Sebastian Pop [Thu, 11 Feb 2010 19:43:05 +0000 (19:43 +0000)]
Add testcase for PR42930.
2010-02-10 Sebastian Pop <sebastian.pop@amd.com>
* g++.dg/graphite/pr42930.C: New.
From-SVN: r156715
Sebastian Pop [Thu, 11 Feb 2010 19:42:57 +0000 (19:42 +0000)]
Fix email address in ChangeLog.graphite.
From-SVN: r156714
Sebastian Pop [Thu, 11 Feb 2010 19:42:51 +0000 (19:42 +0000)]
re PR middle-end/42930 ([graphite] crash when compiling scummvm on Ubuntu 9.10/amd64 with -floop-block)
Fix PR42930.
2010-02-10 Sebastian Pop <seb@napoca>
PR middle-end/42930
* graphite-scop-detection.c (graphite_can_represent_scev): Call
graphite_can_represent_init for MULT_EXPR.
From-SVN: r156713
Sebastian Pop [Thu, 11 Feb 2010 19:42:45 +0000 (19:42 +0000)]
Fix PR42914 and PR42530.
2010-02-10 Sebastian Pop <seb@napoca>
PR middle-end/42914
PR middle-end/42530
* graphite-sese-to-poly.c (remove_phi): New.
(translate_scalar_reduction_to_array): Call remove_phi.
* gcc.dg/graphite/pr42530.c: New.
* gcc.dg/graphite/pr42914.c: New.
From-SVN: r156712
Sebastian Pop [Thu, 11 Feb 2010 19:42:38 +0000 (19:42 +0000)]
re PR tree-optimization/42771 ([graphite] ICE: in graphite_loop_normal_form, at graphite-sese-to-poly.c (2))
Fix PR42771.
2010-02-10 Sebastian Pop <seb@napoca>
PR middle-end/42771
* graphite-clast-to-gimple.c (gloog): Call rename_sese_parameters.
* graphite-clast-to-gimple.h (gloog): Update declaration.
* graphite-poly.c (new_scop): Clear POLY_SCOP_P.
* graphite-poly.h (struct poly_bb): Add missing comments.
(struct scop): Add poly_scop_p field.
(POLY_SCOP_P): New.
* graphite-sese-to-poly.c (build_poly_scop): Set POLY_SCOP_P.
* graphite.c (graphite_transform_loops): Build the polyhedral
representation for each scop before code generation.
* sese.c (rename_variables_in_operand): Removed.
(rename_variables_in_expr): Return the renamed expression.
(rename_sese_parameters): New.
* sese.h (rename_sese_parameters): Declared.
* gcc.dg/graphite/pr42771.c: New.
From-SVN: r156711
Jakub Jelinek [Thu, 11 Feb 2010 19:31:35 +0000 (20:31 +0100)]
re PR fortran/43029 (enum_5.f90 ICE under valgrind)
PR fortran/43029
* decl.c (enumerator_decl): Don't call gfc_free_enum_history
here.
(gfc_match_enumerator_def): But here whenever enumerator_decl returns
MATCH_ERROR.
From-SVN: r156709
Jason Merrill [Thu, 11 Feb 2010 19:00:13 +0000 (14:00 -0500)]
thread (~_Impl_base): Declare inline.
* include/std/thread (~_Impl_base): Declare inline.
* libsupc++/nested_exception.h (~nested_exception): Likewise.
From-SVN: r156707
Paolo Carlini [Thu, 11 Feb 2010 18:11:01 +0000 (18:11 +0000)]
PR libstdc++/41975, DR 579
2010-02-11 Paolo Carlini <paolo.carlini@oracle.com>
PR libstdc++/41975, DR 579
* include/bits/hashtable.h (_Hashtable<>::_M_erase_node): Remove.
(erase(const_iterator), erase(const_iterator, const_iterator)):
Change return type to void.
* include/debug/unordered_map: Adjust.
* include/debug/unordered_set: Likewise.
* testsuite/util/exception/safety.h: Likewise.
* testsuite/23_containers/unordered_map/erase/1.cc: Likewise.
* testsuite/23_containers/unordered_map/erase/24061-map.cc: Likewise.
* testsuite/23_containers/unordered_set/erase/1.cc: Likewise.
* testsuite/23_containers/unordered_set/erase/24061-map.cc: Likewise.
* testsuite/23_containers/unordered_multimap/erase/1.cc: Likewise.
* testsuite/23_containers/unordered_multimap/erase/24061-map.cc:
Likewise.
* testsuite/23_containers/unordered_multiset/erase/1.cc: Likewise.
* testsuite/23_containers/unordered_multiset/erase/24061-map.cc:
Likewise.
From-SVN: r156705
Richard Guenther [Thu, 11 Feb 2010 17:01:21 +0000 (17:01 +0000)]
re PR tree-optimization/42998 (ICE: error: verify_stmts failed)
2010-02-11 Richard Guenther <rguenther@suse.de>
PR tree-optimization/42998
* tree-ssa-pre.c (create_expression_by_pieces): Treat
POINTER_PLUS_EXPR properly.
* gcc.c-torture/compile/pr42998.c: New testcase.
From-SVN: r156704
Sebastian Pop [Thu, 11 Feb 2010 15:45:27 +0000 (15:45 +0000)]
re PR regression/40886 (No loop counter reversal for simple loops anymore)
Fix PR40886.
2010-02-11 Sebastian Pop <sebastian.pop@amd.com>
Changpeng Fang <changpeng.fang@amd.com>
PR middle-end/40886
* tree-ssa-loop-ivopts.c (determine_use_iv_cost_condition): Decrement
the cost of an IV candidate when the IV is used in a test against zero.
* gcc.dg/tree-ssa/ivopts-3.c: New.
Co-Authored-By: Changpeng Fang <changpeng.fang@amd.com>
From-SVN: r156701
Richard Guenther [Thu, 11 Feb 2010 15:38:59 +0000 (15:38 +0000)]
re PR lto/41664 (FAIL: gfortran.dg/lto/pr40725 f_lto_pr40725_0.o-f_lto_pr40725_1.o execute -O2 -fwhopr and -flto)
2010-02-11 Richard Guenther <rguenther@suse.de>
PR lto/41664
* tree-ssa-alias.c (refs_may_alias_p_1): Canonicalize
pointer-vs-decl case by swapping refs. Handle some cases
of pointer-vs-decl disambiguations more conservatively.
* cfgexpand.c (gimple_expand_cfg): Set gimple_df->in_ssa_p
to false after expanding.
From-SVN: r156699
Paolo Carlini [Thu, 11 Feb 2010 14:59:46 +0000 (14:59 +0000)]
gnu.ver: Avoid exporting inadvertently std::nested_exception symbols at GLIBCXX_3.4.
2010-02-11 Paolo Carlini <paolo.carlini@oracle.com>
* config/abi/pre/gnu.ver: Avoid exporting inadvertently
std::nested_exception symbols at GLIBCXX_3.4.
From-SVN: r156698
Richard Guenther [Thu, 11 Feb 2010 12:23:08 +0000 (12:23 +0000)]
re PR driver/43021 (Driver no longer handles fancy names)
2010-02-11 Richard Guenther <rguenther@suse.de>
PR driver/43021
* gcc.c (process_command): Handle LTO file@offset case more
appropriately.
lto/
* lto-elf.c (lto_elf_file_open): Handle file@offset case more
appropriately.
From-SVN: r156695
Jakub Jelinek [Thu, 11 Feb 2010 09:17:54 +0000 (10:17 +0100)]
reload1.c (eliminate_regs_1): If insn is DEBUG_INSN, avoid any modifications outside of the DEBUG_INSN.
* reload1.c (eliminate_regs_1): If insn is DEBUG_INSN, avoid any
modifications outside of the DEBUG_INSN. Accept CLOBBERs inside
of DEBUG_INSNs.
(eliminate_regs_in_insn): Eliminate regs even in DEBUG_INSNs.
From-SVN: r156693
Jakub Jelinek [Thu, 11 Feb 2010 09:16:38 +0000 (10:16 +0100)]
dwarf2out.c (mem_loc_descriptor): Use DW_OP_deref_size if MEM's mode size isn't DWARF2_ADDR_SIZE.
* dwarf2out.c (mem_loc_descriptor) <case MEM>: Use DW_OP_deref_size
if MEM's mode size isn't DWARF2_ADDR_SIZE.
(mem_loc_descriptor) <do_scompare>: Allow also VOIDmode arguments.
Optimize eq/ne comparisons when both arguments are known to be
zero-extended.
(mem_loc_descriptor) <do_ucompare>: Allow also VOIDmode arguments.
Don't mask operands unnecessarily if they are known to be already
zero-extended.
From-SVN: r156692
Jason Merrill [Thu, 11 Feb 2010 02:12:53 +0000 (21:12 -0500)]
re PR c++/41896 ([c++0x] Segfault because of a nested lambda function)
PR c++/41896
* semantics.c (outer_lambda_capture_p): Revert.
(add_capture): Only finish_member_declaration if
we're in the lambda class.
(register_capture_members): New.
* cp-tree.h: Declare it.
* parser.c (cp_parser_lambda_expression): Call it.
From-SVN: r156678
GCC Administrator [Thu, 11 Feb 2010 00:17:24 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r156677
Jason Merrill [Wed, 10 Feb 2010 22:45:07 +0000 (17:45 -0500)]
re PR c++/41896 ([c++0x] Segfault because of a nested lambda function)
PR c++/41896
* semantics.c (outer_lambda_capture_p): Use current_function_decl
instead of current_class_type.
From-SVN: r156673
Jason Merrill [Wed, 10 Feb 2010 21:48:35 +0000 (16:48 -0500)]
PR c++/42983, core issue 906
PR c++/42983, core issue 906
* method.c (defaultable_fn_check): Check virtualness.
* include/std/thread (~_Impl_base): Move default out of line.
* libsupc++/nested_exception.h (~nested_exception): Likewise.
From-SVN: r156672
Jason Merrill [Wed, 10 Feb 2010 21:48:25 +0000 (16:48 -0500)]
re PR c++/43016 ([C++0x] Inappropriate multiple definition error for lambda function when inside inline functions)
PR c++/43016
* semantics.c (maybe_add_lambda_conv_op): Set DECL_INTERFACE_KNOWN.
From-SVN: r156671
Vladimir Makarov [Wed, 10 Feb 2010 20:52:31 +0000 (20:52 +0000)]
ira-conflicts.c (add_insn_allocno_copies): Use find_reg_note instead of loop.
2010-02-10 Vladimir Makarov <vmakarov@redhat.com>
* ira-conflicts.c (add_insn_allocno_copies): Use find_reg_note
instead of loop.
From-SVN: r156670
Paolo Carlini [Wed, 10 Feb 2010 19:14:33 +0000 (19:14 +0000)]
hashtable.h: Fold in include/tr1_impl/hashtable.h contents.
2010-02-10 Paolo Carlini <paolo.carlini@oracle.com>
* include/tr1/hashtable.h: Fold in include/tr1_impl/hashtable.h
contents.
* include/tr1_impl/hashtable.h: Remove.
* include/tr1_impl/hashtable_policy.h: Move and rename...
* include/tr1/hashtable_policy.h: ... here.
* include/tr1_impl/unordered_map: Move and rename...
* include/tr1/unordered_map.h: ... here.
* include/tr1_impl/unordered_set: Move and rename...
* include/tr1/unordered_set.h: ... here.
* include/tr1_impl/unordered_map: Move and rename...
* include/tr1/unordered_map.h: ... here.
* include/tr1/unordered_map: Adjust.
* include/tr1/unordered_set: Likewise.
* include/Makefile.am: Adjust.
* include/Makefile.in: Regenerate.
* include/std/c++0x_warning.h: Move...
* include/bits/c++0x_warning.h: ... here.
* include/debug/unordered_map: Adjust.
* include/debug/unordered_set: Likewise.
* include/std/tuple: Likewise.
* include/std/system_error: Likewise.
* include/std/forward_list: Likewise.
* include/std/thread: Likewise.
* include/std/type_traits: Likewise.
* include/std/future: Likewise.
* include/std/regex: Likewise.
* include/std/atomic: Likewise.
* include/std/ratio: Likewise.
* include/std/random: Likewise.
* include/std/unordered_map: Likewise.
* include/std/chrono: Likewise.: Likewise.
* include/std/condition_variable: Likewise.
* include/std/mutex: Likewise.
* include/std/unordered_set: Likewise.
* include/std/array: Likewise.
* include/profile/unordered_map: Likewise.
* include/profile/unordered_set: Likewise.
* include/c_global/cinttypes: Likewise.
* include/c_global/ctgmath: Likewise.
* include/c_global/cfenv: Likewise.
* include/c_global/cstdint: Likewise.
* include/c_global/cstdbool: Likewise.
* include/c_global/ccomplex: Likewise.
* include/bits/shared_ptr.h: Do not include <bits/c++0x_warning.h>
at all, the file is not meant to be included by the users.
* include/bits/functional_hash.h: Likewise.
* include/bits/unique_ptr.h: Likewise.
* include/bits/forward_list.h: Likewise.
* testsuite/29_atomics/atomic/cons/assign_neg.cc: Adjust dg-*
line numbers.
* testsuite/29_atomics/atomic/cons/copy_neg.cc: Likewise.
* testsuite/23_containers/forward_list/requirements/dr438/
assign_neg.cc: Likewise.
* testsuite/23_containers/forward_list/requirements/dr438/
insert_neg.cc: Likewise.
* testsuite/23_containers/forward_list/requirements/dr438/
constructor_1_neg.cc: Likewise.
* testsuite/23_containers/forward_list/requirements/dr438/
constructor_2_neg.cc: Likewise.
* testsuite/20_util/unique_ptr/modifiers/reset_neg.cc: Likewise.
* testsuite/20_util/unique_ptr/assign/assign_neg.cc: Likewise.
From-SVN: r156667
Richard Guenther [Wed, 10 Feb 2010 16:52:07 +0000 (16:52 +0000)]
re PR tree-optimization/43017 (VRP miscompiles python with -fwrapv, II)
2010-02-10 Richard Guenther <rguenther@suse.de>
PR tree-optimization/43017
* tree-vrp.c (vrp_int_const_binop): Trust int_const_binop
for wrapping signed arithmetic.
* gcc.dg/torture/pr43017.c: New testcase.
From-SVN: r156666
Tobias Burnus [Wed, 10 Feb 2010 16:48:24 +0000 (17:48 +0100)]
[multiple changes]
2010-02-10 Joost VandeVondele <jv244@cam.ac.uk>
Tobias Burnus <burnus@net-b.de>
PR fortran/40823
* decl.c (gfc_match_subroutine): Explicitly set
* sym->declared_at.
2010-02-10 Tobias Burnus <burnus@net-b.de>
PR fortran/40823
* gfortran.dg/private_type_1.f90: Update error location.
* gfortran.dg/invalid_interface_assignment.f90: Ditto.
* gfortran.dg/typebound_operator_2.f03: Ditto.
* gfortran.dg/assignment_2.f90: Ditto.
* gfortran.dg/redefined_intrinsic_assignment.f90: Ditto.
* gfortran.dg/binding_label_tests_9.f03: Ditto.
From-SVN: r156665
Tobias Burnus [Wed, 10 Feb 2010 16:43:22 +0000 (17:43 +0100)]
re PR fortran/43015 (ICE with BIND(C) and -fbounds-check in mingw-w64 cross-compiler)
2010-02-10 Tobias Burnus <burnus@net-b.de>
PR fortran/43015
* trans-decl.c (gfc_generate_function_code): Only check
actual-vs.-dummy character bounds if not bind(C).
2010-02-10 Tobias Burnus <burnus@net-b.de>
PR fortran/43015
* gfortran.dg/bind_c_usage_20.f90: New test.
From-SVN: r156663
Paolo Carlini [Wed, 10 Feb 2010 16:09:42 +0000 (16:09 +0000)]
hashtable.h: Fold in include/tr1_impl/hashtable.h for C++0x use.
2010-02-10 Paolo Carlini <paolo.carlini@oracle.com>
* include/bits/hashtable.h: Fold in include/tr1_impl/hashtable.h
for C++0x use.
* include/bits/hashtable_policy.h: New, copy and adjust for
C++0x use, include/tr1_impl/hashtable_policy.h; fix erase and
insert member functions per n3000.
* include/bits/unordered_map.h: Likewise for include/tr1_impl/
unordered_map.
* include/bits/unordered_set.h: Likewise for include/tr1_impl/
unordered_set.
* include/Makefile.am: Adjust.
* include/Makefile.in: Regenerate.
* include/tr1/unordered_map: Adjust.
* include/tr1/unordered_set: Likewise.
* include/tr1_impl/unordered_map: Adjust, now used only by tr1.
* include/tr1_impl/hashtable: Likewise.
* include/tr1_impl/hashtable_policy.h: Likewise.
* include/tr1_impl/unordered_set: Likewise.
* include/std/unordered_map: Adjust and simplify includes.
* include/std/unordered_set: Likewise.
* include/debug/unordered_map: Adjuse erase and insert members.
* include/debug/unordered_set: Likewise.
* include/profile/unordered_map: Likewise.
* include/profile/unordered_set: Likewise.
* testsuite/util/exception/safety.h: Fix for the updated erase and
insert member functions of the unordered_containers.
* testsuite/23_containers/unordered_map/erase/1.cc: New.
* testsuite/23_containers/unordered_map/erase/24061-map.cc: Likewise.
* testsuite/23_containers/unordered_map/insert/map_single.cc:
Likewise.
* testsuite/23_containers/unordered_map/insert/array_syntax.cc:
Likewise.
* testsuite/23_containers/unordered_map/insert/24061-map.cc: Likewise.
* testsuite/23_containers/unordered_map/insert/map_range.cc: Likewise.
* testsuite/23_containers/set/operators/1_neg.cc: Likewise.
* testsuite/23_containers/unordered_multimap/erase/1.cc: Likewise.
* testsuite/23_containers/unordered_multimap/erase/
24061-multimap.cc: Likewise.
* testsuite/23_containers/unordered_multimap/insert/
24061-multimap.cc: Likewise.
* testsuite/23_containers/unordered_multimap/insert/
multimap_range.cc: Likewise.
* testsuite/23_containers/unordered_multimap/insert/
multimap_single.cc: Likewise.
* testsuite/23_containers/unordered_set/erase/1.cc: Likewise.
* testsuite/23_containers/unordered_set/erase/24061-set.cc: Likewise.
* testsuite/23_containers/unordered_set/insert/set_single.cc: Likewise.
* testsuite/23_containers/unordered_set/insert/24061-set.cc: Likewise.
* testsuite/23_containers/unordered_set/insert/set_range.cc: Likewise.
* testsuite/23_containers/unordered_multiset/erase/1.cc: Likewise.
* testsuite/23_containers/unordered_multiset/erase/
24061-multiset.cc: Likewise.
* testsuite/23_containers/unordered_multiset/insert/
24061-multiset.cc: Likewise.
* testsuite/23_containers/unordered_multiset/insert/
multiset_range.cc: Likewise.
* testsuite/23_containers/unordered_multiset/insert/
multiset_single.cc: Likewise.
* testsuite/23_containers/set/operators/1_neg.cc: Tweak dg-errors
to avoid spurious fails in debug-mode.
* testsuite/23_containers/map/operators/1_neg.cc: Likewise.
From-SVN: r156661
Jakub Jelinek [Wed, 10 Feb 2010 15:10:53 +0000 (16:10 +0100)]
re PR fortran/42309 (Problem with a pointer array passed to a subroutine)
PR fortran/42309
* trans-expr.c (gfc_conv_subref_array_arg): Avoid accessing
info->dimen after info has been freed.
From-SVN: r156659
Jakub Jelinek [Wed, 10 Feb 2010 15:02:56 +0000 (16:02 +0100)]
re PR debug/43010 (ICE with -femit-struct-debug-baseonly)
PR debug/43010
* dwarf2out.c (retry_incomplete_types): Don't call gen_type_die
if no debug info should be emitted for it.
* g++.dg/debug/pr43010.C: New test.
From-SVN: r156657
Kaz Kojima [Wed, 10 Feb 2010 13:13:49 +0000 (13:13 +0000)]
sh.c (find_barrier): Skip call insn with a REG_EH_REGION note when flag_exceptions is set.
* config/sh/sh.c (find_barrier): Skip call insn with a REG_EH_REGION
note when flag_exceptions is set.
From-SVN: r156656
Duncan Sands [Wed, 10 Feb 2010 11:59:11 +0000 (12:59 +0100)]
Add debug.h to PLUGIN_HEADERS.
From-SVN: r156655
Richard Guenther [Wed, 10 Feb 2010 11:54:14 +0000 (11:54 +0000)]
re PR c/43007 (No longer folds (unsigned int) ((long long unsigned int) spi_bias / 1008))
2010-02-10 Richard Guenther <rguenther@suse.de>
PR c/43007
* tree.c (get_unwidened): Handle constants.
* convert.c (convert_to_integer): Handle TRUNC_DIV_EXPR.
* gcc.c-torture/execute/
20100209-1.c: New testcase.
* gcc.dg/fold-div-3.c: Likewise.
From-SVN: r156653
Martin Jambor [Wed, 10 Feb 2010 11:22:55 +0000 (12:22 +0100)]
re PR lto/42985 (Internal compiler error: in ipcp_iterate_stage with different opt level)
2010-02-10 Martin Jambor <mjambor@suse.cz>
PR lto/42985
* ipa-prop.c (ipa_update_after_lto_read): Count parameters and
check for variable argument counts independently.
From-SVN: r156651
Christian Bruel [Wed, 10 Feb 2010 10:09:31 +0000 (11:09 +0100)]
sh.c (find_barrier): Increase length for non delayed conditional branches.
* config/sh/sh.c (find_barrier): Increase length for non delayed
conditional branches.
From-SVN: r156649
Christian Bruel [Wed, 10 Feb 2010 10:08:01 +0000 (11:08 +0100)]
sh.c (find_barrier): Don't emit a CP inside the GP setting.
* config/sh/sh.c (find_barrier): Don't emit a CP inside the GP setting.
From-SVN: r156648
Jakub Jelinek [Wed, 10 Feb 2010 09:21:12 +0000 (10:21 +0100)]
builtins.c (set_builtin_user_assembler_name): Also handle ffs if int is smaller than word.
* builtins.c (set_builtin_user_assembler_name): Also handle
ffs if int is smaller than word.
* gcc.dg/builtin-ffs-1.c: New test.
From-SVN: r156646
Shujing Zhao [Wed, 10 Feb 2010 07:37:37 +0000 (07:37 +0000)]
Make-lang.in (cp/cvt.o, [...]): Depend on intl.h.
2010-02-10 Shujing Zhao <pearly.zhao@oracle.com>
* Make-lang.in (cp/cvt.o, cp/parser.o, cp/search.o): Depend on intl.h.
* cvt.c (warn_ref_binding): Wrap the messages into G_() for easy
translation.
* parser.c (cp_parser_postfix_expression, cp_parser_new_type_id)
(cp_parser_cast_expression, cp_parser_condition, cp_parser_decltype)
(cp_parser_parameter_declaration)
(cp_parser_exception_specification_opt)
(cp_parser_exception_declaration): Likewise.
* pt.c (check_default_tmpl_args): Likewise.
* search.c (lookup_field_r): Likewise.
From-SVN: r156645
Benjamin Kosnik [Wed, 10 Feb 2010 05:45:24 +0000 (05:45 +0000)]
streambuf: Adjust doxygen group markup.
2010-02-09 Benjamin Kosnik <bkoz@redhat.com>
* include/std/streambuf: Adjust doxygen group markup.
* include/std/functional: Same.
* include/ext/vstring.h: Same.
* include/ext/pb_ds/assoc_container.hpp: Same.
* include/ext/stdio_filebuf.h: Same.
* include/ext/stdio_sync_filebuf.h: Same.
* include/bits/localefwd.h: Same.
* include/bits/functional_hash.h: Same.
* include/bits/locale_classes.h: Same.
* include/bits/locale_facets.h: Same.
* include/bits/ios_base.h: Same.
* include/bits/codecvt.h: Same.
* include/bits/locale_facets_nonio.h: Same.
* testsuite/27_io/ios_base/cons/assign_neg.cc: Adjust line numbers.
* testsuite/27_io/ios_base/cons/copy_neg.cc: Same.
From-SVN: r156644
Jerry DeLisle [Wed, 10 Feb 2010 03:34:33 +0000 (03:34 +0000)]
re PR fortran/42999 (bogus error: Parameter 'i' at (1) has not been declared or is a variable, which does not reduce to a constant expression)
2010-02-09 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/42999
* gfortran.dg/array_constructor_35.f90: New test.
From-SVN: r156643
Jerry DeLisle [Wed, 10 Feb 2010 03:31:02 +0000 (03:31 +0000)]
re PR fortran/42999 (bogus error: Parameter 'i' at (1) has not been declared or is a variable, which does not reduce to a constant expression)
2010-02-09 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/42999
* array.c (gfc_constant_ac): Do not prevent expansion of constructors
with iterators.
From-SVN: r156642
Vladimir Makarov [Wed, 10 Feb 2010 02:07:22 +0000 (02:07 +0000)]
re PR middle-end/42973 (IRA apparently systematically making reload too busy on 2 address instructions with 3 operands)
2010-02-09 Vladimir Makarov <vmakarov@redhat.com>
PR middle-end/42973
* ira-conflicts.c (get_dup): Remove.
(process_reg_shuffles): Add new parameter. Use it as an
additional guard for copy generation.
(add_insn_allocno_copies): Rewrite.
From-SVN: r156641
GCC Administrator [Wed, 10 Feb 2010 00:17:01 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r156640
Kaveh R. Ghazi [Tue, 9 Feb 2010 23:50:48 +0000 (23:50 +0000)]
inline-4.c: Bind pic locally.
* gcc.dg/tree-ssa/inline-4.c: Bind pic locally.
From-SVN: r156636
Jason Merrill [Tue, 9 Feb 2010 22:06:23 +0000 (17:06 -0500)]
re PR c++/42399 ([c++0x] assigning lambda causes segfault from null dereference at libiberty/splay-tree.c:148)
PR c++/42399
* pt.c (tsubst_copy_and_build): Propagate LAMBDA_EXPR_LOCATION.
From-SVN: r156635
Jason Merrill [Tue, 9 Feb 2010 20:05:51 +0000 (15:05 -0500)]
re PR c++/42370 ([C++0x][lambda] in a void function: "Warning: control reaches end of non-void function")
PR c++/42370
* decl2.c (change_return_type): New fn.
* semantics.c (apply_lambda_return_type): Use it.
* cp-tree.h: Declare it.
From-SVN: r156634
Jakub Jelinek [Tue, 9 Feb 2010 18:42:17 +0000 (19:42 +0100)]
* module.c (fix_mio_expr): Declare sym.
From-SVN: r156632
Tobias Burnus [Tue, 9 Feb 2010 17:32:53 +0000 (18:32 +0100)]
[multiple changes]
2010-02-09 Paul Thomas <pault@gcc.gnu.org>
PR fortran/41869
* module.c (fix_mio_expr): Fix for private generic procedures.
2010-02-09 Tobias Burnus <burnus@net-b.de>
PR fortran/41869
* gfortran.dg/module_write_1.f90: New test.
From-SVN: r156631
Tobias Burnus [Tue, 9 Feb 2010 17:04:57 +0000 (18:04 +0100)]
re PR libfortran/42996 (Incorrect length returned from get_command_argument intrinsic)
2010-02-09 Tobias Burnus <burnus@net-b.de>
PR fortran/42996
* intrinsics/args.c (get_command_argument_i4): Always return
commandline-argument length for length parameter.
From-SVN: r156630
Alexander Monakov [Tue, 9 Feb 2010 16:51:30 +0000 (19:51 +0300)]
common.opt (fsched2-use-traces): Preserved for backward compatibility.
2010-02-09 Alexander Monakov <amonakov@ispras.ru>
gcc/ChangeLog:
* common.opt (fsched2-use-traces): Preserved for backward
compatibility.
* doc/invoke.texi: Remove the documentation about option
-fsched2-use-traces.
* sched-rgn.c (rest_of_handle_sched2): Remove usage of
flag_sched2_use_traces.
* opts.c (common_handle_option): Add OPT_fsched2_use_traces to
the backward compatibility flag section.
gcc/testsuite/ChangeLog:
* gcc.dg/pr19340.c: Adjust.
From-SVN: r156629
Richard Guenther [Tue, 9 Feb 2010 16:11:34 +0000 (16:11 +0000)]
re PR tree-optimization/43008 (Attribute malloc not handled correctly)
2010-02-09 Richard Guenther <rguenther@suse.de>
PR tree-optimization/43008
* tree-ssa-structalias.c (handle_lhs_call): Pass in the fndecl,
make HEAP variables initialized from global memory if they
are not known builtin functions.
(find_func_aliases): Adjust.
* gcc.c-torture/execute/pr43008.c: New testcase.
From-SVN: r156628
Joseph Myers [Tue, 9 Feb 2010 13:07:30 +0000 (13:07 +0000)]
* vi.po: Update.
From-SVN: r156622
Richard Guenther [Tue, 9 Feb 2010 11:34:28 +0000 (11:34 +0000)]
re PR tree-optimization/43000 (VRP miscompiles python with -fwrapv)
2010-02-09 Richard Guenther <rguenther@suse.de>
PR tree-optimization/43000
* tree-vrp.c (vrp_int_const_binop): Only handle unsigned
arithmetic manually.
* gcc.dg/torture/pr43000.c: New testcase.
* gcc.dg/torture/pr43002.c: Likewise.
From-SVN: r156621
Daniel Kraft [Tue, 9 Feb 2010 10:44:33 +0000 (11:44 +0100)]
re PR fortran/39171 (Misleading warning for negative character length)
2010-02-09 Daniel Kraft <d@domob.eu>
PR fortran/39171
* resolve.c (resolve_charlen): Change warning about negative CHARACTER
length to be correct and issue only with -Wsurprising.
* invoke.texi (Wsurprising): Mention this new warning that is
turned on by -Wsurprising.
2010-02-09 Daniel Kraft <d@domob.eu>
PR fortran/39171
* gfortran.dg/char_length_2.f90: Change warning expectations accordingly
and pass -Wsurprising as necessary.
From-SVN: r156620
Daniel Kraft [Tue, 9 Feb 2010 07:27:47 +0000 (08:27 +0100)]
re PR fortran/41507 (Documentation misleading on MAXLOC/MINLOC array type requirement)
2010-02-09 Daniel Kraft <d@domob.eu>
PR fortran/41507
* intrinsic.texi (MAXVAL): Remove wrong claim that array argument
can be CHARACTER type.
(MINVAL), (MAXLOC), (MINLOC): Ditto.
From-SVN: r156618
Benjamin Kosnik [Tue, 9 Feb 2010 04:49:49 +0000 (04:49 +0000)]
[multiple changes]
2010-02-08 Benjamin Kosnik <bkoz@redhat.com>
PR libstdc++/42460
* include/tr1_impl/regex: Fix quoting issues in doxygen markup.
* include/bits/random.h: Fix multi-line doxygen function markup.
2010-02-08 Matthias Klose <doko@debian.org>
PR libstdc++/42460
* include/std/istream: Fix '\' quoting in doxygen markup.
From-SVN: r156617
Mike Stump [Tue, 9 Feb 2010 00:57:18 +0000 (00:57 +0000)]
Update.
From-SVN: r156616
GCC Administrator [Tue, 9 Feb 2010 00:17:10 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r156615
Andrew Pinski [Mon, 8 Feb 2010 23:37:19 +0000 (15:37 -0800)]
MAINTAINERS (spu port): Update my email address.
2010-02-08 Andrew Pinski <pinskia@gmail.com>
* MAINTAINERS (spu port): Update my email address.
From-SVN: r156611
Ed Smith-Rowland [Mon, 8 Feb 2010 17:42:12 +0000 (17:42 +0000)]
random.tcc (uniform_int_distribution<>:: operator()(_UniformRandomNumberGenerator&, const param_type&)): Use make_unsigned instead of __add_unsigned and conditional instead of...
2010-02-08 Ed Smith-Rowland <3dw4rd@verizon.net>
* include/bits/random.tcc (uniform_int_distribution<>::
operator()(_UniformRandomNumberGenerator&, const param_type&)):
Use make_unsigned instead of __add_unsigned and conditional
instead of __conditional_type.
* include/std/random: Do not include <ext/type_traits.h> and
<ext/numeric_traits.h>.
From-SVN: r156608
Jakub Jelinek [Mon, 8 Feb 2010 15:49:30 +0000 (16:49 +0100)]
re PR tree-optimization/42890 (Crash in type_like_member_ptr_p in ipa-prop.c:382)
PR tree-optimization/42890
* g++.dg/torture/pr42890.C: New test.
From-SVN: r156604
Jakub Jelinek [Mon, 8 Feb 2010 15:47:01 +0000 (16:47 +0100)]
re PR tree-optimization/42931 (-ftree-loop-linear - valgrind warns about using uninitialized variable)
PR tree-optimization/42931
* tree-loop-linear.c (try_interchange_loops): Don't call
double_int_mul if estimated_loop_iterations failed.
From-SVN: r156602
Martin Jambor [Mon, 8 Feb 2010 13:24:12 +0000 (14:24 +0100)]
re PR middle-end/42898 (volatile structures and compound literal initializers)
2010-02-08 Martin Jambor <mjambor@suse.cz>
PR middle-end/42898
* tree-sra.c (build_accesses_from_assign): Do not mark in
should_scalarize_away_bitmap if stmt has volatile ops.
(sra_modify_assign): Do not process assigns piecemeal if if stmt
has volatile ops.
From-SVN: r156599
Richard Guenther [Mon, 8 Feb 2010 10:41:25 +0000 (10:41 +0000)]
re PR middle-end/42995 (inline-small-functions does not inline simple delegation calls)
2010-02-08 Richard Guenther <rguenther@suse.de>
PR middle-end/42995
* gcc.dg/tree-ssa/inline-4.c: New testcase.
From-SVN: r156598
Joern Rennecke [Mon, 8 Feb 2010 05:48:05 +0000 (05:48 +0000)]
* doc/tm.texi (TARGET_UNWIND_WORD_MODE): Document.
From-SVN: r156595
Adam Nemet [Mon, 8 Feb 2010 04:54:12 +0000 (04:54 +0000)]
mips.md (*<optab>_trunc<mode>_exts): Fix comment before the pattern.
* config/mips/mips.md (*<optab>_trunc<mode>_exts): Fix comment
before the pattern.
From-SVN: r156594
Jie Zhang [Mon, 8 Feb 2010 02:59:44 +0000 (02:59 +0000)]
* MAINTAINERS: Add myself as a maintainer for the bfin port.
From-SVN: r156593
GCC Administrator [Mon, 8 Feb 2010 00:20:21 +0000 (00:20 +0000)]
Daily bump.
From-SVN: r156592
Andrew Pinski [Sun, 7 Feb 2010 23:22:49 +0000 (15:22 -0800)]
re PR rtl-optimization/42946 (bootstrap with --enable-checking=df (or =all) fails)
2010-02-07 Andrew Pinski <pinskia@gmail.com>
PR middle-end/42946
* df-core.c (df_finish_pass): Change type of saved_flags to int.
From-SVN: r156589
Joseph Myers [Sun, 7 Feb 2010 23:02:34 +0000 (23:02 +0000)]
* sv.po: Update.
From-SVN: r156588
Sebastian Pop [Sun, 7 Feb 2010 19:49:35 +0000 (19:49 +0000)]
Add ChangeLog entries.
From-SVN: r156585
Sebastian Pop [Sun, 7 Feb 2010 19:49:26 +0000 (19:49 +0000)]
Call abort for runtime test, always return 0 from main.
2010-02-07 Sebastian Pop <sebastian.pop@amd.com>
* gcc.dg/graphite/block-0.c: Call abort for runtime test. Always
return 0 from main.
* gcc.dg/graphite/block-1.c: Same.
* gcc.dg/graphite/block-3.c: Same.
* gcc.dg/graphite/block-4.c: Same.
* gcc.dg/graphite/block-5.c: Same.
* gcc.dg/graphite/block-6.c: Same.
* gcc.dg/graphite/block-7.c: Same.
* gcc.dg/graphite/interchange-0.c: Same.
* gcc.dg/graphite/interchange-1.c: Same.
* gcc.dg/graphite/interchange-10.c: Same.
* gcc.dg/graphite/interchange-11.c: Same.
* gcc.dg/graphite/interchange-12.c: Same.
* gcc.dg/graphite/interchange-2.c: Same.
* gcc.dg/graphite/interchange-3.c: Same.
* gcc.dg/graphite/interchange-4.c: Same.
* gcc.dg/graphite/interchange-5.c: Same.
* gcc.dg/graphite/interchange-6.c: Same.
* gcc.dg/graphite/interchange-7.c: Same.
* gcc.dg/graphite/interchange-8.c: Same.
* gcc.dg/graphite/interchange-9.c: Same.
* gcc.dg/graphite/interchange-mvt.c: Same.
From-SVN: r156584
Sebastian Pop [Sun, 7 Feb 2010 19:49:15 +0000 (19:49 +0000)]
Add new testcases.
2010-02-07 Sebastian Pop <sebastian.pop@amd.com>
* gfortran.dg/graphite/id-19.f: New.
* gfortran.dg/graphite/pr14741.f90: New.
* gfortran.dg/graphite/pr41924.f90: New.
* gfortran.dg/graphite/run-id-2.f90: New.
From-SVN: r156583
Sebastian Pop [Sun, 7 Feb 2010 19:49:06 +0000 (19:49 +0000)]
Fix PR42988: handle unknown_dependence.
2010-02-07 Sebastian Pop <sebastian.pop@amd.com>
PR middle-end/42988
* graphite-dependences.c (dependence_polyhedron): Set PDDR_KIND
to unknown_dependence.
(graphite_legal_transform_dr): Handle the unknown_dependence.
(graphite_carried_dependence_level_k): Same.
* testsuite/libgomp.graphite/force-parallel-5.c: Un-XFAILed.
* testsuite/libgomp.graphite/pr41118.c: Renamed from pr4118.
From-SVN: r156582
Sebastian Pop [Sun, 7 Feb 2010 19:48:55 +0000 (19:48 +0000)]
Fix format in ChangeLog.graphite
2010-02-07 Sebastian Pop <sebastian.pop@amd.com>
* ChangeLog.graphite: Remove testsuite/ or gcc/testsuite/.
From-SVN: r156581
Paolo Carlini [Sun, 7 Feb 2010 18:36:48 +0000 (18:36 +0000)]
re PR libstdc++/16896 (Use of non-reserved names in stl_list.h)
2010-02-07 Paolo Carlini <paolo.carlini@oracle.com>
PR libstdc++/16896
* include/bits/stl_list.h (_List_node_base::transfer): Rename
to _M_transfer.
(_List_node_base::reverse): Rename to _M_reverse.
(_List_node_base::hook): Rename to _M_hook.
(_List_node_base::unhook): Rename to _M_unhook; adjust callers.
* include/bits/list.tcc: Adjust callers.
* src/list.cc: Adjust.
* src/compatibility.cc: Likewise.
* src/compatibility-list.cc: New.
* src/compatibility-debug_list.cc: Likewise.
* src/compatibility-parallel_list.cc: Likewise.
* src/Makefile.am: Add.
* src/Makefile.in: Regenerate.
* config/abi/pre/gnu.ver: Export _M_* symbols.
* src/hash.cc: Rename to hash-aux.cc.
* src/compatibility-ldbl.cc: Adjust.
* src/compatibility-c++0x.cc: Likewise.
* src/hash_tr1.cc: Likewise.
* src/hashtable.cc: Rename to hashtable-aux.cc.
* src/hashtable_c++0x.cc: Adjust.
* src/hashtable_tr1.cc: Likewise.
* src/limits_c++0x.cc: Fold...
* src/limits.cc... here.
From-SVN: r156578
Joseph Myers [Sun, 7 Feb 2010 18:03:42 +0000 (18:03 +0000)]
* es.po: Update.
From-SVN: r156577
Joseph Myers [Sun, 7 Feb 2010 18:02:05 +0000 (18:02 +0000)]
* es.po, nl.po: Update.
From-SVN: r156576
Richard Guenther [Sun, 7 Feb 2010 17:18:22 +0000 (17:18 +0000)]
re PR debug/42991 (cfgexpand.c:2487: Conditional jump or move depends on uninitialised value(s))
2010-02-07 Richard Guenther <rguenther@suse.de>
PR middle-end/42991
* expr.c (get_inner_reference): Always initialize *pbitsize.
From-SVN: r156575
Richard Guenther [Sun, 7 Feb 2010 13:42:52 +0000 (13:42 +0000)]
re PR tree-optimization/42956 (internal compiler error: Segmentation fault with -O1)
2010-02-07 Richard Guenther <rguenther@suse.de>
PR middle-end/42956
* gimplify.c (gimple_fold_indirect_ref): Avoid generating
new ARRAY_REFs on variable size element or minimal index arrays.
Complete.
* tree-ssa-loop-ivopts.c (find_interesting_uses_address): Use
gimple_fold_indirect_ref.
* gcc.c-torture/compile/pr42956.c: New testcase.
From-SVN: r156571
Paolo Carlini [Sun, 7 Feb 2010 10:29:04 +0000 (10:29 +0000)]
stl_pair.h (pair<>::pair(_U1&&, const _T2&), [...]): Revert to std::is_convertible, basing on DR 811.
2010-02-07 Paolo Carlini <paolo.carlini@oracle.com>
* include/bits/stl_pair.h (pair<>::pair(_U1&&, const _T2&),
pair<>::pair(const _T1&, _U2&&), pair<>::pair(_U1&&, _U2&&)):
Revert to std::is_convertible, basing on DR 811.
From-SVN: r156570
Jerry DeLisle [Sun, 7 Feb 2010 07:50:17 +0000 (07:50 +0000)]
re PR libfortran/42742 (Handle very large format strings correctly)
2010-02-06 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libfortran/42742
* gfortran.dg/fmt_cache_2.f: New test.
From-SVN: r156569
Jerry DeLisle [Sun, 7 Feb 2010 07:45:55 +0000 (07:45 +0000)]
re PR libfortran/42742 (Handle very large format strings correctly)
2010-02-06 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libfortran/42742
* io/format.c (reset_fnode_counters): Use the correct pointer to the
head of the fnode list. (parse_format): Remove previous hack that set
limit on size of format string for caching.
From-SVN: r156568
H.J. Lu [Sun, 7 Feb 2010 04:07:05 +0000 (20:07 -0800)]
Remove empy entry.
From-SVN: r156561
GCC Administrator [Sun, 7 Feb 2010 00:16:47 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r156560
Paolo Carlini [Sat, 6 Feb 2010 21:30:52 +0000 (21:30 +0000)]
re PR libstdc++/42989 (operator= for rvalue reference for wstring disappears in libstdc++.so.6.0.14)
2010-02-06 Paolo Carlini <paolo.carlini@oracle.com>
PR libstdc++/42989
* config/abi/pre/gnu.ver: Fix wstring move assignment operator typo.
From-SVN: r156556
Paolo Carlini [Sat, 6 Feb 2010 20:41:09 +0000 (20:41 +0000)]
re PR libstdc++/27340 (valarray uses __cos which may conflict with libm functions)
2010-02-06 Paolo Carlini <paolo.carlini@oracle.com>
PR libstdc++/27340
* include/bits/valarray_before.h: Uglify the structs wrapping
cmath functions differently, eg, _Cos instead of __cos.
* include/bits/valarray_after.h: Adjust.
From-SVN: r156555
Jerry DeLisle [Sat, 6 Feb 2010 18:34:13 +0000 (18:34 +0000)]
read_no_eor.f90: New test.
2010-02-06 Jerry DeLisle <jvdelisle@gcc.gnu.org>
* gfortran.dg/read_no_eor.f90: New test.
From-SVN: r156553
Sebastian Pop [Sat, 6 Feb 2010 17:41:54 +0000 (17:41 +0000)]
re PR tree-optimization/42637 ([graphite] wrong code for -floop-interchange -ftree-loop-distribution)
Fix PR42637.
2010-02-05 Sebastian Pop <sebastian.pop@amd.com>
Konrad Trifunovic <konrad.trifunovic@inria.fr>
PR middle-end/42637
* graphite-dependences.c (build_lexicographical_constraint): Return
a union of dependence polyhedra.
(dependence_polyhedron_1): Adapt for build_lexicographical_constraint.
* testsuite/gcc.dg/graphite/block-0.c: Enable runtime check. XFAILed.
* testsuite/gcc.dg/graphite/block-4.c: Same.
* testsuite/gcc.dg/graphite/block-7.c: Same.
* testsuite/gcc.dg/graphite/interchange-12.c: Same.
* testsuite/gcc.dg/graphite/interchange-mvt.c: Same.
* testsuite/gfortran.dg/graphite/interchange-1.f: XFAILed.
* testsuite/gfortran.dg/graphite/interchange-3.f90: XFAILed.
* testsuite/gfortran.dg/graphite/run-id-1.f: New testcase for PR42637.
Co-Authored-By: Konrad Trifunovic <konrad.trifunovic@inria.fr>
From-SVN: r156552
Sebastian Pop [Sat, 6 Feb 2010 17:41:47 +0000 (17:41 +0000)]
Do not FAIL in miscompiled runtime tests.
2010-02-03 Sebastian Pop <sebastian.pop@amd.com>
* testsuite/gcc.dg/graphite/interchange-12.c: Return 0 to avoid
the execution check to FAIL. This testcase is miscompiled by Graphite
and the check should be re-enabled.
* testsuite/gcc.dg/graphite/interchange-mvt.c: Same.
* testsuite/gcc.dg/graphite/block-0.c: Same.
* testsuite/gcc.dg/graphite/block-4.c: Same.
* testsuite/gcc.dg/graphite/block-7.c: Same.
From-SVN: r156551
Sebastian Pop [Sat, 6 Feb 2010 17:41:41 +0000 (17:41 +0000)]
Fix testsuite.
2010-02-03 Sebastian Pop <sebastian.pop@amd.com>
* testsuite/gcc.dg/graphite/interchange-mvt.c: Un-XFAILed.
* testsuite/gcc.dg/graphite/scop-0.c: Fix spaces.
* testsuite/gfortran.dg/graphite/block-1.f90: Rewrite expected output.
* testsuite/gfortran.dg/graphite/block-2.f: Same.
* testsuite/gfortran.dg/graphite/block-3.f90: Same.
* testsuite/gfortran.dg/graphite/block-4.f90: Same.
* testsuite/gfortran.dg/graphite/interchange-1.f: Un-XFAILed.
* testsuite/gfortran.dg/graphite/interchange-3.f90: Un-XFAILed.
From-SVN: r156550
Sebastian Pop [Sat, 6 Feb 2010 17:41:34 +0000 (17:41 +0000)]
Correct the lexicographical function, add debug_pddr.
2010-02-03 Sebastian Pop <sebastian.pop@amd.com>
* graphite-dependences.c (new_poly_ddr): Initialize PDDR_KIND and
PDDR_ORIGINAL_SCATTERING_P.
(pddr_is_empty): Rewritten.
(print_dependence_polyhedron_layout): New.
(print_pddr): New.
(debug_pddr): New.
(build_alias_set_powerset): Moved up.
(poly_drs_may_alias_p): Moved up.
(lexicographically_gt_p): Removed.
(build_lexicographically_gt_constraint): Renamed
build_lexicographical_constraint, reimplemented.
(dependence_polyhedron_1): Simplified parameters.
(dependence_polyhedron): Same. Keep along empty PDDRs
to make efficient use of the cache.
(reduction_dr_p): Compute the PBBs don't pass them as parameters.
(graphite_legal_transform_dr): Rewritten.
(graphite_legal_transform_bb): Pass fewer arguments to
graphite_legal_transform_dr.
(graphite_carried_dependence_level_k): Simplified implementation.
(dot_original_deps_stmt_1): Call dependence_polyhedron.
(dot_transformed_deps_stmt_1): Same.
(dot_original_deps): Same.
(dot_transformed_deps): Same.
* graphite-dependences.h (struct poly_ddr): Added a new field
original_scattering_p.
(PDDR_ORIGINAL_SCATTERING_P): New.
(print_pddr): Declared.
(debug_pddr): Declared.
* graphite-ppl.c: Clarify comment.
* sese.h (print_gimple_bb): Removed unused declaration.
(debug_gbb): Same.
From-SVN: r156549
Sebastian Pop [Sat, 6 Feb 2010 17:41:28 +0000 (17:41 +0000)]
Add runtime tests for interchange and blocking.
2010-02-03 Sebastian Pop <sebastian.pop@amd.com>
* testsuite/g++.dg/graphite/graphite.exp: Rewritten.
* testsuite/gcc.dg/graphite/graphite.exp: Rewritten.
* testsuite/gfortran.dg/graphite/graphite.exp: Rewritten.
* gcc/testsuite/gcc.dg/graphite/block-0.c: Added runtime test.
* testsuite/gcc.dg/graphite/block-1.c: Same.
* testsuite/gcc.dg/graphite/block-3.c: Same.
* testsuite/gcc.dg/graphite/block-4.c: Same.
* testsuite/gcc.dg/graphite/block-5.c: Same.
* testsuite/gcc.dg/graphite/block-6.c: Same.
* testsuite/gcc.dg/graphite/block-7.c: Same.
* testsuite/gcc.dg/graphite/interchange-0.c: Same.
* testsuite/gcc.dg/graphite/interchange-1.c: Same.
* testsuite/gcc.dg/graphite/interchange-10.c: Same.
* testsuite/gcc.dg/graphite/interchange-11.c: Same.
* testsuite/gcc.dg/graphite/interchange-12.c: Same.
* testsuite/gcc.dg/graphite/interchange-2.c: Same.
* testsuite/gcc.dg/graphite/interchange-3.c: Same.
* testsuite/gcc.dg/graphite/interchange-4.c: Same.
* testsuite/gcc.dg/graphite/interchange-5.c: Same.
* testsuite/gcc.dg/graphite/interchange-6.c: Same.
* testsuite/gcc.dg/graphite/interchange-7.c: Same.
* testsuite/gcc.dg/graphite/interchange-8.c: Same.
* testsuite/gcc.dg/graphite/interchange-9.c: Same.
* testsuite/gcc.dg/graphite/interchange-mvt.c: Same.
* testsuite/gcc.dg/graphite/run-id-2.c: Remove useless dg-do run.
From-SVN: r156548
Ramakrishna Upadrasta [Sat, 6 Feb 2010 17:41:22 +0000 (18:41 +0100)]
Make calls to dot run in background.
2010-01-22 Ramakrishna Upadrasta <Ramakrishna.Upadrasta@inria.fr>
* graphite-scop-detection.c (dot_all_scops): Make calls to dot run in
background.
From-SVN: r156547
Sebastian Pop [Sat, 6 Feb 2010 17:41:15 +0000 (17:41 +0000)]
Add more testcases.
2010-01-20 Sebastian Pop <sebastian.pop@amd.com>
* testsuite/gcc.dg/graphite/id-17.c: New.
* testsuite/gcc.dg/graphite/pr41888.c: New.
* testsuite/gcc.dg/graphite/scop-22.c: New.
* testsuite/gfortran.dg/graphite/pr40982.f90: New.
From-SVN: r156546
Sebastian Pop [Sat, 6 Feb 2010 17:41:09 +0000 (17:41 +0000)]
Clarify comments and use build_pddr.
2010-01-20 Sebastian Pop <sebastian.pop@amd.com>
* graphite-dependences.c: Clarify comments.
(build_pddr): Add an extra parameter direction.
(graphite_legal_transform_dr): Use build_pddr.
(dot_original_deps_stmt_1): Update use of build_pddr.
(dot_transformed_deps_stmt_1): Same.
(dot_original_deps): Same.
(dot_transformed_deps): Same.
* testsuite/gcc.dg/graphite/interchange-12.c: New.
* gcc/testsuite/gcc.dg/graphite/block-7.c: New.
From-SVN: r156545
Sebastian Pop [Sat, 6 Feb 2010 17:41:03 +0000 (17:41 +0000)]
Fix build_pddr.
2010-01-20 Sebastian Pop <sebastian.pop@amd.com>
* graphite-dependences.c (pddr_original_scattering): Renamed build_pddr.
(pddr_transformed_scattering): Removed.
(graphite_legal_transform_dr): Use build_pddr.
(dot_original_deps_stmt_1): Same.
(dot_transformed_deps_stmt_1): Same.
(dot_original_deps): Same.
(dot_transformed_deps): Same.
From-SVN: r156544
Sebastian Pop [Sat, 6 Feb 2010 17:40:58 +0000 (17:40 +0000)]
Add missing function comments.
2010-01-20 Sebastian Pop <sebastian.pop@amd.com>
* graphite-dependences.c (lexicographically_gt_p): Add comments.
(build_lexicographically_gt_constraint): Same.
(dependence_polyhedron_1): Same.
(dependence_polyhedron): Same.
From-SVN: r156543
Sebastian Pop [Sat, 6 Feb 2010 17:40:52 +0000 (17:40 +0000)]
Factor uses of build_pairwise_scheduling.
2010-01-20 Sebastian Pop <sebastian.pop@amd.com>
* graphite-dependences.c (build_pairwise_scheduling_equality): Removed.
(build_pairwise_scheduling_inequality): Renamed build_pairwise_scheduling.
(lexicographically_gt_p): Change type of direction to int.
(build_lexicographically_gt_constraint): Same.
(dependence_polyhedron_1): Same.
(dependence_polyhedron): Same.
(pddr_original_scattering): Use integers for direction.
(pddr_transformed_scattering): Same.
(graphite_legal_transform_dr): Same.
(graphite_carried_dependence_level_k): Same.
From-SVN: r156542
Sebastian Pop [Sat, 6 Feb 2010 17:40:45 +0000 (17:40 +0000)]
Cleanup build relation.
2010-01-20 Sebastian Pop <sebastian.pop@amd.com>
* graphite-dependences.c (build_pairwise_constraint): Renamed
ppl_build_relation. Moved...
(dr_equality_constraints): Use ppl_build_relation.
(build_pairwise_scheduling_equality): Same.
(build_pairwise_scheduling_inequality): Same.
* graphite-ppl.c (ppl_build_relation): ...here.
* graphite-ppl.h (ppl_build_relation): Declared.
From-SVN: r156541
Jerry DeLisle [Sat, 6 Feb 2010 14:39:51 +0000 (14:39 +0000)]
PR libfortran
2010-02-06 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libfortran
* io/transfer.c (read_sf): Handle EOR and EOF conditions for
ADVANCE="no" with PAD="yes" or PAD="no".
From-SVN: r156540
Richard Earnshaw [Sat, 6 Feb 2010 14:05:27 +0000 (14:05 +0000)]
re PR target/42957 (ARM: Segfault when invalid -mfpu argument is specified.)
PR target/42957
* arm.c (arm_override_options): Just return if the user has specified
an invalid fpu name.
From-SVN: r156539