IRA and LRA prefer to use CR7 (which is first in REG_ALLOC_ORDER) over
CR0, although the latter often is cheaper ("x" vs. "y" constraints).
We should figure out why this is and fix it; but until that is done,
this patch makes CR0 the first allocated register: it improves the
current code, and it is required for later patches to be effective.
(It changes two testcases to no longer look at what CR field is
allocated).
* config/rs6000/rs6000.h (REG_ALLOC_ORDER): Move 68 (that is, CR0) to
be the first CR field allocated.
gcc/testsuite/
* gcc.target/powerpc/safe-indirect-jump-2.c: Do not check assigned CR
field number.
* gcc.target/powerpc/safe-indirect-jump-3.c: Ditto.
From-SVN: r265318
2018-10-19 Richard Biener <rguenther@suse.de>
PR target/87657
* config/i386/i386.c (ix86_builtin_vectorization_cost): Use
TYPE_VECTOR_SUBPARTS and avoid relying on vector mode.
* gcc.target/i386/pr87657.c: New testcase.
From-SVN: r265316
* include/bits/regex_executor.tcc (_Backref_matcher::_M_apply): Use
_GLIBCXX_STD_A to refer to normal mode algorithms.
* testsuite/28_regex/headers/regex/parallel_mode.cc: New test.
* testsuite/28_regex/headers/regex/std_c++0x_neg.cc: Remove empty
whitespace.
From-SVN: r265314
This fixes the following testsuite failures on ia32 when compiled with
-D_GLIBCXX_DEBUG:
FAIL: 23_containers/map/modifiers/erase/dr130-linkage-check.cc
FAIL: 23_containers/multimap/modifiers/erase/dr130-linkage-check.cc
FAIL: 23_containers/multiset/modifiers/erase/dr130-linkage-check.cc
FAIL: 23_containers/set/modifiers/erase/dr130-linkage-check.cc
The normal mode containers already use the abi-tag to mangle these
overloads differently, but the debug mode versions weren't fixed.
* include/debug/map.h (map::erase(iterator)): Add abi-tag so that
C++11 version mangles differently from incompatible C++98 version.
* include/debug/multimap.h (multimap::erase(iterator)): Likewise.
* include/debug/multiset.h (multiset::erase(iterator))
(multiset::erase(const_iterator, const_iterator)): Likewise.
* include/debug/set.h (set::erase(iterator))
(multiset::erase(const_iterator, const_iterator)): Likewise.
From-SVN: r265313
Use "register_operand" in AVX512 FMA with memory broadcast when only
registers are allowed.
* config/i386/sse.md
(*<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name>_bcst_1):
Replace nonimmediate_operand with register_operand.
(*<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name>_bcst_2):
Likewise.
(*<sd_mask_codefor>fma_fmadd_<mode><sd_maskz_name>_bcst_3):
Likewise.
From-SVN: r265310
FROM..TO range might contain NOTE_INSN_DELETED insns, for which the
corresponding entries in lra_insn_recog_data[] are NULLs. Example from
the problematic code from PR87596:
(note 148 154 68 7 NOTE_INSN_DELETED)
lra_insn_recog_data[] is used directly only when the insn in question
is taken from insn_bitmap, which is not the case here. In other
situations lra_get_insn_recog_data () guarded by INSN_P () or other
stricter predicate are used. So we need to do this here as well.
A tiny detail worth noting: I put the INSN_P () check before the
insn_bitmap check, because I believe that insn_bitmap can contain only
real insns anyway.
gcc/ChangeLog:
2018-10-19 Ilya Leoshkevich <iii@linux.ibm.com>
PR rtl-optimization/87596
* lra-constraints.c (spill_hard_reg_in_range): Use INSN_P () +
lra_get_insn_recog_data () instead of lra_insn_recog_data[]
for instructions in FROM..TO range.
gcc/testsuite/ChangeLog:
2018-10-19 Ilya Leoshkevich <iii@linux.ibm.com>
PR rtl-optimization/87596
* gcc.target/i386/pr87596.c: New test.
From-SVN: r265306
* cfgexpand.c (expand_one_var): Use specific wording in error message
for non-local frame variables.
* stor-layout.c (layout_decl): Do not issue a warning for them.
From-SVN: r265305
This patch recomputes the insn priority when a replacement for one of its
dependent insns is applied.
gcc/ChangeLog:
* haifa-sched.c (priority): Add force_recompute parameter.
(apply_replacement): Call priority () with force_recompute = true.
(restore_pattern): Likewise.
From-SVN: r265304
Normally aliases compare as identical to the underlying type. Add a
COMPARE_ALIASES flag to let them compare (and hash) differently. This
will be used by later patches in this series.
Reviewed-on: https://go-review.googlesource.com/c/143021
From-SVN: r265297
Previously when export data referred to a type that was not defined in
a directly imported package, we would write the package name as
additional information in the type's export data. That approach
required all type information to be read in order. This patch changes
the compiler to find all references to indirectly imported packages,
and write them out as an indirectimport line in the import data. This
will permit us to read exported type data out of order.
The type traversal used to find indirect imports is a little more
complicated than necessary in preparation for later patches in this
series.
Reviewed-on: https://go-review.googlesource.com/c/143020
From-SVN: r265296
A single flags parameter replaces the Cmp_tags and errors_are_identical
parameters. The existing behavior is unchanged.
This is a simplification step for future work that will add a new flag.
Reviewed-on: https://go-review.googlesource.com/c/143019
From-SVN: r265293
Since mask of vec_merge is in HOST_WIDE_INT, HOST_BITS_PER_WIDE_INT is
the maximum number of vector elements.
* simplify-rtx.c (simplify_subreg): Limit mask of vec_merge to
HOST_BITS_PER_WIDE_INT.
(test_vector_ops_duplicate): Likewise.
From-SVN: r265290
* testsuite/20_util/duration/cons/2.cc: Add -ffloat-store to fix
failure when compiled without optimisation.
* testsuite/ext/profile/mutex_extensions_neg.cc: Prune additional
errors caused by C++17 std::pmr alias templates.
From-SVN: r265287
If a locale's THOUSANDS_SEP or MON_THOUSANDS_SEP string is not a
single character we either need to narrow it to a single char or
ignore it (and therefore disable digit grouping for that facet).
PR libstdc++/87642
* config/locale/gnu/monetary_members.cc
(moneypunct<char, true>::_M_initialize_moneypunct): Use
__narrow_multibyte_chars to convert multibyte thousands separators
to a single char.
* config/locale/gnu/numeric_members.cc
(numpunct<char>::_M_initialize_numpunct): Likewise.
(__narrow_multibyte_chars): New function.
From-SVN: r265286
The export data, which is approximately readable and looks something
like Go, was first implemented back when Go still used semicolons.
Drop the semicolons, to make it look slightly more Go like and make it
slightly smaller.
This updates the compiler and the gccgoimporter package.
This introduces a new version of the export data. There are going to
be more changes to the export data, so this version is still subject
to change.
Reviewed-on: https://go-review.googlesource.com/c/143018
From-SVN: r265284
* udivmodsi4.c (__udivmodsi4): Rename to conform to coding
standard.
* divmod.c: Update references to __udivmodsi4.
* udivmod.c: Ditto.
* udivhi3.c: New file.
* udivmodhi4.c: New file.
* config/pdp11/t-pdp11 (LIB2ADD): Add the new files.
From-SVN: r265277
* config/i386/i386.c (ix86_emit_fp_unordered_jump):
Set JUMP_LABEL to the jump insn.
(ix86_emit_i387_log1p): Use ix86_expand_branch to expand branch.
Predict emitted jump and add label to jump insn.
From-SVN: r265274
In r264887 I broke the build of 502.gcc_r due to an ICE.
The ICE occurs when generating a location for an sprintf warning within
a string literal, where the sprintf call is in a macro.
The root cause is a bug in the original commit of substring locations
(r239175). get_substring_ranges_for_loc has code to handle the case
where the string literal is in a very long source line that exceeds the
length that the current linemap can represent: the start of the token
is in one line map, but then another line map is started, and the end
of the token is in the new linemap. get_substring_ranges_for_loc handles
this by using the linemap of the end-point when building location_t
values within the string. When extracting the linemap for the endpoint
in r239175 I erroneously used LRK_MACRO_EXPANSION_POINT, which should
have instead been LRK_SPELLING_LOCATION.
I believe this bug was dormant due to rejecting macro locations earlier
in the function, but in r264887 I allowed some macro locations in order
to deal with locations coming from the C++ lexer, and this uncovered
the bug: if a string literal was defined in a macro, locations within
the string literal would be looked up using the linemap of the expansion
point of the macro, rather than of the spelling point. This would lead
to garbage location_t values, and, depending on the precise line numbers
of the two locations, an assertion failure (which was causing the build
failure in 502.gcc_r).
This patch fixes the bug by using LRK_SPELLING_LOCATION, and adds some
bulletproofing to the "two linemaps" case.
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu
(g++.sum gained 5 PASS results; gcc.sum gained 3 PASS results).
I also verified that this fixes the build of 502.gcc_r.
gcc/ChangeLog:
PR tree-optimization/87562
* input.c (get_substring_ranges_for_loc): Use
LRK_SPELLING_LOCATION rather than LRK_MACRO_EXPANSION_POINT when
getting the linemap for the endpoint. Verify that it's either
in the same linemap as the start point's spelling location, or
at least in the same file.
gcc/testsuite/ChangeLog:
PR tree-optimization/87562
* c-c++-common/substring-location-PR-87562-1-a.h: New file.
* c-c++-common/substring-location-PR-87562-1-b.h: New file.
* c-c++-common/substring-location-PR-87562-1.c: New test.
* gcc.dg/plugin/diagnostic-test-string-literals-1.c: Add test for
PR 87562.
* gcc.dg/plugin/pr87562-a.h: New file.
* gcc.dg/plugin/pr87562-b.h: New file.
From-SVN: r265271
Use the value of the first element as the initial value of the
__valarray_sum accumulator. Value-initialization might not create the
additive identity for the value type.
Make a similar change to __valarray_product even though it's only ever
used internally with a value_type of size_t.
PR libstdc++/87641
* include/bits/valarray_array.h (__valarray_sum): Use first element
to initialize accumulator instead of value-initializing it.
(__valarray_product<_Tp>): Move to ...
* src/c++98/valarray.cc (__valarray_product<_Tp>): Here. Use first
element to initialize accumulator.
(__valarray_product(const valarray<size_t>&)): Remove const_cast made
unnecessary by LWG 389.
* testsuite/26_numerics/valarray/87641.cc: New test.
From-SVN: r265270
2018-10-18 Paul Thomas <pault@gcc.gnu.org>
PR fortran/58618
* trans-stmt.c (trans_associate_var): All strings that return
as pointer types can be assigned directly to the associate
name so remove 'attr' and the condition that uses it.
2018-10-18 Paul Thomas <pault@gcc.gnu.org>
PR fortran/58618
* gfortran.dg/associate_45.f90 : New test.
From-SVN: r265264
2018-10-18 Paul Thomas <pault@gcc.gnu.org>
PR fortran/58618
* trans-decl.c (gfc_get_symbol_decl): Deal correctly with the
initialization with NULL() of a deferred length pointer.
2018-10-18 Paul Thomas <pault@gcc.gnu.org>
PR fortran/58618
* gfortran.dg/deferred_character_30.f90 : New test.
From-SVN: r265263
We can simplify
(subreg (vec_merge (vec_duplicate X)
(vector)
(const_int ((1 << N) | M)))
(N * sizeof (X)))
to X when mode of X is the same as of mode of subreg.
gcc/
PR target/87537
* simplify-rtx.c (simplify_subreg): Simplify subreg of vec_merge
of vec_duplicate.
(test_vector_ops_duplicate): Add test for a scalar subreg of a
VEC_MERGE of a VEC_DUPLICATE.
gcc/testsuite/
PR target/87537
* gcc.target/i386/pr87537-1.c: New test.
From-SVN: r265260
Now new features are starting to be added to a C2X draft (in the C2x
branch of the C standard git repository, no public WG14 document yet),
it's time to add -std=c2x and associated options to GCC for use in
enabling C2X features.
This patch adds the expected set of options: -std=c2x, -std=gnu2x,
-Wc11-c2x-compat. A first C2X feature is added (the only one so far
in the repository that's obviously relevant to GCC): support (as in
C++) for the string constant to be omitted in _Static_assert. This
feature is duly also supported as an extension in earlier standard
modes (diagnosed with -pedantic, unless -Wno-c11-c2x-compat is given,
or with -Wc11-c2x-compat even in C2X mode).
Bootstrapped with no regressions on x86_64-pc-linux-gnu.
gcc/
* doc/cpp.texi (__STDC_VERSION__): Document C2X handling.
* doc/invoke.texi (-std=c2x, -std=gnu2x): Document new options.
* doc/standards.texi (C Language): Document C2X.
* dwarf2out.c (highest_c_language), config/rl78/rl78.c
(rl78_option_override): Handle "GNU C2X" language name.
gcc/c/
* c-errors.c (pedwarn_c11): New function.
* c-parser.c (disable_extension_diagnostics): Save
warn_c11_c2x_compat and set it to 0.
(restore_extension_diagnostics): Restore warn_c11_c2x_compat.
(c_parser_static_assert_declaration_no_semi): Handle
_Static_assert without string constant.
* c-tree.h (pedwarn_c11): New prototype.
gcc/c-family/
* c-common.c (flag_isoc2x): New variable.
* c-common.h (clk_c): Update comment to reference C2X.
(flag_isoc99, flag_isoc11): Update comments to reference future
standard versions in general.
(flag_isoc2x): Declare.
* c-opts.c (set_std_c2x): New function.
(c_common_handle_option): Handle -std=c2x and -std=gnu2x.
(set_std_c89, set_std_c99, set_std_c11, set_std_c17): Set
flag_isoc2x to 0.
* c.opt (Wc11-c2x-compat, std=c2x, std=gnu2x): New options.
gcc/testsuite/
* gcc.dg/c11-static-assert-7.c, gcc.dg/c11-static-assert-8.c,
gcc.dg/c11-static-assert-9.c, gcc.dg/c2x-static-assert-1.c,
gcc.dg/c2x-static-assert-2.c, gcc.dg/c99-static-assert-2.c,
gcc.dg/gnu2x-static-assert-1.c: New tests.
* gcc.dg/missing-symbol-3.c: Update expected fix-it text.
libcpp/
* include/cpplib.h (enum c_lang): Add CLK_GNUC2X and CLK_STDC2X.
* init.c (lang_defaults): Add GNUC2X and STDC2X entries.
(cpp_init_builtins): Define __STDC_VERSION__ to 202000L for C2X.
From-SVN: r265251
This patch makes references in the manual to C17 reflect it having
been published in July 2018. (For the reasons it took so long to get
to ballot and publication, see the WG14 convenor's report to the last
SC22 plenary - SC22 N5297 - where it references "A troubling new trend
where ISO CS has begun enforcing undocumented rules that are approved
neither by the TMB nor by the WG 14 editors, leading to a decrease in
the usefulness of standards documents.".)
Tested with "make info html pdf".
* doc/invoke.texi (-std=c17), doc/standards.texi (C Language):
Document C17 as published in 2018.
From-SVN: r265249
One target file (config/c6x/t-elf) lists _printf and _gcc_bcmp in
LIB2FUNCS_EXCLUDE, but that does not have any effect, since those are
not filtered away from LIB2FUNCS_ST. Another option is to do as in
config/rl78/t-rl78, which explicitly sets LIB2FUNCS_ST
# Remove __gcc_bcmp from LIB2FUNCS_ST
LIB2FUNCS_ST = _eprintf
but honouring LIB2FUNCS_EXCLUDE also for LIB2FUNCS_ST seems more
natural.
From-SVN: r265246
PR middle-end/87623
* fold-const.c (fold_truth_andor_1): If the right side is not constant,
bail out if both sides do not have the same storage order.
From-SVN: r265242
* wide-int-range.h (wide_int_range_shift_undefined_p): Adjust to
use sign as argument.
* tree-vrp.c (extract_range_from_binary_expr_1): Pass sign to
wide_int_range_shift_undefined_p.
From-SVN: r265237
