* c-cppbuiltin.c (c_cpp_builtins): Change _OPENMP value to

	200805.
	* langhooks.h (struct lang_hooks_for_decls): Add omp_finish_clause.
	Add omp_private_outer_ref hook, add another argument to
	omp_clause_default_ctor hook.
	* langhooks-def.h (LANG_HOOKS_OMP_FINISH_CLAUSE): Define.
	(LANG_HOOKS_OMP_PRIVATE_OUTER_REF): Define.
	(LANG_HOOKS_OMP_CLAUSE_DEFAULT_CTOR): Change to
	hook_tree_tree_tree_tree_null.
	(LANG_HOOKS_DECLS): Add LANG_HOOKS_OMP_FINISH_CLAUSE and
	LANG_HOOKS_OMP_PRIVATE_OUTER_REF.
	* hooks.c (hook_tree_tree_tree_tree_null): New function.
	* hooks.h (hook_tree_tree_tree_tree_null): New prototype.
	* tree.def (OMP_TASK): New tree code.
	* tree.h (OMP_TASK_COPYFN, OMP_TASK_ARG_SIZE, OMP_TASK_ARG_ALIGN,
	OMP_CLAUSE_PRIVATE_OUTER_REF, OMP_CLAUSE_LASTPRIVATE_STMT,
	OMP_CLAUSE_COLLAPSE_ITERVAR, OMP_CLAUSE_COLLAPSE_COUNT,
	OMP_TASKREG_CHECK, OMP_TASKREG_BODY, OMP_TASKREG_CLAUSES,
	OMP_TASKREG_FN, OMP_TASKREG_DATA_ARG, OMP_TASK_BODY,
	OMP_TASK_CLAUSES, OMP_TASK_FN, OMP_TASK_DATA_ARG,
	OMP_CLAUSE_COLLAPSE_EXPR): Define.
	(enum omp_clause_default_kind): Add OMP_CLAUSE_DEFAULT_FIRSTPRIVATE.
	(OMP_DIRECTIVE_P): Add OMP_TASK.
	(OMP_CLAUSE_COLLAPSE, OMP_CLAUSE_UNTIED): New clause codes.
	(OMP_CLAUSE_SCHEDULE_AUTO): New schedule kind.
	* tree.c (omp_clause_code_name): Add OMP_CLAUSE_COLLAPSE
	and OMP_CLAUSE_UNTIED entries.
	(omp_clause_num_ops): Likewise.  Increase OMP_CLAUSE_LASTPRIVATE
	num_ops to 2.
	(walk_tree_1): Handle OMP_CLAUSE_COLLAPSE and OMP_CLAUSE_UNTIED.
	Walk OMP_CLAUSE_LASTPRIVATE_STMT.
	* tree-pretty-print.c (dump_omp_clause): Handle
	OMP_CLAUSE_SCHEDULE_AUTO, OMP_CLAUSE_UNTIED, OMP_CLAUSE_COLLAPSE,
	OMP_CLAUSE_DEFAULT_FIRSTPRIVATE.
	(dump_generic_node): Handle OMP_TASK and collapsed OMP_FOR loops.
	* c-omp.c (c_finish_omp_for): Allow pointer iterators.  Remove
	warning about unsigned iterators.  Change decl/init/cond/incr
	arguments to TREE_VECs, check arguments for all collapsed loops.
	(c_finish_omp_taskwait): New function.
	(c_split_parallel_clauses): Put OMP_CLAUSE_COLLAPSE clause to
	ws_clauses.
	* c-parser.c (c_parser_omp_for_loop): Parse collapsed loops.  Call
	default_function_array_conversion on init.  Add par_clauses argument.
	If decl is present in parallel's lastprivate clause, change it to
	shared and add lastprivate clause for decl to OMP_FOR_CLAUSES.
	Add clauses argument, on success set OMP_FOR_CLAUSES to it.  Look up
	collapse count in clauses.
	(c_parser_omp_for, c_parser_omp_parallel): Adjust
	c_parser_omp_for_loop callers.
	(OMP_FOR_CLAUSE_MASK): Add 1 << PRAGMA_OMP_CLAUSE_COLLAPSE.
	(c_parser_pragma): Handle PRAGMA_OMP_TASKWAIT.
	(c_parser_omp_clause_name): Handle collapse and untied clauses.
	(c_parser_omp_clause_collapse, c_parser_omp_clause_untied): New
	functions.
	(c_parser_omp_clause_schedule): Handle schedule(auto).
	Include correct location in the error message.
	(c_parser_omp_all_clauses): Handle PRAGMA_OMP_CLAUSE_COLLAPSE
	and PRAGMA_OMP_CLAUSE_UNTIED.
	(OMP_TASK_CLAUSE_MASK): Define.
	(c_parser_omp_task, c_parser_omp_taskwait): New functions.
	(c_parser_omp_construct): Handle PRAGMA_OMP_TASK.
	* tree-nested.c (convert_nonlocal_omp_clauses,
	convert_local_omp_clauses): Handle OMP_CLAUSE_LASTPRIVATE_STMT,
	OMP_CLAUSE_REDUCTION_INIT, OMP_CLAUSE_REDUCTION_MERGE,
	OMP_CLAUSE_COLLAPSE and OMP_CLAUSE_UNTIED.
	Don't handle TREE_STATIC or DECL_EXTERNAL VAR_DECLs in
	OMP_CLAUSE_DECL.
	(conver_nonlocal_reference, convert_local_reference,
	convert_call_expr): Handle OMP_TASK the same as OMP_PARALLEL.  Use
	OMP_TASKREG_* macros rather than OMP_PARALLEL_*.
	(walk_omp_for): Adjust for OMP_FOR_{INIT,COND,INCR} changes.
	* tree-gimple.c (is_gimple_stmt): Handle OMP_TASK.
	* c-tree.h (c_begin_omp_task, c_finish_omp_task): New prototypes.
	* c-pragma.h (PRAGMA_OMP_TASK, PRAGMA_OMP_TASKWAIT): New.
	(PRAGMA_OMP_CLAUSE_COLLAPSE, PRAGMA_OMP_CLAUSE_UNTIED): New.
	* c-typeck.c (c_begin_omp_task, c_finish_omp_task): New functions.
	(c_finish_omp_clauses): Handle OMP_CLAUSE_COLLAPSE and
	OMP_CLAUSE_UNTIED.
	* c-pragma.c (init_pragma): Init omp task and omp taskwait pragmas.
	* c-common.h (c_finish_omp_taskwait): New prototype.
	* gimple-low.c (lower_stmt): Handle OMP_TASK.
	* tree-parloops.c (create_parallel_loop): Create 1 entry
	vectors for OMP_FOR_{INIT,COND,INCR}.
	* tree-cfg.c (remove_useless_stmts_1): Handle OMP_* containers.
	(make_edges): Handle OMP_TASK.
	* tree-ssa-operands.c (get_expr_operands): Handle collapsed OMP_FOR
	loops, adjust for OMP_FOR_{INIT,COND,INCR} changes.
	* tree-inline.c (estimate_num_insns_1): Handle OMP_TASK.
	* builtin-types.def (BT_PTR_ULONGLONG, BT_PTR_FN_VOID_PTR_PTR,
	BT_FN_BOOL_ULONGLONGPTR_ULONGLONGPTR,
	BT_FN_BOOL_BOOL_ULL_ULL_ULL_ULLPTR_ULLPTR,
	BT_FN_BOOL_BOOL_ULL_ULL_ULL_ULL_ULLPTR_ULLPTR,
	BT_FN_VOID_OMPFN_PTR_OMPCPYFN_LONG_LONG_BOOL_UINT): New.
	* omp-builtins.def (BUILT_IN_GOMP_TASK, BUILT_IN_GOMP_TASKWAIT,
	BUILT_IN_GOMP_LOOP_ULL_STATIC_START,
	BUILT_IN_GOMP_LOOP_ULL_DYNAMIC_START,
	BUILT_IN_GOMP_LOOP_ULL_GUIDED_START,
	BUILT_IN_GOMP_LOOP_ULL_RUNTIME_START,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_STATIC_START,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_DYNAMIC_START,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_GUIDED_START,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_RUNTIME_START,
	BUILT_IN_GOMP_LOOP_ULL_STATIC_NEXT,
	BUILT_IN_GOMP_LOOP_ULL_DYNAMIC_NEXT,
	BUILT_IN_GOMP_LOOP_ULL_GUIDED_NEXT,
	BUILT_IN_GOMP_LOOP_ULL_RUNTIME_NEXT,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_STATIC_NEXT,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_DYNAMIC_NEXT,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_GUIDED_NEXT,
	BUILT_IN_GOMP_LOOP_ULL_ORDERED_RUNTIME_NEXT): New builtins.
	* gimplify.c (gimplify_omp_for): Allow pointer type for decl,
	handle POINTER_PLUS_EXPR.  If loop counter has been replaced and
	original iterator is present in lastprivate clause or if
	collapse > 1, set OMP_CLAUSE_LASTPRIVATE_STMT.  Handle collapsed
	OMP_FOR loops, adjust for OMP_FOR_{INIT,COND,INCR} changes.
	(gimplify_expr): Handle OMP_SECTIONS_SWITCH and OMP_TASK.
	(enum gimplify_omp_var_data): Add GOVD_PRIVATE_OUTER_REF.
	(omp_notice_variable): Set GOVD_PRIVATE_OUTER_REF if needed,
	if it is set, lookup var in outer contexts too.  Handle
	OMP_CLAUSE_DEFAULT_FIRSTPRIVATE.  Handle vars that are supposed
	to be implicitly determined firstprivate for task regions.
	(gimplify_scan_omp_clauses): Set GOVD_PRIVATE_OUTER_REF if needed,
	if it is set, lookup var in outer contexts too.  Set
	OMP_CLAUSE_PRIVATE_OUTER_REF if GOVD_PRIVATE_OUTER_REF is set.
	Handle OMP_CLAUSE_LASTPRIVATE_STMT, OMP_CLAUSE_COLLAPSE and
	OMP_CLAUSE_UNTIED.  Take region_type as last argument
	instead of in_parallel and in_combined_parallel.
	(gimplify_omp_parallel, gimplify_omp_for, gimplify_omp_workshare):
	Adjust callers.
	(gimplify_adjust_omp_clauses_1): Set OMP_CLAUSE_PRIVATE_OUTER_REF if
	GOVD_PRIVATE_OUTER_REF is set.  Call omp_finish_clause
	langhook.
	(new_omp_context): Set default_kind to
	OMP_CLAUSE_DEFAULT_UNSPECIFIED for OMP_TASK regions.
	(omp_region_type): New enum.
	(struct gimplify_omp_ctx): Remove is_parallel and is_combined_parallel
	fields, add region_type.
	(new_omp_context): Take region_type as argument instead of is_parallel
	and is_combined_parallel.
	(gimple_add_tmp_var, omp_firstprivatize_variable, omp_notice_variable,
	omp_is_private, omp_check_private): Adjust ctx->is_parallel and
	ctx->is_combined_parallel checks.
	(gimplify_omp_task): New function.
	(gimplify_adjust_omp_clauses): Handle OMP_CLAUSE_COLLAPSE and
	OMP_CLAUSE_UNTIED.
	* omp-low.c (extract_omp_for_data): Use schedule(static)
	for schedule(auto).  Handle pointer and unsigned iterators.
	Compute fd->iter_type.  Handle POINTER_PLUS_EXPR increments.
	Add loops argument.  Extract data for collapsed OMP_FOR loops.
	(expand_parallel_call): Assert sched_kind isn't auto,
	map runtime schedule to index 3.
	(struct omp_for_data_loop): New type.
	(struct omp_for_data): Remove v, n1, n2, step, cond_code fields.
	Add loop, loops, collapse and iter_type fields.
	(workshare_safe_to_combine_p): Disallow combined for if
	iter_type is unsigned long long.  Don't combine collapse > 1 loops
	unless all bounds and steps are constant.  Adjust extract_omp_for_data
	caller.
	(expand_omp_for_generic): Handle pointer, unsigned and long long
	iterators.  Handle collapsed OMP_FOR loops.  Adjust
	for struct omp_for_data changes.  If libgomp function doesn't return
	boolean_type_node, add comparison of the return value with 0.
	(expand_omp_for_static_nochunk, expand_omp_for_static_chunk): Handle
	pointer, unsigned and long long iterators.  Adjust for struct
	omp_for_data changes.
	(expand_omp_for): Assert sched_kind isn't auto, map runtime schedule
	to index 3.  Use GOMP_loop_ull*{start,next} if iter_type is
	unsigned long long.  Allocate loops array, pass it to
	extract_omp_for_data.  For collapse > 1 loops use always
	expand_omp_for_generic.
	(omp_context): Add sfield_map and srecord_type fields.
	(is_task_ctx, lookup_sfield): New functions.
	(use_pointer_for_field): Use is_task_ctx helper.  Change first
	argument's type from const_tree to tree.  Clarify comment.
	In OMP_TASK disallow copy-in/out sharing.
	(build_sender_ref): Call lookup_sfield instead of lookup_field.
	(install_var_field): Add mask argument.  Populate both record_type
	and srecord_type if needed.
	(delete_omp_context): Destroy sfield_map, clear DECL_ABSTRACT_ORIGIN
	in srecord_type.
	(fixup_child_record_type): Also remap FIELD_DECL's DECL_SIZE{,_UNIT}
	and DECL_FIELD_OFFSET.
	(scan_sharing_clauses): Adjust install_var_field callers.  For
	firstprivate clauses on explicit tasks allocate the var by value in
	record_type unconditionally, rather than by reference.
	Handle OMP_CLAUSE_PRIVATE_OUTER_REF.  Scan OMP_CLAUSE_LASTPRIVATE_STMT.
	Use is_taskreg_ctx instead of is_parallel_ctx.
	Handle OMP_CLAUSE_COLLAPSE and OMP_CLAUSE_UNTIED.
	(create_omp_child_function_name): Add task_copy argument, use
	*_omp_cpyfn* names if it is true.
	(create_omp_child_function): Add task_copy argument, if true create
	*_omp_cpyfn* helper function.
	(scan_omp_parallel): Adjust create_omp_child_function callers.
	Rename parallel_nesting_level to taskreg_nesting_level.
	(scan_omp_task): New function.
	(lower_rec_input_clauses): Don't run constructors for firstprivate
	explicit task vars which are initialized by *_omp_cpyfn*.  
	Pass outer var ref to omp_clause_default_ctor hook if
	OMP_CLAUSE_PRIVATE_OUTER_REF or OMP_CLAUSE_LASTPRIVATE.
	Replace OMP_CLAUSE_REDUCTION_PLACEHOLDER decls in
	OMP_CLAUSE_REDUCTION_INIT.
	(lower_send_clauses): Clear DECL_ABSTRACT_ORIGIN if in task to
	avoid duplicate setting of fields.  Handle
	OMP_CLAUSE_PRIVATE_OUTER_REF.
	(lower_send_shared_vars): Use srecord_type if non-NULL.  Don't
	copy-out if TREE_READONLY, only copy-in.
	(expand_task_copyfn): New function.
	(expand_task_call): New function.
	(struct omp_taskcopy_context): New type.
	(task_copyfn_copy_decl, task_copyfn_remap_type, create_task_copyfn):
	New functions.
	(lower_omp_parallel): Rename to...
	(lower_omp_taskreg): ... this.  Use OMP_TASKREG_* macros where needed.
	Call create_task_copyfn if srecord_type is needed.  Adjust
	sender_decl type.
	(task_shared_vars): New variable.
	(check_omp_nesting_restrictions): Warn if work-sharing,
	barrier, master or ordered region is closely nested inside OMP_TASK.
	Add warnings for barrier if closely nested inside of work-sharing,
	ordered, or master region.
	(scan_omp_1): Call check_omp_nesting_restrictions even for
	GOMP_barrier calls.  Rename parallel_nesting_level to
	taskreg_nesting_level.  Handle OMP_TASK.
	(lower_lastprivate_clauses): Even if some lastprivate is found on a
	work-sharing construct, continue looking for them on parent parallel
	construct.
	(lower_omp_for_lastprivate): Add lastprivate clauses
	to the beginning of dlist rather than end.  Adjust for struct
	omp_for_data changes.
	(lower_omp_for): Add rec input clauses before OMP_FOR_PRE_BODY,
	not after it.  Handle collapsed OMP_FOR loops, adjust for
	OMP_FOR_{INIT,COND,INCR} changes, adjust extract_omp_for_data
	caller.
	(get_ws_args_for): Adjust extract_omp_for_data caller.
	(scan_omp_for): Handle collapsed OMP_FOR
	loops, adjust for OMP_FOR_{INIT,COND,INCR} changes.
	(lower_omp_single_simple): If libgomp function doesn't return
	boolean_type_node, add comparison of the return value with 0.
	(diagnose_sb_1, diagnose_sb_2): Handle collapsed OMP_FOR
	loops, adjust for OMP_FOR_{INIT,COND,INCR} changes.  Handle OMP_TASK.
	(parallel_nesting_level): Rename to...
	(taskreg_nesting_level): ... this.
	(is_taskreg_ctx): New function.
	(build_outer_var_ref, omp_copy_decl): Use is_taskreg_ctx instead
	of is_parallel_ctx.
	(execute_lower_omp): Rename parallel_nesting_level to
	taskreg_nesting_level.
	(expand_omp_parallel): Rename to...
	(expand_omp_taskreg): ... this.  Use OMP_TASKREG_* macros where needed.
	Call omp_task_call for OMP_TASK regions.
	(expand_omp): Adjust caller, handle OMP_TASK.
	(lower_omp_1): Adjust lower_omp_taskreg caller, handle OMP_TASK.

	* bitmap.c (bitmap_default_obstack_depth): New variable.
	(bitmap_obstack_initialize, bitmap_obstack_release): Do nothing
	if argument is NULL and bitmap_default_obstack is already initialized.
	* ipa-struct-reorg.c (do_reorg_1): Call bitmap_obstack_release
	at the end.
	* matrix-reorg.c (matrix_reorg): Likewise.
cp/
	* cp-tree.h (cxx_omp_finish_clause, cxx_omp_create_clause_info,
	dependent_omp_for_p, begin_omp_task, finish_omp_task,
	finish_omp_taskwait): New prototypes.
	(cxx_omp_clause_default_ctor): Add outer argument.
	(finish_omp_for): Add new clauses argument.
	* cp-gimplify.c (cxx_omp_finish_clause): New function.
	(cxx_omp_predetermined_sharing): Moved from semantics.c, rewritten.
	(cxx_omp_clause_default_ctor): Add outer argument.
	(cp_genericize_r): Walk OMP_CLAUSE_LASTPRIVATE_STMT.
	* cp-objcp-common.h (LANG_HOOKS_OMP_FINISH_CLAUSE): Define.
	* parser.c (cp_parser_omp_for_loop): Parse collapsed for loops.
	Add par_clauses argument.  If decl is present in parallel's
	lastprivate clause, change that clause to shared and add
	a lastprivate clause for decl to OMP_FOR_CLAUSES.
	Fix wording of error messages.  Adjust finish_omp_for caller.
	Add clauses argument.  Parse loops with random access iterators.
	(cp_parser_omp_clause_collapse, cp_parser_omp_clause_untied): New
	functions.
	(cp_parser_omp_for, cp_parser_omp_parallel): Adjust
	cp_parser_omp_for_loop callers.
	(cp_parser_omp_for_cond, cp_parser_omp_for_incr): New helper
	functions.
	(cp_parser_omp_clause_name): Handle collapse and untied
	clauses.
	(cp_parser_omp_clause_schedule): Handle auto schedule.
	(cp_parser_omp_all_clauses): Handle PRAGMA_OMP_CLAUSE_COLLAPSE
	and PRAGMA_OMP_CLAUSE_UNTIED.
	(OMP_FOR_CLAUSE_MASK): Add PRAGMA_OMP_CLAUSE_COLLAPSE.
	(OMP_TASK_CLAUSE_MASK): Define.
	(cp_parser_omp_task, cp_parser_omp_taskwait): New functions.
	(cp_parser_omp_construct): Handle PRAGMA_OMP_TASK.
	(cp_parser_pragma): Handle PRAGMA_OMP_TASK and
	PRAGMA_OMP_TASKWAIT.
	* pt.c (tsubst_omp_clauses): Handle OMP_CLAUSE_COLLAPSE and
	OMP_CLAUSE_UNTIED.  Handle OMP_CLAUSE_LASTPRIVATE_STMT.
	(tsubst_omp_for_iterator): New function.
	(dependent_omp_for_p): New function.
	(tsubst_expr) <case OMP_FOR>: Use it.  Handle collapsed OMP_FOR
	loops.  Adjust finish_omp_for caller.  Handle loops with random
	access iterators.  Adjust for OMP_FOR_{INIT,COND,INCR} changes.
	(tsubst_expr): Handle OMP_TASK.
	* semantics.c (cxx_omp_create_clause_info): New function.
	(finish_omp_clauses): Call it.  Handle OMP_CLAUSE_UNTIED and
	OMP_CLAUSE_COLLAPSE.
	(cxx_omp_predetermined_sharing): Removed.
	* semantics.c (finish_omp_for): Allow pointer iterators.  Use
	handle_omp_for_class_iterator and dependent_omp_for_p.  Handle
	collapsed for loops.  Adjust c_finish_omp_for caller.  Add new
	clauses argument.  Fix check for type dependent cond or incr.
	Set OMP_FOR_CLAUSES to clauses.  Use cp_convert instead of
	fold_convert to convert incr amount to difference_type.  Only
	fold if not in template.  If decl is mentioned in lastprivate
	clause, set OMP_CLAUSE_LASTPRIVATE_STMT.  Handle loops with random
	access iterators.  Adjust for OMP_FOR_{INIT,COND,INCR}
	changes.
	(finish_omp_threadprivate): Allow static class members of the
	current class.
	(handle_omp_for_class_iterator, begin_omp_task, finish_omp_task,
	finish_omp_taskwait): New functions.

	* parser.c (cp_parser_binary_expression): Add prec argument.
	(cp_parser_assignment_expression): Adjust caller.
	* cp-tree.h (outer_curly_brace_block): New prototype.
	* decl.c (outer_curly_brace_block): No longer static.
fortran/
	* scanner.c (skip_free_comments, skip_fixed_comments): Handle tabs.
	* parse.c (next_free): Allow tab after !$omp.
	(decode_omp_directive): Handle !$omp task, !$omp taskwait
	and !$omp end task.
	(case_executable): Add ST_OMP_TASKWAIT.
	(case_exec_markers): Add ST_OMP_TASK.
	(gfc_ascii_statement): Handle ST_OMP_TASK, ST_OMP_END_TASK and
	ST_OMP_TASKWAIT.
	(parse_omp_structured_block, parse_executable): Handle ST_OMP_TASK.
	* gfortran.h (gfc_find_sym_in_expr): New prototype.
	(gfc_statement): Add ST_OMP_TASK, ST_OMP_END_TASK and ST_OMP_TASKWAIT.
	(gfc_omp_clauses): Add OMP_SCHED_AUTO to sched_kind,
	OMP_DEFAULT_FIRSTPRIVATE to default_sharing.  Add collapse and
	untied fields.
	(gfc_exec_op): Add EXEC_OMP_TASK and EXEC_OMP_TASKWAIT.
	* f95-lang.c (LANG_HOOKS_OMP_CLAUSE_COPY_CTOR,
	LANG_HOOKS_OMP_CLAUSE_ASSIGN_OP, LANG_HOOKS_OMP_CLAUSE_DTOR,
	LANG_HOOKS_OMP_PRIVATE_OUTER_REF): Define.
	* trans.h (gfc_omp_clause_default_ctor): Add another argument.
	(gfc_omp_clause_copy_ctor, gfc_omp_clause_assign_op,
	gfc_omp_clause_dtor, gfc_omp_private_outer_ref): New prototypes.
	* types.def (BT_ULONGLONG, BT_PTR_ULONGLONG,
	BT_FN_BOOL_ULONGLONGPTR_ULONGLONGPTR,
	BT_FN_BOOL_BOOL_ULL_ULL_ULL_ULLPTR_ULLPTR,
	BT_FN_BOOL_BOOL_ULL_ULL_ULL_ULL_ULLPTR_ULLPTR,
	BT_FN_VOID_PTR_PTR, BT_PTR_FN_VOID_PTR_PTR,
	BT_FN_VOID_OMPFN_PTR_OMPCPYFN_LONG_LONG_BOOL_UINT): New.
	(BT_BOOL): Use integer type with BOOL_TYPE_SIZE rather
	than boolean_type_node.
	* dump-parse-tree.c (gfc_show_omp_node): Handle EXEC_OMP_TASK,
	EXEC_OMP_TASKWAIT, OMP_SCHED_AUTO, OMP_DEFAULT_FIRSTPRIVATE,
	untied and collapse clauses.
	(gfc_show_code_node): Handle EXEC_OMP_TASK and EXEC_OMP_TASKWAIT.
	* trans.c (gfc_trans_code): Handle EXEC_OMP_TASK and
	EXEC_OMP_TASKWAIT.
	* st.c (gfc_free_statement): Likewise.
	* resolve.c (gfc_resolve_blocks, resolve_code): Likewise.
	(find_sym_in_expr): Rename to...
	(gfc_find_sym_in_expr): ... this.  No longer static.
	(resolve_allocate_expr, resolve_ordinary_assign): Adjust caller.
	* match.h (gfc_match_omp_task, gfc_match_omp_taskwait): New
	prototypes.
	* openmp.c (resolve_omp_clauses): Allow allocatable arrays in
	firstprivate, lastprivate, reduction, copyprivate and copyin
	clauses.
	(omp_current_do_code): Made static.
	(omp_current_do_collapse): New variable.
	(gfc_resolve_omp_do_blocks): Compute omp_current_do_collapse,
	clear omp_current_do_code and omp_current_do_collapse on return.
	(gfc_resolve_do_iterator): Handle collapsed do loops.
	(resolve_omp_do): Likewise, diagnose errorneous collapsed do loops.
	(OMP_CLAUSE_COLLAPSE, OMP_CLAUSE_UNTIED): Define.
	(gfc_match_omp_clauses): Handle default (firstprivate),
	schedule (auto), untied and collapse (n) clauses.
	(OMP_DO_CLAUSES): Add OMP_CLAUSE_COLLAPSE.
	(OMP_TASK_CLAUSES): Define.
	(gfc_match_omp_task, gfc_match_omp_taskwait): New functions.
	* trans-openmp.c (gfc_omp_private_outer_ref): New function.
	(gfc_omp_clause_default_ctor): Add outer argument.  For allocatable
	arrays allocate them with the bounds of the outer var if outer
	var is allocated.
	(gfc_omp_clause_copy_ctor, gfc_omp_clause_assign_op,
	gfc_omp_clause_dtor): New functions.
	(gfc_trans_omp_array_reduction): If decl is allocatable array,
	allocate it with outer var's bounds in OMP_CLAUSE_REDUCTION_INIT
	and deallocate it in OMP_CLAUSE_REDUCTION_MERGE.
	(gfc_omp_predetermined_sharing): Return OMP_CLAUSE_DEFAULT_SHARED
	for assumed-size arrays.
	(gfc_trans_omp_do): Add par_clauses argument.  If dovar is
	present in lastprivate clause and do loop isn't simple,
	set OMP_CLAUSE_LASTPRIVATE_STMT.  If dovar is present in
	parallel's lastprivate clause, change it to shared and add
	lastprivate clause to OMP_FOR_CLAUSES.  Handle collapsed do loops.
	(gfc_trans_omp_directive): Adjust gfc_trans_omp_do callers.
	(gfc_trans_omp_parallel_do): Likewise.  Move collapse clause to
	OMP_FOR from OMP_PARALLEL.
	(gfc_trans_omp_clauses): Handle OMP_SCHED_AUTO,
	OMP_DEFAULT_FIRSTPRIVATE, untied and collapse clauses.
	(gfc_trans_omp_task, gfc_trans_omp_taskwait): New functions.
	(gfc_trans_omp_directive): Handle EXEC_OMP_TASK and
	EXEC_OMP_TASKWAIT.
gcc/testsuite/
	* gcc.dg/gomp/collapse-1.c: New test.
	* gcc.dg/gomp/nesting-1.c: New test.
	* g++.dg/gomp/task-1.C: New test.
	* g++.dg/gomp/predetermined-1.C: New test.
	* g++.dg/gomp/tls-4.C: New test.
	* gfortran.dg/gomp/collapse1.f90: New test.
	* gfortran.dg/gomp/sharing-3.f90: New test.
	* gcc.dg/gomp/pr27499.c (foo): Remove is unsigned dg-warning.
	* g++.dg/gomp/pr27499.C (foo): Likewise.
	* g++.dg/gomp/for-16.C (foo): Likewise.
	* g++.dg/gomp/tls-3.C: Remove dg-error, add S::s definition.
	* g++.dg/gomp/pr34607.C: Adjust dg-error location.
	* g++.dg/gomp/for-16.C (foo): Add a new dg-error.
	* gcc.dg/gomp/appendix-a/a.35.4.c: Add dg-warning.
	* gcc.dg/gomp/appendix-a/a.35.6.c: Likewise.
	* gfortran.dg/gomp/appendix-a/a.35.4.f90: Likewise.
	* gfortran.dg/gomp/appendix-a/a.35.6.f90: Likewise.
	* gfortran.dg/gomp/omp_parse1.f90: Remove !$omp tab test.
	* gfortran.dg/gomp/appendix-a/a.33.4.f90: Remove dg-error
	about allocatable array.
	* gfortran.dg/gomp/reduction1.f90: Likewise.
libgomp/
	* configure.ac (LIBGOMP_GNU_SYMBOL_VERSIONING): New AC_DEFINE.
	Substitute also OMP_*LOCK_25*.
	* configure: Regenerated.
	* config.h.in: Regenerated.
	* Makefile.am (libgomp_la_SOURCES): Add loop_ull.c, iter_ull.c,
	ptrlock.c and task.c.
	* Makefile.in: Regenerated.
	* testsuite/Makefile.in: Regenerated.
	* task.c: New file.
	* loop_ull.c: New file.
	* iter_ull.c: New file.
	* libgomp.h: Include ptrlock.h.
	(enum gomp_task_kind): New type.
	(struct gomp_team): Add task_lock, task_queue, task_count,
	task_running_count, single_count fields.  Add
	work_share_list_free_lock ifndef HAVE_SYNC_BUILTINS.
	Remove work_share_lock, generation_mask,
	oldest_live_gen, num_live_gen and init_work_shares fields, add
	work work_share_list_alloc, work_share_list_free and work_share_chunk
	fields.  Change work_shares from pointer to pointers into an array.
	Change ordered_release field into gomp_sem_t ** from flexible array
	member.  Add implicit_task and initial_work_shares fields.
	Move close to the end of the struct.
	(struct gomp_team_state): Add single_count, last_work_share,
	active_level and level fields, remove work_share_generation.
	(gomp_barrier_handle_tasks): New prototype.
	(gomp_finish_task): New inline function.
	(struct gomp_work_share): Move chunk_size, end, incr into
	transparent union/struct, add chunk_size_ull, end_ll, incr_ll and
	next_ll fields.  Reshuffle fields.  Add next_alloc,
	next_ws, next_free and inline_ordered_team_ids fields, change
	ordered_team_ids into pointer from flexible array member.
	Add mode field.  Put lock and next into a different cache line
	from most of the write-once fields.
	(gomp_iter_ull_static_next, gomp_iter_ull_dynamic_next_locked,
	gomp_iter_ull_guided_next_locked, gomp_iter_ull_dynamic_next,
	gomp_iter_ull_guided_next): New prototypes.
	(gomp_new_icv): New prototype.
	(struct gomp_thread): Add thread_pool and task fields.
	(struct gomp_thread_pool): New type.
	(gomp_new_team): New prototype.
	(gomp_team_start): Change type of last argument.
	(gomp_new_work_share): Removed.
	(gomp_init_work_share, gomp_fini_work_share): New prototypes.
	(gomp_work_share_init_done): New static inline.
	(gomp_throttled_spin_count_var, gomp_available_cpus,
	gomp_managed_threads): New extern decls.
	(gomp_init_task): New prototype.
	(gomp_spin_count_var): New extern var decl.
	(LIBGOMP_GNU_SYMBOL_VERSIONING): Undef if no visibility
	or no alias support, or if not PIC.
	(gomp_init_lock_30, gomp_destroy_lock_30, gomp_set_lock_30,
	gomp_unset_lock_30, gomp_test_lock_30, gomp_init_nest_lock_30,
	gomp_destroy_nest_lock_30, gomp_set_nest_lock_30,
	gomp_unset_nest_lock_30, gomp_test_nest_lock_30, gomp_init_lock_25,
	gomp_destroy_lock_25, gomp_set_lock_25, gomp_unset_lock_25,
	gomp_test_lock_25, gomp_init_nest_lock_25, gomp_destroy_nest_lock_25,
	gomp_set_nest_lock_25, gomp_unset_nest_lock_25,
	gomp_test_nest_lock_25): New prototypes.
	(omp_lock_symver, strong_alias): Define.
	(gomp_remaining_threads_count, gomp_remaining_threads_lock): New
	decls.
	(gomp_end_task): New.
	(struct gomp_task_icv, gomp_global_icv): New.
	(gomp_thread_limit_var, gomp_max_active_levels_var): New.
	(struct gomp_task): New.
	(gomp_nthreads_var, gomp_dyn_var, gomp_nest_var,
	gomp_run_sched_var, gomp_run_sched_chunk): Remove.
	(gomp_icv): New.
	(gomp_schedule_type): Reorder enum to match
	omp_sched_t.
	* team.c (struct gomp_thread_start_data): Add thread_pool and task
	fields.
	(gomp_thread_start): Add gomp_team_barrier_wait call.
	For non-nested case remove clearing of docked thread thr fields.
	Use pool fields instead of global gomp_* variables.  Use
	gomp_barrier_wait_last when needed.  Initialize ts.active_level.
	Create tasks for each member thread.
	(free_team): Only destroy team barrier, task_lock here and free it.
	(gomp_free_thread): Free last_team if non-NULL.
	(gomp_team_end): Call gomp_team_barrier_wait instead of
	gomp_barrier_wait.  For nested case call one extra
	gomp_barrier_wait.  Move here some destruction from free_team.
	Call free_team on pool->last_team if any, rather than freeing
	current team.  Destroy work_share_list_free_lock ifndef
	HAVE_SYNC_BUILTINS.
	(gomp_new_icv): New function.
	(gomp_threads, gomp_threads_size, gomp_threads_used,
	gomp_threads_dock): Removed.
	(gomp_thread_destructor): New variable.
	(gomp_new_thread_pool, gomp_free_pool_helper, gomp_free_thread): New
	functions.
	(gomp_team_start): Create new pool if current thread doesn't have
	one.  Use pool fields instead of global gomp_* variables. 
	Initialize thread_pool field for new threads.  Clear single_count.
	Change last argument from ws to team, don't create
	new team, set ts.work_share to &team->work_shares[0] and clear
	ts.last_work_share.  Don't clear ts.work_share_generation.
	If number of threads changed, adjust atomically gomp_managed_threads.
	Use gomp_init_task instead of gomp_new_task,
	set thr->task to the corresponding implicit_task array entry.
	Create tasks for each member thread.  Initialize ts.level.
	(initialize_team): Call pthread_key_create on
	gomp_thread_destructor.
	(team_destructor): New function.
	(new_team): Removed.
	(gomp_new_team): New function.
	(free_team): Free gomp_work_share blocks chained through next_alloc,
	instead of freeing work_shares and destroying work_share_lock.
	(gomp_team_end): Call gomp_fini_work_share.  If number of threads
	changed, adjust atomically gomp_managed_threads.  Use gomp_end_task.
	* barrier.c (GOMP_barrier): Call gomp_team_barrier_wait instead
	of gomp_barrier_wait.
	* single.c (GOMP_single_copy_start): Call gomp_team_barrier_wait
	instead of gomp_barrier_wait.  Call gomp_work_share_init_done
	if gomp_work_share_start returned true.  Don't unlock ws->lock.
	(GOMP_single_copy_end): Call gomp_team_barrier_wait instead
	of gomp_barrier_wait.
	(GOMP_single_start): Rewritten if HAVE_SYNC_BUILTINS.  Call
	gomp_work_share_init_done if gomp_work_share_start returned true.
	Don't unlock ws->lock.
	* work.c: Include stddef.h.
	(free_work_share): Use work_share_list_free_lock instead
	of atomic chaining ifndef HAVE_SYNC_BUILTINS.  Add team argument.
	Call gomp_fini_work_share and then either free ws if orphaned, or
	put it into work_share_list_free list of the current team.
	(alloc_work_share, gomp_init_work_share, gomp_fini_work_share): New
	functions.
	(gomp_work_share_start, gomp_work_share_end,
	gomp_work_share_end_nowait): Rewritten.
	* omp_lib.f90.in Change some tabs to spaces to prevent warnings.
	(openmp_version): Set to 200805.
	(omp_sched_kind, omp_sched_static, omp_sched_dynamic,
	omp_sched_guided, omp_sched_auto): New parameters.
	(omp_set_schedule, omp_get_schedule, omp_get_thread_limit,
	omp_set_max_active_levels, omp_get_max_active_levels,
	omp_get_level, omp_get_ancestor_thread_num, omp_get_team_size,
	omp_get_active_level): New interfaces.
	* omp_lib.h.in (openmp_version): Set to 200805.
	(omp_sched_kind, omp_sched_static, omp_sched_dynamic,
	omp_sched_guided, omp_sched_auto): New parameters.
	(omp_set_schedule, omp_get_schedule, omp_get_thread_limit,
	omp_set_max_active_levels, omp_get_max_active_levels,
	omp_get_level, omp_get_ancestor_thread_num, omp_get_team_size,
	omp_get_active_level): New externals.
	* loop.c: Include limits.h.
	(GOMP_loop_runtime_next, GOMP_loop_ordered_runtime_next): Handle
	GFS_AUTO.
	(GOMP_loop_runtime_start, GOMP_loop_ordered_runtime_start):
	Likewise.  Use gomp_icv.
	(gomp_loop_static_start, gomp_loop_dynamic_start): Clear
	ts.static_trip here.
	(gomp_loop_static_start, gomp_loop_ordered_static_start): Call
	gomp_work_share_init_done after gomp_loop_init.  Don't unlock ws->lock.
	(gomp_loop_dynamic_start, gomp_loop_guided_start): Call
	gomp_work_share_init_done after gomp_loop_init.  If HAVE_SYNC_BUILTINS,
	don't unlock ws->lock, otherwise lock it.
	(gomp_loop_ordered_dynamic_start, gomp_loop_ordered_guided_start): Call
	gomp_work_share_init_done after gomp_loop_init.  Lock ws->lock.
	(gomp_parallel_loop_start): Call gomp_new_team instead of
	gomp_new_work_share.  Call gomp_loop_init on &team->work_shares[0].
	Adjust gomp_team_start caller.  Pass 0 as second argument to
	gomp_resolve_num_threads.
	(gomp_loop_init): For GFS_DYNAMIC, multiply ws->chunk_size by incr.
	If adding ws->chunk_size nthreads + 1 times after end won't
	overflow, set ws->mode to 1.
	* libgomp_g.h (GOMP_loop_ull_static_start, GOMP_loop_ull_dynamic_start,
	GOMP_loop_ull_guided_start, GOMP_loop_ull_runtime_start,
	GOMP_loop_ull_ordered_static_start,
	GOMP_loop_ull_ordered_dynamic_start,
	GOMP_loop_ull_ordered_guided_start,
	GOMP_loop_ull_ordered_runtime_start, GOMP_loop_ull_static_next,
	GOMP_loop_ull_dynamic_next, GOMP_loop_ull_guided_next,
	GOMP_loop_ull_runtime_next, GOMP_loop_ull_ordered_static_next,
	GOMP_loop_ull_ordered_dynamic_next, GOMP_loop_ull_ordered_guided_next,
	GOMP_loop_ull_ordered_runtime_next, GOMP_task, GOMP_taskwait): New
	prototypes.
	* libgomp.map: Export lock routines also @@OMP_2.0.
	(GOMP_loop_ordered_dynamic_first,
	GOMP_loop_ordered_guided_first, GOMP_loop_ordered_runtime_first,
	GOMP_loop_ordered_static_first): Remove.
	(GOMP_loop_ull_dynamic_next, GOMP_loop_ull_dynamic_start,
	GOMP_loop_ull_guided_next, GOMP_loop_ull_guided_start,
	GOMP_loop_ull_ordered_dynamic_next,
	GOMP_loop_ull_ordered_dynamic_start,
	GOMP_loop_ull_ordered_guided_next,
	GOMP_loop_ull_ordered_guided_start,
	GOMP_loop_ull_ordered_runtime_next,
	GOMP_loop_ull_ordered_runtime_start,
	GOMP_loop_ull_ordered_static_next,
	GOMP_loop_ull_ordered_static_start,
	GOMP_loop_ull_runtime_next, GOMP_loop_ull_runtime_start,
	GOMP_loop_ull_static_next, GOMP_loop_ull_static_start,
	GOMP_task, GOMP_taskwait): Export @@GOMP_2.0.
	(omp_set_schedule, omp_get_schedule,
	omp_get_thread_limit, omp_set_max_active_levels,
	omp_get_max_active_levels, omp_get_level,
	omp_get_ancestor_thread_num, omp_get_team_size, omp_get_active_level,
	omp_set_schedule_, omp_set_schedule_8_,
	omp_get_schedule_, omp_get_schedule_8_, omp_get_thread_limit_,
	omp_set_max_active_levels_, omp_set_max_active_levels_8_,
	omp_get_max_active_levels_, omp_get_level_,
	omp_get_ancestor_thread_num_, omp_get_ancestor_thread_num_8_,
	omp_get_team_size_, omp_get_team_size_8_, omp_get_active_level_):
	New exports @@OMP_3.0.
	* omp.h.in (omp_sched_t): New type.
	(omp_set_schedule, omp_get_schedule, omp_get_thread_limit,
	omp_set_max_active_levels, omp_get_max_active_levels,
	omp_get_level, omp_get_ancestor_thread_num, omp_get_team_size,
	omp_get_active_level): New prototypes.
	* env.c (gomp_spin_count_var, gomp_throttled_spin_count_var,
	gomp_available_cpus, gomp_managed_threads, gomp_max_active_levels_var,
	gomp_thread_limit_var, gomp_remaining_threads_count,
	gomp_remaining_threads_lock): New variables.
	(parse_spincount): New function.
	(initialize_env): Call gomp_init_num_threads unconditionally.
	Initialize gomp_available_cpus.  Call parse_spincount,
	initialize gomp_{,throttled_}spin_count_var
	depending on presence and value of OMP_WAIT_POLICY and
	GOMP_SPINCOUNT env vars.  Handle GOMP_BLOCKTIME env var.
	Handle OMP_WAIT_POLICY, OMP_MAX_ACTIVE_LEVELS,
	OMP_THREAD_LIMIT, OMP_STACKSIZE env vars.  Handle unit specification
	for GOMP_STACKSIZE.  Initialize gomp_remaining_threads_count and
	gomp_remaining_threads_lock if needed.  Use gomp_global_icv.
	(gomp_nthreads_var, gomp_dyn_var, gomp_nest_var,
	gomp_run_sched_var, gomp_run_sched_chunk): Remove.
	(gomp_global_icv): New.
	(parse_schedule): Use it.  Parse "auto".
	(omp_set_num_threads): Use gomp_icv.
	(omp_set_dynamic, omp_get_dynamic, omp_set_nested, omp_get_nested):
	Likewise.
	(omp_get_max_threads): Move from parallel.c.
	(omp_set_schedule, omp_get_schedule, omp_get_thread_limit,
	omp_set_max_active_levels, omp_get_max_active_levels): New functions,
	add ialias.
	(parse_stacksize, parse_wait_policy): New functions.
	* fortran.c: Rewrite lock wrappers, if symbol versioning provide
	both wrappers for compatibility and new locks.
	(omp_set_schedule, omp_get_schedule,
	omp_get_thread_limit, omp_set_max_active_levels,
	omp_get_max_active_levels, omp_get_level,
	omp_get_ancestor_thread_num, omp_get_team_size,
	omp_get_active_level): New ialias_redirect.
	(omp_set_schedule_, omp_set_schedule_8_,
	omp_get_schedule_, omp_get_schedule_8_, omp_get_thread_limit_,
	omp_set_max_active_levels_, omp_set_max_active_levels_8_,
	omp_get_max_active_levels_, omp_get_level_,
	omp_get_ancestor_thread_num_, omp_get_ancestor_thread_num_8_,
	omp_get_team_size_, omp_get_team_size_8_, omp_get_active_level_):
	New functions.
	* parallel.c: Include limits.h.
	(gomp_resolve_num_threads): Add count argument.  Rewritten.
	(GOMP_parallel_start): Call gomp_new_team and pass that as last
	argument to gomp_team_start.  Pass 0 as second argument to
	gomp_resolve_num_threads.
	(GOMP_parallel_end): Decrease gomp_remaining_threads_count
	if gomp_thread_limit_var != ULONG_MAX.
	(omp_in_parallel): Implement using ts.active_level.
	(omp_get_max_threads): Move to env.c.
	(omp_get_level, omp_get_ancestor_thread_num,
	omp_get_team_size, omp_get_active_level): New functions,
	add ialias.
	* sections.c (GOMP_sections_start): Call gomp_work_share_init_done
	after gomp_sections_init.  If HAVE_SYNC_BUILTINS, call
	gomp_iter_dynamic_next instead of the _locked variant and don't take
	lock around it, otherwise acquire it before calling
	gomp_iter_dynamic_next_locked.
	(GOMP_sections_next): If HAVE_SYNC_BUILTINS, call
	gomp_iter_dynamic_next instead of the _locked variant and don't take
	lock around it.
	(GOMP_parallel_sections_start): Call gomp_new_team instead of
	gomp_new_work_share.  Call gomp_sections_init on &team->work_shares[0].
	Adjust gomp_team_start caller.  Pass count as second argument to
	gomp_resolve_num_threads, don't adjust num_threads after the call.
	Use gomp_icv.
	* iter.c (gomp_iter_dynamic_next_locked): Don't multiply
	ws->chunk_size by incr.
	(gomp_iter_dynamic_next): Likewise.  If ws->mode, use more efficient
	code.
	* libgomp_f.h.in (omp_lock_25_arg_t, omp_nest_lock_25_arg_t): New
	types.
	(omp_lock_25_arg, omp_nest_lock_25_arg): New macros.
	(omp_check_defines): Check even the compat defines.
	* config/linux/ptrlock.c: New file.
	* config/linux/ptrlock.h: New file.
	* config/linux/wait.h: New file.
	* config/posix/ptrlock.c: New file.
	* config/posix/ptrlock.h: New file.
	* config/linux/bar.h (gomp_team_barrier_wait,
	gomp_team_barrier_wait_end, gomp_team_barrier_wake): New prototypes.
	(gomp_team_barrier_set_task_pending,
	gomp_team_barrier_clear_task_pending,
	gomp_team_barrier_set_waiting_for_tasks,
	gomp_team_barrier_waiting_for_tasks,
	gomp_team_barrier_done): New inlines.
	(gomp_barrier_t): Rewritten.
	(gomp_barrier_state_t): New typedef.
	(gomp_barrier_init, gomp_barrier_reinit, gomp_barrier_destroy,
	gomp_barrier_wait_start): Rewritten.
	(gomp_barrier_wait_end): Change second argument to
	gomp_barrier_state_t.
	(gomp_barrier_last_thread, gomp_barrier_wait_last): New static
	inlines.
	* config/linux/bar.c: Include wait.h instead of libgomp.h and
	futex.h.
	(gomp_barrier_wait_end): Rewritten.
	(gomp_team_barrier_wait, gomp_team_barrier_wait_end,
	gomp_team_barrier_wake, gomp_barrier_wait_last): New functions.
	* config/posix/bar.h (gomp_barrier_t): Add generation field.
	(gomp_barrier_state_t): New typedef.
	(gomp_team_barrier_wait,
	gomp_team_barrier_wait_end, gomp_team_barrier_wake): New prototypes.
	(gomp_barrier_wait_start): Or all but low 2 bits from generation
	into the return value.  Return gomp_barrier_state_t.
	(gomp_team_barrier_set_task_pending,
	gomp_team_barrier_clear_task_pending,
	gomp_team_barrier_set_waiting_for_tasks,
	gomp_team_barrier_waiting_for_tasks,
	gomp_team_barrier_done): New inlines.
	(gomp_barrier_wait_end): Change second argument to
	gomp_barrier_state_t.
	(gomp_barrier_last_thread, gomp_barrier_wait_last): New static
	inlines.
	* config/posix/bar.c (gomp_barrier_init): Clear generation field.
	(gomp_barrier_wait_end): Change second argument to
	gomp_barrier_state_t. 
	(gomp_team_barrier_wait, gomp_team_barrier_wait_end,
	gomp_team_barrier_wake): New functions.
	* config/linux/mutex.c: Include wait.h instead of libgomp.h and
	futex.h.
	(gomp_futex_wake, gomp_futex_wait): New variables.
	(gomp_mutex_lock_slow): Call do_wait instead of futex_wait.
	* config/linux/lock.c: Rewrite to make locks task owned,
	for backwards compatibility provide the old entrypoints
	if symbol versioning.  Include wait.h instead of libgomp.h and
	futex.h.
	(gomp_set_nest_lock_25): Call do_wait instead of futex_wait.
	* config/posix95/lock.c: Rewrite to make locks task owned,
	for backwards compatibility provide the old entrypoints
	if symbol versioning.
	* config/posix/lock.c: Rewrite to make locks task owned,
	for backwards compatibility provide the old entrypoints
	if symbol versioning.
	* config/linux/proc.c (gomp_init_num_threads): Use gomp_global_icv.
	(get_num_procs, gomp_dynamic_max_threads): Use gomp_icv.
	* config/posix/proc.c, config/mingw32/proc.c: Similarly.
	* config/linux/powerpc/futex.h (FUTEX_WAIT, FUTEX_WAKE): Remove.
	(sys_futex0): Return error code.
	(futex_wake, futex_wait): If ENOSYS was returned, clear
	FUTEX_PRIVATE_FLAG in gomp_futex_wa{ke,it} and retry.
	(cpu_relax, atomic_write_barrier): New static inlines.
	* config/linux/alpha/futex.h (FUTEX_WAIT, FUTEX_WAKE): Remove.
	(futex_wake, futex_wait): If ENOSYS was returned, clear
	FUTEX_PRIVATE_FLAG in gomp_futex_wa{ke,it} and retry.
	(cpu_relax, atomic_write_barrier): New static inlines.
	* config/linux/x86/futex.h (FUTEX_WAIT, FUTEX_WAKE): Remove.
	(sys_futex0): Return error code.
	(futex_wake, futex_wait): If ENOSYS was returned, clear
	FUTEX_PRIVATE_FLAG in gomp_futex_wa{ke,it} and retry.
	(cpu_relax, atomic_write_barrier): New static inlines.
	* config/linux/s390/futex.h (FUTEX_WAIT, FUTEX_WAKE): Remove.
	(sys_futex0): Return error code.
	(futex_wake, futex_wait): If ENOSYS was returned, clear
	FUTEX_PRIVATE_FLAG in gomp_futex_wa{ke,it} and retry.
	(cpu_relax, atomic_write_barrier): New static inlines.
	* config/linux/ia64/futex.h (FUTEX_WAIT, FUTEX_WAKE): Remove.
	(sys_futex0): Return error code.
	(futex_wake, futex_wait): If ENOSYS was returned, clear
	FUTEX_PRIVATE_FLAG in gomp_futex_wa{ke,it} and retry.
	(cpu_relax, atomic_write_barrier): New static inlines.
	* config/linux/sparc/futex.h (FUTEX_WAIT, FUTEX_WAKE): Remove.
	(sys_futex0): Return error code.
	(futex_wake, futex_wait): If ENOSYS was returned, clear
	FUTEX_PRIVATE_FLAG in gomp_futex_wa{ke,it} and retry.
	(cpu_relax, atomic_write_barrier): New static inlines.
	* config/linux/sem.c: Include wait.h instead of libgomp.h and
	futex.h.
	(gomp_sem_wait_slow): Call do_wait instead of futex_wait.
	* config/linux/affinity.c: Assume HAVE_SYNC_BUILTINS.
	* config/linux/omp-lock.h (omp_lock_25_t, omp_nest_lock_25_t): New
	types.
	(omp_nest_lock_t): Change owner into void *, add lock field.
	* config/posix95/omp-lock.h: Include semaphore.h.
	(omp_lock_25_t, omp_nest_lock_25_t): New types.
	(omp_lock_t): Use sem_t instead of mutex if semaphores
	aren't broken.
	(omp_nest_lock_t): Likewise.  Change owner to void *.
	* config/posix/omp-lock.h: Include semaphore.h.
	(omp_lock_25_t, omp_nest_lock_25_t): New types.
	(omp_lock_t): Use sem_t instead of mutex if semaphores
	aren't broken.
	(omp_nest_lock_t): Likewise.  Add owner field.

	* testsuite/libgomp.c/collapse-1.c: New test.
	* testsuite/libgomp.c/collapse-2.c: New test.
	* testsuite/libgomp.c/collapse-3.c: New test.
	* testsuite/libgomp.c/icv-1.c: New test.
	* testsuite/libgomp.c/icv-2.c: New test.
	* testsuite/libgomp.c/lib-2.c: New test.
	* testsuite/libgomp.c/lock-1.c: New test.
	* testsuite/libgomp.c/lock-2.c: New test.
	* testsuite/libgomp.c/lock-3.c: New test.
	* testsuite/libgomp.c/loop-4.c: New test.
	* testsuite/libgomp.c/loop-5.c: New test.
	* testsuite/libgomp.c/loop-6.c: New test.
	* testsuite/libgomp.c/loop-7.c: New test.
	* testsuite/libgomp.c/loop-8.c: New test.
	* testsuite/libgomp.c/loop-9.c: New test.
	* testsuite/libgomp.c/nested-3.c: New test.
	* testsuite/libgomp.c/nestedfn-6.c: New test.
	* testsuite/libgomp.c/sort-1.c: New test.
	* testsuite/libgomp.c/task-1.c: New test.
	* testsuite/libgomp.c/task-2.c: New test.
	* testsuite/libgomp.c/task-3.c: New test.
	* testsuite/libgomp.c/task-4.c: New test.
	* testsuite/libgomp.c++/c++.exp: Add libstdc++-v3 build includes
	to C++ testsuite default compiler options.
	* testsuite/libgomp.c++/collapse-1.C: New test.
	* testsuite/libgomp.c++/collapse-2.C: New test.
	* testsuite/libgomp.c++/ctor-10.C: New test.
	* testsuite/libgomp.c++/for-1.C: New test.
	* testsuite/libgomp.c++/for-2.C: New test.
	* testsuite/libgomp.c++/for-3.C: New test.
	* testsuite/libgomp.c++/for-4.C: New test.
	* testsuite/libgomp.c++/for-5.C: New test.
	* testsuite/libgomp.c++/loop-8.C: New test.
	* testsuite/libgomp.c++/loop-9.C: New test.
	* testsuite/libgomp.c++/loop-10.C: New test.
	* testsuite/libgomp.c++/task-1.C: New test.
	* testsuite/libgomp.c++/task-2.C: New test.
	* testsuite/libgomp.c++/task-3.C: New test.
	* testsuite/libgomp.c++/task-4.C: New test.
	* testsuite/libgomp.c++/task-5.C: New test.
	* testsuite/libgomp.c++/task-6.C: New test.
	* testsuite/libgomp.fortran/allocatable1.f90: New test.
	* testsuite/libgomp.fortran/allocatable2.f90: New test.
	* testsuite/libgomp.fortran/allocatable3.f90: New test.
	* testsuite/libgomp.fortran/allocatable4.f90: New test.
	* testsuite/libgomp.fortran/collapse1.f90: New test.
	* testsuite/libgomp.fortran/collapse2.f90: New test.
	* testsuite/libgomp.fortran/collapse3.f90: New test.
	* testsuite/libgomp.fortran/collapse4.f90: New test.
	* testsuite/libgomp.fortran/lastprivate1.f90: New test.
	* testsuite/libgomp.fortran/lastprivate2.f90: New test.
	* testsuite/libgomp.fortran/lib4.f90: New test.
	* testsuite/libgomp.fortran/lock-1.f90: New test.
	* testsuite/libgomp.fortran/lock-2.f90: New test.
	* testsuite/libgomp.fortran/nested1.f90: New test.
	* testsuite/libgomp.fortran/nestedfn4.f90: New test.
	* testsuite/libgomp.fortran/strassen.f90: New test.
	* testsuite/libgomp.fortran/tabs1.f90: New test.
	* testsuite/libgomp.fortran/tabs2.f: New test.
	* testsuite/libgomp.fortran/task1.f90: New test.
	* testsuite/libgomp.fortran/task2.f90: New test.
	* testsuite/libgomp.fortran/vla4.f90: Add dg-warning.
	* testsuite/libgomp.fortran/vla5.f90: Likewise.
	* testsuite/libgomp.c/pr26943-2.c: Likewise.
	* testsuite/libgomp.c/pr26943-3.c: Likewise.
	* testsuite/libgomp.c/pr26943-4.c: Likewise.


git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@136433 138bc75d-0d04-0410-961f-82ee72b054a4

66 files changed, 6078 insertions, 9 deletions

diff --git a/libgomp/testsuite/Makefile.in b/libgomp/testsuite/Makefile.in
index 9c6163ba2bf..ae1806fb2da 100644
--- a/libgomp/testsuite/Makefile.in
+++ b/libgomp/testsuite/Makefile.in
@@ -112,9 +112,15 @@ MAINTAINER_MODE_TRUE = @MAINTAINER_MODE_TRUE@
 MAKEINFO = @MAKEINFO@
 NM = @NM@
 OBJEXT = @OBJEXT@
+OMP_LOCK_25_ALIGN = @OMP_LOCK_25_ALIGN@
+OMP_LOCK_25_KIND = @OMP_LOCK_25_KIND@
+OMP_LOCK_25_SIZE = @OMP_LOCK_25_SIZE@
 OMP_LOCK_ALIGN = @OMP_LOCK_ALIGN@
 OMP_LOCK_KIND = @OMP_LOCK_KIND@
 OMP_LOCK_SIZE = @OMP_LOCK_SIZE@
+OMP_NEST_LOCK_25_ALIGN = @OMP_NEST_LOCK_25_ALIGN@
+OMP_NEST_LOCK_25_KIND = @OMP_NEST_LOCK_25_KIND@
+OMP_NEST_LOCK_25_SIZE = @OMP_NEST_LOCK_25_SIZE@
 OMP_NEST_LOCK_ALIGN = @OMP_NEST_LOCK_ALIGN@
 OMP_NEST_LOCK_KIND = @OMP_NEST_LOCK_KIND@
 OMP_NEST_LOCK_SIZE = @OMP_NEST_LOCK_SIZE@
diff --git a/libgomp/testsuite/libgomp.c++/c++.exp b/libgomp/testsuite/libgomp.c++/c++.exp
index f11482c7315..f3f42de6619 100644
--- a/libgomp/testsuite/libgomp.c++/c++.exp
+++ b/libgomp/testsuite/libgomp.c++/c++.exp
@@ -31,8 +31,15 @@ if { $lang_test_file_found } {
     set ld_library_path "$always_ld_library_path:${blddir}/${lang_library_path}"
     set_ld_library_path_env_vars
 
+    set flags_file "${blddir}/../libstdc++-v3/scripts/testsuite_flags"
+    if { [file exists $flags_file] } {
+	set libstdcxx_includes [exec sh $flags_file --build-includes]
+    } else {
+	set libstdcxx_includes ""
+    }
+
     # Main loop.
-    gfortran-dg-runtest $tests ""
+    gfortran-dg-runtest $tests $libstdcxx_includes
 }
 
 # All done.
diff --git a/libgomp/testsuite/libgomp.c++/collapse-1.C b/libgomp/testsuite/libgomp.c++/collapse-1.C
new file mode 100644
index 00000000000..132d35cf41d
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/collapse-1.C
@@ -0,0 +1,29 @@
+// { dg-do run }
+
+#include <string.h>
+#include <stdlib.h>
+
+int
+main ()
+{
+  int i, j, k, l = 0;
+  int a[3][3][3];
+
+  memset (a, '\0', sizeof (a));
+  #pragma omp parallel for collapse(4 - 1) schedule(static, 4)
+    for (i = 0; i < 2; i++)
+      for (j = 0; j < 2; j++)
+	for (k = 0; k < 2; k++)
+	  a[i][j][k] = i + j * 4 + k * 16;
+  #pragma omp parallel
+    {
+      #pragma omp for collapse(2) reduction(|:l) private (k)
+	for (i = 0; i < 2; i++)
+	  for (j = 0; j < 2; j++)
+	    for (k = 0; k < 2; k++)
+	      if (a[i][j][k] != i + j * 4 + k * 16)
+		l = 1;
+    }
+  if (l)
+    abort ();
+}
diff --git a/libgomp/testsuite/libgomp.c++/collapse-2.C b/libgomp/testsuite/libgomp.c++/collapse-2.C
new file mode 100644
index 00000000000..a42a1f07ffd
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/collapse-2.C
@@ -0,0 +1,371 @@
+// { dg-do run }
+
+#include <omp.h>
+typedef __PTRDIFF_TYPE__ ptrdiff_t;
+extern "C" void abort ();
+
+template <typename T>
+class I
+{
+public:
+  typedef ptrdiff_t difference_type;
+  I ();
+  ~I ();
+  I (T *);
+  I (const I &);
+  T &operator * ();
+  T *operator -> ();
+  T &operator [] (const difference_type &) const;
+  I &operator = (const I &);
+  I &operator ++ ();
+  I operator ++ (int);
+  I &operator -- ();
+  I operator -- (int);
+  I &operator += (const difference_type &);
+  I &operator -= (const difference_type &);
+  I operator + (const difference_type &) const;
+  I operator - (const difference_type &) const;
+  template <typename S> friend bool operator == (I<S> &, I<S> &);
+  template <typename S> friend bool operator == (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator < (I<S> &, I<S> &);
+  template <typename S> friend bool operator < (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator <= (I<S> &, I<S> &);
+  template <typename S> friend bool operator <= (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator > (I<S> &, I<S> &);
+  template <typename S> friend bool operator > (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator >= (I<S> &, I<S> &);
+  template <typename S> friend bool operator >= (const I<S> &, const I<S> &);
+  template <typename S> friend typename I<S>::difference_type operator - (I<S> &, I<S> &);
+  template <typename S> friend typename I<S>::difference_type operator - (const I<S> &, const I<S> &);
+  template <typename S> friend I<S> operator + (typename I<S>::difference_type , const I<S> &);
+private:
+  T *p;
+};
+template <typename T> I<T>::I () : p (0) {}
+template <typename T> I<T>::~I () { p = (T *) 0; }
+template <typename T> I<T>::I (T *x) : p (x) {}
+template <typename T> I<T>::I (const I &x) : p (x.p) {}
+template <typename T> T &I<T>::operator * () { return *p; }
+template <typename T> T *I<T>::operator -> () { return p; }
+template <typename T> T &I<T>::operator [] (const difference_type &x) const { return p[x]; }
+template <typename T> I<T> &I<T>::operator = (const I &x) { p = x.p; return *this; }
+template <typename T> I<T> &I<T>::operator ++ () { ++p; return *this; }
+template <typename T> I<T> I<T>::operator ++ (int) { return I (p++); }
+template <typename T> I<T> &I<T>::operator -- () { --p; return *this; }
+template <typename T> I<T> I<T>::operator -- (int) { return I (p--); }
+template <typename T> I<T> &I<T>::operator += (const difference_type &x) { p += x; return *this; }
+template <typename T> I<T> &I<T>::operator -= (const difference_type &x) { p -= x; return *this; }
+template <typename T> I<T> I<T>::operator + (const difference_type &x) const { return I (p + x); }
+template <typename T> I<T> I<T>::operator - (const difference_type &x) const { return I (p - x); }
+template <typename T> bool operator == (I<T> &x, I<T> &y) { return x.p == y.p; }
+template <typename T> bool operator == (const I<T> &x, const I<T> &y) { return x.p == y.p; }
+template <typename T> bool operator != (I<T> &x, I<T> &y) { return !(x == y); }
+template <typename T> bool operator != (const I<T> &x, const I<T> &y) { return !(x == y); }
+template <typename T> bool operator < (I<T> &x, I<T> &y) { return x.p < y.p; }
+template <typename T> bool operator < (const I<T> &x, const I<T> &y) { return x.p < y.p; }
+template <typename T> bool operator <= (I<T> &x, I<T> &y) { return x.p <= y.p; }
+template <typename T> bool operator <= (const I<T> &x, const I<T> &y) { return x.p <= y.p; }
+template <typename T> bool operator > (I<T> &x, I<T> &y) { return x.p > y.p; }
+template <typename T> bool operator > (const I<T> &x, const I<T> &y) { return x.p > y.p; }
+template <typename T> bool operator >= (I<T> &x, I<T> &y) { return x.p >= y.p; }
+template <typename T> bool operator >= (const I<T> &x, const I<T> &y) { return x.p >= y.p; }
+template <typename T> typename I<T>::difference_type operator - (I<T> &x, I<T> &y) { return x.p - y.p; }
+template <typename T> typename I<T>::difference_type operator - (const I<T> &x, const I<T> &y) { return x.p - y.p; }
+template <typename T> I<T> operator + (typename I<T>::difference_type x, const I<T> &y) { return I<T> (x + y.p); }
+
+template <typename T>
+class J
+{
+public:
+  J(const I<T> &x, const I<T> &y) : b (x), e (y) {}
+  const I<T> &begin ();
+  const I<T> &end ();
+private:
+  I<T> b, e;
+};
+
+template <typename T> const I<T> &J<T>::begin () { return b; }
+template <typename T> const I<T> &J<T>::end () { return e; }
+
+int results[2000];
+
+void
+f1 (J<int> x, J<int> y, J<int> z)
+{
+  I<int> i, j, k;
+  int l, f = 0, n = 0, m = 0;
+#pragma omp parallel shared (i, j, k, l) firstprivate (f) \
+		     reduction (+:n, m) num_threads (8)
+  {
+  #pragma omp for lastprivate (i, j, k, l) schedule (static, 9) \
+		  collapse (4)
+    for (i = x.begin (); i < x.end (); ++i)
+      for (j = y.begin (); j <= y.end (); j += 1)
+	for (l = 0; l < 1; l++)
+	  for (k = z.begin () + 3; k < z.end () - 3; k++)
+	    if (omp_get_num_threads () == 8
+		&& ((*i + 2) * 12 + (*j + 5) * 4 + (*k - 13)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+  }
+  if (n || i != x.end () || j != y.end () + 1 || k != z.end () - 3
+      || m != 72 || l != 1)
+    abort ();
+}
+
+void
+f2 (J<int> x, J<int> y, J<int> z)
+{
+  int f = 0, n = 0, m = 0;
+#pragma omp parallel for firstprivate (f) reduction (+:n, m) \
+			 num_threads (8) schedule (static, 9) \
+			 collapse (6 - 2)
+  for (I<int> i = x.end () - 1; i >= x.begin (); --i)
+    for (int l = -131; l >= -131; l--)
+      for (I<int> j = y.end (); j > y.begin () - 1; j -= 1)
+	{
+	  for (I<int> k = z.end () - 4; k >= z.begin () + 3; k--)
+	    if (omp_get_num_threads () == 8
+		&& ((3 - *i) * 12 + (-3 - *j) * 4 + (16 - *k)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+	}
+  if (n || m != 72)
+    abort ();
+}
+
+template <typename T>
+void
+f3 (J<int> x, J<int> y, J<int> z)
+{
+  I<int> i, j, k;
+  int l, f = 0, n = 0, m = 0;
+#pragma omp parallel shared (i, j, k, l) firstprivate (f) \
+		     reduction (+:n, m) num_threads (8)
+  {
+  #pragma omp for lastprivate (i, j, k, l) schedule (static, 9) \
+		  collapse (4)
+    for (i = x.begin (); i < x.end (); ++i)
+      for (j = y.begin (); j <= y.end (); j += 1)
+	for (k = z.begin () + 3; k < z.end () - 3; k++)
+	  for (l = 7; l <= 7; l++)
+	    if (omp_get_num_threads () == 8
+		&& ((*i + 2) * 12 + (*j + 5) * 4 + (*k - 13)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+  }
+  if (n || i != x.end () || j != y.end () + 1 || k != z.end () - 3
+      || m != 72 || l != 8)
+    abort ();
+}
+
+template <typename T>
+void
+f4 (J<int> x, J<int> y, J<int> z)
+{
+  int f = 0, n = 0, m = 0;
+#pragma omp parallel for firstprivate (f) reduction (+:n, m) \
+			 num_threads (8) schedule (static, 9) \
+			 collapse (5 - 2)
+  for (I<int> i = x.end () - 1; i >= x.begin (); --i)
+    {
+      for (I<int> j = y.end (); j > y.begin () - 1; j -= 1)
+	{
+	  for (I<int> k = z.end () - 4; k >= z.begin () + 3; k--)
+	    if (omp_get_num_threads () == 8
+		&& ((3 - *i) * 12 + (-3 - *j) * 4 + (16 - *k)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+	}
+    }
+  if (n || m != 72)
+    abort ();
+}
+
+template <typename T>
+void
+f5 (J<int> x, J<int> y, J<int> z)
+{
+  I<int> i, j, k;
+  int f = 0, n = 0, m = 0;
+#pragma omp parallel shared (i, j, k) firstprivate (f) \
+		     reduction (+:n, m) num_threads (8)
+  {
+  #pragma omp for lastprivate (i, j, k) schedule (static, 9) \
+		  collapse (3)
+    for (i = x.begin (); i < x.end (); ++i)
+      for (j = y.begin (); j <= y.end (); j += (T) 1)
+	{
+	  for (k = z.begin () + 3; k < z.end () - 3; k++)
+	    if (omp_get_num_threads () == 8
+		&& ((*i + 2) * 12 + (*j + 5) * 4 + (*k - 13)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+	}
+  }
+  if (n || i != x.end () || j != y.end () + 1 || k != z.end () - 3
+      || m != 72)
+    abort ();
+}
+
+template <typename T>
+void
+f6 (J<int> x, J<int> y, J<int> z)
+{
+  int f = 0, n = 0, m = 0;
+#pragma omp parallel for firstprivate (f) reduction (+:n, m) \
+			 num_threads (8) schedule (static, 9) \
+			 collapse (5 - 2)
+  for (I<int> i = x.end () - 1; i >= x.begin (); --i)
+    {
+      for (I<int> j = y.end (); j > y.begin () - 1; j -= 1)
+	{
+	  for (I<int> k = z.end () - 4; k >= z.begin () + (T) 3; k--)
+	    if (omp_get_num_threads () == 8
+		&& ((3 - *i) * 12 + (-3 - *j) * 4 + (16 - *k)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+	}
+    }
+  if (n || m != 72)
+    abort ();
+}
+
+template <typename T>
+void
+f7 (J<T> x, J<T> y, J<T> z)
+{
+  I<T> i, j, k, o = y.begin ();
+  T l, f = 0, n = 0, m = 0;
+#pragma omp parallel shared (i, j, k, l) firstprivate (f) \
+		     reduction (+:n, m) num_threads (8)
+  {
+  #pragma omp for lastprivate (i, j, k, l) schedule (static, 9) \
+		  collapse (4)
+    for (i = x.begin (); i < x.end (); ++i)
+      for (j = y.begin (); j <= y.end (); j += 1)
+	for (l = *o; l <= *o; l = 1 + l)
+	  for (k = z.begin () + 3; k < z.end () - 3; k++)
+	    if (omp_get_num_threads () == 8
+		&& ((*i + 2) * 12 + (*j + 5) * 4 + (*k - 13)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+  }
+  if (n || i != x.end () || j != y.end () + 1 || k != z.end () - 3
+      || m != 72 || l != *o + 1)
+    abort ();
+}
+
+template <typename T>
+void
+f8 (J<T> x, J<T> y, J<T> z)
+{
+  T f = 0, n = 0, m = 0;
+#pragma omp parallel for firstprivate (f) reduction (+:n, m) \
+			 num_threads (8) schedule (static, 9) \
+			 collapse (6 - 2)
+  for (I<T> i = x.end () - 1; i >= x.begin (); --i)
+    for (T l = 0; l < 1; l++)
+      for (I<T> j = y.end (); j > y.begin () - 1; j -= 1)
+	{
+	  for (I<T> k = z.end () - 4; k >= z.begin () + 3; k--)
+	    if (omp_get_num_threads () == 8
+		&& ((3 - *i) * 12 + (-3 - *j) * 4 + (16 - *k)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+	}
+  if (n || m != 72)
+    abort ();
+}
+
+template <typename S, typename T>
+void
+f9 (J<T> x, J<T> y, J<T> z)
+{
+  S i, j, k, o = y.begin ();
+  T l, f = 0, n = 0, m = 0;
+#pragma omp parallel shared (i, j, k, l) firstprivate (f) \
+		     reduction (+:n, m) num_threads (8)
+  {
+  #pragma omp for lastprivate (i, j, k, l) schedule (static, 9) \
+		  collapse (4)
+    for (i = x.begin (); i < x.end (); ++i)
+      for (j = y.begin (); j <= y.end (); j += 1)
+	for (l = *o; l <= *o; l = 1 + l)
+	  for (k = z.begin () + 3; k < z.end () - 3; k++)
+	    if (omp_get_num_threads () == 8
+		&& ((*i + 2) * 12 + (*j + 5) * 4 + (*k - 13)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+  }
+  if (n || i != x.end () || j != y.end () + 1 || k != z.end () - 3
+      || m != 72 || l != *o + 1)
+    abort ();
+}
+
+template <typename S, typename T>
+void
+f10 (J<T> x, J<T> y, J<T> z)
+{
+  T f = 0, n = 0, m = 0;
+#pragma omp parallel for firstprivate (f) reduction (+:n, m) \
+			 num_threads (8) schedule (static, 9) \
+			 collapse (6 - 2)
+  for (S i = x.end () - 1; i >= x.begin (); --i)
+    for (T l = 0; l < 1; l++)
+      for (S j = y.end (); j > y.begin () - 1; j -= 1)
+	{
+	  for (S k = z.end () - 4; k >= z.begin () + 3; k--)
+	    if (omp_get_num_threads () == 8
+		&& ((3 - *i) * 12 + (-3 - *j) * 4 + (16 - *k)
+		    != (omp_get_thread_num () * 9 + f++)))
+	      n++;
+	    else
+	      m++;
+	}
+  if (n || m != 72)
+    abort ();
+}
+
+int
+main ()
+{
+  int a[2000];
+  long b[2000];
+  for (int i = 0; i < 2000; i++)
+    {
+      a[i] = i - 1000;
+      b[i] = i - 1000;
+    }
+  J<int> x (&a[998], &a[1004]);
+  J<int> y (&a[995], &a[997]);
+  J<int> z (&a[1010], &a[1020]);
+  f1 (x, y, z);
+  f2 (x, y, z);
+  f3 <int> (x, y, z);
+  f4 <int> (x, y, z);
+  f5 <int> (x, y, z);
+  f6 <int> (x, y, z);
+  f7 <int> (x, y, z);
+  f8 <int> (x, y, z);
+  f9 <I<int>, int> (x, y, z);
+  f10 <I<int>, int> (x, y, z);
+}
diff --git a/libgomp/testsuite/libgomp.c++/ctor-10.C b/libgomp/testsuite/libgomp.c++/ctor-10.C
new file mode 100644
index 00000000000..f46e45ec418
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/ctor-10.C
@@ -0,0 +1,78 @@
+// { dg-do run }
+// { dg-require-effective-target tls_runtime }
+
+#include <omp.h>
+#include <assert.h>
+
+#define N 10
+#define THR 4
+
+struct B
+{
+  B();
+  B(const B &);
+  ~B();
+  B& operator=(const B &);
+  void doit();
+  static B *base;
+  static B *threadbase;
+#pragma omp threadprivate(threadbase)
+};
+
+B *B::base;
+B *B::threadbase;
+static unsigned cmask[THR];
+static unsigned dmask[THR];
+
+B::B()
+{
+  assert (base == 0);
+}
+
+B::B(const B &b)
+{
+  unsigned index = &b - base;
+  assert (index < N);
+  cmask[omp_get_thread_num()] |= 1u << index;
+}
+
+B::~B()
+{
+  if (threadbase)
+    {
+      unsigned index = this - threadbase;
+      assert (index < N);
+      dmask[omp_get_thread_num()] |= 1u << index;
+    }
+}
+
+void foo()
+{
+  B b[N];
+
+  B::base = b;
+
+  #pragma omp parallel firstprivate(b)
+    {
+      assert (omp_get_num_threads () == THR);
+      B::threadbase = b;
+    }
+
+  B::threadbase = 0;
+}
+
+int main()
+{
+  omp_set_dynamic (0);
+  omp_set_num_threads (THR);
+  foo();
+
+  for (int i = 0; i < THR; ++i)
+    {
+      unsigned xmask = (1u << N) - 1;
+      assert (cmask[i] == xmask);
+      assert (dmask[i] == xmask);
+    }
+
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c++/for-1.C b/libgomp/testsuite/libgomp.c++/for-1.C
new file mode 100644
index 00000000000..1c713464ebe
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/for-1.C
@@ -0,0 +1,291 @@
+// { dg-do run }
+
+typedef __PTRDIFF_TYPE__ ptrdiff_t;
+extern "C" void abort ();
+
+template <typename T>
+class I
+{
+public:
+  typedef ptrdiff_t difference_type;
+  I ();
+  ~I ();
+  I (T *);
+  I (const I &);
+  T &operator * ();
+  T *operator -> ();
+  T &operator [] (const difference_type &) const;
+  I &operator = (const I &);
+  I &operator ++ ();
+  I operator ++ (int);
+  I &operator -- ();
+  I operator -- (int);
+  I &operator += (const difference_type &);
+  I &operator -= (const difference_type &);
+  I operator + (const difference_type &) const;
+  I operator - (const difference_type &) const;
+  template <typename S> friend bool operator == (I<S> &, I<S> &);
+  template <typename S> friend bool operator == (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator < (I<S> &, I<S> &);
+  template <typename S> friend bool operator < (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator <= (I<S> &, I<S> &);
+  template <typename S> friend bool operator <= (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator > (I<S> &, I<S> &);
+  template <typename S> friend bool operator > (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator >= (I<S> &, I<S> &);
+  template <typename S> friend bool operator >= (const I<S> &, const I<S> &);
+  template <typename S> friend typename I<S>::difference_type operator - (I<S> &, I<S> &);
+  template <typename S> friend typename I<S>::difference_type operator - (const I<S> &, const I<S> &);
+  template <typename S> friend I<S> operator + (typename I<S>::difference_type , const I<S> &);
+private:
+  T *p;
+};
+template <typename T> I<T>::I () : p (0) {}
+template <typename T> I<T>::~I () {}
+template <typename T> I<T>::I (T *x) : p (x) {}
+template <typename T> I<T>::I (const I &x) : p (x.p) {}
+template <typename T> T &I<T>::operator * () { return *p; }
+template <typename T> T *I<T>::operator -> () { return p; }
+template <typename T> T &I<T>::operator [] (const difference_type &x) const { return p[x]; }
+template <typename T> I<T> &I<T>::operator = (const I &x) { p = x.p; return *this; }
+template <typename T> I<T> &I<T>::operator ++ () { ++p; return *this; }
+template <typename T> I<T> I<T>::operator ++ (int) { return I (p++); }
+template <typename T> I<T> &I<T>::operator -- () { --p; return *this; }
+template <typename T> I<T> I<T>::operator -- (int) { return I (p--); }
+template <typename T> I<T> &I<T>::operator += (const difference_type &x) { p += x; return *this; }
+template <typename T> I<T> &I<T>::operator -= (const difference_type &x) { p -= x; return *this; }
+template <typename T> I<T> I<T>::operator + (const difference_type &x) const { return I (p + x); }
+template <typename T> I<T> I<T>::operator - (const difference_type &x) const { return I (p - x); }
+template <typename T> bool operator == (I<T> &x, I<T> &y) { return x.p == y.p; }
+template <typename T> bool operator == (const I<T> &x, const I<T> &y) { return x.p == y.p; }
+template <typename T> bool operator != (I<T> &x, I<T> &y) { return !(x == y); }
+template <typename T> bool operator != (const I<T> &x, const I<T> &y) { return !(x == y); }
+template <typename T> bool operator < (I<T> &x, I<T> &y) { return x.p < y.p; }
+template <typename T> bool operator < (const I<T> &x, const I<T> &y) { return x.p < y.p; }
+template <typename T> bool operator <= (I<T> &x, I<T> &y) { return x.p <= y.p; }
+template <typename T> bool operator <= (const I<T> &x, const I<T> &y) { return x.p <= y.p; }
+template <typename T> bool operator > (I<T> &x, I<T> &y) { return x.p > y.p; }
+template <typename T> bool operator > (const I<T> &x, const I<T> &y) { return x.p > y.p; }
+template <typename T> bool operator >= (I<T> &x, I<T> &y) { return x.p >= y.p; }
+template <typename T> bool operator >= (const I<T> &x, const I<T> &y) { return x.p >= y.p; }
+template <typename T> typename I<T>::difference_type operator - (I<T> &x, I<T> &y) { return x.p - y.p; }
+template <typename T> typename I<T>::difference_type operator - (const I<T> &x, const I<T> &y) { return x.p - y.p; }
+template <typename T> I<T> operator + (typename I<T>::difference_type x, const I<T> &y) { return I<T> (x + y.p); }
+
+template <typename T>
+class J
+{
+public:
+  J(const I<T> &x, const I<T> &y) : b (x), e (y) {}
+  const I<T> &begin ();
+  const I<T> &end ();
+private:
+  I<T> b, e;
+};
+
+template <typename T> const I<T> &J<T>::begin () { return b; }
+template <typename T> const I<T> &J<T>::end () { return e; }
+
+int results[2000];
+
+template <typename T>
+void
+baz (I<T> &i)
+{
+  if (*i < 0 || *i >= 2000)
+    abort ();
+  results[*i]++;
+}
+
+void
+f1 (const I<int> &x, const I<int> &y)
+{
+#pragma omp parallel for
+  for (I<int> i = x; i <= y; i += 6)
+    baz (i);
+}
+
+void
+f2 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel for private(i)
+  for (i = x; i < y - 1; i = 1 - 6 + 7 + i)
+    baz (i);
+}
+
+template <typename T>
+void
+f3 (const I<int> &x, const I<int> &y)
+{
+#pragma omp parallel for
+  for (I<int> i = x; i <= y; i = i + 9 - 8)
+    baz (i);
+}
+
+template <typename T>
+void
+f4 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel for lastprivate(i)
+  for (i = x + 2000 - 64; i > y + 10; --i)
+    baz (i);
+}
+
+void
+f5 (const I<int> &x, const I<int> &y)
+{
+#pragma omp parallel for
+  for (I<int> i = x + 2000 - 64; i > y + 10; i -= 10)
+    baz (i);
+}
+
+template <int N>
+void
+f6 (const I<int> &x, const I<int> &y)
+{
+#pragma omp parallel for
+  for (I<int> i = x + 2000 - 64; i > y + 10; i = i - 12 + 2)
+    {
+      I<int> j = i + N;
+      baz (j);
+    }
+}
+
+template <int N>
+void
+f7 (I<int> i, const I<int> &x, const I<int> &y)
+{
+#pragma omp parallel for
+  for (i = x - 10; i <= y + 10; i += N)
+    baz (i);
+}
+
+template <int N>
+void
+f8 (J<int> j)
+{
+  I<int> i;
+#pragma omp parallel for
+  for (i = j.begin (); i <= j.end () + N; i += 2)
+    baz (i);
+}
+
+template <typename T, int N>
+void
+f9 (const I<T> &x, const I<T> &y)
+{
+#pragma omp parallel for
+  for (I<T> i = x; i <= y; i = i + N)
+    baz (i);
+}
+
+template <typename T, int N>
+void
+f10 (const I<T> &x, const I<T> &y)
+{
+  I<T> i;
+#pragma omp parallel for
+  for (i = x; i > y; i = i + N)
+    baz (i);
+}
+
+template <typename T>
+void
+f11 (const T &x, const T &y)
+{
+#pragma omp parallel
+  {
+#pragma omp for nowait
+    for (T i = x; i <= y; i += 3)
+      baz (i);
+#pragma omp single
+    {
+      T j = y + 3;
+      baz (j);
+    }
+  }
+}
+
+template <typename T>
+void
+f12 (const T &x, const T &y)
+{
+  T i;
+#pragma omp parallel for
+  for (i = x; i > y; --i)
+    baz (i);
+}
+
+template <int N>
+struct K
+{
+  template <typename T>
+  static void
+  f13 (const T &x, const T &y)
+  {
+#pragma omp parallel for
+    for (T i = x; i <= y + N; i += N)
+      baz (i);
+  }
+};
+
+#define check(expr) \
+  for (int i = 0; i < 2000; i++)			\
+    if (expr)						\
+      {							\
+	if (results[i] != 1)				\
+	  abort ();					\
+	results[i] = 0;					\
+      }							\
+    else if (results[i])				\
+      abort ()
+
+int
+main ()
+{
+  int a[2000];
+  long b[2000];
+  for (int i = 0; i < 2000; i++)
+    {
+      a[i] = i;
+      b[i] = i;
+    }
+  f1 (&a[10], &a[1990]);
+  check (i >= 10 && i <= 1990 && (i - 10) % 6 == 0);
+  f2 (&a[0], &a[1999]);
+  check (i < 1998 && (i & 1) == 0);
+  f3<char> (&a[20], &a[1837]);
+  check (i >= 20 && i <= 1837);
+  f4<int> (&a[0], &a[30]);
+  check (i > 40 && i <= 2000 - 64);
+  f5 (&a[0], &a[100]);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f6<-10> (&a[10], &a[110]);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f7<6> (I<int> (), &a[12], &a[1800]);
+  check (i >= 2 && i <= 1808 && (i - 2) % 6 == 0);
+  f8<121> (J<int> (&a[14], &a[1803]));
+  check (i >= 14 && i <= 1924 && (i & 1) == 0);
+  f9<int, 7> (&a[33], &a[1967]);
+  check (i >= 33 && i <= 1967 && (i - 33) % 7 == 0);
+  f10<int, -7> (&a[1939], &a[17]);
+  check (i >= 21 && i <= 1939 && (i - 21) % 7 == 0);
+  f11<I<int> > (&a[16], &a[1981]);
+  check (i >= 16 && i <= 1984 && (i - 16) % 3 == 0);
+  f12<I<int> > (&a[1761], &a[37]);
+  check (i > 37 && i <= 1761);
+  K<5>::f13<I<int> > (&a[1], &a[1935]);
+  check (i >= 1 && i <= 1936 && (i - 1) % 5 == 0);
+  f9<long, 7> (&b[33], &b[1967]);
+  check (i >= 33 && i <= 1967 && (i - 33) % 7 == 0);
+  f10<long, -7> (&b[1939], &b[17]);
+  check (i >= 21 && i <= 1939 && (i - 21) % 7 == 0);
+  f11<I<long> > (&b[16], &b[1981]);
+  check (i >= 16 && i <= 1984 && (i - 16) % 3 == 0);
+  f12<I<long> > (&b[1761], &b[37]);
+  check (i > 37 && i <= 1761);
+  K<5>::f13<I<long> > (&b[1], &b[1935]);
+  check (i >= 1 && i <= 1936 && (i - 1) % 5 == 0);
+}
diff --git a/libgomp/testsuite/libgomp.c++/for-2.C b/libgomp/testsuite/libgomp.c++/for-2.C
new file mode 100644
index 00000000000..98ffa1ae6f0
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/for-2.C
@@ -0,0 +1,182 @@
+// { dg-do run }
+
+extern "C" void abort ();
+
+template <typename T>
+class J
+{
+public:
+  J(T x, T y) : b (x), e (y) {}
+  T begin ();
+  T end ();
+private:
+  T b, e;
+};
+
+template <typename T> T J<T>::begin () { return b; }
+template <typename T> T J<T>::end () { return e; }
+
+int results[2000];
+
+void
+baz (int i)
+{
+  if (i < 0 || i >= 2000)
+    abort ();
+  results[i]++;
+}
+
+void
+f1 (int x, int y)
+{
+#pragma omp parallel for
+  for (int i = x; i <= y; i += 6)
+    baz (i);
+}
+
+void
+f2 (int x, int y)
+{
+  int i;
+#pragma omp parallel for private(i)
+  for (i = x; i < y - 1; i = 1 - 6 + 7 + i)
+    baz (i);
+}
+
+template <typename T>
+void
+f3 (int x, int y)
+{
+#pragma omp parallel for
+  for (int i = x; i <= y; i = i + 9 - 8)
+    baz (i);
+}
+
+template <typename T>
+void
+f4 (int x, int y)
+{
+  int i;
+#pragma omp parallel for lastprivate(i)
+  for (i = x + 2000 - 64; i > y + 10; --i)
+    baz (i);
+}
+
+void
+f5 (int x, int y)
+{
+#pragma omp parallel for
+  for (int i = x + 2000 - 64; i > y + 10L; i -= 10L)
+    baz (i);
+}
+
+template <int N>
+void
+f6 (int x, int y)
+{
+#pragma omp parallel for
+  for (int i = x + 2000 - 64; i > y + 10L; i = i - 12 + 2L)
+    baz (i + N);
+}
+
+template <long N>
+void
+f7 (int i, int x, int y)
+{
+#pragma omp parallel for
+  for (i = x - 10; i <= y + 10; i += N)
+    baz (i);
+}
+
+template <long N>
+void
+f8 (J<int> j)
+{
+  int i;
+#pragma omp parallel for
+  for (i = j.begin (); i <= j.end () + N; i += 2)
+    baz (i);
+}
+
+template <typename T, long N>
+void
+f9 (T x, T y)
+{
+#pragma omp parallel for
+  for (T i = x; i <= y; i = i + N)
+    baz (i);
+}
+
+template <typename T, long N>
+void
+f10 (T x, T y)
+{
+  T i;
+#pragma omp parallel for
+  for (i = x; i > y; i = i + N)
+    baz (i);
+}
+
+template <typename T>
+void
+f11 (T x, long y)
+{
+#pragma omp parallel
+  {
+#pragma omp for nowait
+    for (T i = x; i <= y; i += 3L)
+      baz (i);
+#pragma omp single
+    baz (y + 3);
+  }
+}
+
+template <typename T>
+void
+f12 (T x, T y)
+{
+  T i;
+#pragma omp parallel for
+  for (i = x; i > y; --i)
+    baz (i);
+}
+
+#define check(expr) \
+  for (int i = 0; i < 2000; i++)			\
+    if (expr)						\
+      {							\
+	if (results[i] != 1)				\
+	  abort ();					\
+	results[i] = 0;					\
+      }							\
+    else if (results[i])				\
+      abort ()
+
+int
+main ()
+{
+  f1 (10, 1990);
+  check (i >= 10 && i <= 1990 && (i - 10) % 6 == 0);
+  f2 (0, 1999);
+  check (i < 1998 && (i & 1) == 0);
+  f3<char> (20, 1837);
+  check (i >= 20 && i <= 1837);
+  f4<int> (0, 30);
+  check (i > 40 && i <= 2000 - 64);
+  f5 (0, 100);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f6<-10> (10, 110);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f7<6> (0, 12, 1800);
+  check (i >= 2 && i <= 1808 && (i - 2) % 6 == 0);
+  f8<121> (J<int> (14, 1803));
+  check (i >= 14 && i <= 1924 && (i & 1) == 0);
+  f9<int, 7> (33, 1967);
+  check (i >= 33 && i <= 1967 && (i - 33) % 7 == 0);
+  f10<int, -7> (1939, 17);
+  check (i >= 21 && i <= 1939 && (i - 21) % 7 == 0);
+  f11<int> (16, 1981);
+  check (i >= 16 && i <= 1984 && (i - 16) % 3 == 0);
+  f12<int> (1761, 37);
+  check (i > 37 && i <= 1761);
+}
diff --git a/libgomp/testsuite/libgomp.c++/for-3.C b/libgomp/testsuite/libgomp.c++/for-3.C
new file mode 100644
index 00000000000..235f83875ea
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/for-3.C
@@ -0,0 +1,239 @@
+// { dg-do run }
+
+#include <vector>
+#include <cstdlib>
+
+template <typename T>
+class J
+{
+public:
+  typedef typename std::vector<T>::const_iterator const_iterator;
+  J(const const_iterator &x, const const_iterator &y) : b (x), e (y) {}
+  const const_iterator &begin ();
+  const const_iterator &end ();
+private:
+  const_iterator b, e;
+};
+
+template <typename T>
+const typename std::vector<T>::const_iterator &J<T>::begin () { return b; }
+template <typename T>
+const typename std::vector<T>::const_iterator &J<T>::end () { return e; }
+
+int results[2000];
+
+template <typename T>
+void
+baz (T &i)
+{
+  if (*i < 0 || *i >= 2000)
+    std::abort ();
+  results[*i]++;
+}
+
+void
+f1 (const std::vector<int>::const_iterator &x,
+    const std::vector<int>::const_iterator &y)
+{
+#pragma omp parallel for
+  for (std::vector<int>::const_iterator i = x; i <= y; i += 6)
+    baz (i);
+}
+
+void
+f2 (const std::vector<int>::const_iterator &x,
+    const std::vector<int>::const_iterator &y)
+{
+  std::vector<int>::const_iterator i;
+#pragma omp parallel for private(i)
+  for (i = x; i < y - 1; i = 1 - 6 + 7 + i)
+    baz (i);
+}
+
+template <typename T>
+void
+f3 (const std::vector<int>::const_iterator &x,
+    const std::vector<int>::const_iterator &y)
+{
+#pragma omp parallel for schedule (dynamic, 6)
+  for (std::vector<int>::const_iterator i = x; i <= y; i = i + 9 - 8)
+    baz (i);
+}
+
+template <typename T>
+void
+f4 (const std::vector<int>::const_iterator &x,
+    const std::vector<int>::const_iterator &y)
+{
+  std::vector<int>::const_iterator i;
+#pragma omp parallel for lastprivate(i)
+  for (i = x + 2000 - 64; i > y + 10; --i)
+    baz (i);
+}
+
+void
+f5 (const std::vector<int>::const_iterator &x,
+    const std::vector<int>::const_iterator &y)
+{
+#pragma omp parallel for schedule (static, 10)
+  for (std::vector<int>::const_iterator i = x + 2000 - 64; i > y + 10; i -= 10)
+    baz (i);
+}
+
+template <int N>
+void
+f6 (const std::vector<int>::const_iterator &x,
+    const std::vector<int>::const_iterator &y)
+{
+#pragma omp parallel for schedule (runtime)
+  for (std::vector<int>::const_iterator i = x + 2000 - 64;
+       i > y + 10; i = i - 12 + 2)
+    {
+      std::vector<int>::const_iterator j = i + N;
+      baz (j);
+    }
+}
+
+template <int N>
+void
+f7 (std::vector<int>::const_iterator i,
+    const std::vector<int>::const_iterator &x,
+    const std::vector<int>::const_iterator &y)
+{
+#pragma omp parallel for schedule (dynamic, 6)
+  for (i = x - 10; i <= y + 10; i += N)
+    baz (i);
+}
+
+template <int N>
+void
+f8 (J<int> j)
+{
+  std::vector<int>::const_iterator i;
+#pragma omp parallel for schedule (dynamic, 40)
+  for (i = j.begin (); i <= j.end () + N; i += 2)
+    baz (i);
+}
+
+template <typename T, int N>
+void
+f9 (const typename std::vector<T>::const_iterator &x,
+    const typename std::vector<T>::const_iterator &y)
+{
+#pragma omp parallel for schedule (static, 25)
+  for (typename std::vector<T>::const_iterator i = x; i <= y; i = i + N)
+    baz (i);
+}
+
+template <typename T, int N>
+void
+f10 (const typename std::vector<T>::const_iterator &x,
+     const typename std::vector<T>::const_iterator &y)
+{
+  typename std::vector<T>::const_iterator i;
+#pragma omp parallel for
+  for (i = x; i > y; i = i + N)
+    baz (i);
+}
+
+template <typename T>
+void
+f11 (const T &x, const T &y)
+{
+#pragma omp parallel
+  {
+#pragma omp for nowait schedule (static, 2)
+    for (T i = x; i <= y; i += 3)
+      baz (i);
+#pragma omp single
+    {
+      T j = y + 3;
+      baz (j);
+    }
+  }
+}
+
+template <typename T>
+void
+f12 (const T &x, const T &y)
+{
+  T i;
+#pragma omp parallel for schedule (dynamic, 130)
+  for (i = x; i > y; --i)
+    baz (i);
+}
+
+template <int N>
+struct K
+{
+  template <typename T>
+  static void
+  f13 (const T &x, const T &y)
+  {
+#pragma omp parallel for schedule (runtime)
+    for (T i = x; i <= y + N; i += N)
+      baz (i);
+  }
+};
+
+#define check(expr) \
+  for (int i = 0; i < 2000; i++)			\
+    if (expr)						\
+      {							\
+	if (results[i] != 1)				\
+	  std::abort ();				\
+	results[i] = 0;					\
+      }							\
+    else if (results[i])				\
+      std::abort ()
+
+int
+main ()
+{
+  std::vector<int> a(2000);
+  std::vector<long> b(2000);
+  for (int i = 0; i < 2000; i++)
+    {
+      a[i] = i;
+      b[i] = i;
+    }
+  f1 (a.begin () + 10, a.begin () + 1990);
+  check (i >= 10 && i <= 1990 && (i - 10) % 6 == 0);
+  f2 (a.begin () + 0, a.begin () + 1999);
+  check (i < 1998 && (i & 1) == 0);
+  f3<char> (a.begin () + 20, a.begin () + 1837);
+  check (i >= 20 && i <= 1837);
+  f4<int> (a.begin () + 0, a.begin () + 30);
+  check (i > 40 && i <= 2000 - 64);
+  f5 (a.begin () + 0, a.begin () + 100);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f6<-10> (a.begin () + 10, a.begin () + 110);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f7<6> (std::vector<int>::const_iterator (), a.begin () + 12,
+	 a.begin () + 1800);
+  check (i >= 2 && i <= 1808 && (i - 2) % 6 == 0);
+  f8<121> (J<int> (a.begin () + 14, a.begin () + 1803));
+  check (i >= 14 && i <= 1924 && (i & 1) == 0);
+  f9<int, 7> (a.begin () + 33, a.begin () + 1967);
+  check (i >= 33 && i <= 1967 && (i - 33) % 7 == 0);
+  f10<int, -7> (a.begin () + 1939, a.begin () + 17);
+  check (i >= 21 && i <= 1939 && (i - 21) % 7 == 0);
+  f11<std::vector<int>::const_iterator > (a.begin () + 16, a.begin () + 1981);
+  check (i >= 16 && i <= 1984 && (i - 16) % 3 == 0);
+  f12<std::vector<int>::const_iterator > (a.begin () + 1761, a.begin () + 37);
+  check (i > 37 && i <= 1761);
+  K<5>::f13<std::vector<int>::const_iterator > (a.begin () + 1,
+						a.begin () + 1935);
+  check (i >= 1 && i <= 1936 && (i - 1) % 5 == 0);
+  f9<long, 7> (b.begin () + 33, b.begin () + 1967);
+  check (i >= 33 && i <= 1967 && (i - 33) % 7 == 0);
+  f10<long, -7> (b.begin () + 1939, b.begin () + 17);
+  check (i >= 21 && i <= 1939 && (i - 21) % 7 == 0);
+  f11<std::vector<long>::const_iterator > (b.begin () + 16, b.begin () + 1981);
+  check (i >= 16 && i <= 1984 && (i - 16) % 3 == 0);
+  f12<std::vector<long>::const_iterator > (b.begin () + 1761, b.begin () + 37);
+  check (i > 37 && i <= 1761);
+  K<5>::f13<std::vector<long>::const_iterator > (b.begin () + 1,
+						 b.begin () + 1935);
+  check (i >= 1 && i <= 1936 && (i - 1) % 5 == 0);
+}
diff --git a/libgomp/testsuite/libgomp.c++/for-4.C b/libgomp/testsuite/libgomp.c++/for-4.C
new file mode 100644
index 00000000000..c528ef9d1fa
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/for-4.C
@@ -0,0 +1,225 @@
+// { dg-do run }
+
+#include <string>
+#include <cstdlib>
+
+template <typename T>
+class J
+{
+public:
+  typedef typename std::basic_string<T>::iterator iterator;
+  J(const iterator &x, const iterator &y) : b (x), e (y) {}
+  const iterator &begin ();
+  const iterator &end ();
+private:
+  iterator b, e;
+};
+
+template <typename T>
+const typename std::basic_string<T>::iterator &J<T>::begin () { return b; }
+template <typename T>
+const typename std::basic_string<T>::iterator &J<T>::end () { return e; }
+
+template <typename T>
+void
+baz (T &i)
+{
+  if (*i < L'a' || *i >= L'a' + 2000)
+    std::abort ();
+  (*i)++;
+}
+
+void
+f1 (const std::basic_string<wchar_t>::iterator &x,
+    const std::basic_string<wchar_t>::iterator &y)
+{
+#pragma omp parallel for
+  for (std::basic_string<wchar_t>::iterator i = x; i <= y; i += 6)
+    baz (i);
+}
+
+void
+f2 (const std::basic_string<wchar_t>::iterator &x,
+    const std::basic_string<wchar_t>::iterator &y)
+{
+  std::basic_string<wchar_t>::iterator i;
+#pragma omp parallel for private(i)
+  for (i = x; i < y - 1; i = 1 - 6 + 7 + i)
+    baz (i);
+}
+
+template <typename T>
+void
+f3 (const std::basic_string<wchar_t>::iterator &x,
+    const std::basic_string<wchar_t>::iterator &y)
+{
+#pragma omp parallel for schedule (dynamic, 6)
+  for (std::basic_string<wchar_t>::iterator i = x; i <= y; i = i + 9 - 8)
+    baz (i);
+}
+
+template <typename T>
+void
+f4 (const std::basic_string<wchar_t>::iterator &x,
+    const std::basic_string<wchar_t>::iterator &y)
+{
+  std::basic_string<wchar_t>::iterator i;
+#pragma omp parallel for lastprivate(i)
+  for (i = x + 2000 - 64; i > y + 10; --i)
+    baz (i);
+}
+
+void
+f5 (const std::basic_string<wchar_t>::iterator &x,
+    const std::basic_string<wchar_t>::iterator &y)
+{
+#pragma omp parallel for schedule (static, 10)
+  for (std::basic_string<wchar_t>::iterator i = x + 2000 - 64;
+       i > y + 10; i -= 10)
+    baz (i);
+}
+
+template <int N>
+void
+f6 (const std::basic_string<wchar_t>::iterator &x,
+    const std::basic_string<wchar_t>::iterator &y)
+{
+#pragma omp parallel for schedule (runtime)
+  for (std::basic_string<wchar_t>::iterator i = x + 2000 - 64;
+       i > y + 10; i = i - 12 + 2)
+    {
+      std::basic_string<wchar_t>::iterator j = i + N;
+      baz (j);
+    }
+}
+
+template <int N>
+void
+f7 (std::basic_string<wchar_t>::iterator i,
+    const std::basic_string<wchar_t>::iterator &x,
+    const std::basic_string<wchar_t>::iterator &y)
+{
+#pragma omp parallel for schedule (dynamic, 6)
+  for (i = x - 10; i <= y + 10; i += N)
+    baz (i);
+}
+
+template <wchar_t N>
+void
+f8 (J<wchar_t> j)
+{
+  std::basic_string<wchar_t>::iterator i;
+#pragma omp parallel for schedule (dynamic, 40)
+  for (i = j.begin (); i <= j.end () + N; i += 2)
+    baz (i);
+}
+
+template <typename T, int N>
+void
+f9 (const typename std::basic_string<T>::iterator &x,
+    const typename std::basic_string<T>::iterator &y)
+{
+#pragma omp parallel for schedule (static, 25)
+  for (typename std::basic_string<T>::iterator i = x; i <= y; i = i + N)
+    baz (i);
+}
+
+template <typename T, int N>
+void
+f10 (const typename std::basic_string<T>::iterator &x,
+     const typename std::basic_string<T>::iterator &y)
+{
+  typename std::basic_string<T>::iterator i;
+#pragma omp parallel for
+  for (i = x; i > y; i = i + N)
+    baz (i);
+}
+
+template <typename T>
+void
+f11 (const T &x, const T &y)
+{
+#pragma omp parallel
+  {
+#pragma omp for nowait schedule (static, 2)
+    for (T i = x; i <= y; i += 3)
+      baz (i);
+#pragma omp single
+    {
+      T j = y + 3;
+      baz (j);
+    }
+  }
+}
+
+template <typename T>
+void
+f12 (const T &x, const T &y)
+{
+  T i;
+#pragma omp parallel for schedule (dynamic, 130)
+  for (i = x; i > y; --i)
+    baz (i);
+}
+
+template <int N>
+struct K
+{
+  template <typename T>
+  static void
+  f13 (const T &x, const T &y)
+  {
+#pragma omp parallel for schedule (runtime)
+    for (T i = x; i <= y + N; i += N)
+      baz (i);
+  }
+};
+
+#define check(expr) \
+  for (int i = 0; i < 2000; i++)			\
+    if (expr)						\
+      {							\
+	if (a[i] != L'a' + i + 1)			\
+	  std::abort ();				\
+	a[i] = L'a' + i;				\
+      }							\
+    else if (a[i] != L'a' + i)				\
+      std::abort ()
+
+int
+main ()
+{
+  std::basic_string<wchar_t> a = L"";
+  for (int i = 0; i < 2000; i++)
+    a += L'a' + i;
+  f1 (a.begin () + 10, a.begin () + 1990);
+  check (i >= 10 && i <= 1990 && (i - 10) % 6 == 0);
+  f2 (a.begin () + 0, a.begin () + 1999);
+  check (i < 1998 && (i & 1) == 0);
+  f3<char> (a.begin () + 20, a.begin () + 1837);
+  check (i >= 20 && i <= 1837);
+  f4<int> (a.begin () + 0, a.begin () + 30);
+  check (i > 40 && i <= 2000 - 64);
+  f5 (a.begin () + 0, a.begin () + 100);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f6<-10> (a.begin () + 10, a.begin () + 110);
+  check (i >= 116 && i <= 2000 - 64 && (i - 116) % 10 == 0);
+  f7<6> (std::basic_string<wchar_t>::iterator (), a.begin () + 12,
+	 a.begin () + 1800);
+  check (i >= 2 && i <= 1808 && (i - 2) % 6 == 0);
+  f8<121> (J<wchar_t> (a.begin () + 14, a.begin () + 1803));
+  check (i >= 14 && i <= 1924 && (i & 1) == 0);
+  f9<wchar_t, 7> (a.begin () + 33, a.begin () + 1967);
+  check (i >= 33 && i <= 1967 && (i - 33) % 7 == 0);
+  f10<wchar_t, -7> (a.begin () + 1939, a.begin () + 17);
+  check (i >= 21 && i <= 1939 && (i - 21) % 7 == 0);
+  f11<std::basic_string<wchar_t>::iterator > (a.begin () + 16,
+					      a.begin () + 1981);
+  check (i >= 16 && i <= 1984 && (i - 16) % 3 == 0);
+  f12<std::basic_string<wchar_t>::iterator > (a.begin () + 1761,
+					      a.begin () + 37);
+  check (i > 37 && i <= 1761);
+  K<5>::f13<std::basic_string<wchar_t>::iterator > (a.begin () + 1,
+						    a.begin () + 1935);
+  check (i >= 1 && i <= 1936 && (i - 1) % 5 == 0);
+}
diff --git a/libgomp/testsuite/libgomp.c++/for-5.C b/libgomp/testsuite/libgomp.c++/for-5.C
new file mode 100644
index 00000000000..9b75bf379ce
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/for-5.C
@@ -0,0 +1,303 @@
+// { dg-do run }
+
+typedef __PTRDIFF_TYPE__ ptrdiff_t;
+extern "C" void abort ();
+
+template <typename T>
+class I
+{
+public:
+  typedef ptrdiff_t difference_type;
+  I ();
+  ~I ();
+  I (T *);
+  I (const I &);
+  T &operator * ();
+  T *operator -> ();
+  T &operator [] (const difference_type &) const;
+  I &operator = (const I &);
+  I &operator ++ ();
+  I operator ++ (int);
+  I &operator -- ();
+  I operator -- (int);
+  I &operator += (const difference_type &);
+  I &operator -= (const difference_type &);
+  I operator + (const difference_type &) const;
+  I operator - (const difference_type &) const;
+  template <typename S> friend bool operator == (I<S> &, I<S> &);
+  template <typename S> friend bool operator == (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator < (I<S> &, I<S> &);
+  template <typename S> friend bool operator < (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator <= (I<S> &, I<S> &);
+  template <typename S> friend bool operator <= (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator > (I<S> &, I<S> &);
+  template <typename S> friend bool operator > (const I<S> &, const I<S> &);
+  template <typename S> friend bool operator >= (I<S> &, I<S> &);
+  template <typename S> friend bool operator >= (const I<S> &, const I<S> &);
+  template <typename S> friend typename I<S>::difference_type operator - (I<S> &, I<S> &);
+  template <typename S> friend typename I<S>::difference_type operator - (const I<S> &, const I<S> &);
+  template <typename S> friend I<S> operator + (typename I<S>::difference_type , const I<S> &);
+private:
+  T *p;
+};
+template <typename T> I<T>::I () : p (0) {}
+template <typename T> I<T>::~I () { p = (T *) 0; }
+template <typename T> I<T>::I (T *x) : p (x) {}
+template <typename T> I<T>::I (const I &x) : p (x.p) {}
+template <typename T> T &I<T>::operator * () { return *p; }
+template <typename T> T *I<T>::operator -> () { return p; }
+template <typename T> T &I<T>::operator [] (const difference_type &x) const { return p[x]; }
+template <typename T> I<T> &I<T>::operator = (const I &x) { p = x.p; return *this; }
+template <typename T> I<T> &I<T>::operator ++ () { ++p; return *this; }
+template <typename T> I<T> I<T>::operator ++ (int) { return I (p++); }
+template <typename T> I<T> &I<T>::operator -- () { --p; return *this; }
+template <typename T> I<T> I<T>::operator -- (int) { return I (p--); }
+template <typename T> I<T> &I<T>::operator += (const difference_type &x) { p += x; return *this; }
+template <typename T> I<T> &I<T>::operator -= (const difference_type &x) { p -= x; return *this; }
+template <typename T> I<T> I<T>::operator + (const difference_type &x) const { return I (p + x); }
+template <typename T> I<T> I<T>::operator - (const difference_type &x) const { return I (p - x); }
+template <typename T> bool operator == (I<T> &x, I<T> &y) { return x.p == y.p; }
+template <typename T> bool operator == (const I<T> &x, const I<T> &y) { return x.p == y.p; }
+template <typename T> bool operator != (I<T> &x, I<T> &y) { return !(x == y); }
+template <typename T> bool operator != (const I<T> &x, const I<T> &y) { return !(x == y); }
+template <typename T> bool operator < (I<T> &x, I<T> &y) { return x.p < y.p; }
+template <typename T> bool operator < (const I<T> &x, const I<T> &y) { return x.p < y.p; }
+template <typename T> bool operator <= (I<T> &x, I<T> &y) { return x.p <= y.p; }
+template <typename T> bool operator <= (const I<T> &x, const I<T> &y) { return x.p <= y.p; }
+template <typename T> bool operator > (I<T> &x, I<T> &y) { return x.p > y.p; }
+template <typename T> bool operator > (const I<T> &x, const I<T> &y) { return x.p > y.p; }
+template <typename T> bool operator >= (I<T> &x, I<T> &y) { return x.p >= y.p; }
+template <typename T> bool operator >= (const I<T> &x, const I<T> &y) { return x.p >= y.p; }
+template <typename T> typename I<T>::difference_type operator - (I<T> &x, I<T> &y) { return x.p - y.p; }
+template <typename T> typename I<T>::difference_type operator - (const I<T> &x, const I<T> &y) { return x.p - y.p; }
+template <typename T> I<T> operator + (typename I<T>::difference_type x, const I<T> &y) { return I<T> (x + y.p); }
+
+template <typename T>
+class J
+{
+public:
+  J(const I<T> &x, const I<T> &y) : b (x), e (y) {}
+  const I<T> &begin ();
+  const I<T> &end ();
+private:
+  I<T> b, e;
+};
+
+template <typename T> const I<T> &J<T>::begin () { return b; }
+template <typename T> const I<T> &J<T>::end () { return e; }
+
+int results[2000];
+
+template <typename T>
+void
+baz (I<T> &i)
+{
+  if (*i < 0 || *i >= 2000)
+    abort ();
+  results[*i]++;
+}
+
+I<int>
+f1 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel shared (i)
+  {
+  #pragma omp for lastprivate (i) schedule(runtime)
+    for (i = x; i < y - 1; ++i)
+      baz (i);
+  #pragma omp single
+    i += 3;
+  }
+  return I<int> (i);
+}
+
+I<int>
+f2 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel for lastprivate (i)
+  for (i = x; i < y - 1; i = 1 - 6 + 7 + i)
+    baz (i);
+  return I<int> (i);
+}
+
+template <typename T>
+I<int>
+f3 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel
+  #pragma omp for lastprivate (i)
+    for (i = x + 1000 - 64; i <= y - 10; i++)
+      baz (i);
+  return i;
+}
+
+template <typename T>
+I<int>
+f4 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel for lastprivate (i)
+  for (i = x + 2000 - 64; i > y + 10; --i)
+    baz (i);
+  return I<int> (i);
+}
+
+template <typename T>
+I<int>
+f5 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel for lastprivate (i)
+  for (i = x; i > y + T (6); i--)
+    baz (i);
+  return i;
+}
+
+template <typename T>
+I<int>
+f6 (const I<int> &x, const I<int> &y)
+{
+  I<int> i;
+#pragma omp parallel for lastprivate (i)
+  for (i = x - T (7); i > y; i -= T (2))
+    baz (i);
+  return I<int> (i);
+}
+
+template <int N>
+I<int>
+f7 (I<int> i, const I<int> &x, const I<int> &y)
+{
+#pragma omp parallel for lastprivate (i)
+  for (i = x - 10; i <= y + 10; i += N)
+    baz (i);
+  return I<int> (i);
+}
+
+template <int N>
+I<int>
+f8 (J<int> j)
+{
+  I<int> i;
+#pragma omp parallel shared (i)
+  #pragma omp for lastprivate (i)
+    for (i = j.begin (); i <= j.end () + N; i += 2)
+      baz (i);
+  return i;
+}
+
+I<int> i9;
+
+template <long N>
+I<int> &
+f9 (J<int> j)
+{
+#pragma omp parallel for lastprivate (i9)
+  for (i9 = j.begin () + N; i9 <= j.end () - N; i9 = i9 - N)
+    baz (i9);
+  return i9;
+}
+
+template <typename T, int N>
+I<T>
+f10 (const I<T> &x, const I<T> &y)
+{
+  I<T> i;
+#pragma omp parallel for lastprivate (i)
+  for (i = x; i > y; i = i + N)
+    baz (i);
+  return i;
+}
+
+template <typename T, typename U>
+T
+f11 (T i, const T &x, const T &y)
+{
+#pragma omp parallel
+  #pragma omp for lastprivate (i)
+  for (i = x + U (2); i <= y + U (1); i = U (2) + U (3) + i)
+    baz (i);
+  return T (i);
+}
+
+template <typename T>
+T
+f12 (const T &x, const T &y)
+{
+  T i;
+#pragma omp parallel for lastprivate (i)
+  for (i = x; i > y; --i)
+    baz (i);
+  return i;
+}
+
+#define check(expr) \
+  for (int i = 0; i < 2000; i++)			\
+    if (expr)						\
+      {							\
+	if (results[i] != 1)				\
+	  abort ();					\
+	results[i] = 0;					\
+      }							\
+    else if (results[i])				\
+      abort ()
+
+int
+main ()
+{
+  int a[2000];
+  long b[2000];
+  for (int i = 0; i < 2000; i++)
+    {
+      a[i] = i;
+      b[i] = i;
+    }
+  if (*f1 (&a[10], &a[1873]) != 1875)
+    abort ();
+  check (i >= 10 && i < 1872);
+  if (*f2 (&a[0], &a[1998]) != 1998)
+    abort ();
+  check (i < 1997 && (i & 1) == 0);
+  if (*f3<int> (&a[10], &a[1971]) != 1962)
+    abort ();
+  check (i >= 946 && i <= 1961);
+  if (*f4<int> (&a[0], &a[30]) != 40)
+    abort ();
+  check (i > 40 && i <= 2000 - 64);
+  if (*f5<short> (&a[1931], &a[17]) != 23)
+    abort ();
+  check (i > 23 && i <= 1931);
+  if (*f6<long> (&a[1931], &a[17]) != 16)
+    abort ();
+  check (i > 17 && i <= 1924 && (i & 1) == 0);
+  if (*f7<6> (I<int> (), &a[12], &a[1800]) != 1814)
+    abort ();
+  check (i >= 2 && i <= 1808 && (i - 2) % 6 == 0);
+  if (*f8<121> (J<int> (&a[14], &a[1803])) != 1926)
+    abort ();
+  check (i >= 14 && i <= 1924 && (i & 1) == 0);
+  if (*f9<-3L> (J<int> (&a[27], &a[1761])) != 1767)
+    abort ();
+  check (i >= 24 && i <= 1764 && (i % 3) == 0);
+  if (*f10<int, -7> (&a[1939], &a[17]) != 14)
+    abort ();
+  check (i >= 21 && i <= 1939 && i % 7 == 0);
+  if (*f11<I<int>, short> (I<int> (), &a[71], &a[1941]) != 1943)
+    abort ();
+  check (i >= 73 && i <= 1938 && (i - 73) % 5 == 0);
+  if (*f12<I<int> > (&a[1761], &a[37]) != 37)
+    abort ();
+  check (i > 37 && i <= 1761);
+  if (*f10<long, -7> (&b[1939], &b[17]) != 14)
+    abort ();
+  check (i >= 21 && i <= 1939 && i % 7 == 0);
+  if (*f11<I<long>, short> (I<long> (), &b[71], &b[1941]) != 1943)
+    abort ();
+  check (i >= 73 && i <= 1938 && (i - 73) % 5 == 0);
+  if (*f12<I<long> > (&b[1761], &b[37]) != 37)
+    abort ();
+  check (i > 37 && i <= 1761);
+}
diff --git a/libgomp/testsuite/libgomp.c++/loop-10.C b/libgomp/testsuite/libgomp.c++/loop-10.C
new file mode 100644
index 00000000000..9c0de25d56f
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/loop-10.C
@@ -0,0 +1,105 @@
+// { dg-do run }
+
+#include <omp.h>
+
+extern "C" void abort (void);
+
+#define LLONG_MAX __LONG_LONG_MAX__
+#define ULLONG_MAX (LLONG_MAX * 2ULL + 1)
+#define INT_MAX __INT_MAX__
+
+int v;
+
+int
+test1 (void)
+{
+  int e = 0, cnt = 0;
+  long long i;
+  unsigned long long j;
+  char buf[6], *p;
+
+  #pragma omp for schedule(dynamic,1) collapse(2) nowait
+  for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      if ((i != LLONG_MAX - 30001
+	   && i != LLONG_MAX - 20001
+	   && i != LLONG_MAX - 10001)
+	  || j != 20)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(guided,1) collapse(2) nowait
+  for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      if ((i != -LLONG_MAX + 30000
+	   && i != -LLONG_MAX + 20000
+	   && i != -LLONG_MAX + 10000)
+	  || j != ULLONG_MAX - 3)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(static,1) collapse(2) nowait
+  for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+    for (j = 20; j <= LLONG_MAX - 70 + v; j += LLONG_MAX + 50ULL)
+      if ((i != LLONG_MAX - 30001
+	   && i != LLONG_MAX - 20001
+	   && i != LLONG_MAX - 10001)
+	  || j != 20)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(static) collapse(2) nowait
+  for (i = -LLONG_MAX + 30000 + v; i >= -LLONG_MAX + 10000; i -= 10000)
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      if ((i != -LLONG_MAX + 30000
+	   && i != -LLONG_MAX + 20000
+	   && i != -LLONG_MAX + 10000)
+	  || j != ULLONG_MAX - 3)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(runtime) collapse(2) nowait
+  for (i = 10; i < 30; i++)
+    for (p = buf; p <= buf + 4; p += 2)
+      if (i < 10 || i >= 30 || (p != buf && p != buf + 2 && p != buf + 4))
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 60)
+    abort ();
+  else
+    cnt = 0;
+
+  return 0;
+}
+
+int
+main (void)
+{
+  if (2 * sizeof (int) != sizeof (long long))
+    return 0;
+  asm volatile ("" : "+r" (v));
+  omp_set_schedule (omp_sched_dynamic, 1);
+  test1 ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c++/loop-8.C b/libgomp/testsuite/libgomp.c++/loop-8.C
new file mode 100644
index 00000000000..bc20c68a167
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/loop-8.C
@@ -0,0 +1,276 @@
+#include <omp.h>
+#include <stdlib.h>
+#include <string.h>
+
+int
+test1 ()
+{
+  short int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+test2 ()
+{
+  int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+test3 ()
+{
+  int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+test4 ()
+{
+  int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+main ()
+{
+  test1 ();
+  test2 ();
+  test3 ();
+  omp_set_schedule (omp_sched_static, 0);
+  test4 ();
+  omp_set_schedule (omp_sched_static, 3);
+  test4 ();
+  omp_set_schedule (omp_sched_dynamic, 5);
+  test4 ();
+  omp_set_schedule (omp_sched_guided, 2);
+  test4 ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c++/loop-9.C b/libgomp/testsuite/libgomp.c++/loop-9.C
new file mode 100644
index 00000000000..35daf2276e8
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/loop-9.C
@@ -0,0 +1,387 @@
+// { dg-do run }
+
+#include <omp.h>
+
+extern "C" void abort ();
+
+#define LLONG_MAX __LONG_LONG_MAX__
+#define ULLONG_MAX (LLONG_MAX * 2ULL + 1)
+#define INT_MAX __INT_MAX__
+
+int arr[6 * 5];
+
+void
+set (int loopidx, int idx)
+{
+#pragma omp atomic
+  arr[loopidx * 5 + idx]++;
+}
+
+#define check(var, val, loopidx, idx) \
+  if (var == (val)) set (loopidx, idx); else
+#define test(loopidx, count) \
+  for (idx = 0; idx < 5; idx++) \
+    if (arr[loopidx * 5 + idx] != idx < count) \
+      abort (); \
+    else \
+      arr[loopidx * 5 + idx] = 0
+
+int
+test1 ()
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(dynamic,1) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test2 ()
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(guided,1) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test3 ()
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(static) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test4 ()
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(static,1) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test5 ()
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(runtime) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+main ()
+{
+  if (2 * sizeof (int) != sizeof (long long))
+    return 0;
+  test1 ();
+  test2 ();
+  test3 ();
+  test4 ();
+  omp_set_schedule (omp_sched_static, 0);
+  test5 ();
+  omp_set_schedule (omp_sched_static, 3);
+  test5 ();
+  omp_set_schedule (omp_sched_dynamic, 5);
+  test5 ();
+  omp_set_schedule (omp_sched_guided, 2);
+  test5 ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c++/task-1.C b/libgomp/testsuite/libgomp.c++/task-1.C
new file mode 100644
index 00000000000..535a8287b0c
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/task-1.C
@@ -0,0 +1,83 @@
+extern "C" void abort ();
+
+int a = 18;
+
+void
+f1 (int i, int j, int k)
+{
+  int l = 6, m = 7, n = 8;
+#pragma omp task private(j, m) shared(k, n)
+  {
+    j = 6;
+    m = 5;
+    if (++a != 19 || ++i != 9 || j != 6 || ++l != 7 || m != 5 || ++n != 9)
+      #pragma omp atomic
+	k++;
+  }
+#pragma omp taskwait
+  if (a != 19 || i != 8 || j != 26 || k != 0 || l != 6 || m != 7 || n != 9)
+    abort ();
+}
+
+int v1 = 1, v2 = 2, v5 = 5;
+int err;
+
+void
+f2 (void)
+{
+  int v3 = 3;
+#pragma omp sections private (v1) firstprivate (v2)
+  {
+  #pragma omp section
+    {
+      int v4 = 4;
+      v1 = 7;
+      #pragma omp task
+	{
+	  if (++v1 != 8 || ++v2 != 3 || ++v3 != 4 || ++v4 != 5 || ++v5 != 6)
+	    err = 1;
+	}
+      #pragma omp taskwait
+      if (v1 != 7 || v2 != 2 || v3 != 3 || v4 != 4 || v5 != 6)
+	abort ();
+      if (err)
+	abort ();
+    }
+  }
+}
+
+void
+f3 (int i, int j, int k)
+{
+  int l = 6, m = 7, n = 8;
+#pragma omp task private(j, m) shared(k, n) untied
+  {
+    j = 6;
+    m = 5;
+    if (++a != 19 || ++i != 9 || j != 6 || ++l != 7 || m != 5 || ++n != 9)
+      #pragma omp atomic
+	k++;
+  }
+#pragma omp taskwait
+  if (a != 19 || i != 8 || j != 26 || k != 0 || l != 6 || m != 7 || n != 9)
+    abort ();
+}
+
+int
+main ()
+{
+  f1 (8, 26, 0);
+  f2 ();
+  a = 18;
+  f3 (8, 26, 0);
+  a = 18;
+#pragma omp parallel num_threads(4)
+  {
+    #pragma omp master
+      {
+	f1 (8, 26, 0);
+	a = 18;
+	f3 (8, 26, 0);
+      }
+  }
+}
diff --git a/libgomp/testsuite/libgomp.c++/task-2.C b/libgomp/testsuite/libgomp.c++/task-2.C
new file mode 100644
index 00000000000..a198cc721b5
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/task-2.C
@@ -0,0 +1,70 @@
+// { dg-do run }
+
+#include <omp.h>
+extern "C" void abort ();
+
+int l = 5;
+
+int
+foo (int i)
+{
+  int j = 7;
+  const int k = 8;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp taskwait
+  return (i != 8 * omp_get_thread_num () + 4
+	  || j != 4 * i - 3
+	  || k != 8);
+}
+
+int
+main (void)
+{
+  int r = 0;
+  #pragma omp parallel num_threads (4) reduction(+:r)
+    if (omp_get_num_threads () != 4)
+      {
+	#pragma omp master
+	  l = 133;
+      }
+    else if (foo (8 * omp_get_thread_num ()))
+      r++;
+  if (r || l != 133)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c++/task-3.C b/libgomp/testsuite/libgomp.c++/task-3.C
new file mode 100644
index 00000000000..e1ecb49654a
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/task-3.C
@@ -0,0 +1,90 @@
+// { dg-do run }
+
+extern "C" void abort ();
+
+struct A
+{
+  A ();
+  ~A ();
+  A (const A &);
+  unsigned long l;
+};
+
+int e;
+
+A::A ()
+{
+  l = 17;
+}
+
+A::~A ()
+{
+  if (l > 30)
+    #pragma omp atomic
+      e++;
+}
+
+A::A (const A &r)
+{
+  l = r.l;
+}
+
+void
+check (int i, A &a, int j, A &b)
+{
+  if (i != 6 || a.l != 21 || j != 0 || b.l != 23)
+    #pragma omp atomic
+      e++;
+}
+
+A b;
+int j;
+
+void
+foo (int i)
+{
+  A a;
+  a.l = 21;
+  #pragma omp task firstprivate (i, a, j, b)
+    check (i, a, j, b);
+}
+
+void
+bar (int i, A a)
+{
+  a.l = 21;
+  #pragma omp task firstprivate (i, a, j, b)
+    check (i, a, j, b);
+}
+
+A
+baz ()
+{
+  A a, c;
+  a.l = 21;
+  c.l = 23;
+  #pragma omp task firstprivate (a, c)
+    check (6, a, 0, c);
+  return a;
+}
+
+int
+main ()
+{
+  b.l = 23;
+  foo (6);
+  bar (6, A ());
+  baz ();
+  #pragma omp parallel num_threads (4)
+    {
+      #pragma omp single
+	for (int i = 0; i < 64; i++)
+	  {
+	    foo (6);
+	    bar (6, A ());
+	    baz ();
+	  }
+    }
+  if (e)
+    abort ();
+}
diff --git a/libgomp/testsuite/libgomp.c++/task-4.C b/libgomp/testsuite/libgomp.c++/task-4.C
new file mode 100644
index 00000000000..f2e786a2fdd
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/task-4.C
@@ -0,0 +1,37 @@
+#include <omp.h>
+extern "C" void *memset (void *, int, __SIZE_TYPE__);
+extern "C" void abort (void);
+
+int e;
+
+void
+baz (int i, int *p, int j, int *q)
+{
+  if (p[0] != 1 || p[i] != 3 || q[0] != 2 || q[j] != 4)
+    #pragma omp atomic
+      e++;
+}
+
+void
+foo (int i, int j)
+{
+  int p[i + 1];
+  int q[j + 1];
+  memset (p, 0, sizeof (p));
+  memset (q, 0, sizeof (q));
+  p[0] = 1;
+  p[i] = 3;
+  q[0] = 2;
+  q[j] = 4;
+  #pragma omp task firstprivate (p, q)
+    baz (i, p, j, q);
+}
+
+int
+main ()
+{
+  #pragma omp parallel num_threads (4)
+    foo (5 + omp_get_thread_num (), 7 + omp_get_thread_num ());
+  if (e)
+    abort ();
+}
diff --git a/libgomp/testsuite/libgomp.c++/task-5.C b/libgomp/testsuite/libgomp.c++/task-5.C
new file mode 100644
index 00000000000..c882bfe1517
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/task-5.C
@@ -0,0 +1,90 @@
+// { dg-do run }
+
+extern "C" void abort ();
+
+struct A
+{
+  A ();
+  ~A ();
+  A (const A &);
+  unsigned long l;
+};
+
+int e;
+
+A::A ()
+{
+  l = 17;
+}
+
+A::~A ()
+{
+  if (l > 130)
+    #pragma omp atomic
+      e++;
+}
+
+A::A (const A &r)
+{
+  l = r.l + 64;
+}
+
+void
+check (int i, A &a, int j, A &b)
+{
+  if (i != 6 || a.l != 21 + 64 || j != 0 || b.l != 23 + 64)
+    #pragma omp atomic
+      e++;
+}
+
+A b;
+int j;
+
+void
+foo (int i)
+{
+  A a;
+  a.l = 21;
+  #pragma omp task firstprivate (j, b)
+    check (i, a, j, b);
+}
+
+void
+bar (int i, A a)
+{
+  a.l = 21;
+  #pragma omp task firstprivate (j, b)
+    check (i, a, j, b);
+}
+
+A
+baz ()
+{
+  A a, c;
+  a.l = 21;
+  c.l = 23;
+  #pragma omp task firstprivate (a, c)
+    check (6, a, 0, c);
+  return a;
+}
+
+int
+main ()
+{
+  b.l = 23;
+  foo (6);
+  bar (6, A ());
+  baz ();
+  #pragma omp parallel num_threads (4)
+    {
+      #pragma omp single
+	for (int i = 0; i < 64; i++)
+	  {
+	    foo (6);
+	    bar (6, A ());
+	    baz ();
+	  }
+    }
+  if (e)
+    abort ();
+}
diff --git a/libgomp/testsuite/libgomp.c++/task-6.C b/libgomp/testsuite/libgomp.c++/task-6.C
new file mode 100644
index 00000000000..cc9072b9d1c
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c++/task-6.C
@@ -0,0 +1,86 @@
+extern "C" void abort ();
+
+int a = 18;
+
+template <typename T>
+void
+f1 (T i, T j, T k)
+{
+  T l = 6, m = 7, n = 8;
+#pragma omp task private(j, m) shared(k, n)
+  {
+    j = 6;
+    m = 5;
+    if (++a != 19 || ++i != 9 || j != 6 || ++l != 7 || m != 5 || ++n != 9)
+      #pragma omp atomic
+	k++;
+  }
+#pragma omp taskwait
+  if (a != 19 || i != 8 || j != 26 || k != 0 || l != 6 || m != 7 || n != 9)
+    abort ();
+}
+
+int v1 = 1, v2 = 2, v5 = 5;
+int err;
+
+template <typename T>
+void
+f2 (void)
+{
+  T v3 = 3;
+#pragma omp sections private (v1) firstprivate (v2)
+  {
+  #pragma omp section
+    {
+      T v4 = 4;
+      v1 = 7;
+      #pragma omp task
+	{
+	  if (++v1 != 8 || ++v2 != 3 || ++v3 != 4 || ++v4 != 5 || ++v5 != 6)
+	    err = 1;
+	}
+      #pragma omp taskwait
+      if (v1 != 7 || v2 != 2 || v3 != 3 || v4 != 4 || v5 != 6)
+	abort ();
+      if (err)
+	abort ();
+    }
+  }
+}
+
+template <typename T>
+void
+f3 (T i, T j, T k)
+{
+  T l = 6, m = 7, n = 8;
+#pragma omp task private(j, m) shared(k, n) untied
+  {
+    j = 6;
+    m = 5;
+    if (++a != 19 || ++i != 9 || j != 6 || ++l != 7 || m != 5 || ++n != 9)
+      #pragma omp atomic
+	k++;
+  }
+#pragma omp taskwait
+  if (a != 19 || i != 8 || j != 26 || k != 0 || l != 6 || m != 7 || n != 9)
+    abort ();
+}
+
+int
+main ()
+{
+  f1 <int> (8, 26, 0);
+  f2 <int> ();
+  a = 18;
+  f3 <int> (8, 26, 0);
+  a = 18;
+#pragma omp parallel num_threads(4)
+  {
+    #pragma omp master
+      {
+	f1 <int> (8, 26, 0);
+	a = 18;
+	f3 <int> (8, 26, 0);
+      }
+  }
+}
diff --git a/libgomp/testsuite/libgomp.c/collapse-1.c b/libgomp/testsuite/libgomp.c/collapse-1.c
new file mode 100644
index 00000000000..82becfa7952
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/collapse-1.c
@@ -0,0 +1,30 @@
+/* { dg-do run } */
+
+#include <string.h>
+#include <stdlib.h>
+
+int
+main (void)
+{
+  int i, j, k, l = 0;
+  int a[3][3][3];
+
+  memset (a, '\0', sizeof (a));
+  #pragma omp parallel for collapse(4 - 1) schedule(static, 4)
+    for (i = 0; i < 2; i++)
+      for (j = 0; j < 2; j++)
+	for (k = 0; k < 2; k++)
+	  a[i][j][k] = i + j * 4 + k * 16;
+  #pragma omp parallel
+    {
+      #pragma omp for collapse(2) reduction(|:l)
+	for (i = 0; i < 2; i++)
+	  for (j = 0; j < 2; j++)
+	    for (k = 0; k < 2; k++)
+	      if (a[i][j][k] != i + j * 4 + k * 16)
+		l = 1;
+    }
+  if (l)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/collapse-2.c b/libgomp/testsuite/libgomp.c/collapse-2.c
new file mode 100644
index 00000000000..b5c77d46143
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/collapse-2.c
@@ -0,0 +1,30 @@
+/* { dg-do run } */
+
+#include <stdlib.h>
+#include <omp.h>
+
+int
+main (void)
+{
+  int i, j, k, l = 0, f = 0;
+  int m1 = 4, m2 = -5, m3 = 17;
+
+  #pragma omp parallel for num_threads (8) collapse(3) \
+		       schedule(static, 9) reduction(+:l) \
+		       firstprivate(f)
+    for (i = -2; i < m1; i++)
+      for (j = m2; j < -2; j++)
+	{
+	  for (k = 13; k < m3; k++)
+	    {
+	      if (omp_get_num_threads () == 8
+		  && ((i + 2) * 12 + (j + 5) * 4 + (k - 13)
+		      != (omp_get_thread_num () * 9
+			  + f++)))
+		l++;
+	    }
+	}
+  if (l)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/collapse-3.c b/libgomp/testsuite/libgomp.c/collapse-3.c
new file mode 100644
index 00000000000..4674f83f4b6
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/collapse-3.c
@@ -0,0 +1,31 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -std=gnu99" } */
+
+#include <string.h>
+#include <stdlib.h>
+
+int
+main (void)
+{
+  int i2, l = 0;
+  int a[3][3][3];
+
+  memset (a, '\0', sizeof (a));
+  #pragma omp parallel for collapse(4 - 1) schedule(static, 4)
+    for (int i = 0; i < 2; i++)
+      for (int j = 0; j < 2; j++)
+	for (int k = 0; k < 2; k++)
+	  a[i][j][k] = i + j * 4 + k * 16;
+  #pragma omp parallel
+    {
+      #pragma omp for collapse(2) reduction(|:l)
+	for (i2 = 0; i2 < 2; i2++)
+	  for (int j = 0; j < 2; j++)
+	    for (int k = 0; k < 2; k++)
+	      if (a[i2][j][k] != i2 + j * 4 + k * 16)
+		l = 1;
+    }
+  if (l)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/icv-1.c b/libgomp/testsuite/libgomp.c/icv-1.c
new file mode 100644
index 00000000000..99708f82306
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/icv-1.c
@@ -0,0 +1,33 @@
+#include <omp.h>
+#include <stdlib.h>
+
+int
+main (void)
+{
+  int err = 0;
+
+  omp_set_num_threads (4);
+  if (omp_get_max_threads () != 4)
+    abort ();
+  #pragma omp parallel reduction(|: err) num_threads(1)
+  {
+    if (omp_get_max_threads () != 4)
+      err |= 1;
+    omp_set_num_threads (6);
+    #pragma omp task if(0) shared(err)
+    {
+      if (omp_get_max_threads () != 6)
+	err |= 2;
+      omp_set_num_threads (5);
+      if (omp_get_max_threads () != 5)
+	err |= 4;
+    }
+    if (omp_get_max_threads () != 6)
+      err |= 8;
+  }
+  if (err)
+    abort ();
+  if (omp_get_max_threads () != 4)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/icv-2.c b/libgomp/testsuite/libgomp.c/icv-2.c
new file mode 100644
index 00000000000..326f8eb404a
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/icv-2.c
@@ -0,0 +1,46 @@
+/* { dg-do run { target *-*-linux* } } */
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE 1
+#endif
+#include <pthread.h>
+#include <omp.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+pthread_barrier_t bar;
+
+void *tf (void *p)
+{
+  int l;
+  if (p)
+    omp_set_num_threads (3);
+  pthread_barrier_wait (&bar);
+  if (!p)
+    omp_set_num_threads (6);
+  pthread_barrier_wait (&bar);
+  omp_set_dynamic (0);
+  if (omp_get_max_threads () != (p ? 3 : 6))
+    abort ();
+  l = 0;
+  #pragma omp parallel num_threads (6) reduction (|:l)
+    {
+      l |= omp_get_max_threads () != (p ? 3 : 6);
+      omp_set_num_threads ((p ? 3 : 6) + omp_get_thread_num ());
+      l |= omp_get_max_threads () != ((p ? 3 : 6) + omp_get_thread_num ());
+    }
+  if (l)
+    abort ();
+  return NULL;
+}
+
+int
+main (void)
+{
+  pthread_t th;
+  pthread_barrier_init (&bar, NULL, 2);
+  pthread_create (&th, NULL, tf, NULL);
+  tf ("");
+  pthread_join (th, NULL);
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/lib-2.c b/libgomp/testsuite/libgomp.c/lib-2.c
new file mode 100644
index 00000000000..3a3b3f65517
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/lib-2.c
@@ -0,0 +1,25 @@
+#include <stdlib.h>
+#include <omp.h>
+
+int
+main (void)
+{
+  omp_sched_t kind;
+  int modifier;
+
+  omp_set_schedule (omp_sched_static, 32);
+  omp_get_schedule (&kind, &modifier);
+  if (kind != omp_sched_static || modifier != 32)
+    abort ();
+  omp_set_schedule (omp_sched_guided, 4);
+  omp_get_schedule (&kind, &modifier);
+  if (kind != omp_sched_guided || modifier != 4)
+    abort ();
+  if (omp_get_thread_limit () < 0)
+    abort ();
+  omp_set_max_active_levels (6);
+  if (omp_get_max_active_levels () != 6)
+    abort ();
+
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/lock-1.c b/libgomp/testsuite/libgomp.c/lock-1.c
new file mode 100644
index 00000000000..e09645dbc3f
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/lock-1.c
@@ -0,0 +1,31 @@
+#include <omp.h>
+#include <stdlib.h>
+
+int
+main (void)
+{
+  int l = 0;
+  omp_nest_lock_t lock;
+  omp_init_nest_lock (&lock);
+  if (omp_test_nest_lock (&lock) != 1)
+    abort ();
+  if (omp_test_nest_lock (&lock) != 2)
+    abort ();
+#pragma omp parallel if (0) reduction (+:l)
+  {
+    /* In OpenMP 2.5 this was supposed to return 3,
+       but in OpenMP 3.0 the parallel region has a different
+       task and omp_*_lock_t are owned by tasks, not by threads.  */
+    if (omp_test_nest_lock (&lock) != 0)
+      l++;
+  }
+  if (l)
+    abort ();
+  if (omp_test_nest_lock (&lock) != 3)
+    abort ();
+  omp_unset_nest_lock (&lock);
+  omp_unset_nest_lock (&lock);
+  omp_unset_nest_lock (&lock);
+  omp_destroy_nest_lock (&lock);
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/lock-2.c b/libgomp/testsuite/libgomp.c/lock-2.c
new file mode 100644
index 00000000000..9009b12fe5d
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/lock-2.c
@@ -0,0 +1,32 @@
+#include <omp.h>
+#include <stdlib.h>
+
+int
+main (void)
+{
+  int l = 0;
+  omp_nest_lock_t lock;
+  omp_init_nest_lock (&lock);
+#pragma omp parallel reduction (+:l) num_threads (1)
+  {
+    if (omp_test_nest_lock (&lock) != 1)
+      l++;
+    if (omp_test_nest_lock (&lock) != 2)
+      l++;
+  #pragma omp task if (0) shared (lock, l)
+    {
+      if (omp_test_nest_lock (&lock) != 0)
+	l++;
+    }
+  #pragma omp taskwait
+    if (omp_test_nest_lock (&lock) != 3)
+      l++;
+    omp_unset_nest_lock (&lock);
+    omp_unset_nest_lock (&lock);
+    omp_unset_nest_lock (&lock);
+  }
+  if (l)
+    abort ();
+  omp_destroy_nest_lock (&lock);
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/lock-3.c b/libgomp/testsuite/libgomp.c/lock-3.c
new file mode 100644
index 00000000000..1fc83726d18
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/lock-3.c
@@ -0,0 +1,60 @@
+/* { dg-do run { target *-*-linux* } } */
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE 1
+#endif
+#include <pthread.h>
+#include <omp.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+pthread_barrier_t bar;
+omp_nest_lock_t lock;
+
+void *tf (void *p)
+{
+  int l;
+  if (p)
+    {
+      if (omp_test_nest_lock (&lock) != 1)
+	abort ();
+      if (omp_test_nest_lock (&lock) != 2)
+	abort ();
+    }
+  pthread_barrier_wait (&bar);
+  if (!p && omp_test_nest_lock (&lock) != 0)
+    abort ();
+  pthread_barrier_wait (&bar);
+  if (p)
+    {
+      if (omp_test_nest_lock (&lock) != 3)
+	abort ();
+      omp_unset_nest_lock (&lock);
+      omp_unset_nest_lock (&lock);
+      omp_unset_nest_lock (&lock);
+    }
+  pthread_barrier_wait (&bar);
+  if (!p)
+    {
+      if (omp_test_nest_lock (&lock) != 1)
+	abort ();
+      if (omp_test_nest_lock (&lock) != 2)
+	abort ();
+      omp_unset_nest_lock (&lock);
+      omp_unset_nest_lock (&lock);
+    }
+  return NULL;
+}
+
+int
+main (void)
+{
+  pthread_t th;
+  omp_init_nest_lock (&lock);
+  pthread_barrier_init (&bar, NULL, 2);
+  pthread_create (&th, NULL, tf, NULL);
+  tf ("");
+  pthread_join (th, NULL);
+  omp_destroy_nest_lock (&lock);
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/loop-4.c b/libgomp/testsuite/libgomp.c/loop-4.c
new file mode 100644
index 00000000000..bc57c043aad
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/loop-4.c
@@ -0,0 +1,28 @@
+/* { dg-do run } */
+
+extern void abort (void);
+
+int
+main (void)
+{
+  int e = 0;
+#pragma omp parallel num_threads (4) reduction(+:e)
+  {
+    long i;
+    #pragma omp for schedule(dynamic,1)
+    for (i = __LONG_MAX__ - 30001; i <= __LONG_MAX__ - 10001; i += 10000)
+      if (i != __LONG_MAX__ - 30001
+	  && i != __LONG_MAX__ - 20001
+	  && i != __LONG_MAX__ - 10001)
+	e = 1;
+    #pragma omp for schedule(dynamic,1)
+    for (i = -__LONG_MAX__ + 30000; i >= -__LONG_MAX__ + 10000; i -= 10000)
+      if (i != -__LONG_MAX__ + 30000
+	  && i != -__LONG_MAX__ + 20000
+	  && i != -__LONG_MAX__ + 10000)
+	e = 1;
+  }
+  if (e)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/loop-5.c b/libgomp/testsuite/libgomp.c/loop-5.c
new file mode 100644
index 00000000000..3a5c7cf4556
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/loop-5.c
@@ -0,0 +1,276 @@
+#include <omp.h>
+#include <stdlib.h>
+#include <string.h>
+
+int
+test1 (void)
+{
+  short int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+test2 (void)
+{
+  int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (static, 3)
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+test3 (void)
+{
+  int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (dynamic, 3)
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+test4 (void)
+{
+  int buf[64], *p;
+  int i;
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[10]; p < &buf[54]; p++)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[3]; p <= &buf[63]; p += 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[16]; p < &buf[51]; p = 4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[16]; p <= &buf[40]; p = p + 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[53]; p > &buf[9]; --p)
+    *p = 5;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 5 * (i >= 10 && i < 54))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[63]; p >= &buf[3]; p -= 2)
+    p[-2] = 6;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 6 * ((i & 1) && i <= 61))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[48]; p > &buf[15]; p = -4 + p)
+    p[2] = 7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != 7 * ((i & 3) == 2 && i >= 18 && i < 53))
+      abort ();
+  memset (buf, '\0', sizeof (buf));
+#pragma omp parallel for schedule (runtime)
+  for (p = &buf[40]; p >= &buf[16]; p = p - 4ULL)
+    p[2] = -7;
+  for (i = 0; i < 64; i++)
+    if (buf[i] != -7 * ((i & 3) == 2 && i >= 18 && i <= 42))
+      abort ();
+  return 0;
+}
+
+int
+main (void)
+{
+  test1 ();
+  test2 ();
+  test3 ();
+  omp_set_schedule (omp_sched_static, 0);
+  test4 ();
+  omp_set_schedule (omp_sched_static, 3);
+  test4 ();
+  omp_set_schedule (omp_sched_dynamic, 5);
+  test4 ();
+  omp_set_schedule (omp_sched_guided, 2);
+  test4 ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/loop-6.c b/libgomp/testsuite/libgomp.c/loop-6.c
new file mode 100644
index 00000000000..9029e181bd2
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/loop-6.c
@@ -0,0 +1,387 @@
+/* { dg-do run } */
+
+#include <omp.h>
+
+extern void abort (void);
+
+#define LLONG_MAX __LONG_LONG_MAX__
+#define ULLONG_MAX (LLONG_MAX * 2ULL + 1)
+#define INT_MAX __INT_MAX__
+
+int arr[6 * 5];
+
+void
+set (int loopidx, int idx)
+{
+#pragma omp atomic
+  arr[loopidx * 5 + idx]++;
+}
+
+#define check(var, val, loopidx, idx) \
+  if (var == (val)) set (loopidx, idx); else
+#define test(loopidx, count) \
+  for (idx = 0; idx < 5; idx++) \
+    if (arr[loopidx * 5 + idx] != idx < count) \
+      abort (); \
+    else \
+      arr[loopidx * 5 + idx] = 0
+
+int
+test1 (void)
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(dynamic,1) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(dynamic,1) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test2 (void)
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(guided,1) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(guided,1) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test3 (void)
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(static) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(static) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test4 (void)
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(static,1) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(static,1) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+test5 (void)
+{
+  int e = 0, idx;
+
+#pragma omp parallel reduction(+:e)
+  {
+    long long i;
+    unsigned long long j;
+    #pragma omp for schedule(runtime) nowait
+    for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+      {
+	check (i, LLONG_MAX - 30001, 0, 0)
+	check (i, LLONG_MAX - 20001, 0, 1)
+	check (i, LLONG_MAX - 10001, 0, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+      {
+	check (i, -LLONG_MAX + 30000, 1, 0)
+	check (i, -LLONG_MAX + 20000, 1, 1)
+	check (i, -LLONG_MAX + 10000, 1, 2)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      {
+	check (j, 20, 2, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      {
+	check (j, ULLONG_MAX - 3, 3, 0)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (j = LLONG_MAX - 20000ULL; j <= LLONG_MAX + 10000ULL; j += 10000ULL)
+      {
+	check (j, LLONG_MAX - 20000ULL, 4, 0)
+	check (j, LLONG_MAX - 10000ULL, 4, 1)
+	check (j, LLONG_MAX, 4, 2)
+	check (j, LLONG_MAX + 10000ULL, 4, 3)
+	e = 1;
+      }
+    #pragma omp for schedule(runtime) nowait
+    for (i = -3LL * INT_MAX - 20000LL; i <= INT_MAX + 10000LL; i += INT_MAX + 200LL)
+      {
+	check (i, -3LL * INT_MAX - 20000LL, 5, 0)
+	check (i, -2LL * INT_MAX - 20000LL + 200LL, 5, 1)
+	check (i, -INT_MAX - 20000LL + 400LL, 5, 2)
+	check (i, -20000LL + 600LL, 5, 3)
+	check (i, INT_MAX - 20000LL + 800LL, 5, 4)
+	e = 1;
+      }
+  }
+  if (e)
+    abort ();
+  test (0, 3);
+  test (1, 3);
+  test (2, 1);
+  test (3, 1);
+  test (4, 4);
+  test (5, 5);
+  return 0;
+}
+
+int
+main (void)
+{
+  if (2 * sizeof (int) != sizeof (long long))
+    return 0;
+  test1 ();
+  test2 ();
+  test3 ();
+  test4 ();
+  omp_set_schedule (omp_sched_static, 0);
+  test5 ();
+  omp_set_schedule (omp_sched_static, 3);
+  test5 ();
+  omp_set_schedule (omp_sched_dynamic, 5);
+  test5 ();
+  omp_set_schedule (omp_sched_guided, 2);
+  test5 ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/loop-7.c b/libgomp/testsuite/libgomp.c/loop-7.c
new file mode 100644
index 00000000000..fc97f4a2907
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/loop-7.c
@@ -0,0 +1,105 @@
+/* { dg-do run } */
+
+#include <omp.h>
+
+extern void abort (void);
+
+#define LLONG_MAX __LONG_LONG_MAX__
+#define ULLONG_MAX (LLONG_MAX * 2ULL + 1)
+#define INT_MAX __INT_MAX__
+
+int v;
+
+int
+test1 (void)
+{
+  int e = 0, cnt = 0;
+  long long i;
+  unsigned long long j;
+  char buf[6], *p;
+
+  #pragma omp for schedule(dynamic,1) collapse(2) nowait
+  for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+    for (j = 20; j <= LLONG_MAX - 70; j += LLONG_MAX + 50ULL)
+      if ((i != LLONG_MAX - 30001
+	   && i != LLONG_MAX - 20001
+	   && i != LLONG_MAX - 10001)
+	  || j != 20)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(guided,1) collapse(2) nowait
+  for (i = -LLONG_MAX + 30000; i >= -LLONG_MAX + 10000; i -= 10000)
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      if ((i != -LLONG_MAX + 30000
+	   && i != -LLONG_MAX + 20000
+	   && i != -LLONG_MAX + 10000)
+	  || j != ULLONG_MAX - 3)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(static,1) collapse(2) nowait
+  for (i = LLONG_MAX - 30001; i <= LLONG_MAX - 10001; i += 10000)
+    for (j = 20; j <= LLONG_MAX - 70 + v; j += LLONG_MAX + 50ULL)
+      if ((i != LLONG_MAX - 30001
+	   && i != LLONG_MAX - 20001
+	   && i != LLONG_MAX - 10001)
+	  || j != 20)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(static) collapse(2) nowait
+  for (i = -LLONG_MAX + 30000 + v; i >= -LLONG_MAX + 10000; i -= 10000)
+    for (j = ULLONG_MAX - 3; j >= LLONG_MAX + 70ULL; j -= LLONG_MAX + 50ULL)
+      if ((i != -LLONG_MAX + 30000
+	   && i != -LLONG_MAX + 20000
+	   && i != -LLONG_MAX + 10000)
+	  || j != ULLONG_MAX - 3)
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 3)
+    abort ();
+  else
+    cnt = 0;
+
+  #pragma omp for schedule(runtime) collapse(2) nowait
+  for (i = 10; i < 30; i++)
+    for (p = buf; p <= buf + 4; p += 2)
+      if (i < 10 || i >= 30 || (p != buf && p != buf + 2 && p != buf + 4))
+	e = 1;
+      else
+	cnt++;
+  if (e || cnt != 60)
+    abort ();
+  else
+    cnt = 0;
+
+  return 0;
+}
+
+int
+main (void)
+{
+  if (2 * sizeof (int) != sizeof (long long))
+    return 0;
+  asm volatile ("" : "+r" (v));
+  omp_set_schedule (omp_sched_dynamic, 1);
+  test1 ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/loop-8.c b/libgomp/testsuite/libgomp.c/loop-8.c
new file mode 100644
index 00000000000..25db25c3b43
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/loop-8.c
@@ -0,0 +1,27 @@
+extern void abort (void);
+
+int buf[256];
+
+void __attribute__((noinline))
+foo (void)
+{
+  int i;
+  #pragma omp for schedule (auto)
+    for (i = 0; i < 256; i++)
+      buf[i] += i;
+}
+
+int
+main (void)
+{
+  int i;
+  #pragma omp parallel for schedule (auto)
+    for (i = 0; i < 256; i++)
+      buf[i] = i;
+  #pragma omp parallel num_threads (4)
+    foo ();
+  for (i = 0; i < 256; i++)
+    if (buf[i] != 2 * i)
+      abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/loop-9.c b/libgomp/testsuite/libgomp.c/loop-9.c
new file mode 100644
index 00000000000..1f789e12ecb
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/loop-9.c
@@ -0,0 +1,18 @@
+extern void abort (void);
+
+char buf[8] = "01234567";
+char buf2[8] = "23456789";
+
+int
+main (void)
+{
+  char *p, *q;
+  int sum = 0;
+  #pragma omp parallel for collapse (2) reduction (+:sum) lastprivate (p, q)
+  for (p = buf; p < &buf[8]; p++)
+    for (q = &buf2[0]; q <= buf2 + 7; q++)
+      sum += (*p - '0') + (*q - '0');
+  if (p != &buf[8] || q != buf2 + 8 || sum != 576)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/nested-3.c b/libgomp/testsuite/libgomp.c/nested-3.c
new file mode 100644
index 00000000000..618600633ac
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/nested-3.c
@@ -0,0 +1,89 @@
+#include <omp.h>
+#include <stdlib.h>
+#include <string.h>
+
+int
+main (void)
+{
+  int e[3];
+
+  memset (e, '\0', sizeof (e));
+  omp_set_nested (1);
+  omp_set_dynamic (0);
+  if (omp_in_parallel ()
+      || omp_get_level () != 0
+      || omp_get_ancestor_thread_num (0) != 0
+      || omp_get_ancestor_thread_num (-1) != -1
+      || omp_get_ancestor_thread_num (1) != -1
+      || omp_get_team_size (0) != 1
+      || omp_get_team_size (-1) != -1
+      || omp_get_team_size (1) != -1
+      || omp_get_active_level () != 0)
+    abort ();
+#pragma omp parallel num_threads (4)
+  {
+    int tn1 = omp_get_thread_num ();
+    if (omp_in_parallel () != 1
+	|| omp_get_num_threads () != 4
+	|| tn1 >= 4 || tn1 < 0
+	|| omp_get_level () != 1
+	|| omp_get_ancestor_thread_num (0) != 0
+	|| omp_get_ancestor_thread_num (1) != tn1
+	|| omp_get_ancestor_thread_num (-1) != -1
+	|| omp_get_ancestor_thread_num (2) != -1
+	|| omp_get_team_size (0) != 1
+	|| omp_get_team_size (1) != omp_get_num_threads ()
+	|| omp_get_team_size (-1) != -1
+	|| omp_get_team_size (2) != -1
+	|| omp_get_active_level () != 1)
+      #pragma omp atomic
+	e[0] += 1;
+    #pragma omp parallel if (0) num_threads(5) firstprivate(tn1)
+    {
+      int tn2 = omp_get_thread_num ();
+      if (omp_in_parallel () != 1
+	  || omp_get_num_threads () != 1
+	  || tn2 != 0
+	  || omp_get_level () != 2
+	  || omp_get_ancestor_thread_num (0) != 0
+	  || omp_get_ancestor_thread_num (1) != tn1
+	  || omp_get_ancestor_thread_num (2) != tn2
+	  || omp_get_ancestor_thread_num (-1) != -1
+	  || omp_get_ancestor_thread_num (3) != -1
+	  || omp_get_team_size (0) != 1
+	  || omp_get_team_size (1) != 4
+	  || omp_get_team_size (2) != 1
+	  || omp_get_team_size (-1) != -1
+	  || omp_get_team_size (3) != -1
+	  || omp_get_active_level () != 1)
+	#pragma omp atomic
+	  e[1] += 1;
+      #pragma omp parallel num_threads(2) firstprivate(tn1, tn2)
+      {
+	int tn3 = omp_get_thread_num ();
+	if (omp_in_parallel () != 1
+	    || omp_get_num_threads () != 2
+	    || tn3 > 1 || tn3 < 0
+	    || omp_get_level () != 3
+	    || omp_get_ancestor_thread_num (0) != 0
+	    || omp_get_ancestor_thread_num (1) != tn1
+	    || omp_get_ancestor_thread_num (2) != tn2
+	    || omp_get_ancestor_thread_num (3) != tn3
+	    || omp_get_ancestor_thread_num (-1) != -1
+	    || omp_get_ancestor_thread_num (4) != -1
+	    || omp_get_team_size (0) != 1
+	    || omp_get_team_size (1) != 4
+	    || omp_get_team_size (2) != 1
+	    || omp_get_team_size (3) != 2
+	    || omp_get_team_size (-1) != -1
+	    || omp_get_team_size (4) != -1
+	    || omp_get_active_level () != 2)
+	  #pragma omp atomic
+	    e[2] += 1;
+      }
+    }
+  }
+  if (e[0] || e[1] || e[2])
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/nestedfn-6.c b/libgomp/testsuite/libgomp.c/nestedfn-6.c
new file mode 100644
index 00000000000..c0ace6b3fb8
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/nestedfn-6.c
@@ -0,0 +1,21 @@
+extern void abort (void);
+
+int j;
+
+int
+main (void)
+{
+  int i;
+  void nested (void) { i = 0; }
+#pragma omp parallel for lastprivate (i)
+  for (i = 0; i < 50; i += 3)
+    ;
+  if (i != 51)
+    abort ();
+#pragma omp parallel for lastprivate (j)
+  for (j = -50; j < 70; j += 7)
+    ;
+  if (j != 76)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/pr26943-2.c b/libgomp/testsuite/libgomp.c/pr26943-2.c
index 778048492f6..c052e811288 100644
--- a/libgomp/testsuite/libgomp.c/pr26943-2.c
+++ b/libgomp/testsuite/libgomp.c/pr26943-2.c
@@ -20,7 +20,7 @@ main (void)
     {
       if (a != 8 || b != 12 || e[0] != 'a' || f[0] != 'b')
 	j++;
-#pragma omp barrier
+#pragma omp barrier	/* { dg-warning "may not be closely nested" } */
 #pragma omp atomic
       a += i;
       b += i;
@@ -31,7 +31,7 @@ main (void)
       f[0] += i;
       g[0] = 'g' + i;
       h[0] = 'h' + i;
-#pragma omp barrier
+#pragma omp barrier	/* { dg-warning "may not be closely nested" } */
       if (a != 8 + 6 || b != 12 + i || c != i || d != i)
 	j += 8;
       if (e[0] != 'a' + 6 || f[0] != 'b' + i || g[0] != 'g' + i)
diff --git a/libgomp/testsuite/libgomp.c/pr26943-3.c b/libgomp/testsuite/libgomp.c/pr26943-3.c
index be93cb479d1..dc3d5010da1 100644
--- a/libgomp/testsuite/libgomp.c/pr26943-3.c
+++ b/libgomp/testsuite/libgomp.c/pr26943-3.c
@@ -26,7 +26,7 @@ main (void)
 	{
 	  if (a != 8 || b != 12 || e[0] != 'a' || f[0] != 'b')
 	    j++;
-#pragma omp barrier
+#pragma omp barrier	/* { dg-warning "may not be closely nested" } */
 #pragma omp atomic
 	  a += i;
 	  b += i;
@@ -37,7 +37,7 @@ main (void)
 	  f[0] += i;
 	  g[0] = 'g' + i;
 	  h[0] = 'h' + i;
-#pragma omp barrier
+#pragma omp barrier	/* { dg-warning "may not be closely nested" } */
 	  if (a != 8 + 6 || b != 12 + i || c != i || d != i)
 	    j += 8;
 	  if (e[0] != 'a' + 6 || f[0] != 'b' + i || g[0] != 'g' + i)
diff --git a/libgomp/testsuite/libgomp.c/pr26943-4.c b/libgomp/testsuite/libgomp.c/pr26943-4.c
index 33d368583dd..0f1d4197a5f 100644
--- a/libgomp/testsuite/libgomp.c/pr26943-4.c
+++ b/libgomp/testsuite/libgomp.c/pr26943-4.c
@@ -27,7 +27,7 @@ main (void)
 	{
 	  if (a != 8 || b != 12 || e[0] != 'a' || f[0] != 'b')
 	    j++;
-#pragma omp barrier
+#pragma omp barrier	/* { dg-warning "may not be closely nested" } */
 #pragma omp atomic
 	  a += i;
 	  b += i;
@@ -38,7 +38,7 @@ main (void)
 	  f[0] += i;
 	  g[0] = 'g' + i;
 	  h[0] = 'h' + i;
-#pragma omp barrier
+#pragma omp barrier	/* { dg-warning "may not be closely nested" } */
 	  if (a != 8 + 6 || b != 12 + i || c != i || d != i)
 	    j += 8;
 	  if (e[0] != 'a' + 6 || f[0] != 'b' + i || g[0] != 'g' + i)
diff --git a/libgomp/testsuite/libgomp.c/sort-1.c b/libgomp/testsuite/libgomp.c/sort-1.c
new file mode 100644
index 00000000000..269d69da12c
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/sort-1.c
@@ -0,0 +1,379 @@
+/* Test and benchmark of a couple of parallel sorting algorithms.
+   Copyright (C) 2008 Free Software Foundation, Inc.
+
+   GCC is free software; you can redistribute it and/or modify it under
+   the terms of the GNU General Public License as published by the Free
+   Software Foundation; either version 3, or (at your option) any later
+   version.
+
+   GCC is distributed in the hope that it will be useful, but WITHOUT ANY
+   WARRANTY; without even the implied warranty of MERCHANTABILITY or
+   FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+   for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with GCC; see the file COPYING3.  If not see
+   <http://www.gnu.org/licenses/>.  */
+
+#include <limits.h>
+#include <omp.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+int failures;
+
+#define THRESHOLD 100
+
+static void
+verify (const char *name, double stime, int *array, int count)
+{
+  int i;
+  double etime = omp_get_wtime ();
+
+  printf ("%s: %g\n", name, etime - stime);
+  for (i = 1; i < count; i++)
+    if (array[i] < array[i - 1])
+      {
+	printf ("%s: incorrectly sorted\n", name);
+	failures = 1;
+      }
+}
+
+static void
+insertsort (int *array, int s, int e)
+{
+  int i, j, val;
+  for (i = s + 1; i <= e; i++)
+    {
+      val = array[i];
+      j = i;
+      while (j-- > s && val < array[j])
+	array[j + 1] = array[j];
+      array[j + 1] = val;
+    }
+}
+
+struct int_pair
+{
+  int lo;
+  int hi;
+};
+
+struct int_pair_stack
+{
+  struct int_pair *top;
+#define STACK_SIZE 4 * CHAR_BIT * sizeof (int)
+  struct int_pair arr[STACK_SIZE];
+};
+
+static inline void
+init_int_pair_stack (struct int_pair_stack *stack)
+{
+  stack->top = &stack->arr[0];
+}
+
+static inline void
+push_int_pair_stack (struct int_pair_stack *stack, int lo, int hi)
+{
+  stack->top->lo = lo;
+  stack->top->hi = hi;
+  stack->top++;
+}
+
+static inline void
+pop_int_pair_stack (struct int_pair_stack *stack, int *lo, int *hi)
+{
+  stack->top--;
+  *lo = stack->top->lo;
+  *hi = stack->top->hi;
+}
+
+static inline int
+size_int_pair_stack (struct int_pair_stack *stack)
+{
+  return stack->top - &stack->arr[0];
+}
+
+static inline void
+busy_wait (void)
+{
+#if defined __i386__ || defined __x86_64__
+  __asm volatile ("rep; nop" : : : "memory");
+#elif defined __ia64__
+  __asm volatile ("hint @pause" : : : "memory");
+#elif defined __sparc__ && (defined __arch64__ || defined __sparc_v9__)
+  __asm volatile ("membar #LoadLoad" : : : "memory");
+#else
+  __asm volatile ("" : : : "memory");
+#endif
+}
+
+static inline void
+swap (int *array, int a, int b)
+{
+  int val = array[a];
+  array[a] = array[b];
+  array[b] = val;
+}
+
+static inline int
+choose_pivot (int *array, int lo, int hi)
+{
+  int mid = (lo + hi) / 2;
+
+  if (array[mid] < array[lo])
+    swap (array, lo, mid);
+  if (array[hi] < array[mid])
+    {
+      swap (array, mid, hi);
+      if (array[mid] < array[lo])
+	swap (array, lo, mid);
+    }
+  return array[mid];
+}
+
+static inline int
+partition (int *array, int lo, int hi)
+{
+  int pivot = choose_pivot (array, lo, hi);
+  int left = lo;
+  int right = hi;
+
+  for (;;)
+    {
+      while (array[++left] < pivot);
+      while (array[--right] > pivot);
+      if (left >= right)
+	break;
+      swap (array, left, right);
+    }
+  return left;
+}
+
+static void
+sort1 (int *array, int count)
+{
+  omp_lock_t lock;
+  struct int_pair_stack global_stack;
+  int busy = 1;
+  int num_threads;
+
+  omp_init_lock (&lock);
+  init_int_pair_stack (&global_stack);
+  #pragma omp parallel firstprivate (array, count)
+  {
+    int lo = 0, hi = 0, mid, next_lo, next_hi;
+    bool idle = true;
+    struct int_pair_stack local_stack;
+
+    init_int_pair_stack (&local_stack);
+    if (omp_get_thread_num () == 0)
+      {
+	num_threads = omp_get_num_threads ();
+	hi = count - 1;
+	idle = false;
+      }
+
+    for (;;)
+      {
+	if (hi - lo < THRESHOLD)
+	  {
+	    insertsort (array, lo, hi);
+	    lo = hi;
+	  }
+	if (lo >= hi)
+	  {
+	    if (size_int_pair_stack (&local_stack) == 0)
+	      {
+	      again:
+		omp_set_lock (&lock);
+		if (size_int_pair_stack (&global_stack) == 0)
+		  {
+		    if (!idle)
+		      busy--;
+		    if (busy == 0)
+		      {
+			omp_unset_lock (&lock);
+			break;
+		      }
+		    omp_unset_lock (&lock);
+		    idle = true;
+		    while (size_int_pair_stack (&global_stack) == 0
+			   && busy)
+		      busy_wait ();
+		    goto again;
+		  }
+		if (idle)
+		  busy++;
+		pop_int_pair_stack (&global_stack, &lo, &hi);
+		omp_unset_lock (&lock);
+		idle = false;
+	      }
+	    else
+	      pop_int_pair_stack (&local_stack, &lo, &hi);
+	  }
+
+	mid = partition (array, lo, hi);
+	if (mid - lo < hi - mid)
+	  {
+	    next_lo = mid;
+	    next_hi = hi;
+	    hi = mid - 1;
+	  }
+	else
+	  {
+	    next_lo = lo;
+	    next_hi = mid - 1;
+	    lo = mid;
+	  }
+
+	if (next_hi - next_lo < THRESHOLD)
+	  insertsort (array, next_lo, next_hi);
+	else
+	  {
+	    if (size_int_pair_stack (&global_stack) < num_threads - 1)
+	      {
+		int size;
+
+		omp_set_lock (&lock);
+		size = size_int_pair_stack (&global_stack);
+		if (size < num_threads - 1 && size < STACK_SIZE)
+		  push_int_pair_stack (&global_stack, next_lo, next_hi);
+		else
+		  push_int_pair_stack (&local_stack, next_lo, next_hi);
+		omp_unset_lock (&lock);
+	      }
+	    else
+	      push_int_pair_stack (&local_stack, next_lo, next_hi);
+	  }
+      }
+    }
+  omp_destroy_lock (&lock);
+}
+
+static void
+sort2_1 (int *array, int lo, int hi, int num_threads, int *busy)
+{
+  int mid;
+
+  if (hi - lo < THRESHOLD)
+    {
+      insertsort (array, lo, hi);
+      return;
+    }
+
+  mid = partition (array, lo, hi);
+
+  if (*busy >= num_threads)
+    {
+      sort2_1 (array, lo, mid - 1, num_threads, busy);
+      sort2_1 (array, mid, hi, num_threads, busy);
+      return;
+    }
+
+  #pragma omp atomic
+    *busy += 1;
+
+  #pragma omp parallel num_threads (2) \
+		       firstprivate (array, lo, hi, mid, num_threads, busy)
+  {
+    if (omp_get_thread_num () == 0)
+      sort2_1 (array, lo, mid - 1, num_threads, busy);
+    else
+      {
+	sort2_1 (array, mid, hi, num_threads, busy);
+	#pragma omp atomic
+	  *busy -= 1;
+      }
+  }
+}
+
+static void
+sort2 (int *array, int count)
+{
+  int num_threads;
+  int busy = 1;
+
+  #pragma omp parallel
+    #pragma omp single nowait
+      num_threads = omp_get_num_threads ();
+
+  sort2_1 (array, 0, count - 1, num_threads, &busy);
+}
+
+#if _OPENMP >= 200805
+static void
+sort3_1 (int *array, int lo, int hi)
+{
+  int mid;
+
+  if (hi - lo < THRESHOLD)
+    {
+      insertsort (array, lo, hi);
+      return;
+    }
+
+  mid = partition (array, lo, hi);
+  #pragma omp task
+    sort3_1 (array, lo, mid - 1);
+  sort3_1 (array, mid, hi);
+}
+
+static void
+sort3 (int *array, int count)
+{
+  #pragma omp parallel
+    #pragma omp single
+      sort3_1 (array, 0, count - 1);
+}
+#endif
+
+int
+main (int argc, char **argv)
+{
+  int i, count = 1000000;
+  double stime;
+  int *unsorted, *sorted, num_threads;
+  if (argc >= 2)
+    count = strtoul (argv[1], NULL, 0);
+
+  unsorted = malloc (count * sizeof (int));
+  sorted = malloc (count * sizeof (int));
+  if (unsorted == NULL || sorted == NULL)
+    {
+      puts ("allocation failure");
+      exit (1);
+    }
+
+  srand (0xdeadbeef);
+  for (i = 0; i < count; i++)
+    unsorted[i] = rand ();
+
+  omp_set_nested (1);
+  omp_set_dynamic (0);
+  #pragma omp parallel
+    #pragma omp single nowait
+      num_threads = omp_get_num_threads ();
+  printf ("Threads: %d\n", num_threads);
+
+  memcpy (sorted, unsorted, count * sizeof (int));
+  stime = omp_get_wtime ();
+  sort1 (sorted, count);
+  verify ("sort1", stime, sorted, count);
+
+  memcpy (sorted, unsorted, count * sizeof (int));
+  stime = omp_get_wtime ();
+  sort2 (sorted, count);
+  verify ("sort2", stime, sorted, count);
+
+#if _OPENMP >= 200805
+  memcpy (sorted, unsorted, count * sizeof (int));
+  stime = omp_get_wtime ();
+  sort3 (sorted, count);
+  verify ("sort3", stime, sorted, count);
+#endif
+
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/task-1.c b/libgomp/testsuite/libgomp.c/task-1.c
new file mode 100644
index 00000000000..66f58a29b87
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/task-1.c
@@ -0,0 +1,84 @@
+extern void abort (void);
+
+int a = 18;
+
+void
+f1 (int i, int j, int k)
+{
+  int l = 6, m = 7, n = 8;
+#pragma omp task private(j, m) shared(k, n)
+  {
+    j = 6;
+    m = 5;
+    if (++a != 19 || ++i != 9 || j != 6 || ++l != 7 || m != 5 || ++n != 9)
+      #pragma omp atomic
+	k++;
+  }
+#pragma omp taskwait
+  if (a != 19 || i != 8 || j != 26 || k != 0 || l != 6 || m != 7 || n != 9)
+    abort ();
+}
+
+int v1 = 1, v2 = 2, v5 = 5;
+int err;
+
+void
+f2 (void)
+{
+  int v3 = 3;
+#pragma omp sections private (v1) firstprivate (v2)
+  {
+  #pragma omp section
+    {
+      int v4 = 4;
+      v1 = 7;
+      #pragma omp task
+	{
+	  if (++v1 != 8 || ++v2 != 3 || ++v3 != 4 || ++v4 != 5 || ++v5 != 6)
+	    err = 1;
+	}
+      #pragma omp taskwait
+      if (v1 != 7 || v2 != 2 || v3 != 3 || v4 != 4 || v5 != 6)
+	abort ();
+      if (err)
+	abort ();
+    }
+  }
+}
+
+void
+f3 (int i, int j, int k)
+{
+  int l = 6, m = 7, n = 8;
+#pragma omp task private(j, m) shared(k, n) untied
+  {
+    j = 6;
+    m = 5;
+    if (++a != 19 || ++i != 9 || j != 6 || ++l != 7 || m != 5 || ++n != 9)
+      #pragma omp atomic
+	k++;
+  }
+#pragma omp taskwait
+  if (a != 19 || i != 8 || j != 26 || k != 0 || l != 6 || m != 7 || n != 9)
+    abort ();
+}
+
+int
+main (void)
+{
+  f1 (8, 26, 0);
+  f2 ();
+  a = 18;
+  f3 (8, 26, 0);
+  a = 18;
+#pragma omp parallel num_threads(4)
+  {
+    #pragma omp master
+      {
+	f1 (8, 26, 0);
+	a = 18;
+	f3 (8, 26, 0);
+      }
+  }
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/task-2.c b/libgomp/testsuite/libgomp.c/task-2.c
new file mode 100644
index 00000000000..ed6a09c3557
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/task-2.c
@@ -0,0 +1,53 @@
+extern void abort (void);
+
+int
+f1 (void)
+{
+  int a = 6, e = 0;
+  int nested (int x)
+  {
+    return x + a;
+  }
+  #pragma omp task
+  {
+    int n = nested (5);
+    if (n != 11)
+      #pragma omp atomic
+	e += 1;
+  }
+  #pragma omp taskwait
+  return e;
+}
+
+int
+f2 (void)
+{
+  int a = 6, e = 0;
+  int nested (int x)
+  {
+    return x + a;
+  }
+  a = nested (4);
+  #pragma omp task
+  {
+    if (a != 10)
+      #pragma omp atomic
+	e += 1;
+  }
+  #pragma omp taskwait
+  return e;
+}
+
+int
+main (void)
+{
+  int e = 0;
+  #pragma omp parallel num_threads(4) reduction(+:e)
+  {
+    e += f1 ();
+    e += f2 ();
+  }
+  if (e)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/task-3.c b/libgomp/testsuite/libgomp.c/task-3.c
new file mode 100644
index 00000000000..5657346bd15
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/task-3.c
@@ -0,0 +1,70 @@
+/* { dg-do run } */
+
+#include <omp.h>
+extern void abort ();
+
+int l = 5;
+
+int
+foo (int i)
+{
+  int j = 7;
+  const int k = 8;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp task firstprivate (i) shared (j, l)
+  {
+    #pragma omp critical
+      {
+	j += i;
+	l += k;
+      }
+  }
+  i++;
+  #pragma omp taskwait
+  return (i != 8 * omp_get_thread_num () + 4
+	  || j != 4 * i - 3
+	  || k != 8);
+}
+
+int
+main (void)
+{
+  int r = 0;
+  #pragma omp parallel num_threads (4) reduction(+:r)
+    if (omp_get_num_threads () != 4)
+      {
+	#pragma omp master
+	  l = 133;
+      }
+    else if (foo (8 * omp_get_thread_num ()))
+      r++;
+  if (r || l != 133)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.c/task-4.c b/libgomp/testsuite/libgomp.c/task-4.c
new file mode 100644
index 00000000000..18435930019
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/task-4.c
@@ -0,0 +1,40 @@
+/* { dg-do run } */
+
+#include <omp.h>
+#include <stdlib.h>
+#include <string.h>
+
+int e;
+
+void __attribute__((noinline))
+baz (int i, int *p, int j, int *q)
+{
+  if (p[0] != 1 || p[i] != 3 || q[0] != 2 || q[j] != 4)
+    #pragma omp atomic
+      e++;
+}
+
+void __attribute__((noinline))
+foo (int i, int j)
+{
+  int p[i + 1];
+  int q[j + 1];
+  memset (p, 0, sizeof (p));
+  memset (q, 0, sizeof (q));
+  p[0] = 1;
+  p[i] = 3;
+  q[0] = 2;
+  q[j] = 4;
+  #pragma omp task firstprivate (p, q)
+    baz (i, p, j, q);
+}
+
+int
+main (void)
+{
+  #pragma omp parallel num_threads (4)
+    foo (5 + omp_get_thread_num (), 7 + omp_get_thread_num ());
+  if (e)
+    abort ();
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.fortran/allocatable1.f90 b/libgomp/testsuite/libgomp.fortran/allocatable1.f90
new file mode 100644
index 00000000000..1efe2abe959
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/allocatable1.f90
@@ -0,0 +1,81 @@
+! { dg-do run }
+!$ use omp_lib
+
+  integer, allocatable :: a(:, :)
+  integer :: b(6, 3)
+  integer :: i, j
+  logical :: k, l
+  b(:, :) = 16
+  l = .false.
+  if (allocated (a)) call abort
+!$omp parallel private (a, b) reduction (.or.:l)
+  l = l.or.allocated (a)
+  allocate (a(3, 6))
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.18.or.size(a,1).ne.3.or.size(a,2).ne.6
+  a(3, 2) = 1
+  b(3, 2) = 1
+  deallocate (a)
+  l = l.or.allocated (a)
+!$omp end parallel
+  if (allocated (a).or.l) call abort
+  allocate (a(6, 3))
+  a(:, :) = 3
+  if (.not.allocated (a)) call abort
+  l = l.or.size(a).ne.18.or.size(a,1).ne.6.or.size(a,2).ne.3
+  if (l) call abort
+!$omp parallel private (a, b) reduction (.or.:l)
+  l = l.or..not.allocated (a)
+  a(3, 2) = 1
+  b(3, 2) = 1
+!$omp end parallel
+  if (l.or..not.allocated (a)) call abort
+!$omp parallel firstprivate (a, b) reduction (.or.:l)
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.18.or.size(a,1).ne.6.or.size(a,2).ne.3
+  do i = 1, 6
+    l = l.or.(a(i, 1).ne.3).or.(a(i, 2).ne.3)
+    l = l.or.(a(i, 3).ne.3).or.(b(i, 1).ne.16)
+    l = l.or.(b(i, 2).ne.16).or.(b(i, 3).ne.16)
+  end do
+  a(:, :) = omp_get_thread_num ()
+  b(:, :) = omp_get_thread_num ()
+!$omp end parallel
+  if (any (a.ne.3).or.any (b.ne.16).or.l) call abort
+  k = .true.
+!$omp parallel do firstprivate (a, b, k) lastprivate (a, b) &
+!$omp & reduction (.or.:l)
+  do i = 1, 36
+    l = l.or..not.allocated (a)
+    l = l.or.size(a).ne.18.or.size(a,1).ne.6.or.size(a,2).ne.3
+    if (k) then
+      do j = 1, 6
+        l = l.or.(a(j, 1).ne.3).or.(a(j, 2).ne.3)
+        l = l.or.(a(j, 3).ne.3).or.(b(j, 1).ne.16)
+	l = l.or.(b(j, 2).ne.16).or.(b(j, 3).ne.16)
+      end do
+      k = .false.
+    end if
+    a(:, :) = i + 2
+    b(:, :) = i
+  end do
+  if (any (a.ne.38).or.any (b.ne.36).or.l) call abort
+  deallocate (a)
+  if (allocated (a)) call abort
+  allocate (a (0:1, 0:3))
+  a(:, :) = 0
+!$omp parallel do reduction (+:a) reduction (.or.:l) &
+!$omp & num_threads(3) schedule(static)
+  do i = 0, 7
+    l = l.or..not.allocated (a)
+    l = l.or.size(a).ne.8.or.size(a,1).ne.2.or.size(a,2).ne.4
+    a(modulo (i, 2), i / 2) = a(modulo (i, 2), i / 2) + i
+    a(i / 4, modulo (i, 4)) = a(i / 4, modulo (i, 4)) + i
+  end do
+  if (l) call abort
+  do i = 0, 1
+    do j = 0, 3
+      if (a(i, j) .ne. (5*i + 3*j)) call abort
+    end do
+  end do
+end
diff --git a/libgomp/testsuite/libgomp.fortran/allocatable2.f90 b/libgomp/testsuite/libgomp.fortran/allocatable2.f90
new file mode 100644
index 00000000000..a37616b04b1
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/allocatable2.f90
@@ -0,0 +1,47 @@
+! { dg-do run }
+! { dg-require-effective-target tls_runtime }
+!$ use omp_lib
+
+  integer, save, allocatable :: a(:, :)
+  integer, allocatable :: b(:, :)
+  integer :: n
+  logical :: l
+!$omp threadprivate (a)
+  if (allocated (a)) call abort
+  call omp_set_dynamic (.false.)
+  l = .false.
+!$omp parallel num_threads (4) reduction(.or.:l)
+  allocate (a(-1:1, 7:10))
+  a(:, :) = omp_get_thread_num () + 6
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.12.or.size(a,1).ne.3.or.size(a,2).ne.4
+!$omp end parallel
+  if (l.or.any(a.ne.6)) call abort ()
+!$omp parallel num_threads (4) copyin (a) reduction(.or.:l) private (b)
+  l = l.or.allocated (b)
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.12.or.size(a,1).ne.3.or.size(a,2).ne.4
+  l = l.or.any(a.ne.6)
+  allocate (b(1, 3))
+  a(:, :) = omp_get_thread_num () + 36
+  b(:, :) = omp_get_thread_num () + 66
+  !$omp single
+    n = omp_get_thread_num ()
+  !$omp end single copyprivate (a, b)
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.12.or.size(a,1).ne.3.or.size(a,2).ne.4
+  l = l.or.any(a.ne.(n + 36))
+  l = l.or..not.allocated (b)
+  l = l.or.size(b).ne.3.or.size(b,1).ne.1.or.size(b,2).ne.3
+  l = l.or.any(b.ne.(n + 66))
+  deallocate (b)
+  l = l.or.allocated (b)
+!$omp end parallel
+  if (n.lt.0 .or. n.ge.4) call abort
+  if (l.or.any(a.ne.(n + 36))) call abort
+!$omp parallel num_threads (4) reduction(.or.:l)
+  deallocate (a)
+  l = l.or.allocated (a)
+!$omp end parallel
+  if (l.or.allocated (a)) call abort
+end
diff --git a/libgomp/testsuite/libgomp.fortran/allocatable3.f90 b/libgomp/testsuite/libgomp.fortran/allocatable3.f90
new file mode 100644
index 00000000000..fe3714a2b1f
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/allocatable3.f90
@@ -0,0 +1,21 @@
+! { dg-do run }
+
+  integer, allocatable :: a(:)
+  integer :: i
+  logical :: l
+  l = .false.
+  if (allocated (a)) call abort
+!$omp parallel private (a) reduction (.or.:l)
+  allocate (a (-7:-5))
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.3.or.size(a,1).ne.3
+  a(:) = 0
+  !$omp do private (a)
+  do i = 1, 7
+    a(:) = i
+    l = l.or.any (a.ne.i)
+  end do
+  l = l.or.any (a.ne.0)
+  deallocate (a)
+!$omp end parallel
+end
diff --git a/libgomp/testsuite/libgomp.fortran/allocatable4.f90 b/libgomp/testsuite/libgomp.fortran/allocatable4.f90
new file mode 100644
index 00000000000..996578c94fa
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/allocatable4.f90
@@ -0,0 +1,47 @@
+! { dg-do run }
+
+  integer, allocatable :: a(:, :)
+  integer :: b(6, 3)
+  integer :: i, j
+  logical :: k, l
+  b(:, :) = 16
+  l = .false.
+  if (allocated (a)) call abort
+!$omp task private (a, b) shared (l)
+  l = l.or.allocated (a)
+  allocate (a(3, 6))
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.18.or.size(a,1).ne.3.or.size(a,2).ne.6
+  a(3, 2) = 1
+  b(3, 2) = 1
+  deallocate (a)
+  l = l.or.allocated (a)
+!$omp end task
+!$omp taskwait
+  if (allocated (a).or.l) call abort
+  allocate (a(6, 3))
+  a(:, :) = 3
+  if (.not.allocated (a)) call abort
+  l = l.or.size(a).ne.18.or.size(a,1).ne.6.or.size(a,2).ne.3
+  if (l) call abort
+!$omp task private (a, b) shared (l)
+  l = l.or..not.allocated (a)
+  a(3, 2) = 1
+  b(3, 2) = 1
+!$omp end task
+!$omp taskwait
+  if (l.or..not.allocated (a)) call abort
+!$omp task firstprivate (a, b) shared (l)
+  l = l.or..not.allocated (a)
+  l = l.or.size(a).ne.18.or.size(a,1).ne.6.or.size(a,2).ne.3
+  do i = 1, 6
+    l = l.or.(a(i, 1).ne.3).or.(a(i, 2).ne.3)
+    l = l.or.(a(i, 3).ne.3).or.(b(i, 1).ne.16)
+    l = l.or.(b(i, 2).ne.16).or.(b(i, 3).ne.16)
+  end do
+  a(:, :) = 7
+  b(:, :) = 8
+!$omp end task
+!$omp taskwait
+  if (any (a.ne.3).or.any (b.ne.16).or.l) call abort
+end
diff --git a/libgomp/testsuite/libgomp.fortran/collapse1.f90 b/libgomp/testsuite/libgomp.fortran/collapse1.f90
new file mode 100644
index 00000000000..1ecfa0c9365
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/collapse1.f90
@@ -0,0 +1,26 @@
+! { dg-do run }
+
+program collapse1
+  integer :: i, j, k, a(1:3, 4:6, 5:7)
+  logical :: l
+  l = .false.
+  a(:, :, :) = 0
+  !$omp parallel do collapse(4 - 1) schedule(static, 4)
+    do i = 1, 3
+      do j = 4, 6
+        do k = 5, 7
+          a(i, j, k) = i + j + k
+        end do
+      end do
+    end do
+  !$omp parallel do collapse(2) reduction(.or.:l)
+    do i = 1, 3
+      do j = 4, 6
+        do k = 5, 7
+          if (a(i, j, k) .ne. (i + j + k)) l = .true.
+        end do
+      end do
+    end do
+  !$omp end parallel do
+  if (l) call abort
+end program collapse1
diff --git a/libgomp/testsuite/libgomp.fortran/collapse2.f90 b/libgomp/testsuite/libgomp.fortran/collapse2.f90
new file mode 100644
index 00000000000..77e0dee8260
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/collapse2.f90
@@ -0,0 +1,53 @@
+! { dg-do run }
+
+program collapse2
+  call test1
+  call test2
+contains
+  subroutine test1
+    integer :: i, j, k, a(1:3, 4:6, 5:7)
+    logical :: l
+    l = .false.
+    a(:, :, :) = 0
+    !$omp parallel do collapse(4 - 1) schedule(static, 4)
+      do 164 i = 1, 3
+        do 164 j = 4, 6
+          do 164 k = 5, 7
+            a(i, j, k) = i + j + k
+164      end do
+    !$omp parallel do collapse(2) reduction(.or.:l)
+firstdo: do i = 1, 3
+        do j = 4, 6
+          do k = 5, 7
+            if (a(i, j, k) .ne. (i + j + k)) l = .true.
+          end do
+        end do
+      end do firstdo
+    !$omp end parallel do
+    if (l) call abort
+  end subroutine test1
+
+  subroutine test2
+    integer :: a(3,3,3), k, kk, kkk, l, ll, lll
+    !$omp do collapse(3)
+      do 115 k=1,3
+  dokk: do kk=1,3
+          do kkk=1,3
+            a(k,kk,kkk) = 1
+          enddo
+        enddo dokk
+115   continue
+    if (any(a(1:3,1:3,1:3).ne.1)) call abort
+
+    !$omp do collapse(3)
+ dol: do 120 l=1,3
+  doll: do ll=1,3
+          do lll=1,3
+            a(l,ll,lll) = 2
+          enddo
+        enddo doll
+120   end do dol
+    if (any(a(1:3,1:3,1:3).ne.2)) call abort
+  end subroutine test2
+
+end program collapse2
diff --git a/libgomp/testsuite/libgomp.fortran/collapse3.f90 b/libgomp/testsuite/libgomp.fortran/collapse3.f90
new file mode 100644
index 00000000000..eac9eac651b
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/collapse3.f90
@@ -0,0 +1,204 @@
+! { dg-do run }
+
+program collapse3
+  call test1
+  call test2 (2, 6, -2, 4, 13, 18)
+  call test3 (2, 6, -2, 4, 13, 18, 1, 1, 1)
+  call test4
+  call test5 (2, 6, -2, 4, 13, 18)
+  call test6 (2, 6, -2, 4, 13, 18, 1, 1, 1)
+contains
+  subroutine test1
+    integer :: i, j, k, a(1:7, -3:5, 12:19), m
+    logical :: l
+    l = .false.
+    a(:, :, :) = 0
+    !$omp parallel do collapse (3) lastprivate (i, j, k, m) reduction (.or.:l)
+      do i = 2, 6
+        do j = -2, 4
+          do k = 13, 18
+            l = l.or.i.lt.2.or.i.gt.6.or.j.lt.-2.or.j.gt.4
+            l = l.or.k.lt.13.or.k.gt.18
+            if (.not.l) a(i, j, k) = a(i, j, k) + 1
+            m = i * 100 + j * 10 + k
+          end do
+        end do
+      end do
+    if (i.ne.7.or.j.ne.5.or.k.ne.19) call abort
+    if (m.ne.(600+40+18)) call abort
+    do i = 1, 7
+      do j = -3, 5
+        do k = 12, 19
+          if (i.eq.1.or.i.eq.7.or.j.eq.-3.or.j.eq.5.or.k.eq.12.or.k.eq.19) then
+            if (a(i, j, k).ne.0) print *, i, j, k
+          else
+            if (a(i, j, k).ne.1) print *, 'kk', i, j, k, a(i, j, k)
+          end if
+        end do
+      end do
+    end do
+  end subroutine test1
+
+  subroutine test2(v1, v2, v3, v4, v5, v6)
+    integer :: i, j, k, a(1:7, -3:5, 12:19), m
+    integer :: v1, v2, v3, v4, v5, v6
+    logical :: l
+    l = .false.
+    a(:, :, :) = 0
+    !$omp parallel do collapse (3) lastprivate (i, j, k, m) reduction (.or.:l)
+      do i = v1, v2
+        do j = v3, v4
+          do k = v5, v6
+            l = l.or.i.lt.2.or.i.gt.6.or.j.lt.-2.or.j.gt.4
+            l = l.or.k.lt.13.or.k.gt.18
+            if (.not.l) a(i, j, k) = a(i, j, k) + 1
+            m = i * 100 + j * 10 + k
+          end do
+        end do
+      end do
+    if (i.ne.7.or.j.ne.5.or.k.ne.19) call abort
+    if (m.ne.(600+40+18)) call abort
+    do i = 1, 7
+      do j = -3, 5
+        do k = 12, 19
+          if (i.eq.1.or.i.eq.7.or.j.eq.-3.or.j.eq.5.or.k.eq.12.or.k.eq.19) then
+            if (a(i, j, k).ne.0) print *, i, j, k
+          else
+            if (a(i, j, k).ne.1) print *, 'kk', i, j, k, a(i, j, k)
+          end if
+        end do
+      end do
+    end do
+  end subroutine test2
+
+  subroutine test3(v1, v2, v3, v4, v5, v6, v7, v8, v9)
+    integer :: i, j, k, a(1:7, -3:5, 12:19), m
+    integer :: v1, v2, v3, v4, v5, v6, v7, v8, v9
+    logical :: l
+    l = .false.
+    a(:, :, :) = 0
+    !$omp parallel do collapse (3) lastprivate (i, j, k, m) reduction (.or.:l)
+      do i = v1, v2, v7
+        do j = v3, v4, v8
+          do k = v5, v6, v9
+            l = l.or.i.lt.2.or.i.gt.6.or.j.lt.-2.or.j.gt.4
+            l = l.or.k.lt.13.or.k.gt.18
+            if (.not.l) a(i, j, k) = a(i, j, k) + 1
+            m = i * 100 + j * 10 + k
+          end do
+        end do
+      end do
+    if (i.ne.7.or.j.ne.5.or.k.ne.19) call abort
+    if (m.ne.(600+40+18)) call abort
+    do i = 1, 7
+      do j = -3, 5
+        do k = 12, 19
+          if (i.eq.1.or.i.eq.7.or.j.eq.-3.or.j.eq.5.or.k.eq.12.or.k.eq.19) then
+            if (a(i, j, k).ne.0) print *, i, j, k
+          else
+            if (a(i, j, k).ne.1) print *, 'kk', i, j, k, a(i, j, k)
+          end if
+        end do
+      end do
+    end do
+  end subroutine test3
+
+  subroutine test4
+    integer :: i, j, k, a(1:7, -3:5, 12:19), m
+    logical :: l
+    l = .false.
+    a(:, :, :) = 0
+    !$omp parallel do collapse (3) lastprivate (i, j, k, m) reduction (.or.:l) &
+    !$omp& schedule (dynamic, 5)
+      do i = 2, 6
+        do j = -2, 4
+          do k = 13, 18
+            l = l.or.i.lt.2.or.i.gt.6.or.j.lt.-2.or.j.gt.4
+            l = l.or.k.lt.13.or.k.gt.18
+            if (.not.l) a(i, j, k) = a(i, j, k) + 1
+            m = i * 100 + j * 10 + k
+          end do
+        end do
+      end do
+    if (i.ne.7.or.j.ne.5.or.k.ne.19) call abort
+    if (m.ne.(600+40+18)) call abort
+    do i = 1, 7
+      do j = -3, 5
+        do k = 12, 19
+          if (i.eq.1.or.i.eq.7.or.j.eq.-3.or.j.eq.5.or.k.eq.12.or.k.eq.19) then
+            if (a(i, j, k).ne.0) print *, i, j, k
+          else
+            if (a(i, j, k).ne.1) print *, 'kk', i, j, k, a(i, j, k)
+          end if
+        end do
+      end do
+    end do
+  end subroutine test4
+
+  subroutine test5(v1, v2, v3, v4, v5, v6)
+    integer :: i, j, k, a(1:7, -3:5, 12:19), m
+    integer :: v1, v2, v3, v4, v5, v6
+    logical :: l
+    l = .false.
+    a(:, :, :) = 0
+    !$omp parallel do collapse (3) lastprivate (i, j, k, m) reduction (.or.:l) &
+    !$omp & schedule (guided)
+      do i = v1, v2
+        do j = v3, v4
+          do k = v5, v6
+            l = l.or.i.lt.2.or.i.gt.6.or.j.lt.-2.or.j.gt.4
+            l = l.or.k.lt.13.or.k.gt.18
+            if (.not.l) a(i, j, k) = a(i, j, k) + 1
+            m = i * 100 + j * 10 + k
+          end do
+        end do
+      end do
+    if (i.ne.7.or.j.ne.5.or.k.ne.19) call abort
+    if (m.ne.(600+40+18)) call abort
+    do i = 1, 7
+      do j = -3, 5
+        do k = 12, 19
+          if (i.eq.1.or.i.eq.7.or.j.eq.-3.or.j.eq.5.or.k.eq.12.or.k.eq.19) then
+            if (a(i, j, k).ne.0) print *, i, j, k
+          else
+            if (a(i, j, k).ne.1) print *, 'kk', i, j, k, a(i, j, k)
+          end if
+        end do
+      end do
+    end do
+  end subroutine test5
+
+  subroutine test6(v1, v2, v3, v4, v5, v6, v7, v8, v9)
+    integer :: i, j, k, a(1:7, -3:5, 12:19), m
+    integer :: v1, v2, v3, v4, v5, v6, v7, v8, v9
+    logical :: l
+    l = .false.
+    a(:, :, :) = 0
+    !$omp parallel do collapse (3) lastprivate (i, j, k, m) reduction (.or.:l) &
+    !$omp & schedule (dynamic)
+      do i = v1, v2, v7
+        do j = v3, v4, v8
+          do k = v5, v6, v9
+            l = l.or.i.lt.2.or.i.gt.6.or.j.lt.-2.or.j.gt.4
+            l = l.or.k.lt.13.or.k.gt.18
+            if (.not.l) a(i, j, k) = a(i, j, k) + 1
+            m = i * 100 + j * 10 + k
+          end do
+        end do
+      end do
+    if (i.ne.7.or.j.ne.5.or.k.ne.19) call abort
+    if (m.ne.(600+40+18)) call abort
+    do i = 1, 7
+      do j = -3, 5
+        do k = 12, 19
+          if (i.eq.1.or.i.eq.7.or.j.eq.-3.or.j.eq.5.or.k.eq.12.or.k.eq.19) then
+            if (a(i, j, k).ne.0) print *, i, j, k
+          else
+            if (a(i, j, k).ne.1) print *, 'kk', i, j, k, a(i, j, k)
+          end if
+        end do
+      end do
+    end do
+  end subroutine test6
+
+end program collapse3
diff --git a/libgomp/testsuite/libgomp.fortran/collapse4.f90 b/libgomp/testsuite/libgomp.fortran/collapse4.f90
new file mode 100644
index 00000000000..f19b0f6c695
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/collapse4.f90
@@ -0,0 +1,12 @@
+! { dg-do run }
+
+  integer :: i, j, k
+  !$omp parallel do lastprivate (i, j, k) collapse (3)
+    do i = 0, 17
+      do j = 0, 6
+        do k = 0, 5
+        end do
+      end do
+    end do
+  if (i .ne. 18 .or. j .ne. 7 .or. k .ne. 6) call abort
+end
diff --git a/libgomp/testsuite/libgomp.fortran/lastprivate1.f90 b/libgomp/testsuite/libgomp.fortran/lastprivate1.f90
new file mode 100644
index 00000000000..91bb96ca75a
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/lastprivate1.f90
@@ -0,0 +1,126 @@
+program lastprivate
+  integer :: i
+  common /c/ i
+  !$omp parallel num_threads (4)
+  call test1
+  !$omp end parallel
+  if (i .ne. 21) call abort
+  !$omp parallel num_threads (4)
+  call test2
+  !$omp end parallel
+  if (i .ne. 64) call abort
+  !$omp parallel num_threads (4)
+  call test3
+  !$omp end parallel
+  if (i .ne. 14) call abort
+  call test4
+  call test5
+  call test6
+  call test7
+  call test8
+  call test9
+  call test10
+  call test11
+  call test12
+contains
+  subroutine test1
+    integer :: i
+    common /c/ i
+    !$omp do lastprivate (i)
+    do i = 1, 20
+    end do
+  end subroutine test1
+  subroutine test2
+    integer :: i
+    common /c/ i
+    !$omp do lastprivate (i)
+    do i = 7, 61, 3
+    end do
+  end subroutine test2
+  function ret3 ()
+    integer :: ret3
+    ret3 = 3
+  end function ret3
+  subroutine test3
+    integer :: i
+    common /c/ i
+    !$omp do lastprivate (i)
+    do i = -10, 11, ret3 ()
+    end do
+  end subroutine test3
+  subroutine test4
+    integer :: j
+    !$omp parallel do lastprivate (j) num_threads (4) default (none)
+    do j = 1, 20
+    end do
+    if (j .ne. 21) call abort
+  end subroutine test4
+  subroutine test5
+    integer :: j
+    !$omp parallel do lastprivate (j) num_threads (4) default (none)
+    do j = 7, 61, 3
+    end do
+    if (j .ne. 64) call abort
+  end subroutine test5
+  subroutine test6
+    integer :: j
+    !$omp parallel do lastprivate (j) num_threads (4) default (none)
+    do j = -10, 11, ret3 ()
+    end do
+    if (j .ne. 14) call abort
+  end subroutine test6
+  subroutine test7
+    integer :: i
+    common /c/ i
+    !$omp parallel do lastprivate (i) num_threads (4) default (none)
+    do i = 1, 20
+    end do
+    if (i .ne. 21) call abort
+  end subroutine test7
+  subroutine test8
+    integer :: i
+    common /c/ i
+    !$omp parallel do lastprivate (i) num_threads (4) default (none)
+    do i = 7, 61, 3
+    end do
+    if (i .ne. 64) call abort
+  end subroutine test8
+  subroutine test9
+    integer :: i
+    common /c/ i
+    !$omp parallel do lastprivate (i) num_threads (4) default (none)
+    do i = -10, 11, ret3 ()
+    end do
+    if (i .ne. 14) call abort
+  end subroutine test9
+  subroutine test10
+    integer :: i
+    common /c/ i
+    !$omp parallel num_threads (4) default (none) shared (i)
+    !$omp do lastprivate (i)
+    do i = 1, 20
+    end do
+    !$omp end parallel
+    if (i .ne. 21) call abort
+  end subroutine test10
+  subroutine test11
+    integer :: i
+    common /c/ i
+    !$omp parallel num_threads (4) default (none) shared (i)
+    !$omp do lastprivate (i)
+    do i = 7, 61, 3
+    end do
+    !$omp end parallel
+    if (i .ne. 64) call abort
+  end subroutine test11
+  subroutine test12
+    integer :: i
+    common /c/ i
+    !$omp parallel num_threads (4) default (none) shared (i)
+    !$omp do lastprivate (i)
+    do i = -10, 11, ret3 ()
+    end do
+    !$omp end parallel
+    if (i .ne. 14) call abort
+  end subroutine test12
+end program lastprivate
diff --git a/libgomp/testsuite/libgomp.fortran/lastprivate2.f90 b/libgomp/testsuite/libgomp.fortran/lastprivate2.f90
new file mode 100644
index 00000000000..6d7e11eab00
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/lastprivate2.f90
@@ -0,0 +1,141 @@
+program lastprivate
+  integer :: i, k
+  common /c/ i, k
+  !$omp parallel num_threads (4)
+  call test1
+  !$omp end parallel
+  if (i .ne. 21 .or. k .ne. 20) call abort
+  !$omp parallel num_threads (4)
+  call test2
+  !$omp end parallel
+  if (i .ne. 64 .or. k .ne. 61) call abort
+  !$omp parallel num_threads (4)
+  call test3
+  !$omp end parallel
+  if (i .ne. 14 .or. k .ne. 11) call abort
+  call test4
+  call test5
+  call test6
+  call test7
+  call test8
+  call test9
+  call test10
+  call test11
+  call test12
+contains
+  subroutine test1
+    integer :: i, k
+    common /c/ i, k
+    !$omp do lastprivate (i, k)
+    do i = 1, 20
+      k = i
+    end do
+  end subroutine test1
+  subroutine test2
+    integer :: i, k
+    common /c/ i, k
+    !$omp do lastprivate (i, k)
+    do i = 7, 61, 3
+      k = i
+    end do
+  end subroutine test2
+  function ret3 ()
+    integer :: ret3
+    ret3 = 3
+  end function ret3
+  subroutine test3
+    integer :: i, k
+    common /c/ i, k
+    !$omp do lastprivate (i, k)
+    do i = -10, 11, ret3 ()
+      k = i
+    end do
+  end subroutine test3
+  subroutine test4
+    integer :: j, l
+    !$omp parallel do lastprivate (j, l) num_threads (4)
+    do j = 1, 20
+      l = j
+    end do
+    if (j .ne. 21 .or. l .ne. 20) call abort
+  end subroutine test4
+  subroutine test5
+    integer :: j, l
+    l = 77
+    !$omp parallel do lastprivate (j, l) num_threads (4) firstprivate (l)
+    do j = 7, 61, 3
+      l = j
+    end do
+    if (j .ne. 64 .or. l .ne. 61) call abort
+  end subroutine test5
+  subroutine test6
+    integer :: j, l
+    !$omp parallel do lastprivate (j, l) num_threads (4)
+    do j = -10, 11, ret3 ()
+      l = j
+    end do
+    if (j .ne. 14 .or. l .ne. 11) call abort
+  end subroutine test6
+  subroutine test7
+    integer :: i, k
+    common /c/ i, k
+    !$omp parallel do lastprivate (i, k) num_threads (4)
+    do i = 1, 20
+      k = i
+    end do
+    if (i .ne. 21 .or. k .ne. 20) call abort
+  end subroutine test7
+  subroutine test8
+    integer :: i, k
+    common /c/ i, k
+    !$omp parallel do lastprivate (i, k) num_threads (4)
+    do i = 7, 61, 3
+      k = i
+    end do
+    if (i .ne. 64 .or. k .ne. 61) call abort
+  end subroutine test8
+  subroutine test9
+    integer :: i, k
+    common /c/ i, k
+    k = 77
+    !$omp parallel do lastprivate (i, k) num_threads (4) firstprivate (k)
+    do i = -10, 11, ret3 ()
+      k = i
+    end do
+    if (i .ne. 14 .or. k .ne. 11) call abort
+  end subroutine test9
+  subroutine test10
+    integer :: i, k
+    common /c/ i, k
+    !$omp parallel num_threads (4)
+    !$omp do lastprivate (i, k)
+    do i = 1, 20
+      k = i
+    end do
+    !$omp end parallel
+    if (i .ne. 21 .or. k .ne. 20) call abort
+  end subroutine test10
+  subroutine test11
+    integer :: i, k
+    common /c/ i, k
+    !$omp parallel num_threads (4)
+    !$omp do lastprivate (i, k)
+    do i = 7, 61, 3
+      k = i
+    end do
+    !$omp end parallel
+    if (i .ne. 64 .or. k .ne. 61) call abort
+  end subroutine test11
+  subroutine test12
+    integer :: i, k
+    common /c/ i, k
+    k = 77
+    !$omp parallel num_threads (4)
+    !$omp do lastprivate (i, k) firstprivate (k)
+    do i = -10, 11, ret3 ()
+      k = i
+    end do
+    !$omp end parallel
+    if (i .ne. 14 .or. k .ne. 11) call abort
+  end subroutine test12
+end program lastprivate
diff --git a/libgomp/testsuite/libgomp.fortran/lib4.f90 b/libgomp/testsuite/libgomp.fortran/lib4.f90
new file mode 100644
index 00000000000..cbb984574ff
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/lib4.f90
@@ -0,0 +1,16 @@
+! { dg-do run }
+
+program lib4
+  use omp_lib
+  integer (omp_sched_kind) :: kind
+  integer :: modifier
+  call omp_set_schedule (omp_sched_static, 32)
+  call omp_get_schedule (kind, modifier)
+  if (kind.ne.omp_sched_static.or.modifier.ne.32) call abort
+  call omp_set_schedule (omp_sched_dynamic, 4)
+  call omp_get_schedule (kind, modifier)
+  if (kind.ne.omp_sched_dynamic.or.modifier.ne.4) call abort
+  if (omp_get_thread_limit ().lt.0) call abort
+  call omp_set_max_active_levels (6)
+  if (omp_get_max_active_levels ().ne.6) call abort
+end program lib4
diff --git a/libgomp/testsuite/libgomp.fortran/lock-1.f90 b/libgomp/testsuite/libgomp.fortran/lock-1.f90
new file mode 100644
index 00000000000..d7d3e3fd6cc
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/lock-1.f90
@@ -0,0 +1,24 @@
+! { dg-do run }
+
+  use omp_lib
+
+  integer (kind = omp_nest_lock_kind) :: lock
+  logical :: l
+
+  l = .false.
+  call omp_init_nest_lock (lock)
+  if (omp_test_nest_lock (lock) .ne. 1) call abort
+  if (omp_test_nest_lock (lock) .ne. 2) call abort
+!$omp parallel if (.false.) reduction (.or.:l)
+  ! In OpenMP 2.5 this was supposed to return 3,
+  ! but in OpenMP 3.0 the parallel region has a different
+  ! task and omp_*_lock_t are owned by tasks, not by threads.
+  if (omp_test_nest_lock (lock) .ne. 0) l = .true.
+!$omp end parallel
+  if (l) call abort
+  if (omp_test_nest_lock (lock) .ne. 3) call abort
+  call omp_unset_nest_lock (lock)
+  call omp_unset_nest_lock (lock)
+  call omp_unset_nest_lock (lock)
+  call omp_destroy_nest_lock (lock)
+end
diff --git a/libgomp/testsuite/libgomp.fortran/lock-2.f90 b/libgomp/testsuite/libgomp.fortran/lock-2.f90
new file mode 100644
index 00000000000..9965139b9ba
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/lock-2.f90
@@ -0,0 +1,24 @@
+! { dg-do run }
+
+  use omp_lib
+
+  integer (kind = omp_nest_lock_kind) :: lock
+  logical :: l
+
+  l = .false.
+  call omp_init_nest_lock (lock)
+!$omp parallel num_threads (1) reduction (.or.:l)
+  if (omp_test_nest_lock (lock) .ne. 1) call abort
+  if (omp_test_nest_lock (lock) .ne. 2) call abort
+!$omp task if (.false.) shared (lock, l)
+  if (omp_test_nest_lock (lock) .ne. 0) l = .true.
+!$omp end task
+!$omp taskwait
+  if (omp_test_nest_lock (lock) .ne. 3) l = .true.
+  call omp_unset_nest_lock (lock)
+  call omp_unset_nest_lock (lock)
+  call omp_unset_nest_lock (lock)
+!$omp end parallel
+  if (l) call abort
+  call omp_destroy_nest_lock (lock)
+end
diff --git a/libgomp/testsuite/libgomp.fortran/nested1.f90 b/libgomp/testsuite/libgomp.fortran/nested1.f90
new file mode 100644
index 00000000000..98c4322d0bf
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/nested1.f90
@@ -0,0 +1,87 @@
+! { dg-do run }
+program nested1
+  use omp_lib
+  integer :: e1, e2, e3, e
+  integer :: tn1, tn2, tn3
+  e1 = 0
+  e2 = 0
+  e3 = 0
+  call omp_set_nested (.true.)
+  call omp_set_dynamic (.false.)
+  if (omp_in_parallel ()) call abort
+  if (omp_get_num_threads ().ne.1) call abort
+  if (omp_get_level ().ne.0) call abort
+  if (omp_get_ancestor_thread_num (0).ne.0) call abort
+  if (omp_get_ancestor_thread_num (-1).ne.-1) call abort
+  if (omp_get_ancestor_thread_num (1).ne.-1) call abort
+  if (omp_get_team_size (0).ne.1) call abort
+  if (omp_get_team_size (-1).ne.-1) call abort
+  if (omp_get_team_size (1).ne.-1) call abort
+  if (omp_get_active_level ().ne.0) call abort
+!$omp parallel num_threads (4) private (e, tn1)
+  e = 0
+  tn1 = omp_get_thread_num ()
+  if (.not.omp_in_parallel ()) e = e + 1
+  if (omp_get_num_threads ().ne.4) e = e + 1
+  if (tn1.lt.0.or.tn1.ge.4) e = e + 1
+  if (omp_get_level ().ne.1) e = e + 1
+  if (omp_get_ancestor_thread_num (0).ne.0) e = e + 1
+  if (omp_get_ancestor_thread_num (1).ne.tn1) e = e + 1
+  if (omp_get_ancestor_thread_num (-1).ne.-1) e = e + 1
+  if (omp_get_ancestor_thread_num (2).ne.-1) e = e + 1
+  if (omp_get_team_size (0).ne.1) e = e + 1
+  if (omp_get_team_size (1).ne.4) e = e + 1
+  if (omp_get_team_size (-1).ne.-1) e = e + 1
+  if (omp_get_team_size (2).ne.-1) e = e + 1
+  if (omp_get_active_level ().ne.1) e = e + 1
+  !$omp atomic
+    e1 = e1 + e
+!$omp parallel num_threads (5) if (.false.) firstprivate (tn1) &
+!$omp& private (e, tn2)
+  e = 0
+  tn2 = omp_get_thread_num ()
+  if (.not.omp_in_parallel ()) e = e + 1
+  if (omp_get_num_threads ().ne.1) e = e + 1
+  if (tn2.ne.0) e = e + 1
+  if (omp_get_level ().ne.2) e = e + 1
+  if (omp_get_ancestor_thread_num (0).ne.0) e = e + 1
+  if (omp_get_ancestor_thread_num (1).ne.tn1) e = e + 1
+  if (omp_get_ancestor_thread_num (2).ne.tn2) e = e + 1
+  if (omp_get_ancestor_thread_num (-1).ne.-1) e = e + 1
+  if (omp_get_ancestor_thread_num (3).ne.-1) e = e + 1
+  if (omp_get_team_size (0).ne.1) e = e + 1
+  if (omp_get_team_size (1).ne.4) e = e + 1
+  if (omp_get_team_size (2).ne.1) e = e + 1
+  if (omp_get_team_size (-1).ne.-1) e = e + 1
+  if (omp_get_team_size (3).ne.-1) e = e + 1
+  if (omp_get_active_level ().ne.1) e = e + 1
+  !$omp atomic
+    e2 = e2 + e
+!$omp parallel num_threads (2) firstprivate (tn1, tn2) &
+!$omp& private (e, tn3)
+  e = 0
+  tn3 = omp_get_thread_num ()
+  if (.not.omp_in_parallel ()) e = e + 1
+  if (omp_get_num_threads ().ne.2) e = e + 1
+  if (tn3.lt.0.or.tn3.ge.2) e = e + 1
+  if (omp_get_level ().ne.3) e = e + 1
+  if (omp_get_ancestor_thread_num (0).ne.0) e = e + 1
+  if (omp_get_ancestor_thread_num (1).ne.tn1) e = e + 1
+  if (omp_get_ancestor_thread_num (2).ne.tn2) e = e + 1
+  if (omp_get_ancestor_thread_num (3).ne.tn3) e = e + 1
+  if (omp_get_ancestor_thread_num (-1).ne.-1) e = e + 1
+  if (omp_get_ancestor_thread_num (4).ne.-1) e = e + 1
+  if (omp_get_team_size (0).ne.1) e = e + 1
+  if (omp_get_team_size (1).ne.4) e = e + 1
+  if (omp_get_team_size (2).ne.1) e = e + 1
+  if (omp_get_team_size (3).ne.2) e = e + 1
+  if (omp_get_team_size (-1).ne.-1) e = e + 1
+  if (omp_get_team_size (4).ne.-1) e = e + 1
+  if (omp_get_active_level ().ne.2) e = e + 1
+  !$omp atomic
+    e3 = e3 + e
+!$omp end parallel
+!$omp end parallel
+!$omp end parallel
+  if (e1.ne.0.or.e2.ne.0.or.e3.ne.0) call abort
+end program nested1
diff --git a/libgomp/testsuite/libgomp.fortran/nestedfn4.f90 b/libgomp/testsuite/libgomp.fortran/nestedfn4.f90
new file mode 100644
index 00000000000..c987bf440b0
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/nestedfn4.f90
@@ -0,0 +1,41 @@
+program foo
+  integer :: i, j, k
+  integer :: a(10), c(10)
+  k = 2
+  a(:) = 0
+  call test1
+  call test2
+  do i = 1, 10
+    if (a(i) .ne. 10 * i) call abort
+  end do
+  !$omp parallel do reduction (+:c)
+  do i = 1, 10
+    c = c + a
+  end do
+  do i = 1, 10
+    if (c(i) .ne. 10 * a(i)) call abort
+  end do
+  !$omp parallel do lastprivate (j)
+  do j = 1, 10, k
+  end do
+  if (j .ne. 11) call abort
+contains
+  subroutine test1
+    integer :: i
+    integer :: b(10)
+    do i = 1, 10
+      b(i) = i
+    end do
+    c(:) = 0
+    !$omp parallel do reduction (+:a)
+    do i = 1, 10
+      a = a + b
+    end do
+  end subroutine test1
+  subroutine test2
+    !$omp parallel do lastprivate (j)
+    do j = 1, 10, k
+    end do
+    if (j .ne. 11) call abort
+  end subroutine test2
+end program foo
diff --git a/libgomp/testsuite/libgomp.fortran/strassen.f90 b/libgomp/testsuite/libgomp.fortran/strassen.f90
new file mode 100644
index 00000000000..b44982665a6
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/strassen.f90
@@ -0,0 +1,75 @@
+! { dg-options "-O2" }
+
+program strassen_matmul
+  use omp_lib
+  integer, parameter :: N = 1024
+  double precision, save :: A(N,N), B(N,N), C(N,N), D(N,N)
+  double precision :: start, end
+
+  call random_seed
+  call random_number (A)
+  call random_number (B)
+  start = omp_get_wtime ()
+  C = matmul (A, B)
+  end = omp_get_wtime ()
+  write(*,'(a, f10.6)') ' Time for matmul      = ', end - start
+  D = 0
+  start = omp_get_wtime ()
+  call strassen (A, B, D, N)
+  end = omp_get_wtime ()
+  write(*,'(a, f10.6)') ' Time for Strassen    = ', end - start
+  if (sqrt (sum ((C - D) ** 2)) / N .gt. 0.1) call abort
+  D = 0
+  start = omp_get_wtime ()
+!$omp parallel
+!$omp single
+  call strassen (A, B, D, N)
+!$omp end single nowait
+!$omp end parallel
+  end = omp_get_wtime ()
+  write(*,'(a, f10.6)') ' Time for Strassen MP = ', end - start
+  if (sqrt (sum ((C - D) ** 2)) / N .gt. 0.1) call abort
+
+contains
+
+  recursive subroutine strassen (A, B, C, N)
+    integer, intent(in) :: N
+    double precision, intent(in) :: A(N,N), B(N,N)
+    double precision, intent(out) :: C(N,N)
+    double precision :: T(N/2,N/2,7)
+    integer :: K, L
+
+    if (iand (N,1) .ne. 0 .or. N < 64) then
+      C = matmul (A, B)
+      return
+    end if
+    K = N / 2
+    L = N / 2 + 1
+!$omp task shared (A, B, T)
+    call strassen (A(:K,:K) + A(L:,L:), B(:K,:K) + B(L:,L:), T(:,:,1), K)
+!$omp end task
+!$omp task shared (A, B, T)
+    call strassen (A(L:,:K) + A(L:,L:), B(:K,:K), T(:,:,2), K)
+!$omp end task
+!$omp task shared (A, B, T)
+    call strassen (A(:K,:K), B(:K,L:) - B(L:,L:), T(:,:,3), K)
+!$omp end task
+!$omp task shared (A, B, T)
+    call strassen (A(L:,L:), B(L:,:K) - B(:K,:K), T(:,:,4), K)
+!$omp end task
+!$omp task shared (A, B, T)
+    call strassen (A(:K,:K) + A(:K,L:), B(L:,L:), T(:,:,5), K)
+!$omp end task
+!$omp task shared (A, B, T)
+    call strassen (A(L:,:K) - A(:K,:K), B(:K,:K) + B(:K,L:), T(:,:,6), K)
+!$omp end task
+!$omp task shared (A, B, T)
+    call strassen (A(:K,L:) - A(L:,L:), B(L:,:K) + B(L:,L:), T(:,:,7), K)
+!$omp end task
+!$omp taskwait
+    C(:K,:K) = T(:,:,1) + T(:,:,4) - T(:,:,5) + T(:,:,7)
+    C(L:,:K) = T(:,:,2) + T(:,:,4)
+    C(:K,L:) = T(:,:,3) + T(:,:,5)
+    C(L:,L:) = T(:,:,1) - T(:,:,2) + T(:,:,3) + T(:,:,6)
+  end subroutine strassen
+end
diff --git a/libgomp/testsuite/libgomp.fortran/tabs1.f90 b/libgomp/testsuite/libgomp.fortran/tabs1.f90
new file mode 100644
index 00000000000..4f3d4f5b435
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/tabs1.f90
@@ -0,0 +1,12 @@
+	if (b().ne.2) call abort
+contains
+subroutine a
+!$omp parallel
+	!$omp	end	parallel
+	end subroutine a
+function b()
+	integer :: b
+	b = 1
+	!$	b = 2
+end function b
+	end
diff --git a/libgomp/testsuite/libgomp.fortran/tabs2.f b/libgomp/testsuite/libgomp.fortran/tabs2.f
new file mode 100644
index 00000000000..7aed5498d34
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/tabs2.f
@@ -0,0 +1,13 @@
+! { dg-options "-ffixed-form" }
+      if (b().ne.2) call abort
+      contains
+      subroutine a
+!$omp parallel
+!$omp	end	parallel
+	end subroutine a
+      function b()
+      integer :: b
+	b = 1
+!$	b = 2
+      end function b
+      end
diff --git a/libgomp/testsuite/libgomp.fortran/task1.f90 b/libgomp/testsuite/libgomp.fortran/task1.f90
new file mode 100644
index 00000000000..df57cb83168
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/task1.f90
@@ -0,0 +1,27 @@
+! { dg-do run }
+
+program tasktest
+  use omp_lib
+  integer :: i, j
+  common /tasktest_j/ j
+  j = 0
+  !$omp parallel private (i)
+    i = omp_get_thread_num ()
+    if (i.lt.2) then
+      !$omp task if (.false.) default(firstprivate)
+        call subr (i + 1)
+      !$omp end task
+    end if
+  !$omp end parallel
+  if (j.gt.0) call abort
+contains
+  subroutine subr (i)
+    use omp_lib
+    integer :: i, j
+    common /tasktest_j/ j
+    if (omp_get_thread_num ().ne.(i - 1)) then
+    !$omp atomic
+      j = j + 1
+    end if
+  end subroutine subr
+end program tasktest
diff --git a/libgomp/testsuite/libgomp.fortran/task2.f90 b/libgomp/testsuite/libgomp.fortran/task2.f90
new file mode 100644
index 00000000000..24ffee53ac8
--- /dev/null
+++ b/libgomp/testsuite/libgomp.fortran/task2.f90
@@ -0,0 +1,142 @@
+  integer :: err
+  err = 0
+!$omp parallel num_threads (4) default (none) shared (err)
+!$omp single
+  call test
+!$omp end single
+!$omp end parallel
+  if (err.ne.0) call abort
+contains
+  subroutine check (x, y, l)
+    integer :: x, y
+    logical :: l
+    l = l .or. x .ne. y
+  end subroutine check
+
+  subroutine foo (c, d, e, f, g, h, i, j, k, n)
+    use omp_lib
+    integer :: n
+    character (len = *) :: c
+    character (len = n) :: d
+    integer, dimension (2, 3:5, n) :: e
+    integer, dimension (2, 3:n, n) :: f
+    character (len = *), dimension (5, 3:n) :: g
+    character (len = n), dimension (5, 3:n) :: h
+    real, dimension (:, :, :) :: i
+    double precision, dimension (3:, 5:, 7:) :: j
+    integer, dimension (:, :, :) :: k
+    logical :: l
+    integer :: p, q, r
+    character (len = n) :: s
+    integer, dimension (2, 3:5, n) :: t
+    integer, dimension (2, 3:n, n) :: u
+    character (len = n), dimension (5, 3:n) :: v
+    character (len = 2 * n + 24) :: w
+    integer :: x, z
+    character (len = 1) :: y
+    s = 'PQRSTUV'
+    forall (p = 1:2, q = 3:5, r = 1:7) t(p, q, r) = -10 + p - q + 2 * r
+    forall (p = 1:2, q = 3:7, r = 1:7) u(p, q, r) = 30 - p + q - 2 * r
+    forall (p = 1:5, q = 3:7, p + q .le. 8) v(p, q) = '_+|/Oo_'
+    forall (p = 1:5, q = 3:7, p + q .gt. 8) v(p, q) = '///|||!'
+!$omp task default (none) firstprivate (c, d, e, f, g, h, i, j, k) &
+!$omp & firstprivate (s, t, u, v) private (l, p, q, r, w, x, y) shared (err)
+    l = .false.
+    l = l .or. c .ne. 'abcdefghijkl'
+    l = l .or. d .ne. 'ABCDEFG'
+    l = l .or. s .ne. 'PQRSTUV'
+    do 100, p = 1, 2
+      do 100, q = 3, 7
+	do 100, r = 1, 7
+	  if (q .lt. 6) l = l .or. e(p, q, r) .ne. 5 + p + q + 2 * r
+	  l = l .or. f(p, q, r) .ne. 25 + p + q + 2 * r
+	  if (r .lt. 6 .and. q + r .le. 8) l = l .or. g(r, q) .ne. '0123456789AB'
+	  if (r .lt. 6 .and. q + r .gt. 8) l = l .or. g(r, q) .ne. '9876543210ZY'
+	  if (r .lt. 6 .and. q + r .le. 8) l = l .or. h(r, q) .ne. '0123456'
+	  if (r .lt. 6 .and. q + r .gt. 8) l = l .or. h(r, q) .ne. '9876543'
+	  if (q .lt. 6) l = l .or. t(p, q, r) .ne. -10 + p - q + 2 * r
+	  l = l .or. u(p, q, r) .ne. 30 - p + q - 2 * r
+	  if (r .lt. 6 .and. q + r .le. 8) l = l .or. v(r, q) .ne. '_+|/Oo_'
+	  if (r .lt. 6 .and. q + r .gt. 8) l = l .or. v(r, q) .ne. '///|||!'
+100 continue
+    do 101, p = 3, 5
+      do 101, q = 2, 6
+	do 101, r = 1, 7
+	  l = l .or. i(p - 2, q - 1, r) .ne. 7.5 * p * q * r
+	  l = l .or. j(p, q + 3, r + 6) .ne. 9.5 * p * q * r
+101 continue
+    do 102, p = 1, 5
+      do 102, q = 4, 6
+	l = l .or. k(p, 1, q - 3) .ne. 19 + p + 7 + 3 * q
+102 continue
+    call check (size (e, 1), 2, l)
+    call check (size (e, 2), 3, l)
+    call check (size (e, 3), 7, l)
+    call check (size (e), 42, l)
+    call check (size (f, 1), 2, l)
+    call check (size (f, 2), 5, l)
+    call check (size (f, 3), 7, l)
+    call check (size (f), 70, l)
+    call check (size (g, 1), 5, l)
+    call check (size (g, 2), 5, l)
+    call check (size (g), 25, l)
+    call check (size (h, 1), 5, l)
+    call check (size (h, 2), 5, l)
+    call check (size (h), 25, l)
+    call check (size (i, 1), 3, l)
+    call check (size (i, 2), 5, l)
+    call check (size (i, 3), 7, l)
+    call check (size (i), 105, l)
+    call check (size (j, 1), 4, l)
+    call check (size (j, 2), 5, l)
+    call check (size (j, 3), 7, l)
+    call check (size (j), 140, l)
+    call check (size (k, 1), 5, l)
+    call check (size (k, 2), 1, l)
+    call check (size (k, 3), 3, l)
+    call check (size (k), 15, l)
+    if (l) then
+!$omp atomic
+      err = err + 1
+    end if
+!$omp end task
+  c = ''
+  d = ''
+  e(:, :, :) = 199
+  f(:, :, :) = 198
+  g(:, :) = ''
+  h(:, :) = ''
+  i(:, :, :) = 7.0
+  j(:, :, :) = 8.0
+  k(:, :, :) = 9
+  s = ''
+  t(:, :, :) = 10
+  u(:, :, :) = 11
+  v(:, :) = ''
+  end subroutine foo
+
+  subroutine test
+    character (len = 12) :: c
+    character (len = 7) :: d
+    integer, dimension (2, 3:5, 7) :: e
+    integer, dimension (2, 3:7, 7) :: f
+    character (len = 12), dimension (5, 3:7) :: g
+    character (len = 7), dimension (5, 3:7) :: h
+    real, dimension (3:5, 2:6, 1:7) :: i
+    double precision, dimension (3:6, 2:6, 1:7) :: j
+    integer, dimension (1:5, 7:7, 4:6) :: k
+    integer :: p, q, r
+    c = 'abcdefghijkl'
+    d = 'ABCDEFG'
+    forall (p = 1:2, q = 3:5, r = 1:7) e(p, q, r) = 5 + p + q + 2 * r
+    forall (p = 1:2, q = 3:7, r = 1:7) f(p, q, r) = 25 + p + q + 2 * r
+    forall (p = 1:5, q = 3:7, p + q .le. 8) g(p, q) = '0123456789AB'
+    forall (p = 1:5, q = 3:7, p + q .gt. 8) g(p, q) = '9876543210ZY'
+    forall (p = 1:5, q = 3:7, p + q .le. 8) h(p, q) = '0123456'
+    forall (p = 1:5, q = 3:7, p + q .gt. 8) h(p, q) = '9876543'
+    forall (p = 3:5, q = 2:6, r = 1:7) i(p, q, r) = 7.5 * p * q * r
+    forall (p = 3:6, q = 2:6, r = 1:7) j(p, q, r) = 9.5 * p * q * r
+    forall (p = 1:5, q = 7:7, r = 4:6) k(p, q, r) = 19 + p + q + 3 * r
+    call foo (c, d, e, f, g, h, i, j, k, 7)
+  end subroutine test
+end
diff --git a/libgomp/testsuite/libgomp.fortran/vla4.f90 b/libgomp/testsuite/libgomp.fortran/vla4.f90
index 58caabc6248..cdd4849b6ad 100644
--- a/libgomp/testsuite/libgomp.fortran/vla4.f90
+++ b/libgomp/testsuite/libgomp.fortran/vla4.f90
@@ -94,7 +94,7 @@ contains
     forall (p = 1:2, q = 3:7, r = 1:7) u(p, q, r) = 30 - x - p + q - 2 * r
     forall (p = 1:5, q = 3:7, p + q .le. 8) v(p, q) = w(1:7)
     forall (p = 1:5, q = 3:7, p + q .gt. 8) v(p, q) = w(20:26)
-!$omp barrier
+!$omp barrier		! { dg-warning "may not be closely nested" }
     y = ''
     if (x .eq. 0) y = '0'
     if (x .eq. 1) y = '1'
diff --git a/libgomp/testsuite/libgomp.fortran/vla5.f90 b/libgomp/testsuite/libgomp.fortran/vla5.f90
index 5c889f9923a..9b611505219 100644
--- a/libgomp/testsuite/libgomp.fortran/vla5.f90
+++ b/libgomp/testsuite/libgomp.fortran/vla5.f90
@@ -66,7 +66,7 @@ contains
     forall (p = 1:2, q = 3:7, r = 1:7) u(p, q, r) = 30 - x - p + q - 2 * r
     forall (p = 1:5, q = 3:7, p + q .le. 8) v(p, q) = w(1:7)
     forall (p = 1:5, q = 3:7, p + q .gt. 8) v(p, q) = w(20:26)
-!$omp barrier
+!$omp barrier		! { dg-warning "may not be closely nested" }
     y = ''
     if (x .eq. 0) y = '0'
     if (x .eq. 1) y = '1'