Jakub Jelinek [Fri, 17 Apr 2015 17:09:20 +0000 (19:09 +0200)]
re PR bootstrap/62077 (--with-build-config=bootstrap-lto fails)
PR bootstrap/62077
* configure.ac (--enable-stage1-checking): Default to
release,misc,gimple,rtlflag,tree,types if --disable-checking
or --enable-checking is not specified and DEV-PHASE is not
experimental.
* configure: Regenerated.
Jakub Jelinek [Fri, 17 Apr 2015 16:43:28 +0000 (18:43 +0200)]
re PR target/65689 ([AArch64] S constraint fails for inline asm at -O0)
PR target/65689
* genpreds.c (struct constraint_data): Add maybe_allows_reg and
maybe_allows_mem bitfields.
(maybe_allows_none_start, maybe_allows_none_end,
maybe_allows_reg_start, maybe_allows_reg_end, maybe_allows_mem_start,
maybe_allows_mem_end): New variables.
(compute_maybe_allows): New function.
(add_constraint): Use it to initialize maybe_allows_reg and
maybe_allows_mem fields.
(choose_enum_order): Sort the non-is_register/is_const_int/is_memory/
is_address constraints such that those that allow neither mem nor
reg come first, then those that only allow reg but not mem, then
those that only allow mem but not reg, then the rest.
(write_allows_reg_mem_function): New function.
(write_tm_preds_h): Call it.
* stmt.c (parse_output_constraint, parse_input_constraint): Use
the generated insn_extra_constraint_allows_reg_mem function
instead of always setting *allows_reg = true; *allows_mem = true;
for unknown extra constraints.
* gcc.target/aarch64/c-output-template-4.c: New test.
H.J. Lu [Fri, 17 Apr 2015 16:23:24 +0000 (16:23 +0000)]
Properly handle uninitialized common symbol
Uninitialized common symbol behavior in executables is target and linker
dependent. default_binds_local_p_3 is made public and updated to take an
argument to indicate if the linker can guarantee that an uninitialized
common symbol in the executable will still be defined (through COPY
relocation) in the executable. If common symbol is local to executable,
default_binds_local_p_3 will treat non-external variable as defined
locally. default_binds_local_p_2 is changed to treat common symbol as
local for non-PIE binaries.
For i386, common symbol is local only for non-PIE binaries. For x86-64,
common symbol is local only for non-PIE binaries or linker supports copy
reloc in PIE binaries. If a target treats common symbol as local only
for non-PIE binaries, it can define TARGET_BINDS_LOCAL_P as
default_binds_local_p_2.
gcc/
PR target/65780
* output.h (default_binds_local_p_3): New.
* varasm.c (default_binds_local_p_3): Make it public. Take an
argument to indicate if common symbol may be local. If common
symbol may be local, treat non-external variable as defined
locally.
(default_binds_local_p_2): Pass !flag_pic to default_binds_local_p_3.
(default_binds_local_p_1): Pass false to default_binds_local_p_3.
* config/i386/i386.c (ix86_binds_local_p): New.
(TARGET_BINDS_LOCAL_P): Replace default_binds_local_p_2 with
ix86_binds_local_p.
gcc/testsuite/
PR target/65780
* gcc.dg/pr65780-1.c: New test.
* gcc.dg/pr65780-2.c: Likewise.
* gcc.target/i386/pr32219-9.c: Likewise.
* gcc.target/i386/pr32219-1.c (xxx): Make it initialized common
symbol.
* gcc.target/i386/pr64317.c (c): Initialize.
Martin Liska [Fri, 17 Apr 2015 13:56:15 +0000 (15:56 +0200)]
Fix IPA memory leaks.
* ipa-cp.c (ipcp_driver): Release prev_edge_clone.
* ipa-icf.c (sem_item_optimizer::subdivide_classes_by_sensitive_refs):
Release symbol_compare_collection.
* ipa-reference.c: Add TODO that a vector should be released.
H.J. Lu [Fri, 17 Apr 2015 12:58:07 +0000 (12:58 +0000)]
Hide __cpu_indicator_init/__cpu_model from linker
We shouldn't call external function, __cpu_indicator_init, while an object
is being relocated since its .got.plt section hasn't been updated. It
works for non-PIE since no update on .got.plt section is required. This
patch creates libgcc.so as a linker script, hides __cpu_indicator_init
and __cpu_model in libgcc.so.1 from linker, forces linker to resolve
__cpu_indicator_init and __cpu_model to their hidden definitions in
libgcc.a while providing backward binary compatibility.
Tom de Vries [Fri, 17 Apr 2015 09:27:08 +0000 (09:27 +0000)]
Set PROP_gimple_lva for functions without IFN_VA_ARG
2015-04-17 Tom de Vries <tom@codesourcery.com>
* gimplify.c (gimplify_function_tree): Tentatively set PROP_gimple_lva
in cfun->curr_properties.
(gimplify_va_arg_expr): Clear PROP_gimple_lva in cfun->curr_properties
if we generate an IFN_VA_ARG.
* tree-inline.c (expand_call_inline): Reset PROP_gimple_lva in dest
function if PROP_gimple_lva is not set in src function.
Tom de Vries [Fri, 17 Apr 2015 09:26:33 +0000 (09:26 +0000)]
Add gimple_find_sub_bbs
2015-04-17 Tom de Vries <tom@codesourcery.com>
Michael Matz <matz@suse.de>
* tree-cfg.c (make_blocks_1): Factor out of ...
(make_blocks): ... here.
(make_edges_bb): Factor out of ...
(make_edges): ... here.
(gimple_find_sub_bbs): New function.
* tree-cfg.h (gimple_find_sub_bbs): Declare.
Co-Authored-By: Michael Matz <matz@suse.de>
From-SVN: r222170
* config/i386/predicates.md (register_mixssei387nonimm_operand): New.
* config/i386/i386.md (*fop_<mode>_1_mixed): Merge with
*fop_<mode>_1_sse using enabled attribute. Use
register_mixssei387nonimm_operand operand 1 predicate. Change
alternative 3 constraints from "x" to "v".
Nick Clifton [Thu, 16 Apr 2015 07:57:56 +0000 (07:57 +0000)]
rl78-opts.h (enum rl78_mul_types): Add MUL_G14 and MUL_UNINIT.
* config/rl78/rl78-opts.h (enum rl78_mul_types): Add MUL_G14 and
MUL_UNINIT.
(enum rl78_cpu_type): New.
* config/rl78/rl78-virt.md (attr valloc): Add divhi and divsi.
(umulhi3_shift_virt): Remove m constraint from operand 1.
(umulqihi3_virt): Likewise.
* config/rl78/rl78.c (rl78_option_override): Add code to process
-mcpu and -mmul options.
(rl78_alloc_physical_registers): Add code to handle divhi and
divsi valloc attributes.
(set_origin): Likewise.
* config/rl78/rl78.h (RL78_MUL_G14): Define.
(TARGET_G10, TARGET_G13, TARGET_G14): Define.
(TARGET_CPU_CPP_BUILTINS): Define __RL78_MUL_xxx__ and
__RL78_Gxx__.
(ASM_SPEC): Pass -mcpu on to assembler.
* config/rl78/rl78.md (mulqi3): Add a clobber of AX.
(mulqi3_rl78): Likewise.
(mulhi3_g13): Likewise.
(mulhi3): Generate the G13 or G14 versions of the insn directly.
(mulsi3): Likewise.
(mulhi3_g14): Add clobbers of AX and BC.
(mulsi3_g14): Likewise.
(mulsi3_g13): Likewise.
(udivmodhi4, udivmodhi4_g14, udivmodsi4): New patterns.
(udivmodsi4_g14, udivmodsi4_g13): New patterns.
* config/rl78/rl78.opt (mmul): Initialise value to
RL78_MUL_UNINIT.
(mcpu): New option.
(m13, m14, mrl78): New option aliases.
* config/rl78/t-rl78 (MULTILIB_OPTIONS): Add mg13 and mg14.
(MULTILIB_DIRNAMES): Add g13 and g14.
* doc/invoke.texi: Document -mcpu and -mmul options.
* config/rl78/divmodhi.S: Add G14 and G13 versions of the __divhi3
and __modhi3 functions.
* config/rl78/divmodso.S: Add G14 and G13 versions of the
__divsi3, __udivsi3, __modsi3 and __umodsi3 functions.
Richard Biener [Thu, 16 Apr 2015 07:55:57 +0000 (07:55 +0000)]
tree-ssa-ccp.c (likely_value): See if we have operands that are marked as never simulate again and return...
2015-04-16 Richard Biener <rguenther@suse.de>
* tree-ssa-ccp.c (likely_value): See if we have operands that
are marked as never simulate again and return CONSTANT in this
case.
* tree-ssa-propagate.c (simulate_stmt): Mark stmts that do
not have any operands that will be simulated again as
not being simulated again.
* constexpr.c (cxx_eval_store_expression): Ignore clobbers.
(build_constexpr_constructor_member_initializers): Loop to find
the BIND_EXPR.
* decl.c (start_preparsed_function): Clobber the object at the
beginning of a constructor.
Jason Merrill [Wed, 15 Apr 2015 21:17:21 +0000 (17:17 -0400)]
decl.c (grokmethod): Only set DECL_COMDAT if TREE_PUBLIC is set.
* decl.c (grokmethod): Only set DECL_COMDAT if TREE_PUBLIC is set.
* method.c (implicitly_declare_fn): Likewise.
* decl2.c (vague_linkage_p): Check TREE_PUBLIC first.
Jason Merrill [Wed, 15 Apr 2015 21:17:03 +0000 (17:17 -0400)]
re PR c++/65727 (Segfault With Decltype In Lambda Expression Used To Initialize Static Class Member)
PR c++/65727
* lambda.c (lambda_expr_this_capture): In unevaluated context go
through the normal loop, just don't capture.
(maybe_resolve_dummy): Handle null return.
Co-Authored-By: Marek Polacek <polacek@redhat.com>
From-SVN: r222132
i386.md (*cmpi<FPCMP:unord><MODEF:mode>_mixed): Merge with *cmpi<FPCMP:unord><MODEF:mode>_sse using enabled attribute.
* config/i386/i386.md (*cmpi<FPCMP:unord><MODEF:mode>_mixed):
Merge with *cmpi<FPCMP:unord><MODEF:mode>_sse using enabled attribute.
(*extendsfdf2_mixed): Merge with *extendsfdf2_sse using enabled
attribute.
(*truncdfsf_fast_mixed): Merge with *truncdfsf_fast_sse using
enabled attribute.
(*float<SWI48:mode><MODEF:mode>2_mixed): Rename from
*float<SWI48:mode><MODEF:mode>2_sse.
(*absneg<mode>2_mixed): Merge with *absneg<mode>2_sse using
enabled attribute.
(*fop_<mode>_comm_mixed): Merge with *fop_<mode>_comm_sse using
enabled attribute.
Jeff Law [Wed, 15 Apr 2015 18:51:49 +0000 (12:51 -0600)]
re PR tree-optimization/47679 (Strange uninitialized warning after SRA)
PR tree-optimization/47679
* tree-ssa-dom.c (build_and_record_new_cond): Moved to avoid
need for forward declaration in upcoming changes.
(record_conditions, record_edge_info): Likewise.
Jakub Jelinek [Wed, 15 Apr 2015 11:47:44 +0000 (13:47 +0200)]
re PR ipa/65765 (Compiling Firefox with GCC 5 leads to broken javascript engine on x86-64)
PR ipa/65765
* ipa-icf-gimple.c (func_checker::compare_bb): For GIMPLE_NOP
and GIMPLE_PREDICT use break instead of return true. For
GIMPLE_EH_DISPATCH, compare dispatch region.
Matthew Wahab [Wed, 15 Apr 2015 08:35:53 +0000 (08:35 +0000)]
extend.texi (__sync Builtins): Simplify some text.
* doc/extend.texi (__sync Builtins): Simplify some text. Update
details about the implementation. Make clear preference for
__atomic builtins. Reduce possibility of future change.
Nick Clifton [Wed, 15 Apr 2015 07:41:07 +0000 (07:41 +0000)]
rx.opt (mallow-string-insns): New option.
* config/rx/rx.opt (mallow-string-insns): New option.
* config/rx/rx.c (RX_BUILTIN_RMPA): Disable the use of this
builtin if string instructions are denied.
* config/rx/rx.h (TARGET_CPU_CPP_BUILTINS): Define
__RX_ALLOW_STRING_INSNS__ or __RX_DISALLOW_STRING_INSNS__, as
appropriate.
(ASM_SPEC): Pass -mno-allow-string-insns on to the assembler.
* config/rx/rx.md (movstr): Enable pattern only if string
instructions are allowed.
(rx_movstr, rx_strend, movmemsi, rx_movmem): Likewise.
(cmpstrnsi, cmpstrsi, rx_cmpstrn, rmpa): Likewise.
* config/rx/t-rx (MULTILIB_OPTIONS): Add mno-allow-string-insns.
(MULTILIB_DIRNAMES): Add no-strings.
* doc/invoke.texi: Document -mno-allow-string-insns.
* gcc.target/rx/builtins.c: Disable RMPA test if string
instructions are not allowed.
Nick Clifton [Wed, 15 Apr 2015 07:24:44 +0000 (07:24 +0000)]
rl78.c (rl78_expand_prologue): Mark large stack decrement instruction as being frame related.
* config/rl78/rl78.c (rl78_expand_prologue): Mark large stack
decrement instruction as being frame related.
(rl78_print_operand_1): Handle 'p' modifier to add +0 to HL
based addresses.
If zero extending a function address enclose the operation in
%code(...).
(rl78_preferred_reload_class): New function.
(TARGET_PREFERRED_RELOAD_CLASS): Define.
* config/rl78/rl78.md: Remove useless constraints in expanders.
(mulqi3_rl78): Remove + qualifier on input-only operand 1.
(mulhi3_rl78): Likewise.
(mulhi3_g13): Likewise.
(mulsi3_rl78): Likewise.
(es_addr): Move to before the multiply patterns.
Jerry DeLisle [Wed, 15 Apr 2015 01:27:03 +0000 (01:27 +0000)]
re PR fortran/65089 (FAIL: gfortran.dg/io_real_boz(2|_[45]).f90 when tested with -fsanitize=address)
2015-04-14 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libgfortran/65089
* io/format.h (free_format): New function to free memory
allocated for building format error messages.
* io/format.c (format_error): Add checks before freeing memory
to avoid potential segfaults and free formatting data when
needed on error conditions. Always allocate and NULL terminate
the string.
* io/transfer.c (st_read_done, st_write_done): Use new
free_format function to clean up memory allocations when done.
* config/i386/i386.h (LEGACY_INT_REG_P): New define.
(LEGACY_INT_REGNO_P): Ditto.
(GENERAL_REGNO_P): Use LEGACY_INT_REGNO_P.
(ANY_MASK_REG_P): Remove.
(BND_REG_P): Rename from ANY_BND_REG_P.
* config/i386/i386.c (print_reg): Use LEGACY_INT_REG_P to print
legacy integer registers. Do not handle MMX_REG_P in a special way.
Merge 64byte and 32byte SSE handling.
re PR c++/60994 (gcc does not recognize hidden/shadowed enumeration as valid nested-name-specifier)
PR c++/60994
* parser.c (cp_parser_class_name): Add enum_ok parameter.
(cp_parser_qualifying_entity): Use it instead of cp_parser_type_name.
(cp_parser_diagnose_invalid_type_name): Don't assume a template is
a class template.
Co-Authored-By: Jason Merrill <jason@redhat.com>
From-SVN: r222094
Marc Glisse [Tue, 14 Apr 2015 11:02:48 +0000 (13:02 +0200)]
re PR libstdc++/61347 (std::distance(list.first(),list.end()) in O(1))
2015-04-14 Marc Glisse <marc.glisse@inria.fr>
PR libstdc++/61347
* include/bits/stl_iterator_base_funcs.h (_List_iterator,
_List_const_iterator): Declare.
(__distance): Declare new overloads for _List_iterator and
_List_const_iterator.
* include/bits/stl_list.h (__distance): New overloads for
_List_iterator and _List_const_iterator.
* testsuite/23_containers/list/61347.cc: New testcase.
predicates.md (any_QIreg_operand): Rename from q_regs_operand.
* config/i386/predicates.md (any_QIreg_operand): Rename from
q_regs_operand. Do not process subregs.
(QIreg_operand): Use QI_REGNO_P predicate.
(ext_QIreg_operand): Ditto.
(ext_register_operand): Ditto.
* config/i386/i386.md (TEST splitters): Use QIreg_operand predicate.
(AND splitters): Ditto.
(AND with -65536 splitter): Add SWI48 mode for operand 0.
(AND with -256 splitter): Use any_QIreg_operand predicate and
SWI248 mode for operand 0.
(AND with -65281 splitter): Use QIreg_operand predicate and SWI248
mode for operand 0.
(SETCC + MOVZBL peepholes): Update for renamed any_QIreg_operand.
Jan Hubicka [Mon, 13 Apr 2015 14:16:41 +0000 (16:16 +0200)]
nonzero-3.C: New testcase.
* g++.dg/tree-ssa/nonzero-3.C: New testcase.
* tree-vrp.c (nonnull_arg_p): THIS pointers and references are non-zero.
(gimple_stmt_nonzero_warnv_p): Reference return values are non-zero.
Jakub Jelinek [Mon, 13 Apr 2015 08:44:56 +0000 (10:44 +0200)]
gcc_release: Fix up thinko in the last commit, improve error message.
* gcc_release: Fix up thinko in the last commit, improve error
message. Commit all ChangeLog, DEV-PHASE and BASE-VER changes
together, rather than doing dozens of separate commits.
Joseph Myers [Sun, 12 Apr 2015 21:35:17 +0000 (22:35 +0100)]
Don't enable -fprefetch-loop-arrays for TARGET_SOFTWARE_PREFETCHING_BENEFICIAL if -Os.
Testing an x86_64 toolchain configured for a processor with
TARGET_SOFTWARE_PREFETCHING_BENEFICIAL produces failures
FAIL: g++.dg/pr60518.C -std=gnu++98 (test for excess errors)
FAIL: g++.dg/pr60518.C -std=gnu++11 (test for excess errors)
FAIL: g++.dg/pr60518.C -std=gnu++14 (test for excess errors)
where the failure is from the message "pr60518.C:1:0: warning:
-fprefetch-loop-arrays is not supported with -Os".
Given that lack of support, it seems appropriate for the back end not
to enable this option in the -Os case; this patch implements that.
Bootstrapped with no regressions on x86_64-unknown-linux-gnu; tested
for a cross to x86_64-linux-gnu --with-arch=btver2, where those test
failures duly disappear.
* config/i386/i386.c (ix86_option_override_internal): Don't set
-fprefetch-loop-arrays if optimizing for size.
Jakub Jelinek [Sun, 12 Apr 2015 20:49:37 +0000 (22:49 +0200)]
gcc_release: For RELEASE_MAJOR 5+ only use the major number in branch name.
* gcc_release: For RELEASE_MAJOR 5+ only use the major
number in branch name. When doing final release of 5+,
if BASE-VER doesn't match release, but is *.*.1 for the
one smaller minor version, bump BASE-VER.