codegen.cc (build_delegate_cst): Give anonymous delegate field
names same as per ABI spec.

2 years agoanalyzer: fix direction of -Wanalyzer-out-of-bounds note [PR106626]
David Malcolm [Mon, 15 Aug 2022 18:47:03 +0000 (14:47 -0400)]
analyzer: fix direction of -Wanalyzer-out-of-bounds note [PR106626]

Fix a read/write typo.

Also, add more test coverage of -Wanalyzer-out-of-bounds to help
establish a baseline for experiments on tweaking the wording of
the warning (PR analyzer/106626).

gcc/analyzer/ChangeLog:
PR analyzer/106626
* region-model.cc (buffer_overread::emit): Fix copy&paste error in
direction of the access in the note.

gcc/testsuite/ChangeLog:
PR analyzer/106626
* gcc.dg/analyzer/out-of-bounds-read-char-arr.c: New test.
* gcc.dg/analyzer/out-of-bounds-read-int-arr.c: New test.
* gcc.dg/analyzer/out-of-bounds-write-char-arr.c: New test.
* gcc.dg/analyzer/out-of-bounds-write-int-arr.c: New test.

Signed-off-by: David Malcolm <dmalcolm@redhat.com>
2 years agoanalyzer: better fix for -Wanalyzer-use-of-uninitialized-value [PR106573]
David Malcolm [Mon, 15 Aug 2022 18:47:02 +0000 (14:47 -0400)]
analyzer: better fix for -Wanalyzer-use-of-uninitialized-value [PR106573]

gcc/analyzer/ChangeLog:
PR analyzer/106573
* region-model.cc (region_model::on_call_pre): Use check_call_args
when ensuring that we call get_arg_svalue on all args.  Remove
redundant call from handling for stdio builtins.

Signed-off-by: David Malcolm <dmalcolm@redhat.com>
2 years agoCheck for undefined and varying first.
Andrew MacLeod [Mon, 15 Aug 2022 14:16:23 +0000 (10:16 -0400)]
Check for undefined and varying first.

Rearrange order in irange:set to ensure all POLY_INTs map to varying.

PR tree-optimization/106621
gcc/
* value-range.cc (irange::set): Check for POLY_INT_CST early.

gcc/testsuite/
* gcc.dg/pr106621.c

2 years agoanalyzer: fix for ICE in sm-fd.cc [PR106551]
Immad Mir [Mon, 15 Aug 2022 16:51:27 +0000 (22:21 +0530)]
analyzer: fix for ICE in sm-fd.cc [PR106551]

This patch fixes the ICE caused by valid_to_unchecked_state
in sm-fd.cc by exiting early if first argument of any "dup"
functions is invalid.

gcc/analyzer/ChangeLog:
PR analyzer/106551
* sm-fd.cc (check_for_dup): exit early if first
argument is invalid for all dup functions.

gcc/testsuite/ChangeLog:
PR analyzer/106551
* gcc.dg/analyzer/fd-dup-1.c: New testcase.

Signed-off-by: Immad Mir <mirimmad@outlook.com>
2 years agoSupport shifts and rotates by integer constants in TImode STV on x86_64.
Roger Sayle [Mon, 15 Aug 2022 16:46:38 +0000 (17:46 +0100)]
Support shifts and rotates by integer constants in TImode STV on x86_64.

This patch adds support for converting 128-bit TImode shifts and rotates
to SSE equivalents using V1TImode during the TImode STV pass.
Previously, only logical shifts by multiples of 8 were handled
(from my patch earlier this month).

As an example of the benefits, the following rotate by 32-bits:

unsigned __int128 a, b;
void rot32() { a = (b >> 32) | (b << 96); }

when compiled on x86_64 with -O2 previously generated:

        movq    b(%rip), %rax
        movq    b+8(%rip), %rdx
        movq    %rax, %rcx
        shrdq   $32, %rdx, %rax
        shrdq   $32, %rcx, %rdx
        movq    %rax, a(%rip)
        movq    %rdx, a+8(%rip)
        ret

with this patch, now generates:

        movdqa  b(%rip), %xmm0
        pshufd  $57, %xmm0, %xmm0
        movaps  %xmm0, a(%rip)
        ret

[which uses a V4SI permutation for those that don't read SSE].
This should help 128-bit cryptography codes, that interleave XORs
with rotations (but that don't use additions or subtractions).

2022-08-15  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
* config/i386/i386-features.cc
(timode_scalar_chain::compute_convert_gain): Provide costs for
shifts and rotates.
(timode_scalar_chain::convert_insn): Handle ASHIFTRT, ROTATERT
and ROTATE just like existing ASHIFT and LSHIFTRT cases.
(timode_scalar_to_vector_candidate_p): Handle all shifts and
rotates by integer constants between 0 and 127.

gcc/testsuite/ChangeLog
* gcc.target/i386/sse4_1-stv-9.c: New test case.

2 years agoImproved gain calculation for COMPARE to 0 or -1 in TImode STV on x86_64.
Roger Sayle [Mon, 15 Aug 2022 16:43:02 +0000 (17:43 +0100)]
Improved gain calculation for COMPARE to 0 or -1 in TImode STV on x86_64.

This patch tweaks timode_scalar_chain::compute_convert_gain to provide
more accurate costs for converting TImode comparisons against zero or
minus 1 to V1TImode equivalents.

2022-08-15  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
* config/i386/i386-features.cc
(timode_scalar_chain::compute_convert_gain): Provide gains for
comparisons against 0/-1, including "*testti" patterns.

2 years agoPR tree-optimization/64992: (B << 2) != 0 is B when B is Boolean.
Roger Sayle [Mon, 15 Aug 2022 16:39:47 +0000 (17:39 +0100)]
PR tree-optimization/64992: (B << 2) != 0 is B when B is Boolean.

This patch resolves both PR tree-optimization/64992 and PR
tree-optimization/98956 which are missed optimization enhancement
request, for which Andrew Pinski already has a proposed solution
(related to a fix for PR tree-optimization/98954).  Yesterday,
I proposed an alternate improved patch for PR98954, which although
superior in most respects, alas didn't address this case [which
doesn't include a BIT_AND_EXPR], hence this follow-up fix.

For many functions, F(B), of a (zero-one) Boolean value B, the
expression F(B) != 0 can often be simplified to just B.  Hence
"(B * 5) != 0" is B, "-B != 0" is B, "bswap(B) != 0" is B,
"(B >>r 3) != 0" is B.  These are all currently optimized by GCC,
with the strange exception of left shifts by a constant (possibly
due to the undefined/implementation defined behaviour when the
shift constant is larger than the first operand's precision).
This patch adds support for this particular case, when the shift
constant is valid.

2022-08-15  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
PR tree-optimization/64992
PR tree-optimization/98956
* match.pd (ne (lshift @0 @1) 0): Simplify (X << C) != 0 to X
when X is zero_one_valued_p and the shift constant C is valid.
(eq (lshift @0 @1) 0): Likewise, simplify (X << C) == 0 to !X
when X is zero_one_valued_p and the shift constant C is valid.

gcc/testsuite/ChangeLog
PR tree-optimization/64992
* gcc.dg/pr64992.c: New test case.

2 years agoPR tree-optimization/71343: Optimize (X<<C)&(Y<<C) as (X&Y)<<C.
Roger Sayle [Mon, 15 Aug 2022 16:32:26 +0000 (17:32 +0100)]
PR tree-optimization/71343: Optimize (X<<C)&(Y<<C) as (X&Y)<<C.

This patch is the first part of a solution to PR tree-optimization/71343,
a missed-optimization enhancement request where GCC fails to see that
(a<<2)+(b<<2) == a*4+b*4.

This piece is that (X<<C) op (Y<<C) can be simplified to (X op Y) << C,
for many binary operators, including AND, IOR, XOR, and (if overflow
isn't an issue) PLUS and MINUS.  Likewise, the right shifts (both logical
and arithmetic) and bit-wise logical operators can be simplified in a
similar fashion.  These all reduce the number of GIMPLE binary operations
from 3 to 2, by combining/eliminating a shift operation.

2022-08-15  Roger Sayle  <roger@nextmovesoftware.com>
    Richard Biener  <rguenther@suse.de>

gcc/ChangeLog
PR tree-optimization/71343
* match.pd (op (lshift @0 @1) (lshift @2 @1)): Optimize the
expression (X<<C) + (Y<<C) to (X+Y)<<C for multiple operators.
(op (rshift @0 @1) (rshift @2 @1)): Likewise, simplify (X>>C)^(Y>>C)
to (X^Y)>>C for binary logical operators, AND, IOR and XOR.

gcc/testsuite/ChangeLog
PR tree-optimization/71343
* gcc.dg/pr71343-1.c: New test case.

2 years agoc++: Fix module line no testcase
Nathan Sidwell [Mon, 15 Aug 2022 14:19:36 +0000 (07:19 -0700)]
c++: Fix module line no testcase

Not all systems have the same injected headers, leading to line
location table differences that are immaterial to the test.  Fix the
regexp more robustly.

gcc/testsuite/
* g++.dg/modules/loc-prune-4.C: Adjust regexp

2 years agoc++: Extend -Wpessimizing-move for class prvalues [PR106276]
Marek Polacek [Fri, 22 Jul 2022 22:10:30 +0000 (18:10 -0400)]
c++: Extend -Wpessimizing-move for class prvalues [PR106276]

We already have a warning that warns about pessimizing std::move
in a return statement, when it prevents the NRVO:

  T fn()
  {
    T t;
    return std::move (t); // warning \o/
  }

However, the warning doesn't warn when what we are returning is a class
prvalue, that is, when std::move prevents the RVO:

  T fn()
  {
    T t;
    return std::move (T{}); // no warning :-(
  }

This came up recently in GCC:
<https://gcc.gnu.org/pipermail/gcc-patches/2022-July/598177.html>.

This patch fixes that.  I would like to extend the warning further, so
that it warns in more contexts, e.g.:

  T t = std::move(T());

or

  void foo (T);
  foo (std::move(T()));

PR c++/106276

gcc/cp/ChangeLog:

* typeck.cc (can_do_rvo_p): New.
(maybe_warn_pessimizing_move): Warn when moving a temporary object
in a return statement prevents copy elision.

gcc/testsuite/ChangeLog:

* g++.dg/cpp0x/Wpessimizing-move7.C: New test.

2 years agoSimplify range_on_path_entry
Richard Biener [Mon, 15 Aug 2022 11:03:54 +0000 (13:03 +0200)]
Simplify range_on_path_entry

I've noticed that range_on_path_entry does mightly complicated things
that don't make sense to me and the commentary might just be
out of date.  For the sake of it I replaced it with range_on_entry
and statistics show we thread _more_ jumps with that, so better
not do magic there.

* gimple-range-path.cc (range_on_path_entry): Just
call range_on_entry.

2 years agoi386 PIE: testsuite: cope with default pie on ia32
Alexandre Oliva [Mon, 15 Aug 2022 12:21:33 +0000 (09:21 -0300)]
i386 PIE: testsuite: cope with default pie on ia32

This patch continues the effort of cleaning up the testsuite for
--enable-default-pie; the focus herein is mostly 32-bit x86.

As much as I tried to avoid it, most of the changes to the testsuite
simply disable PIC/PIE, for reasons I'm going to detail below.

static-cdtor1.C gets new patterns to match PIE output.  Some
avx512fp16 tests change only in register allocation, because of the
register used to hold the GOT base address.  Interrupt tests changed
in this regard as well, but here it also affected register saving and
restoring.

The previous patch modified cet-sjlj tests, mentioning a single regexp
covering PIC and nonPIC got incorrect match counts.  I found out that
adding ?: to parenthesized subpatterns avoids miscounting matches.
Other tests that count certain kinds of insns needed adjustment over
insns in get_pc_thunk, extra loads from the GOT, or extra adds to
compute addresses.  In one case, namely stack-check-12, it is nonPIC
that had extra insns, that PIC gets rid of, or rather, pushing and
popping the PIC register obviates the dummy push and matching pop used
for stack probing in nonpic.

pr95126 tests were supposed to optimize loads into known constants,
but the @GOTOFF addresses prevent that for reasons I have not
investigated, but that would be clearly desirable, so I've XFAILed
these.  pr95852 is another case of missed optimization: sibcalls are
not possible when the PIC register needs to be set up for the call,
which prevents the expected constant propagation to the return block;
I have adjusted the codegen expectations of these tests.

As for tests that disable PIE...  Some are judgment calls, that fail
for similar reasons as tests described above, but I chose not to
adjust their expectations; others are just not possible with PIC, or
not worth the effort of adjusting.

anon[14].C check for no global or comdat symbols, respectively, but
-fPIE outputs get_pc_thunk, as global hidden comdat.
initlist-const1.C wants .rodata and checks for no .data, but PIC
outputs constant data that needs relocations in .data.rel.ro.local.
no-stack-protector-attr-3.C and stackprotectexplicit2.C count
stack_check_fail matches; -fPIE calls stack_check_fail_local instead,
which matches the pattern, but this symbol is also marked as .hidden,
so the match count needs to be adjusted.

pr71694.C checks for no movl, but get_pc_thunk contains one.
pr102892-1.c is a missed optimization, ivopts creates an induction
variable because the array address can't be part of an indexing base
address with PIE, and that ends up stopping a load from being resolved
to a constant as expected.  sibcall-11.c needs @PLT for the call,
which requires the PIC register, which makes sibcalling impossible.
builtin-self.c, in turn, expects no calls, but PIC calls get_pc_thunk.

avx* vector tests that had PIE disabled were affected in that the need
for GOT-based addressing modes changed instruction selection in ways
that deviated from the expectations of the tests.  Ditto other vector
tests: pr100865*, pr101796-1, pr101846, pr101989-broadcast-1, and
pr102021, pr54855-[37], and pr90773-17.

pr15184* tests need a PIC register to access global variables, which
affects register allocation, so the patterns would have to be
adjusted.  pr27971 can't use the expected addressing mode to
dereference the array with PIC, so it ends up selecting an indexed
addressing mode, obviating the expected separate shift insn.

pr70263-2 is another case that implicitly expects a sibcall,
impossible because of the need for the PIC register; without a
sibcall, the expected REG_EQUIV for the reuse of the stack slot of an
incoming argument does not occur.  pr78035 duplicates the final
compare in both then and else blocks with PIE, which deviates from the
expected cmp count.  pr81736-[57] test for no frame pointer, but the
PIC register assignment to a call-saved register forces a frame; the
former ends up not using the PIC register, but it's only optimized out
after committing to a stack frame to preserve it.  pr85620-6 also
expects a tail call in a situation that is impossible on ia32 PIC.

pr85667-6 doesn't expect the movl in get_pc_thunk.  pr93492-5 tests
-mfentry, not available with PIC on ia32.  pr96539 expects a
tail-call, to avoid copying a large-ish struct argument, but the call
requires the PIC register, so no tail-call.  stack-prot-sym.c expects
a nonpic addressing mode.

for  gcc/testsuite/ChangeLog

* g++.dg/abi/anon1.C: Disable pie on ia32.
* g++.dg/abi/anon4.C: Likewise.
* g++.dg/cpp0x/initlist-const1.C: Likewise.
* g++.dg/no-stack-protector-attr-3.C: Likewise.
* g++.dg/stackprotectexplicit2.C: Likewise.
* g++.dg/pr71694.C: Likewise.
* gcc.dg/pr102892-1.c: Likewise.
* gcc.dg/sibcall-11.c: Likewise.
* gcc.dg/torture/builtin-self.c: Likewise.
* gcc.target/i386/avx2-dest-false-dep-for-glc.c: Likewise.
* gcc.target/i386/avx512bf16-cvtsbh2ss-1.c: Likewise.
* gcc.target/i386/avx512f-broadcast-pr87767-1.c: Likewise.
* gcc.target/i386/avx512f-broadcast-pr87767-3.c: Likewise.
* gcc.target/i386/avx512f-broadcast-pr87767-5.c: Likewise.
* gcc.target/i386/avx512f-broadcast-pr87767-7.c: Likewise.
* gcc.target/i386/avx512fp16-broadcast-1.c: Likewise.
* gcc.target/i386/avx512fp16-pr101846.c: Likewise.
* gcc.target/i386/avx512vl-broadcast-pr87767-1.c: Likewise.
* gcc.target/i386/avx512vl-broadcast-pr87767-3.c: Likewise.
* gcc.target/i386/avx512vl-broadcast-pr87767-5.c: Likewise.
* gcc.target/i386/pr100865-2.c: Likewise.
* gcc.target/i386/pr100865-3.c: Likewise.
* gcc.target/i386/pr100865-4a.c: Likewise.
* gcc.target/i386/pr100865-4b.c: Likewise.
* gcc.target/i386/pr100865-5a.c: Likewise.
* gcc.target/i386/pr100865-5b.c: Likewise.
* gcc.target/i386/pr100865-6a.c: Likewise.
* gcc.target/i386/pr100865-6b.c: Likewise.
* gcc.target/i386/pr100865-6c.c: Likewise.
* gcc.target/i386/pr100865-7b.c: Likewise.
* gcc.target/i386/pr101796-1.c: Likewise.
* gcc.target/i386/pr101846-2.c: Likewise.
* gcc.target/i386/pr101989-broadcast-1.c: Likewise.
* gcc.target/i386/pr102021.c: Likewise.
* gcc.target/i386/pr90773-17.c: Likewise.
* gcc.target/i386/pr54855-3.c: Likewise.
* gcc.target/i386/pr54855-7.c: Likewise.
* gcc.target/i386/pr15184-1.c: Likewise.
* gcc.target/i386/pr15184-2.c: Likewise.
* gcc.target/i386/pr27971.c: Likewise.
* gcc.target/i386/pr70263-2.c: Likewise.
* gcc.target/i386/pr78035.c: Likewise.
* gcc.target/i386/pr81736-5.c: Likewise.
* gcc.target/i386/pr81736-7.c: Likewise.
* gcc.target/i386/pr85620-6.c: Likewise.
* gcc.target/i386/pr85667-6.c: Likewise.
* gcc.target/i386/pr93492-5.c: Likewise.
* gcc.target/i386/pr96539.c: Likewise.
PR target/81708 (%gs:my_guard)
* gcc.target/i386/stack-prot-sym.c: Likewise.
* g++.dg/init/static-cdtor1.C: Add alternate patterns for PIC.
* gcc.target/i386/avx512fp16-vcvtsh2si-1a.c: Extend patterns
for PIC/PIE register allocation.
* gcc.target/i386/pr100704-3.c: Likewise.
* gcc.target/i386/avx512fp16-vcvtsh2usi-1a.c: Likewise.
* gcc.target/i386/avx512fp16-vcvttsh2si-1a.c: Likewise.
* gcc.target/i386/avx512fp16-vcvttsh2usi-1a.c: Likewise.
* gcc.target/i386/avx512fp16-vmovsh-1a.c: Likewise.
* gcc.target/i386/interrupt-11.c: Likewise, allowing for
preservation of the PIC register.
* gcc.target/i386/interrupt-12.c: Likewise.
* gcc.target/i386/interrupt-13.c: Likewise.
* gcc.target/i386/interrupt-15.c: Likewise.
* gcc.target/i386/interrupt-16.c: Likewise.
* gcc.target/i386/interrupt-17.c: Likewise.
* gcc.target/i386/interrupt-8.c: Likewise.
* gcc.target/i386/cet-sjlj-6a.c: Combine patterns from
previous change.
* gcc.target/i386/cet-sjlj-6b.c: Likewise.
* gcc.target/i386/pad-10.c: Accept insns in get_pc_thunk.
* gcc.target/i386/pr70321.c: Likewise.
* gcc.target/i386/pr81563.c: Likewise.
* gcc.target/i386/pr84278.c: Likewise.
* gcc.target/i386/pr90773-2.c: Likewise, plus extra loads from
the GOT.
* gcc.target/i386/pr90773-3.c: Likewise.
* gcc.target/i386/pr94913-2.c: Accept additional PIC insns.
* gcc.target/i386/stack-check-17.c: Likewise.
* gcc.target/i386/stack-check-12.c: Do not require dummy stack
probing obviated with PIC.
* gcc.target/i386/pr95126-m32-1.c: Expect missed optimization
with PIC.
* gcc.target/i386/pr95126-m32-2.c: Likewise.
* gcc.target/i386/pr95852-2.c: Accept different optimization
with PIC.
* gcc.target/i386/pr95852-4.c: Likewise.

2 years agoifcvt: Fix up noce_convert_multiple_sets [PR106590]
Jakub Jelinek [Mon, 15 Aug 2022 11:56:57 +0000 (13:56 +0200)]
ifcvt: Fix up noce_convert_multiple_sets [PR106590]

The following testcase is miscompiled on x86_64-linux.
The problem is in the noce_convert_multiple_sets optimization.
We essentially have:
if (g == 1)
  {
    g = 1;
    f = 23;
  }
else
  {
    g = 2;
    f = 20;
  }
and for each insn try to create a conditional move sequence.
There is code to detect overlap with the regs used in the condition
and the destinations, so we actually try to construct:
tmp_g = g == 1 ? 1 : 2;
f = g == 1 ? 23 : 20;
g = tmp_g;
which is fine.  But, we actually try to create two different
conditional move sequences in each case, seq1 with the whole
(eq (reg/v:HI 82 [ g ]) (const_int 1 [0x1]))
condition and seq2 with cc_cmp
(eq (reg:CCZ 17 flags) (const_int 0 [0]))
to rely on the earlier present comparison.  In each case, we
compare the rtx costs and choose the cheaper sequence (seq1 if both
have the same cost).
The problem is that with the skylake tuning,
tmp_g = g == 1 ? 1 : 2;
is actually expanded as
tmp_g = (g == 1) + 1;
in seq1 (which clobbers (reg 17 flags)) and as a cmov in seq2
(which doesn't).  The tuning says both have the same cost, so we
pick seq1.  Next we check sequences for
f = g == 1 ? 23 : 20; and here the seq2 cmov is cheaper, but it
uses (reg 17 flags) which has been clobbered earlier.

The following patch fixes that by detecting if we in the chosen
sequence clobber some register mentioned in cc_cmp or rev_cc_cmp,
and if yes, arranges for only seq1 (i.e. sequences that emit the
comparison itself) to be used after that.

2022-08-15  Jakub Jelinek  <jakub@redhat.com>

PR rtl-optimization/106590
* ifcvt.cc (check_for_cc_cmp_clobbers): New function.
(noce_convert_multiple_sets_1): If SEQ sets or clobbers any regs
mentioned in cc_cmp or rev_cc_cmp, don't consider seq2 for any
further conditional moves.

* gcc.dg/torture/pr106590.c: New test.

2 years agox86: Enable __bf16 type for TARGET_SSE2 and above
konglin1 [Mon, 15 Aug 2022 06:39:22 +0000 (14:39 +0800)]
x86: Enable __bf16 type for TARGET_SSE2 and above

gcc/ChangeLog:

* config/i386/i386-builtin-types.def (BFLOAT16): New primitive type.
* config/i386/i386-builtins.cc : Support __bf16 type for i386 backend.
(ix86_register_bf16_builtin_type): New function.
(ix86_bf16_type_node): New.
(ix86_bf16_ptr_type_node): Ditto.
(ix86_init_builtin_types): Add ix86_register_bf16_builtin_type function call.
* config/i386/i386-modes.def (FLOAT_MODE): Add BFmode.
(ADJUST_FLOAT_FORMAT): Ditto.
* config/i386/i386.cc (classify_argument): Handle BFmode.
(construct_container): Ditto.
(function_value_32): Return __bf16 by %xmm0.
(function_value_64): Return __bf16 by SSE register.
(ix86_output_ssemov): Handle BFmode.
(ix86_legitimate_constant_p): Disable BFmode constant double.
(ix86_secondary_reload): Require gpr as intermediate register
to store __bf16 from sse register when sse4 is not available.
(ix86_scalar_mode_supported_p): Enable __bf16 under sse2.
(ix86_mangle_type): Add manlging for __bf16 type.
(ix86_invalid_conversion): New function for target hook.
(ix86_invalid_unary_op): Ditto.
(ix86_invalid_binary_op): Ditto.
(TARGET_INVALID_CONVERSION): New define for target hook.
(TARGET_INVALID_UNARY_OP): Ditto.
(TARGET_INVALID_BINARY_OP): Ditto.
* config/i386/i386.h (host_detect_local_cpu): Add BFmode.
* config/i386/i386.md ("mode"): Add BFmode.
(MODE_SIZE): Ditto.
(X87MODEFH): Ditto.
(HFBF): Add new define_mode_iterator.
(*pushhf_rex64): Change for BFmode.
(*push<mode>_rex64): Ditto.
(*pushhf): Ditto.
(*push<mode>): Ditto.
(MODESH): Ditto.
(hfbfconstf): Add new define_mode_attr.
(*mov<mode>_internal): Add BFmode.

gcc/testsuite/ChangeLog:

* g++.target/i386/bfloat_cpp_typecheck.C: New test.
* gcc.target/i386/bfloat16-1.c: Ditto.
* gcc.target/i386/sse2-bfloat16-1.c: Ditto.
* gcc.target/i386/sse2-bfloat16-2.c: Ditto.
* gcc.target/i386/sse2-bfloat16-scalar-typecheck.c: Ditto.

2 years agoDaily bump.
GCC Administrator [Mon, 15 Aug 2022 00:16:42 +0000 (00:16 +0000)]
Daily bump.

2 years agoDaily bump.
GCC Administrator [Sun, 14 Aug 2022 00:16:56 +0000 (00:16 +0000)]
Daily bump.

2 years agoMove V1TI shift/rotate lowering from expand to pre-reload split on x86_64.
Roger Sayle [Sat, 13 Aug 2022 13:52:45 +0000 (14:52 +0100)]
Move V1TI shift/rotate lowering from expand to pre-reload split on x86_64.

This patch moves the lowering of 128-bit V1TImode shifts and rotations by
constant bit counts to sequences of SSE operations from the RTL expansion
pass to the pre-reload split pass.  Postponing this splitting of shifts
and rotates enables (will enable) the TImode equivalents of these operations/
instructions to be considered as candidates by the (TImode) STV pass.
Technically, this patch changes the existing expanders to continue to
lower shifts by variable amounts, but constant operands become RTL
instructions, specified by define_insn_and_split that are triggered by
x86_pre_reload_split.  The one minor complication is that logical shifts
by multiples of eight, don't get split, but are handled by existing insn
patterns, such as sse2_ashlv1ti3 and sse2_lshrv1ti3.  There should be no
changes in generated code with this patch, which just adjusts the pass
in which transformations get applied.

2022-08-13  Roger Sayle  <roger@nextmovesoftware.com>
    Uroš Bizjak  <ubizjak@gmail.com>

gcc/ChangeLog
* config/i386/predicates.md (const_0_to_255_not_mul_8_operand):
New predicate for values between 0/1 and 255, not multiples of 8.
* config/i386/sse.md (ashlv1ti3): Delay lowering of logical left
shifts by constant bit counts.
(*ashlvti3_internal): New define_insn_and_split that lowers
logical left shifts by constant bit counts, that aren't multiples
of 8, before reload.
(lshrv1ti3): Delay lowering of logical right shifts by constant.
(*lshrv1ti3_internal): New define_insn_and_split that lowers
logical right shifts by constant bit counts, that aren't multiples
of 8, before reload.
(ashrv1ti3):: Delay lowering of arithmetic right shifts by
constant bit counts.
(*ashrv1ti3_internal): New define_insn_and_split that lowers
arithmetic right shifts by constant bit counts before reload.
(rotlv1ti3): Delay lowering of rotate left by constant.
(*rotlv1ti3_internal): New define_insn_and_split that lowers
rotate left by constant bits counts before reload.
(rotrv1ti3): Delay lowering of rotate right by constant.
(*rotrv1ti3_internal): New define_insn_and_split that lowers
rotate right by constant bits counts before reload.

2 years agotestsuite: Disable out-of-bounds checker in analyzer/torture/pr93451.c
Tim Lange [Sat, 13 Aug 2022 11:31:45 +0000 (13:31 +0200)]
testsuite: Disable out-of-bounds checker in analyzer/torture/pr93451.c

This patch disables Wanalyzer-out-of-bounds for analyzer/torture/pr93451.c
and makes the test case pass when compiled with -m32.

The emitted warning is a true positive but only occurs if
sizeof (long int) is less than sizeof (double). I've already discussed a
similar case with Dave in the context of pr96764.c and we came to the
conclusion that we just disable the checker in such cases.

Committed under the "obvious fix" rule.

gcc/testsuite/ChangeLog:

* gcc.dg/analyzer/torture/pr93451.c:
Disable Wanalyzer-out-of-bounds.

2 years agoDaily bump.
GCC Administrator [Sat, 13 Aug 2022 00:16:54 +0000 (00:16 +0000)]
Daily bump.