+2010-05-15 Richard Guenther <rguenther@suse.de>
+
+ PR tree-optimization/44038
+ * tree-ssa-forwprop.c (forward_propagate_addr_expr_1): Avoid
+ taking the address of a V_C_E of a constant.
+
+2010-05-14 Jan Hubicka <jh@suse.cz>
+
+ * tree.h (memory_identifier_string): Remove.
+ * ipa-reference.c: Update comment; do not include gt-ipa-reference.h
+ (ipa_reference_global_vars_info_d): Remove statics_not_read and
+ statics_not_written.
+ (ipa_reference_optimization_summary_d): New structure.
+ (ipa_reference_optimization_summary_t): New type and vector.
+ (ipa_reference_vars_info_d): Embedd structures instead of using
+ pointers.
+ (reference_vars_to_consider): Remove out of GGC space.
+ (module_statics_escape): Remove.
+ (global_info_obstack): Rename to ...
+ (optimization_summary_obstack): ... this one.
+ (initialization_status_t): Remove.
+ (memory_identifier_string): Remove.
+ (get_reference_vars_info): Fix indenting.
+ (set_reference_vars_info): Likewise.
+ (get_reference_optimization_summary): New.
+ (set_reference_optimization_summary): New.
+ (get_global_reference_vars_info): Remove.
+ (ipa_reference_get_read_global): Remove.
+ (ipa_reference_get_written_global): Remove.
+ (ipa_reference_get_not_read_global): Update.
+ (ipa_reference_get_not_written_global): Update.
+ (is_proper_for_analysis): Outlaw addressable.
+ (propagate_bits): Update for new datastructures.
+ (analyze_variable): Remove.
+ (init_function_info): Update for new datastructures.
+ (clean_function_local_data): Remove.
+ (clean_function): Remove.
+ (copy_global_bitmap): Use optimizations_summary_obstack.
+ (duplicate_node_data): Duplicate optimization summary only.
+ (remove_node_data): Remove optimization summary only.
+ (generate_summary): Do not analyze variables; do not compute
+ module_statics_escape; do not prune solutions by it.
+ (read_write_all_from_decl): Fix typos in comments.
+ (propagate): Doscover readonly and nonaddressable first;
+ update for new datastructures; share global bitmaps.
+ * ipa-reference.h (ipa_reference_get_read_global,
+ ipa_reference_get_written_global): Remove.
+ * ipa-pure-const.c (check_stmt): Do not use memory_identifier_string.
+ * Makefile.in: Remove ipa-refereference from GT files.
+
+2010-05-14 Jakub Jelinek <jakub@redhat.com>
+
+ PR debug/44112
+ * dwarf2out.c (resolve_one_addr): Check TREE_ASM_WRITTEN
+ for all SYMBOL_REF_DECLs.
+
+2010-05-14 Jan Hubicka <jh@suse.cz>
+
+ * cgraph.h (ipa_discover_readonly_nonaddressable_vars): Declare.
+ (varpool_all_refs_explicit_p): New inline function.
+ * ipa-reference.c: Update comment.
+ (module_statics_written): Remove.
+ (get_static_decl): Remove.
+ (ipa_init): Do not initialize module_statics_written.
+ (analyze_function): Likewise.
+ (generate_summary): Likewise; do not compute module_statics_readonly
+ and do not update variable flags.
+ (propagate): Call ipa_discover_readonly_nonaddressable_vars.
+ * ipa.c: Inlucde flags.h
+ (cgraph_local_node_p): New.
+ (cgraph_remove_unreachable_nodes): Return early when not optimizing;
+ promote functions to local.
+ (ipa_discover_readonly_nonaddressable_vars): New function.
+ (function_and_variable_visibility): Use cgraph_local_node_p.
+ * varpool.c (varpool_finalize_decl): Set force_output for
+ DECL_PRESERVE_P vars.
+
+2010-05-14 Jan Hubicka <jh@suse.cz>
+
+ * ipa.c (cgraph_remove_unreachable_nodes): Revert accidental commit.
+
+2010-05-14 Richard Guenther <rguenther@suse.de>
+
+ PR tree-optimization/44119
+ * tree-ssa-pre.c (eliminate): Properly mark replacement of
+ a PHI node necessary.
+
+2010-05-14 Eric Botcazou <ebotcazou@adacore.com>
+
+ * tree.h (TREE_ADDRESSABLE): Remove bogus usage for FIELD_DECL.
+
+2010-05-14 Jason Merrill <jason@redhat.com>
+
+ PR c++/44127
+ * gimple.h (enum gf_mask): Add GF_CALL_NOTHROW.
+ (gimple_call_set_nothrow): New.
+ * gimple.c (gimple_build_call_from_tree): Call it.
+ (gimple_call_flags): Set ECF_NOTHROW from GF_CALL_NOTHROW.
+
+ PR c++/44127
+ * gimplify.c (gimplify_seq_add_stmt): No longer static.
+ * gimple.h: Declare it.
+ * gimple.c (gimple_build_eh_filter): No ops.
+
+2010-05-14 Jan Hubicka <jh@suse.cz>
+
+ * ipa.c (enqueue_cgraph_node): Update comment; do not re-enqueue
+ nodes already in queue.
+ (cgraph_remove_unreachable_nodes): Cleanup; fix problem with
+ re-enqueueing node.
+
+2010-05-14 Jakub Jelinek <jakub@redhat.com>
+
+ PR debug/44136
+ * cfgexpand.c (expand_debug_expr): If non-memory op0
+ has BLKmode, return NULL.
+
+2010-05-14 Harsha Jagasia <harsha.jagasia@amd.com>
+
+ * config.gcc: Add support for --with-cpu option for bdver1.
+ * config/i386/i386.h (TARGET_BDVER1): New macro.
+ (ix86_tune_indices): Change SSE_UNALIGNED_MOVE_OPTIMAL
+ to SSE_UNALIGNED_LOAD_OPTIMAL. Add SSE_UNALIGNED_STORE_OPTIMAL.
+ (ix86_tune_features) :Change SSE_UNALIGNED_MOVE_OPTIMAL
+ to SSE_UNALIGNED_LOAD_OPTIMAL. Add SSE_UNALIGNED_STORE_OPTIMAL.
+ Add SSE_PACKED_SINGLE_INSN_OPTIMAL.
+ (TARGET_CPU_DEFAULT_NAMES): Add bdver1.
+ (processor_type): Add PROCESSOR_BDVER1.
+ * config/i386/i386.md: Add bdver1 as a new cpu attribute to match
+ processor_type in config/i386/i386.h.
+ Add check for TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL to emit
+ movaps <reg, reg> instead of movapd <reg, reg> when replacing
+ movsd <reg, reg> or movss <reg, reg> for SSE and AVX.
+ Add check for TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL
+ to emit packed xor instead of packed double/packed integer
+ xor for SSE and AVX when moving a zero value.
+ * config/i386/sse.md: Add check for
+ TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL to emit movaps instead of
+ movapd/movdqa for SSE and AVX.
+ Add check for TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL to emit packed
+ single logical operations i.e and, or and xor instead of packed double
+ logical operations for SSE and AVX.
+ * config/i386/i386-c.c:
+ (ix86_target_macros_internal): Add PROCESSOR_BDVER1.
+ * config/i386/driver-i386.c: Turn on -mtune=native for BDVER1.
+ (has_fma4, has_xop): New.
+ * config/i386/i386.c (bdver1_cost): New variable.
+ (m_BDVER1): New macro.
+ (m_AMD_MULTIPLE): Add m_BDVER1.
+ (x86_tune_use_leave, x86_tune_push_memory, x86_tune_unroll_strlen,
+ x86_tune_deep_branch_prediction, x86_tune_use_sahf, x86_tune_movx,
+ x86_tune_use_simode_fiop, x86_tune_promote_qimode,
+ x86_tune_add_esp_8, x86_tune_tune_sub_esp_4, x86_tune_sub_esp_8,
+ x86_tune_integer_dfmode_moves, x86_tune_partial_reg_dependency,
+ x86_tune_sse_partial_reg_dependency,
+ x86_tune_sse_unaligned_load_optimal,
+ x86_tune_sse_unaligned_store_optimal, x86_tune_sse_typeless_stores,
+ x86_tune_memory_mismatch_stall, x86_tune_use_ffreep,
+ x86_tune_inter_unit_moves, x86_tune_inter_unit_conversions,
+ x86_tune_use_bt, x86_tune_pad_returns, x86_tune_slow_imul_imm32_mem,
+ x86_tune_slow_imul_imm8, x86_tune_fuse_cmp_and_branch):
+ Enable/disable for bdver1.
+ (processor_target_table): Add bdver1_cost.
+ (cpu_names): Add bdver1.
+ (override_options): Set up PROCESSOR_BDVER1 for bdver1 entry in
+ processor_alias_table.
+ (ix86_expand_vector_move_misalign): Change.
+ TARGET_SSE_UNALIGNED_MOVE_OPTIMAL to TARGET_SSE_UNALIGNED_LOAD_OPTIMAL.
+ Check for TARGET_SSE_UNALIGNED_STORE_OPTIMAL.
+ Check for TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL to emit movups instead
+ of movupd/movdqu for SSE and AVX.
+ (ix86_tune_issue_rate): Add PROCESSOR_BDVER1.
+ (ix86_tune_adjust_cost): Add code for bdver1.
+ (standard_sse_constant_opcode): Add check for
+ TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL to emit packed single xor instead
+ of packed double xor for SSE and AVX.
+
+2010-05-14 Pat Haugen <pthaugen@us.ibm.com>
+
+ * tree-ssa-loop.prefetch.c (prune_ref_by_group_reuse): Cast abs()
+ result to unsigned.
+
+2010-05-14 Tristan Gingold <gingold@adacore.com>
+
+ * toplev.c (default_debug_hooks): Remove this variable.
+ (process_options): Remove assignments to default_debug_hooks.
+
+2010-05-14 Martin Jambor <mjambor@suse.cz>
+
+ * langhooks-def.h (LANG_HOOKS_FOLD_OBJ_TYPE_REF): Remove.
+ (LANG_HOOKS_INITIALIZER): Remove LANG_HOOKS_FOLD_OBJ_TYPE_REF.
+ * langhooks.h (struct lang_hooks_for_decls): Removed field
+ fold_obj_type_ref.
+ * tree.c (free_lang_data): Remove assignment to
+ lang_hooks.fold_obj_type_ref.
+ * tree.def (OBJ_TYPE_REF): Update comment.
+
+2010-05-14 Richard Guenther <rguenther@suse.de>
+
+ PR tree-optimization/44124
+ * tree-ssa-sccvn.c (vn_nary_may_trap): Fix invalid memory access.
+
+2010-05-14 Alan Modra <amodra@gmail.com>
+
+ PR target/44075
+ * config/rs6000/rs6000.c (struct machine_function): Reorder
+ fields for better packing. Add lr_save_state.
+ (rs6000_ra_ever_killed): Return lr_save_state if set.
+ (rs6000_emit_eh_reg_restore): Set lr_save_state.
+
+2010-05-13 Jan Hubicka <jh@suse.cz>
+
+ * varpool.c (decide_is_variable_needed): Drop code checking
+ TREE_SYMBOL_REFERENCED.
+
+2010-05-13 Jan Hubicka <jh@suse.cz>
+
+ * final.c (output_addr_const): Do not call mark_decl_referenced.
+ * cgraphunit.c (process_function_and_variable_attributes): Use
+ mark_needed_node dirrectly.
+ (assemble_thunk): Do not call mark_decl_referenced.
+
+2010-05-13 Anatoly Sokolov <aesok@post.ru>
+
+ * targhooks.c (default_mode_dependent_address_p): Cast 'addr' to rtx.
+
+2010-05-13 Jeff Law <law@redhat.com>
+
+ * ira-conflicts.c (print_allocno_conflicts): New function broken out
+ from...
+ (print_conflicts): Call print_allocno_conflicts.
+
+2010-05-13 Jakub Jelinek <jakub@redhat.com>
+
+ PR debug/44104
+ * dwarf2out.c (modified_type_die): Don't dereference mod_type_die
+ if it is NULL.
+
+2010-05-13 Kai Tietz <kai.tietz@onevision.com>
+
+ * config.gcc (i[34567]86-*-mingw* | x86_64-*-mingw*): Choose
+ t-mingw-w64 or t-mingw-w32 for multilib configuration.
+ * config/i386/t-mingw-w32: New.
+ * config/i386/t-mingw-w64 (MULTILIB_OSDIRNAMES): Change lib64 to lib.
+
+2010-05-13 Martin Jambor <mjambor@suse.cz>
+
+ * gimple.c (gimple_fold_obj_type_ref): Removed (a replacement moved to
+ gimple-fold.c).
+ * gimple-fold.c (get_base_binfo_for_type): New function.
+ (gimple_get_relevant_ref_binfo): Likewise.
+ (gimple_fold_obj_type_ref_known_binfo): Likewise.
+ (gimple_fold_obj_type_ref): Likewise.
+ (fold_gimple_call): Simplify condition for folding virtual calls
+ and call gimple_fold_obj_type_ref.
+ * gimple.h (gimple_get_relevant_ref_binfo): Declare.
+ (gimple_fold_obj_type_ref_known_binfo): Likewise.
+
+2010-05-13 Andreas Schwab <schwab@linux-m68k.org>
+
+ * config/rs6000/rs6000-protos.h
+ (rs6000_mode_dependent_address_ptr): Change argument to const_rtx.
+ * config/rs6000/rs6000.c (rs6000_mode_dependent_address)
+ (rs6000_debug_mode_dependent_address)
+ (rs6000_mode_dependent_address_ptr): Likewise.
+
+2010-05-13 Jakub Jelinek <jakub@redhat.com>
+
+ PR debug/43983
+ * var-tracking.c (track_expr_p): Allow tracking of variables optimized
+ by SRA.
+ * Makefile.in (dwarf2out.o): Depend on $(TREE_FLOW_H).
+ * tree-sra.c (create_access_replacement): Call unshare_expr before
+ passing expr to SET_DECL_DEBUG_EXPR, and remove any SSA_NAMEs from
+ it.
+ * dwarf2out.c: Include tree-flow.h.
+ (struct var_loc_node): Rename var_loc_note field to loc, add comment.
+ (size_of_loc_descr, output_loc_operands, output_loc_operands_raw):
+ Handle DW_OP_bit_piece.
+ (decl_piece_bitsize, decl_piece_varloc_ptr, decl_piece_node,
+ construct_piece_list, adjust_piece_list): New functions.
+ (add_var_loc_to_decl): Handle SRA optimized variables.
+ Adjust for var_loc_note to loc field renaming.
+ (dw_loc_list_1): For WANT_ADDRESS == 2 prefer DECL_MODE of decl
+ in VAR_LOCATION note.
+ (new_loc_descr_op_bit_piece): New function.
+ (dw_sra_loc_expr): New function.
+ (dw_loc_list): Use it. Don't handle the last range after the
+ loop, handle it inside of the loop. Adjust for var_loc_note
+ to loc field renaming.
+ (add_location_or_const_value_attribute): Only special case
+ single entry loc lists if loc is NOTE_P. Adjust for
+ var_loc_note to loc field renaming.
+ (dwarf2out_var_location): Don't set newloc->var_loc_note
+ and newloc->next here.
+
2010-05-12 Jan Hubicka <jh@suse.cz>
- * cgraph.c (cgraph_mark_address_taken_node): No longer imply needed flag.
+ * cgraph.c (cgraph_mark_address_taken_node): No longer imply needed
+ flag.
* cgraph.h (cgraph_only_called_directly_p,
cgraph_can_remove_if_no_direct_calls_p): test address_taken flag.
(cgraph_can_remove_if_no_direct_calls_and_refs_p): New function.
* cgraphunit.c (cgraph_mark_functions_to_output): Test address_taken.
(assemble
* ipa.c (cgraph_remove_unreachable_nodes): Use
- cgraph_can_remove_if_no_direct_calls_and_refs_p; clear address_taken flags.
+ cgraph_can_remove_if_no_direct_calls_and_refs_p; clear address_taken
+ flags.
* tree-inline.c (copy_bb): Check address_taken flag.
* tree-profile.c (tree_gen_ic_func_profiler): Check address_taken and
externally_visible flag.
2010-05-11 Jan Hubicka <jh@suse.cz>
PR tree-optimize/44063
- * ipa-inline.c (cgraph_edge_badness): Move always inlines to top of queue.
+ * ipa-inline.c (cgraph_edge_badness): Move always inlines to top of
+ queue.
(cgraph_decide_inlining_of_small_function): Skip check when disrgarding
limits.
(estimate_function_body_sizes): Compute sizes even when disregarding.
2010-05-11 Christian Borntraeger <borntraeger@de.ibm.com>
- * tree-ssa-loop-prefetch.c (prune_ref_by_group_reuse): Reset
+ * tree-ssa-loop-prefetch.c (prune_ref_by_group_reuse): Reset
prefetch_before to PREFETCH_ALL if to accesses "meet" beyond
cache size.
2010-05-11 Christian Borntraeger <borntraeger@de.ibm.com>
- * tree-ssa-loop-prefetch.c: Add debug for dropped prefetches.
+ * tree-ssa-loop-prefetch.c: Add debug for dropped prefetches.
2010-05-11 Jakub Jelinek <jakub@redhat.com>
2010-05-10 Jan Hubicka <jh@suse.cz>
- * lto-stramer-out.c (produce_asm_for_decls): Correct accidentally commited change.
+ * lto-stramer-out.c (produce_asm_for_decls): Correct accidentally
+ commited change.
2010-05-10 Jan Hubicka <jh@suse.cz>
- * passes.c (ipa_write_summaries_1, ipa_write_optimization_summaries): Allocate
- encoders.
+ * passes.c (ipa_write_summaries_1, ipa_write_optimization_summaries):
+ Allocate encoders.
* lto-section-out.c (lto_new_out_decl_state): Do not allocate it here.
* lto-streamer.c (lto_streamer_cache_insert_1): Use alloc pool.
(lto_streamer_cache_create): Init alloc pool.
2010-04-08 Wolfgang Gellerich <gellerich@de.ibm.com>
- * config/s390/s390.c (override_options): Adjust the z10
- defaults for max-unroll-times, max-completely-peeled-insns
- and max-completely-peel-times.
+ * config/s390/s390.c (override_options): Adjust the z10 defaults
+ for max-unroll-times, max-completely-peeled-insns
+ and max-completely-peel-times.
2010-04-08 Andreas Krebbel <Andreas.Krebbel@de.ibm.com>