gcc/
* hard-reg-set.h: Include hash-table.h.
(target_hard_regs): Add a finalize method and a x_simplifiable_subregs
field.
* target-globals.c (target_globals::~target_globals): Call
hard_regs->finalize.
* rtl.h (subreg_shape): New structure.
(shape_of_subreg): New function.
(simplifiable_subregs): Declare.
* reginfo.c (simplifiable_subreg): New structure.
(simplifiable_subregs_hasher): Likewise.
(simplifiable_subregs): New function.
(invalid_mode_changes): Delete.
(alid_mode_changes, valid_mode_changes_obstack): New variables.
(record_subregs_of_mode): Remove subregs_of_mode parameter.
Record valid mode changes in valid_mode_changes.
(find_subregs_of_mode): Remove subregs_of_mode parameter.
Update calls to record_subregs_of_mode.
(init_subregs_of_mode): Remove invalid_mode_changes and bitmap
handling. Initialize new variables. Update call to
find_subregs_of_mode.
(invalid_mode_change_p): Check new variables instead of
invalid_mode_changes.
(finish_subregs_of_mode): Finalize new variables instead of
invalid_mode_changes.
(target_hard_regs::finalize): New function.
* ira-costs.c (print_allocno_costs): Call invalid_mode_change_p
even when CLASS_CANNOT_CHANGE_MODE is undefined.
gcc/
* hard-reg-set.h (COPY_HARD_REG_SET, COMPL_HARD_REG_SET)
(AND_HARD_REG_SET, AND_COMPL_HARD_REG_SET, IOR_HARD_REG_SET)
(IOR_COMPL_HARD_REG_SET): Allow the "from" set to be constant.
configure.ac (target_header_dir): Move block defining this to before the block setting inhibit_libc.
* configure.ac (target_header_dir): Move block defining
this to before the block setting inhibit_libc.
(inhibit_libc): When considering $with_headers, just
check it it's explicitly "no". If not, also check if
$target_header_dir/stdio.h is present. If not, set
inhibit_libc=true.
* configure: Regenerate.
* config/rs6000/rs6000.md (div<mode>3): Fix comment. Use a different
insn for divides by integer powers of two.
(div<mode>3_sra, *div<mode>3_sra_dot, *div<mode>3_sra_dot2): New.
(mod<mode>3): Fix formatting.
(three anonymous define_insn and two define_split): Delete.
* config/rs6000/rs6000.md (strlensi): Don't use subsi3 with a
constant, use addsi3 directly.
(three anonymous define_insn, two define_split): Delete.
(sub<mode>3): Move. Do not allow constant second operand.
Generate different insn for constant first operand.
(*subf<mode>3, *subf<mode>3_dot, *subf<mode>3_dot2): New.
(subf<mode>3_imm): New.
(ctz<mode>2, ffs<mode>2): Clobber CA_REGNO where required.
(*plus_ltu<mode>): Only handle registers.
(*plus_ltu<mode>_1): New. Handle integer third operand.
(*plus_gtu<mode>): Only handle registers.
(*plus_gtu<mode>_1): New. Handle integer third operand.
* config/rs6000/predicates.md (ca_operand): Allow subregs.
(input_operand): Do not allow ca_operand.
* config/rs6000/rs6000.c (rs6000_hard_regno_mode_ok): For the
carry bit, allow SImode and Pmode.
(rs6000_init_hard_regno_mode_ok): Make the carry bit class NO_REGS.
Jan Hubicka [Fri, 19 Sep 2014 23:56:52 +0000 (01:56 +0200)]
re PR c++/61825 (g++.dg/cpp0x/static_assert9.C FAILs)
PR c++/61825
* c-family/c-common.c (handle_alias_ifunc_attribute): Check
that visibility change is possible
(handle_weakref_attribute): Likewise.
* cgraph.h (symtab_node): Add method get_create and
field refuse_visibility_changes.
(symtab_node::get_create): New method.
* fold-const.c (tree_single_nonzero_warnv_p): Use get_create.
* varasm.c (mark_weak): Verify that visibility change is
possible.
* gcc.dg/tree-ssa/nonzero-1.c: Require error to be output.
Michael Meissner [Fri, 19 Sep 2014 19:36:57 +0000 (19:36 +0000)]
predicates.md (fusion_gpr_mem_load): Move testing for base_reg_operand to be common between LO_SUM and PLUS.
2014-09-19 Michael Meissner <meissner@linux.vnet.ibm.com>
* config/rs6000/predicates.md (fusion_gpr_mem_load): Move testing
for base_reg_operand to be common between LO_SUM and PLUS.
(fusion_gpr_mem_combo): New predicate to match a fused address
that combines the addis and memory offset address.
* config/rs6000/rs6000.c (fusion_gpr_load_p): Change calling
signature to pass each argument separately, rather than
using an operands array. Rewrite the insns found by peephole2 to
be a single insn, rather than hoping the insns will still be
together when the peephole pass is done. Drop being called via a
normal peephole.
(emit_fusion_gpr_load): Change calling signature to be called from
the fusion_gpr_load_<mode> insns with a combined memory address
instead of the peephole pass passing the addis and offset
separately.
* config/rs6000/rs6000.md (UNSPEC_FUSION_GPR): New unspec for GPR
fusion.
(power8 fusion peephole): Drop support for doing power8 via a
normal peephole that was created by the peephole2 pass.
(power8 fusion peephole2): Create a new insn with the fused
address, so that the fused operation is kept together after
register allocation is done.
(fusion_gpr_load_<mode>): Likewise.
Fix mcount test cases to only run on supported targets
* gcc.dg/pg-override.c: Only run on x86 Linux.
* gcc.dg/pg.c: Dito.
* gcc.target/i386/fentry-override.c: Exclude for PIC.
* gcc.target/i386/fentry.c: Dito.
rs6000.c (rs6000_special_adjust_field_align_p): Update GCC version name to GCC 5.
* config/rs6000/rs6000.c (rs6000_special_adjust_field_align_p):
Update GCC version name to GCC 5.
(rs6000_function_arg_boundary): Likewise.
(rs6000_function_arg): Likewise.
Ian Lance Taylor [Fri, 19 Sep 2014 15:24:56 +0000 (15:24 +0000)]
simple-object-elf.c (simple_object_elf_write_ehdr): Correctly handle objects with more than SHN_LORESERVE sections.
* simple-object-elf.c (simple_object_elf_write_ehdr): Correctly
handle objects with more than SHN_LORESERVE sections.
(simple_object_elf_write_shdr): Add sh_link parameter.
(simple_object_elf_write_to_file): Correctly handle objects with
more than SHN_LORESERVE sections.
James Greenhalgh [Fri, 19 Sep 2014 10:37:40 +0000 (10:37 +0000)]
[Patch] Teach genrecog/genoutput that scratch registers require write constraint modifiers
gcc/
* doc/md.texi (Modifiers): Consistently use "read/write"
nomenclature rather than "input/output".
* genrecog.c (constraints_supported_in_insn_p): New.
(validate_pattern): If needed, also check constraints on
MATCH_SCRATCH operands.
* genoutput.c (validate_insn_alternatives): Catch earlyclobber
operands with no '=' or '+' modifier.
Andreas Krebbel [Fri, 19 Sep 2014 09:14:59 +0000 (09:14 +0000)]
re PR target/62662 (Miscompilation of Qt on s390x)
2014-09-19 Andreas Krebbel <Andreas.Krebbel@de.ibm.com>
PR target/62662
* config/s390/s390.c (s390_emit_epilogue): When doing the return
address load optimization force s390_optimize_prologue to leave it
that way. Only do the optimization if we already decided to push
r14 into a stack slot.
Joseph Myers [Thu, 18 Sep 2014 23:27:26 +0000 (00:27 +0100)]
Remove LIBGCC2_TF_CEXT target macro.
This patch removes the (undocumented) LIBGCC2_TF_CEXT target macro,
replacing it by -fbuilding-libgcc predefines (and thereby gets rid of
another LIBGCC2_LONG_DOUBLE_TYPE_SIZE conditional, though some more
patches are needed before that target macro can be eliminated). This
macro indicated the suffix used on __builtin_huge_val,
__builtin_copysign, __builtin_fabs built-in function names to produce
the names for a given floating-point mode.
Predefines are added for all floating-point modes supported for
libgcc, not just TFmode. These are fully accurate for modes
corresponding to float, double and long double. For other modes, the
suffix for *constants* is determined by the targetm.c.mode_for_suffix
hook (the limit to two possible suffixes 'w' and 'q' being hardcoded
in various places). This is in fact the suffix for built-in functions
as well where such functions exist.
* For i386, the *q functions always exist (whether or not TFmode is
used for long double). The *w functions never exist (but this
doesn't matter for libgcc, since no i386 configuration treats XFmode
as a supported scalar mode if long double is TFmode; if __float80
were to be supported for 64-bit Android, properly such functions
ought to be added).
* For ia64, the *q functions exist for non-HP-UX (under HP-UX, long
double is TFmode, so they aren't needed). The *w functions never
exist. This is an issue for this libgcc code for the XFmode complex
functions in libgcc on HP-UX; as I understand it, right now those
will accidentally be using TFmode versions of those three functions,
so involving unnecessary conversions, while the sanity check on CEXT
accidentally passes because all it tests is the sizes of the types.
Because of the lack of 'w' functions, the patch uses 'l' when the
constant suffix is 'w', matching what the existing libgcc code would
do for IA64 HP-UX in that case.
Ideally there would be generic code to create such built-in functions
for all supported floating-point types. That may be something to
consider if support for TS 18661-3 (standard bindings for IEEE
754-2008, defining names such as _Float128, and function names such as
copysignf128) is added in future.
Bootstrapped with no regressions on x86_64-unknown-linux-gnu.
Jakub Jelinek [Thu, 18 Sep 2014 16:43:28 +0000 (18:43 +0200)]
re PR c++/63248 (Crash when OpenMP target's array section handling is used with templates)
PR c++/63248
* semantics.c (finish_omp_clauses): Don't call cp_omp_mappable_type
on type of type dependent expressions, and don't call it if
handle_omp_array_sections has kept TREE_LIST because something
was type dependent.
* pt.c (tsubst_expr) <case OMP_TARGET, case OMP_TARGET_DATA>:
Use keep_next_level, begin_omp_structured_block and
finish_omp_structured_block instead of push_stmt_list and
pop_stmt_list.
libgomp/
* testsuite/libgomp.c++/pr63248.C: New test.
Joseph Myers [Thu, 18 Sep 2014 12:00:21 +0000 (13:00 +0100)]
Fix i386 FP_TRAPPING_EXCEPTIONS.
The i386 sfp-machine.h defines FP_TRAPPING_EXCEPTIONS in a way that is
always wrong: it treats a set bit as indicating the exception is
trapping, when actually a set bit (both for 387 and SSE floating
point) indicates it is masked, and a clear bit indicates it is
trapping. This patch fixes this bug.
Bootstrapped with no regressions on x86_64-unknown-linux-gnu.
libgcc:
* config/i386/sfp-machine.h (FP_TRAPPING_EXCEPTIONS): Treat clear
bits not set bits as indicating trapping exceptions.
gcc/testsuite:
* gcc.dg/torture/float128-exact-underflow.c: New test.
Jakub Jelinek [Wed, 17 Sep 2014 19:05:49 +0000 (21:05 +0200)]
re PR debug/63284 (-fcompare-debug issue due to redirection to __builtin_unreachable ())
PR debug/63284
* tree-cfgcleanup.c (fixup_noreturn_call): Don't split block
if there are only debug stmts after the noreturn call, instead
remove the debug stmts.
Jan Hubicka [Wed, 17 Sep 2014 16:15:29 +0000 (18:15 +0200)]
ipa-devirt.c (type_pair, [...]): New types.
* ipa-devirt.c (type_pair, default_hashset_traits): New types.
(odr_types_equivalent_p): Use pair hash.
(odr_subtypes_equivalent_p): Likewise, do structural compare
on ODR types that may be mismatched.
(warn_odr): Support warning when only one field is given.
(odr_types_equivalent_p): Strenghten comparsions made;
support VOIDtype.
(add_type_duplicate): Update VISITED hash set.
projects / thirdparty / gcc.git / log
