gcc/testsuite/
* gcc.c-torture/execute/builtins/memops-asm-lib.c: Do the copy or
set before checking whether it was allowed.
* gcc.c-torture/execute/builtins/lib/memset.c: Do the memset
before checking inside_main.
From-SVN: r121746
* calls.c (store_one_arg): Pass correct alignment to
emit_push_insn for non-BLKmode values.
* expr.c (emit_push_insn): If STRICT_ALIGNMENT, copy to an
unaligned stack slot via a suitably aligned slot.
From-SVN: r121736
2007-02-08 Paul Brook <paul@codesourcery.com>
* config/arm/lib1funcs.asm (RETLDM): Pop directly into PC when no
special interworking needed.
From-SVN: r121727
* config/i386/xmmintrin.h: Make inclusion of emmintrin.h
conditional to __SSE2__.
(Entries below should have been added to first ChangeLog
entry for amdfam10 dated 2007-02-05)
* config/i386/emmintrin.h: Generate #error if __SSE2__ is not
defined.
* config/i386/pmmintrin.h: Generate #error if __SSE3__ is not
defined.
* config/i386/tmmintrin.h: Generate #error if __SSSE3__ is not
defined.
From-SVN: r121726
* config/m32c/m32c-protos.h (m32c_illegal_subreg_p): New.
* config/m32c/m32c.c (legal_subregs): New.
(m32c_illegal_subreg_p): New.
* config/m32c/predicates.md (m32c_any_operand): Use it to reject
unsupported subregs of hard regs.
From-SVN: r121725
* simplify-rtx.c (simplify_unary_operation_1) <POPCOUNT>: We can
strip zero_extend, bswap and rotates from POCOUNT's argument.
<PARITY>: Likewise, we can strip not, bswap, sign_extend,
zero_extend and rotates from PARITY's argument.
<BSWAP>: A byte-swap followed by a byte-swap is an identity.
(simplify_const_unary_operation) <BSWAP>: Evaluate the byte-swap
of an integer constant at compile-time.
* gcc.target/i386/builtin-bswap-2.c: New test case.
From-SVN: r121716
PR 30562
* tree-flow.h (struct var_ann_d): Remove field 'is_used'.
Update all users.
* tree-ssa-alias.c (compute_is_aliased): Remove. Update all
users.
(init_alias_info):
* tree-ssa-live.c (remove_unused_locals): Do not remove
TREE_ADDRESSABLE variables.
* tree-ssa-structalias.c (compute_points_to_sets): Tidy.
* tree-ssa-operands.c (add_virtual_operand): Remove argument
FOR_CLOBBER. Update all users.
If VAR has an associated alias set, add a virtual operand for
it if no alias is found to conflict with the memory reference.
From-SVN: r121715
* trans-stmt.c (gfc_trans_forall_1): Optimize the cases where the
mask expression is a compile-time constant (".true." or ".false.").
* gfortran.dg/forall_8.f90: New test case.
* gfortran.dg/forall_9.f90: Likewise.
From-SVN: r121714
PR c++/30703
* gimplify.c (gimplify_scan_omp_clauses): Remove special casing
of INDIRECT_REF <RESULT_DECL>.
* cp-gimplify.c (cp_genericize_r): Don't dereference invisiref
parameters and result decls in omp clauses.
(cxx_omp_privatize_by_reference): Pass also invisiref PARM_DECLs
by reference.
* testsuite/libgomp.c++/pr30703.C: New test.
From-SVN: r121688
2007-02-07 Andreas Krebbel <krebbel1@de.ibm.com>
* raise-gcc.c (get_region_description_for, get_call_site_action_for,
get_action_description_for): Replaced _Unwind_Word with _uleb128_t
and _Unwind_SWord with _sleb128_t.
From-SVN: r121681
* doc/loop.texi: Document possibility not to perform disambiguation
of loops with multiple latches.
* cfgloopmanip.c (alp_enum_p): Removed.
(add_loop): Handle subloops. Use get_loop_body_with_size.
(create_preheader): Do not allow ENTRY_BLOCK_PTR to be preheader.
* cfghooks.c (redirect_edge_and_branch_force): Set dominator for
the new forwarder block.
(make_forwarder_block): Only call new_bb_cbk if it is not NULL.
Handle the case latch is NULL.
* tree-ssa-dom.c (tree_ssa_dominator_optimize): Avoid cfg modifications
when marking loop exits.
* ifcvt.c (if_convert): Ditto. Mark loop exits even if cfg cannot
be modified.
* loop-init.c (loop_optimizer_init): Do not modify cfg. Call
disambiguate_loops_with_multiple_latches.
* tree-cfgcleanup.c (cleanup_tree_cfg_loop): Calculate dominators
before fix_loop_structure.
* cfgloop.c: Include pointer-set.h and output.h.
(canonicalize_loop_headers, HEADER_BLOCK, LATCH_EDGE,
update_latch_info, mfb_keep_just, mfb_keep_nonlatch): Removed.
(get_loop_latch_edges, find_subloop_latch_edge_by_profile,
find_subloop_latch_edge_by_ivs, find_subloop_latch_edge,
mfb_redirect_edges_in_set, form_subloop, merge_latch_edges,
disambiguate_multiple_latches, get_loop_body_with_size,
disambiguate_loops_with_multiple_latches): New functions.
(flow_loop_dump): Dump multiple latch edges.
(flow_loop_nodes_find): Handle loops with multiple latches.
(flow_loops_find): Ditto. Do not call canonicalize_loop_headers.
(glb_enum_p): Modified.
(get_loop_body): Use get_loop_body_with_size.
* cfgloop.h (LOOPS_HAVE_RECORDED_EXITS): New flag.
(AVOID_CFG_MODIFICATIONS): New constant.
(disambiguate_loops_with_multiple_latches, add_loop,
get_loop_body_with_size): Declare.
* Makefile.in (cfgloop.o): Add pointer-set.h and output.h.
* gcc.dg/tree-ssa/loop-25.c: New test.
From-SVN: r121670
2006-02-06 Paolo Bonzini <bonzini@gnu.org>
* Makefile.in (tree-ssa-loop-ivopts.o): Add pointer-set.h dependency.
(tree-ssa-reassoc.o): Add pointer-set.h dependency.
(tree-cfg.o): Remove hashtab.h dependency.
* tree-ssa-loop-ivopts.c: Include pointer-set.h.
(struct ivopts_data): Change niters to pointer_map_t.
(struct nfe_cache_elt, nfe_hash, nfe_eq): Delete.
(niter_for_exit): Create pointer_map on demand. Change for
pointer_map API.
(tree_ssa_iv_optimize_init): Initialize data->niters to NULL.
(free_loop_data): Destroy data->niters if created and reset field.
(tree_ssa_iv_optimize_finalize): Don't delete data->niters here.
(tree_ssa_iv_optimize_loop): Check for presence of stale data.
* tree-ssa-reassoc.c: Include pointer-set.h.
(bb_rank): Change to long *.
(operand_rank): Change to pointer_map_t.
(find_operand_rank): Return long, -1 if not found. Declare as inline.
(insert_operand_rank): Accept long.
(operand_entry_hash, operand_entry_eq): Remove.
(get_rank): Return long. Adjust for changes above.
(init_reassoc): Change rank type to long. Adjust creation of bb_rank
and operand_rank.
(fini_reassoc): Delete operand_rank with pointer_map_destroy.
* tree-ssa-structalias.c (vi_for_tree): Change to pointer_map.
(struct tree_vi, tree_vi_t, tree_vi_hash, tree_vi_eq): Delete.
(insert_vi_for_tree): Rewrite for pointer_map API. Assert argument
is not NULL.
(lookup_vi_for_tree): Rewrite for pointer_map API. Return varinfo_t
directly since it cannot be NULL.
(get_vi_for_tree): Rewrite for pointer_map API.
(find_what_p_points_to): Adjust for change to lookup_vi_for_tree.
(init_alias_vars): Create vi_for_tree as pointer_map.
(delete_points_to_sets): Delete vi_for_tree using pointer_map_destroy.
* tree-cfg.c: Don't include hashtab.h.
(edge_to_cases): Declare as pointer_map.
(struct edge_to_cases_elt, edge_to_cases_hash, edge_to_cases_eq):
Delete.
(edge_to_cases_cleanup): Rewrite as pointer_map_traverse callback.
(start_recording_case_labels): Create edge_to_cases as pointer_map.
(end_recoding_case_labels): Cleanup edge_to_cases manually before
destroying it.
(record_switch_edge): Delete.
(get_cases_for_edge): Adjust for pointer_map API, inline
record_switch_edge (rewritten for new API), remove goto.
From-SVN: r121648
2007-02-06 Dorit Nuzman <dorit@il.ibm.com>
Victor Kaplansky <victork@il.ibm.com>
* tree-vectorizer.c (vect_is_simple_use): Support induction.
(vect_is_simple_reduction): Support reduction with induction as
one of the operands.
(vect_is_simple_iv_evolution): Fix formatting.
* tree-vect-analyze.c (vect_mark_stmts_to_be_vectorized): Fix
formatting. Don't mark induction phis for vectorization.
(vect_analyze_scalar_cycles): Analyze all inductions, then reductions.
* tree-vect-transform.c (get_initial_def_for_induction): New function.
(vect_get_vec_def_for_operand): Support induction.
(vect_get_vec_def_for_stmt_copy): Fix formatting and add check for
induction case.
(vectorizable_reduction): Support reduction with induction as one of
the operands.
(vectorizable_type_demotion): Use def-type of stmt argument rather
than dummy def-type.
* tree-ssa-loop.c (gate_scev_const_prop): Return the value of
flag_tree_scev_cprop.
* common.opt (tree-scev-cprop): New flag.
* tree-vect-transform.c (vect_create_destination_var): Use 'kind' in
call to vect_get_new_vect_var.
Co-Authored-By: Victor Kaplansky <victork@il.ibm.com>
From-SVN: r121643
