This patch uses df_read_modify_subreg_p to check whether writing
to a subreg would preserve some of the existing contents.
This has the effect of putting more emphasis on the
REGMODE_NATURAL_SIZE-based definition of whether something can be
partially modified, instead of using UNITS_PER_WORD unconditionally.
This becomes important for SVE, where UNITS_PER_WORD has no
significance for subregs of multi-register LD2/ST2, LD3/ST3 and
LD4/ST4 tuples.
2017-10-26 Richard Sandiford <richard.sandiford@linaro.org>
Alan Hayward <alan.hayward@arm.com>
David Sherwood <david.sherwood@arm.com>
Stop print_hex from printing bits above the precision
2017-10-26 Richard Sandiford <richard.sandiford@linaro.org>
gcc/
* wide-int-print.cc (print_hex): Loop based on extract_uhwi.
Don't print any bits outside the precision of the value.
* wide-int.cc (test_printing): Add some new tests.
James Greenhalgh [Thu, 26 Oct 2017 14:17:40 +0000 (14:17 +0000)]
[obvious][arm testsuite] Fixup expected location in require-pic-register-loc.c
After r254010 we now add -gcolumn-info by default, that means the tests
in gcc.target/arm/require-pic-register-loc.c need adjusting to not expect
to see column zero.
gcc/testsuite/
* gcc.target/arm/require-pic-register-loc.c: Use wider regex for
column information.
David Malcolm [Wed, 25 Oct 2017 23:53:41 +0000 (23:53 +0000)]
C: detect more missing semicolons (PR c/7356)
c_parser_declaration_or_fndef has logic for parsing what might be
either a declaration or a function definition.
This patch adds a test to detect cases where a semicolon would have
terminated the decls as a declaration, where the token that follows
would start a new declaration specifier, and updates the error message
accordingly, with a fix-it hint.
This addresses PR c/7356, fixing the case of a stray token before a
#include that previously gave inscrutable output, and improving e.g.:
int i
int j;
from:
t.c:2:1: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'int'
int j;
^~~
to:
t.c:1:6: error: expected ';' before 'int'
int i
^
;
int j;
~~~
gcc.dg/noncompile/920923-1.c needs a slight update, as the output for
the first line changes from:
gcc/testsuite/ChangeLog:
PR c/7356
PR c/44515
* c-c++-common/pr44515.c: New test case.
* gcc.dg/pr7356-2.c: New test case.
* gcc.dg/pr7356.c: New test case.
* gcc.dg/spellcheck-typenames.c: Update the "singed" char "TODO"
case to reflect changes to output.
* gcc.dg/noncompile/920923-1.c: Add dg-warning to reflect changes
to output.
Palmer Dabbelt [Wed, 25 Oct 2017 22:45:55 +0000 (22:45 +0000)]
RISC-V: Add Sign/Zero extend patterns for PIC loads
Loads on RISC-V are sign-extending by default, but we weren't telling
GCC this in our PIC load patterns. This corrects the problem, and adds
a zero-extending pattern as well.
gcc/ChangeLog
2017-10-25 Palmer Dabbelt <palmer@dabbelt.com>
* config/riscv/riscv.md (ZERO_EXTEND_LOAD): Define.
* config/riscv/pic.md (local_pic_load): Rename to local_pic_load_s,
mark as a sign-extending load.
(local_pic_load_u): Define.
Eric Botcazou [Wed, 25 Oct 2017 21:53:21 +0000 (21:53 +0000)]
re PR middle-end/82062 (simple conditional expressions no longer folded)
PR middle-end/82062
* fold-const.c (operand_equal_for_comparison_p): Also return true
if ARG0 is a simple variant of ARG1 with narrower precision.
(fold_ternary_loc): Always pass unstripped operands to the predicate.
Richard Biener [Wed, 25 Oct 2017 10:20:37 +0000 (10:20 +0000)]
tree-ssa-sccvn.h (vn_eliminate): Declare.
2017-10-25 Richard Biener <rguenther@suse.de>
* tree-ssa-sccvn.h (vn_eliminate): Declare.
* tree-ssa-pre.c (class eliminate_dom_walker, eliminate,
class pass_fre): Move to ...
* tree-ssa-sccvn.c (class eliminate_dom_walker, vn_eliminate,
class pass_fre): ... here and adjust for statistics.
Jakub Jelinek [Wed, 25 Oct 2017 08:05:58 +0000 (10:05 +0200)]
re PR libstdc++/81706 (std::sin vectorization bug)
PR libstdc++/81706
* attribs.c (attribute_value_equal): Use omp_declare_simd_clauses_equal
for comparison of OMP_CLAUSEs regardless of flag_openmp{,_simd}.
(duplicate_one_attribute, copy_attributes_to_builtin): New functions.
* attribs.h (duplicate_one_attribute, copy_attributes_to_builtin): New
declarations.
* c-decl.c (merge_decls): Copy "omp declare simd" attributes from
newdecl to corresponding __builtin_ if any.
* decl.c (duplicate_decls): Copy "omp declare simd" attributes from
newdecl to corresponding __builtin_ if any.
* gcc.target/i386/pr81706.c: New test.
* g++.dg/ext/pr81706.C: New test.
* tree-ssa-pre.c (need_eh_cleanup, need_ab_cleanup, el_to_remove,
el_to_fixup, el_todo, el_avail, el_avail_stack, eliminate_avail,
eliminate_push_avail, eliminate_insert): Move inside...
(class eliminate_dom_walker): ... this class in preparation
of move.
(fini_eliminate): Remove by merging with ...
(eliminate): ... this function. Adjust for class changes.
(pass_pre::execute): Remove fini_eliminate call.
(pass_fre::execute): Likewise.
Jakub Jelinek [Tue, 24 Oct 2017 19:35:37 +0000 (21:35 +0200)]
re PR target/82460 (AVX512: choose between vpermi2d and vpermt2d to save mov instructions. Also, fails to optimize away shifts before shuffle)
PR target/82460
* config/i386/sse.md (UNSPEC_VPERMI2, UNSPEC_VPERMI2_MASK): Remove.
(VPERMI2, VPERMI2I): New mode iterators.
(<avx512>_vpermi2var<mode>3_maskz): Remove 3 define_expand patterns.
(<avx512>_vpermi2var<mode>3<sd_maskz_name>): Remove 3 define_insn
patterns.
(<avx512>_vpermi2var<mode>3_mask): New define_expand using VPERMI2
mode iterator. Remove 3 old define_insn patterns.
(*<avx512>_vpermi2var<mode>3_mask): 2 new define_insn patterns.
(<avx512>_vpermt2var<mode>3_maskz): Adjust 1 define_expand to use
VPERMI2 mode iterator, remove the other two expanders.
(<avx512>_vpermt2var<mode>3<sd_maskz_name>): Adjust 1 define_insn
to use VPERMI2 mode iterator, add another alternative for vpermi2*
instructions, remove the other two patterns.
(<avx512>_vpermt2var<mode>3_mask): Adjust 1 define_insn to use VPERMI2
mode iterator, remove the other two patterns.
* config/i386/i386.c (ix86_expand_vec_perm_vpermi2): Renamed to ...
(ix86_expand_vec_perm_vpermt2): ... this. Swap mask and op0
arguments, use gen_*vpermt2* expanders instead of gen_*vpermi2*
and adjust argument order accordingly.
(ix86_expand_vec_perm): Adjust caller.
(expand_vec_perm_1): Likewise.
(expand_vec_perm_vpermi2_vpshub2): Rename to ...
(expand_vec_perm_vpermt2_vpshub2): ... this.
(ix86_expand_vec_perm_const_1): Adjust caller.
(ix86_vectorize_vec_perm_const_ok): Adjust comments.
* gcc.target/i386/pr82460-1.c: New test.
* gcc.target/i386/pr82460-2.c: New test.
* gcc.target/i386/avx512f-vpermt2pd-1.c: Adjust scan-assembler*
regexps to allow vpermt2* to vpermi2* replacement or vice versa
where possible.
* gcc.target/i386/avx512vl-vpermt2pd-1.c: Likewise.
* gcc.target/i386/avx512f-vpermt2d-1.c: Likewise.
* gcc.target/i386/vect-pack-trunc-2.c: Likewise.
* gcc.target/i386/avx512vl-vpermt2ps-1.c: Likewise.
* gcc.target/i386/avx512vl-vpermt2q-1.c: Likewise.
* gcc.target/i386/avx512f-vpermt2ps-1.c: Likewise.
* gcc.target/i386/avx512vl-vpermt2d-1.c: Likewise.
* gcc.target/i386/avx512bw-vpermt2w-1.c: Likewise.
* gcc.target/i386/avx512vbmi-vpermt2b-1.c: Likewise.
* gcc.target/i386/avx512f-vpermt2q-1.c: Likewise.
Wilco Dijkstra [Tue, 24 Oct 2017 17:21:19 +0000 (17:21 +0000)]
Cleanup autopref scheduling
r253236 broke AArch64 bootstrap. Earlier revision r253071 changed scheduling
behaviour on AArch64 as autopref scheduling no longer checks the base.
This patch fixes the bootstrap failure and cleans up autopref scheduling.
The code is greatly simplified. Sort accesses on the offset first, and
only if the offsets are the same fall back to other comparisons in
rank_for_schedule. This doesn't at all restore the original behaviour
since we no longer compare the base address, but it now defines a total
sorting order. More work will be required to improve the sorting so
that only loads/stores with the same base are affected.
Wilco Dijkstra [Tue, 24 Oct 2017 16:58:02 +0000 (16:58 +0000)]
PR60580: Fix frame pointer option magic
To fix PR60580 simplify the logic in aarch64_override_options_after_change_1 ().
If the frame pointer is enabled, set it to a special value that behaves similar
to frame pointer omission. If we don't do this all leaf functions will get a
frame pointer even if flag_omit_leaf_frame_pointer is set.
If flag_omit_frame_pointer has this special value, we must force the frame
pointer if not in a leaf function. We also need to force it in a leaf function
if flag_omit_frame_pointer is not set or if LR is used.
Doing this allows both -fomit-frame-pointer and -fomit-leaf-frame-pointer to be
independently set and changed in each function with the expected behaviour.
gcc/
PR middle-end/60580
* config/aarch64/aarch64.c (aarch64_frame_pointer_required)
Check special value of flag_omit_frame_pointer.
(aarch64_can_eliminate): Likewise.
(aarch64_override_options_after_change_1): Simplify handling of
-fomit-frame-pointer and -fomit-leaf-frame-pointer.
Alan Modra [Tue, 24 Oct 2017 12:45:01 +0000 (23:15 +1030)]
PR82687, g++.dg/asan/default-options-1.C fails with PR82575 fix
The problem with making discarded symbols hidden is that the
non-default visibility is sticky. When symbols other than the
__gnu_lto ones are discarded that turns out to be a bad idea.
PR lto/82687
PR lto/82575
* simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
Only make __gnu_lto symbols hidden. Delete outdated comment.
Silence ISO C warning.
H.J. Lu [Tue, 24 Oct 2017 10:52:50 +0000 (10:52 +0000)]
i386: Don't insert ENDBR at function entrance when called directly
There is no need to insert ENDBR instruction at function entrance if
function is only called directly.
gcc/
PR target/82659
* config/i386/i386.c (rest_of_insert_endbranch): Don't insert
ENDBR instruction at function entrance if function is only
called directly.
Jakub Jelinek [Tue, 24 Oct 2017 10:44:56 +0000 (12:44 +0200)]
re PR rtl-optimization/82628 (wrong code at -Os on x86_64-linux-gnu in the 32-bit mode)
PR target/82628
* config/i386/i386.md (addcarry<mode>, subborrow<mode>): Change
patterns to better describe from which operation the CF is computed.
(addcarry<mode>_0, subborrow<mode>_0): New patterns.
* config/i386/i386.c (ix86_expand_builtin) <case handlecarry>: Pass
one LTU with [DT]Imode and another one with [SD]Imode. If arg0
is 0, use _0 suffixed expanders instead of emitting a comparison
before it.
David Malcolm [Mon, 23 Oct 2017 20:25:58 +0000 (20:25 +0000)]
Add INCLUDE_UNIQUE_PTR and use it (PR bootstrap/82610)
gcc/ChangeLog:
PR bootstrap/82610
* system.h: Conditionally include "unique-ptr.h" if
INCLUDE_UNIQUE_PTR is defined.
* unique-ptr-tests.cc: Remove include of "unique-ptr.h" in favor
of defining INCLUDE_UNIQUE_PTR before including "system.h".
include/ChangeLog:
* unique-ptr.h: Make include of <memory> conditional on C++11 or
later.
Sebastian Perta [Mon, 23 Oct 2017 17:30:22 +0000 (13:30 -0400)]
Forgot to add the new file :-P 2017-10-13 Sebastian Perta <sebastian.perta@renesas.com>
Forgot to add the new file :-P
2017-10-13 Sebastian Perta <sebastian.perta@renesas.com>
* config/rl78/adddi3.S: New assembly file.
* config/rl78/t-rl78: Added adddi3.S to LIB2ADD.
Jan Hubicka [Mon, 23 Oct 2017 15:10:09 +0000 (17:10 +0200)]
i386.c (dimode_scalar_chain::compute_convert_gain): Use xmm_move instead of sse_move.
* i386.c (dimode_scalar_chain::compute_convert_gain): Use
xmm_move instead of sse_move.
(sse_store_index): New function.
(ix86_register_move_cost): Be more sensible about mismatch stall;
model AVX moves correctly; make difference between sse->integer and
integer->sse.
(ix86_builtin_vectorization_cost): Model correctly aligned and unaligned
moves; make difference between SSE and AVX.
* i386.h (processor_costs): Remove sse_move; add xmm_move, ymm_move
and zmm_move. Increase size of sse load and store tables;
add unaligned load and store tables; add ssemmx_to_integer.
* x86-tune-costs.h: Update all entries according to real
move latencies from Agner Fog's manual and chip documentation.
Jakub Jelinek [Mon, 23 Oct 2017 14:58:23 +0000 (16:58 +0200)]
re PR rtl-optimization/82628 (wrong code at -Os on x86_64-linux-gnu in the 32-bit mode)
PR target/82628
* config/i386/predicates.md (x86_64_dwzext_immediate_operand): New.
* config/i386/constraints.md (Wf): New constraint.
* config/i386/i386.md (UNSPEC_SBB): New unspec.
(cmp<dwi>_doubleword): Removed.
(sub<mode>3_carry_ccc, *sub<mode>3_carry_ccc_1): New patterns.
(sub<mode>3_carry_ccgz): Use unspec instead of compare.
* config/i386/i386.c (ix86_expand_branch) <case E_TImode>: Don't
expand with cmp<dwi>_doubleword. For LTU and GEU use
sub<mode>3_carry_ccc instead of sub<mode>3_carry_ccgz and use CCCmode.
Jonathan Wakely [Mon, 23 Oct 2017 12:11:22 +0000 (13:11 +0100)]
Implement C++17 Filesystem library
Based on Filesystem TS implementation, with the changes applied by:
- P0219R1 Relative Paths for Filesystem
- P0317R1 Directory Entry Caching for Filesystem
- P0492R2 Resolution of C++17 National Body Comments
Where appropriate code is shared between the TS and C++17
implementations.
* include/Makefile.am: Add new headers for C++17 filesystem library.
* include/Makefile.in: Regenerate.
* include/bits/fs_dir.h: New header, based on Filesystem TS code in
include/experimental/bits directory.
* include/bits/fs_fwd.h: Likewise.
* include/bits/fs_ops.h: Likewise.
* include/bits/fs_path.h: Likewise.
* include/experimental/bits/fs_dir.h: Rename Doxygen group.
* include/experimental/bits/fs_fwd.h: Likewise.
* include/experimental/bits/fs_ops.h: Likewise.
* include/experimental/bits/fs_path.h: Likewise.
* include/experimental/filesystem (filesystem_error::_M_gen_what):
Remove inline definition.
* include/precompiled/stdc++.h: Add <filesystem> to precompiled
header.
* include/std/filesystem: New header.
* python/libstdcxx/v6/printers.py: Enable printer for std::filesystem
paths.
* src/filesystem/Makefile.am: Add new files. Compile as C++17.
* src/filesystem/Makefile.in: Regenerate.
* src/filesystem/cow-dir.cc: Update comment.
* src/filesystem/cow-ops.cc: Likewise.
* src/filesystem/cow-path.cc: Likewise.
* src/filesystem/cow-std-dir.cc: New file.
* src/filesystem/cow-std-ops.cc: New file.
* src/filesystem/cow-std-path.cc: New file.
* src/filesystem/dir-common.h (_Dir_base, get_file_type): New header
for common code.
* src/filesystem/dir.cc (_Dir): Derive from _Dir_base.
(open_dir): Move to _Dir_base constructor.
(get_file_type): Move to dir-common.h.
(recurse): Move to _Dir_base::should_recurse.
* src/filesystem/ops-common.h: New header for common code.
* src/filesystem/ops.cc (is_set, make_file_type, make_file_status)
(is_not_found_errno, file_time, do_copy_file): Move to ops-common.h.
* src/filesystem/path.cc (filesystem_error::_M_gen_what): Define.
* src/filesystem/std-dir.cc: New file, based on Filesystem TS code.
* src/filesystem/std-ops.cc: Likewise.
* src/filesystem/std-dir.cc: Likewise.
* testsuite/27_io/filesystem/iterators/directory_iterator.cc: New
test.
* testsuite/27_io/filesystem/iterators/pop.cc: New test.
* testsuite/27_io/filesystem/iterators/recursive_directory_iterator.cc:
New test.
* testsuite/27_io/filesystem/operations/absolute.cc: New test.
* testsuite/27_io/filesystem/operations/canonical.cc: New test.
* testsuite/27_io/filesystem/operations/copy.cc: New test.
* testsuite/27_io/filesystem/operations/copy_file.cc: New test.
* testsuite/27_io/filesystem/operations/create_directories.cc: New
test.
* testsuite/27_io/filesystem/operations/create_directory.cc: New test.
* testsuite/27_io/filesystem/operations/create_symlink.cc: New test.
* testsuite/27_io/filesystem/operations/current_path.cc: New test.
* testsuite/27_io/filesystem/operations/equivalent.cc: New test.
* testsuite/27_io/filesystem/operations/exists.cc: New test.
* testsuite/27_io/filesystem/operations/file_size.cc: New test.
* testsuite/27_io/filesystem/operations/is_empty.cc: New test.
* testsuite/27_io/filesystem/operations/last_write_time.cc: New test.
* testsuite/27_io/filesystem/operations/permissions.cc: New test.
* testsuite/27_io/filesystem/operations/proximate.cc: New test.
* testsuite/27_io/filesystem/operations/read_symlink.cc: New test.
* testsuite/27_io/filesystem/operations/relative.cc: New test.
* testsuite/27_io/filesystem/operations/remove_all.cc: New test.
* testsuite/27_io/filesystem/operations/space.cc: New test.
* testsuite/27_io/filesystem/operations/status.cc: New test.
* testsuite/27_io/filesystem/operations/symlink_status.cc: New test.
* testsuite/27_io/filesystem/operations/temp_directory_path.cc: New
test.
* testsuite/27_io/filesystem/operations/weakly_canonical.cc: New test.
* testsuite/27_io/filesystem/path/append/path.cc: New test.
* testsuite/27_io/filesystem/path/assign/assign.cc: New test.
* testsuite/27_io/filesystem/path/assign/copy.cc: New test.
* testsuite/27_io/filesystem/path/compare/compare.cc: New test.
* testsuite/27_io/filesystem/path/compare/path.cc: New test.
* testsuite/27_io/filesystem/path/compare/strings.cc: New test.
* testsuite/27_io/filesystem/path/concat/path.cc: New test.
* testsuite/27_io/filesystem/path/concat/strings.cc: New test.
* testsuite/27_io/filesystem/path/construct/copy.cc: New test.
* testsuite/27_io/filesystem/path/construct/default.cc: New test.
* testsuite/27_io/filesystem/path/construct/locale.cc: New test.
* testsuite/27_io/filesystem/path/construct/range.cc: New test.
* testsuite/27_io/filesystem/path/construct/string_view.cc: New test.
* testsuite/27_io/filesystem/path/decompose/extension.cc: New test.
* testsuite/27_io/filesystem/path/decompose/filename.cc: New test.
* testsuite/27_io/filesystem/path/decompose/parent_path.cc: New test.
* testsuite/27_io/filesystem/path/decompose/relative_path.cc: New
test.
* testsuite/27_io/filesystem/path/decompose/root_directory.cc: New
test.
* testsuite/27_io/filesystem/path/decompose/root_name.cc: New test.
* testsuite/27_io/filesystem/path/decompose/root_path.cc: New test.
* testsuite/27_io/filesystem/path/decompose/stem.cc: New test.
* testsuite/27_io/filesystem/path/generation/normal.cc: New test.
* testsuite/27_io/filesystem/path/generation/proximate.cc: New test.
* testsuite/27_io/filesystem/path/generation/relative.cc: New test.
* testsuite/27_io/filesystem/path/generic/generic_string.cc: New test.
* testsuite/27_io/filesystem/path/itr/traversal.cc: New test.
* testsuite/27_io/filesystem/path/modifiers/clear.cc: New test.
* testsuite/27_io/filesystem/path/modifiers/make_preferred.cc: New
test.
* testsuite/27_io/filesystem/path/modifiers/remove_filename.cc: New
test.
* testsuite/27_io/filesystem/path/modifiers/replace_extension.cc: New
test.
* testsuite/27_io/filesystem/path/modifiers/replace_filename.cc: New
test.
* testsuite/27_io/filesystem/path/modifiers/swap.cc: New test.
* testsuite/27_io/filesystem/path/native/string.cc: New test.
* testsuite/27_io/filesystem/path/nonmember/hash_value.cc: New test.
* testsuite/27_io/filesystem/path/query/empty.cc: New test.
* testsuite/27_io/filesystem/path/query/has_extension.cc: New test.
* testsuite/27_io/filesystem/path/query/has_filename.cc: New test.
* testsuite/27_io/filesystem/path/query/has_parent_path.cc: New test.
* testsuite/27_io/filesystem/path/query/has_relative_path.cc: New
test.
* testsuite/27_io/filesystem/path/query/has_root_directory.cc: New
test.
* testsuite/27_io/filesystem/path/query/has_root_name.cc: New test.
* testsuite/27_io/filesystem/path/query/has_root_path.cc: New test.
* testsuite/27_io/filesystem/path/query/has_stem.cc: New test.
* testsuite/27_io/filesystem/path/query/is_relative.cc: New test.
* testsuite/experimental/filesystem/path/construct/string_view.cc:
Define USE_FILESYSTEM_TS.
* testsuite/util/testsuite_fs.h: Allow use with C++17 paths as well
as Filesystem TS.
* tree-ssa-pre.c (bitmap_remove_from_set): Rename to...
(bitmap_remove_expr_from_set): ... this. All callers call this
for non-constant values.
(bitmap_set_subtract): Rename to...
(bitmap_set_subtract_expressions): ... this. Adjust and
optimize.
(bitmap_set_contains_value): Remove superfluous check.
(bitmap_set_replace_value): Inline into single caller ...
(bitmap_value_replace_in_set): ... here and simplify.
(dependent_clean): Merge into ...
(clean): ... this using an overload. Adjust.
(prune_clobbered_mems): Adjust.
(compute_antic_aux): Likewise.
(compute_partial_antic_aux): Likewise.
Jonathan Wakely [Mon, 23 Oct 2017 11:10:28 +0000 (12:10 +0100)]
PR libstdc++/82644 document IS 29124 support
Also fix declarations of special functions in C++17, to import them into
the global namespace in <math.h>, and to prevent defining the
non-standard hypergeometric functions in strict mode.
PR libstdc++/82644
* doc/xml/manual/intro.xml: Include new section.
* doc/xml/manual/status_cxxis29124.xml: New section on IS 29124
status.
* include/bits/specfun.h [__STRICT_ANSI__] (hyperg, hypergf, hypergl)
(conf_hyperg, conf_hypergf, conf_hypergl): Don't declare.
* include/c_compatibility/math.h: Import special functions into
global namespace for C++17.
* testsuite/26_numerics/headers/cmath/82644.cc: New test.
* testsuite/26_numerics/headers/cmath/functions_global_c++17.cc: New
test.
I took the documentation of the FRAME_GROWS_DOWNWARD behaviour from the
version that was in most header files, since the one in the manual seemed
less clear.
The patch deliberately keeps FIRST_PARM_OFFSET(FNDECL) in
microblaze_starting_frame_offset; this seems to be a port-local
convention and takes advantage of the fact that FIRST_PARM_OFFSET
doesn't read FNDECL.
2017-10-23 Richard Sandiford <richard.sandiford@linaro.org>
gcc/
* target.def (starting_frame_offset): New hook.
* doc/tm.texi (STARTING_FRAME_OFFSET): Remove in favor of...
(TARGET_STARTING_FRAME_OFFSET): ...this new hook.
* doc/tm.texi.in: Regenerate.
* hooks.h (hook_hwi_void_0): Declare.
* hooks.c (hook_hwi_void_0): New function.
* doc/rtl.texi: Refer to TARGET_STARTING_FRAME_OFFSET instead of
STARTING_FRAME_OFFSET.
* builtins.c (expand_builtin_setjmp_receiver): Likewise.
* reload1.c (reload): Likewise.
* cfgexpand.c (expand_used_vars): Use targetm.starting_frame_offset
instead of STARTING_FRAME_OFFSET.
* function.c (try_fit_stack_local): Likewise.
(assign_stack_local_1): Likewise
(instantiate_virtual_regs): Likewise.
* rtlanal.c (rtx_addr_can_trap_p_1): Likewise.
* config/avr/avr.md (nonlocal_goto_receiver): Likewise.
* config/aarch64/aarch64.h (STARTING_FRAME_OFFSET): Delete.
* config/alpha/alpha.h (STARTING_FRAME_OFFSET): Likewise.
* config/arc/arc.h (STARTING_FRAME_OFFSET): Likewise.
* config/arm/arm.h (STARTING_FRAME_OFFSET): Likewise.
* config/bfin/bfin.h (STARTING_FRAME_OFFSET): Likewise.
* config/c6x/c6x.h (STARTING_FRAME_OFFSET): Likewise.
* config/cr16/cr16.h (STARTING_FRAME_OFFSET): Likewise.
* config/cris/cris.h (STARTING_FRAME_OFFSET): Likewise.
* config/fr30/fr30.h (STARTING_FRAME_OFFSET): Likewise.
* config/frv/frv.h (STARTING_FRAME_OFFSET): Likewise.
* config/ft32/ft32.h (STARTING_FRAME_OFFSET): Likewise.
* config/h8300/h8300.h (STARTING_FRAME_OFFSET): Likewise.
* config/i386/i386.h (STARTING_FRAME_OFFSET): Likewise.
* config/ia64/ia64.h (STARTING_FRAME_OFFSET): Likewise.
* config/m32c/m32c.h (STARTING_FRAME_OFFSET): Likewise.
* config/m68k/m68k.h (STARTING_FRAME_OFFSET): Likewise.
* config/mcore/mcore.h (STARTING_FRAME_OFFSET): Likewise.
* config/mn10300/mn10300.h (STARTING_FRAME_OFFSET): Likewise.
* config/moxie/moxie.h (STARTING_FRAME_OFFSET): Likewise.
* config/msp430/msp430.h (STARTING_FRAME_OFFSET): Likewise.
* config/nds32/nds32.h (STARTING_FRAME_OFFSET): Likewise.
* config/nios2/nios2.h (STARTING_FRAME_OFFSET): Likewise.
* config/nvptx/nvptx.h (STARTING_FRAME_OFFSET): Likewise.
* config/pdp11/pdp11.h (STARTING_FRAME_OFFSET): Likewise.
* config/riscv/riscv.h (STARTING_FRAME_OFFSET): Likewise.
* config/rl78/rl78.h (STARTING_FRAME_OFFSET): Likewise.
* config/rx/rx.h (STARTING_FRAME_OFFSET): Likewise.
* config/s390/s390.h (STARTING_FRAME_OFFSET): Likewise.
* config/sh/sh.h (STARTING_FRAME_OFFSET): Likewise.
* config/sparc/sparc.c (sparc_compute_frame_size): Likewise.
* config/sparc/sparc.h (STARTING_FRAME_OFFSET): Likewise.
* config/spu/spu.h (STARTING_FRAME_OFFSET): Likewise.
* config/stormy16/stormy16.h (STARTING_FRAME_OFFSET): Likewise.
* config/tilegx/tilegx.h (STARTING_FRAME_OFFSET): Likewise.
* config/tilepro/tilepro.h (STARTING_FRAME_OFFSET): Likewise.
* config/v850/v850.h (STARTING_FRAME_OFFSET): Likewise.
* config/visium/visium.h (STARTING_FRAME_OFFSET): Likewise.
* config/avr/avr.h (STARTING_FRAME_OFFSET): Likewise.
* config/avr/avr-protos.h (avr_starting_frame_offset): Likewise.
* config/avr/avr.c (avr_starting_frame_offset): Make static and
return a HOST_WIDE_INT.
(avr_builtin_setjmp_frame_value): Use it instead of
STARTING_FRAME_OFFSET.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/epiphany/epiphany.h (STARTING_FRAME_OFFSET): Delete.
* config/epiphany/epiphany.c (epiphany_starting_frame_offset):
New function.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/iq2000/iq2000.h (STARTING_FRAME_OFFSET): Delete.
* config/iq2000/iq2000.c (iq2000_starting_frame_offset): New function.
(TARGET_CONSTANT_ALIGNMENT): Redefine.
* config/lm32/lm32.h (STARTING_FRAME_OFFSET): Delete.
* config/lm32/lm32.c (lm32_starting_frame_offset): New function.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/m32r/m32r.h (STARTING_FRAME_OFFSET): Delete.
* config/m32r/m32r.c (m32r_starting_frame_offset): New function.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/microblaze/microblaze.h (STARTING_FRAME_OFFSET): Delete.
* config/microblaze/microblaze.c (microblaze_starting_frame_offset):
New function.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/mips/mips.h (STARTING_FRAME_OFFSET): Delete.
* config/mips/mips.c (mips_compute_frame_info): Refer to
TARGET_STARTING_FRAME_OFFSET instead of STARTING_FRAME_OFFSET.
(mips_starting_frame_offset): New function.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/mmix/mmix.h (STARTING_FRAME_OFFSET): Delete.
* config/mmix/mmix-protos.h (mmix_starting_frame_offset): Delete.
* config/mmix/mmix.c (mmix_starting_frame_offset): Make static
and return a HOST_WIDE_INT.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
(mmix_initial_elimination_offset): Refer to
TARGET_STARTING_FRAME_OFFSET instead of STARTING_FRAME_OFFSET.
* config/pa/pa.h (STARTING_FRAME_OFFSET): Delete.
* config/pa/pa.c (pa_starting_frame_offset): New function.
(pa_compute_frame_size): Use it instead of STARTING_FRAME_OFFSET.
(pa_expand_prologue): Likewise.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/powerpcspe/aix.h (STARTING_FRAME_OFFSET): Split out
!FRAME_GROWS_DOWNWARD handling to...
(RS6000_STARTING_FRAME_OFFSET): ...this new macro.
* config/powerpcspe/darwin.h (STARTING_FRAME_OFFSET): Split out
!FRAME_GROWS_DOWNWARD handling to...
(RS6000_STARTING_FRAME_OFFSET): ...this new macro.
* config/powerpcspe/powerpcspe.h (STARTING_FRAME_OFFSET): Split out
!FRAME_GROWS_DOWNWARD handling to...
(RS6000_STARTING_FRAME_OFFSET): ...this new macro.
* config/powerpcspe/powerpcspe.c (TARGET_STARTING_FRAME_OFFSET):
Redefine.
(rs6000_starting_frame_offset): New function.
* config/rs6000/aix.h (STARTING_FRAME_OFFSET): Split out
!FRAME_GROWS_DOWNWARD handling to...
(RS6000_STARTING_FRAME_OFFSET): ...this new macro.
* config/rs6000/darwin.h (STARTING_FRAME_OFFSET): Split out
!FRAME_GROWS_DOWNWARD handling to...
(RS6000_STARTING_FRAME_OFFSET): ...this new macro.
* config/rs6000/rs6000.h (STARTING_FRAME_OFFSET): Split out
!FRAME_GROWS_DOWNWARD handling to...
(RS6000_STARTING_FRAME_OFFSET): ...this new macro.
* config/rs6000/rs6000.c (TARGET_STARTING_FRAME_OFFSET): Refine.
(rs6000_starting_frame_offset): New function.
* config/vax/elf.h (STARTING_FRAME_OFFSET): Delete.
* config/vax/vax.h (STARTING_FRAME_OFFSET): Delete.
* config/vax/vax.c (vax_starting_frame_offset): New function.
(vax_expand_prologue): Use it instead of STARTING_FRAME_OFFSET.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* config/xtensa/xtensa.h (STARTING_FRAME_OFFSET): Delete.
* config/xtensa/xtensa.c (xtensa_starting_frame_offset): New function.
(TARGET_STARTING_FRAME_OFFSET): Redefine.
* system.h (STARTING_FRAME_OFFSET): Poison.
Richard Biener [Mon, 23 Oct 2017 09:20:14 +0000 (09:20 +0000)]
re PR tree-optimization/82129 (ICE in compute_antic, at tree-ssa-pre.c:2447)
2017-10-23 Richard Biener <rguenther@suse.de>
PR tree-optimization/82129
* tree-ssa-pre.c (bitmap_set_and): Remove.
(compute_antic_aux): Compute ANTIC_OUT intersection in a way
canonicalizing expressions in the set to those with lowest
ID rather than taking that from the first edge.
But subreg_lowpart_offset returns an unsigned int and
adjust_address_nv takes a HWI, so a subreg offset of 4 would
give a memory offset of 0x00000000fffffffffc.
2017-10-23 Richard Sandiford <richard.sandiford@linaro.org>
gcc/
* combine.c (rtx_equal_for_field_assignment_p): Use
byte_lowpart_offset.
Jan Hubicka [Sun, 22 Oct 2017 22:09:47 +0000 (22:09 +0000)]
i386.c (ix86_builtin_vectorization_cost): Use existing rtx_cost latencies instead of having separate table...
* i386.c (ix86_builtin_vectorization_cost): Use existing rtx_cost
latencies instead of having separate table; make difference between
integer and float costs.
* i386.h (processor_costs): Remove scalar_stmt_cost,
scalar_load_cost, scalar_store_cost, vec_stmt_cost, vec_to_scalar_cost,
scalar_to_vec_cost, vec_align_load_cost, vec_unalign_load_cost,
vec_store_cost.
* x86-tune-costs.h: Remove entries which has been removed in
procesor_costs from all tables; make cond_taken_branch_cost
and cond_not_taken_branch_cost COST_N_INSNS based.
SUBREG_PROMOTED_VAR_P handling in expand_direct_optab_fn
This is needed by the later SVE LAST reductions, where an 8-bit
or 16-bit result is zero- rather than sign-extended to 32 bits.
I think it could occur in other situations too.
2017-09-19 Richard Sandiford <richard.sandiford@linaro.org>
Alan Hayward <alan.hayward@arm.com>
David Sherwood <david.sherwood@arm.com>
gcc/
* internal-fn.c (expand_direct_optab_fn): Don't assign directly
to a SUBREG_PROMOTED_VAR.
Co-Authored-By: Alan Hayward <alan.hayward@arm.com> Co-Authored-By: David Sherwood <david.sherwood@arm.com>
From-SVN: r253992
This patch is like the earlier GET_MODE_UNIT_SIZE one,
but for precisions rather than sizes. There is one behavioural
change in expand_debug_expr: we shouldn't use lowpart subregs
for non-scalar truncations, since that would just reinterpret
some of the scalars and drop the rest. (This probably doesn't
trigger in practice.) Using TRUNCATE is fine for scalars,
since simplify_gen_unary knows when a subreg can be used.
2017-10-22 Richard Sandiford <richard.sandiford@linaro.org>
Alan Hayward <alan.hayward@arm.com>
David Sherwood <david.sherwood@arm.com>
This patch uses HWI_COMPUTABLE_MODE_P (X) instead of
GET_MODE_PRECISION (X) <= HOST_BITS_PER_WIDE_INT in cases
where X also needs to be a scalar integer.
2017-10-22 Richard Sandiford <richard.sandiford@linaro.org>
Alan Hayward <alan.hayward@arm.com>
David Sherwood <david.sherwood@arm.com>
This patch uses subreg_size_lowpart_offset in places that open-coded
the calculation. The reload use (and the LRA one that was based on it)
seemed to ignore the BYTES_BIG_ENDIAN != WORDS_BIG_ENDIAN case; it's not
obvious whether that was deliberate or an oversight.
2017-10-22 Richard Sandiford <richard.sandiford@linaro.org>
Alan Hayward <alan.hayward@arm.com>
David Sherwood <david.sherwood@arm.com>
Add wide_int version of inchash::hash::add_wide_int
This patch adds an inchash hasher for wide_int-based types.
It means that hash_tree no longer hashes TREE_INT_CST_EXT_NUNITS,
but that was redundant with hashing the type.
2017-10-22 Richard Sandiford <richard.sandiford@linaro.org>
gcc/
* inchash.h (inchash::hash::add_wide_int): New function.
* lto-streamer-out.c (hash_tree): Use it.
Igor Tsimbalist [Sat, 21 Oct 2017 21:16:32 +0000 (23:16 +0200)]
Add x86 CET documentation.
gcc/doc/
* extend.texi: Add x86 specific to 'nocf_check' attribute.
List CET intrinsics.
* invoke.texi: Add -mcet, -mibt, -mshstk options. Add x86
specific to -fcf-protection option.
projects / thirdparty / gcc.git / log
