diff options
author | Jan Hubicka <hubicka@ucw.cz> | 2017-11-03 16:42:30 +0100 |
---|---|---|
committer | Jan Hubicka <hubicka@gcc.gnu.org> | 2017-11-03 15:42:30 +0000 |
commit | e7a740068ed3cc5961101f07012314d940a97ae5 (patch) | |
tree | 60ff28d76be31e01dc2ff68a6aba99743046eba5 /gcc/predict.c | |
parent | 13494fcb363e8a901db7768a851a9eed1dea62e3 (diff) | |
download | gcc-e7a740068ed3cc5961101f07012314d940a97ae5.zip gcc-e7a740068ed3cc5961101f07012314d940a97ae5.tar.gz gcc-e7a740068ed3cc5961101f07012314d940a97ae5.tar.bz2 |
asan.c (create_cond_insert_point): Maintain profile.
* asan.c (create_cond_insert_point): Maintain profile.
* ipa-utils.c (ipa_merge_profiles): Be sure only IPA profiles are
merged.
* basic-block.h (struct basic_block_def): Remove frequency.
(EDGE_FREQUENCY): Use to_frequency
* bb-reorder.c (push_to_next_round_p): Use only IPA counts for global
heuristics.
(find_traces): Update to use to_frequency.
(find_traces_1_round): Likewise; use only IPA counts.
(bb_to_key): Likewise.
(connect_traces): Use IPA counts only.
(copy_bb_p): Update to use to_frequency.
(fix_up_crossing_landing_pad): Likewise.
(sanitize_hot_paths): Likewise.
* bt-load.c (basic_block_freq): Likewise.
* cfg.c (init_flow): Set count_max to uninitialized.
(check_bb_profile): Remove frequencies; check counts.
(dump_bb_info): Do not dump frequencies.
(update_bb_profile_for_threading): Update counts only.
(scale_bbs_frequencies_int): Likewise.
(MAX_SAFE_MULTIPLIER): Remove.
(scale_bbs_frequencies_gcov_type): Update counts only.
(scale_bbs_frequencies_profile_count): Update counts only.
(scale_bbs_frequencies): Update counts only.
* cfg.h (struct control_flow_graph): Add count-max.
(update_bb_profile_for_threading): Update prototype.
* cfgbuild.c (find_bb_boundaries): Do not update frequencies.
(find_many_sub_basic_blocks): Likewise.
* cfgcleanup.c (try_forward_edges): Likewise.
(try_crossjump_to_edge): Likewise.
* cfgexpand.c (expand_gimple_cond): Likewise.
(expand_gimple_tailcall): Likewise.
(construct_init_block): Likewise.
(construct_exit_block): Likewise.
* cfghooks.c (verify_flow_info): Check consistency of counts.
(dump_bb_for_graph): Do not dump frequencies.
(split_block_1): Do not update frequencies.
(split_edge): Do not update frequencies.
(make_forwarder_block): Do not update frequencies.
(duplicate_block): Do not update frequencies.
(account_profile_record): Do not update frequencies.
* cfgloop.c (find_subloop_latch_edge_by_profile): Use IPA counts
for global heuristics.
* cfgloopanal.c (average_num_loop_insns): Update to use to_frequency.
(expected_loop_iterations_unbounded): Use counts only.
* cfgloopmanip.c (scale_loop_profile): Simplify.
(create_empty_loop_on_edge): Simplify
(loopify): Simplify
(duplicate_loop_to_header_edge): Simplify
* cfgrtl.c (force_nonfallthru_and_redirect): Update profile.
(update_br_prob_note): Take care of removing note when profile
becomes undefined.
(relink_block_chain): Do not dump frequency.
(rtl_account_profile_record): Use to_frequency.
* cgraph.c (symbol_table::create_edge): Convert count to ipa count.
(cgraph_edge::redirect_call_stmt_to_calle): Conver tcount to ipa count.
(cgraph_update_edges_for_call_stmt_node): Likewise.
(cgraph_edge::verify_count_and_frequency): Update.
(cgraph_node::verify_node): Temporarily disable frequency verification.
* cgraphbuild.c (compute_call_stmt_bb_frequency): Use
to_cgraph_frequency.
(cgraph_edge::rebuild_edges): Convert to ipa counts.
* cgraphunit.c (init_lowered_empty_function): Do not initialize
frequencies.
(cgraph_node::expand_thunk): Update profile.
* except.c (dw2_build_landing_pads): Do not update frequency.
* final.c (compute_alignments): Use to_frequency.
(dump_basic_block_info): Do not dump frequency.
* gimple-pretty-print.c (dump_profile): Do not dump frequency.
(dump_gimple_bb_header): Do not dump frequency.
* gimple-ssa-isolate-paths.c (isolate_path): Do not update frequency;
do update count.
* gimple-streamer-in.c (input_bb): Do not stream frequency.
* gimple-streamer-out.c (output_bb): Do not stream frequency.
* haifa-sched.c (sched_pressure_start_bb): Use to_freuqency.
(init_before_recovery): Do not update frequency.
(sched_create_recovery_edges): Do not update frequency.
* hsa-gen.c (convert_switch_statements): Do not update frequency.
* ipa-cp.c (ipcp_propagate_stage): Update search for max_count.
(ipa_cp_c_finalize): Set max_count to uninitialized.
* ipa-fnsummary.c (get_minimal_bb): Use counts.
(param_change_prob): Use counts.
* ipa-profile.c (ipa_profile_generate_summary): Do not summarize
local profiles.
* ipa-split.c (consider_split): Use to_frequency.
(split_function): Use to_frequency.
* ira-build.c (loop_compare_func): Likewise.
(mark_loops_for_removal): Likewise.
(mark_all_loops_for_removal): Likewise.
* loop-doloop.c (doloop_modify): Do not update frequency.
* loop-unroll.c (unroll_loop_runtime_iterations): Do not update
frequency.
* lto-streamer-in.c (input_function): Update count_max.
* omp-expand.c (expand_omp_taskreg): Update count_max.
* omp-simd-clone.c (simd_clone_adjust): Update profile.
* predict.c (maybe_hot_frequency_p): Use to_frequency.
(maybe_hot_count_p): Use ipa counts only.
(maybe_hot_bb_p): Simplify.
(maybe_hot_edge_p): Simplify.
(probably_never_executed): Do not take frequency argument.
(probably_never_executed_bb_p): Do not pass frequency.
(probably_never_executed_edge_p): Likewise.
(combine_predictions_for_bb): Check that profile is nonzero.
(propagate_freq): Do not set frequency.
(drop_profile): Simplify.
(counts_to_freqs): Simplify.
(expensive_function_p): Use to_frequency.
(propagate_unlikely_bbs_forward): Simplify.
(determine_unlikely_bbs): Simplify.
(estimate_bb_frequencies): Add hack to silence graphite issues.
(compute_function_frequency): Use ipa counts.
(pass_profile::execute): Update.
(rebuild_frequencies): Use counts only.
(force_edge_cold): Use counts only.
* profile-count.c (profile_count::dump): Dump new count types.
(profile_count::differs_from_p): Check compatiblity.
(profile_count::to_frequency): New function.
(profile_count::to_cgraph_frequency): New function.
* profile-count.h (struct function): Declare.
(enum profile_quality): Add profile_guessed_local and
profile_guessed_global0.
(class profile_proability): Decrease number of bits to 29;
update from_reg_br_prob_note and to_reg_br_prob_note.
(class profile_count: Update comment; decrease number of bits
to 61. Check compatibility.
(profile_count::compatible_p): New private member function.
(profile_count::ipa_p): New member function.
(profile_count::operator<): Handle global zero correctly.
(profile_count::operator>): Handle global zero correctly.
(profile_count::operator<=): Handle global zero correctly.
(profile_count::operator>=): Handle global zero correctly.
(profile_count::nonzero_p): New member function.
(profile_count::force_nonzero): New member function.
(profile_count::max): New member function.
(profile_count::apply_scale): Handle IPA scalling.
(profile_count::guessed_local): New member function.
(profile_count::global0): New member function.
(profile_count::ipa): New member function.
(profile_count::to_frequency): Declare.
(profile_count::to_cgraph_frequency): Declare.
* profile.c (OVERLAP_BASE): Delete.
(compute_frequency_overlap): Delete.
(compute_branch_probabilities): Do not use compute_frequency_overlap.
* regs.h (REG_FREQ_FROM_BB): Use to_frequency.
* sched-ebb.c (rank): Use counts only.
* shrink-wrap.c (handle_simple_exit): Use counts only.
(try_shrink_wrapping): Use counts only.
(place_prologue_for_one_component): Use counts only.
* tracer.c (find_best_predecessor): Use to_frequency.
(find_trace): Use to_frequency.
(tail_duplicate): Use to_frequency.
* trans-mem.c (expand_transaction): Do not update frequency.
* tree-call-cdce.c: Do not update frequency.
* tree-cfg.c (gimple_find_sub_bbs): Likewise.
(gimple_merge_blocks): Likewise.
(gimple_split_edge): Likewise.
(gimple_duplicate_sese_region): Likewise.
(gimple_duplicate_sese_tail): Likewise.
(move_sese_region_to_fn): Likewise.
(gimple_account_profile_record): Likewise.
(insert_cond_bb): Likewise.
* tree-complex.c (expand_complex_div_wide): Likewise.
* tree-eh.c (lower_resx): Update profile.
* tree-inline.c (copy_bb): Simplify count scaling; do not scale
frequencies.
(initialize_cfun): Do not initialize frequencies
(freqs_to_counts): Delete.
(copy_cfg_body): Ignore count parameter.
(copy_body): Update.
(expand_call_inline): Update count_max.
(optimize_inline_calls): Update count_max.
(tree_function_versioning): Update count_max.
* tree-ssa-coalesce.c (coalesce_cost_bb): Use to_frequency.
* tree-ssa-ifcombine.c (update_profile_after_ifcombine): Do not update
frequency.
* tree-ssa-loop-im.c (execute_sm_if_changed): Use counts only.
* tree-ssa-loop-ivcanon.c (unloop_loops): Do not update freuqency.
(try_peel_loop): Likewise.
* tree-ssa-loop-ivopts.c (get_scaled_computation_cost_at): Use
to_frequency.
* tree-ssa-loop-manip.c (niter_for_unrolled_loop): Pass -1.
(tree_transform_and_unroll_loop): Do not use frequencies
* tree-ssa-loop-niter.c (estimate_numbers_of_iterations):
Use reliable prediction only.
* tree-ssa-loop-unswitch.c (hoist_guard): Do not use frequencies.
* tree-ssa-sink.c (select_best_block): Use to_frequency.
* tree-ssa-tail-merge.c (replace_block_by): Temporarily disable
probability scaling.
* tree-ssa-threadupdate.c (create_block_for_threading): Do
not update frequency
(any_remaining_duplicated_blocks): Likewise.
(update_profile): Likewise.
(estimated_freqs_path): Delete.
(freqs_to_counts_path): Delete.
(clear_counts_path): Delete.
(ssa_fix_duplicate_block_edges): Likewise.
(duplicate_thread_path): Likewise.
* tree-switch-conversion.c (gen_inbound_check): Use counts.
* tree-tailcall.c (decrease_profile): Do not update frequency.
(eliminate_tail_call): Likewise.
* tree-vect-loop-manip.c (vect_do_peeling): Likewise.
* tree-vect-loop.c (scale_profile_for_vect_loop): Likewise.
(optimize_mask_stores): Likewise.
* tree-vect-stmts.c (vectorizable_simd_clone_call): Likewise.
* ubsan.c (ubsan_expand_null_ifn): Update profile.
(ubsan_expand_ptr_ifn): Update profile.
* value-prof.c (gimple_ic): Simplify.
* value-prof.h (gimple_ic): Update prototype.
* ipa-inline-transform.c (inline_transform): Fix scaling conditoins.
* ipa-inline.c (compute_uninlined_call_time): Be sure that
counts are nonzero.
(want_inline_self_recursive_call_p): Likewise.
(resolve_noninline_speculation): Only cummulate defined counts.
(inline_small_functions): Use nonzero_p.
(ipa_inline): Do not access freed node.
Unknown ChangeLog:
2017-11-02 Jan Hubicka <hubicka@ucw.cz>
* testsuite/gcc.dg/no-strict-overflow-3.c (foo): Update magic
value to not clash with frequency.
* testsuite/gcc.dg/strict-overflow-3.c (foo): Likewise.
* testsuite/gcc.dg/tree-ssa/builtin-sprintf-2.c: Update template.
* testsuite/gcc.dg/tree-ssa/dump-2.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-10.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-11.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-12.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-20040816-1.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-20040816-2.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-5.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-8.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-9.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-cd.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-pr56541.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-pr68583.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-pr69489-1.c: Update template.
* testsuite/gcc.dg/tree-ssa/ifc-pr69489-2.c: Update template.
* testsuite/gcc.target/i386/pr61403.c: Update template.
From-SVN: r254379
Diffstat (limited to 'gcc/predict.c')
-rw-r--r-- | gcc/predict.c | 147 |
1 files changed, 81 insertions, 66 deletions
diff --git a/gcc/predict.c b/gcc/predict.c index 0a85d0b..cf42ccb 100644 --- a/gcc/predict.c +++ b/gcc/predict.c @@ -137,12 +137,12 @@ maybe_hot_frequency_p (struct function *fun, int freq) if (profile_status_for_fn (fun) == PROFILE_ABSENT) return true; if (node->frequency == NODE_FREQUENCY_EXECUTED_ONCE - && freq < (ENTRY_BLOCK_PTR_FOR_FN (fun)->frequency * 2 / 3)) + && freq < (ENTRY_BLOCK_PTR_FOR_FN (fun)->count.to_frequency (cfun) * 2 / 3)) return false; if (PARAM_VALUE (HOT_BB_FREQUENCY_FRACTION) == 0) return false; if (freq * PARAM_VALUE (HOT_BB_FREQUENCY_FRACTION) - < ENTRY_BLOCK_PTR_FOR_FN (fun)->frequency) + < ENTRY_BLOCK_PTR_FOR_FN (fun)->count.to_frequency (cfun)) return false; return true; } @@ -175,10 +175,14 @@ set_hot_bb_threshold (gcov_type min) /* Return TRUE if frequency FREQ is considered to be hot. */ bool -maybe_hot_count_p (struct function *, profile_count count) +maybe_hot_count_p (struct function *fun, profile_count count) { if (!count.initialized_p ()) return true; + if (!count.ipa_p ()) + return maybe_hot_frequency_p (fun, count.to_frequency (fun)); + if (count.ipa () == profile_count::zero ()) + return false; /* Code executed at most once is not hot. */ if (count <= MAX (profile_info ? profile_info->runs : 1, 1)) return false; @@ -192,9 +196,7 @@ bool maybe_hot_bb_p (struct function *fun, const_basic_block bb) { gcc_checking_assert (fun); - if (!maybe_hot_count_p (fun, bb->count)) - return false; - return maybe_hot_frequency_p (fun, bb->frequency); + return maybe_hot_count_p (fun, bb->count); } /* Return true in case BB can be CPU intensive and should be optimized @@ -203,9 +205,7 @@ maybe_hot_bb_p (struct function *fun, const_basic_block bb) bool maybe_hot_edge_p (edge e) { - if (!maybe_hot_count_p (cfun, e->count ())) - return false; - return maybe_hot_frequency_p (cfun, EDGE_FREQUENCY (e)); + return maybe_hot_count_p (cfun, e->count ()); } /* Return true if profile COUNT and FREQUENCY, or function FUN static @@ -213,7 +213,7 @@ maybe_hot_edge_p (edge e) static bool probably_never_executed (struct function *fun, - profile_count count, int) + profile_count count) { gcc_checking_assert (fun); if (count == profile_count::zero ()) @@ -238,7 +238,7 @@ probably_never_executed (struct function *fun, bool probably_never_executed_bb_p (struct function *fun, const_basic_block bb) { - return probably_never_executed (fun, bb->count, bb->frequency); + return probably_never_executed (fun, bb->count); } @@ -259,7 +259,7 @@ probably_never_executed_edge_p (struct function *fun, edge e) { if (unlikely_executed_edge_p (e)) return true; - return probably_never_executed (fun, e->count (), EDGE_FREQUENCY (e)); + return probably_never_executed (fun, e->count ()); } /* Return true when current function should always be optimized for size. */ @@ -1289,7 +1289,8 @@ combine_predictions_for_bb (basic_block bb, bool dry_run) } clear_bb_predictions (bb); - if (!bb->count.initialized_p () && !dry_run) + if ((!bb->count.nonzero_p () || !first->probability.initialized_p ()) + && !dry_run) { first->probability = profile_probability::from_reg_br_prob_base (combined_probability); @@ -3014,10 +3015,7 @@ propagate_freq (basic_block head, bitmap tovisit) BLOCK_INFO (bb)->npredecessors = count; /* When function never returns, we will never process exit block. */ if (!count && bb == EXIT_BLOCK_PTR_FOR_FN (cfun)) - { - bb->count = profile_count::zero (); - bb->frequency = 0; - } + bb->count = profile_count::zero (); } BLOCK_INFO (head)->frequency = 1; @@ -3050,7 +3048,10 @@ propagate_freq (basic_block head, bitmap tovisit) * BLOCK_INFO (e->src)->frequency / REG_BR_PROB_BASE); */ - sreal tmp = e->probability.to_reg_br_prob_base (); + /* FIXME: Graphite is producing edges with no profile. Once + this is fixed, drop this. */ + sreal tmp = e->probability.initialized_p () ? + e->probability.to_reg_br_prob_base () : 0; tmp *= BLOCK_INFO (e->src)->frequency; tmp *= real_inv_br_prob_base; frequency += tmp; @@ -3082,7 +3083,10 @@ propagate_freq (basic_block head, bitmap tovisit) = ((e->probability * BLOCK_INFO (bb)->frequency) / REG_BR_PROB_BASE); */ - sreal tmp = e->probability.to_reg_br_prob_base (); + /* FIXME: Graphite is producing edges with no profile. Once + this is fixed, drop this. */ + sreal tmp = e->probability.initialized_p () ? + e->probability.to_reg_br_prob_base () : 0; tmp *= BLOCK_INFO (bb)->frequency; EDGE_INFO (e)->back_edge_prob = tmp * real_inv_br_prob_base; } @@ -3196,10 +3200,26 @@ drop_profile (struct cgraph_node *node, profile_count call_count) } basic_block bb; - FOR_ALL_BB_FN (bb, fn) + push_cfun (DECL_STRUCT_FUNCTION (node->decl)); + if (flag_guess_branch_prob) { - bb->count = profile_count::uninitialized (); + bool clear_zeros + = ENTRY_BLOCK_PTR_FOR_FN + (DECL_STRUCT_FUNCTION (node->decl))->count.nonzero_p (); + FOR_ALL_BB_FN (bb, fn) + if (clear_zeros || !(bb->count == profile_count::zero ())) + bb->count = bb->count.guessed_local (); + DECL_STRUCT_FUNCTION (node->decl)->cfg->count_max = + DECL_STRUCT_FUNCTION (node->decl)->cfg->count_max.guessed_local (); } + else + { + FOR_ALL_BB_FN (bb, fn) + bb->count = profile_count::uninitialized (); + DECL_STRUCT_FUNCTION (node->decl)->cfg->count_max + = profile_count::uninitialized (); + } + pop_cfun (); struct cgraph_edge *e; for (e = node->callees; e; e = e->next_caller) @@ -3300,33 +3320,16 @@ handle_missing_profiles (void) bool counts_to_freqs (void) { - gcov_type count_max; - profile_count true_count_max = profile_count::zero (); + profile_count true_count_max = profile_count::uninitialized (); basic_block bb; - /* Don't overwrite the estimated frequencies when the profile for - the function is missing. We may drop this function PROFILE_GUESSED - later in drop_profile (). */ - if (!ENTRY_BLOCK_PTR_FOR_FN (cfun)->count.initialized_p () - || ENTRY_BLOCK_PTR_FOR_FN (cfun)->count == profile_count::zero ()) - return false; - FOR_BB_BETWEEN (bb, ENTRY_BLOCK_PTR_FOR_FN (cfun), NULL, next_bb) - if (bb->count > true_count_max) - true_count_max = bb->count; - - /* If we have no counts to base frequencies on, keep those that are - already there. */ - if (!(true_count_max > 0)) - return false; + if (!(bb->count < true_count_max)) + true_count_max = true_count_max.max (bb->count); - count_max = true_count_max.to_gcov_type (); - - FOR_ALL_BB_FN (bb, cfun) - if (bb->count.initialized_p ()) - bb->frequency = RDIV (bb->count.to_gcov_type () * BB_FREQ_MAX, count_max); + cfun->cfg->count_max = true_count_max; - return true; + return true_count_max.nonzero_p (); } /* Return true if function is likely to be expensive, so there is no point to @@ -3348,11 +3351,11 @@ expensive_function_p (int threshold) /* Frequencies are out of range. This either means that function contains internal loop executing more than BB_FREQ_MAX times or profile feedback is available and function has not been executed at all. */ - if (ENTRY_BLOCK_PTR_FOR_FN (cfun)->frequency == 0) + if (ENTRY_BLOCK_PTR_FOR_FN (cfun)->count.to_frequency (cfun) == 0) return true; /* Maximally BB_FREQ_MAX^2 so overflow won't happen. */ - limit = ENTRY_BLOCK_PTR_FOR_FN (cfun)->frequency * threshold; + limit = ENTRY_BLOCK_PTR_FOR_FN (cfun)->count.to_frequency (cfun) * threshold; FOR_EACH_BB_FN (bb, cfun) { rtx_insn *insn; @@ -3360,7 +3363,7 @@ expensive_function_p (int threshold) FOR_BB_INSNS (bb, insn) if (active_insn_p (insn)) { - sum += bb->frequency; + sum += bb->count.to_frequency (cfun); if (sum > limit) return true; } @@ -3409,7 +3412,6 @@ propagate_unlikely_bbs_forward (void) "Basic block %i is marked unlikely by forward prop\n", bb->index); bb->count = profile_count::zero (); - bb->frequency = 0; } else bb->aux = NULL; @@ -3440,9 +3442,6 @@ determine_unlikely_bbs () bb->count = profile_count::zero (); } - if (bb->count == profile_count::zero ()) - bb->frequency = 0; - FOR_EACH_EDGE (e, ei, bb->succs) if (!(e->probability == profile_probability::never ()) && unlikely_executed_edge_p (e)) @@ -3497,7 +3496,6 @@ determine_unlikely_bbs () "Basic block %i is marked unlikely by backward prop\n", bb->index); bb->count = profile_count::zero (); - bb->frequency = 0; FOR_EACH_EDGE (e, ei, bb->preds) if (!(e->probability == profile_probability::never ())) { @@ -3554,8 +3552,13 @@ estimate_bb_frequencies (bool force) FOR_EACH_EDGE (e, ei, bb->succs) { - EDGE_INFO (e)->back_edge_prob - = e->probability.to_reg_br_prob_base (); + /* FIXME: Graphite is producing edges with no profile. Once + this is fixed, drop this. */ + if (e->probability.initialized_p ()) + EDGE_INFO (e)->back_edge_prob + = e->probability.to_reg_br_prob_base (); + else + EDGE_INFO (e)->back_edge_prob = REG_BR_PROB_BASE / 2; EDGE_INFO (e)->back_edge_prob *= real_inv_br_prob_base; } } @@ -3564,16 +3567,28 @@ estimate_bb_frequencies (bool force) to outermost to examine frequencies for back edges. */ estimate_loops (); + bool global0 = ENTRY_BLOCK_PTR_FOR_FN (cfun)->count.initialized_p () + && ENTRY_BLOCK_PTR_FOR_FN (cfun)->count.ipa_p (); + freq_max = 0; FOR_EACH_BB_FN (bb, cfun) if (freq_max < BLOCK_INFO (bb)->frequency) freq_max = BLOCK_INFO (bb)->frequency; freq_max = real_bb_freq_max / freq_max; + cfun->cfg->count_max = profile_count::uninitialized (); FOR_BB_BETWEEN (bb, ENTRY_BLOCK_PTR_FOR_FN (cfun), NULL, next_bb) { sreal tmp = BLOCK_INFO (bb)->frequency * freq_max + real_one_half; - bb->frequency = tmp.to_int (); + profile_count count = profile_count::from_gcov_type (tmp.to_int ()); + + /* If we have profile feedback in which this function was never + executed, then preserve this info. */ + if (global0) + bb->count = count.global0 (); + else if (!(bb->count == profile_count::zero ())) + bb->count = count.guessed_local (); + cfun->cfg->count_max = cfun->cfg->count_max.max (bb->count); } free_aux_for_blocks (); @@ -3598,7 +3613,8 @@ compute_function_frequency (void) if (profile_status_for_fn (cfun) != PROFILE_READ) { int flags = flags_from_decl_or_type (current_function_decl); - if (ENTRY_BLOCK_PTR_FOR_FN (cfun)->count == profile_count::zero () + if ((ENTRY_BLOCK_PTR_FOR_FN (cfun)->count.ipa_p () + && ENTRY_BLOCK_PTR_FOR_FN (cfun)->count.ipa() == profile_count::zero ()) || lookup_attribute ("cold", DECL_ATTRIBUTES (current_function_decl)) != NULL) { @@ -3717,7 +3733,7 @@ pass_profile::execute (function *fun) { struct loop *loop; FOR_EACH_LOOP (loop, LI_FROM_INNERMOST) - if (loop->header->frequency) + if (loop->header->count.initialized_p ()) fprintf (dump_file, "Loop got predicted %d to iterate %i times.\n", loop->num, (int)expected_loop_iterations_unbounded (loop)); @@ -3843,15 +3859,12 @@ rebuild_frequencies (void) which may also lead to frequencies incorrectly reduced to 0. There is less precision in the probabilities, so we only do this for small max counts. */ - profile_count count_max = profile_count::zero (); + cfun->cfg->count_max = profile_count::uninitialized (); basic_block bb; FOR_BB_BETWEEN (bb, ENTRY_BLOCK_PTR_FOR_FN (cfun), NULL, next_bb) - if (bb->count > count_max) - count_max = bb->count; + cfun->cfg->count_max = cfun->cfg->count_max.max (bb->count); - if (profile_status_for_fn (cfun) == PROFILE_GUESSED - || (!flag_auto_profile && profile_status_for_fn (cfun) == PROFILE_READ - && count_max < REG_BR_PROB_BASE / 10)) + if (profile_status_for_fn (cfun) == PROFILE_GUESSED) { loop_optimizer_init (0); add_noreturn_fake_exit_edges (); @@ -4017,17 +4030,19 @@ force_edge_cold (edge e, bool impossible) after loop transforms. */ if (!(prob_sum > profile_probability::never ()) && count_sum == profile_count::zero () - && single_pred_p (e->src) && e->src->frequency > (impossible ? 0 : 1)) + && single_pred_p (e->src) && e->src->count.to_frequency (cfun) + > (impossible ? 0 : 1)) { - int old_frequency = e->src->frequency; + int old_frequency = e->src->count.to_frequency (cfun); if (dump_file && (dump_flags & TDF_DETAILS)) fprintf (dump_file, "Making bb %i %s.\n", e->src->index, impossible ? "impossible" : "cold"); - e->src->frequency = MIN (e->src->frequency, impossible ? 0 : 1); + int new_frequency = MIN (e->src->count.to_frequency (cfun), + impossible ? 0 : 1); if (impossible) e->src->count = profile_count::zero (); else - e->src->count = e->count ().apply_scale (e->src->frequency, + e->src->count = e->count ().apply_scale (new_frequency, old_frequency); force_edge_cold (single_pred_edge (e->src), impossible); } |