H.J. Lu [Fri, 24 Aug 2018 23:37:53 +0000 (23:37 +0000)]
Set start_location to 0 if we ran out of line map space
With profiledbootstrap and --with-build-config=bootstrap-lto, linemap_add
may create a macro map when we run out of line map space. This patch
changes start_location to UNKNOWN_LOCATION (0) in this case.
Tested with profiledbootstrap and --with-build-config=bootstrap-lto on
Linux/x86-64.
PR bootstrap/86872
* line-map.c (pure_location_p): Return true if linemap_lookup
returns NULL.
(linemap_add): Set start_location to 0 if we run out of line map
space.
From-SVN: r263845
David Malcolm [Fri, 24 Aug 2018 21:17:48 +0000 (21:17 +0000)]
diagnostics: tweaks to line-spans vs line numbering (PR 87091)
This patch tweaks how line numbers are printed for a diagnostic
containing multiple line spans.
With this patch, rather than printing span headers:
../x86_64-pc-linux-gnu/libstdc++-v3/include/vector:87:22: note: message
../x86_64-pc-linux-gnu/libstdc++-v3/include/vector:74:1:
++ |+#include <vector>
74 | #endif
../x86_64-pc-linux-gnu/libstdc++-v3/include/vector:87:22:
87 | using vector = std::vector<_Tp, polymorphic_allocator<_Tp>>;
| ^~~
we now print:
../x86_64-pc-linux-gnu/libstdc++-v3/include/vector:87:22: note: message
+++ |+#include <vector>
74 | #endif
....
87 | using vector = std::vector<_Tp, polymorphic_allocator<_Tp>>;
| ^~~
and for sufficiently close lines, rather than print a gap:
+ |+#include <stdio.h>
1 | test (int ch)
..
3 | putchar (ch);
| ^~~~~~~
we print the line itself:
+ |+#include <stdio.h>
1 | test (int ch)
2 | {
3 | putchar (ch);
| ^~~~~~~
gcc/ChangeLog:
PR 87091
* diagnostic-show-locus.c (layout::layout): Ensure the margin is
wide enough for jumps in the line-numbering to be visible.
(layout::print_gap_in_line_numbering): New member function.
(layout::calculate_line_spans): When using line numbering, merge
line spans that are only 1 line apart.
(diagnostic_show_locus): When printing line numbers, show gaps in
line numbering directly, rather than printing headers.
(selftest::test_diagnostic_show_locus_fixit_lines): Add test of
line-numbering with multiple line spans.
(selftest::test_fixit_insert_containing_newline_2): Add test of
line-numbering, in which the spans are close enough to be merged.
gcc/testsuite/ChangeLog:
PR 87091
* gcc.dg/missing-header-fixit-3.c: Update for changes to how
line spans are printed with -fdiagnostics-show-line-numbers.
From-SVN: r263843
Aldy Hernandez [Fri, 24 Aug 2018 18:37:51 +0000 (18:37 +0000)]
gimple-ssa-evrp-analyze.c (set_ssa_range_info): Pass value_range to range_includes_zero_p.
* gimple-ssa-evrp-analyze.c (set_ssa_range_info): Pass value_range
to range_includes_zero_p. Do not special case VR_ANTI_RANGE.
* tree-vrp.c (range_is_nonnull): Remove.
(range_includes_zero_p): Accept value_range instead of min/max.
(extract_range_from_binary_expr_1): Do not early bail on
POINTER_PLUS_EXPR.
Use range_includes_zero_p instead of range_is_nonnull.
(extract_range_from_unary_expr): Use range_includes_zero_p instead
of range_is_nonnull.
(vrp_meet_1): Pass value_range to range_includes_zero_p. Do not
special case VR_ANTI_RANGE.
(vrp_finalize): Same.
* tree-vrp.h (range_includes_zero_p): Pass value_range as argument
instead of min/max.
(range_is_nonnull): Remove.
* vr-values.c (vrp_stmt_computes_nonzero): Use
range_includes_zero_p instead of range_is_nonnull.
(extract_range_basic): Pass value_range to range_includes_zero_p
instead of range_is_nonnull.
From-SVN: r263842
Marc Glisse [Fri, 24 Aug 2018 18:24:42 +0000 (20:24 +0200)]
new(nothrow) is malloc-like
2018-08-24 Marc Glisse <marc.glisse@inria.fr>
PR libstdc++/86822
* libsupc++/new (operator new(size_t, nothrow_t), operator
new[](size_t, nothrow_t), operator new(size_t, align_val_t, nothrow_t),
operator new[](size_t, align_val_t, nothrow_t)): Add malloc attribute.
From-SVN: r263841
Ian Lance Taylor [Fri, 24 Aug 2018 18:15:04 +0000 (18:15 +0000)]
runtime: remove the dummy arg of getcallersp
This is a port of https://golang.org/cl/109596 to the gofrontend, in
preparation for updating libgo to 1.11.
Original CL description:
getcallersp is intrinsified, and so the dummy arg is no longer
needed. Remove it, as well as a few dummy args that are solely
to feed getcallersp.
Reviewed-on: https://go-review.googlesource.com/131116
From-SVN: r263840
Jonathan Wakely [Fri, 24 Aug 2018 17:43:18 +0000 (18:43 +0100)]
Forward declare debug containers so std::pmr aliases work
Prior to this change, including a <debug/xxx> header when _GLIBCXX_DEBUG
is also defined would fail to compile in C++17 or later. The <debug/xxx>
header would include the standard <xxx> header which defined
std::pmr::xxx as an alias for std::xxx. But in Debug Mode std::xxx
refers to std::__debug::xxx which has not been defined yet (because it
is in <debug/xxx> after the inclusion of <xxx>).
This adds declarations of the debug containers before including the
non-Debug Mode <xxx> header, so that the std::pmr::xxx aliases work.
* include/debug/deque (std::__debug::deque): Declare.
* include/debug/forward_list (std::__debug::forward_list): Declare.
* include/debug/list (std::__debug::list): Declare.
* include/debug/map (std::__debug::map): Declare.
* include/debug/set (std::__debug::set): Declare.
* include/debug/unordered_map (std::__debug::unordered_map): Declare.
* include/debug/unordered_set (std::__debug::unordered_set): Declare.
* include/debug/vector (std::__debug::vector): Declare.
* testsuite/23_containers/deque/types/pmr_typedefs_debug.cc: New test.
* testsuite/23_containers/forward_list/pmr_typedefs_debug.cc: New
test.
* testsuite/23_containers/list/pmr_typedefs_debug.cc: New test.
* testsuite/23_containers/map/pmr_typedefs_debug.cc: New test.
* testsuite/23_containers/multimap/pmr_typedefs_debug.cc: New test.
* testsuite/23_containers/multiset/pmr_typedefs_debug.cc: New test.
* testsuite/23_containers/set/pmr_typedefs_debug.cc: New test.
* testsuite/23_containers/unordered_map/pmr_typedefs_debug.cc: New
test.
* testsuite/23_containers/unordered_multimap/pmr_typedefs_debug.cc:
New test.
* testsuite/23_containers/unordered_multiset/pmr_typedefs_debug.cc:
New test.
* testsuite/23_containers/unordered_set/pmr_typedefs_debug.cc: New
test.
* testsuite/23_containers/vector/cons/destructible_debug_neg.cc:
Adjust dg-error lineno.
* testsuite/23_containers/vector/types/pmr_typedefs_debug.cc: New
test.
From-SVN: r263839
Thomas Koenig [Fri, 24 Aug 2018 17:26:57 +0000 (17:26 +0000)]
re PR fortran/86837 (Optimization breaks an unformatted read with implicit loop)
2018-08-24 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/86837
* frontend-passes.c (var_in_expr_callback): New function.
(var_in_expr): New function.
(traverse_io_block): Use var_in_expr instead of
gfc_check_dependency for checking if the variable depends on the
previous interators.
2018-08-24 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/86837
* gfortran.dg/implied_do_io_6.f90: New test.
From-SVN: r263838
H.J. Lu [Fri, 24 Aug 2018 16:58:27 +0000 (16:58 +0000)]
Add a testcase for PR middle-end/87092
PR middle-end/87092
* gcc.dg/pr87092.c: New test.
From-SVN: r263837
Marek Polacek [Fri, 24 Aug 2018 15:48:43 +0000 (15:48 +0000)]
re PR c++/67012 (decltype(auto) with trailing return type)
PR c++/67012
PR c++/86942
* decl.c (grokdeclarator): Disallow functions with trailing return
type with decltype(auto) as its type. Also check the function if
it's inner declarator doesn't exist
* g++.dg/cpp0x/auto52.C: New test.
* g++.dg/cpp1y/auto-fn52.C: New test.
* g++.dg/cpp1y/auto-fn53.C: New test.
* g++.dg/cpp1y/auto-fn54.C: New test.
From-SVN: r263836
Uros Bizjak [Fri, 24 Aug 2018 14:42:32 +0000 (16:42 +0200)]
emit-rtl.c (init_emit_once): Do not emit MODE_POINTER_BOUNDS RTXes.
* emit-rtl.c (init_emit_once): Do not emit MODE_POINTER_BOUNDS RTXes.
* emit-rtl.h (rtl_data): Remove return_bnd.
* explow.c (trunc_int_for_mode): Do not handle POINTER_BOUNDS_MODE_P.
* function.c (diddle_return_value): Do not handle crtl->return_bnd.
* genmodes.c (complete_mode): Do not handle MODE_POINTER_BOUNDS.
(POINTER_BOUNDS_MODE): Remove definition.
(make_pointer_bounds_mode): Remove.
(get_mode_class): Do not handle MODE_POINTER_BOUNDS.
* machmode.h (POINTER_BOUNDS_MODE_P): Remove definition.
(scalare_mode::includes_p): Do not handle MODE_POINTER_BOUNDS.
* mode-classes.def: Do not define MODE_POINTER_BOUNDS.
* stor-layout.c (int_mode_for_mode): Do not handle MODE_POINTER_BOUNDS.
* tree-core.h (enum tree_index): Remove TI_POINTER_BOUNDS_TYPE.
* varasm.c (output_constant_pool_2): Do not handle MODE_POINTER_BOUNDS.
* config/i386/i386-modes.def (BND32, BND64): Remove.
* config/i386/i386.c (dbx_register_map): Remove bound registers.
(dbx64_register_map): Ditto.
(svr4_dbx_register_map): Ditto.
(indirect_thunk_bnd_needed): Remove.
(indirect_thunks_bnd_used): Ditto.
(indirect_return_bnd_needed): Ditto.
(indirect_return_via_cx_bnd): Ditto.
(enum indirect_thunk_prefix): Remove indirect_thunk_prefix_bnd.
(indirect_thunk_name): Remove handling of indirect_thunk_prefix_bnd.
(output_indirect_thunk): Ditto. Remove need_prefix argument.
(output_indirect_thunk_function): Remove handling of
indirect_return_bnd_needed, indirect_return_via_cx_bnd,
indirect_thunk_bnd_needed and indirect_thunks_bnd_used variables.
(ix86_save_reg): Remove handling of crtl->return_bnd.
(ix86_legitimate_constant_p): Remove handling of POINTER_BOUNDS_MODE_P.
(ix86_print_operand_address_as): Remove handling of UNSPEC_BNDMK_ADDR
and UNSPEC_BNDLX_ADDR.
(ix86_output_indirect_branch_via_reg): Remove handling of
indirect_thunk_prefix_bnd.
(ix86_output_indirect_branch_via_push): Ditto.
(ix86_output_function_return): Ditto.
(ix86_output_indirect_function_return): Ditto.
(avoid_func_arg_motion): Do not handle UNSPEC_BNDSTX.
* config/i386/i386.h (FIXED_REGISTERS): Remove bound registers.
(CALL_USED_REGISTERS): Ditto.
(REG_ALLOC_ORDER): Update for removal of bound registers.
(HI_REGISTER_NAMES): Ditto.
* config/i386/i386.md (UNSPEC_BNDMK, UNSPEC_BNDMK_ADDR, UNSPEC_BNDSTX)
(UNSPEC_BNDLDX, UNSPEC_BNDLDX_ADDR, UNSPEC_BNDCL, UNSPEC_BNDCU)
(UNSPEC_BNDCN, UNSPEC_MPX_FENCE): Remove.
(BND0_REG, BND1_REG, BND2_REG, BND3_REG): Remove
(FIRST_PSEUDO_REG): Update.
(BND): Remove mode iterator.
* config/i386/predicates.md (bnd_mem_operator): Remove.
From-SVN: r263835
Richard Sandiford [Fri, 24 Aug 2018 13:06:56 +0000 (13:06 +0000)]
Define vect_perm for variable-length SVE
Variable-length SVE now supports enough permutes to define vect_perm.
The change to vect_perm_supported is currently a no-op because the
function is only called with a count of 3.
2018-08-24 Richard Sandiford <richard.sandiford@arm.com>
gcc/testsuite/
* lib/target-supports.exp (vect_perm_supported): Only return
false for variable-length vectors if the permute size is not
a power of 2.
(check_effective_target_vect_perm)
(check_effective_target_vect_perm_byte)
(check_effective_target_vect_perm_short): Remove check for
variable-length vectors.
* gcc.dg/vect/slp-23.c: Add an XFAIL for variable-length SVE.
* gcc.dg/vect/slp-perm-10.c: Likewise.
* gcc.dg/vect/slp-perm-9.c: Add an XFAIL for variable-length vectors.
From-SVN: r263834
Richard Sandiford [Fri, 24 Aug 2018 13:06:09 +0000 (13:06 +0000)]
Avoid is_constant calls in vectorizable_bswap
The "new" VEC_PERM_EXPR handling makes it easy to support bswap
for variable-length vectors.
2018-08-24 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* tree-vect-stmts.c (vectorizable_bswap): Handle variable-length
vectors.
gcc/testsuite/
* gcc.target/aarch64/sve/bswap_1.c: New test.
* gcc.target/aarch64/sve/bswap_2.c: Likewise.
* gcc.target/aarch64/sve/bswap_3.c: Likewise.
From-SVN: r263833
Richard Sandiford [Fri, 24 Aug 2018 13:05:36 +0000 (13:05 +0000)]
Handle SLP permutations for variable-length vectors
The SLP code currently punts for all variable-length permutes.
This patch makes it handle the easy case of N->N permutes in which
the number of vector lanes is a multiple of N. Every permute then
uses the same mask, and that mask repeats (with a stride) every
N elements.
The patch uses the same path for constant-length vectors,
since it should be slightly cheaper in terms of compile time.
2018-08-24 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* tree-vect-slp.c (vect_transform_slp_perm_load): Separate out
the case in which the permute needs only a single element and
repeats for every vector of the result. Extend that case to
handle variable-length vectors.
* tree-vect-stmts.c (vectorizable_load): Update accordingly.
gcc/testsuite/
* gcc.target/aarch64/sve/slp_perm_1.c: New test.
* gcc.target/aarch64/sve/slp_perm_2.c: Likewise.
* gcc.target/aarch64/sve/slp_perm_3.c: Likewise.
* gcc.target/aarch64/sve/slp_perm_4.c: Likewise.
* gcc.target/aarch64/sve/slp_perm_5.c: Likewise.
* gcc.target/aarch64/sve/slp_perm_6.c: Likewise.
* gcc.target/aarch64/sve/slp_perm_7.c: Likewise.
From-SVN: r263832
H.J. Lu [Fri, 24 Aug 2018 11:28:40 +0000 (11:28 +0000)]
DWARF: Call set_indirect_string on DW_MACINFO_start_file
Since -gsplit-dwarf -g3 will output filename as indirect string, call
set_indirect_string on DW_MACINFO_start_file for -gsplit-dwarf -g3.
gcc/
PR debug/79342
* dwarf2out.c (save_macinfo_strings): Call set_indirect_string
on DW_MACINFO_start_file for -gsplit-dwarf -g3.
gcc/testsuite/
PR debug/79342
* gcc.dg/pr79342.: New test.
From-SVN: r263831
Richard Biener [Fri, 24 Aug 2018 11:17:16 +0000 (11:17 +0000)]
cfg.h (struct control_flow_graph): Add edge_flags_allocated and bb_flags_allocated members.
2018-08-24 Richard Biener <rguenther@suse.de>
* cfg.h (struct control_flow_graph): Add edge_flags_allocated and
bb_flags_allocated members.
(auto_flag): New RAII class for allocating flags.
(auto_edge_flag): New RAII class for allocating edge flags.
(auto_bb_flag): New RAII class for allocating bb flags.
* cfgloop.c (verify_loop_structure): Allocate temporary edge
flag dynamically.
* cfganal.c (dfs_enumerate_from): Remove use of visited sbitmap
in favor of temporarily allocated BB flag.
* hsa-brig.c: Re-order includes.
* hsa-dump.c: Likewise.
* hsa-regalloc.c: Likewise.
* print-rtl.c: Likewise.
* profile-count.c: Likewise.
From-SVN: r263830
Segher Boessenkool [Fri, 24 Aug 2018 10:49:27 +0000 (12:49 +0200)]
rs6000: Check that the base of a TOCREL is the TOC (PR86989)
There currently is nothing that prevents replacing the TOC_REGISTER in
a TOCREL unspec with something else, like a pseudo, or a memory ref.
This of course does not work. Fix that.
Tested on powerpc64-linux {-m32,-m64}; committing.
Segher
2018-08-24 Segher Boessenkool <segher@kernel.crashing.org>
PR target/86989
* config/rs6000/rs6000.c (toc_relative_expr_p): Check that the base is
the TOC register.
---
gcc/config/rs6000/rs6000.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/gcc/config/rs6000/rs6000.c b/gcc/config/rs6000/rs6000.c
index
a967912..
ed33912 100644
--- a/gcc/config/rs6000/rs6000.c
+++ b/gcc/config/rs6000/rs6000.c
@@ -7932,7 +7932,9 @@ toc_relative_expr_p (const_rtx op, bool strict, const_rtx *tocrel_base_ret,
*tocrel_offset_ret = tocrel_offset;
return (GET_CODE (tocrel_base) == UNSPEC
- && XINT (tocrel_base, 1) == UNSPEC_TOCREL);
+ && XINT (tocrel_base, 1) == UNSPEC_TOCREL
+ && REG_P (XVECEXP (tocrel_base, 0, 1))
+ && REGNO (XVECEXP (tocrel_base, 0, 1)) == TOC_REGISTER);
}
/* Return true if X is a constant pool address, and also for cmodel=medium
--
1.8.3.1
From-SVN: r263829
Aldy Hernandez [Fri, 24 Aug 2018 08:06:06 +0000 (08:06 +0000)]
PR 87073/bootstrap
PR 87073/bootstrap
* wide-int-range.cc (wide_int_range_div): Do not ignore result
from wide_int_range_multiplicative_op.
From-SVN: r263828
GCC Administrator [Fri, 24 Aug 2018 00:16:48 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r263827
Martin Sebor [Thu, 23 Aug 2018 19:39:18 +0000 (19:39 +0000)]
PR tree-optimization/87072 - false warning: array subscript is above array bounds
gcc/testsuite/ChangeLog:
* gcc.dg/Warray-bounds-35.c: New test.
From-SVN: r263822
David Edelsohn [Thu, 23 Aug 2018 19:33:45 +0000 (19:33 +0000)]
* async.h (ASYNC_IO): Revert _AIX test.
From-SVN: r263821
Jonathan Wakely [Thu, 23 Aug 2018 19:27:54 +0000 (20:27 +0100)]
Fix failure when -fno-rtti test is run in C++17 or later
* testsuite/util/testsuite_allocator.h (__gnu_test::memory_resource):
Only define when RTTI is enabled.
From-SVN: r263820
Prathamesh Kulkarni [Thu, 23 Aug 2018 18:22:10 +0000 (18:22 +0000)]
tree-vect-data-refs.c (vect_grouped_store_supported): Fix typo "permutaion".
2018-08-23 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
* tree-vect-data-refs.c (vect_grouped_store_supported): Fix typo
"permutaion".
From-SVN: r263819
Giuliano Belinassi [Thu, 23 Aug 2018 17:01:09 +0000 (17:01 +0000)]
Fix typo 'exapnded' to 'expanded'
2018-08-23 Giuliano Belinassi <giuliano.belinassi@usp.br>
gcc/
* genmatch.c (parser::parse_operation): Fix typo 'exapnded'
to 'expanded'.
gcc/ada/
* exp_unst.ads: Fix typo 'exapnded' to 'expanded'.
From-SVN: r263818
Alexander Monakov [Thu, 23 Aug 2018 16:01:12 +0000 (19:01 +0300)]
scev: dump final value replacement expressions
* tree-scalar-evolution.c (final_value_replacement_loop): Dump
full GENERIC expression used for replacement.
From-SVN: r263817
Jonathan Wakely [Thu, 23 Aug 2018 15:23:31 +0000 (16:23 +0100)]
Define debug mode function for C++98
This function is declared unconditionally but was only defined for C++11
and later, leading to linker errors when the testsuite was run with
-std=gnu++98 -D_GLIBCXX_DEBUG added to the flags.
* include/debug/vector (__niter_base): Define for C++98.
From-SVN: r263816
Jonathan Wakely [Thu, 23 Aug 2018 15:10:45 +0000 (16:10 +0100)]
Fix C++98 tests to not use C++11 features.
* testsuite/25_algorithms/partial_sort_copy/debug/irreflexive_neg.cc:
Fix C++98 test to not use C++11 features.
* testsuite/25_algorithms/fill_n/2.cc: Likewise.
From-SVN: r263815
Jonathan Wakely [Thu, 23 Aug 2018 15:10:35 +0000 (16:10 +0100)]
Fix comments in testsuite script
* scripts/check_compile: Fix comments.
From-SVN: r263814
Aldy Hernandez [Thu, 23 Aug 2018 13:25:36 +0000 (13:25 +0000)]
tree-vrp.c (abs_extent_range): Remove.
* tree-vrp.c (abs_extent_range): Remove.
(extract_range_into_wide_ints): Pass wide ints by reference.
(extract_range_from_binary_expr_1): Rewrite the *DIV_EXPR code.
Pass wide ints by reference in all calls to
extract_range_into_wide_ints.
* wide-int-range.cc (wide_int_range_div): New.
* wide-int-range.h (wide_int_range_div): New.
(wide_int_range_includes_zero_p): New.
(wide_int_range_zero_p): New.
From-SVN: r263813
Matthew Malcomson [Thu, 23 Aug 2018 12:43:17 +0000 (12:43 +0000)]
[AARCH64] use "arch_enabled" attribute for aarch64.
arm.md has some attributes "arch" and "arch_enabled" to aid enabling
and disabling insn alternatives based on the architecture being
targeted. This patch introduces a similar attribute in the aarch64
backend. The new attribute will be used to enable a new alternative
for the atomic_store insn in a future patch, but is an atomic change
in itself.
The new attribute has values "any", "fp", "fp16", "simd", and "sve".
These attribute values have been taken from the pre-existing
attributes "fp", "fp16", "simd", and "sve".
The standalone "fp" attribute has been reintroduced in terms of the
"arch" attribute as it's needed for the xgene1.md scheduling file --
the use in this file can't be changed to check for `(eq_attr "arch"
"fp")` as the file is reused by the arm.md machine description whose
'arch' attribute doesn't have an 'fp' value.
2018-08-23 Matthew Malcomson <matthew.malcomson@arm.com>
* config/aarch64/aarch64.md (arches): New enum.
(arch): New enum attr.
(arch_enabled): New attr.
(enabled): Now uses arch_enabled only.
(simd, sve, fp16): Removed attribute.
(fp): Attr now defined in terms of 'arch'.
(*mov<mode>_aarch64, *movsi_aarch64, *movdi_aarch64, *movti_aarch64,
*movhf_aarch64, <optab><fcvt_target><GPF:mode>2,
<FCVT_F2FIXED:fcvt_fixed_insn><GPF:mode>3,
<FCVT_FIXED2F:fcvt_fixed_insn><GPI:mode>3): Merge 'fp' and 'simd'
attributes into 'arch'.
(*movsf_aarch64, *movdf_aarch64, *movtf_aarch64, *add<mode>3_aarch64,
subdi3, neg<mode>2, <optab><mode>3, one_cmpl<mode>2,
*<NLOGICAL:optab>_one_cmpl<mode>3, *xor_one_cmpl<mode>3,
*aarch64_ashl_sisd_or_int_<mode>3, *aarch64_lshr_sisd_or_int_<mode>3,
*aarch64_ashr_sisd_or_int_<mode>3, *aarch64_sisd_ushl): Convert use of
'simd' attribute into 'arch'.
(load_pair_sw_<SX:mode><SX2:mode>, load_pair_dw_<DX:mode><DX2:mode>,
store_pair_sw_<SX:mode><SX2:mode>, store_pair_dw_<DX:mode><DX2:mode>):
Convert use of 'fp' attribute to 'arch'.
* config/aarch64/aarch64-simd.md (move_lo_quad_internal_<mode>,
move_lo_quad_internal_<mode>): (different modes) Merge 'fp' and 'simd'
into 'arch'.
(move_lo_quad_internal_be_<mode>, move_lo_quad_internal_be_<mode>):
(different modes) Merge 'fp' and 'simd' into 'arch'.
(*aarch64_combinez<mode>, *aarch64_combinez_be<mode>): Merge 'fp' and
'simd' into 'arch'.
From-SVN: r263811
Segher Boessenkool [Thu, 23 Aug 2018 12:40:14 +0000 (14:40 +0200)]
Fix recent bug in canonicalize_comparison (PR87026)
The new code testing which way a comparison is best expressed creates
a pseudoregister (by hand) and creates some insns with that. Such
insns will no longer recog() when pseudo-registers are no longer
aloowed (after reload). But we have an ifcvt pass after reload (ce3).
This patch simply returns if we cannot create pseudos.
PR rtl-optimization/87026
* expmed.c (canonicalize_comparison): If we can no longer create
pseudoregisters, don't.
From-SVN: r263810
Jonathan Wakely [Thu, 23 Aug 2018 11:21:17 +0000 (12:21 +0100)]
Fix C++11-ism in C++98 member function
* include/debug/string (insert(__const_iterator, _InIter, _InIter)):
[!_GLIBCXX_USE_CXX11_ABI]: Replace use of C++11-only cbegin() with
begin(), for C++98 compatibility.
From-SVN: r263809
Jonathan Wakely [Thu, 23 Aug 2018 10:51:52 +0000 (11:51 +0100)]
Fix testsuite failures for __gnu_debug::string with old ABI
The __gnu_debug string (mostly) implements the C++11 API, but when it
wraps the old COW string many of the member functions in the base class
have the wrong parameter types or return types. This makes the
__gnu_debug::string type adapt itself to the base class API. This
actually makes the debug string slightly more conforming than the
underlying string type when using the old ABI.
* include/bits/basic_string.h [_GLIBCXX_USE_CXX11_ABI]
(basic_string::__const_iterator): Change access to protected.
[!_GLIBCXX_USE_CXX11_ABI] (basic_string::__const_iterator): Define
as typedef for iterator.
* include/debug/string (__const_iterator): Use typedef from base.
(insert(const_iterator, _CharT))
(replace(const_iterator, const_iterator, const basic_string&))
(replace(const_iterator, const_iterator, const _CharT*, size_type))
(replace(const_iterator, const_iterator, const CharT*))
(replace(const_iterator, const_iterator, size_type, _CharT))
(replace(const_iterator, const_iterator, _InputIter, _InputIter))
(replace(const_iterator, const_iterator, initializer_list<_CharT>)):
Change const_iterator parameters to __const_iterator.
(insert(iterator, size_type, _CharT)): Add C++98 overload.
(insert(const_iterator, _InputIterator, _InputIterator)): Change
const_iterator parameter to __const_iterator.
[!_GLIBCXX_USE_CXX11_ABI]: Add workaround for incorrect return type
of base's member function.
(insert(const_iterator, size_type, _CharT)) [!_GLIBCXX_USE_CXX11_ABI]:
Likewise.
(insert(const_iterator, initializer_list<_CharT>))
[!_GLIBCXX_USE_CXX11_ABI]: Likewise.
* testsuite/21_strings/basic_string/init-list.cc: Remove effective
target directive.
From-SVN: r263808
Nathan Sidwell [Thu, 23 Aug 2018 10:21:21 +0000 (10:21 +0000)]
[libiberty patch] Fix PGO bootstrap
https://gcc.gnu.org/ml/gcc-patches/2018-08/msg01386.html
PR driver/87056
* pex-unix.c (pex_unix_exec_child): Duplicate bad_fn into local
scopes to avoid potential clobber.
Co-Authored-By: Martin Liska <mliska@suse.cz>
From-SVN: r263807
Richard Earnshaw [Thu, 23 Aug 2018 09:47:34 +0000 (09:47 +0000)]
PR target/86951 arm - Handle speculation barriers on pre-armv7 CPUs
The AArch32 instruction sets prior to Armv7 do not define the ISB and
DSB instructions that are needed to form a speculation barrier. While
I do not know of any instances of cores based on those instruction
sets being vulnerable to speculative side channel attacks it is
possible to run code built for those ISAs on more recent hardware
where they would become vulnerable.
This patch works around this by using a library call added to libgcc.
That code can then take any platform-specific actions necessary to
ensure safety.
For the moment I've only handled two cases: the library code being
built for armv7 or later anyway and running on Linux.
On Linux we can handle this by calling the kernel function that will
flush a small amount of cache. Such a sequence ends with a ISB+DSB
sequence if running on an Armv7 or later CPU.
gcc:
PR target/86951
* config/arm/arm-protos.h (arm_emit_speculation_barrier): New
prototype.
* config/arm/arm.c (speculation_barrier_libfunc): New static
variable.
(arm_init_libfuncs): Initialize it.
(arm_emit_speculation_barrier): New function.
* config/arm/arm.md (speculation_barrier): Call
arm_emit_speculation_barrier for architectures that do not have
DSB or ISB.
(speculation_barrier_insn): Only match on Armv7 or later.
libgcc:
PR target/86951
* config/arm/lib1funcs.asm (speculation_barrier): New function.
* config/arm/t-arm (LIB1ASMFUNCS): Add it to list of functions
to build.
From-SVN: r263806
Richard Biener [Thu, 23 Aug 2018 09:29:45 +0000 (09:29 +0000)]
re PR ipa/87024 (ICE in fold_stmt_1)
2018-08-23 Richard Biener <rguenther@suse.de>
PR middle-end/87024
* tree-inline.c (copy_bb): Drop unused __builtin_va_arg_pack_len
calls.
* gcc.dg/pr87024.c: New testcase.
From-SVN: r263805
Richard Sandiford [Thu, 23 Aug 2018 08:57:17 +0000 (08:57 +0000)]
[AArch64] Improve SVE handling of single-vector permutes
aarch64_vectorize_vec_perm_const was failing to set one_vector_p
if the permute had only a single input. This in turn was hiding
a problem in the SVE TBL handling: it accepted single-vector
variable-length permutes, but sent them through the general
two-vector aarch64_expand_sve_vec_perm, which is only set up
to handle constant-length permutes.
2018-08-23 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* config/aarch64/aarch64.c (aarch64_evpc_sve_tbl): Fix handling
of single-vector TBLs.
(aarch64_vectorize_vec_perm_const): Set one_vector_p when only
one input is given.
gcc/testsuite/
* gcc.dg/vect/no-vfa-vect-depend-2.c: Remove XFAIL.
* gcc.dg/vect/no-vfa-vect-depend-3.c: Likewise.
* gcc.dg/vect/pr65947-13.c: Update for vect_fold_extract_last.
* gcc.dg/vect/pr80631-2.c: Likewise.
From-SVN: r263804
Richard Sandiford [Thu, 23 Aug 2018 08:51:38 +0000 (08:51 +0000)]
Fix aarch64_evpc_tbl guard (PR 85910)
This patch fixes a typo in aarch64_expand_vec_perm_const_1 that I
introduced as part of the SVE changes. I don't know of any cases in
which it has any practical effect, since we'll eventually try to use
TBL as a variable permute instead. Having the code is still an
important part of defining the interface properly and so we shouldn't
simply drop it.
2018-08-23 Richard Sandiford <richard.sandiford@arm.com>
gcc/
PR target/85910
* config/aarch64/aarch64.c (aarch64_expand_vec_perm_const_1): Fix
aarch64_evpc_tbl guard.
From-SVN: r263803
Janne Blomqvist [Thu, 23 Aug 2018 08:35:17 +0000 (11:35 +0300)]
Signed zero for {max,min}val intrinsics
The Fortran standard specifies (e.g. F2018 7.4.3.2) that intrinsic
procedures shall treat positive and negative real zero as equivalent,
unless it is explicitly specified otherwise. For {max,min}val there
is no such explicit mention. Thus, remove code to handle signed
zeros.
2018-08-23 Janne Blomqvist <blomqvist.janne@gmail.com>
* trans-intrinsic.c (gfc_conv_intrinsic_minmaxval): Delete
HONOR_SIGNED_ZEROS checks.
From-SVN: r263802
Jonathan Wakely [Thu, 23 Aug 2018 08:26:22 +0000 (09:26 +0100)]
Remove C++14-isms from C++11 tests
* testsuite/20_util/reference_wrapper/lwg2993.cc: Fix C++11 test to
not use C++14 feature.
* testsuite/23_containers/list/68222_neg.cc: Likewise.
From-SVN: r263801
Jonathan Wakely [Thu, 23 Aug 2018 08:26:16 +0000 (09:26 +0100)]
Skip tests that depend on the cxx11 std::string
* testsuite/21_strings/basic_string/init-list.cc:
Require cxx11-abi.
* testsuite/experimental/polymorphic_allocator/pmr_typedefs_match.cc:
Likewise.
* testsuite/experimental/polymorphic_allocator/pmr_typedefs_string.cc:
Likewise.
From-SVN: r263800
Paul Thomas [Thu, 23 Aug 2018 06:27:54 +0000 (06:27 +0000)]
re PR fortran/86863 ([OOP][F2008] type-bound module procedure name not recognized)
2017-08-23 Paul Thomas <pault@gcc.gnu.org>
PR fortran/86863
* resolve.c (resolve_typebound_call): If the TBP is not marked
as a subroutine, check the specific symbol.
2017-08-23 Paul Thomas <pault@gcc.gnu.org>
PR fortran/86863
* gfortran.dg/submodule_32.f08: New test.
From-SVN: r263799
GCC Administrator [Thu, 23 Aug 2018 00:17:18 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r263798
Bernd Edlinger [Wed, 22 Aug 2018 23:43:44 +0000 (23:43 +0000)]
tree-ssa-dse.c (compute_trims): Avoid folding away undefined behaviour.
* tree-ssa-dse.c (compute_trims): Avoid folding away undefined
behaviour.
From-SVN: r263793
Jonathan Wakely [Wed, 22 Aug 2018 23:43:43 +0000 (00:43 +0100)]
Fix tests that fail in C++98 mode
* testsuite/23_containers/deque/capacity/max_size.cc: Fix test for
C++98 mode.
* testsuite/23_containers/deque/modifiers/assign/1.cc: Likewise.
* testsuite/23_containers/list/modifiers/assign/1.cc: Likewise.
* testsuite/23_containers/vector/bool/modifiers/assign/1.cc: Likewise.
* testsuite/23_containers/vector/capacity/max_size.cc: Likewise.
* testsuite/23_containers/vector/modifiers/assign/1.cc: Likewise.
From-SVN: r263792
Jonathan Wakely [Wed, 22 Aug 2018 22:54:33 +0000 (23:54 +0100)]
PR libstdc++/87061 remove pmr type aliases for COW strings
The pmr aliases for basic_string and match_results are incompatible with
the gcc4-compatible ABI because the Copy-On-Write basic_string class
doesn't support C++11 allocators.
PR libstdc++/87061
* include/experimental/regex [!_GLIBCXX_USE_CXX11_ABI]
(experimental::pmr::match_results, experimental::pmr::cmatch)
(experimental::pmr::smatch, experimental::pmr::wcmatch)
(experimental::pmr::wsmatch): Do not declare for gcc4-compatible ABI,
because COW strings don't support C++11 allocator model.
* include/experimental/string [!_GLIBCXX_USE_CXX11_ABI]
(experimental::pmr::basic_string, experimental::pmr::string)
(experimental::pmr::u16string, experimental::pmr::u32string)
(experimental::pmr::wstring): Likewise.
* include/std/regex [!_GLIBCXX_USE_CXX11_ABI] (pmr::match_results)
(pmr::cmatch, pmr::smatch, pmr::wcmatch, pmr::wsmatch): Likewise.
* include/std/string [!_GLIBCXX_USE_CXX11_ABI] (pmr::basic_string)
(pmr::string, pmr::u16string, pmr::u32string, pmr::wstring): Likewise.
* testsuite/21_strings/basic_string/types/pmr_typedefs.cc: Require
cxx11-abi.
* testsuite/28_regex/match_results/pmr_typedefs.cc: Likewise.
From-SVN: r263791
Jonathan Wakely [Wed, 22 Aug 2018 22:22:40 +0000 (23:22 +0100)]
PR libstdc++/78448 limit vector::max_size and deque::max_size
The container requirements imply that max_size() can't exceed the
maximum value of the container's difference_type. Enforce this for
std::vector and std::deque, and add checks to ensure the container
doesn't grow larger than that.
PR libstdc++/78448
* include/bits/deque.tcc (deque::_M_range_initialize): Use
_S_check_init_len to check size.
(deque::_M_push_back_aux, deque::_M_push_front_aux): Throw length
error if size would exceed max_size().
* include/bits/stl_deque.h (_Deque_base::size_type): Remove typedef.
(_Deque_base(_Deque_base&&, const allocator_type&, size_t)): Use
size_t instead of size_type.
(deq(size_type, const allocator_type&)
(deq(size_type, const value_type&, const allocator_type&)
(deque::_M_initialize_dispatch): Use _S_check_init_len to check size.
(deque::max_size): Call _S_max_size.
(deque::_S_check_init_len, deque::_S_max_size): New functions.
* include/bits/stl_vector.h (vector(size_type, const allocator_type&))
(vector(size_type, const value_type&, const allocator_type&))
(vector::_M_initialize_dispatch, vector::_M_range_initialize): Use
_S_check_init_len to check size.
(vector::max_size): Call _S_max_size.
(vector::_M_check_len): Prevent max from being expanded as a
function-like macro.
(vector::_S_check_init_len, vector::_S_max_size): New functions.
* include/bits/vector.tcc (vector::_M_assign_aux): Use
_S_check_init_len to check size.
* testsuite/23_containers/deque/capacity/max_size.cc: New test.
* testsuite/23_containers/vector/capacity/max_size.cc: New test.
From-SVN: r263789
Thomas Koenig [Wed, 22 Aug 2018 21:29:49 +0000 (21:29 +0000)]
gfortran.texi: Mention that asynchronous I/O does not work on systems which lack condition...
2018-08-22 Thomas Koenig <tkoenig@gcc.gnu.org>
* gfortran.texi: Mention that asynchronous I/O does
not work on systems which lack condition variables, such
as AIX.
2018-08-22 Thomas Koenig <tkoenig@gcc.gnu.org>
* async.h: Set ASYNC_IO to zero if _AIX is defined.
(struct adv_cond): If ASYNC_IO is zero, the struct has no members.
(async_unit): If ASYNC_IO is zero, remove unneeded members.
From-SVN: r263788
Janus Weil [Wed, 22 Aug 2018 19:31:40 +0000 (21:31 +0200)]
re PR fortran/86935 (Bad locus in ASSOCIATE statement)
fix PR 86935
2018-08-22 Janus Weil <janus@gcc.gnu.org>
PR fortran/86935
* match.c (gfc_match_associate): Improve diagnostics for the ASSOCIATE
statement.
2018-08-22 Janus Weil <janus@gcc.gnu.org>
PR fortran/86935
* gfortran.dg/associate_3.f90: Update error message.
* gfortran.dg/associate_39.f90: New test case.
From-SVN: r263787
François Dumont [Wed, 22 Aug 2018 18:51:25 +0000 (18:51 +0000)]
re PR libstdc++/68222 (_Safe_iterator provides operators the wrapped iterator can't actually support)
2018-08-22 François Dumont <fdumont@gcc.gnu.org>
PR libstdc++/68222
* include/debug/safe_iterator.h
(_Safe_iterator<_It, _Sq, _Cat>): Add category template parameter.
(_Safe_iterator<>::_Const_iterator): Remove.
(_Safe_iterator<>::_IsConstant): New.
(_Safe_iterator<>::_OtherIterator): New.
(_Safe_iterator<_It, _Sq, _Cat>::_Safe_iterator<_MutIte>(
const _Safe_iterator<_MutIte, _Sq, _Cat>&)): Add _IsConstant::__value in
__gnu_cxx::__enable_if condition.
(_Safe_iterator<_It, _Sq, _Cat>::_M_get_distance_to): New.
(_Safe_iterator<_It, _Sq, _Cat>::_M_get_distance_from_begin): New.
(_Safe_iterator<_It, _Sq, _Cat>::_M_get_distance_to_end): New.
(_Safe_iterator<_It, _Sq, std::bidirectional_iterator_tag>): New.
(_Safe_iterator<_It, _Sq, _Cat>::operator--()): Move...
(_Safe_iterator<_It, _Sq, std::bidirectional_iterator_tag>
::operator--()): ...here.
(_Safe_iterator<_It, _Sq, _Cat>::operator--(int)): Move...
(_Safe_iterator<_It, _Sq, std::bidirectional_iterator_tag>
::operator--(int)): ...here.
(_Safe_iterator<_It, _Sq, _Cat>::_M_decrementable()): Move...
(_Safe_iterator<_It, _Sq, std::bidirectional_iterator_tag>
::_M_decrementable()): ...here.
(_Safe_iterator<_It, _Sq, std::random_access_iterator_tag>): New.
(_Safe_iterator<_It, _Sq, _Cat>::operator[](const difference_type&)):
Move...
(_Safe_iterator<_It, _Sq, std::random_access_iterator_tag>
::operator[](const difference_type&)): ...here.
(_Safe_iterator<_It, _Sq, _Cat>::operator+=(const difference_type&)):
Move...
(_Safe_iterator<_It, _Sq, std::random_access_iterator_tag>
::operator+=(const difference_type&)): ...here.
(_Safe_iterator<_It, _Sq, _Cat>::operator+(const difference_type&)):
Move...
(_Safe_iterator<_It, _Sq, std::random_access_iterator_tag>
::operator+(const difference_type&)): ...here.
(_Safe_iterator<_It, _Sq, _Cat>::operator-=(const difference_type&)):
Move...
(_Safe_iterator<_It, _Sq, std::random_access_iterator_tag>
::operator-=(const difference_type&)): ...here.
(_Safe_iterator<_It, _Sq, _Cat>::operator-(const difference_type&)):
Move...
(_Safe_iterator<_It, _Sq, std::random_access_iterator_tag>
::operator-(const difference_type&)): ...here.
(operator<(const _Safe_iterator<>&, const _Safe_iterator<>&)):
Constraint to random access iterators.
(operator<=(const _Safe_iterator<>&, const _Safe_iterator<>&)):
Likewise.
(operator>(const _Safe_iterator<>&, const _Safe_iterator<>&)): Likewise.
(operator>=(const _Safe_iterator<>&, const _Safe_iterator<>&)):
Likewise.
(operator-(const _Safe_iterator<>&, const _Safe_iterator<>&)): Likewise.
(operator+(const difference_type&, const _Safe_iterator<>&)): Likewise.
(__check_dereferenceable(const _Safe_iterator<>&)): Remove.
(__get_distance): Remove.
(__get_distance_from_begin): Remove.
(__get_distance_to_end): Remove.
(struct __is_safe_random_iterator<_Safe_iterator<>>): Remove partial
specialization.
(__base(const _Safe_iterator<>&, std::input_iterator_tag)): Remove.
(__base(const _Safe_iterator<>&, std::random_access_iterator_tag)): Remove.
(__base(const _Safe_iterator<>&)): Constraint to random access iterator.
* include/debug/safe_iterator.tcc
(_Safe_iterator<>::_M_get_distance_from_begin()): New.
(_Safe_iterator<>::_M_get_distance_to_end()): New.
(_Safe_iterator<>::_M_get_distance_to(const _Safe_iterator<>&)): New.
(_Safe_iterator<_It, _Seq, std::random_access_iterator_tag>
::_M_valid_range): New.
* include/debug/safe_local_iterator.h
(_Safe_local_iterator<>::_Const_local_iterator): Remove.
(_Safe_local_iterator<>::_IsConstant): New.
(_Safe_local_iterator<>::_OtherIterator): New.
(_Safe_local_iterator<_It, _Cont>::_Safe_local_iterator<_MutIte, _Cont>(
const _Safe_local_iterator<_MutIte, _Seq>&)): Add _IsConstant::__value
in __gnu_cxx::__enable_if condition. If singular compare base iterator
with _MutIte rather than _It.
(_Safe_local_iterator<>::_S_constant): Make constexpr.
(_Safe_local_iterator<>::_M_get_distance_to): New.
(__check_dereferenceable(const _Safe_local_iterator<>&)): Remove.
(__get_distance(const _Safe_local_iterator<>&,
const _Safe_local_iterator<>&, std::input_iterator_tag)): Remove.
(__valid_range(const _Safe_local_iterator<>&,
const _Safe_local_iterator<>&)): New.
* include/debug/safe_local_iterator.tcc
(_Safe_local_iterator<>::_M_get_distance_to): New.
* include/debug/deque (std::__debug::deque<>): Add
::__gnu_debug::_Safe_iterator<> friend declaration.
* include/debug/forward_list (std::__debug::forward_list<>): Likewise.
* include/debug/list (std::__debug::list<>): Likewise.
* include/debug/map.h (std::__debug::map<>): Likewise.
* include/debug/multimap.h (std::__debug::multimap<>): Likewise.
* include/debug/set.h (std::__debug::set<>): Likewise.
* include/debug/multiset.h (std::__debug::multiset<>): Likewise.
* include/debug/string (std::__debug::basic_string<>): Likewise.
* include/debug/unordered_map (std::__debug::unordered_map<>): Likewise
and add ::__gnu_debug::_Safe_local_iterator<> friend declaration.
(std::__debug::unordered_multimap<>): Likewise.
* include/debug/unordered_set (std::__debug::unordered_set<>): Likewise.
(std::__debug::unordered_multiset<>): Likewise.
* include/debug/formatter.h: Adapt.
* include/debug/helper_functions.h
(__gnu_debug::_Safe_local_iterator<>): Add declaration.
(__get_distance<_Ite>(_Ite, _Ite, std::random_access_iterator_tag):
Pass parameter by copy.
(__get_distance<_Ite>(_Ite, _Ite, std::input_iterator_tag): Likewise.
(__get_distance<_Ite>(_Ite, _Ite): Likewise.
(__valid_range_aux<_Integral>): Pass _Integral by copy.
(__valid_range<_InputIterator>): Pass _InputIterator by copy.
(__valid_range<>(const _Safe_iterator<>&,
const _Safe_iterator<>&, typename _Distance_traits<>::__type&)):
Declare.
(__valid_range(const _Safe_local_iterator<>&,
const _Safe_local_iterator<>&, typename _Distance_traits<>::__type&)):
Declare.
(__valid_range<>(const _Safe_iterator<>&, const _Safe_iterator<>&)):
Declare.
(__valid_range(const _Safe_local_iterator<>&, const _Safe_local_iterator<>&)):
Declare.
(__can_advance): Adapt.
(struct __is_safe_random_iterator<>): Remove.
(struct _SIter_base<>): Remove.
* include/debug/functions.h: Include <bits/stl_iterator.h>.
(__check_dereferenceable): Remove.
(__foreign_iterator_aux4, __foreign_iterator_aux3): Adapt.
(__foreign_iterator_aux2, __foreign_iterator_aux): Adapt.
(__foreign_iterator): Adapt.
* include/debug/stl_iterator.h
(__is_safe_random_iterator<std::reverse_iterator<>>): Remove.
(__base(const std::reverse_iterator<_Safe_iterator<_It, _Sq>)):
Constraint for random access iterators.
(__niter_base): Adapt.
* testsuite/util/testsuite_containers.h:
Include <bits/boost_concept_check.h>.
(iterator_concept_checks<_It, _Mutable, _Category>): New.
(citerator<_Cont>::forward_members::forward_members()): Instantiate
latter for container iterator and const_iterator.
* testsuite/23_containers/list/68222_neg.cc: New.
* testsuite/23_containers/vector/cons/destructible_debug_neg.cc: Adapt
line number.
* testsuite/23_containers/unordered_set/debug/debug_functions.cc:
(test01): Remove.
* testsuite/23_containers/vector/debug/debug_functions.cc (test01):
Remove.
From-SVN: r263786
Andrew Benson [Wed, 22 Aug 2018 17:47:53 +0000 (17:47 +0000)]
module.c (load_generic_interfaces): Move call to find_symbol() so that only occurs if actually needed.
2018-08-22 Andrew Benson <abensonca@gmail.com>
* module.c (load_generic_interfaces): Move call to find_symbol()
so that only occurs if actually needed.
From-SVN: r263784
Andrew Benson [Wed, 22 Aug 2018 17:40:55 +0000 (17:40 +0000)]
Update write after approval list
2018-08-22 Andrew Benson <abensonca@gmail.com>
* MAINTAINERS: Add self to write after approval list.
From-SVN: r263783
Janus Weil [Wed, 22 Aug 2018 17:10:00 +0000 (19:10 +0200)]
re PR fortran/86888 ([F08] allocatable components of indirectly recursive type)
fix PR 86888
2018-08-22 Janus Weil <janus@gcc.gnu.org>
PR fortran/86888
* decl.c (gfc_match_data_decl): Allow allocatable components of
indirectly recursive type.
* resolve.c (resolve_component): Remove two errors messages ...
(resolve_fl_derived): ... and replace them by a new one.
2018-08-22 Janus Weil <janus@gcc.gnu.org>
PR fortran/86888
* gfortran.dg/alloc_comp_basics_6.f90: Update an error message and add
an additional case.
* gfortran.dg/alloc_comp_basics_7.f90: New test case.
* gfortran.dg/class_17.f03: Update error message.
* gfortran.dg/class_55.f90: Ditto.
* gfortran.dg/dtio_11.f90: Update error messages.
* gfortran.dg/implicit_actual.f90: Add an error message.
* gfortran.dg/typebound_proc_12.f90: Update error message.
From-SVN: r263782
Martin Sebor [Wed, 22 Aug 2018 16:42:56 +0000 (16:42 +0000)]
PR middle-end/87052 - STRING_CST printing incomplete in Gimple dumps
gcc/testsuite/ChangeLog:
PR middle-end/87052
* gcc.dg/pr87052.c: New test.
* gcc.dg/tree-ssa/dump-3.c: Adjust.
gcc/ChangeLog:
PR middle-end/87052
* tree-pretty-print.c (pretty_print_string): Add argument.
(dump_generic_node): Call to pretty_print_string with string size.
From-SVN: r263781
Segher Boessenkool [Wed, 22 Aug 2018 16:04:09 +0000 (18:04 +0200)]
[ fix changelog ]
2018-08-22 Segher Boessenkool <segher@kernel.crashing.org>
PR rtl-optimization/86771
* combine.c (try_combine): Do not allow splitting a resulting PARALLEL
of two SETs into those two SETs, one to be placed at i2, if that SETs
destination is modified between i2 and i3.
From-SVN: r263780
Szabolcs Nagy [Wed, 22 Aug 2018 15:12:05 +0000 (15:12 +0000)]
Change AArch64 specific FMAX/FMIN tests into generic MAX_EXPR/MIN_EXPR tests
gfortran now always uses MAX_EXPR/MIN_EXPR for MAX/MIN intrinsics, so the
AArch64 specific FMAX/FMIN tests are no longer valid.
2018-08-22 Szabolcs Nagy <szabolcs.nagy@arm.com>
* gfortran.dg/max_fmax_aarch64.f90: Rename to...
* gfortran.dg/max_expr.f90: ...this.
* gfortran.dg/min_fmin_aarch64.f90: Rename to...
* gfortran.dg/min_expr.f90: ...this.
From-SVN: r263778
Segher Boessenkool [Wed, 22 Aug 2018 13:41:05 +0000 (15:41 +0200)]
combine: Do another check before splitting a parallel (PR86771)
When combine splits a resulting parallel into its two SETs, it has to
place one at i2, and the other stays at i3. This does not work if the
destination of the SET that will be placed at i2 is modified between
i2 and i3. This patch fixes it.
* combine.c (try_combine): Do not allow splitting a resulting PARALLEL
of two SETs into those two SETs, one to be placed at i2, if that SETs
destination is modified between i2 and i3.
From-SVN: r263776
Richard Sandiford [Wed, 22 Aug 2018 13:02:48 +0000 (13:02 +0000)]
[2/2] Fix bogus inner induction (PR 86725)
This patch is the second part of the fix for PR 86725. The problem
in the original test is that for:
outer1:
x_1 = PHI <x_4(outer2), ...>;
...
inner:
x_2 = PHI <x_1(outer1), x_3(...)>;
...
x_3 = ...;
...
outer2:
x_4 = PHI <x_3(inner)>;
...
there are corner cases in which it is possible to classify the
inner phi as an induction but not the outer phi. The -4.c test
is a more direct example.
After failing to classify x_1 as an induction, we go on to
classify it as a double reduction (which is basically true).
But we still classified the inner phi as an induction rather
than as part of a reduction, leading to an ICE when trying
to vectorise the outer phi.
We analyse the phis for outer loops first, so the simplest
fix is not to classify the phi as an induction if outer loop
analysis said that it should be a reduction.
The -2.c test is from the original PR. The -3.c test is a
version in which "wo" really is used a reduction; this was
already correctly rejected, but for the wrong reason ("inner-loop
induction only used outside of the outer vectorized loop").
The -4.c test is another way of tickling the original problem
without relying on the undefinedness of signed overflow.
The -5.c test shows an (uninteresting) example in which the
patch prevents a spurious failure to vectorise the outer loop.
2018-08-22 Richard Sandiford <richard.sandiford@arm.com>
gcc/
PR tree-optimization/86725
* tree-vect-loop.c (vect_inner_phi_in_double_reduction_p): New
function.
(vect_analyze_scalar_cycles_1): Check it.
gcc/testsuite/
PR tree-optimization/86725
* gcc.dg/vect/no-scevccp-pr86725-2.c: New test.
* gcc.dg/vect/no-scevccp-pr86725-3.c: Likewise.
* gcc.dg/vect/no-scevccp-pr86725-4.c: Likewise.
* gcc.dg/vect/no-scevccp-pr86725-5.c: Likewise.
From-SVN: r263774
Richard Sandiford [Wed, 22 Aug 2018 13:02:39 +0000 (13:02 +0000)]
[1/2] Fix bogus double reduction (PR 86725)
This patch is the first part of the fix for PR 86725. We would
treat x_1 in:
outer1:
x_1 = PHI <x_4(outer2), ...>;
...
inner:
x_2 = ...x_1...;
...
x_3 = ...;
...
outer2:
x_4 = PHI <x_3(inner)>;
...
as a double reduction without checking what kind of statement x_2 is.
In practice it has to be a phi, since for other x_2, x_1 would simply
be a loop invariant that gets used for every inner loop iteration.
The idea with doing this patch first is that, by checking x_2 really
is a phi, we can hand off the validation of the rest of the reduction
to the phi analysis in the inner loop.
The test case is a variant of the one in the PR.
2018-08-22 Richard Sandiford <richard.sandiford@arm.com>
gcc/
PR tree-optimization/86725
* tree-vect-loop.c (vect_is_simple_reduction): When treating
an outer loop phi as a double reduction, make sure that the
single user of the phi result is an inner loop phi.
gcc/testsuite/
PR tree-optimization/86725
* gcc.dg/vect/no-scevccp-pr86725-1.c: New test.
From-SVN: r263773
Richard Sandiford [Wed, 22 Aug 2018 12:59:08 +0000 (12:59 +0000)]
Make the vectoriser drop to strided accesses for stores with gaps
We could vectorise:
for (...)
{
a[0] = ...;
a[1] = ...;
a[2] = ...;
a[3] = ...;
a += stride;
}
(including the case when stride == 8) but not:
for (...)
{
a[0] = ...;
a[1] = ...;
a[2] = ...;
a[3] = ...;
a += 8;
}
(where the stride is always 8). The former was treated as a "grouped
and strided" store, while the latter was treated as a grouped store
with gaps, which we don't support.
This patch makes us treat groups of stores with gaps at the end as
strided groups too. I tried to go through all uses of STMT_VINFO_STRIDED_P
and all vector uses of DR_STEP to see whether there were any hard-baked
assumptions, but couldn't see any. I wondered whether we should relax:
/* We do not have to consider dependences between accesses that belong
to the same group, unless the stride could be smaller than the
group size. */
if (DR_GROUP_FIRST_ELEMENT (stmtinfo_a)
&& (DR_GROUP_FIRST_ELEMENT (stmtinfo_a)
== DR_GROUP_FIRST_ELEMENT (stmtinfo_b))
&& !STMT_VINFO_STRIDED_P (stmtinfo_a))
return false;
for cases in which the step is constant and the absolute step is known
to be greater than the group size, but data dependence analysis should
already return chrec_known for those cases.
The new test is a version of vect-avg-15.c with the variable step
replaced by a constant one.
A natural follow-on would be to do the same for groups with gaps in
the middle:
/* Check that the distance between two accesses is equal to the type
size. Otherwise, we have gaps. */
diff = (TREE_INT_CST_LOW (DR_INIT (data_ref))
- TREE_INT_CST_LOW (prev_init)) / type_size;
if (diff != 1)
{
[...]
if (DR_IS_WRITE (data_ref))
{
if (dump_enabled_p ())
dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
"interleaved store with gaps\n");
return false;
}
But I think we should do that separately and see what the fallout
from this change is first.
2018-08-22 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* tree-vect-data-refs.c (vect_analyze_group_access_1): Convert
grouped stores with gaps to a strided group.
gcc/testsuite/
* gcc.dg/vect/vect-avg-16.c: New test.
* gcc.dg/vect/slp-37.c: Expect the loop to be vectorized.
* gcc.dg/vect/vect-strided-u8-i8-gap4.c,
* gcc.dg/vect/vect-strided-u8-i8-gap4-big-array.c: Likewise for
the second loop in main1.
From-SVN: r263772
Richard Sandiford [Wed, 22 Aug 2018 12:58:53 +0000 (12:58 +0000)]
Only test STMT_VINFO_STRIDED_P for the first statement in a group
get_load_store_type & co were testing STMT_VINFO_STRIDED_P on individual
statements in a group instead of the first. This has no effect on
its own, but is needed by a later patch.
2018-08-22 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* tree-vect-stmts.c (get_group_load_store_type)
(get_load_store_type): Only test STMT_VINFO_STRIDED_P for the
first statement in a group.
From-SVN: r263771
Iain Sandoe [Wed, 22 Aug 2018 12:31:23 +0000 (12:31 +0000)]
Udate Darwin's LTO specs to match what's in gcc/gcc.c
gcc/
* config/darwin.h (LINK_COMMAND_SPEC_A): Sync LTO options with
the sequence used in gcc/gcc.c.
From-SVN: r263770
Iain Sandoe [Wed, 22 Aug 2018 12:22:07 +0000 (12:22 +0000)]
stringify __USER_LABEL_PREFIX__ in pr85248 testcase.
gcc/testsuite
* gcc.dg/lto/pr85248_0.c (test_alias):
Stringify __USER_LABEL_PREFIX__.
(test_noreturn): Likewise.
From-SVN: r263769
Iain Sandoe [Wed, 22 Aug 2018 12:12:46 +0000 (12:12 +0000)]
Make the gcc-ar,nm, strip tools respond correctly to --help and --version
when there's no plugin built.
gcc/
PR other/704
* gcc-ar.c (main): Don’t try to invoke the plug-in if we’re not
building it.
From-SVN: r263768
Jonathan Wakely [Wed, 22 Aug 2018 12:11:17 +0000 (13:11 +0100)]
PR libstdc++/77854 document size_type for containers
PR libstdc++/77854
* doc/xml/manual/status_cxx1998.xml: Document size_type and
difference_type for containers.
* doc/html/*: Regenerate.
From-SVN: r263767
Iain Sandoe [Wed, 22 Aug 2018 12:05:45 +0000 (12:05 +0000)]
commit missed change from r263764
From-SVN: r263766
Iain Sandoe [Wed, 22 Aug 2018 11:58:43 +0000 (11:58 +0000)]
Move Darwin10 unwinder fix to a crt shim.
gcc/
* config/darwin10.h (LINK_GCC_C_SEQUENCE_SPEC): Adjust to use the
Darwin10-specific unwinder-shim.
* config/darwin12.h (LINK_GCC_C_SEQUENCE_SPEC): Remove.
* config/rs6000/darwin.h (DARWIN_CRT1_SPEC, DARWIN_DYLIB1_SPEC):
New to cater for Darwin10 Rosetta.
libgcc/
* config/unwind-dw2-fde-darwin.c
(_darwin10_Unwind_FindEnclosingFunction): move from here ...
* config/darwin10-unwind-find-enc-func.c: … to here.
* config/t-darwin: Build Darwin10 unwinder shim crt.
* libgcc/config.host: Add the Darwin10 unwinder shim.
From-SVN: r263765
Iain Sandoe [Wed, 22 Aug 2018 11:42:53 +0000 (11:42 +0000)]
Allw for Darwin to operate with an assembler supporting gotoff in data.
gcc/
* config/i386/i386.c (ix86_output_addr_diff_elt): Move the MACH-O
specific test before the one for HAVE_AS_GOTOFF_IN_DATA.
From-SVN: r263764
Iain Sandoe [Wed, 22 Aug 2018 11:37:02 +0000 (11:37 +0000)]
Fix FDE labels for Darwin
gcc/
PR bootstrap/81033
PR target/81733
PR target/52795
* gcc/dwarf2out.c (FUNC_SECOND_SECT_LABEL): New.
(dwarf2out_switch_text_section): Generate a local label for the second
function sub-section and apply it as the second FDE start label.
* gcc/final.c (final_scan_insn_1): Emit second FDE label after the second
sub-section start.
From-SVN: r263763
Richard Biener [Wed, 22 Aug 2018 11:01:58 +0000 (11:01 +0000)]
re PR tree-optimization/86988 (ICE: tree check: expected integer_cst, have var_decl in get_len, at tree.h:5563)
2018-08-22 Richard Biener <rguenther@suse.de>
PR tree-optimization/86988
* tree-vrp.c (vrp_prop::check_mem_ref): Bail out on VLAs.
* g++.dg/pr86988.C: New testcase.
From-SVN: r263762
Richard Biener [Wed, 22 Aug 2018 08:07:36 +0000 (08:07 +0000)]
re PR tree-optimization/86945 (BUG with optimisation of select case statement in gfortran)
2018-08-22 Richard Biener <rguenther@suse.de>
PR tree-optimization/86945
* tree-cfg.c (generate_range_test): Use unsigned arithmetic.
* gcc.dg/torture/pr86945.c: New testcase.
From-SVN: r263761
Alexandre Oliva [Wed, 22 Aug 2018 03:48:00 +0000 (03:48 +0000)]
rs6000 small data: add note on sdata2/r2
for gcc/ChangeLog
* config/rs6000/rs6000.c (SMALL_DATA_RELOC, SMALL_DATA_REG): Add
a comment about how uses of r2 for .sdata2 come about.
From-SVN: r263760
Alexandre Oliva [Wed, 22 Aug 2018 03:47:48 +0000 (03:47 +0000)]
fix reassoc cut&pasto
for gcc/ChangeLog
* tree-ssa-reassoc.c (is_reassociable_op): Fix cut&pasto.
From-SVN: r263759
GCC Administrator [Wed, 22 Aug 2018 00:16:29 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r263758
François Dumont [Tue, 21 Aug 2018 19:44:41 +0000 (19:44 +0000)]
P0646R1 Improving the Return Value of Erase-Like Algorithms I
2018-08-21 François Dumont <fdumont@gcc.gnu.org>
P0646R1 Improving the Return Value of Erase-Like Algorithms I
* include/debug/forward_list (forward_list::__remove_return_type):
Define typedef as size_type or void, according to __cplusplus value.
(_GLIBCXX_FWDLIST_REMOVE_RETURN_TYPE_TAG): Define macro as abi-tag or
empty, according to __cplusplus value.
(_GLIBCXX20_ONLY): Define macro.
(forward_list::remove, forward_list::unique): Use typedef and macro
to change return type and add abi-tag for C++2a. Return number of
removed elements for C++2a.
(forward_list::remove_if<Pred>, forward_list::unique<BinPred>): Use
typedef to change return type for C++2a. Return number of removed
elements for C++2a.
* include/debug/list (list::__remove_return_type): Define typedef as
size_type or void, according to __cplusplus value.
(_GLIBCXX_LIST_REMOVE_RETURN_TYPE_TAG): Define macro as abi-tag or
empty, according to __cplusplus value.
(_GLIBCXX20_ONLY): Define macro.
(list::remove, list::unique): Use typedef and macro to change return
type and add abi-tag for C++2a. Return number of removed elements for
C++2a.
(list::remove_if<Predicate>, list::unique<BinaryPredicate>): Use typedef
to change return type for C++2a. Return number of removed elements for
C++2a.
From-SVN: r263752
Janne Blomqvist [Tue, 21 Aug 2018 19:36:13 +0000 (22:36 +0300)]
Unconditionally use MAX_EXPR/MIN_EXPR for MAX/MIN intrinsics
For floating point types, the question is what MAX(a, NaN) or MIN(a,
NaN) should return (where "a" is a normal number). There are valid
usecases for returning either one, but the Fortran standard doesn't
specify which one should be chosen. Also, there is no consensus among
other tested compilers. In short, it's a mess. So lets just do
whatever is fastest, which is using MAX_EXPR/MIN_EXPR which are not
defined to do anything in particular if one of the operands is a NaN.
gcc/fortran/ChangeLog:
2018-08-21 Janne Blomqvist <jb@gcc.gnu.org>
* trans-intrinsic.c (gfc_conv_intrinsic_minmax): Use
MAX_EXPR/MIN_EXPR unconditionally for real arguments.
* gfortran.texi (Compiler Characteristics): Document MAX/MIN
behavior wrt NaN.
gcc/testsuite/ChangeLog:
2018-08-21 Janne Blomqvist <jb@gcc.gnu.org>
* gfortran.dg/nan_1.f90: Remove tests that test MAX/MIN with NaNs.
From-SVN: r263751
Nicolas Koenig [Tue, 21 Aug 2018 18:48:59 +0000 (18:48 +0000)]
re PR fortran/25829 ([F03] Asynchronous IO support)
2018-08-21 Nicolas Koenig <koenigni@gcc.gnu.org>
Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/25829
* gfortran.texi: Add description of asynchronous I/O.
* trans-decl.c (gfc_finish_var_decl): Treat asynchronous variables
as volatile.
* trans-io.c (gfc_build_io_library_fndecls): Rename st_wait to
st_wait_async and change argument spec from ".X" to ".w".
(gfc_trans_wait): Pass ID argument via reference.
2018-08-21 Nicolas Koenig <koenigni@gcc.gnu.org>
Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/25829
* gfortran.dg/f2003_inquire_1.f03: Add write statement.
* gfortran.dg/f2003_io_1.f03: Add wait statement.
2018-08-21 Nicolas Koenig <koenigni@gcc.gnu.org>
Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/25829
* Makefile.am: Add async.c to gfor_io_src.
Add async.h to gfor_io_headers.
* Makefile.in: Regenerated.
* gfortran.map: Add _gfortran_st_wait_async.
* io/async.c: New file.
* io/async.h: New file.
* io/close.c: Include async.h.
(st_close): Call async_wait for an asynchronous unit.
* io/file_pos.c (st_backspace): Likewise.
(st_endfile): Likewise.
(st_rewind): Likewise.
(st_flush): Likewise.
* io/inquire.c: Add handling for asynchronous PENDING
and ID arguments.
* io/io.h (st_parameter_dt): Add async bit.
(st_parameter_wait): Correct.
(gfc_unit): Add au pointer.
(st_wait_async): Add prototype.
(transfer_array_inner): Likewise.
(st_write_done_worker): Likewise.
* io/open.c: Include async.h.
(new_unit): Initialize asynchronous unit.
* io/transfer.c (async_opt): New struct.
(wrap_scalar_transfer): New function.
(transfer_integer): Call wrap_scalar_transfer to do the work.
(transfer_real): Likewise.
(transfer_real_write): Likewise.
(transfer_character): Likewise.
(transfer_character_wide): Likewise.
(transfer_complex): Likewise.
(transfer_array_inner): New function.
(transfer_array): Call transfer_array_inner.
(transfer_derived): Call wrap_scalar_transfer.
(data_transfer_init): Check for asynchronous I/O.
Perform a wait operation on any pending asynchronous I/O
if the data transfer is synchronous. Copy PDT and enqueue
thread for data transfer.
(st_read_done_worker): New function.
(st_read_done): Enqueue transfer or call st_read_done_worker.
(st_write_done_worker): New function.
(st_write_done): Enqueue transfer or call st_read_done_worker.
(st_wait): Document as no-op for compatibility reasons.
(st_wait_async): New function.
* io/unit.c (insert_unit): Use macros LOCK, UNLOCK and TRYLOCK;
add NOTE where necessary.
(get_gfc_unit): Likewise.
(init_units): Likewise.
(close_unit_1): Likewise. Call async_close if asynchronous.
(close_unit): Use macros LOCK and UNLOCK.
(finish_last_advance_record): Likewise.
(newunit_alloc): Likewise.
* io/unix.c (find_file): Likewise.
(flush_all_units_1): Likewise.
(flush_all_units): Likewise.
* libgfortran.h (generate_error_common): Add prototype.
* runtime/error.c: Include io.h and async.h.
(generate_error_common): New function.
2018-08-21 Nicolas Koenig <koenigni@gcc.gnu.org>
Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/25829
* testsuite/libgomp.fortran/async_io_1.f90: New test.
* testsuite/libgomp.fortran/async_io_2.f90: New test.
* testsuite/libgomp.fortran/async_io_3.f90: New test.
* testsuite/libgomp.fortran/async_io_4.f90: New test.
* testsuite/libgomp.fortran/async_io_5.f90: New test.
* testsuite/libgomp.fortran/async_io_6.f90: New test.
* testsuite/libgomp.fortran/async_io_7.f90: New test.
Co-Authored-By: Thomas Koenig <tkoenig@gcc.gnu.org>
From-SVN: r263750
Marek Polacek [Tue, 21 Aug 2018 18:37:23 +0000 (18:37 +0000)]
re PR c++/86499 (lambda-expressions with capture-default are allowed at namespace scope)
PR c++/86499
* parser.c (cp_parser_lambda_introducer): Give error if a non-local
lambda has a capture-default.
* g++.dg/cpp0x/lambda/lambda-non-local.C: New test.
* g++.dg/cpp0x/lambda/lambda-this10.C: Adjust dg-error.
From-SVN: r263749
Paolo Carlini [Tue, 21 Aug 2018 15:49:58 +0000 (15:49 +0000)]
decl.c (check_static_variable_definition): Change to return void.
2018-08-21 Paolo Carlini <paolo.carlini@oracle.com>
* decl.c (check_static_variable_definition): Change to return void.
From-SVN: r263742
Marek Polacek [Tue, 21 Aug 2018 15:38:36 +0000 (15:38 +0000)]
PR c++/86981, Implement -Wpessimizing-move.
* c.opt (Wpessimizing-move): New option.
* typeck.c (decl_in_std_namespace_p): New.
(is_std_move_p): New.
(maybe_warn_pessimizing_move): New.
(can_do_nrvo_p): New, factored out of ...
(check_return_expr): ... here. Warn about potentially harmful
std::move in a return statement.
* doc/invoke.texi: Document -Wpessimizing-move.
* g++.dg/cpp0x/Wpessimizing-move1.C: New test.
* g++.dg/cpp0x/Wpessimizing-move2.C: New test.
* g++.dg/cpp0x/Wpessimizing-move3.C: New test.
* g++.dg/cpp0x/Wpessimizing-move4.C: New test.
* g++.dg/cpp1z/Wpessimizing-move1.C: New test.
From-SVN: r263741
Marek Polacek [Tue, 21 Aug 2018 15:25:17 +0000 (15:25 +0000)]
re PR c++/65043 (Expected narrowing conversion during list initialization of bool from double)
PR c++/65043
* call.c (standard_conversion): Set check_narrowing.
* typeck2.c (check_narrowing): Use CP_INTEGRAL_TYPE_P rather
than comparing with INTEGER_TYPE.
* g++.dg/concepts/pr67595.C: Add dg-warning.
* g++.dg/cpp0x/Wnarrowing11.C: New test.
* g++.dg/cpp0x/Wnarrowing12.C: New test.
* g++.dg/cpp0x/rv-cast5.C: Add static_cast.
From-SVN: r263739
Marek Polacek [Tue, 21 Aug 2018 15:10:51 +0000 (15:10 +0000)]
* cp-tree.h: Fix typo.
From-SVN: r263738
Hristian Kirtchev [Tue, 21 Aug 2018 14:50:03 +0000 (14:50 +0000)]
[Ada] Minor reformattings
2018-08-21 Hristian Kirtchev <kirtchev@adacore.com>
gcc/ada/
* checks.adb, contracts.adb, exp_aggr.adb, exp_attr.adb,
exp_ch6.adb, exp_ch7.adb, exp_ch9.adb, exp_unst.adb,
exp_util.adb, freeze.adb, gnatlink.adb, layout.adb,
lib-writ.adb, lib-xref-spark_specific.adb, sem_ch13.adb,
sem_ch3.adb, sem_ch6.adb, sem_res.adb, sem_util.adb, sinfo.ads,
sprint.adb: Minor reformatting.
From-SVN: r263737
Jerome Lambourg [Tue, 21 Aug 2018 14:49:49 +0000 (14:49 +0000)]
[Ada] Add a new gnat tool vxlink
VxLink is a helper tool used as a wrapper around g++/gcc to build
VxWorks DKM (Downloadable Kernel Modules).
Such DKM is a partially linked object that includes entry points for
constructors and destructors.
This tool thus uses g++ to generate an intermediate partially linked
object, retrieves the list of constructors and destructors in it and
produces a C file that lists those ctors/dtors in a way that is
understood be VxWorks kernel. It then links this file with the
intermediate object to produce a valid DKM.
2018-08-21 Jerome Lambourg <lambourg@adacore.com>
gcc/ada/
* vxlink-bind.adb, vxlink-bind.ads, vxlink-link.adb,
vxlink-link.ads, vxlink-main.adb, vxlink.adb, vxlink.ads: Add a
new tool vxlink to handle VxWorks constructors in DKMs.
* gcc-interface/Makefile.in: add rules to build vxlink
From-SVN: r263736
Ed Schonberg [Tue, 21 Aug 2018 14:49:42 +0000 (14:49 +0000)]
[Ada] Crash on expression function and tagged types
This patch fixes a compiler abort on an expression function whose
expression includes tagged types that have not been frozen before the
generated body of the function is analyzed, even though that body is
inserted at the end of the current declarative part.
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_ch6.adb (Analyze_Subprogram_Body_Helper, Mask_Type):
Refine the handling of freezing types for expression functions
that are not completions, when analyzing the generated body for
the function: the body is inserted at the end of the enclosing
declarative part, and its analysis may freeze types declared in
the same scope that have not been frozen yet.
gcc/testsuite/
* gnat.dg/expr_func7.adb, gnat.dg/expr_func7.ads: New testcase.
From-SVN: r263735
Ed Schonberg [Tue, 21 Aug 2018 14:49:34 +0000 (14:49 +0000)]
[Ada] Spurious crash on expression function as completion with contracts
This patch fixes a compiler abort on an expression function that is a
completion of a subprogram with preconditions. The problem is caused by
the presence of types in the precondition that are not frozen when the
subprogram body constructed for the expression function receives the
code that enforces the precondition. These types must be frozen before
the contract is expanded, so the freeze nodes for these types appear in
the proper scope. This is analogous to what is done with type references
that appear in the original expression of the expression function.
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_ch6.adb: Remove Freeze_Expr_Types.
* freeze.ads, freeze.adb (Freeze_Expr_Types): Moved from
sem_ch6.adb, and extended to handle other expressions that may
contain unfrozen types that must be frozen in their proper
scopes.
* contracts.adb (Analyze_Entry_Or_Subprogram_Contract): If the
contract is for the generated body of an expression function
that is a completion, traverse the expressions for pre- and
postconditions to freeze all types before adding the contract
code within the subprogram body.
gcc/testsuite/
* gnat.dg/expr_func6.adb, gnat.dg/expr_func6.ads: New testcase.
From-SVN: r263734
Hristian Kirtchev [Tue, 21 Aug 2018 14:49:26 +0000 (14:49 +0000)]
[Ada] Retention of with clauses for ignored Ghost units
This patch ensures that with clauses that mention ignored Ghost units are
retained in the tree. The retention is necessary for several reasons:
* The with clauses allow the new elaboration order mechanism to
produce the same library edges regardless of whether the Ghost unit
is checked or ignored. This ensures that the elaboration order
remains consistent.
* The with clauses allow the unnesting mechanism to properly
recognize that all units have been examined for unnesing purposes.
No observable impact, no test needed.
2018-08-21 Hristian Kirtchev <kirtchev@adacore.com>
gcc/ada/
* sem_ch10.adb: Remove the with and use clause for unit Ghost.
(Analyze_With_Clause): Do not mark a with clause which mentions
an ignored Ghost code for elimination.
From-SVN: r263733
Javier Miranda [Tue, 21 Aug 2018 14:48:35 +0000 (14:48 +0000)]
[Ada] Crash processing SPARK annotate aspect
The compiler blows up writing the ALI file of a package that has a ghost
subprogram with an annotate contract.
2018-08-21 Javier Miranda <miranda@adacore.com>
gcc/ada/
* lib-writ.adb (Write_Unit_Information): Handle pragmas removed
by the expander.
gcc/testsuite/
* gnat.dg/spark2.adb, gnat.dg/spark2.ads: New testcase.
From-SVN: r263732
Ed Schonberg [Tue, 21 Aug 2018 14:48:03 +0000 (14:48 +0000)]
[Ada] Spurious error on overriding protected function in instance
The conformance between an overriding protected operation with
progenitors and the overridden interface operation requires subtype
conformance; requiring equality of return types in the case of a
function is too restrictive and leads to spurious errors when the return
type is a generic actual.
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_ch6.adb (Check_Synchronized_Overriding): The conformance
between an overriding protected operation and the overridden
abstract progenitor operation requires subtype conformance;
requiring equality of return types in the case of a function is
too restrictive and leads to spurious errors when the return
type is a generic actual.
gcc/testsuite/
* gnat.dg/prot6.adb, gnat.dg/prot6.ads: New testcase.
From-SVN: r263731
Ed Schonberg [Tue, 21 Aug 2018 14:47:56 +0000 (14:47 +0000)]
[Ada] Properly set scope of artificial entities in blocks
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* exp_ch9.adb (Expand_N_Timed_Entry_Call,
Expand_Conditional_Entry_Call): Use Reset_Scopes_Of to set
properly the scope of all entities created in blocks generated
by the expansion of these constructs.
From-SVN: r263730
Ed Schonberg [Tue, 21 Aug 2018 14:47:50 +0000 (14:47 +0000)]
[Ada] Fix scope computation for loop statements
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_ch13.adb (Build_Predicate_Functioss): Apply
Reset_Quantified_Variables_Scope after predicate function has
been analyzed, so that the scope can be reset on the generated
loop statements that have replaced the quantified expressions.
From-SVN: r263729
Bob Duff [Tue, 21 Aug 2018 14:47:44 +0000 (14:47 +0000)]
[Ada] Einfo: remove obsolete code
2018-08-21 Bob Duff <duff@adacore.com>
gcc/ada
* einfo.ads, einfo.adb (Private_View, Shadow_Entities): Remove
obsolete code.
From-SVN: r263728
Maroua Maalej [Tue, 21 Aug 2018 14:47:38 +0000 (14:47 +0000)]
[Ada] Update for Ownership rules for access types according to AI12-0240
The implementation of these Ownership rules for safe pointers and
automatic memory management is still a prototype at an experimental
stage.
To activate the checks, the code should be compiled with the debug flag
-gnatdF and the flag -gnatd.F for setting the context for formal
verification of SPARK.
These changes do not affect compilation.
2018-08-21 Maroua Maalej <maalej@adacore.com>
gcc/ada/
* sem_spark.adb (Check_Call_Statement): Check global and formal
parameter permissions at call sites.
(Check_Callable_Body): Assume permissions on globals and
parameters depending on their modes then analyse the body
operations.
(Check_Declaration): Consider both deep (including elementary
access) object declarations and normal variables. First check
whether the deep object is of Ownership Aspec True or not, then,
depending on its initialization, assign the appropriate state.
Check related to non access type variables deal with
initialization value permissions.
(Check_Expression): Check nodes used in the expression being
analyzed.
(Check_Globals): Call by Check_Call_Statement to perform the
check on globals.
(Check_List): Call Check_Node on each element of the list.
(Check_Loop_Statement): Check the Iteration_Scheme and loop
statements.
(Check_Node): Main traversal procedure to check safe pointer usage.
(Check_Package_Body): Check subprogram's body.
(Check_Param_In): Take a formal and an actual parameter and
Check the permission of every in-mode parameter.
(Check_Param_Out): Take a formal and an actual parameter and
check the state of out-mode and in out-mode parameters.
(Check_Statement): Check statements other than procedure call.
(Get_Perm, Get_Perm_Or_Tree, Get_Perm_Tree): Find out the state
related to the given name.
(Is_Deep): Return True if an object is of access type or has
subfields of access type.
(Perm_Error, Perm_Error_Subprogram_End): Add an error message
whenever the found state on the given name is different from the
one expected (in the statement being analyzed).
(Process_Path): Given an operation and a current state, call
Perm_Error if there is any mismatch.
(Return_Declarations, Return_Globals, Return_The_Global): Check
the state of a given name at the end of the subprogram. These
procedures may change depending on how we shall finally deal
with globals and the rhs state in a move operation.
(Set_Perm_Extensions, Set_Perm_Prefixes_Borrow,
Set_Perm_Prefixes, Setup_Globals, Setup_Parameter_Or_Global,
Setup_Parameters): Set up the new states to the given node and
up and down the tree after an operation.
(Has_Ownership_Aspect_True): This function may disappear later
when the Ownership Aspect will be implemented in the FE.
From-SVN: r263727
Ed Schonberg [Tue, 21 Aug 2018 14:47:32 +0000 (14:47 +0000)]
[Ada] Spurious ambiguity error on call returning an access type
If F is a function with a single defaulted parameter that returns an
access_to_array type, then F (I) may designate either the return type or
an indexing of the result of the call, after implicit dereferencing. If
the component type C of the array type AR is accces AR this is ambiguous
in a context whose expected type is C. If F is parameterless the call is
not ambiguous.
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_res.adb (Resolve_Call): Resolve correctly a parameterless
call that returns an access type whose designated type is the
component type of an array, when the function has no defaulted
parameters.
gcc/testsuite/
* gnat.dg/access5.adb, gnat.dg/access5.ads: New testcase.
From-SVN: r263726
Yannick Moy [Tue, 21 Aug 2018 14:47:25 +0000 (14:47 +0000)]
[Ada] Document entries of the target parametrization file
2018-08-21 Yannick Moy <moy@adacore.com>
gcc/ada/
* doc/gnat_ugn/building_executable_programs_with_gnat.rst:
Document entries of the target parametrization file.
* gnat_ugn.texi: Regenerate.
From-SVN: r263725
Ed Schonberg [Tue, 21 Aug 2018 14:47:18 +0000 (14:47 +0000)]
[Ada] Set scope of elaboration flag for 'Access
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_attr.adb: Set scope of elaboration flag for 'Access.
From-SVN: r263724
Ed Schonberg [Tue, 21 Aug 2018 14:47:12 +0000 (14:47 +0000)]
[Ada] Sprint: add guard on printing aspects
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sprint.adb: Add guard on printing aspects.
From-SVN: r263723
Javier Miranda [Tue, 21 Aug 2018 14:47:05 +0000 (14:47 +0000)]
[Ada] Crash compiling SPARK ghost functions with callgraph info
The compiler blows up generating the callgraph output of SPARK ghost
subprograms whose contracts invoke ghost functions.
2018-08-21 Javier Miranda <miranda@adacore.com>
gcc/ada/
* exp_cg.adb (Generate_CG_Output): Handle calls removed by the
expander.
From-SVN: r263722
Ed Schonberg [Tue, 21 Aug 2018 14:46:59 +0000 (14:46 +0000)]
[Ada] Unnesting: do not set size of access subprograms
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* layout.adb: Do not set size of access subprogram if unnesting.
From-SVN: r263721
Ed Schonberg [Tue, 21 Aug 2018 14:46:54 +0000 (14:46 +0000)]
[Ada] Unnesting: do not emit warnings for access to subprograms
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* freeze.adb: Remove warnings for access to subprograms when
unnesting is active.
From-SVN: r263720
Ed Schonberg [Tue, 21 Aug 2018 14:46:49 +0000 (14:46 +0000)]
[Ada] Improper copying of limited arrays with default initialization
This patch fixes an improper expansion of aggregates for limited array
types in an object declaration. Prior to this patch, The presence of the
aggregate (which can only consist of box initializations) would create a
temporary that was then assigned to the object in the declaration.
Apart from a violation of the semantics of build-in-place limited
objects, this can also lead to out-of-scope access in LLVM.
Executing the following;
gcc -c -gnatDG nocopy.adb
grep quintet nocopy.adb.dg | wc -l
must yield:
5
----
procedure NoCopy is
-- Task used in this example to test that the limited component
-- is properly initialized.
task type T_Task (Disc : Natural);
task body T_Task is
begin
null;
end T_Task;
type My_Rec (D : Natural := 9999) is record
-- Components initialized by means of the current value
-- of the record discriminant
T : T_Task (D);
end record;
type TR is array (1 .. 5) of My_Rec;
Quintet : TR := (others => (others => <>));
begin
null;
end NoCopy;
2018-08-21 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* exp_aggr.adb (Expand_Array_Aggregate): If the component type
is limited, the array must be constructed in place, so set flag
In_Place_Assign_OK_For_Declaration accordingly. This prevents
improper copying of an array of tasks during initialization.
From-SVN: r263719