This test was copied from 27_io/filesystem/path/query/is_absolute.cc but
should have been modified to test the path type from the TS instead of
std::filesystem::path.
* testsuite/experimental/filesystem/path/query/is_absolute.cc: Fix
test to use TS, not C++17.
From-SVN: r266957
2018-12-10 Richard Biener <rguenther@suse.de>
* tree-affine.c (tree_to_aff_combination): Remove unreachable
MEM_REF case.
(aff_combination_expand): Cache on SSA names, not possibly
on conversion trees. Avoid expanding cases we do not handle.
From-SVN: r266956
2018-12-10 Richard Biener <rguenther@suse.de>
PR tree-optimization/88427
* vr-values.c (vr_values::extract_range_from_phi_node):
Handle symbolic ranges conservatively when trying to drop
to Inf +- 1.
* gcc.dg/pr88427.c: New testcase.
From-SVN: r266955
2018-12-10 Martin Jambor <mjambor@suse.cz>
PR ipa/88214
* ipa-prop.c (determine_locally_known_aggregate_parts): Make sure
we check pointers against pointers.
testsuite/
* gcc.dg/ipa/pr88214.c: New test.
From-SVN: r266953
2018-12-10 Richard Biener <rguenther@suse.de>
PR middle-end/88415
* gimple.c (gimple_assign_set_rhs_with_ops): Transfer EH
info to a newly allocated stmt.
* gcc.dg/gomp/pr88415.c: New testcase.
From-SVN: r266951
* c-c++-common/patchable_function_entry-decl.c: Pass -mcpu=gr6 for
Visium and remove other specific handling.
* c-c++-common/patchable_function_entry-default.c: Likewise.
* c-c++-common/patchable_function_entry-definition.c: Likewise.
From-SVN: r266948
2018-12-10 Fredrik Nyström <fredrik@lysator.liu.se>
PR bootstrap/65725
* config/sol2.h: Only use libgcc-unwind.map if
ENABLE_SHARED_LIBGCC.
From-SVN: r266946
2018-12-09 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88205
* io.c (gfc_match_open): Move NEWUNIT checks to after STATUS checks.
2018-12-09 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88205
* gfortran.dg/pr88205.f90: New unit.
From-SVN: r266936
The tests in gdc.test/compilable/ddoc*.d don't require the module to be
compiled all the way down to object code. Instead, only compile the
test sources with -fdoc, and scan the generated html content.
gcc/testsuite/ChangeLog:
PR d/88039
* gdc.test/gdc-test.exp (gdc-convert-args): Handle -D.
(dmd2dg): Check generated html in ddoc tests.
(gdc-do-test): Set dg-do-what-default to compile for ddoc tests.
From-SVN: r266933
2018-12-09 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88206
* match.c (gfc_match_type_spec): REAL can be an intrinsic function.
2018-12-09 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88206
* gfortran.dg/pr88206.f90: New test.
From-SVN: r266930
2018-12-09 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/88411
* io/transfer.c (dta_transfer_init): Do not treat as an
asynchronous statement unless the statement has
ASYNCHRONOUS="YES".
(st_write_done): Likewise.
(st_read_done): Do not perform async_wait for synchronous I/O
on an async unit.
(st_read_done): Likewise.
2018-12-09 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/88411
* testsuite/libgomp.fortran/async_io_8.f90: New test.
From-SVN: r266929
2018-12-09 Fritz Reese <fritzoreese@gmail.com>
PR fortran/88228
* resolve.c (resolve_operator): Do not call resolve_function.
Break like other cases.
2018-12-09 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88228
* gfortran.dg/pr88228.f90: New test.
From-SVN: r266926
Backports VRP fixes from the D front-end implementation to the C++ port,
and fixes errors reported by ubsan build where the conversion from D
didn't include adjusting integer suffixes from 'UL' to 'ULL'.
Fixes https://gcc.gnu.org/PR88366
Reviewed-on: https://github.com/dlang/dmd/pull/9046
From-SVN: r266925
gcc/fortran/
* trans-openmp.c (gfc_trans_oacc_combined_directive): Set the
location of combined acc loops.
gcc/testsuite/
* gfortran.dg/goacc/combined-directives-3.f90: New file.
Reviewed-by: Thomas Schwinge <thomas@codesourcery.com>
From-SVN: r266924
The C++ FE doesn't set the expr_location of the split acc loop in combined acc
parallel/kernels loop directives. This only happens for with combined
directives, otherwise cp_parser_omp_construct would be responsible for setting
the location. After fixing this bug, I was able to resolve a couple of long
standing diagnostics discrepancies between the C/C++ FEs in the test suite.
gcc/cp/
* parser.c (cp_parser_oacc_kernels_parallel): Adjust EXPR_LOCATION
on the combined acc loop.
gcc/testsuite/
* c-c++-common/goacc/combined-directives-3.c: New test.
* c-c++-common/goacc/loop-2-kernels.c (void K): Adjust test.
* c-c++-common/goacc/loop-2-parallel.c (void P): Adjust test.
* c-c++-common/goacc/loop-3.c (void p2): Adjust test.
Reviewed-by: Thomas Schwinge <thomas@codesourcery.com>
From-SVN: r266923
Don't duplicate in the Fortran front end what's generically being checked in
the middle end.
gcc/fortran/
* openmp.c (resolve_oacc_loop_blocks): Remove checking of OpenACC
loop clauses.
gcc/testsuite/
* gfortran.dg/goacc/loop-2-kernels.f95: Update.
* gfortran.dg/goacc/loop-2-parallel.f95: Likewise.
* gfortran.dg/goacc/nested-parallelism.f90: Likewise.
From-SVN: r266922
gcc/testsuite/
* gfortran.dg/goacc/loop-2.f95: Split into...
* gfortran.dg/goacc/loop-2-kernels-nested.f95: ... this new
file...
* gfortran.dg/goacc/loop-2-kernels-tile.f95: ..., and this new
file...
* gfortran.dg/goacc/loop-2-kernels.f95: ..., and this new file...
* gfortran.dg/goacc/loop-2-parallel-3.f95: ..., and this new
file...
* gfortran.dg/goacc/loop-2-parallel-nested.f95: ..., and this new
file...
* gfortran.dg/goacc/loop-2-parallel-tile.f95: ..., and this new
file...
* gfortran.dg/goacc/loop-2-parallel.f95: ..., and this new file.
From-SVN: r266921
The Fortran front end declares that the OpenACC "Clause SEQ conflicts with
INDEPENDENT". While that combination doesn't make too much sense indeed, it's
still valid; these are orthogonal concepts.
gcc/fortran/
PR fortran/88420
* openmp.c (resolve_oacc_loop_blocks): Remove "Clause SEQ
conflicts with INDEPENDENT" diagnostic.
gcc/testsuite/
PR fortran/88420
* gfortran.dg/goacc/loop-1-2.f95: Update.
* gfortran.dg/goacc/loop-1.f95: Likewise.
From-SVN: r266920
2018-12-08 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88048
* resolve.c (check_data_variable): Named constant cannot be a
data object.
2018-12-08 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88048
* gfortran.dg/pr88048.f90: New test.
From-SVN: r266916
PR fortran/87945
* decl.c (var_element): Inquiry parameters cannit be data objects.
20180-12-08 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/87945
* gfortran.dg/pr87945_1.f90: New test.
* gfortran.dg/pr87945_2.f90: New test.
From-SVN: r266915
2018-12-08 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88025
* expr.c (gfc_apply_init): Remove asserts that cannot trigger.
Check for a NULL pointer.
2018-12-08 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88025
* gfortran.dg/pr88025.f90: New test.
From-SVN: r266913
I noticed I accidentally copied a line too many from the "volatile"
handling to the "inline" handling. This fixes it.
gcc/c/
* c-parser (c_parser_asm_statement) [RID_INLINE]: Delete stray line
setting "quals".
From-SVN: r266909
2018-12-08 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/88357
* class.c (insert_component_ref): Check for NULL pointer and
previous error message issued.
* parse.c (parse_associate): Check for NULL pointer.
* resolve.c (resolve_assoc_var): Check for NULL pointer.
2018-12-08 Steven G. Kargl <kargl@gcc.gnu.org>
* gfortran.dg/pr88357_1.f90: New test.
* gfortran.dg/pr88357_2.f90: New test.
From-SVN: r266908
PR fortran/88304
* tree-nested.c (convert_local_reference_stmt): Handle clobbers where
lhs is not a decl normally, don't call use_pointer_in_frame on that
lhs.
* gfortran.fortran-torture/compile/pr88304-2.f90: New test.
From-SVN: r266907
PR rtl-optimization/88390
* dwarf2cfi.c (struct dw_cfi_row): Add window_save field.
(cfi_row_equal_p): Compare it.
(dwarf2out_frame_debug_cfa_window_save): Add FAKE parameter.
If FAKE is false, set window_save of the current row.
(dwarf2out_frame_debug) <REG_CFA_TOGGLE_RA_MANGLE>: Call above
function with FAKE parameter set to true.
<REG_CFA_WINDOW_SAVE>: Likewise but with FAKE parameter set to false.
(change_cfi_row): Emit a DW_CFA_GNU_window_save if necessary.
From-SVN: r266906
PR libgomp/87995
* testsuite/libgomp.c-c++-common/cancel-taskgroup-3.c: Require
tls_runtime effective target.
(t): New threadprivate variable.
(main): Set t in threads which execute iterations of the worksharing
loop. Propagate that to the task after the loop and don't abort
if the current taskgroup hasn't been cancelled.
From-SVN: r266904
/cp
2018-12-07 Paolo Carlini <paolo.carlini@oracle.com>
* decl2.c (grokbitfield): Use DECL_SOURCE_LOCATION in error messages
about bit-fields with function type, warn_if_not_aligned type, and
static bit-fields; avoid DECL_NAME for unnamed declarations.
/testsuite
2018-12-07 Paolo Carlini <paolo.carlini@oracle.com>
* g++.dg/other/bitfield7.C: New.
* g++.dg/parse/bitfield8.C: Likewise.
* g++.dg/parse/bitfield9.C: Likewise.
* g++.dg/pr53037-4.C: Test the locations too.
From-SVN: r266900
gcc/
PR target/87496
* config/rs6000/rs6000.c (rs6000_option_override_internal): Disallow
-mabi=ieeelongdouble and -mabi=ibmlongdouble without -mlong-double-128.
Do not error for -mabi=ibmlongdouble and no ISA 2.06 support.
* doc/invoke.texi: Document -mabi=ibmlongdouble and -mabi=ieeelongdouble
require -mlong-double-128.
gcc/testsuite/
PR target/87496
* gcc.target/powerpc/pr87496.c: Rename from this...
* gcc.target/powerpc/pr87496-1.c: ...to this. Update comment.
* gcc.target/powerpc/pr87496-2.c: New test.
* gcc.target/powerpc/pr87496-3.c: New test.
From-SVN: r266899
On top of the previous patch that implements TARGET_ESTIMATED_POLY_VALUE
and adds an sve_width tuning field to the CPU structs, this patch implements
an -moverride knob to adjust this sve_width field to allow for experimentation.
Again, reminder that this only has an effect when compiling for VLA-SVE that is,
without msve-vector-bits=<foo>. This just adjusts tuning heuristics in the compiler,,
like profitability thresholds for vectorised versioned loops, and others.
It can be used, for example like -moverride=sve_width=256 to set the sve_width
tuning field to 256. Widths outside of the accepted SVE widths [128 - 2048] are rejected
as you'd expect.
* config/aarch64/aarch64.c (aarch64_tuning_override_functions): Add
sve_width entry.
(aarch64_parse_sve_width_string): Define.
* gcc.target/aarch64/sve/override_sve_width_1.c: New test.
From-SVN: r266898
* gimple-ssa-evrp-analyze.h (class evrp_range_analyzer): Add
m_update_global_ranges member. Add corresponding argument to ctor.
* gimple-ssa-evrp-analyze.c
(evrp_range_analyzer::evrp_range_analyzer): Add new argument and
initialize m_update_global_ranges.
(evrp_range_analyzer::set_ssa_range_info): Assert that we are
updating global ranges.
(evrp_range_analyzer::record_ranges_from_incoming_edge): Only
update global ranges if explicitly requested.
(evrp_range_analyzer::record_ranges_from_phis): Similarly.
(evrp_range_analyzer::record_ranges_from_stmt): Similarly.
* gimple-ssa-evrp.c (evrp_dom_walker): Pass new argument to
evrp_range_analyzer ctor.
* gimple-ssa-sprintf.c (sprintf_dom_walker): Similarly.
* tree-ssa-dom.c (dom_opt_dom_walker): Similarly.
* gcc.c-torture/builtins/strnlen.x: New file to filter -Og from
options to test.
From-SVN: r266897
The hook TARGET_ESTIMATED_POLY_VALUE allows a target to give an estimate for a poly_int run-time value.
It is used exclusively in tuning decisions, things like estimated loop iterations, probabilities etc.
It is not relied on for correctness.
If we know the SVE width implemented in hardware we can make more more
informed decisions in the implementation of TARGET_ESTIMATED_POLY_VALUE,
even when compiling for VLA vectorisation.
This patch adds an sve_width field to our tuning structs and sets it for
the current CPU tunings.
A new value is introduced to the aarch64_sve_vector_bits_enum enum that indicates
that SVE is not available: SVE_NOT_IMPLEMENTED. I set it to the same value as SVE_SCALABLE
so that parts of the aarch64 backend that follow the pattern:
if (vector_width == SVE_SCALABLE)
do_vla_friendly_action ()
else
assume_specific_width_for_correctness ()
continue to work without change, but the CPU tuning structs can use a more
appropriate moniker for indicating the absence of SVE.
This sets sve_width to SVE_NOT_IMPLEMENTED for all cores.
I aim to add an -moverride switch in the next patch that allows a power user to experiment
with different values of it for investigations.
* config/aarch64/aarch64-opts.h (aarch64_sve_vector_bits_enum):
Add SVE_NOT_IMPLEMENTED value.
* config/aarch64/aarch64-protos.h (struct tune_params): Add sve_width
field.
* config/aarch64/aarch64.c (generic_tunings,cortexa35_tunings,
cortexa53_tunings, cortexa57_tunings, cortexa72_tunings,
cortexa73_tunings, exynosm1_tunings, thunderx_tunings,
thunderx_tunings, tsv110_tunings, xgene1_tunings, qdf24xx_tunings,
saphira_tunings, thunderx2t99_tunings, emag_tunings):
Specify sve_width.
(aarch64_estimated_poly_value): Define.
(TARGET_ESTIMATED_POLY_VALUE): Define.
From-SVN: r266896
A recent patch inadvertently added the use of "vector" to mmintrin.h
when all such uses should be "__vector".
[gcc]
2018-12-07 Paul A. Clarke <pc@us.ibm.com>
PR target/88408
* config/rs6000/mmintrin.h (_mm_packs_pu16): Correctly use "__vector".
From-SVN: r266895
PR c++/86669
* call.c (make_temporary_var_for_ref_to_temp): Call pushdecl even for
automatic vars.
* g++.dg/cpp0x/initlist105.C: New test.
* g++.dg/cpp0x/initlist106.C: New test.
* g++.dg/other/pr86669.C: New test.
From-SVN: r266893
When using the unpredicated immediate forms of MUL, LSL, LSR and ASR,
the rtl patterns would still have the predicate operand we created for
the other forms. This patch splits the patterns after reload in order
to get rid of the predicate, like we already do for WHILE.
2018-12-07 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* config/aarch64/aarch64-sve.md (*mul<mode>3, *v<optab><mode>3):
Split the patterns after reload if we don't need the predicate
operand.
(*post_ra_mul<mode>3, *post_ra_v<optab><mode>3): New patterns.
gcc/testsuite/
* gcc.target/aarch64/sve/pred_elim_2.c: New test.
From-SVN: r266892
