2018-03-11 Paul Thomas <pault@gcc.gnu.org>
PR fortran/84546
* trans-array.c (structure_alloc_comps): Make sure that the
vptr is copied and that the unlimited polymorphic _len is used
to compute the size to be allocated.
* trans-expr.c (gfc_get_class_array_ref): If unlimited, use the
unlimited polymorphic _len for the offset to the element.
(gfc_copy_class_to_class): Set the new 'unlimited' argument.
* trans.h : Add the boolean 'unlimited' to the prototype.
2018-03-11 Paul Thomas <pault@gcc.gnu.org>
PR fortran/84546
* gfortran.dg/unlimited_polymorphic_29.f90 : New test.
From-SVN: r258438
2018-03-11 Steven G. Kargl <kargls@gcc.gnu.org>
* check.c (gfc_check_kill): Check pid and sig are scalar.
(gfc_check_kill_sub): Restrict kind to 4 and 8.
* intrinsic.c (add_function): Sort keyword list. Add pid and sig
keywords for KILL. Remove redundant *back="back" in favor of the
original *bck="back".
(add_subroutines): Sort keyword list. Add pid and sig keywords
for KILL.
* intrinsic.texi: Fix documentation to consistently use pid and sig.
* iresolve.c (gfc_resolve_kill): Kind can only be 4 or 8. Choose the
correct function.
(gfc_resolve_rename_sub): Add comment.
From-SVN: r258436
PR debug/58150
* dwarf2out.c (gen_enumeration_type_die): Don't guard adding
DW_AT_declaration for ENUM_IS_OPAQUE on -gdwarf-4 or -gno-strict-dwarf,
but on TYPE_SIZE. Don't do anything for ENUM_IS_OPAQUE if not creating
a new die. Don't set TREE_ASM_WRITTEN if ENUM_IS_OPAQUE. Guard
addition of most attributes on !orig_type_die or the attribute not
being present already. Assert TYPE_VALUES is NULL for ENUM_IS_OPAQUE.
* g++.dg/debug/dwarf2/enum2.C: New test.
From-SVN: r258434
2018-03-09 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/84734
* arith.c (check_result, eval_intrinsic): If result overflows, pass
the expression up the chain instead of a NULL pointer.
2018-03-09 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/84734
* gfortran.dg/pr84734.f90: New test.
From-SVN: r258416
2018-03-10 Vladimir Makarov <vmakarov@redhat.com>
Reverting patch:
2018-03-09 Vladimir Makarov <vmakarov@redhat.com>
PR target/83712
* lra-assigns.c (assign_by_spills): Return a flag of reload
assignment failure. Do not process the reload assignment
failures. Do not spill other reload pseudos if they has the same
reg class.
(lra_assign): Add a return arg. Set up from the result of
assign_by_spills call.
(find_reload_regno_insns, lra_split_hard_reg_for): New functions.
* lra-constraints.c (split_reg): Add a new arg. Use it instead of
usage_insns if it is not NULL.
(spill_hard_reg_in_range): New function.
(split_if_necessary, inherit_in_ebb): Pass a new arg to split_reg.
* lra-int.h (spill_hard_reg_in_range, lra_split_hard_reg_for): New
function prototypes.
(lra_assign): Change prototype.
* lra.c (lra): Add code to deal with fails by splitting hard reg
live ranges.
From-SVN: r258415
When outputting entry views in symbolic mode, we used to use a lbl_id,
but that outputs the view as an addr, perhaps even in an indirect one,
which is all excessive and undesirable for a small assembler-computed
constant.
Introduce a new value class for symbolic views, so that we can output
the labels as constant data, using as narrow forms as possible, but
wide enough for any symbolic views output in the compilation. We
don't know exactly where the assembler will reset views, but we count
the symbolic views since known reset points and use that as an upper
bound for view numbers.
Ideally, we'd use uleb128, but then the compiler would have to defer
.debug_info offset computation to the assembler. I'm not going there
for now, so a symbolic uleb128 assembler constant in an attribute is
not something GCC can deal with ATM.
for gcc/ChangeLog
PR debug/84620
* dwarf2out.h (dw_val_class): Add dw_val_class_symview.
(dw_val_node): Add val_symbolic_view.
* dwarf2out.c (dw_line_info_table): Add symviews_since_reset.
(symview_upper_bound): New.
(new_line_info_table): Initialize symviews_since_reset.
(dwarf2out_source_line): Count symviews_since_reset and set
symview_upper_bound.
(dw_val_equal_p): Handle symview.
(add_AT_symview): New.
(print_dw_val): Handle symview.
(attr_checksum, attr_checksum_ordered): Likewise.
(same_dw_val_p, size_of_die): Likewise.
(value_format, output_die): Likewise.
(add_high_low_attributes): Use add_AT_symview for entry_view.
(dwarf2out_finish): Reset symview_upper_bound, clear
zero_view_p.
From-SVN: r258411
gcc/
PR target/83969
* config/rs6000/rs6000.c (rs6000_offsettable_memref_p): New prototype.
Add strict argument and use it.
(rs6000_split_multireg_move): Update for new strict argument.
(mem_operand_gpr): Disallow all non-offsettable addresses.
* config/rs6000/rs6000.md (*movdi_internal64): Use YZ constraint.
gcc/testsuite/
PR target/83969
* gcc.target/powerpc/pr83969.c: New test.
From-SVN: r258400
PR c++/84076
* call.c (convert_arg_to_ellipsis): Instead of cp_build_addr_expr
build ADDR_EXPR with REFERENCE_TYPE.
(build_over_call): For purposes of check_function_arguments, if
argarray[j] is ADDR_EXPR with REFERENCE_TYPE created above, use
its operand rather than the argument itself.
* g++.dg/warn/Wformat-2.C: New test.
From-SVN: r258397
PR c++/84767
* tree-inline.c (copy_tree_body_r): For INDIRECT_REF of a remapped
decl, use remap_type if we want to use the type.
* g++.dg/ext/vla18.C: New test.
From-SVN: r258395
gcc/ChangeLog:
PR tree-optimization/84526
* gimple-ssa-warn-restrict.c (builtin_memref::set_base_and_offset):
Remove dead code.
(builtin_access::generic_overlap): Be prepared to handle non-array
base objects.
gcc/testsuite/ChangeLog:
PR tree-optimization/84526
* gcc.dg/Wrestrict-10.c: New test.
* gcc.dg/Wrestrict-11.c: New test.
From-SVN: r258394
Makefile: add internal/trace to noinst_DATA
The internal/trace package is only imported by tests (specifically the
tests in runtime/trace) so it must be in noinst_DATA to ensure that it
is built before running the tests.
This was mostly working because internal/trace has tests itself, and
is listed in check-packages.txt before runtime/trace, so typical
invocations of make would build internal/trace for checking purposes
before checking runtime/trace. But we need this change to make that
reliable.
Reviewed-on: https://go-review.googlesource.com/99836
From-SVN: r258392
PR c++/84724
* decl.c (duplicate_decls): Don't override __* prefixed builtins
except for __[^b]*_chk, instead issue permerror and for -fpermissive
also a note and return olddecl.
* g++.dg/ext/pr84724.C: New test.
From-SVN: r258391
2018-03-09 Vladimir Makarov <vmakarov@redhat.com>
PR target/83712
* lra-assigns.c (assign_by_spills): Return a flag of reload
assignment failure. Do not process the reload assignment
failures. Do not spill other reload pseudos if they has the same
reg class.
(lra_assign): Add a return arg. Set up from the result of
assign_by_spills call.
(find_reload_regno_insns, lra_split_hard_reg_for): New functions.
* lra-constraints.c (split_reg): Add a new arg. Use it instead of
usage_insns if it is not NULL.
(spill_hard_reg_in_range): New function.
(split_if_necessary, inherit_in_ebb): Pass a new arg to split_reg.
* lra-int.h (spill_hard_reg_in_range, lra_split_hard_reg_for): New
function prototypes.
(lra_assign): Change prototype.
* lra.c (lra): Add code to deal with fails by splitting hard reg
live ranges.
2018-03-09 Vladimir Makarov <vmakarov@redhat.com>
PR target/83712
* gcc.target/arm/pr83712.c: New.
From-SVN: r258390
Currently when handling an invalid -march or -mcpu option on a toolchain without an explicit --with-mode configuration
and compiling without an explicit -mthumb or -marm the arm specs end up calling arm_target_thumb_only to determine
the "thumbness" of the target, which involves parsing the architecture or cpu name. But the functions doing that
parsing also emit error messages and hints on invalid arguments. Later when we parse the architecture or cpu string to
as part of the canonicalisation process (arm_canon_arch_option) we end up emitting the errors again.
The solution in this patch is to silence the errors during the arm_target_thumb_only processing so that they are not emitted
twice. arm_canon_arch_option is guaranteed to run as well, so it can emit the errors and hints that it needs.
Bootstrapped and tested on arm-none-linux-gnueabihf.
Checked that we emit the arch/cpu hints for invalid -march/-mcpu options only once when no "thumbness" options were specified
during configuration or invocation.
PR target/83193
* common/config/arm/arm-common.c (arm_parse_arch_option_name):
Accept complain bool parameter. Only emit errors if it is true.
(arm_parse_cpu_option_name): Likewise.
(arm_target_thumb_only): Adjust callers of the above.
* config/arm/arm-protos.h (arm_parse_cpu_option_name): Adjust
prototype to take a default true bool parameter.
(arm_parse_arch_option_name): Likewise.
From-SVN: r258389
The compiler builtin will use the hardware instruction cdsg if the
memory operand is properly aligned and will fall back to the
library call otherwise.
In case the compiler for one part is able to detect that the
location is aligned and fails to do so for another usage of the hw
instruction and the sw fall back would be mixed on the same memory
location. To avoid this the library fall back also has to use the
hardware instruction if possible.
libatomic/ChangeLog:
2018-03-09 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* config/s390/exch_n.c: New file.
* configure.tgt: Add the config directory for s390.
From-SVN: r258384
The create_dir helper was calling the throwing form of
filesystem::is_directory instead of passing the error_code argument.
Since std::filesystem::create_directory(const path&, error_code&) is
noexcept, it would call std::terminate if an error occurred in
is_directory.
Passing the error_code also takes care of clearing it in the case where
is_directory returns true.
src/filesystem/ops.cc (create_dir): Pass error_code to is_directory.
src/filesystem/std-ops.cc (create_dir): Likewise.
From-SVN: r258375
PR c++/80598
* call.c (build_over_call): In templates set TREE_USED (first_fn) when
not calling mark_used for the benefit of -Wunused-function warning.
* g++.dg/warn/Wunused-function4.C: New test.
From-SVN: r258370
PR inline-asm/84742
* recog.c (asm_operand_ok): Return 0 if multi-character constraint
has ',' character inside of it.
* gcc.target/i386/pr84742-1.c: New test.
* gcc.target/i386/pr84742-2.c: New test.
From-SVN: r258369
In this wrong-code PR the combine pass ends up moving a CC-using instruction past a *compare_cstore<mode>_insn
insn_and_split. After reload the *compare_cstore<mode>_insn splitter ends up generating a SUBS instruction that
clobbers the condition flags, and things go bad.
The solution is simple, the *compare_cstore<mode>_insn pattern should specify that it clobbers the CC register
so that combine (or any other pass) does not assume that it can move CC-using patterns across it.
This patch does that and fixes the testcase.
The testcase FAILs on GCC 8 only, but the buggy pattern is in GCC 6 onwards, so we should backport this as
a latent bug fix after it's had some time to bake in trunk.
Bootstrapped and tested on aarch64-none-linux-gnu.
PR target/84748
* config/aarch64/aarch64.md (*compare_cstore<mode>_insn): Mark pattern
as clobbering CC_REGNUM.
* gcc.c-torture/execute/pr84748.c: New test.
From-SVN: r258366
2018-03-08 Richard Biener <rguenther@suse.de>
PR middle-end/84552
* tree-scalar-evolution.c: Include tree-into-ssa.h.
(follow_copies_to_constant): Do not follow SSA names registered
for update.
* gcc.dg/graphite/pr84552.c: New testcase.
From-SVN: r258365
PR tree-optimization/84178 reports a couple of source files that ICE inside
ifcvt when compiled with -03 -fno-tree-forwprop (trunk and gcc 7).
Both cases involve problems with ifcvt's per-BB gimplified predicates.
Testcase 1 fails this assertion within release_bb_predicate during cleanup:
283 if (flag_checking)
284 for (gimple_stmt_iterator i = gsi_start (stmts);
285 !gsi_end_p (i); gsi_next (&i))
286 gcc_assert (! gimple_use_ops (gsi_stmt (i)));
The testcase contains a division in the loop, which leads to
if_convertible_loop_p returning false (due to gimple_could_trap_p being true
for the division). This happens *after* the per-BB gimplified predicates
have been created in predicate_bbs (loop).
Hence tree_if_conversion bails out to "cleanup", but the gimplified predicates
exist and make use of SSA names; for example this conjunction for two BB
conditions:
_4 = h4.1_112 != 0;
_175 = (signed char) _117;
_176 = _175 >= 0;
_174 = _4 & _176;
is using SSA names.
This assertion was added in r236498 (aka c3deca2519d97c55876869c57cf11ae1e5c6cf8b):
2016-05-20 Richard Biener <rguenther@suse.de>
* tree-if-conv.c (add_bb_predicate_gimplified_stmts): Use
gimple_seq_add_seq_without_update.
(release_bb_predicate): Assert we have no operands to free.
(if_convertible_loop_p_1): Calculate post dominators later.
Do not free BB predicates here.
(combine_blocks): Do not recompute BB predicates.
(version_loop_for_if_conversion): Save BB predicates around
loop versioning.
* gcc.dg/tree-ssa/ifc-cd.c: Adjust.
The following patch fixes this by adding a call to gimple_seq_discard
to release_bb_predicate. It also updates the assertion, so that
instead of asserting the stmts have no imm uses, instead assert that
they weren't added to a bb before discarding them (otherwise discarding
them would be a bug). We know this is the case because
insert_gimplified_predicates has:
/* Once the sequence is code generated, set it to NULL. */
set_bb_predicate_gimplified_stmts (bb, NULL);
but asserting it seems appropriate as a double-check.
The patch doesn't address the 2nd issue within PR tree-optimization/84178.
gcc/ChangeLog:
PR tree-optimization/84178
* tree-if-conv.c (release_bb_predicate): Remove the
the assertion that the stmts have NULL use_ops.
Discard the statements, asserting that they haven't
yet been added to a BB.
gcc/testsuite/ChangeLog:
PR tree-optimization/84178
* gcc.c-torture/compile/pr84178-1.c: New test.
From-SVN: r258363
