Firstly, it adds back the split conditions that I accidentally removed.
Without it the dot insns are never generated, or rather, always split
back to a separate compare instruction.
Secondly, the shift amount should be SI always, not GPR, or GCC will
insert a zero-extend at expand time that it cannot get rid of later.
The test tests whether dot-form instructions are generated for both
"dot" and "dot2" cases, that is, with just a CC output or also a GPR
output; for all four basic shifts, with a register amount or an
immediate amount. It also tests for superfluous zero-extends. This
also tests if combine "simplifies" the rotates to right-rotates, which
it shouldn't do anymore.
From-SVN: r212267
* arm.md (arch): Add armv6_or_vfpv3.
(arch_enabled): Add test for the above.
* vfp.md (divsf_vfp, divdf_vfp): Add earlyclobber when code can run
on VFP9.
(sqrtsf_vfp, sqrtdf_vfp): Likewise.
From-SVN: r212265
* gcov-io.c (gcov_read_words): Don't call memmove if excess is 0.
* data-streamer-in.c (streamer_read_hwi): Shift UHWI 1 instead of
HWI 1 and negate the unsigned value.
* expmed.c (expand_sdiv_pow2): For modes wider than word always
use AND instead of shift.
* wide-int-print.cc (print_decs): Negate UHWI instead of HWI.
c-family/
* c-ada-spec.c (dump_ada_nodes): Don't call qsort if
comments->count <= 1, as comments->entries might be NULL.
From-SVN: r212264
* config/rs6000/rs6000.c (rs6000_adjust_atomic_subword): Use
BYTES_BIG_ENDIAN rather than WORDS_BIG_ENDIAN to check for byte
endianness.
From-SVN: r212258
Many targets do not have both rotate and rotatert. Of the 47 targets
in the tree, 17 have both, 9 have only rotate, 2 have only rotatert, and
19 have neither (this is based on "grep -wil" so it can be slightly off).
rs6000 has only rotate, and mips has only rotatert. For such targets
simplifying rotate to rotatert and vice versa is not simplifying things
at all. rs6000 has already way too many rotate patterns (some days it
seems like two thousand, but it is somewhat less in reality still); I
would prefer not to double that again.
So, this patch makes genrecog define HAVE_rotate and HAVE_rotatert if
those RTL codes are mentioned anywhere in the machine description, and
then does the transformation in simplify-rtx.c only if both these flags
are set.
From-SVN: r212239
PR go/61620
runtime: Don't free tiny blocks in map deletion.
The memory allocator now has a special case for tiny blocks
(smaller than 16 bytes) and they can not be explicitly freed.
From-SVN: r212233
* decl.c (variable_decl): Reject old style initialization
for derived type components.
* gfortran.dg/oldstyle_5.f: New test.
Co-Authored-By: Fritz Reese <Reese-Fritz@zai.com>
From-SVN: r212227
/cp
2014-07-02 Paolo Carlini <paolo.carlini@oracle.com>
* pt.c (convert_template_argument): Use inform instead of error in
three places.
/testsuite
2014-07-02 Paolo Carlini <paolo.carlini@oracle.com>
* g++.dg/cpp0x/variadic-ex10.C: Adjust for inform instead of error.
* g++.dg/cpp0x/variadic-ex14.C: Likewise.
* g++.dg/parse/error11.C: Likewise.
* g++.old-deja/g++.brendan/template17.C: Likewise.
From-SVN: r212223
* tree.c (decls_same_for_odr, decls_same_for_odr,
types_same_for_odr): Remove.
(type_in_anonymous_namespace_p): Constify argument.
* tree.h (types_same_for_odr, type_in_anonymous_namespace_p):
Constify.
* ipa-devirt.c (odr_type_d): Add ODR_VIOLATED field.
(main_odr_variant): New function.
(hash_type_name): Make static; update assert; do not ICE on
non-records.
(types_same_for_odr): Bring here from tree.c; simplify
and remove old structural comparing code that doesn't work
for templates.
(odr_hasher::equal): Update assert.
(add_type_duplicate): Return true when bases should be computed;
replace incomplete loader by complete; do not output duplicated
warnings; do not ICE on non-records; set odr_violated flag.
(get_odr_type): Be ready to replace incomplete type by complete
one; work on ODR variants instead of main variants; reorder item
in array so bases have still smaller indexes.
(dump_type_inheritance_graph): Be ready for holdes in odr_types
array.
(possible_polymorphic_call_targets): Do not ICE when BINFO is NULL.
From-SVN: r212218
This introduces the "bench" build target, which can be used to run
all benchmarks.
It is also possible to run subsets of benchmarks with the
"package/check" build targets by setting GOBENCH to a matching regex.
From-SVN: r212212
When compiling a small-enough compilation unit that has no address table
entries, but complex enough that -freorder-blocks-and-partition produces
location lists, dwarf2out_finish does not call index_location_lists, but
optimize_location_lists will later assume that the addr_index_table has
been indexed.
Google ref: b/15417905
When resolve_addr_in_expr replaces a CONST_STRING rtx, it directly
updates the pointer to the old expression with the new one. In the
case of a DW_OP_GNU_addr_index or DW_OP_GNU_const_index, that pointer
may be in an address table entry, which is keyed by the rtx. Instead
of directly replacing the pointer, we need to remove the old address
table entry (i.e., decrement its reference count), and add a new one.
Google ref: b/15957101
gcc/
* dwarf2out.c (remove_addr_table_entry): Remove unnecessary hash table
lookup.
(resolve_addr_in_expr): When replacing the rtx in a location list
entry, get a new address table entry.
(dwarf2out_finish): Call index_location_lists even if there are no
addr_index_table entries yet.
From-SVN: r212211
* gcc.target/aarch64/aapcs64/aapcs64.exp:
(additional_flags_for_func_ret): New variable based on $additional_flags
with -fno-use-caller-save.
(func-ret-*.c): Use the new variable.
Co-Authored-By: Yufeng Zhang <yufeng.zhang@arm.com>
From-SVN: r212206
When a system macro is expanded in a non-system file during
out-of-line preprocessing, it can happen that the preprocessor forgets
to emit line markers to express the system-ness status of tokens that
come after the expansion of the macro.
That can lead to situations where the entire non-system file can be
considered as being a system file and thus have its warnings be
discarded during the compilation of the resulting preprocessed file.
My understanding is that this is due to the preprocessor not
systematically detecting (and reporting) the change in system-ness of
tokens.
And this is what this patch does. Each time the system-ness of a
given token is different from the previous token that was emitted by
the preprocessor, it emits a line marker for the sole purpose of
marking the new system-ness of the subsequent tokens to come.
Bootstrapped and tested on x86_64-unknown-linux-gnu against trunk.
gcc/c-family/ChangeLog:
* c-ppoutput.c (struct print::prev_was_system_token): New data
member.
(init_pp_output): Initialize it.
(maybe_print_line_1, maybe_print_line, print_line_1, print_line)
(do_line_change): Return a flag saying if a line marker was
emitted or not.
(scan_translation_unit): Detect if the system-ness of the token we
are about to emit is different from the one of the previously
emitted token. If so, emit a line marker. Avoid emitting
useless adjacent line markers.
(scan_translation_unit_directives_only): Adjust.
gcc/testsuite/ChangeLog:
* gcc.dg/cpp/syshdr{4,5}.{c,h}: New test files.
Signed-off-by: Dodji Seketeli <dodji@redhat.com>
From-SVN: r212194
* cgraph.h (iterate_direct_aliases): New function.
(FOR_EACH_ALIAS): New macro iterates all direct aliases for a node.
* cgraph.c (cgraph_for_node_thunks_and_aliases): Usage of
FOR_EACH_ALIAS added.
(cgraph_for_node_and_aliases): Likewise.
* cgraphunit.c (assemble_thunks_and_aliases): Likewise.
* ipa-inline.c (reset_edge_caches): Likewise.
(update_caller_keys): Likewise.
* trans-mem.c (ipa_tm_execute): Likewise.
*varpool.c (varpool_analyze_node): Likewise.
(varpool_for_node_and_aliases): Likewise.
* ipa-ref.h (first_alias): New function.
(last_alias): Likewise.
(has_aliases_p): Likewise.
* ipa-ref.c (ipa_ref::remove_reference): Removal function
is sensitive to IPA_REF_ALIASes.
* symtab.c (symtab_node::add_reference): Node of IPA_REF_ALIAS type
are put at the beginning of the list.
(symtab_node::iterate_direct_aliases): New function.
* lto-partition.c (add_symbol_to_partition_1): Usage of
FOR_EACH_ALIAS added.
From-SVN: r212191
2014-06-30 David Holsgrove <david.holsgrove@xilinx.com>
* gcc/testsuite/lib/scanasm.exp (dg-function-on-line): Add
MicroBlaze specific pattern.
--This line, and those below,
M ChangeLog
M lib/scanasm.exp
From-SVN: r212189
PR libstdc++/61061
PR libstdc++/61582
* include/bits/regex_automaton.h (_NFA<>::_M_insert_state): Add
a NFA state limit. If it's exceeded, regex_constants::error_space
will be throwed.
* include/bits/regex_automaton.tcc (_StateSeq<>::_M_clone): Use
map (which is sparse) instead of vector. This reduce n times clones'
cost from O(n^2) to O(n).
* include/std/regex: Add map dependency.
* testsuite/28_regex/algorithms/regex_match/ecma/char/61601.cc: New
testcase.
From-SVN: r212185
PR libstdc++/61424
* include/bits/regex.tcc (__regex_algo_impl<>): Use DFS for ECMAScript,
not just regex containing back-references.
* include/bits/regex_compiler.tcc (_Compiler<>::_M_disjunction):
exchange _M_next and _M_alt for alternative operator,
making matching from left to right.
* include/bits/regex_executor.h (_State_info<>::_M_get_sol_pos):
Add position tracking fom DFS.
* include/bits/regex_executor.tcc (_Executor<>::_M_main_dispatch,
_Executor<>::_M_dfs): Likewise.
* include/bits/regex_scanner.h: Remove unused enum entry.
* testsuite/28_regex/algorithms/regex_search/61424.cc: New
testcase from PR.
From-SVN: r212184
Revert:
* tree-streamer-out.c (pack_ts_type_common_value_fields): Stream if type
is complete.
(write_ts_type_common_tree_pointers): Do not stream fields not set for incomplete
types; do not stream duplicated fields for variants; sanity check that variant
and type match.
(write_ts_type_non_common_tree_pointers): Likewise.
* tree-streamer-in.c (unpack_ts_type_common_value_fields): Mark in TYPE_SIZE whether
type is complete.
(lto_input_ts_type_common_tree_pointers): Do same changes as in
write_ts_type_common_tree_pointers
(lto_input_ts_type_non_common_tree_pointers): Likewise.
* lto.c (lto_copy_fields_not_streamed): New function.
(compare_tree_sccs_1): Do not compare fields shared in between type
and variant.
(lto_read_decls): Fixup types first before inserting into hash.
From-SVN: r212179
Jakub Jelinek
Segher Boessenkool
Thomas Schwinge
Richard Earnshaw
Francois-Xavier Coudert
Marek Polacek
Maciej W. Rozycki
Tobias Burnus
Zhenqiang Chen
GCC Administrator
Uros Bizjak
Jan Hubicka
Ian Lance Taylor
Christian Bruel
Christian Bruel
Kyrylo Tkachov
Paolo Carlini
Cary Coutant
James Greenhalgh
Trevor Saunders
Dodji Seketeli
Teresa Johnson
Dodji Seketeli
Martin Liska
David Holsgrove
Tim Shen
Edward Smith-Rowland
Tim Shen
Joseph Myers