2017-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_eval.adb (Check_Expression_Against_Static_Predicate):
If expression is compile-time known and obeys a static predicate
it must be labelled as static, to prevent spurious warnings and
run-time errors, e.g. in case statements. This is relevant when
the expression is the result of constant-folding a type conversion
whose expression is a variable with a known static value.
2017-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* exp_attr.adb, sem_attr.ads: Minor reformatting.
From-SVN: r244135
2017-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch5.adb (Analyze_Loop_Statement): If the loop includes an
iterator specification with a serious syntactic error, transform
construct into an infinite loop in order to continue analysis
and prevent a compiler abort.
2017-01-06 Tristan Gingold <gingold@adacore.com>
* exp_ch9.adb (Expand_N_Protected_Type_Declaration): Do not generate
max_queue_lengths_array if unused.
2017-01-06 Bob Duff <duff@adacore.com>
* errout.adb (Set_Msg_Text): Protect against out-of-bounds
array access, in case "\" is at the end of Text.
* stylesw.adb (Set_Style_Check_Options): Don't include input
characters in the error message template, because they could
be control characters such as "\", which Errout will try to
interpret.
2017-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_ch4.adb (Find_Indexing_Operations, Inspect_Declarations):
For a private type examine the visible declarations that follow
the partial view, not just the private declarations that follow
the full view.
2017-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* exp_ch5.adb, sem_ch3.adb, checks.adb: Minor reformatting and
code cleanup.
From-SVN: r244133
2017-01-06 Ed Schonberg <schonberg@adacore.com>
* exp_ch5.adb (Get_Default_Iterator): For a derived type, the
alias of the inherited op is the parent iterator, no need to
examine dispatch table positions which might not be established
yet if type is not frozen.
* sem_disp.adb (Check_Controlling_Formals): The formal of a
predicate function may be a subtype of a tagged type.
* sem_ch3.adb (Complete_Private_Subtype): Adjust inheritance
of representation items for the completion of a type extension
where a predicate applies to the partial view.
* checks.ads, checks.adb (Apply_Predicate_Check): Add optional
parameter that designates function whose actual receives a
predicate check, to improve warning message when the check will
lead to infinite recursion.
* sem_res.adb (Resolve_Actuals): Pass additional parameter to
Apply_Predicate_Check.
From-SVN: r244132
2017-01-06 Yannick Moy <moy@adacore.com>
* ghost.adb Minor fixing of references to SPARK RM.
(Check_Ghost_Context): Check whether reference is to a lvalue
before issuing an error about violation of SPARK RM 6.9(13)
when declaration has Ghost policy Check and reference has Ghost
policy Ignore.
* sem_util.adb Minor indentation.
* sem_ch10.adb (Analyze_Package_Body_Stub, Analyze_Protected_Body_Stub,
Analyze_Task_Body_Stub): Set Ekind of the defining identifier.
* sem_util.ads (Unique_Defining_Entity): Document the result
for package body stubs.
2017-01-06 Tristan Gingold <gingold@adacore.com>
* raise-gcc.c (abort): Macro to call Abort_Propagation.
* s-tpoben.ads (Protected_Entry_Queue_Max_Access): Make it access
constant.
* exp_ch9.adb (Expand_N_Protected_Type_Declaration):
Do not generate the Entry_Max_Queue_Lengths_Array if all default
values.
* exp_util.adb (Corresponding_Runtime_Package): Consider
Max_Queue_Length pragma.
From-SVN: r244129
2017-01-06 Justin Squirek <squirek@adacore.com>
* exp_ch9.adb (Expand_N_Protected_Type_Declaration):
Remove declaration generation in the case of
System_Tasking_Protected_Objects_Single_Entry being used,
and add a warning message when this is detected to occur.
(Make_Initialize_Protection): Remove reference pass in the case
of System_Tasking_Protected_Objects_Single_Entry.
* rtsfind.ads: Remove RE_Protected_Entry_Queue_Max
* s-tposen.adb (Initialize_Protection_Entry): Remove
Entry_Queue_Max parameter.
* s-tposen.ads: Remove the types use to store the entry queue
maximum.
* sem_prag.adb (Analyze_Pragma): Remove entry families restriction
2017-01-06 Yannick Moy <moy@adacore.com>
* sem_util.adb, sem_util.ads (Get_Enum_Lit_From_Pos): Strengthen
behavior of function, to also accept out of range positions
and raise Constraint_Error in such case, and to copy sloc from
literal if No_Location passed as location.
* uintp.adb, uintp.ads (UI_To_Int, UI_To_CC): Strengthen behavior
of functions to raise Constraint_Error in case of value not in
appropriate range.
2017-01-06 Tristan Gingold <gingold@adacore.com>
* sem_util.adb, s-taprop-linux.adb (Finalize_TCB): Remove call to
Invalidate_Stack_Cache.
2017-01-06 Eric Botcazou <ebotcazou@adacore.com>
* s-os_lib.adb: Minor fix to the signature of Readlink.
2017-01-06 Javier Miranda <miranda@adacore.com>
* sem_ch6.adb (Conforming_Types): Handle another
confusion between views in a nested instance with an actual
private type whose full view is not in scope.
2017-01-06 Arnaud Charlet <charlet@adacore.com>
* exp_ch5.adb (Expand_N_If_Statement): Obey existing comment and
mark a rewritten if statement as explicit (Comes_From_Source).
From-SVN: r244128
2017-01-06 Tristan Gingold <gingold@adacore.com>
* ada.ads, a-unccon.ads: Add pragma No_Elaboration_Code_All.
2017-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* sem_case.adb: Minor reformatting.
2017-01-06 Thomas Quinot <quinot@adacore.com>
* g-socthi-mingw.adb: Remove now extraneous USE TYPE clause
2017-01-06 Justin Squirek <squirek@adacore.com>
* aspects.adb: Register aspect in Canonical_Aspect.
* aspects.ads: Associate qualities of Aspect_Max_Queue_Length
into respective tables.
* einfo.ads, einfo.adb: Add a new attribute for
handling the parameters for Pragma_Max_Entry_Queue
(Entry_Max_Queue_Lengths_Array) in E_Protected_Type. Subprograms
for accessing and setting were added as well.
* par-prag.adb (Prag): Register Pramga_Max_Entry_Queue.
* exp_ch9.adb (Expand_N_Protected_Type_Declaration): Emit
declaration for pramga arguments and store them in the protected
type node.
(Make_Initialize_Protection): Pass a reference to
the Entry_Max_Queue_Lengths_Array in the protected type node to
the runtime.
* rtsfind.adb: Minor grammar fix.
* rtsfind.ads: Register new types taken from the
runtime libraries RE_Protected_Entry_Queue_Max and
RE_Protected_Entry_Queue_Max_Array
* s-tposen.adb, s-tpoben.adb
(Initialize_Protection_Entry/Initialize_Protection_Entries):
Add extra parameter and add assignment to local object.
* s-tposen.ads, s-tpoben.ads: Add new types to
store entry queue maximums and a field to the entry object record.
* sem_ch13.adb (Analyze_Aspect_Specifications): Add case statement
for Aspect_Max_Queue_Length.
(Check_Aspect_At_Freeze_Point):
Add aspect to list of aspects that don't require delayed analysis.
* sem_prag.adb (Analyze_Pragma): Add case statement for
Pragma_Max_Queue_Length, check semantics, and register arugments
in the respective entry nodes.
* sem_util.adb, sem_util.ads Add functions Get_Max_Queue_Length
and Has_Max_Queue_Length
* snames.ads-tmpl: Add constant for the new aspect-name
Name_Max_Queue_Length and corrasponding pragma.
2017-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* exp_util.adb (Is_Controlled_Function_Call):
Reimplemented. Consider any node which has an entity as the
function call may appear in various ways.
From-SVN: r244126
2017-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* exp_attr.adb (Rewrite_Stream_Proc_Call): Use
an unchecked type conversion when performing a view conversion
to/from a private type. In all other cases use a regular type
conversion to ensure that any relevant checks are properly
installed.
2017-01-06 Hristian Kirtchev <kirtchev@adacore.com>
* sem_prag.adb, sem_ch8.adb: Minor reformatting.
2017-01-06 Ed Schonberg <schonberg@adacore.com>
* sem_case.adb (Explain_Non_Static_Bound): Suppress cascaded
error on case expression that is an entity, when coverage is
incomplete and entity has a static value obtained by local
propagation.
(Handle_Static_Predicate): New procedure, subsidiary of
Check_Choices, to handle case alternatives that are either
subtype names or subtype indications involving subtypes that
have static predicates.
2017-01-06 Thomas Quinot <quinot@adacore.com>
* s-oscons-tmplt.c, g-socket.adb, g-socket.ads, g-sothco.ads:
(GNAT.Socket): Add support for Busy_Polling and Generic_Option
2017-01-06 Bob Duff <duff@adacore.com>
* sem_elab.adb (Activate_Elaborate_All_Desirable): Don't add
Elaborate_All(P) to P itself. That could happen in obscure cases,
and always introduced a cycle (P body must be elaborated before
P body).
* lib-writ.ads: Comment clarification.
* ali-util.ads: Minor comment fix.
* ali.adb: Minor reformatting.
2017-01-06 Tristan Gingold <gingold@adacore.com>
* a-exexpr-gcc.adb: Improve comment.
From-SVN: r244125
simplify_transformation_to_array had the nested loop unrolled 7 times,
which is reasonable given that it iterates over arrays of size
GFC_MAX_DIMENSIONS == 7.
The problem is that the last iteration increments the index, tests
that it's less than result->rank, and then accesses the arrays with
the incremented index.
We did not optimize out that part in the 7th iteration, so VRP flagged
the unreachable code as accessing arrays past the end.
It couldn't possibly know that we'd never reach that part, since the
test was on result->rank, and it's not obvious (for the compiler) that
result->rank <= GFC_MAX_DIMENSIONS.
Even an assert to that effect before the enclosing loop didn't avoid
the warning turned to error, though; I suppose there might be some
aliasing at play, because moving the assert into the loop does. An
assert on the index itself would also work, even more efficiently, but
we're just silencing the warning instead.
for gcc/fortran/ChangeLog
* simplify.c (simplify_transformation_to_array): Silence
array bounds warning. Fix whitespace.
From-SVN: r244122
In stage2 of bootstrap-O1, the code that warns if sprintf might
overflow its output buffer cannot tell that an unsigned value narrowed
to 16 bits will fit in 4 bytes with %4x.
Converting the value to 'unsigned short' makes it obvious that it
fits, at least on machines with 16-bit shorts.
for gcc/c-family/ChangeLog
* c-pretty-print.c (pp_c_tree_decl_identifier): Convert 16-bit
value to unsigned short to fit in 4 hex digits without
warnings.
From-SVN: r244121
gcc/ChangeLog:
PR tree-optimization/78910
* gimple-ssa-sprintf.c (tree_digits): Add an argument.
(format_integer): Correct off-by-one error in the handling
of precision with negative numbers in signed conversions..
gcc/testsuite/ChangeLog:
PR tree-optimization/78910
* gcc.dg/tree-ssa/builtin-sprintf-warn-7.c: Adjust text of expected
diagnostics.
* gcc.dg/tree-ssa/builtin-sprintf.c: Add test cases.
* gcc.dg/tree-ssa/pr78910.c: New test.
From-SVN: r244116
PR tree-optimization/71016
* tree-ssa-phiopt.c (tree_ssa_phiopt_worker): Pass cond_stmt to
factor_out_conditional_conversion. Formatting fix.
(factor_out_conditional_conversion): Add cond_stmt argument.
If arg1 is INTEGER_CST, punt if new_arg0 is not any operand of
cond_stmt and if arg0_def_stmt is not the only stmt in its bb.
Formatting fix.
* gcc.target/i386/pr71016.c: New test.
* gcc.target/aarch64/pr71016.c: New test.
* gcc.dg/tree-ssa/pr66726-3.c: New test.
From-SVN: r244114
PR c++/78931
* decl.c (cp_finish_decomp): Remove probe variable, if tt is
REFERENCE_REF_P, set tt to its operand.
* g++.dg/cpp1z/decomp19.C: New test.
From-SVN: r244113
PR c++/78890
* class.c (check_field_decls): Diagnose REFERENCE_TYPE fields in
unions even for C++11 and later.
* g++.dg/init/ref14.C: Expect error even in C++11 and later.
* g++.dg/init/union1.C: Likewise.
* g++.dg/cpp0x/union6.C: Expect errors.
* g++.dg/cpp0x/union8.C: New test.
* g++.dg/cpp0x/pr78890-1.C: New test.
* g++.dg/cpp0x/pr78890-2.C: New test.
From-SVN: r244112
This is the combination of these patches:
- [8a/9] Introduce class function_reader (v8)
- Add ASSERT_RTX_PTR_EQ
- [8b/9] Add target-independent selftests of RTL function reader (v2)
- [8c/9] Add aarch64-specific selftests for RTL function reader (v2)
- [8d/9] Add x86_64-specific selftests for RTL function reader (v2)
gcc/ChangeLog:
* Makefile.in (OBJS): Add read-md.o, read-rtl.o,
read-rtl-function.o, and selftest-rtl.o.
* config/aarch64/aarch64.c: Include selftest.h and
selftest-rtl.h.
(selftest::aarch64_test_loading_full_dump): New function.
(selftest::aarch64_run_selftests): New function.
(TARGET_RUN_TARGET_SELFTESTS): Wire it up to
selftest::aarch64_run_selftests.
* config/i386/i386.c
(selftest::ix86_test_loading_dump_fragment_1): New function.
(selftest::ix86_test_loading_call_insn): New function.
(selftest::ix86_test_loading_full_dump): New function.
(selftest::ix86_test_loading_unspec): New function.
(selftest::ix86_run_selftests): Call the new functions.
* emit-rtl.c (maybe_set_max_label_num): New function.
* emit-rtl.h (maybe_set_max_label_num): New decl.
* function.c (instantiate_decls): Guard call to
instantiate_decls_1 with if (DECL_INITIAL (fndecl)).
* function-tests.c (selftest::verify_three_block_rtl_cfg): Remove
"static".
* gensupport.c (gen_reader::gen_reader): Pass "false"
for new "compact" param of rtx_reader.
* print-rtl.c (rtx_writer::print_rtx_operand): Print "(nil)"
rather than an empty string for NULL strings.
* read-md.c: Potentially include config.h rather than bconfig.h.
Wrap include of errors.h with #ifdef GENERATOR_FILE.
(have_error): New global, copied from errors.c.
(md_reader::read_name): Rename to...
(md_reader::read_name_1): ...this, adding "out_loc" param,
and converting "missing name or number" to returning false, rather
than failing.
(md_reader::read_name): Reimplement in terms of read_name_1.
(md_reader::read_name_or_nil): New function.
(md_reader::read_string): Handle "(nil)" by returning NULL.
(md_reader::md_reader): Add new param "compact".
(md_reader::read_md_files): Wrap with #ifdef GENERATOR_FILE.
(md_reader::read_file): New method.
* read-md.h (md_reader::md_reader): Add new param "compact".
(md_reader::read_file): New method.
(md_reader::is_compact): New accessor.
(md_reader::read_name): Convert return type from void to
file_location.
(md_reader::read_name_or_nil): New decl.
(md_reader::read_name_1): New decl.
(md_reader::m_compact): New field.
(noop_reader::noop_reader): Pass "false" for new "compact" param
of rtx_reader.
(rtx_reader::rtx_reader): Add new "compact" param.
(rtx_reader::read_rtx_operand): Make virtual and convert return
type from void to rtx.
(rtx_reader::read_until): New decl.
(rtx_reader::handle_any_trailing_information): New virtual
function.
(rtx_reader::postprocess): New virtual function.
(rtx_reader::finalize_string): New virtual function.
(rtx_reader::m_in_call_function_usage): New field.
(rtx_reader::m_reuse_rtx_by_id): New field.
* read-rtl-function.c: New file.
* selftest-rtl.c (selftest::assert_rtx_ptr_eq_at): New function.
* selftest-rtl.h (ASSERT_RTX_PTR_EQ): New macro.
(selftest::verify_three_block_rtl_cfg): New decl.
* read-rtl-function.h: New file.
* read-rtl.c: Potentially include config.h rather than bconfig.h.
For host, include function.h, memmodel.h, and emit-rtl.h.
(one_time_initialization): New function.
(struct compact_insn_name): New struct.
(compact_insn_names): New array.
(find_code): Handle insn codes in compact dumps.
(apply_subst_iterator): Wrap with #ifdef GENERATOR_FILE.
(bind_subst_iter_and_attr): Likewise.
(add_condition_to_string): Likewise.
(add_condition_to_rtx): Likewise.
(apply_attribute_uses): Likewise.
(add_current_iterators): Likewise.
(apply_iterators): Likewise.
(initialize_iterators): Guard usage of apply_subst_iterator with
#ifdef GENERATOR_FILE.
(read_conditions): Wrap with #ifdef GENERATOR_FILE.
(md_reader::read_mapping): Likewise.
(add_define_attr_for_define_subst): Likewise.
(add_define_subst_attr): Likewise.
(read_subst_mapping): Likewise.
(check_code_iterator): Likewise.
(rtx_reader::read_rtx): Likewise. Move one-time initialization
logic to...
(one_time_initialization): New function.
(rtx_reader::read_until): New method.
(read_flags): New function.
(parse_reg_note_name): New function.
(rtx_reader::read_rtx_code): Initialize "iterator" to NULL.
Handle reuse_rtx ids.
Wrap iterator lookup within #ifdef GENERATOR_FILE.
Add parsing support for RTL dumps, mirroring the special-cases in
print_rtx, by calling read_flags, reading REG_NOTE names, INSN_UID
values, and calling handle_any_trailing_information.
(rtx_reader::read_rtx_operand): Convert return type from void
to rtx, returning return_rtx. Handle case 'e'. Call
finalize_string on XSTR and XTMPL fields.
(rtx_reader::read_nested_rtx): Handle dumps in which trailing
"(nil)" values were omitted. Call the postprocess vfunc on the
return_rtx.
(rtx_reader::rtx_reader): Add new "compact" param and pass to base
class ctor. Initialize m_in_call_function_usage. Call
one_time_initialization.
* rtl-tests.c (selftest::test_uncond_jump): Call
set_new_first_and_last_insn.
* rtl.h (read_rtx): Wrap decl with #ifdef GENERATOR_FILE.
* selftest-rtl.c: New file.
* selftest-rtl.h (class selftest::rtl_dump_test): New class.
(selftest::get_insn_by_uid): New decl.
* selftest-run-tests.c (selftest::run_tests): Call
read_rtl_function_c_tests.
* selftest.h (selftest::read_rtl_function_c_tests): New decl.
* tree-dfa.c (ssa_default_def): Return NULL_TREE for rtl function
dumps.
gcc/testsuite/ChangeLog:
* selftests/asr_div1.rtl: New file.
* selftests/aarch64: New subdirectory.
* selftests/aarch64/times-two.rtl: New file.
* selftests/bb-index.rtl: New file.
* selftests/cfg-test.rtl: New file.
* selftests/const-int.rtl: New file.
* selftests/example-labels.rtl: New file.
* selftests/insn-with-mode.rtl: New file.
* selftests/jump-to-label-ref.rtl: New file.
* selftests/jump-to-return.rtl: New file.
* selftests/jump-to-simple-return.rtl: New file.
* selftests/mem.rtl: New file.
* selftests/note-insn-deleted.rtl: New file.
* selftests/note_insn_basic_block.rtl: New file.
* selftests/simple-cse.rtl: New file.
* selftests/symbol-ref.rtl: New file.
* selftests/x86_64: New subdirectory.
* selftests/x86_64/call-insn.rtl: New file.
* selftests/x86_64/copy-hard-reg-into-frame.rtl: New file.
* selftests/x86_64/times-two.rtl: New file.
* selftests/x86_64/unspec.rtl: New file.
From-SVN: r244110
* common.opt (fvect-cost-model): Remove RejectNegative flag, use
3 argument Alias with unlimited for the negative form.
(fno-vect-cost-model): Removed.
From-SVN: r244106
2017-01-05 Martin Liska <mliska@suse.cz>
* hsa-gen.c (gen_hsa_divmod): New function.
(gen_hsa_insn_for_internal_fn_call): Use the function
for IFN_DIVMOD.
From-SVN: r244104
cp/
PR c++/78765
* pt.c (convert_nontype_argument): Don't try and see if integral
or enum expressions are constants prematurely.
testsuite/
PR c++/78765
* g++.dg/cpp0x/pr78765.C: New.
From-SVN: r244101
These were provided by Dominik to check more of the corner case in our
memset/memcpy inline code.
gcc/testsuite/ChangeLog:
2017-01-05 Dominik Vogt <vogt@linux.vnet.ibm.com>
* gcc.target/s390/memcpy-2.c: New test.
* gcc.target/s390/memset-2.c: New test.
From-SVN: r244099
See the memset unrolling patch. The very same applies to memcpys with
constant lengths.
2017-01-05 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* config/s390/s390.c (s390_expand_movmem): Unroll MVC loop for
small constant length operands.
gcc/testsuite/ChangeLog:
2017-01-05 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* gcc.target/s390/memcpy-1.c: New test.
From-SVN: r244098
lengths.
When expanding a memset we emit a loop of MVCs/XCs instructions dealing
with 256 byte blocks. This loop used to get unrolled with older GCCs
when using constant length operands. GCC lost this ability probably
when more of the loop unrolling stuff has been moved to tree level.
With this patch the unrolling is done manually when emitting the RTL
insns.
2017-01-05 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* gcc.target/s390/memset-1.c: New test.
gcc/ChangeLog:
2017-01-05 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* config/s390/s390.c (s390_expand_setmem): Unroll the loop for
small constant length operands.
From-SVN: r244097
A memset with a value != 0 is currently implemented using the mvc
instruction propagating the first byte through 256 byte blocks. While
for the first mvc the byte is written with a separate instruction
subsequent MVCs used the last byte of the previous 256 byte block.
Starting with z13 this causes a major performance degradation. With
this patch we always set the first byte with an mvi or stc in order to
avoid the overlapping of the MVC operands between loop iterations.
On older machines this basically makes no measurable difference so the
patch enables the new behavior for all machine levels in order to make
sure that code built for older machine levels runs well when moved to
a z13.
Bootstrapped and regression tested on s390 and s390x using z900 and z13
as default -march level. No regressions.
gcc/ChangeLog:
2017-01-05 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* config/s390/s390.c (s390_expand_setmem): Avoid overlapping bytes
between loop iterations.
From-SVN: r244096
PR tree-optimizatin/78812
* rtl.h (contains_mem_rtx_p): Prototype.
* ifcvt.c (containts_mem_rtx_p): Move from here to...
* rtlanal.c (contains_mem_rtx_p): Here and remvoe static linkage.
* gcse.c (prune_expressions): Use contains_mem_rtx_p to discover
and prune MEMs that are not at the toplevel of a SET_SRC rtx. Look
through ZERO_EXTEND and SIGN_EXTEND when trying to avoid pruning
MEMs.
PR tree-optimization/78812
* g++.dg/torture/pr78812.C: New test.
From-SVN: r244093
Building with the bootstrap-O3 configuration option fails to compile
input.c due to an AFAICT false-positive warning about an uninitialized
use of a variable.
This patch adds a default initializer to silence it.
for gcc/ChangeLog
* input.c (assert_char_at_range): Default-initialize
actual_range.
From-SVN: r244091
This patch fixes a false-positive warning in df-scan, at bootstrap-O3
failed, and enables GCC to optimize out the code that leads to the
warning.
df_ref_create_structure was inlined into the else part of
df_ref_record. Due to the condition of the corresponding if, In the
else part, VRP deduced unsigned regno >= FIRST_PSEUDO_REGISTER.
In df_ref_create_structure, there's another regno variable,
initialized with the same expression and value as the caller's. GCC
can tell as much, but this regno variable is signed. It is used,
shifted right, to index a hard regset bit array within a path that
tests that this signed regno < FIRST_PSEUDO_REGISTER.
GCC warned about the possible out-of-range indexing into the hard
regset array. It shouldn't, after all, the same regno can't possibly
be both < FIRST_PSEUDO_REGISTER and >= FIRST_PSEUDO_REGISTER, can it?
Well, the optimizers correctly decide it could, if it was a negative
int that, when converted to unsigned, became larger than
FIRST_PSEUDO_REGISTER. But GCC doesn't know regno can't be negative,
so the test could not be optimize out. What's more, given the
constraints, VRP correctly concluded the hard regset array would
always be indexed by a value way outside the array index range.
This patch changes the inlined regno to unsigned, like the caller's,
so that we can now tell the conditions can't both hold, so we optimize
out the path containing the would-be out-of-range array indexing.
for gcc/ChangeLog
* df-scan.c (df_ref_create_structure): Make regno unsigned,
to match the caller.
From-SVN: r244090
A debug insn after the final jump of a basic block may cause the
expander to emit a dummy move where the non-debug compile won't
because it finds the jump insn at the end of the insn stream.
Fix the condition so that, instead of requiring the jump as the last
insn, it also matches a jump followed by debug insns.
This fixes the compilation of libgcc/libgcov-profiler.c with
-fcompare-debug on i686-linux-gnu.
for gcc/ChangeLog
* cfgexpand.c (expand_gimple_basic_block): Disregard debug
insns after final jump in test to emit dummy move.
From-SVN: r244089
Various Ada RTS files failed -fcompare-debug compilation because debug
stmts prevented EH cleanups from taking place. Adjusting
cleanup_empty_eh to skip them fixes it.
for gcc/ChangeLog
* gimple-iterator.h (gsi_one_nondebug_before_end_p): New.
* tree-eh.c (cleanup_empty_eh): Skip more debug stmts.
From-SVN: r244088
Building with the bootstrap-O3 configuration option fails to compile
fortran/module.c due to an AFAICT false-positive warning about an
uninitialized use of a variable.
This patch adds a dummy initializer to silence it.
for gcc/fortran/ChangeLog
* module.c (load_omp_udrs): Initialize name.
From-SVN: r244087
Building with the bootstrap-O1 configuration option fails to compile a
number of files due to AFAICT false-positive warnings about uses of
uninitialized variables.
This patch adds dummy initializers to silence them all.
for gcc/ChangeLog
* multiple_target.c (create_dispatcher_calls): Init e_next.
* tree-ssa-loop-split.c (split_loop): Init border.
* tree-vect-loop.c (vect_determine_vectorization_factor): Init
scalar_type.
From-SVN: r244086
[gcc]
2017-01-04 Michael Meissner <meissner@linux.vnet.ibm.com>
PR target/71977
PR target/70568
PR target/78823
* config/rs6000/predicates.md (sf_subreg_operand): New predicate.
(altivec_register_operand): Do not return true if the operand
contains a SUBREG mixing SImode and SFmode.
(vsx_register_operand): Likewise.
(vsx_reg_sfsubreg_ok): New predicate.
(vfloat_operand): Do not return true if the operand contains a
SUBREG mixing SImode and SFmode.
(vint_operand): Likewise.
(vlogical_operand): Likewise.
(gpc_reg_operand): Likewise.
(int_reg_operand): Likewise.
* config/rs6000/rs6000-protos.h (valid_sf_si_move): Add
declaration.
* config/rs6000/rs6000.c (valid_sf_si_move): New function to
determine if a MOVSI or MOVSF operation contains SUBREGs that mix
SImode and SFmode.
(rs6000_emit_move_si_sf_subreg): New helper function.
(rs6000_emit_move): Call rs6000_emit_move_si_sf_subreg to possbily
fixup SUBREGs involving SImode and SFmode.
* config/rs6000/vsx.md (SFBOOL_*): New constants that are operand
numbers for the new peephole2 optimization.
(peephole2 for SFmode unions): New peephole2 to optimize cases in
the GLIBC math library that do AND/IOR/XOR operations on single
precision floating point.
* config/rs6000/rs6000.h (TARGET_NO_SF_SUBREG): New internal
target macros to say whether we need to avoid SUBREGs mixing
SImode and SFmode.
(TARGET_ALLOW_SF_SUBREG): Likewise.
* config/rs6000/rs6000.md (UNSPEC_SF_FROM_SI): New unspecs.
(UNSPEC_SI_FROM_SF): Likewise.
(iorxor): Change spacing.
(and_ior_xor): New iterator for AND, IOR, and XOR.
(movsi_from_sf): New insns for SImode/SFmode SUBREG support.
(movdi_from_sf_zero_ext): Likewise.
(mov<mode>_hardfloat, FMOVE32 iterator): Use register_operand
instead of gpc_reg_operand. Add SImode/SFmode SUBREG support.
(movsf_from_si): New insn for SImode/SFmode SUBREG support.
(fma<mode>4): Use gpc_reg_operand instead of register_operand.
(fms<mode>4): Likewise.
(fnma<mode>4): Likewise.
(fnms<mode>4): Likewise.
(nfma<mode>4): Likewise.
(nfms<mode>4): Likewise.
[gcc/testsuite]
2017-01-04 Michael Meissner <meissner@linux.vnet.ibm.com>
PR target/71977
PR target/70568
PR target/78823
* gcc.target/powerpc/pr71977-1.c: New tests to check whether on
64-bit VSX systems with direct move, whether we optimize common
code sequences in the GLIBC math library for float math functions.
* gcc.target/powerpc/pr71977-2.c: Likewise.
From-SVN: r244084
PR c++/64767
* c.opt (Wpointer-compare): New option.
* c-parser.c (c_parser_postfix_expression): Mark zero character
constants by setting original_type in c_expr.
* c-typeck.c (parser_build_binary_op): Warn when a pointer is compared
with a zero character constant.
(char_type_p): New function.
* typeck.c (cp_build_binary_op): Warn when a pointer is compared with
a zero character literal.
* doc/invoke.texi: Document -Wpointer-compare.
* c-c++-common/Wpointer-compare-1.c: New test.
From-SVN: r244076
PR c++/78949
* typeck.c (cp_build_unary_op): Call mark_rvalue_use on arg if it has
vector type.
* c-c++-common/Wunused-var-16.c: New test.
From-SVN: r244075
