Richard Earnshaw [Fri, 16 Jun 2017 21:02:10 +0000 (21:02 +0000)]
[arm] Allow +opt on arbitrary cpu and architecture
This is the main patch to provide the infrastructure for adding
feature extensions to CPU and architecture specifications. It does not,
however, add all the extensions that we intend to support (just a small
number to permit some basic testing). Now, instead of having specific
entries in the architecture table for variants such as armv8-a+crc, the
crc extension is specified as an optional component of the armv8-a
architecture entry. Similar control can be added to CPU option names.
In both cases the list of permitted options is controlled by the main
architecture or CPU name to prevent arbitrary cross-products of options.
* config/arm/arm-cpus.in (armv8-a): Add options crc, simd crypto and
nofp.
(armv8-a+crc): Delete.
(armv8.1-a): Add options simd, crypto and nofp.
(armv8.2-a): Add options fp16, simd, crypto and nofp.
(armv8.2-a+fp16): Delete.
(armv8-m.main): Add option dsp.
(armv8-m.main+dsp): Delete.
(cortex-a8): Add fpu. Add nofp option.
(cortex-a9): Add fpu. Add nofp and nosimd options.
* config/arm/parsecpu.awk (gen_data): Generate option tables and
link to main cpu and architecture data structures.
(gen_comm_data): Only put isa attributes from the main architecture
in common tables.
(option): New statement for architecture and CPU entries.
* arm.c (struct cpu_option): New structure.
(struct processors): Add entry for options.
(arm_unrecognized_feature): New function.
(arm_parse_arch_cpu_name): Ignore any characters after the first
'+' character.
(arm_parse_arch_cpu_feature): New function.
(arm_configure_build_target): Separate out any CPU and architecture
features and parse separately. Don't error out if -mfpu=auto is
used with only an architecture string.
(arm_print_asm_arch_directives): New function.
(arm_file_start): Call it.
* config/arm/arm-cpu-cdata.h: Regenerated.
* config/arm/arm-cpu-data.h: Likewise.
* config/arm/arm-tables.opt: Likewise.
Richard Earnshaw [Fri, 16 Jun 2017 21:01:51 +0000 (21:01 +0000)]
[arm] Don't pass -mfpu=auto through to the assembler.
The assembler doesn't understand -mfpu=auto. The easiest way to handle this
is to surpress this value from being passed through, while still passing
through legacy values.
* config/arm/elf.h (ASM_SPEC): Only pass -mfpu through to the
assembler when it is not -mfpu=auto.
Richard Earnshaw [Fri, 16 Jun 2017 21:01:29 +0000 (21:01 +0000)]
[arm] Rewrite -march and -mcpu options for passing to
The assembler does not understand all the '+' options accepted by the
compiler. The best solution to this is to simply strip the extensions
and just pass the raw architecture or cpu name through to the
assembler. We will use .arch and .arch_extension directives anyway to
turn on or off individual features. We already do something similar
for big.little combinations and this just extends this principle a bit
further. This patch also fixes a possible bug by ensuring that the
limited string copy is correctly NUL-terminated.
While messing with this code I've also taken the opportunity to clean up
the duplicate definitions of EXTRA_SPEC_FUNCTIONS by moving it outside of
the ifdef wrapper.
* config/arm/arm.h (BIG_LITTLE_SPEC): Delete macro.
(ASM_REWRITE_SPEC_FUNCTIONS): New macro.
(BIG_LITTLE_CPU_SPEC_FUNCTIONS): Delete macro.
(ASM_CPU_SPEC): Rewrite.
(MCPU_MTUNE_NATIVE_FUNCTIONS): New macro.
(EXTRA_SPEC_FUNCTIONS): Move outside of ifdef. Use
MCPU_MTUNE_NATIVE_FUNCTIONS and ASM_REWRITE_SPEC_FUNCTIONS. Remove
reference to BIG_LITTLE_CPU_SPEC_FUNCTIONS.
* common/config/arm/arm-common.c (arm_rewrite_selected_cpu): Ensure
copied string is NUL-terminated. Also strip any characters prefixed
by '+'.
(arm_rewrite_selected_arch): New function.
(arm_rewrite_march): New function.
Richard Earnshaw [Fri, 16 Jun 2017 21:01:22 +0000 (21:01 +0000)]
[arm] Use strings for -march, -mcpu and -mtune options
In order to support more complex specifications for cpus and architectures
we need to move away from using enumerations to represent the set of
permitted options. This basic change just moves the option parsing
infrastructure over to that, but changes nothing more beyond generating
a hint when the specified option does not match a known target (previously
the help option was able to print out all the permitted values, but we
can no-longer do that.
* config/arm/arm.opt (x_arm_arch_string): New TargetSave option.
(x_arm_cpu_string, x_arm_tune_string): Likewise.
(march, mcpu, mtune): Convert to string-based options.
* config/arm/arm.c (arm_print_hint_for_core_or_arch): New function.
(arm_parse_arch_cpu_name): New function.
(arm_configure_build_target): Use arm_parse_arch_cpu_name to
identify selected architecture or CPU.
(arm_option_save): New function.
(TARGET_OPTION_SAVE): Redefine.
(arm_option_restore): Restore string options.
(arm_option_print): Print string options.
Martin Sebor [Fri, 16 Jun 2017 20:52:03 +0000 (20:52 +0000)]
PR tree-optimization/80934 - bzero should be assumed not to escape pointer argument
PR tree-optimization/80934 - bzero should be assumed not to escape pointer argument
PR tree-optimization/80933 - redundant bzero/bcopy calls not eliminated
gcc/ChangeLog:
PR tree-optimization/80933
PR tree-optimization/80934
* builtins.c (fold_builtin_3): Do not handle bcmp here.
* gimple-fold.c (gimple_fold_builtin_bcmp): New function.
(gimple_fold_builtin_bcopy, gimple_fold_builtin_bzero): Likewise.
(gimple_fold_builtin): Call them.
James Greenhalgh [Fri, 16 Jun 2017 17:29:56 +0000 (17:29 +0000)]
[Patch ARM] Fix PR71778
gcc/
PR target/71778
* config/arm/arm-builtins.c (arm_expand_builtin_args): Return TARGET
if given a non-constant argument for an intrinsic which requires a
constant.
libcc1/
* libcp1plugin.cc (plugin_build_decl): Adjust parm building.
(--This line, and those below, will be ignored--
M gcc/cp/parser.c
M gcc/cp/ChangeLog
M gcc/cp/decl.c
M gcc/cp/lambda.c
M gcc/cp/cp-tree.h
M gcc/cp/method.c
M gcc/cp/decl2.c
M libcc1/libcp1plugin.cc
M libcc1/ChangeLog
Nathan Sidwell [Fri, 16 Jun 2017 14:16:20 +0000 (14:16 +0000)]
Make rtti lazier
Make rtti lazier
* rtti.c (enum tinfo_kind): Add TK_DERIVED_TYPES,
TK_VMI_CLASS_TYPES, TK_MAX. Delete TK_FIXED.
(tinfo_names): New.
(typeid_ok_p): Add quotes to error messages. Use get_tinfo_desc.
(get_tinfo_decl): Use get_tinfo_desc.
(get_pseudo_ti_init): Likewise. Adjust VMI construction.
(create_pseudo_type_info): Delete.
(get_pseudo_ti_index): Just determine the index.
(get_tinfo_desc): New. Create all types lazily.
(create_tinfo_types): Just allocate the descriptor array.
(emit_support_tinfos): Use non-inserting type lookup. Set builtin
location.
Jonathan Wakely [Fri, 16 Jun 2017 11:54:59 +0000 (12:54 +0100)]
PR libstdc++/81092 add std::wstring symbols and bump library version
PR libstdc++/81092
* acinclude.m4: Bump libtool_VERSION.
* config/abi/post/i386-linux-gnu/baseline_symbols.txt: Update.
* config/abi/post/x86_64-linux-gnu/32/baseline_symbols.txt: Update.
* config/abi/pre/gnu.ver: Add wstring constructor symbols to
GLIBCXX_3.4.24 version and move random_device::_M_get_entropy() symbol
to new GLIBCXX_3.4.25 version.
* doc/xml/manual/abi.xml: Document new versions.
* doc/html/*: Regenerate.
* testsuite/21_strings/basic_string/cons/char/8.cc: Use base object
constructors to ensure required symbols are exported.
* testsuite/21_strings/basic_string/cons/wchar_t/8.cc: Likewise.
* testsuite/util/testsuite_abi.cc: Add new version.
Maxim Kuvyrkov [Fri, 16 Jun 2017 09:30:43 +0000 (09:30 +0000)]
Add prefetch configuration to aarch64 backend.
* config/aarch64/aarch64-protos.h (struct cpu_prefetch_tune):
New tune structure.
(struct tune_params): Use cpu_prefetch_tune instead of cache_line_size.
[Unrelated to main purpose of the patch] Place the pointer field last
to enable type checking errors when tune structure are wrongly merged.
* config/aarch64/aarch64.c (generic_prefetch_tune,)
(exynosm1_prefetch_tune, qdf24xx_prefetch_tune,)
(thunderx2t99_prefetch_tune): New tune constants.
(tune_params *_tunings): Update all tunings (no functional change).
(aarch64_override_options_internal): Set PARAM_SIMULTANEOUS_PREFETCHES,
PARAM_L1_CACHE_SIZE, PARAM_L1_CACHE_LINE_SIZE, and PARAM_L2_CACHE_SIZE
from tunings structures.
Jakub Jelinek [Fri, 16 Jun 2017 09:13:28 +0000 (11:13 +0200)]
re PR sanitizer/81094 (-fsanitize=object-size does not instrument aggregate call arguments)
PR sanitizer/81094
* ubsan.c (instrument_null): Add T argument, use it instead
of computing it based on IS_LHS.
(instrument_object_size): Likewise.
(pass_ubsan::execute): Adjust instrument_null and
instrument_object_size callers to pass gimple_get_lhs or
gimple_assign_rhs1 result to it. Use instrument_null instead of
calling get_base_address and instrument_mem_ref. Handle
aggregate call arguments for object-size sanitization.
PR c++/80560
* dumpfile.c (dump_register): Avoid calling memset to initialize
a class with a default ctor.
* gcc.c (struct compiler): Remove const qualification.
* genattrtab.c (gen_insn_reserv): Replace memset with initialization.
* hash-table.h: Ditto.
* ipa-cp.c (allocate_and_init_ipcp_value): Replace memset with
assignment.
* ipa-prop.c (ipa_free_edge_args_substructures): Ditto.
* omp-low.c (lower_omp_ordered_clauses): Replace memset with
default ctor.
* params.h (struct param_info): Make struct members non-const.
* tree-switch-conversion.c (emit_case_bit_tests): Replace memset
with default initialization.
* vec.h (vec_copy_construct, vec_default_construct): New helper
functions.
(vec<T>::copy, vec<T>::splice, vec<T>::reserve): Replace memcpy
with vec_copy_construct.
(vect<T>::quick_grow_cleared): Replace memset with default ctor.
(vect<T>::vec_safe_grow_cleared, vec_safe_grow_cleared): Same.
* doc/invoke.texi (-Wclass-memaccess): Document.
libcpp/ChangeLog:
PR c++/80560
* line-map.c (line_maps::~line_maps): Avoid calling htab_delete
with a null pointer.
(linemap_init): Avoid calling memset on an object of a non-trivial
type.
libitm/ChangeLog:
PR c++/80560
* beginend.cc (GTM::gtm_thread::rollback): Avoid calling memset
on an object of a non-trivial type.
(GTM::gtm_transaction_cp::commit): Use assignment instead of memcpy
to copy an object.
* method-ml.cc (orec_iterator::reinit): Avoid -Wclass-memaccess.
gcc/testsuite/ChangeLog:
PR c++/80560
* g++.dg/Wclass-memaccess.C: New test.
crtl->is_leaf has a comment above it referring to "Local Register
Allocation". However this is set by IRA and not LRA since the meaning
of Local Register Allocator in GCC has changed quite drastically since
1999 when this comment was introduced above the variable
current_function_is_leaf. From memory gcc in those days had "local"
and "global" register allocation - however all of this has changed
quite a bit drastically since.
This patch merely updates the comment as I did some head scratching
this morning about how this could be set before Local Register
Allocator especially given the comment in ira.c as below.
/* Determine if the current function is a leaf before running IRA
since this can impact optimizations done by the prologue and
epilogue thus changing register elimination offsets. */
crtl->is_leaf = leaf_function_p ();
Fix problems with the MSP430 port's handling of persistent data.
PR target/78818
gcc * config/msp430/msp430.c (msp430_data_attr): Check that it's possible
for a variable to have a section before checking if the section has a
name.
Set section to.persistent if persistent attribute is set.
Warn if .persistent attribute is used on an automatic variable.
tests * gcc.target/msp430/pr78818-real.c: New template for tests.
* gcc.target/msp430/pr78818-auto.c: New test.
* gcc.target/msp430/pr78818-data-region.c: New test.
* gcc.target/msp430/pr78818-data-sec.c: New test.
* gcc.target/msp430/pr78818-auto-warn.c: New test.
2017-06-15 Thomas Preud'homme <thomas.preudhomme@arm.com>
gcc/testsuite/
* gcc.target/arm/its.c: Check that no IT blocks has more than 2
instructions in it rather than the number of IT blocks being 2.
Transfer scan directive arm_thumb2 restriction to the whole
testcase and restrict further to Thumb-only targets.
Ian Lance Taylor [Wed, 14 Jun 2017 23:42:53 +0000 (23:42 +0000)]
compiler: avoid crash on erroneous type
If there is an error constructing the backend type, the GCC backend
will report that the size is 1. That will then cause construction of
the ptrmask to crash. Avoid that case by just generating an empty
ptrmask.
Noticed while compiling a broken package. The policy I've been
following is to not commit a test case for a compiler crash on invalid
code, so no test case.
gcc/testsuite/
* gcc.target/mips/data-sym-jump.c: Adjust for whitespace changes.
* gcc.target/mips/pr51513-1.c: New test.
* gcc.target/mips/pr51513-2.c: New test.
Max Filippov [Wed, 14 Jun 2017 17:19:44 +0000 (17:19 +0000)]
gcc: xtensa: allow XCHAL_* macros to be non-constant
XCHAL_* macros from the xtensa-config.h are used in a number of places
that require them to be preprocessor constants. Rewrite these places so
that non-constant XCHAL_* definitions could be used there.
2017-06-14 Max Filippov <jcmvbkbc@gmail.com>
gcc/
* config/xtensa/xtensa.c (xtensa_option_override): Append
MASK_CONST16 to target_flags in the absence of TARGET_L32R.
(hwloop_optimize, hwloop_fail, hwloop_pattern_reg,
xtensa_doloop_hooks): Define unconditionally.
(xtensa_reorg_loops): Only call reorg_loops in the presence of
TARGET_LOOPS.
* config/xtensa/xtensa.h (TARGET_L32R): New definition.
(TARGET_DEFAULT): Remove XCHAL_HAVE_L32R condition and account
for it in xtensa_option_override.
(HARD_FRAME_POINTER_IS_FRAME_POINTER,
HARD_FRAME_POINTER_IS_ARG_POINTER): New definitions.
Boris Kolpackov [Wed, 14 Jun 2017 16:05:57 +0000 (16:05 +0000)]
Recognize '-' as special -MF argument (write to stdout)
Sometimes it is useful to generate pre-processed output to a file and
the dependency information to stdout for further analysis/processing.
For example:
g++ -E -MD -fdirectives-only -o test.ii test.cxx
This will generate the dependency information to test.d (as per the
documentation). While changing this behavior is probably unwise, one
traditional (e.g., supported by -o) way to handle this is to recognize
the special '-' file name as an instruction to write to stdout:
Wilco Dijkstra [Wed, 14 Jun 2017 14:51:46 +0000 (14:51 +0000)]
Improve Cortex-A53 FP scheduler
The Cortex-A53 scheduler model of FMAC bypass is not quite right
for FMAC to FMAC forwarding. Experiments also show the latencies of
FP operations are too high as well. Rather than adding more bypasses,
adjust the latencies of FP instructions to get a better schedule on
average. As a result SPECFP2006 is 1.1% faster.
Ian Lance Taylor [Wed, 14 Jun 2017 13:59:02 +0000 (13:59 +0000)]
cmd/cgo: make _cgo_flags consistent across runs
The go tool will pass -I objdir as one of the flags, where objdir is
the temporary build directory. Remove that from _cgo_flags: we don't
need it, and it will be different each time.
Sort the flags to avoid the unpredictable map iteration order.
This matters for gccgo because for a package that uses cgo, the go
tool when building for gccgo will store the _cgo_flags file in the
archive. That means that we want to generate identical _cgo_flags for
every run.
The test for this is the cmd/go testsuite, to follow in a future CL.
Ian Lance Taylor [Wed, 14 Jun 2017 13:47:28 +0000 (13:47 +0000)]
internal/testenv: support testing gccgo
If GO_TESTING_GOTOOLS is set in the environment, permit tests using
gccgo to run the go tool. Like GO_BUILDER_NAME, this should not be set
normally. But it is needed when testing the go tool itself, and will
be set by the gotools Makefile in a future CL.
Richard Biener [Wed, 14 Jun 2017 11:40:20 +0000 (11:40 +0000)]
re PR middle-end/81088 (UBSAN: false positive as a result of reassosiation)
2017-06-14 Richard Biener <rguenther@suse.de>
PR middle-end/81088
* fold-const.c (split_tree): Drop TREE_OVERFLOW flag from
literal constants.
(fold_binary_loc): When associating do not treat pre-existing
TREE_OVERFLOW on literal constants as a reason to allow
TREE_OVERFLOW on associated literal constants.
Eric Botcazou [Wed, 14 Jun 2017 11:22:19 +0000 (11:22 +0000)]
sparc.h (MASK_ISA): Add MASK_LEON and MASK_LEON3.
* config/sparc/sparc.h (MASK_ISA): Add MASK_LEON and MASK_LEON3.
(MASK_FEATURES): New macro.
* config/sparc/sparc.c (sparc_option_override): Remove the special
handling of -mfpu and generalize it to all MASK_FEATURES switches.
Ian Lance Taylor [Tue, 13 Jun 2017 14:26:07 +0000 (14:26 +0000)]
compiler: containing small bfunction mixup in Gogo::write_globals
Fix buglet in Gogo::write_globals-- in a couple of places the
wrong Bfunction was being used for the containing (not target)
function when creating calls for init functions.
Martin Liska [Tue, 13 Jun 2017 13:59:05 +0000 (15:59 +0200)]
Implement no_sanitize function attribute
2017-06-13 Martin Liska <mliska@suse.cz>
PR sanitize/78204
* c-c++-common/ubsan/attrib-2.c (float_cast2): Enhance the
test by adding no_sanitize attribute.
* gcc.dg/asan/use-after-scope-4.c: Likewise.
2017-06-13 Martin Liska <mliska@suse.cz>
PR sanitize/78204
* c-attribs.c (add_no_sanitize_value): New function.
(handle_no_sanitize_attribute): Likewise.
(handle_no_sanitize_address_attribute): Use the function.
(handle_no_sanitize_thread_attribute): New function.
(handle_no_address_safety_analysis_attribute): Use
add_no_sanitize_value.
(handle_no_sanitize_undefined_attribute): Likewise.
* c-common.h: Declare new functions.
* c-ubsan.c (ubsan_instrument_division): Use sanitize_flags_p.
(ubsan_instrument_shift): Likewise.
(ubsan_instrument_bounds): Likewise.
(ubsan_maybe_instrument_array_ref): Likewise.
(ubsan_maybe_instrument_reference_or_call): Likewise.
2017-06-13 Martin Liska <mliska@suse.cz>
PR sanitize/78204
* asan.c (asan_sanitize_stack_p): Use sanitize_flags_p.
(gate_asan): Likewise.
* asan.h (asan_no_sanitize_address_p): Remove the function.
(sanitize_flags_p): New function.
* builtins.def: Fix coding style.
* common.opt: Use renamed enum value.
* convert.c (convert_to_integer_1): Use sanitize_flags_p.
* doc/extend.texi: Document no_sanitize attribute.
* flag-types.h (enum sanitize_code): Rename SANITIZE_NONDEFAULT
to SANITIZE_UNDEFINED_NONDEFAULT.
* gcc.c (sanitize_spec_function): Use the renamed enum value.
* gimple-fold.c (optimize_atomic_compare_exchange_p):
Use sanitize_flags_p.
* gimplify.c (gimplify_function_tree): Likewise.
* ipa-inline.c (sanitize_attrs_match_for_inline_p): Likewise.
* opts.c (parse_no_sanitize_attribute): New function.
(common_handle_option): Use renamed enum value.
* opts.h (parse_no_sanitize_attribute): Declare.
* tree.c (sanitize_flags_p): New function.
* tree.h: Declared here.
* tsan.c: Use sanitize_flags_p.
* ubsan.c (ubsan_expand_null_ifn): Likewise.
(instrument_mem_ref): Likewise.
(instrument_bool_enum_load): Likewise.
(do_ubsan_in_current_function): Remove the function.
(pass_ubsan::execute): Use sanitize_flags_p.
* ubsan.h: Remove do_ubsan_in_current_function
* tree-cfg.c (print_no_sanitize_attr_value): New function.
(dump_function_to_file): Use it here.
2017-06-13 Martin Liska <mliska@suse.cz>
Renlin Li [Tue, 13 Jun 2017 13:57:45 +0000 (13:57 +0000)]
[ARM][Testsuite]make dump_stack function call conditional in cold-lc.c
After r249013, die () and dump_stack () are both in cold section. This makes
the compiler generate bl instruction for the function call, instead of
honoring the -mlong-calls option.
This patch changes the dump_stack function call conditional, which fixes the
regression.
Ian Lance Taylor [Tue, 13 Jun 2017 13:29:26 +0000 (13:29 +0000)]
runtime: don't always show frames with no function in traceback
If there is no function name, the traceback is generally
uninformative. In earlier versions we did not show such frames.
Restore that behavior. These frames can be seen with GOTRACEBACK=system.
Martin Jambor [Tue, 13 Jun 2017 11:40:24 +0000 (13:40 +0200)]
[PR80803 2/2] Diligent queuing in SRA grp_write prop
2017-06-13 Martin Jambor <mjambor@suse.cz>
PR tree-optimization/80803
PR tree-optimization/81063
* tree-sra.c (subtree_mark_written_and_enqueue): Move up in the file.
(propagate_subaccesses_across_link): Enqueue subtree whneve necessary
instead of relying on the caller.
testsuite/
gcc.dg/tree-ssa/pr80803.c: New test.
gcc.dg/tree-ssa/pr81063.c: Likewise.
Martin Jambor [Tue, 13 Jun 2017 11:34:51 +0000 (13:34 +0200)]
[PR80803 1/2] Streamline SRA access enqueuing
2017-06-13 Martin Jambor <mjambor@suse.cz>
* tree-sra.c (add_access_to_work_queue): Only enqueue accesses
that have a first_link.
(sort_and_splice_var_accesses): Do not check first_link before
enquing.
(subtree_mark_written_and_enqueue): Likewise.
(propagate_all_subaccesses): Likewise and do not stop at first
parent with a first_link.
projects / thirdparty / gcc.git / log
