... and add a minimum amount of offloading testing.
(Leaving aside that 'fwrite' to 'stderr' probably wouldn't work anyway) the
'fwrite' calls in 'libgomp/error.c:GOMP_warning', 'libgomp/error.c:GOMP_error'
drag in 'isatty', which isn't provided by my nvptx newlib build at present, so
we get, for example:
[...]
FAIL: libgomp.c/../libgomp.c-c++-common/declare_target-1.c (test for excess errors)
Excess errors:
unresolved symbol isatty
mkoffload: fatal error: [...]/build-gcc/./gcc/x86_64-pc-linux-gnu-accel-nvptx-none-gcc returned 1 exit status
[...]
..., and many more.
Fix up for recent commit 0d973c0a0d
"openmp: Implement the error directive".
libgomp/
* config/nvptx/error.c (fwrite, exit): Override, too.
* testsuite/libgomp.c-c++-common/error-1.c: Add a minimum amount
of offloading testing.
* testsuite/libgomp.fortran/error-1.f90: Likewise.
Since 'Remove obsolete IRIX 6.5 support' [1] we only use
gp-relative jump-tables for PIC code. We can fall back to
default behaviour for asm_function_rodata_section.
[1] https://gcc.gnu.org/ml/libstdc++/2012-03/msg00067.html
2018-06-04 Dragan Mladjenovic <dragan.mladjenovic@rt-rk.com>
gcc/
* config/mips/mips.c (mips_function_rodata_section,
TARGET_ASM_FUNCTION_RODATA_SECTION): Removed.
Special-casing checks for in-tree gas features is unnecessary since
r100007 which made configure-gcc depend on all-gas, and thus making
alternate code path in gcc_GAS_CHECK_FEATURE for in-tree gas
redundant.
Along the way this fixes PR 91602, which is caused by incorrect guess
of leb128 support presence in RISC-V.
First patch removes alternate code path in gcc_GAS_CHECK_FEATURE and
related code, the rest are further cleanups. Patches 2 and 3 in
series make no functional changes, thus configure is unchanged.
gcc/ChangeLog:
PR target/91602
* acinclude.m4 (_gcc_COMPUTE_GAS_VERSION, _gcc_GAS_VERSION_GTE_IFELSE)
(gcc_GAS_VERSION_GTE_IFELSE): Remove.
(gcc_GAS_CHECK_FEATURE): Do not handle in-tree case specially.
* configure.ac: Remove gcc_cv_gas_major_version, gcc_cv_gas_minor_version.
Remove remaining checks for in-tree assembler.
* configure: Regenerate.
gcc/
* config/h8300/h8300.c (shift_alg_hi): Improve arithmetic shift right
by 15 bits for H8/300H and H8/S. Improve logical shifts by 12
bits for H8/S.
(shift_alg_si): Improve arithmetic right shift by 28-30 bits for
H8/300H. Improve arithmetic shift right by 15 bits for H8/S.
Improve logical shifts by 27 bits for H8/S.
(get_shift_alg): Corresponding changes.
(h8300_option_override): Revert to loops for -Os when profitable.
Tests that depend on filesystem permissions FAIL if run on Windows or as
root. Add a helper function to detect those cases, so the tests can skip
those checks gracefully.
Signed-off-by: Jonathan Wakely <jwakely@redhat.com>
libstdc++-v3/ChangeLog:
PR libstdc++/90787
* testsuite/27_io/filesystem/iterators/directory_iterator.cc:
Use new __gnu_test::permissions_are_testable() function.
* testsuite/27_io/filesystem/iterators/recursive_directory_iterator.cc:
Likewise.
* testsuite/27_io/filesystem/operations/exists.cc: Likewise.
* testsuite/27_io/filesystem/operations/is_empty.cc: Likewise.
* testsuite/27_io/filesystem/operations/remove.cc: Likewise.
* testsuite/27_io/filesystem/operations/remove_all.cc: Likewise.
* testsuite/27_io/filesystem/operations/status.cc: Likewise.
* testsuite/27_io/filesystem/operations/symlink_status.cc:
Likewise.
* testsuite/27_io/filesystem/operations/temp_directory_path.cc:
Likewise.
* testsuite/experimental/filesystem/iterators/directory_iterator.cc:
Likewise.
* testsuite/experimental/filesystem/iterators/recursive_directory_iterator.cc:
Likewise.
* testsuite/experimental/filesystem/operations/exists.cc:
Likewise.
* testsuite/experimental/filesystem/operations/is_empty.cc:
Likewise.
* testsuite/experimental/filesystem/operations/remove.cc:
Likewise.
* testsuite/experimental/filesystem/operations/remove_all.cc:
Likewise.
* testsuite/experimental/filesystem/operations/temp_directory_path.cc:
Likewise.
* testsuite/util/testsuite_fs.h (__gnu_test::permissions_are_testable):
New function to guess whether testing permissions will work.
This patch adds support for the 'll' (long double)
and 'w' (HOST_WIDE_INT) length modifiers to the
Fortran FE diagnostic function (gfc_error, gfc_warning, ...)
gcc/c-family/ChangeLog:
* c-format.c (gcc_gfc_length_specs): Add 'll' and 'w'.
(gcc_gfc_char_table): Add T9L_LL and T9L_ULL to
"di" and "u", respecitively; fill with BADLEN to match
size of 'types'.
(get_init_dynamic_hwi): Split off from ...
(init_dynamic_diag_info): ... here. Call it.
(init_dynamic_gfc_info): Call it.
gcc/fortran/ChangeLog:
* error.c
(error_uinteger): Take 'long long unsigned' instead
of 'long unsigned' as argumpent.
(error_integer): Take 'long long' instead of 'long'.
(error_hwuint, error_hwint): New.
(error_print): Update to handle 'll' and 'w'
length modifiers.
* simplify.c (substring_has_constant_len): Use '%wd'
in gfc_error.
This uses the group_id computed to ensure DRs in different BBs do
not get merged into a DR group. To achieve this we seed the
group from the BB index when group_ids are not computed and we
make sure to bump the group_id when advancing to the next BB for
BB SLP analysis.
This paves the way for relaxing the grouping for BB vectorization
by adjusting its group_id computation.
2021-08-20 Richard Biener <rguenther@suse.de>
* tree-vect-data-refs.c (dr_group_sort_cmp): Do not compare
BBs.
(vect_analyze_data_ref_accesses): Likewise. Assign the BB
index as group_id when dataref_groups were not computed.
* tree-vect-slp.c (vect_slp_bbs): Bump current_group when
we advace to the next BB.
This patch implements the error directive. Depending on clauses it is either
a compile time diagnostics (in that case diagnosed right away) or runtime
diagnostics (libgomp API call that diagnoses at runtime), and either fatal
or warning (error or warning at compile time or fatal error vs. error at
runtime) and either has no message or user supplied message (this kind of
e.g. deprecated attribute). The directive is also stand-alone directive
when at runtime while utility (thus disappears from the IL as if it wasn't
there for parsing like nothing directive) at compile time.
There are some clarifications in the works ATM, so this patch doesn't yet
require that for compile time diagnostics the user message must be a constant
string literal, there are uncertainities on what exactly is valid argument
of message clause (whether just const char * type, convertible to const char *,
qualified/unqualified const char * or char * or what else) and what to do
in templates. Currently even in templates it is diagnosed right away for
compile time diagnostics, if we'll need to substitute it, we'd need to queue
something into the IL, have pt.c handle it and diagnose only later.
2021-08-20 Jakub Jelinek <jakub@redhat.com>
gcc/
* omp-builtins.def (BUILT_IN_GOMP_WARNING, BUILT_IN_GOMP_ERROR): New
builtins.
gcc/c-family/
* c-pragma.h (enum pragma_kind): Add PRAGMA_OMP_ERROR.
* c-pragma.c (omp_pragmas): Add error directive.
* c-omp.c (omp_directives): Uncomment error directive entry.
gcc/c/
* c-parser.c (c_parser_omp_error): New function.
(c_parser_pragma): Handle PRAGMA_OMP_ERROR.
gcc/cp/
* parser.c (cp_parser_handle_statement_omp_attributes): Determine if
PRAGMA_OMP_ERROR directive is C_OMP_DIR_STANDALONE.
(cp_parser_omp_error): New function.
(cp_parser_pragma): Handle PRAGMA_OMP_ERROR.
gcc/fortran/
* types.def (BT_FN_VOID_CONST_PTR_SIZE): New DEF_FUNCTION_TYPE_2.
* f95-lang.c (ATTR_COLD_NORETURN_NOTHROW_LEAF_LIST): Define.
gcc/testsuite/
* c-c++-common/gomp/error-1.c: New test.
* c-c++-common/gomp/error-2.c: New test.
* c-c++-common/gomp/error-3.c: New test.
* g++.dg/gomp/attrs-1.C (bar): Add error directive test.
* g++.dg/gomp/attrs-2.C (bar): Add error directive test.
* g++.dg/gomp/attrs-13.C: New test.
* g++.dg/gomp/error-1.C: New test.
libgomp/
* libgomp.map (GOMP_5.1): Add GOMP_error and GOMP_warning.
* libgomp_g.h (GOMP_warning, GOMP_error): Declare.
* error.c (GOMP_warning, GOMP_error): New functions.
* testsuite/libgomp.c-c++-common/error-1.c: New test.
While working on error directive, I've noticed a few spots in OpenMP
parsing where we consume and don't diagnose superfluous commas at the end
(either of depend sink arguments or at the end of requires pragma).
2021-08-20 Jakub Jelinek <jakub@redhat.com>
gcc/c/
* c-parser.c (c_parser_omp_clause_depend_sink): Reject spurious
comma at the end of list.
(c_parser_omp_requires): Likewise.
gcc/cp/
* parser.c (cp_parser_omp_clause_depend_sink): Reject spurious
comma at the end of list. Don't parse closing paren here...
(cp_parser_omp_clause_depend): ... but here instead.
gcc/testsuite/
* c-c++-common/gomp/sink-5.c: New test.
* c-c++-common/gomp/requires-3.c: Add test for spurious comma
at the end of pragma line.
PR gcov-profile/89961
gcc/ChangeLog:
* gcov.c (make_gcov_file_name): Rewrite using std::string.
(mangle_name): Simplify, do not used the second argument.
(strip_extention): New function.
(get_md5sum): Likewise.
(get_gcov_intermediate_filename): Handle properly -p and -x
options.
(output_gcov_file): Use string type.
(generate_results): Likewise.
(md5sum_to_hex): Remove.
I noticed that the xx built-in functions (xxspltiw, xxspltidp, xxsplti32dx,
xxeval, xxblend, and xxpermx) were all defined in altivec.md. However, since
the XX instructions can take both traditional floating point and Altivec
registers, these built-in functions should be in vsx.md.
This patch just moves the insns from altivec.md to vsx.md.
I also moved the VM3 mode iterator and VM3_char mode attribute from altivec.md
to vsx.md, since the only use of these were for the XXBLEND insns.
2021-08-20 Michael Meissner <meissner@linux.ibm.com>
gcc/
* config/rs6000/altivec.md (UNSPEC_XXEVAL): Move to vsx.md.
(UNSPEC_XXSPLTIW): Move to vsx.md.
(UNSPEC_XXSPLTID): Move to vsx.md.
(UNSPEC_XXSPLTI32DX): Move to vsx.md.
(UNSPEC_XXBLEND): Move to vsx.md.
(UNSPEC_XXPERMX): Move to vsx.md.
(VM3): Move to vsx.md.
(VM3_char): Move to vsx.md.
(xxspltiw_v4si): Move to vsx.md.
(xxspltiw_v4sf): Move to vsx.md.
(xxspltiw_v4sf_inst): Move to vsx.md.
(xxspltidp_v2df): Move to vsx.md.
(xxspltidp_v2df_inst): Move to vsx.md.
(xxsplti32dx_v4si_inst): Move to vsx.md.
(xxsplti32dx_v4sf): Move to vsx.md.
(xxsplti32dx_v4sf_inst): Move to vsx.md.
(xxblend_<mode>): Move to vsx.md.
(xxpermx): Move to vsx.md.
(xxpermx_inst): Move to vsx.md.
* config/rs6000/vsx.md (UNSPEC_XXEVAL): Move from altivec.md.
(UNSPEC_XXSPLTIW): Move from altivec.md.
(UNSPEC_XXSPLTID): Move from altivec.md.
(UNSPEC_XXSPLTI32DX): Move from altivec.md.
(UNSPEC_XXBLEND): Move from altivec.md.
(UNSPEC_XXPERMX): Move from altivec.md.
(VM3): Move from altivec.md.
(VM3_char): Move from altivec.md.
(xxspltiw_v4si): Move from altivec.md.
(xxspltiw_v4sf): Move from altivec.md.
(xxspltiw_v4sf_inst): Move from altivec.md.
(xxspltidp_v2df): Move from altivec.md.
(xxspltidp_v2df_inst): Move from altivec.md.
(xxsplti32dx_v4si_inst): Move from altivec.md.
(xxsplti32dx_v4sf): Move from altivec.md.
(xxsplti32dx_v4sf_inst): Move from altivec.md.
(xxblend_<mode>): Move from altivec.md.
(xxpermx): Move from altivec.md.
(xxpermx_inst): Move from altivec.md.
An issue with a backend patch I've been investigating has revealed
a missed optimization opportunity during GCC's vector lowering pass.
An unrecognized insn for "(set (reg:SI) (not:SI (const_int 0))"
revealed that not only was my expander not expecting a NOT with
a constant operand, but also that veclower was producing the
dubious tree expression ~0.
The attached patch replaces a call to gimple_build_assign with a
call to either gimplify_build1 or gimplify_build2 depending upon
whether the operation takes one or two operands. The net effect
is that where GCC previously produced the following optimized
gimple for testsuite/c-c++common/Wunused-var-16.c (notice the ~0
and the "& 0"):
void foo ()
{
V x;
V y;
vector(16) unsigned char _1;
unsigned char _7;
unsigned char _8;
y_2 = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
x_3 = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
_7 = ~0;
_1 = {_7, _7, _7, _7, _7, _7, _7, _7, _7, _7, _7, _7, _7, _7, _7, _7};
_8 = 0 & _7;
y_4 = {_8, _8, _8, _8, _8, _8, _8, _8, _8, _8, _8, _8, _8, _8, _8, _8};
v = y_4;
return;
}
With this patch we now generate:
void foo ()
{
V x;
V y;
vector(16) unsigned char _1;
y_2 = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
x_3 = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
_1 = { 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 2
55, 255 };
y_4 = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
v = y_4;
return;
}
2021-08-20 Roger Sayle <roger@nextmovesoftware.com>
gcc/ChangeLog
* tree-vect-generic.c (expand_vector_operations_1): Use either
gimplify_build1 or gimplify_build2 instead of gimple_build_assign
when constructing scalar splat expressions.
gcc/testsuite/ChangeLog
* c-c++-common/Wunused-var-16.c: Add an extra check that ~0
is optimized away.
PR101849 shows we ICE on a test case when we pass a non __vector_pair *
pointer to the __builtin_vsx_lxvp and __builtin_vsx_stxvp built-ins
that is cast to __vector_pair *. The problem is that when we expand
the built-in, the cast has already been removed from gimple and we are
only given the base pointer. The solution used here (which fixes the ICE)
is to catch this case and convert the pointer to a __vector_pair * pointer
when expanding the built-in.
2021-08-19 Peter Bergner <bergner@linux.ibm.com>
gcc/
PR target/101849
* config/rs6000/rs6000-call.c (rs6000_gimple_fold_mma_builtin): Cast
pointer to __vector_pair *.
gcc/testsuite/
PR target/101849
* gcc.target/powerpc/pr101849.c: New test.
It is intended that the default for the NeXT runtime at ABI 2 is to
check for nil message receivers. This updates this to match the
documented behaviour and to match the behaviour of the system tools.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/objc/ChangeLog:
* objc-next-runtime-abi-02.c (objc_next_runtime_abi_02_init):
Default receiver nilchecks on.
This adds my new SHOW_HEADERFILE option, and removes some obsolete
options.
Signed-off-by: Jonathan Wakely <jwakely@redhat.com>
libstdc++-v3/ChangeLog:
* doc/doxygen/user.cfg.in: Update to Doxygen 1.9.2
An array member cannot be direct-initialized in a ctor-initializer-list,
so use the base class' move constructor, which does the right thing for
both arrays and non-arrays.
This constructor could be defaulted, but that would make it trivial for
some specializations, which would change the argument passing ABI. Do
that for the versioned namespace only.
Signed-off-by: Jonathan Wakely <jwakely@redhat.com>
libstdc++-v3/ChangeLog:
PR libstdc++/101960
* include/std/tuple (_Tuple_impl(_Tuple_impl&&)): Use base
class' move constructor. Define as defaulted for versioned
namespace.
* testsuite/20_util/tuple/cons/101960.cc: New test.
We should document the status of this unimplemented feature.
Signed-off-by: Jonathan Wakely <jwakely@redhat.com>
libstdc++-v3/ChangeLog:
PR libstdc++/100139
* doc/xml/manual/status_cxx2020.xml: Add P1739R4 to status table.
* doc/html/manual/status.html: Regenerate.
The current code assumes that system_clock::duration is nanoseconds, and
also performs a value-changing conversion from nanoseconds::max() to
double (which doesn't matter after dividing by 1e9, but triggers a
warning with Clang nonetheless).
A better solution is to use system_clock::duration::max() and perform
the comparison entirely using the std::chrono types, rather than with
dimensionless arithmetic types.
This doesn't address the FIXME in the function, so the overflow check
still rejects some values that could be represented by the file_clock.
Signed-off-by: Jonathan Wakely <jwakely@redhat.com>
libstdc++-v3/ChangeLog:
* src/filesystem/ops-common.h (filesystem::file_time): Improve
overflow check by using system_clock::duration::max().
As suggested in the PR, the following patch adds two new clrsb
expansion possibilities if target doesn't have clrsb_optab for the
requested nor wider modes, but does have clz_optab for the requested
mode.
One expansion is
clrsb (op0)
expands as
clz (op0 ^ (((stype)op0) >> (prec-1))) - 1
which is usable if CLZ_DEFINED_VALUE_AT_ZERO is 2 with value
of prec, because the clz argument can be 0 and clrsb should give
prec-1 in that case.
The other expansion is
clz (((op0 << 1) ^ (((stype)op0) >> (prec-1))) | 1)
where the clz argument is never 0, but it is one operation longer.
E.g. on x86_64-linux with -O2 -mno-lzcnt, this results for
int foo (int x) { return __builtin_clrsb (x); }
in
- subq $8, %rsp
- movslq %edi, %rdi
- call __clrsbdi2
- addq $8, %rsp
- subl $32, %eax
+ leal (%rdi,%rdi), %eax
+ sarl $31, %edi
+ xorl %edi, %eax
+ orl $1, %eax
+ bsrl %eax, %eax
+ xorl $31, %eax
and with -O2 -mlzcnt:
+ movl %edi, %eax
+ sarl $31, %eax
+ xorl %edi, %eax
+ lzcntl %eax, %eax
+ subl $1, %eax
On armv7hl-linux-gnueabi with -O2:
- push {r4, lr}
- bl __clrsbsi2
- pop {r4, pc}
+ @ link register save eliminated.
+ eor r0, r0, r0, asr #31
+ clz r0, r0
+ sub r0, r0, #1
+ bx lr
As it (at least usually) will make code larger, it is
disabled for -Os or cold instructions.
2021-08-19 Jakub Jelinek <jakub@redhat.com>
PR middle-end/101950
* optabs.c (expand_clrsb_using_clz): New function.
(expand_unop): Use it as another clrsb expansion fallback.
* gcc.target/i386/pr101950-1.c: New test.
* gcc.target/i386/pr101950-2.c: New test.
When working on error directive, I've noticed the C FE ICEs on
#pragma omp requires atomic_default_mem_order (
where it tries to peek 2nd token after the CPP_PRAGMA_EOL (or CPP_EOF)
in there in order to improve error-recovery on say
atomic_default_mem_order (acquire)
or
atomic_default_mem_order (seqcst)
etc. The C++ FE didn't ICE, but it is better to follow the same thing there.
2021-08-19 Jakub Jelinek <jakub@redhat.com>
gcc/c/
* c-parser.c (c_parser_omp_requires): Don't call
c_parser_peek_2nd_token and optionally consume token if current
token is CPP_EOF, CPP_PRAGMA_EOL or CPP_CLOSE_PAREN.
gcc/cp/
* parser.c (cp_parser_omp_requires): Don't call cp_lexer_nth_token_is
and optionally consume token if current token is CPP_EOF,
CPP_PRAGMA_EOL or CPP_CLOSE_PAREN.
gcc/testsuite/
* c-c++-common/gomp/requires-3.c: Add testcase for
atomic_default_mem_order ( at the end of line without corresponding ).
This provides adjusted assembler fragments that are suitable
for x86_64 Mach-O.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/testsuite/ChangeLog:
* jit.dg/test-asm.c: Provide Mach-O fragment.
* jit.dg/test-asm.cc: Likewise.
When working on error directive, I've noticed that while C FE diagnosed
clauses on nothing directive which doesn't allow any, the C++ FE silently
accepted it.
2021-08-19 Jakub Jelinek <jakub@redhat.com>
* parser.c (cp_parser_omp_nothing): Use cp_parser_require_pragma_eol
instead of cp_parser_skip_to_pragma_eol.
* c-c++-common/gomp/nothing-2.c: New test.
The testsuite setup for jit is not compatible with Darwin since it
assumes that all targets support --export-dynamic.
- this is fixed by adding '-rdynamic' conditionally upon target
support for that (-rdynamic will be converted to the appropriate
linker option).
There is also an assumption that a suitable version of dejagnu.h
is present in some default include search path that is usable from
the testsuite. This is not the case for Darwin (dejagnu.h is not
installed, and would not, in general, be found in any default include
search path if installed via one of the main 'distros'). Also the
upstream dejagnu.h has a definition of 'wait()' that clashes with a
libc routines and therefore causes fails in the testsuite.
- This patch imports the header from dejagnu-1.6.2 and
* renames it to 'jit-dejagnu.h'
* patches it to avoid unused variable warnings and the clash
with the libc definition of wait ()
* In accordance with the advice in the expect man page, ensures
that the final output of the 'totals' print is stable.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/jit/ChangeLog:
* docs/examples/tut04-toyvm/toyvm.c: Include jit-dejagnu.h.
* docs/examples/tut04-toyvm/toyvm.cc: Likewise.
* jit-dejagnu.h: New file, imported from dejagnu-1.6.2 and
patched for this application.
gcc/testsuite/ChangeLog:
* jit.dg/harness.h: Include jit-dejagnu.h.
* jit.dg/jit.exp: Use -rdynamic conditionally on target
support, instead of unconditional -Wl,--export-dynamic.
Fix up for r242748 (commit 3615816da8)
"[PR target/78213] Do not ICE on non-empty -fself-test", as made
apparent by recent commit a42467bdb7
"Restore 'gcc.dg/pr78213.c' testing", after the test case had gotten
disabled in r243681 (commit ecfc21ff34)
"Introduce selftest::locate_file" shortly after its original introduction.
gcc/testsuite/
PR testsuite/101969
* gcc.dg/pr78213.c: Fix up for '--enable-checking=release' etc.
Fix directory to enable -fopenmp processing.
gcc/testsuite/
PR testsuite/101963
* gfortran.dg/nothing-1.f90: Moved to ...
* gfortran.dg/gomp/nothing-1.f90: ... here.
* gfortran.dg/nothing-2.f90: Moved to ...
* gfortran.dg/gomp/nothing-2.f90: ... here;
avoid $ issue in $OMP in dg-error.
