Kai Tietz [Wed, 27 Mar 2013 20:54:01 +0000 (21:54 +0100)]
cygwin-stdint.h: Add support for cygwin x64 target.
* config/i386/cygwin-stdint.h: Add support for cygwin x64 target.
* config/i386/t-cygwin-w64: New file.
* config/i386/cygwin-w64.h: New file.
* config/i386/cygwin.h (EXTRA_OS_CPP_BUILTINS): Extend
and add support for x64-cygwin target.
(CPP_SPEC): Likewise.
(CXX_WRAP_SPEC_LIST): Undefine before define.
(LIBGCJ_SONAME): Use 15 as version.
Richard Biener [Wed, 27 Mar 2013 10:38:29 +0000 (10:38 +0000)]
re PR tree-optimization/37021 (Fortran Complex reduction / multiplication not vectorized)
2013-03-27 Richard Biener <rguenther@suse.de>
PR tree-optimization/37021
* tree-vect-data-refs.c (vect_check_strided_load): Allow
REALPART/IMAGPART_EXPRs around the supported refs.
* tree-ssa-structalias.c (find_func_aliases): Assume that
floating-point values are not used to transfer pointers.
* gfortran.dg/vect/fast-math-pr37021.f90: New testcase.
* gcc/target.def (TARGET_HAS_IFUNC_P): New target hook.
* gcc/doc/tm.texi.in (TARGET_HAS_IFUNC_P): New.
* gcc/doc/tm.texi: Regenerate.
* gcc/targhooks.h (default_has_ifunc_p): New.
* gcc/targhooks.c (default_has_ifunc_p): Ditto.
* gcc/config/linux-protos.h: New file.
* gcc/config/linux-android.h (TARGET_HAS_IFUNC_P): Using version of
this hook for linux which disables support of indirect functions in
android.
* gcc/config/linux-android.c: New file.
* gcc/config/t-linux-android.c: Ditto.
* gcc/config.gcc: Added new object file linux-android.o.
* gcc/config/i386/i386.c (ix86_get_function_versions_dispatcher):
Using TARGET_HAS_IFUNC hook instead of HAVE_GNU_INDIRECT_FUNCTION.
* gcc/varasm.c (do_assemble_alias): Likewise.
* configure.ac: Define HAVE_GNU_INDIRECT_FUNCTION as zero if the target
doesn't support indirect functions.
* configure: Regenerate.
+ /* This is used by float.h to define the float_t and double_t data
+ types. For historical reasons both are double on s390 what cannot
+ be changed anymore. */
+ #define TARGET_FLT_EVAL_METHOD 1
+
/* Target machine storage layout. */
Walter Lee [Wed, 27 Mar 2013 06:08:01 +0000 (06:08 +0000)]
tilegx.md (insn_mnz_<mode>): Replaced by ...
* config/tilegx/tilegx.md (insn_mnz_<mode>): Replaced by ...
(insn_mnz_v8qi): ... this ...
(insn_mnz_v4hi): ... and this. Replace (const_int 0) with the
vector equivalent.
(insn_v<n>mnz): Replaced by ...
(insn_v1mnz): ... this ...
(insn_v2mnz): ... and this. Replace (const_int 0) with the vector
equivalent.
(insn_mz_<mode>): Replaced by ...
(insn_mz_v8qi): ... this ...
(insn_mz_v4hi): ... and this. Replace (const_int 0) with the
vector equivalent.
(insn_v<n>mz): Replaced by ...
(insn_v1mz): ... this ...
(insn_v2mz): ... and this. Replace (const_int 0) with the vector
equivalent.
Jason Merrill [Wed, 27 Mar 2013 03:01:41 +0000 (23:01 -0400)]
re PR c++/52597 ([C++11] confusing diagnostics for invalid use of non-static member function in decltype)
PR c++/52597
* typeck.c (invalid_nonstatic_memfn_p): Use get_first_fn. Take tree.
* semantics.c (finish_decltype_type): Check it before type_unknown_p.
* cp-tree.h: Adjust prototype.
Janne Blomqvist [Tue, 26 Mar 2013 22:08:17 +0000 (00:08 +0200)]
PR 25708 Use a temporary buffer when parsing module files.
2013-03-27 Janne Blomqvist <jb@gcc.gnu.org>
PR fortran/25708
* module.c (module_locus): Use long for position.
(module_content): New variable.
(module_pos): Likewise.
(prev_character): Remove.
(bad_module): Free data instead of closing mod file.
(set_module_locus): Use module_pos.
(get_module_locus): Likewise.
(module_char): use buffer rather than stdio file.
(module_unget_char): Likewise.
(read_module_to_tmpbuf): New function.
(gfc_use_module): Call read_module_to_tmpbuf.
Uros Bizjak [Tue, 26 Mar 2013 16:24:48 +0000 (17:24 +0100)]
i386.md (*cmpqi_ext_1): Merge with *cmpqi_ext_1_rex64 using nox64 isa attribute.
* config/i386/i386.md (*cmpqi_ext_1): Merge with *cmpqi_ext_1_rex64
using nox64 isa attribute. Use nonimmediate_x86nomem_operand as
operand 0 predicate.
(*cmpqi_ext_3): Merge with *cmpqi_ext_3_rex64 using nox64 isa
attribute. Use general_x64nomem_operand as operand 1 predicate.
(*movqi_extv_1): Merge with *movqi_extv_1_rex64 using nox64 isa
attribute. Use nonimmediate_x64nomem_operand as operand 0 predicate.
(*movqi_extzv_2): Merge with *movqi_extzv_2_rex64 using nox64 isa
attribute. Use nonimmediate_x64nomem_operand as operand 0 predicate.
(mov<mode>_insv_1): Remove expander. Merge insn with
movsi_insv_1 using SWI48 mode iterator and nox64 isa attribute.
Use general_x64nomem_operand as operand 1 predicate.
(addqi_ext_1): Merge with *addqi_ext_1_rex64 using nox64 isa attribute.
(*testqi_ext_1): Merge with *testqi_ext_1_rex64 using nox64 isa
attribute. Use nonimmediate_x64nomem_operand as operand 1 predicate.
(*andqi_ext_1): Merge with *andqi_ext_1_rex64 using nox64 isa
attribute. Use nonimmediate_x64nomem_operand as operand 2 predicate.
(*<code>qi_ext_1): Merge with *<code>qi_ext_1_rex64 using nox64 isa
attribute. Use nonimmediate_x64nomem_operand as operand 1 predicate.
(*xorqi_cc_ext_1): Merge with *xorqi_cc_ext_1_rex64 using nox64
isa attribute. Use general_x64nomem_operand as operand 2 predicate.
* config/i386/predicates.md (nonimmediate_x64nomem_operand): New.
(general_x64nomem_operand): Ditto.
Richard Biener [Tue, 26 Mar 2013 15:20:38 +0000 (15:20 +0000)]
alias.c (find_base_term): Avoid redundant and not used recursion.
2013-03-26 Richard Biener <rguenther@suse.de>
* alias.c (find_base_term): Avoid redundant and not used
recursion.
(base_alias_check): Get the initial base term from the
caller.
(true_dependence_1): Compute and pass base terms to base_alias_check.
(write_dependence_p): Likewise.
(may_alias_p): Likewise.
Richard Biener [Tue, 26 Mar 2013 09:14:59 +0000 (09:14 +0000)]
emit-rtl.c (set_mem_attributes_minus_bitpos): Remove alignment computations and rely on get_object_alignment_1 for the...
2013-03-26 Richard Biener <rguenther@suse.de>
* emit-rtl.c (set_mem_attributes_minus_bitpos): Remove
alignment computations and rely on get_object_alignment_1
for the !TYPE_P case.
Commonize DECL/COMPONENT_REF handling in the ARRAY_REF path.
Jeff Law [Tue, 26 Mar 2013 04:00:20 +0000 (22:00 -0600)]
tree-ssa-dom.c (record_equivalences_from_incoming_edge): Add missing check for INTEGRAL_TYPE_P that was missing due to checking in wrong...
* tree-ssa-dom.c (record_equivalences_from_incoming_edge): Add missing
check for INTEGRAL_TYPE_P that was missing due to checking in wrong
version of prior patch.
Tilo Schwarz [Mon, 25 Mar 2013 20:11:20 +0000 (20:11 +0000)]
re PR fortran/52512 (Cannot match namelist object name)
2013-03-25 Tilo Schwarz <tilo@tilo-schwarz.de>
PR libfortran/52512
* io/list_read.c (nml_parse_qualifier): To check for a derived type
don't use the namelist head element type but the current element type.
(nml_get_obj_data): Add current namelist element type to
nml_parse_qualifier call.
Jeff Law [Mon, 25 Mar 2013 19:05:57 +0000 (13:05 -0600)]
tree-ssa-dom.c (record_equivalences_from_incoming_edge): Rework slightly to avoid creating and folding useless trees.
* tree-ssa-dom.c (record_equivalences_from_incoming_edge): Rework
slightly to avoid creating and folding useless trees. Simplify
slightly by restricting to INTEGER_CSTs and using int_fits_type_p.
Uros Bizjak [Mon, 25 Mar 2013 18:09:36 +0000 (19:09 +0100)]
i386.md (*zero_extendsidi2): Merge with *zero_extendsidi2_rex64.
* config/i386/i386.md (*zero_extendsidi2): Merge with
*zero_extendsidi2_rex64. Use x64 and nox64 isa attributes.
* config/i386/predicates.md (x86_64_zext_operand): Rename from
x86_64_zext_general_operand. Use nonimmediate_operand on 32bit
targets. Clarify comment.
Martin Jambor [Mon, 25 Mar 2013 16:50:02 +0000 (17:50 +0100)]
ipa-prop.c (ipa_write_jump_function): Stream simple and aritmetic pass-through jump functions differently.
2013-03-25 Martin Jambor <mjambor@suse.cz>
* ipa-prop.c (ipa_write_jump_function): Stream simple and aritmetic
pass-through jump functions differently.
(ipa_read_jump_function): Likewise. Also use setter functions to set
up jump functions.
Martin Jambor [Mon, 25 Mar 2013 16:42:41 +0000 (17:42 +0100)]
ipa-cp.c (ipa_get_indirect_edge_target): Renamed to ipa_get_indirect_edge_target_1...
2013-03-25 Martin Jambor <mjambor@suse.cz>
* ipa-cp.c (ipa_get_indirect_edge_target): Renamed to
ipa_get_indirect_edge_target_1, added parameter agg_reps and ability to
process it.
(ipa_get_indirect_edge_target): New function.
(devirtualization_time_bonus): New parameter known_aggs, pass it to
ipa_get_indirect_edge_target. Update all callers.
(ipcp_discover_new_direct_edges): New parameter aggvals. Pass it to
ipa_get_indirect_edge_target_1 instead of calling
ipa_get_indirect_edge_target.
(create_specialized_node): Pass aggvlas to
ipcp_discover_new_direct_edges.
Kyrylo Tkachov [Mon, 25 Mar 2013 14:55:05 +0000 (14:55 +0000)]
Fix ChangeLog formatting.
gcc/
Fix ChangeLog formatting.
gcc/testsuite
2013-03-25 Kyrylo Tkachov <kyrylo.tkachov at arm.com>
* gcc.target/aarch64/atomic-comp-swap-release-acquire.c: Move test
body from here...
* gcc.target/aarch64/atomic-comp-swap-release-acquire.x: ... to here.
* gcc.target/aarch64/atomic-op-acq_rel.c: Move test body from here...
* gcc.target/aarch64/atomic-op-acq_rel.x: ... to here.
* gcc.target/aarch64/atomic-op-acquire.c: Move test body from here...
* gcc.target/aarch64/atomic-op-acquire.x: ... to here.
* gcc.target/aarch64/atomic-op-char.c: Move test body from here...
* gcc.target/aarch64/atomic-op-char.x: ... to here.
* gcc.target/aarch64/atomic-op-consume.c: Move test body from here...
* gcc.target/aarch64/atomic-op-consume.x: ... to here.
* gcc.target/aarch64/atomic-op-int.c: Move test body from here...
* gcc.target/aarch64/atomic-op-int.x: ... to here.
* gcc.target/aarch64/atomic-op-relaxed.c: Move test body from here...
* gcc.target/aarch64/atomic-op-relaxed.x: ... to here.
* gcc.target/aarch64/atomic-op-release.c: Move test body from here...
* gcc.target/aarch64/atomic-op-release.x: ... to here.
* gcc.target/aarch64/atomic-op-seq_cst.c: Move test body from here...
* gcc.target/aarch64/atomic-op-seq_cst.x: ... to here.
* gcc.target/aarch64/atomic-op-short.c: Move test body from here...
* gcc.target/aarch64/atomic-op-short.x: ... to here.
* gcc.target/arm/atomic-comp-swap-release-acquire.c: New test.
* gcc.target/arm/atomic-op-acq_rel.c: Likewise.
* gcc.target/arm/atomic-op-acquire.c: Likewise.
* gcc.target/arm/atomic-op-char.c: Likewise.
* gcc.target/arm/atomic-op-consume.c: Likewise.
* gcc.target/arm/atomic-op-int.c: Likewise.
* gcc.target/arm/atomic-op-relaxed.c: Likewise.
* gcc.target/arm/atomic-op-release.c: Likewise.
* gcc.target/arm/atomic-op-seq_cst.c: Likewise.
* gcc.target/arm/atomic-op-short.c: Likewise.
Uros Bizjak [Sun, 24 Mar 2013 18:26:48 +0000 (19:26 +0100)]
sse.md (mov<mode>): Merge with movv2sf expander using MMXMODE mode iterator.
* config/i386/sse.md (mov<mode>): Merge with movv2sf expander
using MMXMODE mode iterator.
(*move<mode>_internal): Merge with *movv2sf_internal and
*movv2sf_internal_rex64 using MMXMODE mode iterator.
Andi Kleen [Sun, 24 Mar 2013 00:23:10 +0000 (00:23 +0000)]
Avoid non constant memory model uses in libatomic
x86 ends up using non constant memory models for some of the libatomic
functions. These all end up as __ATOMIC_SEQ_CST. Just use this
directly. This avoids a new warning for non constant memory
models, which broke the bootstrap with -Werror
Passed bootstrap and test on x86_64-linux.
libatomic/:
2013-03-23 Andi Kleen <ak@linux.intel.com>
* gcas.c: (EXACT_INLINE): Use __ATOMIC_SEQ_CST.
* gexch.c: (EXACT_INLINE): Use __ATOMIC_SEQ_CST.
* gload.c: (EXACT_INLINE): Use __ATOMIC_SEQ_CST.
* gstore.c: (EXACT_INLINE): Use __ATOMIC_SEQ_CST.
Steven Bosscher [Sat, 23 Mar 2013 21:40:20 +0000 (21:40 +0000)]
df-core.c (rest_of_handle_df_initialize): Use XCNEWVEC instead of XNEWVEC followed by memset.
* df-core.c (rest_of_handle_df_initialize): Use XCNEWVEC instead
of XNEWVEC followed by memset.
(df_worklist_dataflow): Use XNEWVEC instead of xmalloc with a cast.