/* IRA allocation based on graph coloring. Copyright (C) 2006-2014 Free Software Foundation, Inc. Contributed by Vladimir Makarov . This file is part of GCC. 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 . */ #include "config.h" #include "system.h" #include "coretypes.h" #include "tm.h" #include "rtl.h" #include "tm_p.h" #include "target.h" #include "regs.h" #include "flags.h" #include "sbitmap.h" #include "bitmap.h" #include "hash-table.h" #include "hard-reg-set.h" #include "basic-block.h" #include "expr.h" #include "diagnostic-core.h" #include "reload.h" #include "params.h" #include "df.h" #include "ira-int.h" typedef struct allocno_hard_regs *allocno_hard_regs_t; /* The structure contains information about hard registers can be assigned to allocnos. Usually it is allocno profitable hard registers but in some cases this set can be a bit different. Major reason of the difference is a requirement to use hard register sets that form a tree or a forest (set of trees), i.e. hard register set of a node should contain hard register sets of its subnodes. */ struct allocno_hard_regs { /* Hard registers can be assigned to an allocno. */ HARD_REG_SET set; /* Overall (spilling) cost of all allocnos with given register set. */ HOST_WIDEST_INT cost; }; typedef struct allocno_hard_regs_node *allocno_hard_regs_node_t; /* A node representing allocno hard registers. Such nodes form a forest (set of trees). Each subnode of given node in the forest refers for hard register set (usually allocno profitable hard register set) which is a subset of one referred from given node. */ struct allocno_hard_regs_node { /* Set up number of the node in preorder traversing of the forest. */ int preorder_num; /* Used for different calculation like finding conflict size of an allocno. */ int check; /* Used for calculation of conflict size of an allocno. The conflict size of the allocno is maximal number of given allocno hard registers needed for allocation of the conflicting allocnos. Given allocno is trivially colored if this number plus the number of hard registers needed for given allocno is not greater than the number of given allocno hard register set. */ int conflict_size; /* The number of hard registers given by member hard_regs. */ int hard_regs_num; /* The following member is used to form the final forest. */ bool used_p; /* Pointer to the corresponding profitable hard registers. */ allocno_hard_regs_t hard_regs; /* Parent, first subnode, previous and next node with the same parent in the forest. */ allocno_hard_regs_node_t parent, first, prev, next; }; /* Info about changing hard reg costs of an allocno. */ struct update_cost_record { /* Hard regno for which we changed the cost. */ int hard_regno; /* Divisor used when we changed the cost of HARD_REGNO. */ int divisor; /* Next record for given allocno. */ struct update_cost_record *next; }; /* To decrease footprint of ira_allocno structure we store all data needed only for coloring in the following structure. */ struct allocno_color_data { /* TRUE value means that the allocno was not removed yet from the conflicting graph during colouring. */ unsigned int in_graph_p : 1; /* TRUE if it is put on the stack to make other allocnos colorable. */ unsigned int may_be_spilled_p : 1; /* TRUE if the allocno is trivially colorable. */ unsigned int colorable_p : 1; /* Number of hard registers of the allocno class really available for the allocno allocation. It is number of the profitable hard regs. */ int available_regs_num; /* Allocnos in a bucket (used in coloring) chained by the following two members. */ ira_allocno_t next_bucket_allocno; ira_allocno_t prev_bucket_allocno; /* Used for temporary purposes. */ int temp; /* Used to exclude repeated processing. */ int last_process; /* Profitable hard regs available for this pseudo allocation. It means that the set excludes unavailable hard regs and hard regs conflicting with given pseudo. They should be of the allocno class. */ HARD_REG_SET profitable_hard_regs; /* Record hard regs which has been newly assigned to other pseudo regs which are conflicting with given pseduo when coloring current region. For hardregs in new_conflict_hard_regs, the updated_hard_reg_costs and updated_conflict_hard_reg_costs of the hardregs for given pseudo are stale, so they should not be used for cost update in update_conflict_hard_regno_costs. */ HARD_REG_SET new_conflict_hard_regs; /* The allocno hard registers node. */ allocno_hard_regs_node_t hard_regs_node; /* Array of structures allocno_hard_regs_subnode representing given allocno hard registers node (the 1st element in the array) and all its subnodes in the tree (forest) of allocno hard register nodes (see comments above). */ int hard_regs_subnodes_start; /* The length of the previous array. */ int hard_regs_subnodes_num; /* Records about updating allocno hard reg costs from copies. If the allocno did not get expected hard register, these records are used to restore original hard reg costs of allocnos connected to this allocno by copies. */ struct update_cost_record *update_cost_records; /* Threads. We collect allocnos connected by copies into threads and try to assign hard regs to allocnos by threads. */ /* Allocno representing all thread. */ ira_allocno_t first_thread_allocno; /* Allocnos in thread forms a cycle list through the following member. */ ira_allocno_t next_thread_allocno; /* All thread frequency. Defined only for first thread allocno. */ int thread_freq; }; /* See above. */ typedef struct allocno_color_data *allocno_color_data_t; /* Container for storing allocno data concerning coloring. */ static allocno_color_data_t allocno_color_data; /* Macro to access the data concerning coloring. */ #define ALLOCNO_COLOR_DATA(a) ((allocno_color_data_t) ALLOCNO_ADD_DATA (a)) /* Used for finding allocno colorability to exclude repeated allocno processing and for updating preferencing to exclude repeated allocno processing during assignment. */ static int curr_allocno_process; /* This file contains code for regional graph coloring, spill/restore code placement optimization, and code helping the reload pass to do a better job. */ /* Bitmap of allocnos which should be colored. */ static bitmap coloring_allocno_bitmap; /* Bitmap of allocnos which should be taken into account during coloring. In general case it contains allocnos from coloring_allocno_bitmap plus other already colored conflicting allocnos. */ static bitmap consideration_allocno_bitmap; /* All allocnos sorted according their priorities. */ static ira_allocno_t *sorted_allocnos; /* Vec representing the stack of allocnos used during coloring. */ static vec allocno_stack_vec; /* Helper for qsort comparison callbacks - return a positive integer if X > Y, or a negative value otherwise. Use a conditional expression instead of a difference computation to insulate from possible overflow issues, e.g. X - Y < 0 for some X > 0 and Y < 0. */ #define SORTGT(x,y) (((x) > (y)) ? 1 : -1) /* Definition of vector of allocno hard registers. */ /* Vector of unique allocno hard registers. */ static vec allocno_hard_regs_vec; struct allocno_hard_regs_hasher : typed_noop_remove { typedef allocno_hard_regs value_type; typedef allocno_hard_regs compare_type; static inline hashval_t hash (const value_type *); static inline bool equal (const value_type *, const compare_type *); }; /* Returns hash value for allocno hard registers V. */ inline hashval_t allocno_hard_regs_hasher::hash (const value_type *hv) { return iterative_hash (&hv->set, sizeof (HARD_REG_SET), 0); } /* Compares allocno hard registers V1 and V2. */ inline bool allocno_hard_regs_hasher::equal (const value_type *hv1, const compare_type *hv2) { return hard_reg_set_equal_p (hv1->set, hv2->set); } /* Hash table of unique allocno hard registers. */ static hash_table allocno_hard_regs_htab; /* Return allocno hard registers in the hash table equal to HV. */ static allocno_hard_regs_t find_hard_regs (allocno_hard_regs_t hv) { return allocno_hard_regs_htab.find (hv); } /* Insert allocno hard registers HV in the hash table (if it is not there yet) and return the value which in the table. */ static allocno_hard_regs_t insert_hard_regs (allocno_hard_regs_t hv) { allocno_hard_regs **slot = allocno_hard_regs_htab.find_slot (hv, INSERT); if (*slot == NULL) *slot = hv; return *slot; } /* Initialize data concerning allocno hard registers. */ static void init_allocno_hard_regs (void) { allocno_hard_regs_vec.create (200); allocno_hard_regs_htab.create (200); } /* Add (or update info about) allocno hard registers with SET and COST. */ static allocno_hard_regs_t add_allocno_hard_regs (HARD_REG_SET set, HOST_WIDEST_INT cost) { struct allocno_hard_regs temp; allocno_hard_regs_t hv; gcc_assert (! hard_reg_set_empty_p (set)); COPY_HARD_REG_SET (temp.set, set); if ((hv = find_hard_regs (&temp)) != NULL) hv->cost += cost; else { hv = ((struct allocno_hard_regs *) ira_allocate (sizeof (struct allocno_hard_regs))); COPY_HARD_REG_SET (hv->set, set); hv->cost = cost; allocno_hard_regs_vec.safe_push (hv); insert_hard_regs (hv); } return hv; } /* Finalize data concerning allocno hard registers. */ static void finish_allocno_hard_regs (void) { int i; allocno_hard_regs_t hv; for (i = 0; allocno_hard_regs_vec.iterate (i, &hv); i++) ira_free (hv); allocno_hard_regs_htab.dispose (); allocno_hard_regs_vec.release (); } /* Sort hard regs according to their frequency of usage. */ static int allocno_hard_regs_compare (const void *v1p, const void *v2p) { allocno_hard_regs_t hv1 = *(const allocno_hard_regs_t *) v1p; allocno_hard_regs_t hv2 = *(const allocno_hard_regs_t *) v2p; if (hv2->cost > hv1->cost) return 1; else if (hv2->cost < hv1->cost) return -1; else return 0; } /* Used for finding a common ancestor of two allocno hard registers nodes in the forest. We use the current value of 'node_check_tick' to mark all nodes from one node to the top and then walking up from another node until we find a marked node. It is also used to figure out allocno colorability as a mark that we already reset value of member 'conflict_size' for the forest node corresponding to the processed allocno. */ static int node_check_tick; /* Roots of the forest containing hard register sets can be assigned to allocnos. */ static allocno_hard_regs_node_t hard_regs_roots; /* Definition of vector of allocno hard register nodes. */ /* Vector used to create the forest. */ static vec hard_regs_node_vec; /* Create and return allocno hard registers node containing allocno hard registers HV. */ static allocno_hard_regs_node_t create_new_allocno_hard_regs_node (allocno_hard_regs_t hv) { allocno_hard_regs_node_t new_node; new_node = ((struct allocno_hard_regs_node *) ira_allocate (sizeof (struct allocno_hard_regs_node))); new_node->check = 0; new_node->hard_regs = hv; new_node->hard_regs_num = hard_reg_set_size (hv->set); new_node->first = NULL; new_node->used_p = false; return new_node; } /* Add allocno hard registers node NEW_NODE to the forest on its level given by ROOTS. */ static void add_new_allocno_hard_regs_node_to_forest (allocno_hard_regs_node_t *roots, allocno_hard_regs_node_t new_node) { new_node->next = *roots; if (new_node->next != NULL) new_node->next->prev = new_node; new_node->prev = NULL; *roots = new_node; } /* Add allocno hard registers HV (or its best approximation if it is not possible) to the forest on its level given by ROOTS. */ static void add_allocno_hard_regs_to_forest (allocno_hard_regs_node_t *roots, allocno_hard_regs_t hv) { unsigned int i, start; allocno_hard_regs_node_t node, prev, new_node; HARD_REG_SET temp_set; allocno_hard_regs_t hv2; start = hard_regs_node_vec.length (); for (node = *roots; node != NULL; node = node->next) { if (hard_reg_set_equal_p (hv->set, node->hard_regs->set)) return; if (hard_reg_set_subset_p (hv->set, node->hard_regs->set)) { add_allocno_hard_regs_to_forest (&node->first, hv); return; } if (hard_reg_set_subset_p (node->hard_regs->set, hv->set)) hard_regs_node_vec.safe_push (node); else if (hard_reg_set_intersect_p (hv->set, node->hard_regs->set)) { COPY_HARD_REG_SET (temp_set, hv->set); AND_HARD_REG_SET (temp_set, node->hard_regs->set); hv2 = add_allocno_hard_regs (temp_set, hv->cost); add_allocno_hard_regs_to_forest (&node->first, hv2); } } if (hard_regs_node_vec.length () > start + 1) { /* Create a new node which contains nodes in hard_regs_node_vec. */ CLEAR_HARD_REG_SET (temp_set); for (i = start; i < hard_regs_node_vec.length (); i++) { node = hard_regs_node_vec[i]; IOR_HARD_REG_SET (temp_set, node->hard_regs->set); } hv = add_allocno_hard_regs (temp_set, hv->cost); new_node = create_new_allocno_hard_regs_node (hv); prev = NULL; for (i = start; i < hard_regs_node_vec.length (); i++) { node = hard_regs_node_vec[i]; if (node->prev == NULL) *roots = node->next; else node->prev->next = node->next; if (node->next != NULL) node->next->prev = node->prev; if (prev == NULL) new_node->first = node; else prev->next = node; node->prev = prev; node->next = NULL; prev = node; } add_new_allocno_hard_regs_node_to_forest (roots, new_node); } hard_regs_node_vec.truncate (start); } /* Add allocno hard registers nodes starting with the forest level given by FIRST which contains biggest set inside SET. */ static void collect_allocno_hard_regs_cover (allocno_hard_regs_node_t first, HARD_REG_SET set) { allocno_hard_regs_node_t node; ira_assert (first != NULL); for (node = first; node != NULL; node = node->next) if (hard_reg_set_subset_p (node->hard_regs->set, set)) hard_regs_node_vec.safe_push (node); else if (hard_reg_set_intersect_p (set, node->hard_regs->set)) collect_allocno_hard_regs_cover (node->first, set); } /* Set up field parent as PARENT in all allocno hard registers nodes in forest given by FIRST. */ static void setup_allocno_hard_regs_nodes_parent (allocno_hard_regs_node_t first, allocno_hard_regs_node_t parent) { allocno_hard_regs_node_t node; for (node = first; node != NULL; node = node->next) { node->parent = parent; setup_allocno_hard_regs_nodes_parent (node->first, node); } } /* Return allocno hard registers node which is a first common ancestor node of FIRST and SECOND in the forest. */ static allocno_hard_regs_node_t first_common_ancestor_node (allocno_hard_regs_node_t first, allocno_hard_regs_node_t second) { allocno_hard_regs_node_t node; node_check_tick++; for (node = first; node != NULL; node = node->parent) node->check = node_check_tick; for (node = second; node != NULL; node = node->parent) if (node->check == node_check_tick) return node; return first_common_ancestor_node (second, first); } /* Print hard reg set SET to F. */ static void print_hard_reg_set (FILE *f, HARD_REG_SET set, bool new_line_p) { int i, start; for (start = -1, i = 0; i < FIRST_PSEUDO_REGISTER; i++) { if (TEST_HARD_REG_BIT (set, i)) { if (i == 0 || ! TEST_HARD_REG_BIT (set, i - 1)) start = i; } if (start >= 0 && (i == FIRST_PSEUDO_REGISTER - 1 || ! TEST_HARD_REG_BIT (set, i))) { if (start == i - 1) fprintf (f, " %d", start); else if (start == i - 2) fprintf (f, " %d %d", start, start + 1); else fprintf (f, " %d-%d", start, i - 1); start = -1; } } if (new_line_p) fprintf (f, "\n"); } /* Print allocno hard register subforest given by ROOTS and its LEVEL to F. */ static void print_hard_regs_subforest (FILE *f, allocno_hard_regs_node_t roots, int level) { int i; allocno_hard_regs_node_t node; for (node = roots; node != NULL; node = node->next) { fprintf (f, " "); for (i = 0; i < level * 2; i++) fprintf (f, " "); fprintf (f, "%d:(", node->preorder_num); print_hard_reg_set (f, node->hard_regs->set, false); fprintf (f, ")@" HOST_WIDEST_INT_PRINT_DEC "\n", node->hard_regs->cost); print_hard_regs_subforest (f, node->first, level + 1); } } /* Print the allocno hard register forest to F. */ static void print_hard_regs_forest (FILE *f) { fprintf (f, " Hard reg set forest:\n"); print_hard_regs_subforest (f, hard_regs_roots, 1); } /* Print the allocno hard register forest to stderr. */ void ira_debug_hard_regs_forest (void) { print_hard_regs_forest (stderr); } /* Remove unused allocno hard registers nodes from forest given by its *ROOTS. */ static void remove_unused_allocno_hard_regs_nodes (allocno_hard_regs_node_t *roots) { allocno_hard_regs_node_t node, prev, next, last; for (prev = NULL, node = *roots; node != NULL; node = next) { next = node->next; if (node->used_p) { remove_unused_allocno_hard_regs_nodes (&node->first); prev = node; } else { for (last = node->first; last != NULL && last->next != NULL; last = last->next) ; if (last != NULL) { if (prev == NULL) *roots = node->first; else prev->next = node->first; if (next != NULL) next->prev = last; last->next = next; next = node->first; } else { if (prev == NULL) *roots = next; else prev->next = next; if (next != NULL) next->prev = prev; } ira_free (node); } } } /* Set up fields preorder_num starting with START_NUM in all allocno hard registers nodes in forest given by FIRST. Return biggest set PREORDER_NUM increased by 1. */ static int enumerate_allocno_hard_regs_nodes (allocno_hard_regs_node_t first, allocno_hard_regs_node_t parent, int start_num) { allocno_hard_regs_node_t node; for (node = first; node != NULL; node = node->next) { node->preorder_num = start_num++; node->parent = parent; start_num = enumerate_allocno_hard_regs_nodes (node->first, node, start_num); } return start_num; } /* Number of allocno hard registers nodes in the forest. */ static int allocno_hard_regs_nodes_num; /* Table preorder number of allocno hard registers node in the forest -> the allocno hard registers node. */ static allocno_hard_regs_node_t *allocno_hard_regs_nodes; /* See below. */ typedef struct allocno_hard_regs_subnode *allocno_hard_regs_subnode_t; /* The structure is used to describes all subnodes (not only immediate ones) in the mentioned above tree for given allocno hard register node. The usage of such data accelerates calculation of colorability of given allocno. */ struct allocno_hard_regs_subnode { /* The conflict size of conflicting allocnos whose hard register sets are equal sets (plus supersets if given node is given allocno hard registers node) of one in the given node. */ int left_conflict_size; /* The summary conflict size of conflicting allocnos whose hard register sets are strict subsets of one in the given node. Overall conflict size is left_conflict_subnodes_size + MIN (max_node_impact - left_conflict_subnodes_size, left_conflict_size) */ short left_conflict_subnodes_size; short max_node_impact; }; /* Container for hard regs subnodes of all allocnos. */ static allocno_hard_regs_subnode_t allocno_hard_regs_subnodes; /* Table (preorder number of allocno hard registers node in the forest, preorder number of allocno hard registers subnode) -> index of the subnode relative to the node. -1 if it is not a subnode. */ static int *allocno_hard_regs_subnode_index; /* Setup arrays ALLOCNO_HARD_REGS_NODES and ALLOCNO_HARD_REGS_SUBNODE_INDEX. */ static void setup_allocno_hard_regs_subnode_index (allocno_hard_regs_node_t first) { allocno_hard_regs_node_t node, parent; int index; for (node = first; node != NULL; node = node->next) { allocno_hard_regs_nodes[node->preorder_num] = node; for (parent = node; parent != NULL; parent = parent->parent) { index = parent->preorder_num * allocno_hard_regs_nodes_num; allocno_hard_regs_subnode_index[index + node->preorder_num] = node->preorder_num - parent->preorder_num; } setup_allocno_hard_regs_subnode_index (node->first); } } /* Count all allocno hard registers nodes in tree ROOT. */ static int get_allocno_hard_regs_subnodes_num (allocno_hard_regs_node_t root) { int len = 1; for (root = root->first; root != NULL; root = root->next) len += get_allocno_hard_regs_subnodes_num (root); return len; } /* Build the forest of allocno hard registers nodes and assign each allocno a node from the forest. */ static void form_allocno_hard_regs_nodes_forest (void) { unsigned int i, j, size, len; int start; ira_allocno_t a; allocno_hard_regs_t hv; bitmap_iterator bi; HARD_REG_SET temp; allocno_hard_regs_node_t node, allocno_hard_regs_node; allocno_color_data_t allocno_data; node_check_tick = 0; init_allocno_hard_regs (); hard_regs_roots = NULL; hard_regs_node_vec.create (100); for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (! TEST_HARD_REG_BIT (ira_no_alloc_regs, i)) { CLEAR_HARD_REG_SET (temp); SET_HARD_REG_BIT (temp, i); hv = add_allocno_hard_regs (temp, 0); node = create_new_allocno_hard_regs_node (hv); add_new_allocno_hard_regs_node_to_forest (&hard_regs_roots, node); } start = allocno_hard_regs_vec.length (); EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; allocno_data = ALLOCNO_COLOR_DATA (a); if (hard_reg_set_empty_p (allocno_data->profitable_hard_regs)) continue; hv = (add_allocno_hard_regs (allocno_data->profitable_hard_regs, ALLOCNO_MEMORY_COST (a) - ALLOCNO_CLASS_COST (a))); } SET_HARD_REG_SET (temp); AND_COMPL_HARD_REG_SET (temp, ira_no_alloc_regs); add_allocno_hard_regs (temp, 0); qsort (allocno_hard_regs_vec.address () + start, allocno_hard_regs_vec.length () - start, sizeof (allocno_hard_regs_t), allocno_hard_regs_compare); for (i = start; allocno_hard_regs_vec.iterate (i, &hv); i++) { add_allocno_hard_regs_to_forest (&hard_regs_roots, hv); ira_assert (hard_regs_node_vec.length () == 0); } /* We need to set up parent fields for right work of first_common_ancestor_node. */ setup_allocno_hard_regs_nodes_parent (hard_regs_roots, NULL); EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; allocno_data = ALLOCNO_COLOR_DATA (a); if (hard_reg_set_empty_p (allocno_data->profitable_hard_regs)) continue; hard_regs_node_vec.truncate (0); collect_allocno_hard_regs_cover (hard_regs_roots, allocno_data->profitable_hard_regs); allocno_hard_regs_node = NULL; for (j = 0; hard_regs_node_vec.iterate (j, &node); j++) allocno_hard_regs_node = (j == 0 ? node : first_common_ancestor_node (node, allocno_hard_regs_node)); /* That is a temporary storage. */ allocno_hard_regs_node->used_p = true; allocno_data->hard_regs_node = allocno_hard_regs_node; } ira_assert (hard_regs_roots->next == NULL); hard_regs_roots->used_p = true; remove_unused_allocno_hard_regs_nodes (&hard_regs_roots); allocno_hard_regs_nodes_num = enumerate_allocno_hard_regs_nodes (hard_regs_roots, NULL, 0); allocno_hard_regs_nodes = ((allocno_hard_regs_node_t *) ira_allocate (allocno_hard_regs_nodes_num * sizeof (allocno_hard_regs_node_t))); size = allocno_hard_regs_nodes_num * allocno_hard_regs_nodes_num; allocno_hard_regs_subnode_index = (int *) ira_allocate (size * sizeof (int)); for (i = 0; i < size; i++) allocno_hard_regs_subnode_index[i] = -1; setup_allocno_hard_regs_subnode_index (hard_regs_roots); start = 0; EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; allocno_data = ALLOCNO_COLOR_DATA (a); if (hard_reg_set_empty_p (allocno_data->profitable_hard_regs)) continue; len = get_allocno_hard_regs_subnodes_num (allocno_data->hard_regs_node); allocno_data->hard_regs_subnodes_start = start; allocno_data->hard_regs_subnodes_num = len; start += len; } allocno_hard_regs_subnodes = ((allocno_hard_regs_subnode_t) ira_allocate (sizeof (struct allocno_hard_regs_subnode) * start)); hard_regs_node_vec.release (); } /* Free tree of allocno hard registers nodes given by its ROOT. */ static void finish_allocno_hard_regs_nodes_tree (allocno_hard_regs_node_t root) { allocno_hard_regs_node_t child, next; for (child = root->first; child != NULL; child = next) { next = child->next; finish_allocno_hard_regs_nodes_tree (child); } ira_free (root); } /* Finish work with the forest of allocno hard registers nodes. */ static void finish_allocno_hard_regs_nodes_forest (void) { allocno_hard_regs_node_t node, next; ira_free (allocno_hard_regs_subnodes); for (node = hard_regs_roots; node != NULL; node = next) { next = node->next; finish_allocno_hard_regs_nodes_tree (node); } ira_free (allocno_hard_regs_nodes); ira_free (allocno_hard_regs_subnode_index); finish_allocno_hard_regs (); } /* Set up left conflict sizes and left conflict subnodes sizes of hard registers subnodes of allocno A. Return TRUE if allocno A is trivially colorable. */ static bool setup_left_conflict_sizes_p (ira_allocno_t a) { int i, k, nobj, start; int conflict_size, left_conflict_subnodes_size, node_preorder_num; allocno_color_data_t data; HARD_REG_SET profitable_hard_regs; allocno_hard_regs_subnode_t subnodes; allocno_hard_regs_node_t node; HARD_REG_SET node_set; nobj = ALLOCNO_NUM_OBJECTS (a); conflict_size = 0; data = ALLOCNO_COLOR_DATA (a); subnodes = allocno_hard_regs_subnodes + data->hard_regs_subnodes_start; COPY_HARD_REG_SET (profitable_hard_regs, data->profitable_hard_regs); node = data->hard_regs_node; node_preorder_num = node->preorder_num; COPY_HARD_REG_SET (node_set, node->hard_regs->set); node_check_tick++; for (k = 0; k < nobj; k++) { ira_object_t obj = ALLOCNO_OBJECT (a, k); ira_object_t conflict_obj; ira_object_conflict_iterator oci; FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { int size; ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); allocno_hard_regs_node_t conflict_node, temp_node; HARD_REG_SET conflict_node_set; allocno_color_data_t conflict_data; conflict_data = ALLOCNO_COLOR_DATA (conflict_a); if (! ALLOCNO_COLOR_DATA (conflict_a)->in_graph_p || ! hard_reg_set_intersect_p (profitable_hard_regs, conflict_data ->profitable_hard_regs)) continue; conflict_node = conflict_data->hard_regs_node; COPY_HARD_REG_SET (conflict_node_set, conflict_node->hard_regs->set); if (hard_reg_set_subset_p (node_set, conflict_node_set)) temp_node = node; else { ira_assert (hard_reg_set_subset_p (conflict_node_set, node_set)); temp_node = conflict_node; } if (temp_node->check != node_check_tick) { temp_node->check = node_check_tick; temp_node->conflict_size = 0; } size = (ira_reg_class_max_nregs [ALLOCNO_CLASS (conflict_a)][ALLOCNO_MODE (conflict_a)]); if (ALLOCNO_NUM_OBJECTS (conflict_a) > 1) /* We will deal with the subwords individually. */ size = 1; temp_node->conflict_size += size; } } for (i = 0; i < data->hard_regs_subnodes_num; i++) { allocno_hard_regs_node_t temp_node; temp_node = allocno_hard_regs_nodes[i + node_preorder_num]; ira_assert (temp_node->preorder_num == i + node_preorder_num); subnodes[i].left_conflict_size = (temp_node->check != node_check_tick ? 0 : temp_node->conflict_size); if (hard_reg_set_subset_p (temp_node->hard_regs->set, profitable_hard_regs)) subnodes[i].max_node_impact = temp_node->hard_regs_num; else { HARD_REG_SET temp_set; int j, n, hard_regno; enum reg_class aclass; COPY_HARD_REG_SET (temp_set, temp_node->hard_regs->set); AND_HARD_REG_SET (temp_set, profitable_hard_regs); aclass = ALLOCNO_CLASS (a); for (n = 0, j = ira_class_hard_regs_num[aclass] - 1; j >= 0; j--) { hard_regno = ira_class_hard_regs[aclass][j]; if (TEST_HARD_REG_BIT (temp_set, hard_regno)) n++; } subnodes[i].max_node_impact = n; } subnodes[i].left_conflict_subnodes_size = 0; } start = node_preorder_num * allocno_hard_regs_nodes_num; for (i = data->hard_regs_subnodes_num - 1; i >= 0; i--) { int size, parent_i; allocno_hard_regs_node_t parent; size = (subnodes[i].left_conflict_subnodes_size + MIN (subnodes[i].max_node_impact - subnodes[i].left_conflict_subnodes_size, subnodes[i].left_conflict_size)); parent = allocno_hard_regs_nodes[i + node_preorder_num]->parent; if (parent == NULL) continue; parent_i = allocno_hard_regs_subnode_index[start + parent->preorder_num]; if (parent_i < 0) continue; subnodes[parent_i].left_conflict_subnodes_size += size; } left_conflict_subnodes_size = subnodes[0].left_conflict_subnodes_size; conflict_size += (left_conflict_subnodes_size + MIN (subnodes[0].max_node_impact - left_conflict_subnodes_size, subnodes[0].left_conflict_size)); conflict_size += ira_reg_class_max_nregs[ALLOCNO_CLASS (a)][ALLOCNO_MODE (a)]; data->colorable_p = conflict_size <= data->available_regs_num; return data->colorable_p; } /* Update left conflict sizes of hard registers subnodes of allocno A after removing allocno REMOVED_A with SIZE from the conflict graph. Return TRUE if A is trivially colorable. */ static bool update_left_conflict_sizes_p (ira_allocno_t a, ira_allocno_t removed_a, int size) { int i, conflict_size, before_conflict_size, diff, start; int node_preorder_num, parent_i; allocno_hard_regs_node_t node, removed_node, parent; allocno_hard_regs_subnode_t subnodes; allocno_color_data_t data = ALLOCNO_COLOR_DATA (a); ira_assert (! data->colorable_p); node = data->hard_regs_node; node_preorder_num = node->preorder_num; removed_node = ALLOCNO_COLOR_DATA (removed_a)->hard_regs_node; ira_assert (hard_reg_set_subset_p (removed_node->hard_regs->set, node->hard_regs->set) || hard_reg_set_subset_p (node->hard_regs->set, removed_node->hard_regs->set)); start = node_preorder_num * allocno_hard_regs_nodes_num; i = allocno_hard_regs_subnode_index[start + removed_node->preorder_num]; if (i < 0) i = 0; subnodes = allocno_hard_regs_subnodes + data->hard_regs_subnodes_start; before_conflict_size = (subnodes[i].left_conflict_subnodes_size + MIN (subnodes[i].max_node_impact - subnodes[i].left_conflict_subnodes_size, subnodes[i].left_conflict_size)); subnodes[i].left_conflict_size -= size; for (;;) { conflict_size = (subnodes[i].left_conflict_subnodes_size + MIN (subnodes[i].max_node_impact - subnodes[i].left_conflict_subnodes_size, subnodes[i].left_conflict_size)); if ((diff = before_conflict_size - conflict_size) == 0) break; ira_assert (conflict_size < before_conflict_size); parent = allocno_hard_regs_nodes[i + node_preorder_num]->parent; if (parent == NULL) break; parent_i = allocno_hard_regs_subnode_index[start + parent->preorder_num]; if (parent_i < 0) break; i = parent_i; before_conflict_size = (subnodes[i].left_conflict_subnodes_size + MIN (subnodes[i].max_node_impact - subnodes[i].left_conflict_subnodes_size, subnodes[i].left_conflict_size)); subnodes[i].left_conflict_subnodes_size -= diff; } if (i != 0 || (conflict_size + ira_reg_class_max_nregs[ALLOCNO_CLASS (a)][ALLOCNO_MODE (a)] > data->available_regs_num)) return false; data->colorable_p = true; return true; } /* Return true if allocno A has empty profitable hard regs. */ static bool empty_profitable_hard_regs (ira_allocno_t a) { allocno_color_data_t data = ALLOCNO_COLOR_DATA (a); return hard_reg_set_empty_p (data->profitable_hard_regs); } /* Set up profitable hard registers for each allocno being colored. */ static void setup_profitable_hard_regs (void) { unsigned int i; int j, k, nobj, hard_regno, nregs, class_size; ira_allocno_t a; bitmap_iterator bi; enum reg_class aclass; enum machine_mode mode; allocno_color_data_t data; /* Initial set up from allocno classes and explicitly conflicting hard regs. */ EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; if ((aclass = ALLOCNO_CLASS (a)) == NO_REGS) continue; data = ALLOCNO_COLOR_DATA (a); if (ALLOCNO_UPDATED_HARD_REG_COSTS (a) == NULL && ALLOCNO_CLASS_COST (a) > ALLOCNO_MEMORY_COST (a)) CLEAR_HARD_REG_SET (data->profitable_hard_regs); else { mode = ALLOCNO_MODE (a); COPY_HARD_REG_SET (data->profitable_hard_regs, ira_useful_class_mode_regs[aclass][mode]); nobj = ALLOCNO_NUM_OBJECTS (a); for (k = 0; k < nobj; k++) { ira_object_t obj = ALLOCNO_OBJECT (a, k); AND_COMPL_HARD_REG_SET (data->profitable_hard_regs, OBJECT_TOTAL_CONFLICT_HARD_REGS (obj)); } } /* If loop region is marked as !fp_is_free, all the allocnos inside it will not use fp as a free register. */ if (frame_pointer_partially_needed && !a->loop_tree_node->fp_is_free) CLEAR_HARD_REG_BIT (data->profitable_hard_regs, HARD_FRAME_POINTER_REGNUM); } /* Exclude hard regs already assigned for conflicting objects. */ EXECUTE_IF_SET_IN_BITMAP (consideration_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; if ((aclass = ALLOCNO_CLASS (a)) == NO_REGS || ! ALLOCNO_ASSIGNED_P (a) || (hard_regno = ALLOCNO_HARD_REGNO (a)) < 0) continue; mode = ALLOCNO_MODE (a); nregs = hard_regno_nregs[hard_regno][mode]; nobj = ALLOCNO_NUM_OBJECTS (a); for (k = 0; k < nobj; k++) { ira_object_t obj = ALLOCNO_OBJECT (a, k); ira_object_t conflict_obj; ira_object_conflict_iterator oci; FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); /* We can process the conflict allocno repeatedly with the same result. */ if (nregs == nobj && nregs > 1) { int num = OBJECT_SUBWORD (conflict_obj); if (REG_WORDS_BIG_ENDIAN) CLEAR_HARD_REG_BIT (ALLOCNO_COLOR_DATA (conflict_a)->profitable_hard_regs, hard_regno + nobj - num - 1); else CLEAR_HARD_REG_BIT (ALLOCNO_COLOR_DATA (conflict_a)->profitable_hard_regs, hard_regno + num); } else AND_COMPL_HARD_REG_SET (ALLOCNO_COLOR_DATA (conflict_a)->profitable_hard_regs, ira_reg_mode_hard_regset[hard_regno][mode]); } } } /* Exclude too costly hard regs. */ EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { int min_cost = INT_MAX; int *costs; a = ira_allocnos[i]; if ((aclass = ALLOCNO_CLASS (a)) == NO_REGS || empty_profitable_hard_regs (a)) continue; data = ALLOCNO_COLOR_DATA (a); mode = ALLOCNO_MODE (a); if ((costs = ALLOCNO_UPDATED_HARD_REG_COSTS (a)) != NULL || (costs = ALLOCNO_HARD_REG_COSTS (a)) != NULL) { class_size = ira_class_hard_regs_num[aclass]; for (j = 0; j < class_size; j++) { hard_regno = ira_class_hard_regs[aclass][j]; if (! TEST_HARD_REG_BIT (data->profitable_hard_regs, hard_regno)) continue; if (ALLOCNO_UPDATED_MEMORY_COST (a) < costs[j]) CLEAR_HARD_REG_BIT (data->profitable_hard_regs, hard_regno); else if (min_cost > costs[j]) min_cost = costs[j]; } } else if (ALLOCNO_UPDATED_MEMORY_COST (a) < ALLOCNO_UPDATED_CLASS_COST (a)) CLEAR_HARD_REG_SET (data->profitable_hard_regs); if (ALLOCNO_UPDATED_CLASS_COST (a) > min_cost) ALLOCNO_UPDATED_CLASS_COST (a) = min_cost; } } /* This page contains functions used to choose hard registers for allocnos. */ /* Pool for update cost records. */ static alloc_pool update_cost_record_pool; /* Initiate update cost records. */ static void init_update_cost_records (void) { update_cost_record_pool = create_alloc_pool ("update cost records", sizeof (struct update_cost_record), 100); } /* Return new update cost record with given params. */ static struct update_cost_record * get_update_cost_record (int hard_regno, int divisor, struct update_cost_record *next) { struct update_cost_record *record; record = (struct update_cost_record *) pool_alloc (update_cost_record_pool); record->hard_regno = hard_regno; record->divisor = divisor; record->next = next; return record; } /* Free memory for all records in LIST. */ static void free_update_cost_record_list (struct update_cost_record *list) { struct update_cost_record *next; while (list != NULL) { next = list->next; pool_free (update_cost_record_pool, list); list = next; } } /* Free memory for CURR record in LIST. */ static void free_a_update_cost_record (struct update_cost_record **list, struct update_cost_record *prev, struct update_cost_record *curr) { if (prev == NULL) *list = curr->next; else prev->next = curr->next; pool_free (update_cost_record_pool, curr); } /* Free memory allocated for all update cost records. */ static void finish_update_cost_records (void) { free_alloc_pool (update_cost_record_pool); } /* Array whose element value is TRUE if the corresponding hard register was already allocated for an allocno. */ static bool allocated_hardreg_p[FIRST_PSEUDO_REGISTER]; /* Describes one element in a queue of allocnos whose costs need to be updated. Each allocno in the queue is known to have an allocno class. */ struct update_cost_queue_elem { /* This element is in the queue iff CHECK == update_cost_check. */ int check; /* COST_HOP_DIVISOR**N, where N is the length of the shortest path connecting this allocno to the one being allocated. */ int divisor; /* Copy from which we are chaning costs of connected allocnos. It is used not go back in graph of allocnos connected by copies. */ ira_copy_t from; /* The next allocno in the queue, or null if this is the last element. */ ira_allocno_t next; }; /* The first element in a queue of allocnos whose copy costs need to be updated. Null if the queue is empty. */ static ira_allocno_t update_cost_queue; /* The last element in the queue described by update_cost_queue. Not valid if update_cost_queue is null. */ static struct update_cost_queue_elem *update_cost_queue_tail; /* A pool of elements in the queue described by update_cost_queue. Elements are indexed by ALLOCNO_NUM. */ static struct update_cost_queue_elem *update_cost_queue_elems; /* The current value of update_costs_from_copies call count. */ static int update_cost_check; /* Allocate and initialize data necessary for function update_costs_from_copies. */ static void initiate_cost_update (void) { size_t size; size = ira_allocnos_num * sizeof (struct update_cost_queue_elem); update_cost_queue_elems = (struct update_cost_queue_elem *) ira_allocate (size); memset (update_cost_queue_elems, 0, size); update_cost_check = 0; init_update_cost_records (); } /* Deallocate data used by function update_costs_from_copies. */ static void finish_cost_update (void) { ira_free (update_cost_queue_elems); finish_update_cost_records (); } /* When we traverse allocnos to update hard register costs, the cost divisor will be multiplied by the following macro value for each hop from given allocno to directly connected allocnos. */ #define COST_HOP_DIVISOR 4 /* Start a new cost-updating pass. */ static void start_update_cost (void) { update_cost_check++; update_cost_queue = NULL; } /* Add (ALLOCNO, FROM, DIVISOR) to the end of update_cost_queue, unless ALLOCNO is already in the queue, or has NO_REGS class. */ static inline void queue_update_cost (ira_allocno_t allocno, ira_copy_t from, int divisor) { struct update_cost_queue_elem *elem; elem = &update_cost_queue_elems[ALLOCNO_NUM (allocno)]; if (elem->check != update_cost_check && ALLOCNO_CLASS (allocno) != NO_REGS) { elem->check = update_cost_check; elem->from = from; elem->divisor = divisor; elem->next = NULL; if (update_cost_queue == NULL) update_cost_queue = allocno; else update_cost_queue_tail->next = allocno; update_cost_queue_tail = elem; } } /* Try to remove the first element from update_cost_queue. Return false if the queue was empty, otherwise make (*ALLOCNO, *FROM, *DIVISOR) describe the removed element. */ static inline bool get_next_update_cost (ira_allocno_t *allocno, ira_copy_t *from, int *divisor) { struct update_cost_queue_elem *elem; if (update_cost_queue == NULL) return false; *allocno = update_cost_queue; elem = &update_cost_queue_elems[ALLOCNO_NUM (*allocno)]; *from = elem->from; *divisor = elem->divisor; update_cost_queue = elem->next; return true; } /* Increase costs of HARD_REGNO by UPDATE_COST for ALLOCNO. Return true if we really modified the cost. */ static bool update_allocno_cost (ira_allocno_t allocno, int hard_regno, int update_cost) { int i; enum reg_class aclass = ALLOCNO_CLASS (allocno); i = ira_class_hard_reg_index[aclass][hard_regno]; if (i < 0) return false; ira_allocate_and_set_or_copy_costs (&ALLOCNO_UPDATED_HARD_REG_COSTS (allocno), aclass, ALLOCNO_UPDATED_CLASS_COST (allocno), ALLOCNO_HARD_REG_COSTS (allocno)); ira_allocate_and_set_or_copy_costs (&ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (allocno), aclass, 0, ALLOCNO_CONFLICT_HARD_REG_COSTS (allocno)); ALLOCNO_UPDATED_HARD_REG_COSTS (allocno)[i] += update_cost; ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (allocno)[i] += update_cost; return true; } /* Update (decrease if DECR_P) HARD_REGNO cost of allocnos connected by copies to ALLOCNO to increase chances to remove some copies as the result of subsequent assignment. Record cost updates if RECORD_P is true. CONFLICT_P shows whether this is updating cost for allocno known conflicting with another allocno assigned to HARD_REGNO. */ static void update_costs_from_allocno (ira_allocno_t allocno, int hard_regno, int divisor, bool decr_p, bool record_p, bool conflict_p) { int cost, update_cost; enum machine_mode mode; enum reg_class rclass, aclass; ira_allocno_t another_allocno; ira_copy_t cp, next_cp, from = NULL; rclass = REGNO_REG_CLASS (hard_regno); do { mode = ALLOCNO_MODE (allocno); ira_init_register_move_cost_if_necessary (mode); for (cp = ALLOCNO_COPIES (allocno); cp != NULL; cp = next_cp) { if (cp->first == allocno) { next_cp = cp->next_first_allocno_copy; another_allocno = cp->second; } else if (cp->second == allocno) { next_cp = cp->next_second_allocno_copy; another_allocno = cp->first; } else gcc_unreachable (); if (from != NULL && (another_allocno == from->first || another_allocno == from->second)) continue; /* For a = b * c; if there are copies (a, b) and (a, c), it makes no sense to propagate cost from b to a then to c. */ if (from != NULL && from->insn == cp->insn) continue; /* For a = b * c; if there are copies (a, b) and (a, c), and if we know b is conflicting with hardreg r1, sometimes it is bad to propagate the disfavor of r1 from b to a because the alterate copy (a, c) may still want to choose r1. */ if (conflict_p && find_alternate_copy (cp, another_allocno)) continue; aclass = ALLOCNO_CLASS (another_allocno); if (! TEST_HARD_REG_BIT (reg_class_contents[aclass], hard_regno) || ALLOCNO_ASSIGNED_P (another_allocno)) continue; cost = (cp->second == allocno ? ira_register_move_cost[mode][rclass][aclass] : ira_register_move_cost[mode][aclass][rclass]); if (decr_p) cost = -cost; update_cost = cp->freq * cost / divisor; if (update_cost == 0) continue; if (! update_allocno_cost (another_allocno, hard_regno, update_cost)) continue; queue_update_cost (another_allocno, cp, divisor * COST_HOP_DIVISOR); if (record_p && ALLOCNO_COLOR_DATA (another_allocno) != NULL) ALLOCNO_COLOR_DATA (another_allocno)->update_cost_records = get_update_cost_record (hard_regno, divisor * COST_HOP_DIVISOR, ALLOCNO_COLOR_DATA (another_allocno) ->update_cost_records); } } while (get_next_update_cost (&allocno, &from, &divisor)); } /* Decrease preferred ALLOCNO hard register costs and costs of allocnos connected to ALLOCNO through copy. */ static void update_costs_from_prefs (ira_allocno_t allocno) { ira_pref_t pref; start_update_cost (); for (pref = ALLOCNO_PREFS (allocno); pref != NULL; pref = pref->next_pref) update_costs_from_allocno (allocno, pref->hard_regno, COST_HOP_DIVISOR, true, true, false); } /* Update (decrease if DECR_P) the cost of allocnos connected to ALLOCNO through copies to increase chances to remove some copies as the result of subsequent assignment. ALLOCNO was just assigned to a hard register. Record cost updates if RECORD_P is true. */ static void update_costs_from_copies (ira_allocno_t allocno, bool decr_p, bool record_p) { int hard_regno; hard_regno = ALLOCNO_HARD_REGNO (allocno); ira_assert (hard_regno >= 0 && ALLOCNO_CLASS (allocno) != NO_REGS); start_update_cost (); update_costs_from_allocno (allocno, hard_regno, 1, decr_p, record_p, false); } /* Restore costs of allocnos connected to ALLOCNO by copies as it was before updating costs of these allocnos from given allocno. This is a wise thing to do as if given allocno did not get an expected hard reg, using smaller cost of the hard reg for allocnos connected by copies to given allocno becomes actually misleading. Free all update cost records for ALLOCNO as we don't need them anymore. */ static void restore_costs_from_copies (ira_allocno_t allocno) { struct update_cost_record *records, *curr; if (ALLOCNO_COLOR_DATA (allocno) == NULL) return; records = ALLOCNO_COLOR_DATA (allocno)->update_cost_records; start_update_cost (); for (curr = records; curr != NULL; curr = curr->next) { update_costs_from_allocno (allocno, curr->hard_regno, curr->divisor, false, false, false); } free_update_cost_record_list (records); ALLOCNO_COLOR_DATA (allocno)->update_cost_records = NULL; } /* If HARDREG is assigned to A, pseudo reg B conflicting with A will not be assigned to HARDREG anymore. It is better to take the fact into consideration of cost propagation. So this func revokes the previous HARDREG related updates originated from B, then propagates the disfavor of the HARDREG to other pseudos connecting with B. Note: We only propagate cost to the copies connecting with B. The updated_hard_reg_costs and updated_conflict_hard_reg_costs of B are not updated here. The impact of hardreg conflicting may be little underestimated compared with the impact of preference (For hardreg preference propagation, updated_hard_reg_costs of A is updated, then B connecting with A will be updated. When assign_hard_reg for A, the update to B will be superimposed on the preference of A). In addition, by updating updated_hard_reg_costs and updated_conflict_hard_reg_costs of B, the impact to improve_allocation is unclear. So for now, updated_hard_reg_costs and updated_conflict_hard_reg_costs of B are not updated. new_conflict_hard_regs is created to prevent using updated_hard_reg_costs and updated_conflict_hard_reg_costs of B in update_conflict_hard_regno_costs func. */ static void restore_costs_from_conflicts (ira_allocno_t a, int hardreg) { int word, nwords; enum machine_mode mode; nwords = ALLOCNO_NUM_OBJECTS (a); mode = ALLOCNO_MODE (a); ira_init_register_move_cost_if_necessary (mode); curr_allocno_process++; for (word = 0; word < nwords; word++) { ira_object_t conflict_obj; ira_object_t obj = ALLOCNO_OBJECT (a, word); ira_object_conflict_iterator oci; /* Update conflicting allocnos. */ FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); struct update_cost_record *curr, *prev, *next; allocno_color_data_t data = ALLOCNO_COLOR_DATA (conflict_a); if (!bitmap_bit_p (consideration_allocno_bitmap, ALLOCNO_NUM (conflict_a)) || ALLOCNO_ASSIGNED_P (conflict_a) || (data == NULL) || (data->last_process == curr_allocno_process)) continue; data->last_process = curr_allocno_process; /* Record that hardreg will not be assigned to conflict_a in ira coloring unless improve_allocation adjusts the overall coloring later. */ SET_HARD_REG_BIT (data->new_conflict_hard_regs, hardreg); start_update_cost (); /* Revoke the hardreg related cost update records saved in ALLOCNO_COLOR_DATA (conflict_a), then remove those records from head of data->update_cost_records list. */ curr = data->update_cost_records; prev = data->update_cost_records; while (curr != NULL && curr->hard_regno == hardreg) { update_costs_from_allocno (conflict_a, curr->hard_regno, curr->divisor, false, false, false); /* Delete current record. */ free_a_update_cost_record (&data->update_cost_records, NULL, curr); curr = data->update_cost_records; prev = data->update_cost_records; } if (curr != NULL) curr = curr->next; /* Revoke the hardreg related cost update records saved in ALLOCNO_COLOR_DATA (conflict_a), then remove those records from the rest of data->update_cost_records list. */ for (; curr != NULL; curr = next) { next = curr->next; if (curr->hard_regno == hardreg) { update_costs_from_allocno (conflict_a, curr->hard_regno, curr->divisor, false, false, false); /* Delete current record. */ free_a_update_cost_record (&data->update_cost_records, prev, curr); } else prev = curr; } /* Propagate the disfavor of hardreg from conflict_a to the allocnos connecting with conflict_a via copies. */ update_costs_from_allocno (conflict_a, hardreg, 1, false, true, true); } } } /* This function updates COSTS (decrease if DECR_P) for hard_registers of ACLASS by conflict costs of the unassigned allocnos connected by copies with allocnos in update_cost_queue. This update increases chances to remove some copies. */ static void update_conflict_hard_regno_costs (int *costs, enum reg_class aclass, bool decr_p) { int i, cost, class_size, freq, mult, div, divisor; int index, hard_regno; int *conflict_costs; bool cont_p; enum reg_class another_aclass; ira_allocno_t allocno, another_allocno; ira_copy_t cp, next_cp, from; while (get_next_update_cost (&allocno, &from, &divisor)) for (cp = ALLOCNO_COPIES (allocno); cp != NULL; cp = next_cp) { if (cp->first == allocno) { next_cp = cp->next_first_allocno_copy; another_allocno = cp->second; } else if (cp->second == allocno) { next_cp = cp->next_second_allocno_copy; another_allocno = cp->first; } else gcc_unreachable (); if (from != NULL && (another_allocno == from->first || another_allocno == from->second)) continue; /* For a = b * c; if there are copies (a, b) and (a, c), it makes no sense to propagate cost from b to a then to c. */ if (from != NULL && from->insn == cp->insn) continue; another_aclass = ALLOCNO_CLASS (another_allocno); if (! ira_reg_classes_intersect_p[aclass][another_aclass] || ALLOCNO_ASSIGNED_P (another_allocno) || ALLOCNO_COLOR_DATA (another_allocno)->may_be_spilled_p) continue; class_size = ira_class_hard_regs_num[another_aclass]; ira_allocate_and_copy_costs (&ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (another_allocno), another_aclass, ALLOCNO_CONFLICT_HARD_REG_COSTS (another_allocno)); conflict_costs = ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (another_allocno); if (conflict_costs == NULL) cont_p = true; else { mult = cp->freq; freq = ALLOCNO_FREQ (another_allocno); if (freq == 0) freq = 1; div = freq * divisor; cont_p = false; for (i = class_size - 1; i >= 0; i--) { hard_regno = ira_class_hard_regs[another_aclass][i]; ira_assert (hard_regno >= 0); index = ira_class_hard_reg_index[aclass][hard_regno]; if (index < 0) continue; /* If it is known that hard_regno will not be assigned to another_allocno, don't use conflict_costs[i] of another_allocno to update costs[i]. */ if (!TEST_HARD_REG_BIT (ALLOCNO_COLOR_DATA (another_allocno) ->profitable_hard_regs, hard_regno) || TEST_HARD_REG_BIT (ALLOCNO_COLOR_DATA (another_allocno) ->new_conflict_hard_regs, hard_regno)) continue; cost = (int) ((unsigned) conflict_costs [i] * mult) / div; if (cost == 0) continue; cont_p = true; if (decr_p) cost = -cost; costs[index] += cost; } } /* Probably 5 hops will be enough. */ if (cont_p && divisor <= (COST_HOP_DIVISOR * COST_HOP_DIVISOR * COST_HOP_DIVISOR * COST_HOP_DIVISOR)) queue_update_cost (another_allocno, cp, divisor * COST_HOP_DIVISOR); } } /* Set up conflicting (through CONFLICT_REGS) for each object of allocno A and the start allocno profitable regs (through START_PROFITABLE_REGS). Remember that the start profitable regs exclude hard regs which can not hold value of mode of allocno A. This covers mostly cases when multi-register value should be aligned. */ static inline void get_conflict_and_start_profitable_regs (ira_allocno_t a, bool retry_p, HARD_REG_SET *conflict_regs, HARD_REG_SET *start_profitable_regs) { int i, nwords; ira_object_t obj; nwords = ALLOCNO_NUM_OBJECTS (a); for (i = 0; i < nwords; i++) { obj = ALLOCNO_OBJECT (a, i); COPY_HARD_REG_SET (conflict_regs[i], OBJECT_TOTAL_CONFLICT_HARD_REGS (obj)); } if (retry_p) { COPY_HARD_REG_SET (*start_profitable_regs, reg_class_contents[ALLOCNO_CLASS (a)]); AND_COMPL_HARD_REG_SET (*start_profitable_regs, ira_prohibited_class_mode_regs [ALLOCNO_CLASS (a)][ALLOCNO_MODE (a)]); } else COPY_HARD_REG_SET (*start_profitable_regs, ALLOCNO_COLOR_DATA (a)->profitable_hard_regs); } /* Return true if HARD_REGNO is ok for assigning to allocno A with PROFITABLE_REGS and whose objects have CONFLICT_REGS. */ static inline bool check_hard_reg_p (ira_allocno_t a, int hard_regno, HARD_REG_SET *conflict_regs, HARD_REG_SET profitable_regs) { int j, nwords, nregs; enum reg_class aclass; enum machine_mode mode; aclass = ALLOCNO_CLASS (a); mode = ALLOCNO_MODE (a); if (TEST_HARD_REG_BIT (ira_prohibited_class_mode_regs[aclass][mode], hard_regno)) return false; /* Checking only profitable hard regs. */ if (! TEST_HARD_REG_BIT (profitable_regs, hard_regno)) return false; nregs = hard_regno_nregs[hard_regno][mode]; nwords = ALLOCNO_NUM_OBJECTS (a); for (j = 0; j < nregs; j++) { int k; int set_to_test_start = 0, set_to_test_end = nwords; if (nregs == nwords) { if (REG_WORDS_BIG_ENDIAN) set_to_test_start = nwords - j - 1; else set_to_test_start = j; set_to_test_end = set_to_test_start + 1; } for (k = set_to_test_start; k < set_to_test_end; k++) if (TEST_HARD_REG_BIT (conflict_regs[k], hard_regno + j)) break; if (k != set_to_test_end) break; } return j == nregs; } #ifndef HONOR_REG_ALLOC_ORDER /* Return number of registers needed to be saved and restored at function prologue/epilogue if we allocate HARD_REGNO to hold value of MODE. */ static int calculate_saved_nregs (int hard_regno, enum machine_mode mode) { int i; int nregs = 0; ira_assert (hard_regno >= 0); for (i = hard_regno_nregs[hard_regno][mode] - 1; i >= 0; i--) if (!allocated_hardreg_p[hard_regno + i] && (!TEST_HARD_REG_BIT (call_used_reg_set, hard_regno + i) || (frame_pointer_partially_needed && (hard_regno + i == HARD_FRAME_POINTER_REGNUM))) && !LOCAL_REGNO (hard_regno + i)) nregs++; return nregs; } #endif /* Choose a hard register for allocno A. If RETRY_P is TRUE, it means that the function called from function `ira_reassign_conflict_allocnos' and `allocno_reload_assign'. In this case some allocno data are not defined or updated and we should not touch these data. The function returns true if we managed to assign a hard register to the allocno. To assign a hard register, first of all we calculate all conflict hard registers which can come from conflicting allocnos with already assigned hard registers. After that we find first free hard register with the minimal cost. During hard register cost calculation we take conflict hard register costs into account to give a chance for conflicting allocnos to get a better hard register in the future. If the best hard register cost is bigger than cost of memory usage for the allocno, we don't assign a hard register to given allocno at all. If we assign a hard register to the allocno, we update costs of the hard register for allocnos connected by copies to improve a chance to coalesce insns represented by the copies when we assign hard registers to the allocnos connected by the copies. */ static bool assign_hard_reg (ira_allocno_t a, bool retry_p) { HARD_REG_SET conflicting_regs[2], profitable_hard_regs; int i, j, hard_regno, best_hard_regno, class_size; int cost, mem_cost, min_cost, full_cost, min_full_cost, nwords, word; int *a_costs; enum reg_class aclass; enum machine_mode mode; static int costs[FIRST_PSEUDO_REGISTER], full_costs[FIRST_PSEUDO_REGISTER]; #ifndef HONOR_REG_ALLOC_ORDER int saved_nregs; enum reg_class rclass; int add_cost; #endif #ifdef STACK_REGS bool no_stack_reg_p; #endif ira_assert (! ALLOCNO_ASSIGNED_P (a)); get_conflict_and_start_profitable_regs (a, retry_p, conflicting_regs, &profitable_hard_regs); aclass = ALLOCNO_CLASS (a); class_size = ira_class_hard_regs_num[aclass]; best_hard_regno = -1; memset (full_costs, 0, sizeof (int) * class_size); mem_cost = 0; memset (costs, 0, sizeof (int) * class_size); memset (full_costs, 0, sizeof (int) * class_size); #ifdef STACK_REGS no_stack_reg_p = false; #endif if (! retry_p) start_update_cost (); mem_cost += ALLOCNO_UPDATED_MEMORY_COST (a); ira_allocate_and_copy_costs (&ALLOCNO_UPDATED_HARD_REG_COSTS (a), aclass, ALLOCNO_HARD_REG_COSTS (a)); a_costs = ALLOCNO_UPDATED_HARD_REG_COSTS (a); #ifdef STACK_REGS no_stack_reg_p = no_stack_reg_p || ALLOCNO_TOTAL_NO_STACK_REG_P (a); #endif cost = ALLOCNO_UPDATED_CLASS_COST (a); for (i = 0; i < class_size; i++) if (a_costs != NULL) { costs[i] += a_costs[i]; full_costs[i] += a_costs[i]; } else { costs[i] += cost; full_costs[i] += cost; } nwords = ALLOCNO_NUM_OBJECTS (a); curr_allocno_process++; for (word = 0; word < nwords; word++) { ira_object_t conflict_obj; ira_object_t obj = ALLOCNO_OBJECT (a, word); ira_object_conflict_iterator oci; /* Take preferences of conflicting allocnos into account. */ FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); enum reg_class conflict_aclass; /* Reload can give another class so we need to check all allocnos. */ if (!retry_p && (!bitmap_bit_p (consideration_allocno_bitmap, ALLOCNO_NUM (conflict_a)) || ((!ALLOCNO_ASSIGNED_P (conflict_a) || ALLOCNO_HARD_REGNO (conflict_a) < 0) && !(hard_reg_set_intersect_p (profitable_hard_regs, ALLOCNO_COLOR_DATA (conflict_a)->profitable_hard_regs))))) continue; conflict_aclass = ALLOCNO_CLASS (conflict_a); ira_assert (ira_reg_classes_intersect_p [aclass][conflict_aclass]); if (ALLOCNO_ASSIGNED_P (conflict_a)) { hard_regno = ALLOCNO_HARD_REGNO (conflict_a); if (hard_regno >= 0 && (ira_hard_reg_set_intersection_p (hard_regno, ALLOCNO_MODE (conflict_a), reg_class_contents[aclass]))) { int n_objects = ALLOCNO_NUM_OBJECTS (conflict_a); int conflict_nregs; mode = ALLOCNO_MODE (conflict_a); conflict_nregs = hard_regno_nregs[hard_regno][mode]; if (conflict_nregs == n_objects && conflict_nregs > 1) { int num = OBJECT_SUBWORD (conflict_obj); if (REG_WORDS_BIG_ENDIAN) SET_HARD_REG_BIT (conflicting_regs[word], hard_regno + n_objects - num - 1); else SET_HARD_REG_BIT (conflicting_regs[word], hard_regno + num); } else IOR_HARD_REG_SET (conflicting_regs[word], ira_reg_mode_hard_regset[hard_regno][mode]); if (hard_reg_set_subset_p (profitable_hard_regs, conflicting_regs[word])) goto fail; } } else if (! retry_p && ! ALLOCNO_COLOR_DATA (conflict_a)->may_be_spilled_p /* Don't process the conflict allocno twice. */ && (ALLOCNO_COLOR_DATA (conflict_a)->last_process != curr_allocno_process)) { int k, *conflict_costs; ALLOCNO_COLOR_DATA (conflict_a)->last_process = curr_allocno_process; ira_allocate_and_copy_costs (&ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (conflict_a), conflict_aclass, ALLOCNO_CONFLICT_HARD_REG_COSTS (conflict_a)); conflict_costs = ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (conflict_a); if (conflict_costs != NULL) for (j = class_size - 1; j >= 0; j--) { hard_regno = ira_class_hard_regs[aclass][j]; ira_assert (hard_regno >= 0); k = ira_class_hard_reg_index[conflict_aclass][hard_regno]; if (k < 0) continue; /* If it is known that hard_regno will not be assigned to conflict_a, don't use conflict_costs[i] of conflict_a to update full_costs[i]. */ if (!TEST_HARD_REG_BIT (ALLOCNO_COLOR_DATA (conflict_a) ->profitable_hard_regs, hard_regno) || TEST_HARD_REG_BIT (ALLOCNO_COLOR_DATA (conflict_a) ->new_conflict_hard_regs, hard_regno)) continue; full_costs[j] -= conflict_costs[k]; } queue_update_cost (conflict_a, NULL, COST_HOP_DIVISOR); } } } if (! retry_p) /* Take into account preferences of allocnos connected by copies to the conflict allocnos. */ update_conflict_hard_regno_costs (full_costs, aclass, true); /* Take preferences of allocnos connected by copies into account. */ if (! retry_p) { start_update_cost (); queue_update_cost (a, NULL, COST_HOP_DIVISOR); update_conflict_hard_regno_costs (full_costs, aclass, false); } min_cost = min_full_cost = INT_MAX; /* We don't care about giving callee saved registers to allocnos no living through calls because call clobbered registers are allocated first (it is usual practice to put them first in REG_ALLOC_ORDER). */ mode = ALLOCNO_MODE (a); for (i = 0; i < class_size; i++) { hard_regno = ira_class_hard_regs[aclass][i]; #ifdef STACK_REGS if (no_stack_reg_p && FIRST_STACK_REG <= hard_regno && hard_regno <= LAST_STACK_REG) continue; #endif if (! check_hard_reg_p (a, hard_regno, conflicting_regs, profitable_hard_regs)) continue; cost = costs[i]; full_cost = full_costs[i]; #ifndef HONOR_REG_ALLOC_ORDER if ((saved_nregs = calculate_saved_nregs (hard_regno, mode)) != 0) /* We need to save/restore the hard register in epilogue/prologue. Therefore we increase the cost. */ { rclass = REGNO_REG_CLASS (hard_regno); add_cost = ((ira_memory_move_cost[mode][rclass][0] + ira_memory_move_cost[mode][rclass][1]) * saved_nregs / hard_regno_nregs[hard_regno][mode] - 1); cost += add_cost; full_cost += add_cost; } #endif if (min_cost > cost) min_cost = cost; if (min_full_cost > full_cost) { min_full_cost = full_cost; best_hard_regno = hard_regno; ira_assert (hard_regno >= 0); } } if (min_full_cost > mem_cost) { if (! retry_p && internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "(memory is more profitable %d vs %d) ", mem_cost, min_full_cost); best_hard_regno = -1; } fail: if (best_hard_regno >= 0) { for (i = hard_regno_nregs[best_hard_regno][mode] - 1; i >= 0; i--) allocated_hardreg_p[best_hard_regno + i] = true; } if (! retry_p) restore_costs_from_copies (a); ALLOCNO_HARD_REGNO (a) = best_hard_regno; ALLOCNO_ASSIGNED_P (a) = true; if (best_hard_regno >= 0) update_costs_from_copies (a, true, ! retry_p); if (! retry_p && (best_hard_regno >= 0)) restore_costs_from_conflicts (a, best_hard_regno); ira_assert (ALLOCNO_CLASS (a) == aclass); /* We don't need updated costs anymore: */ ira_free_allocno_updated_costs (a); return best_hard_regno >= 0; } /* An array used to sort copies. */ static ira_copy_t *sorted_copies; /* Return TRUE if live ranges of allocnos A1 and A2 intersect. It is used to find a conflict for new allocnos or allocnos with the different allocno classes. */ static bool allocnos_conflict_by_live_ranges_p (ira_allocno_t a1, ira_allocno_t a2) { rtx reg1, reg2; int i, j; int n1 = ALLOCNO_NUM_OBJECTS (a1); int n2 = ALLOCNO_NUM_OBJECTS (a2); if (a1 == a2) return false; reg1 = regno_reg_rtx[ALLOCNO_REGNO (a1)]; reg2 = regno_reg_rtx[ALLOCNO_REGNO (a2)]; if (reg1 != NULL && reg2 != NULL && ORIGINAL_REGNO (reg1) == ORIGINAL_REGNO (reg2)) return false; for (i = 0; i < n1; i++) { ira_object_t c1 = ALLOCNO_OBJECT (a1, i); for (j = 0; j < n2; j++) { ira_object_t c2 = ALLOCNO_OBJECT (a2, j); if (ira_live_ranges_intersect_p (OBJECT_LIVE_RANGES (c1), OBJECT_LIVE_RANGES (c2))) return true; } } return false; } /* The function is used to sort copies according to their execution frequencies. */ static int copy_freq_compare_func (const void *v1p, const void *v2p) { ira_copy_t cp1 = *(const ira_copy_t *) v1p, cp2 = *(const ira_copy_t *) v2p; int pri1, pri2; pri1 = cp1->freq; pri2 = cp2->freq; if (pri2 - pri1) return pri2 - pri1; /* If freqencies are equal, sort by copies, so that the results of qsort leave nothing to chance. */ return cp1->num - cp2->num; } /* Return true if any allocno from thread of A1 conflicts with any allocno from thread A2. */ static bool allocno_thread_conflict_p (ira_allocno_t a1, ira_allocno_t a2) { ira_allocno_t a, conflict_a; for (a = ALLOCNO_COLOR_DATA (a2)->next_thread_allocno;; a = ALLOCNO_COLOR_DATA (a)->next_thread_allocno) { for (conflict_a = ALLOCNO_COLOR_DATA (a1)->next_thread_allocno;; conflict_a = ALLOCNO_COLOR_DATA (conflict_a)->next_thread_allocno) { if (allocnos_conflict_by_live_ranges_p (a, conflict_a)) return true; if (conflict_a == a1) break; } if (a == a2) break; } return false; } /* Merge two threads given correspondingly by their first allocnos T1 and T2 (more accurately merging T2 into T1). */ static void merge_threads (ira_allocno_t t1, ira_allocno_t t2) { ira_allocno_t a, next, last; gcc_assert (t1 != t2 && ALLOCNO_COLOR_DATA (t1)->first_thread_allocno == t1 && ALLOCNO_COLOR_DATA (t2)->first_thread_allocno == t2); for (last = t2, a = ALLOCNO_COLOR_DATA (t2)->next_thread_allocno;; a = ALLOCNO_COLOR_DATA (a)->next_thread_allocno) { ALLOCNO_COLOR_DATA (a)->first_thread_allocno = t1; if (a == t2) break; last = a; } next = ALLOCNO_COLOR_DATA (t1)->next_thread_allocno; ALLOCNO_COLOR_DATA (t1)->next_thread_allocno = t2; ALLOCNO_COLOR_DATA (last)->next_thread_allocno = next; ALLOCNO_COLOR_DATA (t1)->thread_freq += ALLOCNO_COLOR_DATA (t2)->thread_freq; } /* Create threads by processing CP_NUM copies from sorted)ciopeis. We process the most expensive copies first. */ static void form_threads_from_copies (int cp_num) { ira_allocno_t a, thread1, thread2; ira_copy_t cp; int i, n; qsort (sorted_copies, cp_num, sizeof (ira_copy_t), copy_freq_compare_func); /* Form threads processing copies, most frequently executed first. */ for (; cp_num != 0;) { for (i = 0; i < cp_num; i++) { cp = sorted_copies[i]; thread1 = ALLOCNO_COLOR_DATA (cp->first)->first_thread_allocno; thread2 = ALLOCNO_COLOR_DATA (cp->second)->first_thread_allocno; if (thread1 == thread2) continue; if (! allocno_thread_conflict_p (thread1, thread2)) { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, " Forming thread by copy %d:a%dr%d-a%dr%d (freq=%d):\n", cp->num, ALLOCNO_NUM (cp->first), ALLOCNO_REGNO (cp->first), ALLOCNO_NUM (cp->second), ALLOCNO_REGNO (cp->second), cp->freq); merge_threads (thread1, thread2); if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { thread1 = ALLOCNO_COLOR_DATA (thread1)->first_thread_allocno; fprintf (ira_dump_file, " Result (freq=%d): a%dr%d(%d)", ALLOCNO_COLOR_DATA (thread1)->thread_freq, ALLOCNO_NUM (thread1), ALLOCNO_REGNO (thread1), ALLOCNO_FREQ (thread1)); for (a = ALLOCNO_COLOR_DATA (thread1)->next_thread_allocno; a != thread1; a = ALLOCNO_COLOR_DATA (a)->next_thread_allocno) fprintf (ira_dump_file, " a%dr%d(%d)", ALLOCNO_NUM (a), ALLOCNO_REGNO (a), ALLOCNO_FREQ (a)); fprintf (ira_dump_file, "\n"); } i++; break; } } /* Collect the rest of copies. */ for (n = 0; i < cp_num; i++) { cp = sorted_copies[i]; if (ALLOCNO_COLOR_DATA (cp->first)->first_thread_allocno != ALLOCNO_COLOR_DATA (cp->second)->first_thread_allocno) sorted_copies[n++] = cp; } cp_num = n; } } /* Create threads by processing copies of all alocnos from BUCKET. We process the most expensive copies first. */ static void form_threads_from_bucket (ira_allocno_t bucket) { ira_allocno_t a; ira_copy_t cp, next_cp; int cp_num = 0; for (a = bucket; a != NULL; a = ALLOCNO_COLOR_DATA (a)->next_bucket_allocno) { for (cp = ALLOCNO_COPIES (a); cp != NULL; cp = next_cp) { if (cp->first == a) { next_cp = cp->next_first_allocno_copy; sorted_copies[cp_num++] = cp; } else if (cp->second == a) next_cp = cp->next_second_allocno_copy; else gcc_unreachable (); } } form_threads_from_copies (cp_num); } /* Create threads by processing copies of colorable allocno A. We process most expensive copies first. */ static void form_threads_from_colorable_allocno (ira_allocno_t a) { ira_allocno_t another_a; ira_copy_t cp, next_cp; int cp_num = 0; for (cp = ALLOCNO_COPIES (a); cp != NULL; cp = next_cp) { if (cp->first == a) { next_cp = cp->next_first_allocno_copy; another_a = cp->second; } else if (cp->second == a) { next_cp = cp->next_second_allocno_copy; another_a = cp->first; } else gcc_unreachable (); if ((! ALLOCNO_COLOR_DATA (another_a)->in_graph_p && !ALLOCNO_COLOR_DATA (another_a)->may_be_spilled_p) || ALLOCNO_COLOR_DATA (another_a)->colorable_p) sorted_copies[cp_num++] = cp; } form_threads_from_copies (cp_num); } /* Form initial threads which contain only one allocno. */ static void init_allocno_threads (void) { ira_allocno_t a; unsigned int j; bitmap_iterator bi; EXECUTE_IF_SET_IN_BITMAP (consideration_allocno_bitmap, 0, j, bi) { a = ira_allocnos[j]; /* Set up initial thread data: */ ALLOCNO_COLOR_DATA (a)->first_thread_allocno = ALLOCNO_COLOR_DATA (a)->next_thread_allocno = a; ALLOCNO_COLOR_DATA (a)->thread_freq = ALLOCNO_FREQ (a); } } /* This page contains the allocator based on the Chaitin-Briggs algorithm. */ /* Bucket of allocnos that can colored currently without spilling. */ static ira_allocno_t colorable_allocno_bucket; /* Bucket of allocnos that might be not colored currently without spilling. */ static ira_allocno_t uncolorable_allocno_bucket; /* The current number of allocnos in the uncolorable_bucket. */ static int uncolorable_allocnos_num; /* Return the current spill priority of allocno A. The less the number, the more preferable the allocno for spilling. */ static inline int allocno_spill_priority (ira_allocno_t a) { allocno_color_data_t data = ALLOCNO_COLOR_DATA (a); return (data->temp / (ALLOCNO_EXCESS_PRESSURE_POINTS_NUM (a) * ira_reg_class_max_nregs[ALLOCNO_CLASS (a)][ALLOCNO_MODE (a)] + 1)); } /* Add allocno A to bucket *BUCKET_PTR. A should be not in a bucket before the call. */ static void add_allocno_to_bucket (ira_allocno_t a, ira_allocno_t *bucket_ptr) { ira_allocno_t first_a; allocno_color_data_t data; if (bucket_ptr == &uncolorable_allocno_bucket && ALLOCNO_CLASS (a) != NO_REGS) { uncolorable_allocnos_num++; ira_assert (uncolorable_allocnos_num > 0); } first_a = *bucket_ptr; data = ALLOCNO_COLOR_DATA (a); data->next_bucket_allocno = first_a; data->prev_bucket_allocno = NULL; if (first_a != NULL) ALLOCNO_COLOR_DATA (first_a)->prev_bucket_allocno = a; *bucket_ptr = a; } /* Compare two allocnos to define which allocno should be pushed first into the coloring stack. If the return is a negative number, the allocno given by the first parameter will be pushed first. In this case such allocno has less priority than the second one and the hard register will be assigned to it after assignment to the second one. As the result of such assignment order, the second allocno has a better chance to get the best hard register. */ static int bucket_allocno_compare_func (const void *v1p, const void *v2p) { ira_allocno_t a1 = *(const ira_allocno_t *) v1p; ira_allocno_t a2 = *(const ira_allocno_t *) v2p; int diff, freq1, freq2, a1_num, a2_num; ira_allocno_t t1 = ALLOCNO_COLOR_DATA (a1)->first_thread_allocno; ira_allocno_t t2 = ALLOCNO_COLOR_DATA (a2)->first_thread_allocno; int cl1 = ALLOCNO_CLASS (a1), cl2 = ALLOCNO_CLASS (a2); freq1 = ALLOCNO_COLOR_DATA (t1)->thread_freq; freq2 = ALLOCNO_COLOR_DATA (t2)->thread_freq; if ((diff = freq1 - freq2) != 0) return diff; if ((diff = ALLOCNO_NUM (t2) - ALLOCNO_NUM (t1)) != 0) return diff; /* Push pseudos requiring less hard registers first. It means that we will assign pseudos requiring more hard registers first avoiding creation small holes in free hard register file into which the pseudos requiring more hard registers can not fit. */ if ((diff = (ira_reg_class_max_nregs[cl1][ALLOCNO_MODE (a1)] - ira_reg_class_max_nregs[cl2][ALLOCNO_MODE (a2)])) != 0) return diff; freq1 = ALLOCNO_FREQ (a1); freq2 = ALLOCNO_FREQ (a2); if ((diff = freq1 - freq2) != 0) return diff; a1_num = ALLOCNO_COLOR_DATA (a1)->available_regs_num; a2_num = ALLOCNO_COLOR_DATA (a2)->available_regs_num; if ((diff = a2_num - a1_num) != 0) return diff; return ALLOCNO_NUM (a2) - ALLOCNO_NUM (a1); } /* Sort bucket *BUCKET_PTR and return the result through BUCKET_PTR. */ static void sort_bucket (ira_allocno_t *bucket_ptr, int (*compare_func) (const void *, const void *)) { ira_allocno_t a, head; int n; for (n = 0, a = *bucket_ptr; a != NULL; a = ALLOCNO_COLOR_DATA (a)->next_bucket_allocno) sorted_allocnos[n++] = a; if (n <= 1) return; qsort (sorted_allocnos, n, sizeof (ira_allocno_t), compare_func); head = NULL; for (n--; n >= 0; n--) { a = sorted_allocnos[n]; ALLOCNO_COLOR_DATA (a)->next_bucket_allocno = head; ALLOCNO_COLOR_DATA (a)->prev_bucket_allocno = NULL; if (head != NULL) ALLOCNO_COLOR_DATA (head)->prev_bucket_allocno = a; head = a; } *bucket_ptr = head; } /* Add ALLOCNO to colorable bucket maintaining the order according their priority. ALLOCNO should be not in a bucket before the call. */ static void add_allocno_to_ordered_colorable_bucket (ira_allocno_t allocno) { ira_allocno_t before, after; form_threads_from_colorable_allocno (allocno); for (before = colorable_allocno_bucket, after = NULL; before != NULL; after = before, before = ALLOCNO_COLOR_DATA (before)->next_bucket_allocno) if (bucket_allocno_compare_func (&allocno, &before) < 0) break; ALLOCNO_COLOR_DATA (allocno)->next_bucket_allocno = before; ALLOCNO_COLOR_DATA (allocno)->prev_bucket_allocno = after; if (after == NULL) colorable_allocno_bucket = allocno; else ALLOCNO_COLOR_DATA (after)->next_bucket_allocno = allocno; if (before != NULL) ALLOCNO_COLOR_DATA (before)->prev_bucket_allocno = allocno; } /* Delete ALLOCNO from bucket *BUCKET_PTR. It should be there before the call. */ static void delete_allocno_from_bucket (ira_allocno_t allocno, ira_allocno_t *bucket_ptr) { ira_allocno_t prev_allocno, next_allocno; if (bucket_ptr == &uncolorable_allocno_bucket && ALLOCNO_CLASS (allocno) != NO_REGS) { uncolorable_allocnos_num--; ira_assert (uncolorable_allocnos_num >= 0); } prev_allocno = ALLOCNO_COLOR_DATA (allocno)->prev_bucket_allocno; next_allocno = ALLOCNO_COLOR_DATA (allocno)->next_bucket_allocno; if (prev_allocno != NULL) ALLOCNO_COLOR_DATA (prev_allocno)->next_bucket_allocno = next_allocno; else { ira_assert (*bucket_ptr == allocno); *bucket_ptr = next_allocno; } if (next_allocno != NULL) ALLOCNO_COLOR_DATA (next_allocno)->prev_bucket_allocno = prev_allocno; } /* Put allocno A onto the coloring stack without removing it from its bucket. Pushing allocno to the coloring stack can result in moving conflicting allocnos from the uncolorable bucket to the colorable one. */ static void push_allocno_to_stack (ira_allocno_t a) { enum reg_class aclass; allocno_color_data_t data, conflict_data; int size, i, n = ALLOCNO_NUM_OBJECTS (a); data = ALLOCNO_COLOR_DATA (a); data->in_graph_p = false; allocno_stack_vec.safe_push (a); aclass = ALLOCNO_CLASS (a); if (aclass == NO_REGS) return; size = ira_reg_class_max_nregs[aclass][ALLOCNO_MODE (a)]; if (n > 1) { /* We will deal with the subwords individually. */ gcc_assert (size == ALLOCNO_NUM_OBJECTS (a)); size = 1; } for (i = 0; i < n; i++) { ira_object_t obj = ALLOCNO_OBJECT (a, i); ira_object_t conflict_obj; ira_object_conflict_iterator oci; FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); conflict_data = ALLOCNO_COLOR_DATA (conflict_a); if (conflict_data->colorable_p || ! conflict_data->in_graph_p || ALLOCNO_ASSIGNED_P (conflict_a) || !(hard_reg_set_intersect_p (ALLOCNO_COLOR_DATA (a)->profitable_hard_regs, conflict_data->profitable_hard_regs))) continue; ira_assert (bitmap_bit_p (coloring_allocno_bitmap, ALLOCNO_NUM (conflict_a))); if (update_left_conflict_sizes_p (conflict_a, a, size)) { delete_allocno_from_bucket (conflict_a, &uncolorable_allocno_bucket); add_allocno_to_ordered_colorable_bucket (conflict_a); if (internal_flag_ira_verbose > 4 && ira_dump_file != NULL) { fprintf (ira_dump_file, " Making"); ira_print_expanded_allocno (conflict_a); fprintf (ira_dump_file, " colorable\n"); } } } } } /* Put ALLOCNO onto the coloring stack and remove it from its bucket. The allocno is in the colorable bucket if COLORABLE_P is TRUE. */ static void remove_allocno_from_bucket_and_push (ira_allocno_t allocno, bool colorable_p) { if (colorable_p) delete_allocno_from_bucket (allocno, &colorable_allocno_bucket); else delete_allocno_from_bucket (allocno, &uncolorable_allocno_bucket); if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { fprintf (ira_dump_file, " Pushing"); ira_print_expanded_allocno (allocno); if (colorable_p) fprintf (ira_dump_file, "(cost %d)\n", ALLOCNO_COLOR_DATA (allocno)->temp); else fprintf (ira_dump_file, "(potential spill: %spri=%d, cost=%d)\n", ALLOCNO_BAD_SPILL_P (allocno) ? "bad spill, " : "", allocno_spill_priority (allocno), ALLOCNO_COLOR_DATA (allocno)->temp); } if (! colorable_p) ALLOCNO_COLOR_DATA (allocno)->may_be_spilled_p = true; push_allocno_to_stack (allocno); } /* Put all allocnos from colorable bucket onto the coloring stack. */ static void push_only_colorable (void) { form_threads_from_bucket (colorable_allocno_bucket); sort_bucket (&colorable_allocno_bucket, bucket_allocno_compare_func); for (;colorable_allocno_bucket != NULL;) remove_allocno_from_bucket_and_push (colorable_allocno_bucket, true); } /* Return the frequency of exit edges (if EXIT_P) or entry from/to the loop given by its LOOP_NODE. */ int ira_loop_edge_freq (ira_loop_tree_node_t loop_node, int regno, bool exit_p) { int freq, i; edge_iterator ei; edge e; vec edges; ira_assert (current_loops != NULL && loop_node->loop != NULL && (regno < 0 || regno >= FIRST_PSEUDO_REGISTER)); freq = 0; if (! exit_p) { FOR_EACH_EDGE (e, ei, loop_node->loop->header->preds) if (e->src != loop_node->loop->latch && (regno < 0 || (bitmap_bit_p (df_get_live_out (e->src), regno) && bitmap_bit_p (df_get_live_in (e->dest), regno)))) freq += EDGE_FREQUENCY (e); } else { edges = get_loop_exit_edges (loop_node->loop); FOR_EACH_VEC_ELT (edges, i, e) if (regno < 0 || (bitmap_bit_p (df_get_live_out (e->src), regno) && bitmap_bit_p (df_get_live_in (e->dest), regno))) freq += EDGE_FREQUENCY (e); edges.release (); } return REG_FREQ_FROM_EDGE_FREQ (freq); } /* Calculate and return the cost of putting allocno A into memory. */ static int calculate_allocno_spill_cost (ira_allocno_t a) { int regno, cost; enum machine_mode mode; enum reg_class rclass; ira_allocno_t parent_allocno; ira_loop_tree_node_t parent_node, loop_node; regno = ALLOCNO_REGNO (a); cost = ALLOCNO_UPDATED_MEMORY_COST (a) - ALLOCNO_UPDATED_CLASS_COST (a); if (ALLOCNO_CAP (a) != NULL) return cost; loop_node = ALLOCNO_LOOP_TREE_NODE (a); if ((parent_node = loop_node->parent) == NULL) return cost; if ((parent_allocno = parent_node->regno_allocno_map[regno]) == NULL) return cost; mode = ALLOCNO_MODE (a); rclass = ALLOCNO_CLASS (a); if (ALLOCNO_HARD_REGNO (parent_allocno) < 0) cost -= (ira_memory_move_cost[mode][rclass][0] * ira_loop_edge_freq (loop_node, regno, true) + ira_memory_move_cost[mode][rclass][1] * ira_loop_edge_freq (loop_node, regno, false)); else { ira_init_register_move_cost_if_necessary (mode); cost += ((ira_memory_move_cost[mode][rclass][1] * ira_loop_edge_freq (loop_node, regno, true) + ira_memory_move_cost[mode][rclass][0] * ira_loop_edge_freq (loop_node, regno, false)) - (ira_register_move_cost[mode][rclass][rclass] * (ira_loop_edge_freq (loop_node, regno, false) + ira_loop_edge_freq (loop_node, regno, true)))); } return cost; } /* Used for sorting allocnos for spilling. */ static inline int allocno_spill_priority_compare (ira_allocno_t a1, ira_allocno_t a2) { int pri1, pri2, diff; if (ALLOCNO_BAD_SPILL_P (a1) && ! ALLOCNO_BAD_SPILL_P (a2)) return 1; if (ALLOCNO_BAD_SPILL_P (a2) && ! ALLOCNO_BAD_SPILL_P (a1)) return -1; pri1 = allocno_spill_priority (a1); pri2 = allocno_spill_priority (a2); if ((diff = pri1 - pri2) != 0) return diff; if ((diff = ALLOCNO_COLOR_DATA (a1)->temp - ALLOCNO_COLOR_DATA (a2)->temp) != 0) return diff; return ALLOCNO_NUM (a1) - ALLOCNO_NUM (a2); } /* Used for sorting allocnos for spilling. */ static int allocno_spill_sort_compare (const void *v1p, const void *v2p) { ira_allocno_t p1 = *(const ira_allocno_t *) v1p; ira_allocno_t p2 = *(const ira_allocno_t *) v2p; return allocno_spill_priority_compare (p1, p2); } /* Push allocnos to the coloring stack. The order of allocnos in the stack defines the order for the subsequent coloring. */ static void push_allocnos_to_stack (void) { ira_allocno_t a; int cost; /* Calculate uncolorable allocno spill costs. */ for (a = uncolorable_allocno_bucket; a != NULL; a = ALLOCNO_COLOR_DATA (a)->next_bucket_allocno) if (ALLOCNO_CLASS (a) != NO_REGS) { cost = calculate_allocno_spill_cost (a); /* ??? Remove cost of copies between the coalesced allocnos. */ ALLOCNO_COLOR_DATA (a)->temp = cost; } sort_bucket (&uncolorable_allocno_bucket, allocno_spill_sort_compare); for (;;) { push_only_colorable (); a = uncolorable_allocno_bucket; if (a == NULL) break; remove_allocno_from_bucket_and_push (a, false); } ira_assert (colorable_allocno_bucket == NULL && uncolorable_allocno_bucket == NULL); ira_assert (uncolorable_allocnos_num == 0); } /* Pop the coloring stack and assign hard registers to the popped allocnos. */ static void pop_allocnos_from_stack (void) { ira_allocno_t allocno; enum reg_class aclass; for (;allocno_stack_vec.length () != 0;) { allocno = allocno_stack_vec.pop (); aclass = ALLOCNO_CLASS (allocno); if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { fprintf (ira_dump_file, " Popping"); ira_print_expanded_allocno (allocno); fprintf (ira_dump_file, " -- "); } if (aclass == NO_REGS) { ALLOCNO_HARD_REGNO (allocno) = -1; ALLOCNO_ASSIGNED_P (allocno) = true; ira_assert (ALLOCNO_UPDATED_HARD_REG_COSTS (allocno) == NULL); ira_assert (ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (allocno) == NULL); if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "assign memory\n"); } else if (assign_hard_reg (allocno, false)) { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "assign reg %d\n", ALLOCNO_HARD_REGNO (allocno)); } else if (ALLOCNO_ASSIGNED_P (allocno)) { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "spill%s\n", ALLOCNO_COLOR_DATA (allocno)->may_be_spilled_p ? "" : "!"); } ALLOCNO_COLOR_DATA (allocno)->in_graph_p = true; } } /* Set up number of available hard registers for allocno A. */ static void setup_allocno_available_regs_num (ira_allocno_t a) { int i, n, hard_regno, hard_regs_num, nwords; enum reg_class aclass; allocno_color_data_t data; aclass = ALLOCNO_CLASS (a); data = ALLOCNO_COLOR_DATA (a); data->available_regs_num = 0; if (aclass == NO_REGS) return; hard_regs_num = ira_class_hard_regs_num[aclass]; nwords = ALLOCNO_NUM_OBJECTS (a); for (n = 0, i = hard_regs_num - 1; i >= 0; i--) { hard_regno = ira_class_hard_regs[aclass][i]; /* Checking only profitable hard regs. */ if (TEST_HARD_REG_BIT (data->profitable_hard_regs, hard_regno)) n++; } data->available_regs_num = n; if (internal_flag_ira_verbose <= 2 || ira_dump_file == NULL) return; fprintf (ira_dump_file, " Allocno a%dr%d of %s(%d) has %d avail. regs ", ALLOCNO_NUM (a), ALLOCNO_REGNO (a), reg_class_names[aclass], ira_class_hard_regs_num[aclass], n); print_hard_reg_set (ira_dump_file, data->profitable_hard_regs, false); fprintf (ira_dump_file, ", %snode: ", hard_reg_set_equal_p (data->profitable_hard_regs, data->hard_regs_node->hard_regs->set) ? "" : "^"); print_hard_reg_set (ira_dump_file, data->hard_regs_node->hard_regs->set, false); for (i = 0; i < nwords; i++) { ira_object_t obj = ALLOCNO_OBJECT (a, i); if (nwords != 1) { if (i != 0) fprintf (ira_dump_file, ", "); fprintf (ira_dump_file, " obj %d", i); } fprintf (ira_dump_file, " (confl regs = "); print_hard_reg_set (ira_dump_file, OBJECT_TOTAL_CONFLICT_HARD_REGS (obj), false); fprintf (ira_dump_file, ")"); } fprintf (ira_dump_file, "\n"); } /* Put ALLOCNO in a bucket corresponding to its number and size of its conflicting allocnos and hard registers. */ static void put_allocno_into_bucket (ira_allocno_t allocno) { ALLOCNO_COLOR_DATA (allocno)->in_graph_p = true; setup_allocno_available_regs_num (allocno); if (setup_left_conflict_sizes_p (allocno)) add_allocno_to_bucket (allocno, &colorable_allocno_bucket); else add_allocno_to_bucket (allocno, &uncolorable_allocno_bucket); } /* Map: allocno number -> allocno priority. */ static int *allocno_priorities; /* Set up priorities for N allocnos in array CONSIDERATION_ALLOCNOS. */ static void setup_allocno_priorities (ira_allocno_t *consideration_allocnos, int n) { int i, length, nrefs, priority, max_priority, mult; ira_allocno_t a; max_priority = 0; for (i = 0; i < n; i++) { a = consideration_allocnos[i]; nrefs = ALLOCNO_NREFS (a); ira_assert (nrefs >= 0); mult = floor_log2 (ALLOCNO_NREFS (a)) + 1; ira_assert (mult >= 0); allocno_priorities[ALLOCNO_NUM (a)] = priority = (mult * (ALLOCNO_MEMORY_COST (a) - ALLOCNO_CLASS_COST (a)) * ira_reg_class_max_nregs[ALLOCNO_CLASS (a)][ALLOCNO_MODE (a)]); if (priority < 0) priority = -priority; if (max_priority < priority) max_priority = priority; } mult = max_priority == 0 ? 1 : INT_MAX / max_priority; for (i = 0; i < n; i++) { a = consideration_allocnos[i]; length = ALLOCNO_EXCESS_PRESSURE_POINTS_NUM (a); if (ALLOCNO_NUM_OBJECTS (a) > 1) length /= ALLOCNO_NUM_OBJECTS (a); if (length <= 0) length = 1; allocno_priorities[ALLOCNO_NUM (a)] = allocno_priorities[ALLOCNO_NUM (a)] * mult / length; } } /* Sort allocnos according to the profit of usage of a hard register instead of memory for them. */ static int allocno_cost_compare_func (const void *v1p, const void *v2p) { ira_allocno_t p1 = *(const ira_allocno_t *) v1p; ira_allocno_t p2 = *(const ira_allocno_t *) v2p; int c1, c2; c1 = ALLOCNO_UPDATED_MEMORY_COST (p1) - ALLOCNO_UPDATED_CLASS_COST (p1); c2 = ALLOCNO_UPDATED_MEMORY_COST (p2) - ALLOCNO_UPDATED_CLASS_COST (p2); if (c1 - c2) return c1 - c2; /* If regs are equally good, sort by allocno numbers, so that the results of qsort leave nothing to chance. */ return ALLOCNO_NUM (p1) - ALLOCNO_NUM (p2); } /* We used Chaitin-Briggs coloring to assign as many pseudos as possible to hard registers. Let us try to improve allocation with cost point of view. This function improves the allocation by spilling some allocnos and assigning the freed hard registers to other allocnos if it decreases the overall allocation cost. */ static void improve_allocation (void) { unsigned int i; int j, k, n, hregno, conflict_hregno, base_cost, class_size, word, nwords; int check, spill_cost, min_cost, nregs, conflict_nregs, r, best; bool try_p; enum reg_class aclass; enum machine_mode mode; int *allocno_costs; int costs[FIRST_PSEUDO_REGISTER]; HARD_REG_SET conflicting_regs[2], profitable_hard_regs; ira_allocno_t a; bitmap_iterator bi; /* Clear counts used to process conflicting allocnos only once for each allocno. */ EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) ALLOCNO_COLOR_DATA (ira_allocnos[i])->temp = 0; check = n = 0; /* Process each allocno and try to assign a hard register to it by spilling some its conflicting allocnos. */ EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; ALLOCNO_COLOR_DATA (a)->temp = 0; if (empty_profitable_hard_regs (a)) continue; check++; aclass = ALLOCNO_CLASS (a); allocno_costs = ALLOCNO_UPDATED_HARD_REG_COSTS (a); if (allocno_costs == NULL) allocno_costs = ALLOCNO_HARD_REG_COSTS (a); if ((hregno = ALLOCNO_HARD_REGNO (a)) < 0) base_cost = ALLOCNO_UPDATED_MEMORY_COST (a); else if (allocno_costs == NULL) /* It means that assigning a hard register is not profitable (we don't waste memory for hard register costs in this case). */ continue; else base_cost = allocno_costs[ira_class_hard_reg_index[aclass][hregno]]; try_p = false; get_conflict_and_start_profitable_regs (a, false, conflicting_regs, &profitable_hard_regs); class_size = ira_class_hard_regs_num[aclass]; /* Set up cost improvement for usage of each profitable hard register for allocno A. */ for (j = 0; j < class_size; j++) { hregno = ira_class_hard_regs[aclass][j]; if (! check_hard_reg_p (a, hregno, conflicting_regs, profitable_hard_regs)) continue; ira_assert (ira_class_hard_reg_index[aclass][hregno] == j); k = allocno_costs == NULL ? 0 : j; costs[hregno] = (allocno_costs == NULL ? ALLOCNO_UPDATED_CLASS_COST (a) : allocno_costs[k]); costs[hregno] -= base_cost; if (costs[hregno] < 0) try_p = true; } if (! try_p) /* There is no chance to improve the allocation cost by assigning hard register to allocno A even without spilling conflicting allocnos. */ continue; mode = ALLOCNO_MODE (a); nwords = ALLOCNO_NUM_OBJECTS (a); /* Process each allocno conflicting with A and update the cost improvement for profitable hard registers of A. To use a hard register for A we need to spill some conflicting allocnos and that creates penalty for the cost improvement. */ for (word = 0; word < nwords; word++) { ira_object_t conflict_obj; ira_object_t obj = ALLOCNO_OBJECT (a, word); ira_object_conflict_iterator oci; FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); if (ALLOCNO_COLOR_DATA (conflict_a)->temp == check) /* We already processed this conflicting allocno because we processed earlier another object of the conflicting allocno. */ continue; ALLOCNO_COLOR_DATA (conflict_a)->temp = check; if ((conflict_hregno = ALLOCNO_HARD_REGNO (conflict_a)) < 0) continue; spill_cost = ALLOCNO_UPDATED_MEMORY_COST (conflict_a); k = (ira_class_hard_reg_index [ALLOCNO_CLASS (conflict_a)][conflict_hregno]); ira_assert (k >= 0); if ((allocno_costs = ALLOCNO_UPDATED_HARD_REG_COSTS (conflict_a)) != NULL) spill_cost -= allocno_costs[k]; else if ((allocno_costs = ALLOCNO_HARD_REG_COSTS (conflict_a)) != NULL) spill_cost -= allocno_costs[k]; else spill_cost -= ALLOCNO_UPDATED_CLASS_COST (conflict_a); conflict_nregs = hard_regno_nregs[conflict_hregno][ALLOCNO_MODE (conflict_a)]; for (r = conflict_hregno; r >= 0 && r + hard_regno_nregs[r][mode] > conflict_hregno; r--) if (check_hard_reg_p (a, r, conflicting_regs, profitable_hard_regs)) costs[r] += spill_cost; for (r = conflict_hregno + 1; r < conflict_hregno + conflict_nregs; r++) if (check_hard_reg_p (a, r, conflicting_regs, profitable_hard_regs)) costs[r] += spill_cost; } } min_cost = INT_MAX; best = -1; /* Now we choose hard register for A which results in highest allocation cost improvement. */ for (j = 0; j < class_size; j++) { hregno = ira_class_hard_regs[aclass][j]; if (check_hard_reg_p (a, hregno, conflicting_regs, profitable_hard_regs) && min_cost > costs[hregno]) { best = hregno; min_cost = costs[hregno]; } } if (min_cost >= 0) /* We are in a situation when assigning any hard register to A by spilling some conflicting allocnos does not improve the allocation cost. */ continue; nregs = hard_regno_nregs[best][mode]; /* Now spill conflicting allocnos which contain a hard register of A when we assign the best chosen hard register to it. */ for (word = 0; word < nwords; word++) { ira_object_t conflict_obj; ira_object_t obj = ALLOCNO_OBJECT (a, word); ira_object_conflict_iterator oci; FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); if ((conflict_hregno = ALLOCNO_HARD_REGNO (conflict_a)) < 0) continue; conflict_nregs = hard_regno_nregs[conflict_hregno][ALLOCNO_MODE (conflict_a)]; if (best + nregs <= conflict_hregno || conflict_hregno + conflict_nregs <= best) /* No intersection. */ continue; ALLOCNO_HARD_REGNO (conflict_a) = -1; sorted_allocnos[n++] = conflict_a; if (internal_flag_ira_verbose > 2 && ira_dump_file != NULL) fprintf (ira_dump_file, "Spilling a%dr%d for a%dr%d\n", ALLOCNO_NUM (conflict_a), ALLOCNO_REGNO (conflict_a), ALLOCNO_NUM (a), ALLOCNO_REGNO (a)); } } /* Assign the best chosen hard register to A. */ ALLOCNO_HARD_REGNO (a) = best; if (internal_flag_ira_verbose > 2 && ira_dump_file != NULL) fprintf (ira_dump_file, "Assigning %d to a%dr%d\n", best, ALLOCNO_NUM (a), ALLOCNO_REGNO (a)); } if (n == 0) return; /* We spilled some allocnos to assign their hard registers to other allocnos. The spilled allocnos are now in array 'sorted_allocnos'. There is still a possibility that some of the spilled allocnos can get hard registers. So let us try assign them hard registers again (just a reminder -- function 'assign_hard_reg' assigns hard registers only if it is possible and profitable). We process the spilled allocnos with biggest benefit to get hard register first -- see function 'allocno_cost_compare_func'. */ qsort (sorted_allocnos, n, sizeof (ira_allocno_t), allocno_cost_compare_func); for (j = 0; j < n; j++) { a = sorted_allocnos[j]; ALLOCNO_ASSIGNED_P (a) = false; if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { fprintf (ira_dump_file, " "); ira_print_expanded_allocno (a); fprintf (ira_dump_file, " -- "); } if (assign_hard_reg (a, false)) { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "assign hard reg %d\n", ALLOCNO_HARD_REGNO (a)); } else { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "assign memory\n"); } } } /* Sort allocnos according to their priorities. */ static int allocno_priority_compare_func (const void *v1p, const void *v2p) { ira_allocno_t a1 = *(const ira_allocno_t *) v1p; ira_allocno_t a2 = *(const ira_allocno_t *) v2p; int pri1, pri2; pri1 = allocno_priorities[ALLOCNO_NUM (a1)]; pri2 = allocno_priorities[ALLOCNO_NUM (a2)]; if (pri2 != pri1) return SORTGT (pri2, pri1); /* If regs are equally good, sort by allocnos, so that the results of qsort leave nothing to chance. */ return ALLOCNO_NUM (a1) - ALLOCNO_NUM (a2); } /* Chaitin-Briggs coloring for allocnos in COLORING_ALLOCNO_BITMAP taking into account allocnos in CONSIDERATION_ALLOCNO_BITMAP. */ static void color_allocnos (void) { unsigned int i, n; bitmap_iterator bi; ira_allocno_t a; setup_profitable_hard_regs (); EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { int l, nr; HARD_REG_SET conflict_hard_regs; allocno_color_data_t data; ira_pref_t pref, next_pref; a = ira_allocnos[i]; nr = ALLOCNO_NUM_OBJECTS (a); CLEAR_HARD_REG_SET (conflict_hard_regs); for (l = 0; l < nr; l++) { ira_object_t obj = ALLOCNO_OBJECT (a, l); IOR_HARD_REG_SET (conflict_hard_regs, OBJECT_CONFLICT_HARD_REGS (obj)); } data = ALLOCNO_COLOR_DATA (a); for (pref = ALLOCNO_PREFS (a); pref != NULL; pref = next_pref) { next_pref = pref->next_pref; if (! ira_hard_reg_in_set_p (pref->hard_regno, ALLOCNO_MODE (a), data->profitable_hard_regs)) ira_remove_pref (pref); } } if (flag_ira_algorithm == IRA_ALGORITHM_PRIORITY) { n = 0; EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; if (ALLOCNO_CLASS (a) == NO_REGS) { ALLOCNO_HARD_REGNO (a) = -1; ALLOCNO_ASSIGNED_P (a) = true; ira_assert (ALLOCNO_UPDATED_HARD_REG_COSTS (a) == NULL); ira_assert (ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (a) == NULL); if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { fprintf (ira_dump_file, " Spill"); ira_print_expanded_allocno (a); fprintf (ira_dump_file, "\n"); } continue; } sorted_allocnos[n++] = a; } if (n != 0) { setup_allocno_priorities (sorted_allocnos, n); qsort (sorted_allocnos, n, sizeof (ira_allocno_t), allocno_priority_compare_func); for (i = 0; i < n; i++) { a = sorted_allocnos[i]; if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { fprintf (ira_dump_file, " "); ira_print_expanded_allocno (a); fprintf (ira_dump_file, " -- "); } if (assign_hard_reg (a, false)) { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "assign hard reg %d\n", ALLOCNO_HARD_REGNO (a)); } else { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "assign memory\n"); } } } } else { form_allocno_hard_regs_nodes_forest (); if (internal_flag_ira_verbose > 2 && ira_dump_file != NULL) print_hard_regs_forest (ira_dump_file); EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; if (ALLOCNO_CLASS (a) != NO_REGS && ! empty_profitable_hard_regs (a)) { ALLOCNO_COLOR_DATA (a)->in_graph_p = true; update_costs_from_prefs (a); } else { ALLOCNO_HARD_REGNO (a) = -1; ALLOCNO_ASSIGNED_P (a) = true; /* We don't need updated costs anymore. */ ira_free_allocno_updated_costs (a); if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { fprintf (ira_dump_file, " Spill"); ira_print_expanded_allocno (a); fprintf (ira_dump_file, "\n"); } } } /* Put the allocnos into the corresponding buckets. */ colorable_allocno_bucket = NULL; uncolorable_allocno_bucket = NULL; EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, i, bi) { a = ira_allocnos[i]; if (ALLOCNO_COLOR_DATA (a)->in_graph_p) put_allocno_into_bucket (a); } push_allocnos_to_stack (); pop_allocnos_from_stack (); finish_allocno_hard_regs_nodes_forest (); } improve_allocation (); } /* Output information about the loop given by its LOOP_TREE_NODE. */ static void print_loop_title (ira_loop_tree_node_t loop_tree_node) { unsigned int j; bitmap_iterator bi; ira_loop_tree_node_t subloop_node, dest_loop_node; edge e; edge_iterator ei; if (loop_tree_node->parent == NULL) fprintf (ira_dump_file, "\n Loop 0 (parent -1, header bb%d, depth 0)\n bbs:", NUM_FIXED_BLOCKS); else { ira_assert (current_loops != NULL && loop_tree_node->loop != NULL); fprintf (ira_dump_file, "\n Loop %d (parent %d, header bb%d, depth %d)\n bbs:", loop_tree_node->loop_num, loop_tree_node->parent->loop_num, loop_tree_node->loop->header->index, loop_depth (loop_tree_node->loop)); } for (subloop_node = loop_tree_node->children; subloop_node != NULL; subloop_node = subloop_node->next) if (subloop_node->bb != NULL) { fprintf (ira_dump_file, " %d", subloop_node->bb->index); FOR_EACH_EDGE (e, ei, subloop_node->bb->succs) if (e->dest != EXIT_BLOCK_PTR_FOR_FN (cfun) && ((dest_loop_node = IRA_BB_NODE (e->dest)->parent) != loop_tree_node)) fprintf (ira_dump_file, "(->%d:l%d)", e->dest->index, dest_loop_node->loop_num); } fprintf (ira_dump_file, "\n all:"); EXECUTE_IF_SET_IN_BITMAP (loop_tree_node->all_allocnos, 0, j, bi) fprintf (ira_dump_file, " %dr%d", j, ALLOCNO_REGNO (ira_allocnos[j])); fprintf (ira_dump_file, "\n modified regnos:"); EXECUTE_IF_SET_IN_BITMAP (loop_tree_node->modified_regnos, 0, j, bi) fprintf (ira_dump_file, " %d", j); fprintf (ira_dump_file, "\n border:"); EXECUTE_IF_SET_IN_BITMAP (loop_tree_node->border_allocnos, 0, j, bi) fprintf (ira_dump_file, " %dr%d", j, ALLOCNO_REGNO (ira_allocnos[j])); fprintf (ira_dump_file, "\n Pressure:"); for (j = 0; (int) j < ira_pressure_classes_num; j++) { enum reg_class pclass; pclass = ira_pressure_classes[j]; if (loop_tree_node->reg_pressure[pclass] == 0) continue; fprintf (ira_dump_file, " %s=%d", reg_class_names[pclass], loop_tree_node->reg_pressure[pclass]); } fprintf (ira_dump_file, "\n"); } /* Color the allocnos inside loop (in the extreme case it can be all of the function) given the corresponding LOOP_TREE_NODE. The function is called for each loop during top-down traverse of the loop tree. */ static void color_pass (ira_loop_tree_node_t loop_tree_node) { int regno, hard_regno, index = -1, n; int cost, exit_freq, enter_freq; unsigned int j; bitmap_iterator bi; enum machine_mode mode; enum reg_class rclass, aclass, pclass; ira_allocno_t a, subloop_allocno; ira_loop_tree_node_t subloop_node; ira_assert (loop_tree_node->bb == NULL); if (internal_flag_ira_verbose > 1 && ira_dump_file != NULL) print_loop_title (loop_tree_node); bitmap_copy (coloring_allocno_bitmap, loop_tree_node->all_allocnos); bitmap_copy (consideration_allocno_bitmap, coloring_allocno_bitmap); n = 0; EXECUTE_IF_SET_IN_BITMAP (consideration_allocno_bitmap, 0, j, bi) { a = ira_allocnos[j]; n++; if (! ALLOCNO_ASSIGNED_P (a)) continue; bitmap_clear_bit (coloring_allocno_bitmap, ALLOCNO_NUM (a)); } allocno_color_data = (allocno_color_data_t) ira_allocate (sizeof (struct allocno_color_data) * n); memset (allocno_color_data, 0, sizeof (struct allocno_color_data) * n); curr_allocno_process = 0; n = 0; EXECUTE_IF_SET_IN_BITMAP (consideration_allocno_bitmap, 0, j, bi) { a = ira_allocnos[j]; ALLOCNO_ADD_DATA (a) = allocno_color_data + n; n++; } init_allocno_threads (); /* Color all mentioned allocnos including transparent ones. */ color_allocnos (); /* Process caps. They are processed just once. */ if (flag_ira_region == IRA_REGION_MIXED || flag_ira_region == IRA_REGION_ALL) EXECUTE_IF_SET_IN_BITMAP (loop_tree_node->all_allocnos, 0, j, bi) { a = ira_allocnos[j]; if (ALLOCNO_CAP_MEMBER (a) == NULL) continue; /* Remove from processing in the next loop. */ bitmap_clear_bit (consideration_allocno_bitmap, j); rclass = ALLOCNO_CLASS (a); pclass = ira_pressure_class_translate[rclass]; if (flag_ira_region == IRA_REGION_MIXED && (loop_tree_node->reg_pressure[pclass] <= ira_class_hard_regs_num[pclass])) { mode = ALLOCNO_MODE (a); hard_regno = ALLOCNO_HARD_REGNO (a); if (hard_regno >= 0) { index = ira_class_hard_reg_index[rclass][hard_regno]; ira_assert (index >= 0); } regno = ALLOCNO_REGNO (a); subloop_allocno = ALLOCNO_CAP_MEMBER (a); subloop_node = ALLOCNO_LOOP_TREE_NODE (subloop_allocno); ira_assert (!ALLOCNO_ASSIGNED_P (subloop_allocno)); ALLOCNO_HARD_REGNO (subloop_allocno) = hard_regno; ALLOCNO_ASSIGNED_P (subloop_allocno) = true; if (hard_regno >= 0) update_costs_from_copies (subloop_allocno, true, true); /* We don't need updated costs anymore: */ ira_free_allocno_updated_costs (subloop_allocno); } } /* Update costs of the corresponding allocnos (not caps) in the subloops. */ for (subloop_node = loop_tree_node->subloops; subloop_node != NULL; subloop_node = subloop_node->subloop_next) { ira_assert (subloop_node->bb == NULL); EXECUTE_IF_SET_IN_BITMAP (consideration_allocno_bitmap, 0, j, bi) { a = ira_allocnos[j]; ira_assert (ALLOCNO_CAP_MEMBER (a) == NULL); mode = ALLOCNO_MODE (a); rclass = ALLOCNO_CLASS (a); pclass = ira_pressure_class_translate[rclass]; hard_regno = ALLOCNO_HARD_REGNO (a); /* Use hard register class here. ??? */ if (hard_regno >= 0) { index = ira_class_hard_reg_index[rclass][hard_regno]; ira_assert (index >= 0); } regno = ALLOCNO_REGNO (a); /* ??? conflict costs */ subloop_allocno = subloop_node->regno_allocno_map[regno]; if (subloop_allocno == NULL || ALLOCNO_CAP (subloop_allocno) != NULL) continue; ira_assert (ALLOCNO_CLASS (subloop_allocno) == rclass); ira_assert (bitmap_bit_p (subloop_node->all_allocnos, ALLOCNO_NUM (subloop_allocno))); if ((flag_ira_region == IRA_REGION_MIXED) && (loop_tree_node->reg_pressure[pclass] <= ira_class_hard_regs_num[pclass])) { if (! ALLOCNO_ASSIGNED_P (subloop_allocno)) { ALLOCNO_HARD_REGNO (subloop_allocno) = hard_regno; ALLOCNO_ASSIGNED_P (subloop_allocno) = true; if (hard_regno >= 0) update_costs_from_copies (subloop_allocno, true, true); /* We don't need updated costs anymore: */ ira_free_allocno_updated_costs (subloop_allocno); } continue; } exit_freq = ira_loop_edge_freq (subloop_node, regno, true); enter_freq = ira_loop_edge_freq (subloop_node, regno, false); ira_assert (regno < ira_reg_equiv_len); if (ira_equiv_no_lvalue_p (regno)) { if (! ALLOCNO_ASSIGNED_P (subloop_allocno)) { ALLOCNO_HARD_REGNO (subloop_allocno) = hard_regno; ALLOCNO_ASSIGNED_P (subloop_allocno) = true; if (hard_regno >= 0) update_costs_from_copies (subloop_allocno, true, true); /* We don't need updated costs anymore: */ ira_free_allocno_updated_costs (subloop_allocno); } } else if (hard_regno < 0) { ALLOCNO_UPDATED_MEMORY_COST (subloop_allocno) -= ((ira_memory_move_cost[mode][rclass][1] * enter_freq) + (ira_memory_move_cost[mode][rclass][0] * exit_freq)); } else { aclass = ALLOCNO_CLASS (subloop_allocno); ira_init_register_move_cost_if_necessary (mode); cost = (ira_register_move_cost[mode][rclass][rclass] * (exit_freq + enter_freq)); ira_allocate_and_set_or_copy_costs (&ALLOCNO_UPDATED_HARD_REG_COSTS (subloop_allocno), aclass, ALLOCNO_UPDATED_CLASS_COST (subloop_allocno), ALLOCNO_HARD_REG_COSTS (subloop_allocno)); ira_allocate_and_set_or_copy_costs (&ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (subloop_allocno), aclass, 0, ALLOCNO_CONFLICT_HARD_REG_COSTS (subloop_allocno)); ALLOCNO_UPDATED_HARD_REG_COSTS (subloop_allocno)[index] -= cost; ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (subloop_allocno)[index] -= cost; if (ALLOCNO_UPDATED_CLASS_COST (subloop_allocno) > ALLOCNO_UPDATED_HARD_REG_COSTS (subloop_allocno)[index]) ALLOCNO_UPDATED_CLASS_COST (subloop_allocno) = ALLOCNO_UPDATED_HARD_REG_COSTS (subloop_allocno)[index]; ALLOCNO_UPDATED_MEMORY_COST (subloop_allocno) += (ira_memory_move_cost[mode][rclass][0] * enter_freq + ira_memory_move_cost[mode][rclass][1] * exit_freq); } } } ira_free (allocno_color_data); EXECUTE_IF_SET_IN_BITMAP (consideration_allocno_bitmap, 0, j, bi) { a = ira_allocnos[j]; ALLOCNO_ADD_DATA (a) = NULL; } } /* Initialize the common data for coloring and calls functions to do Chaitin-Briggs and regional coloring. */ static void do_coloring (void) { coloring_allocno_bitmap = ira_allocate_bitmap (); if (internal_flag_ira_verbose > 0 && ira_dump_file != NULL) fprintf (ira_dump_file, "\n**** Allocnos coloring:\n\n"); ira_traverse_loop_tree (false, ira_loop_tree_root, color_pass, NULL); if (internal_flag_ira_verbose > 1 && ira_dump_file != NULL) ira_print_disposition (ira_dump_file); ira_free_bitmap (coloring_allocno_bitmap); } /* Move spill/restore code, which are to be generated in ira-emit.c, to less frequent points (if it is profitable) by reassigning some allocnos (in loop with subloops containing in another loop) to memory which results in longer live-range where the corresponding pseudo-registers will be in memory. */ static void move_spill_restore (void) { int cost, regno, hard_regno, hard_regno2, index; bool changed_p; int enter_freq, exit_freq; enum machine_mode mode; enum reg_class rclass; ira_allocno_t a, parent_allocno, subloop_allocno; ira_loop_tree_node_t parent, loop_node, subloop_node; ira_allocno_iterator ai; for (;;) { changed_p = false; if (internal_flag_ira_verbose > 0 && ira_dump_file != NULL) fprintf (ira_dump_file, "New iteration of spill/restore move\n"); FOR_EACH_ALLOCNO (a, ai) { regno = ALLOCNO_REGNO (a); loop_node = ALLOCNO_LOOP_TREE_NODE (a); if (ALLOCNO_CAP_MEMBER (a) != NULL || ALLOCNO_CAP (a) != NULL || (hard_regno = ALLOCNO_HARD_REGNO (a)) < 0 || loop_node->children == NULL /* don't do the optimization because it can create copies and the reload pass can spill the allocno set by copy although the allocno will not get memory slot. */ || ira_equiv_no_lvalue_p (regno) || !bitmap_bit_p (loop_node->border_allocnos, ALLOCNO_NUM (a))) continue; mode = ALLOCNO_MODE (a); rclass = ALLOCNO_CLASS (a); index = ira_class_hard_reg_index[rclass][hard_regno]; ira_assert (index >= 0); cost = (ALLOCNO_MEMORY_COST (a) - (ALLOCNO_HARD_REG_COSTS (a) == NULL ? ALLOCNO_CLASS_COST (a) : ALLOCNO_HARD_REG_COSTS (a)[index])); ira_init_register_move_cost_if_necessary (mode); for (subloop_node = loop_node->subloops; subloop_node != NULL; subloop_node = subloop_node->subloop_next) { ira_assert (subloop_node->bb == NULL); subloop_allocno = subloop_node->regno_allocno_map[regno]; if (subloop_allocno == NULL) continue; ira_assert (rclass == ALLOCNO_CLASS (subloop_allocno)); /* We have accumulated cost. To get the real cost of allocno usage in the loop we should subtract costs of the subloop allocnos. */ cost -= (ALLOCNO_MEMORY_COST (subloop_allocno) - (ALLOCNO_HARD_REG_COSTS (subloop_allocno) == NULL ? ALLOCNO_CLASS_COST (subloop_allocno) : ALLOCNO_HARD_REG_COSTS (subloop_allocno)[index])); exit_freq = ira_loop_edge_freq (subloop_node, regno, true); enter_freq = ira_loop_edge_freq (subloop_node, regno, false); if ((hard_regno2 = ALLOCNO_HARD_REGNO (subloop_allocno)) < 0) cost -= (ira_memory_move_cost[mode][rclass][0] * exit_freq + ira_memory_move_cost[mode][rclass][1] * enter_freq); else { cost += (ira_memory_move_cost[mode][rclass][0] * exit_freq + ira_memory_move_cost[mode][rclass][1] * enter_freq); if (hard_regno2 != hard_regno) cost -= (ira_register_move_cost[mode][rclass][rclass] * (exit_freq + enter_freq)); } } if ((parent = loop_node->parent) != NULL && (parent_allocno = parent->regno_allocno_map[regno]) != NULL) { ira_assert (rclass == ALLOCNO_CLASS (parent_allocno)); exit_freq = ira_loop_edge_freq (loop_node, regno, true); enter_freq = ira_loop_edge_freq (loop_node, regno, false); if ((hard_regno2 = ALLOCNO_HARD_REGNO (parent_allocno)) < 0) cost -= (ira_memory_move_cost[mode][rclass][0] * exit_freq + ira_memory_move_cost[mode][rclass][1] * enter_freq); else { cost += (ira_memory_move_cost[mode][rclass][1] * exit_freq + ira_memory_move_cost[mode][rclass][0] * enter_freq); if (hard_regno2 != hard_regno) cost -= (ira_register_move_cost[mode][rclass][rclass] * (exit_freq + enter_freq)); } } if (cost < 0) { ALLOCNO_HARD_REGNO (a) = -1; if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) { fprintf (ira_dump_file, " Moving spill/restore for a%dr%d up from loop %d", ALLOCNO_NUM (a), regno, loop_node->loop_num); fprintf (ira_dump_file, " - profit %d\n", -cost); } changed_p = true; } } if (! changed_p) break; } } /* Update current hard reg costs and current conflict hard reg costs for allocno A. It is done by processing its copies containing other allocnos already assigned. */ static void update_curr_costs (ira_allocno_t a) { int i, hard_regno, cost; enum machine_mode mode; enum reg_class aclass, rclass; ira_allocno_t another_a; ira_copy_t cp, next_cp; ira_free_allocno_updated_costs (a); ira_assert (! ALLOCNO_ASSIGNED_P (a)); aclass = ALLOCNO_CLASS (a); if (aclass == NO_REGS) return; mode = ALLOCNO_MODE (a); ira_init_register_move_cost_if_necessary (mode); for (cp = ALLOCNO_COPIES (a); cp != NULL; cp = next_cp) { if (cp->first == a) { next_cp = cp->next_first_allocno_copy; another_a = cp->second; } else if (cp->second == a) { next_cp = cp->next_second_allocno_copy; another_a = cp->first; } else gcc_unreachable (); if (! ira_reg_classes_intersect_p[aclass][ALLOCNO_CLASS (another_a)] || ! ALLOCNO_ASSIGNED_P (another_a) || (hard_regno = ALLOCNO_HARD_REGNO (another_a)) < 0) continue; rclass = REGNO_REG_CLASS (hard_regno); i = ira_class_hard_reg_index[aclass][hard_regno]; if (i < 0) continue; cost = (cp->first == a ? ira_register_move_cost[mode][rclass][aclass] : ira_register_move_cost[mode][aclass][rclass]); ira_allocate_and_set_or_copy_costs (&ALLOCNO_UPDATED_HARD_REG_COSTS (a), aclass, ALLOCNO_CLASS_COST (a), ALLOCNO_HARD_REG_COSTS (a)); ira_allocate_and_set_or_copy_costs (&ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (a), aclass, 0, ALLOCNO_CONFLICT_HARD_REG_COSTS (a)); ALLOCNO_UPDATED_HARD_REG_COSTS (a)[i] -= cp->freq * cost; ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (a)[i] -= cp->freq * cost; } } /* Try to assign hard registers to the unassigned allocnos and allocnos conflicting with them or conflicting with allocnos whose regno >= START_REGNO. The function is called after ira_flattening, so more allocnos (including ones created in ira-emit.c) will have a chance to get a hard register. We use simple assignment algorithm based on priorities. */ void ira_reassign_conflict_allocnos (int start_regno) { int i, allocnos_to_color_num; ira_allocno_t a; enum reg_class aclass; bitmap allocnos_to_color; ira_allocno_iterator ai; allocnos_to_color = ira_allocate_bitmap (); allocnos_to_color_num = 0; FOR_EACH_ALLOCNO (a, ai) { int n = ALLOCNO_NUM_OBJECTS (a); if (! ALLOCNO_ASSIGNED_P (a) && ! bitmap_bit_p (allocnos_to_color, ALLOCNO_NUM (a))) { if (ALLOCNO_CLASS (a) != NO_REGS) sorted_allocnos[allocnos_to_color_num++] = a; else { ALLOCNO_ASSIGNED_P (a) = true; ALLOCNO_HARD_REGNO (a) = -1; ira_assert (ALLOCNO_UPDATED_HARD_REG_COSTS (a) == NULL); ira_assert (ALLOCNO_UPDATED_CONFLICT_HARD_REG_COSTS (a) == NULL); } bitmap_set_bit (allocnos_to_color, ALLOCNO_NUM (a)); } if (ALLOCNO_REGNO (a) < start_regno || (aclass = ALLOCNO_CLASS (a)) == NO_REGS) continue; for (i = 0; i < n; i++) { ira_object_t obj = ALLOCNO_OBJECT (a, i); ira_object_t conflict_obj; ira_object_conflict_iterator oci; FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); ira_assert (ira_reg_classes_intersect_p [aclass][ALLOCNO_CLASS (conflict_a)]); if (!bitmap_set_bit (allocnos_to_color, ALLOCNO_NUM (conflict_a))) continue; sorted_allocnos[allocnos_to_color_num++] = conflict_a; } } } ira_free_bitmap (allocnos_to_color); if (allocnos_to_color_num > 1) { setup_allocno_priorities (sorted_allocnos, allocnos_to_color_num); qsort (sorted_allocnos, allocnos_to_color_num, sizeof (ira_allocno_t), allocno_priority_compare_func); } for (i = 0; i < allocnos_to_color_num; i++) { a = sorted_allocnos[i]; ALLOCNO_ASSIGNED_P (a) = false; update_curr_costs (a); } for (i = 0; i < allocnos_to_color_num; i++) { a = sorted_allocnos[i]; if (assign_hard_reg (a, true)) { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, " Secondary allocation: assign hard reg %d to reg %d\n", ALLOCNO_HARD_REGNO (a), ALLOCNO_REGNO (a)); } } } /* This page contains functions used to find conflicts using allocno live ranges. */ #ifdef ENABLE_IRA_CHECKING /* Return TRUE if live ranges of pseudo-registers REGNO1 and REGNO2 intersect. This should be used when there is only one region. Currently this is used during reload. */ static bool conflict_by_live_ranges_p (int regno1, int regno2) { ira_allocno_t a1, a2; ira_assert (regno1 >= FIRST_PSEUDO_REGISTER && regno2 >= FIRST_PSEUDO_REGISTER); /* Reg info caclulated by dataflow infrastructure can be different from one calculated by regclass. */ if ((a1 = ira_loop_tree_root->regno_allocno_map[regno1]) == NULL || (a2 = ira_loop_tree_root->regno_allocno_map[regno2]) == NULL) return false; return allocnos_conflict_by_live_ranges_p (a1, a2); } #endif /* This page contains code to coalesce memory stack slots used by spilled allocnos. This results in smaller stack frame, better data locality, and in smaller code for some architectures like x86/x86_64 where insn size depends on address displacement value. On the other hand, it can worsen insn scheduling after the RA but in practice it is less important than smaller stack frames. */ /* TRUE if we coalesced some allocnos. In other words, if we got loops formed by members first_coalesced_allocno and next_coalesced_allocno containing more one allocno. */ static bool allocno_coalesced_p; /* Bitmap used to prevent a repeated allocno processing because of coalescing. */ static bitmap processed_coalesced_allocno_bitmap; /* See below. */ typedef struct coalesce_data *coalesce_data_t; /* To decrease footprint of ira_allocno structure we store all data needed only for coalescing in the following structure. */ struct coalesce_data { /* Coalesced allocnos form a cyclic list. One allocno given by FIRST represents all coalesced allocnos. The list is chained by NEXT. */ ira_allocno_t first; ira_allocno_t next; int temp; }; /* Container for storing allocno data concerning coalescing. */ static coalesce_data_t allocno_coalesce_data; /* Macro to access the data concerning coalescing. */ #define ALLOCNO_COALESCE_DATA(a) ((coalesce_data_t) ALLOCNO_ADD_DATA (a)) /* Merge two sets of coalesced allocnos given correspondingly by allocnos A1 and A2 (more accurately merging A2 set into A1 set). */ static void merge_allocnos (ira_allocno_t a1, ira_allocno_t a2) { ira_allocno_t a, first, last, next; first = ALLOCNO_COALESCE_DATA (a1)->first; a = ALLOCNO_COALESCE_DATA (a2)->first; if (first == a) return; for (last = a2, a = ALLOCNO_COALESCE_DATA (a2)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { ALLOCNO_COALESCE_DATA (a)->first = first; if (a == a2) break; last = a; } next = allocno_coalesce_data[ALLOCNO_NUM (first)].next; allocno_coalesce_data[ALLOCNO_NUM (first)].next = a2; allocno_coalesce_data[ALLOCNO_NUM (last)].next = next; } /* Return TRUE if there are conflicting allocnos from two sets of coalesced allocnos given correspondingly by allocnos A1 and A2. We use live ranges to find conflicts because conflicts are represented only for allocnos of the same allocno class and during the reload pass we coalesce allocnos for sharing stack memory slots. */ static bool coalesced_allocno_conflict_p (ira_allocno_t a1, ira_allocno_t a2) { ira_allocno_t a, conflict_a; if (allocno_coalesced_p) { bitmap_clear (processed_coalesced_allocno_bitmap); for (a = ALLOCNO_COALESCE_DATA (a1)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { bitmap_set_bit (processed_coalesced_allocno_bitmap, ALLOCNO_NUM (a)); if (a == a1) break; } } for (a = ALLOCNO_COALESCE_DATA (a2)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { for (conflict_a = ALLOCNO_COALESCE_DATA (a1)->next;; conflict_a = ALLOCNO_COALESCE_DATA (conflict_a)->next) { if (allocnos_conflict_by_live_ranges_p (a, conflict_a)) return true; if (conflict_a == a1) break; } if (a == a2) break; } return false; } /* The major function for aggressive allocno coalescing. We coalesce only spilled allocnos. If some allocnos have been coalesced, we set up flag allocno_coalesced_p. */ static void coalesce_allocnos (void) { ira_allocno_t a; ira_copy_t cp, next_cp; unsigned int j; int i, n, cp_num, regno; bitmap_iterator bi; cp_num = 0; /* Collect copies. */ EXECUTE_IF_SET_IN_BITMAP (coloring_allocno_bitmap, 0, j, bi) { a = ira_allocnos[j]; regno = ALLOCNO_REGNO (a); if (! ALLOCNO_ASSIGNED_P (a) || ALLOCNO_HARD_REGNO (a) >= 0 || ira_equiv_no_lvalue_p (regno)) continue; for (cp = ALLOCNO_COPIES (a); cp != NULL; cp = next_cp) { if (cp->first == a) { next_cp = cp->next_first_allocno_copy; regno = ALLOCNO_REGNO (cp->second); /* For priority coloring we coalesce allocnos only with the same allocno class not with intersected allocno classes as it were possible. It is done for simplicity. */ if ((cp->insn != NULL || cp->constraint_p) && ALLOCNO_ASSIGNED_P (cp->second) && ALLOCNO_HARD_REGNO (cp->second) < 0 && ! ira_equiv_no_lvalue_p (regno)) sorted_copies[cp_num++] = cp; } else if (cp->second == a) next_cp = cp->next_second_allocno_copy; else gcc_unreachable (); } } qsort (sorted_copies, cp_num, sizeof (ira_copy_t), copy_freq_compare_func); /* Coalesced copies, most frequently executed first. */ for (; cp_num != 0;) { for (i = 0; i < cp_num; i++) { cp = sorted_copies[i]; if (! coalesced_allocno_conflict_p (cp->first, cp->second)) { allocno_coalesced_p = true; if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, " Coalescing copy %d:a%dr%d-a%dr%d (freq=%d)\n", cp->num, ALLOCNO_NUM (cp->first), ALLOCNO_REGNO (cp->first), ALLOCNO_NUM (cp->second), ALLOCNO_REGNO (cp->second), cp->freq); merge_allocnos (cp->first, cp->second); i++; break; } } /* Collect the rest of copies. */ for (n = 0; i < cp_num; i++) { cp = sorted_copies[i]; if (allocno_coalesce_data[ALLOCNO_NUM (cp->first)].first != allocno_coalesce_data[ALLOCNO_NUM (cp->second)].first) sorted_copies[n++] = cp; } cp_num = n; } } /* Usage cost and order number of coalesced allocno set to which given pseudo register belongs to. */ static int *regno_coalesced_allocno_cost; static int *regno_coalesced_allocno_num; /* Sort pseudos according frequencies of coalesced allocno sets they belong to (putting most frequently ones first), and according to coalesced allocno set order numbers. */ static int coalesced_pseudo_reg_freq_compare (const void *v1p, const void *v2p) { const int regno1 = *(const int *) v1p; const int regno2 = *(const int *) v2p; int diff; if ((diff = (regno_coalesced_allocno_cost[regno2] - regno_coalesced_allocno_cost[regno1])) != 0) return diff; if ((diff = (regno_coalesced_allocno_num[regno1] - regno_coalesced_allocno_num[regno2])) != 0) return diff; return regno1 - regno2; } /* Widest width in which each pseudo reg is referred to (via subreg). It is used for sorting pseudo registers. */ static unsigned int *regno_max_ref_width; /* Redefine STACK_GROWS_DOWNWARD in terms of 0 or 1. */ #ifdef STACK_GROWS_DOWNWARD # undef STACK_GROWS_DOWNWARD # define STACK_GROWS_DOWNWARD 1 #else # define STACK_GROWS_DOWNWARD 0 #endif /* Sort pseudos according their slot numbers (putting ones with smaller numbers first, or last when the frame pointer is not needed). */ static int coalesced_pseudo_reg_slot_compare (const void *v1p, const void *v2p) { const int regno1 = *(const int *) v1p; const int regno2 = *(const int *) v2p; ira_allocno_t a1 = ira_regno_allocno_map[regno1]; ira_allocno_t a2 = ira_regno_allocno_map[regno2]; int diff, slot_num1, slot_num2; int total_size1, total_size2; if (a1 == NULL || ALLOCNO_HARD_REGNO (a1) >= 0) { if (a2 == NULL || ALLOCNO_HARD_REGNO (a2) >= 0) return regno1 - regno2; return 1; } else if (a2 == NULL || ALLOCNO_HARD_REGNO (a2) >= 0) return -1; slot_num1 = -ALLOCNO_HARD_REGNO (a1); slot_num2 = -ALLOCNO_HARD_REGNO (a2); if ((diff = slot_num1 - slot_num2) != 0) return (frame_pointer_needed || !FRAME_GROWS_DOWNWARD == STACK_GROWS_DOWNWARD ? diff : -diff); total_size1 = MAX (PSEUDO_REGNO_BYTES (regno1), regno_max_ref_width[regno1]); total_size2 = MAX (PSEUDO_REGNO_BYTES (regno2), regno_max_ref_width[regno2]); if ((diff = total_size2 - total_size1) != 0) return diff; return regno1 - regno2; } /* Setup REGNO_COALESCED_ALLOCNO_COST and REGNO_COALESCED_ALLOCNO_NUM for coalesced allocno sets containing allocnos with their regnos given in array PSEUDO_REGNOS of length N. */ static void setup_coalesced_allocno_costs_and_nums (int *pseudo_regnos, int n) { int i, num, regno, cost; ira_allocno_t allocno, a; for (num = i = 0; i < n; i++) { regno = pseudo_regnos[i]; allocno = ira_regno_allocno_map[regno]; if (allocno == NULL) { regno_coalesced_allocno_cost[regno] = 0; regno_coalesced_allocno_num[regno] = ++num; continue; } if (ALLOCNO_COALESCE_DATA (allocno)->first != allocno) continue; num++; for (cost = 0, a = ALLOCNO_COALESCE_DATA (allocno)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { cost += ALLOCNO_FREQ (a); if (a == allocno) break; } for (a = ALLOCNO_COALESCE_DATA (allocno)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { regno_coalesced_allocno_num[ALLOCNO_REGNO (a)] = num; regno_coalesced_allocno_cost[ALLOCNO_REGNO (a)] = cost; if (a == allocno) break; } } } /* Collect spilled allocnos representing coalesced allocno sets (the first coalesced allocno). The collected allocnos are returned through array SPILLED_COALESCED_ALLOCNOS. The function returns the number of the collected allocnos. The allocnos are given by their regnos in array PSEUDO_REGNOS of length N. */ static int collect_spilled_coalesced_allocnos (int *pseudo_regnos, int n, ira_allocno_t *spilled_coalesced_allocnos) { int i, num, regno; ira_allocno_t allocno; for (num = i = 0; i < n; i++) { regno = pseudo_regnos[i]; allocno = ira_regno_allocno_map[regno]; if (allocno == NULL || ALLOCNO_HARD_REGNO (allocno) >= 0 || ALLOCNO_COALESCE_DATA (allocno)->first != allocno) continue; spilled_coalesced_allocnos[num++] = allocno; } return num; } /* Array of live ranges of size IRA_ALLOCNOS_NUM. Live range for given slot contains live ranges of coalesced allocnos assigned to given slot. */ static live_range_t *slot_coalesced_allocnos_live_ranges; /* Return TRUE if coalesced allocnos represented by ALLOCNO has live ranges intersected with live ranges of coalesced allocnos assigned to slot with number N. */ static bool slot_coalesced_allocno_live_ranges_intersect_p (ira_allocno_t allocno, int n) { ira_allocno_t a; for (a = ALLOCNO_COALESCE_DATA (allocno)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { int i; int nr = ALLOCNO_NUM_OBJECTS (a); for (i = 0; i < nr; i++) { ira_object_t obj = ALLOCNO_OBJECT (a, i); if (ira_live_ranges_intersect_p (slot_coalesced_allocnos_live_ranges[n], OBJECT_LIVE_RANGES (obj))) return true; } if (a == allocno) break; } return false; } /* Update live ranges of slot to which coalesced allocnos represented by ALLOCNO were assigned. */ static void setup_slot_coalesced_allocno_live_ranges (ira_allocno_t allocno) { int i, n; ira_allocno_t a; live_range_t r; n = ALLOCNO_COALESCE_DATA (allocno)->temp; for (a = ALLOCNO_COALESCE_DATA (allocno)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { int nr = ALLOCNO_NUM_OBJECTS (a); for (i = 0; i < nr; i++) { ira_object_t obj = ALLOCNO_OBJECT (a, i); r = ira_copy_live_range_list (OBJECT_LIVE_RANGES (obj)); slot_coalesced_allocnos_live_ranges[n] = ira_merge_live_ranges (slot_coalesced_allocnos_live_ranges[n], r); } if (a == allocno) break; } } /* We have coalesced allocnos involving in copies. Coalesce allocnos further in order to share the same memory stack slot. Allocnos representing sets of allocnos coalesced before the call are given in array SPILLED_COALESCED_ALLOCNOS of length NUM. Return TRUE if some allocnos were coalesced in the function. */ static bool coalesce_spill_slots (ira_allocno_t *spilled_coalesced_allocnos, int num) { int i, j, n, last_coalesced_allocno_num; ira_allocno_t allocno, a; bool merged_p = false; bitmap set_jump_crosses = regstat_get_setjmp_crosses (); slot_coalesced_allocnos_live_ranges = (live_range_t *) ira_allocate (sizeof (live_range_t) * ira_allocnos_num); memset (slot_coalesced_allocnos_live_ranges, 0, sizeof (live_range_t) * ira_allocnos_num); last_coalesced_allocno_num = 0; /* Coalesce non-conflicting spilled allocnos preferring most frequently used. */ for (i = 0; i < num; i++) { allocno = spilled_coalesced_allocnos[i]; if (ALLOCNO_COALESCE_DATA (allocno)->first != allocno || bitmap_bit_p (set_jump_crosses, ALLOCNO_REGNO (allocno)) || ira_equiv_no_lvalue_p (ALLOCNO_REGNO (allocno))) continue; for (j = 0; j < i; j++) { a = spilled_coalesced_allocnos[j]; n = ALLOCNO_COALESCE_DATA (a)->temp; if (ALLOCNO_COALESCE_DATA (a)->first == a && ! bitmap_bit_p (set_jump_crosses, ALLOCNO_REGNO (a)) && ! ira_equiv_no_lvalue_p (ALLOCNO_REGNO (a)) && ! slot_coalesced_allocno_live_ranges_intersect_p (allocno, n)) break; } if (j >= i) { /* No coalescing: set up number for coalesced allocnos represented by ALLOCNO. */ ALLOCNO_COALESCE_DATA (allocno)->temp = last_coalesced_allocno_num++; setup_slot_coalesced_allocno_live_ranges (allocno); } else { allocno_coalesced_p = true; merged_p = true; if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, " Coalescing spilled allocnos a%dr%d->a%dr%d\n", ALLOCNO_NUM (allocno), ALLOCNO_REGNO (allocno), ALLOCNO_NUM (a), ALLOCNO_REGNO (a)); ALLOCNO_COALESCE_DATA (allocno)->temp = ALLOCNO_COALESCE_DATA (a)->temp; setup_slot_coalesced_allocno_live_ranges (allocno); merge_allocnos (a, allocno); ira_assert (ALLOCNO_COALESCE_DATA (a)->first == a); } } for (i = 0; i < ira_allocnos_num; i++) ira_finish_live_range_list (slot_coalesced_allocnos_live_ranges[i]); ira_free (slot_coalesced_allocnos_live_ranges); return merged_p; } /* Sort pseudo-register numbers in array PSEUDO_REGNOS of length N for subsequent assigning stack slots to them in the reload pass. To do this we coalesce spilled allocnos first to decrease the number of memory-memory move insns. This function is called by the reload. */ void ira_sort_regnos_for_alter_reg (int *pseudo_regnos, int n, unsigned int *reg_max_ref_width) { int max_regno = max_reg_num (); int i, regno, num, slot_num; ira_allocno_t allocno, a; ira_allocno_iterator ai; ira_allocno_t *spilled_coalesced_allocnos; /* Set up allocnos can be coalesced. */ coloring_allocno_bitmap = ira_allocate_bitmap (); for (i = 0; i < n; i++) { regno = pseudo_regnos[i]; allocno = ira_regno_allocno_map[regno]; if (allocno != NULL) bitmap_set_bit (coloring_allocno_bitmap, ALLOCNO_NUM (allocno)); } allocno_coalesced_p = false; processed_coalesced_allocno_bitmap = ira_allocate_bitmap (); allocno_coalesce_data = (coalesce_data_t) ira_allocate (sizeof (struct coalesce_data) * ira_allocnos_num); /* Initialize coalesce data for allocnos. */ FOR_EACH_ALLOCNO (a, ai) { ALLOCNO_ADD_DATA (a) = allocno_coalesce_data + ALLOCNO_NUM (a); ALLOCNO_COALESCE_DATA (a)->first = a; ALLOCNO_COALESCE_DATA (a)->next = a; } coalesce_allocnos (); ira_free_bitmap (coloring_allocno_bitmap); regno_coalesced_allocno_cost = (int *) ira_allocate (max_regno * sizeof (int)); regno_coalesced_allocno_num = (int *) ira_allocate (max_regno * sizeof (int)); memset (regno_coalesced_allocno_num, 0, max_regno * sizeof (int)); setup_coalesced_allocno_costs_and_nums (pseudo_regnos, n); /* Sort regnos according frequencies of the corresponding coalesced allocno sets. */ qsort (pseudo_regnos, n, sizeof (int), coalesced_pseudo_reg_freq_compare); spilled_coalesced_allocnos = (ira_allocno_t *) ira_allocate (ira_allocnos_num * sizeof (ira_allocno_t)); /* Collect allocnos representing the spilled coalesced allocno sets. */ num = collect_spilled_coalesced_allocnos (pseudo_regnos, n, spilled_coalesced_allocnos); if (flag_ira_share_spill_slots && coalesce_spill_slots (spilled_coalesced_allocnos, num)) { setup_coalesced_allocno_costs_and_nums (pseudo_regnos, n); qsort (pseudo_regnos, n, sizeof (int), coalesced_pseudo_reg_freq_compare); num = collect_spilled_coalesced_allocnos (pseudo_regnos, n, spilled_coalesced_allocnos); } ira_free_bitmap (processed_coalesced_allocno_bitmap); allocno_coalesced_p = false; /* Assign stack slot numbers to spilled allocno sets, use smaller numbers for most frequently used coalesced allocnos. -1 is reserved for dynamic search of stack slots for pseudos spilled by the reload. */ slot_num = 1; for (i = 0; i < num; i++) { allocno = spilled_coalesced_allocnos[i]; if (ALLOCNO_COALESCE_DATA (allocno)->first != allocno || ALLOCNO_HARD_REGNO (allocno) >= 0 || ira_equiv_no_lvalue_p (ALLOCNO_REGNO (allocno))) continue; if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, " Slot %d (freq,size):", slot_num); slot_num++; for (a = ALLOCNO_COALESCE_DATA (allocno)->next;; a = ALLOCNO_COALESCE_DATA (a)->next) { ira_assert (ALLOCNO_HARD_REGNO (a) < 0); ALLOCNO_HARD_REGNO (a) = -slot_num; if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, " a%dr%d(%d,%d)", ALLOCNO_NUM (a), ALLOCNO_REGNO (a), ALLOCNO_FREQ (a), MAX (PSEUDO_REGNO_BYTES (ALLOCNO_REGNO (a)), reg_max_ref_width[ALLOCNO_REGNO (a)])); if (a == allocno) break; } if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "\n"); } ira_spilled_reg_stack_slots_num = slot_num - 1; ira_free (spilled_coalesced_allocnos); /* Sort regnos according the slot numbers. */ regno_max_ref_width = reg_max_ref_width; qsort (pseudo_regnos, n, sizeof (int), coalesced_pseudo_reg_slot_compare); FOR_EACH_ALLOCNO (a, ai) ALLOCNO_ADD_DATA (a) = NULL; ira_free (allocno_coalesce_data); ira_free (regno_coalesced_allocno_num); ira_free (regno_coalesced_allocno_cost); } /* This page contains code used by the reload pass to improve the final code. */ /* The function is called from reload to mark changes in the allocation of REGNO made by the reload. Remember that reg_renumber reflects the change result. */ void ira_mark_allocation_change (int regno) { ira_allocno_t a = ira_regno_allocno_map[regno]; int old_hard_regno, hard_regno, cost; enum reg_class aclass = ALLOCNO_CLASS (a); ira_assert (a != NULL); hard_regno = reg_renumber[regno]; if ((old_hard_regno = ALLOCNO_HARD_REGNO (a)) == hard_regno) return; if (old_hard_regno < 0) cost = -ALLOCNO_MEMORY_COST (a); else { ira_assert (ira_class_hard_reg_index[aclass][old_hard_regno] >= 0); cost = -(ALLOCNO_HARD_REG_COSTS (a) == NULL ? ALLOCNO_CLASS_COST (a) : ALLOCNO_HARD_REG_COSTS (a) [ira_class_hard_reg_index[aclass][old_hard_regno]]); update_costs_from_copies (a, false, false); } ira_overall_cost -= cost; ALLOCNO_HARD_REGNO (a) = hard_regno; if (hard_regno < 0) { ALLOCNO_HARD_REGNO (a) = -1; cost += ALLOCNO_MEMORY_COST (a); } else if (ira_class_hard_reg_index[aclass][hard_regno] >= 0) { cost += (ALLOCNO_HARD_REG_COSTS (a) == NULL ? ALLOCNO_CLASS_COST (a) : ALLOCNO_HARD_REG_COSTS (a) [ira_class_hard_reg_index[aclass][hard_regno]]); update_costs_from_copies (a, true, false); } else /* Reload changed class of the allocno. */ cost = 0; ira_overall_cost += cost; } /* This function is called when reload deletes memory-memory move. In this case we marks that the allocation of the corresponding allocnos should be not changed in future. Otherwise we risk to get a wrong code. */ void ira_mark_memory_move_deletion (int dst_regno, int src_regno) { ira_allocno_t dst = ira_regno_allocno_map[dst_regno]; ira_allocno_t src = ira_regno_allocno_map[src_regno]; ira_assert (dst != NULL && src != NULL && ALLOCNO_HARD_REGNO (dst) < 0 && ALLOCNO_HARD_REGNO (src) < 0); ALLOCNO_DONT_REASSIGN_P (dst) = true; ALLOCNO_DONT_REASSIGN_P (src) = true; } /* Try to assign a hard register (except for FORBIDDEN_REGS) to allocno A and return TRUE in the case of success. */ static bool allocno_reload_assign (ira_allocno_t a, HARD_REG_SET forbidden_regs) { int hard_regno; enum reg_class aclass; int regno = ALLOCNO_REGNO (a); HARD_REG_SET saved[2]; int i, n; n = ALLOCNO_NUM_OBJECTS (a); for (i = 0; i < n; i++) { ira_object_t obj = ALLOCNO_OBJECT (a, i); COPY_HARD_REG_SET (saved[i], OBJECT_TOTAL_CONFLICT_HARD_REGS (obj)); IOR_HARD_REG_SET (OBJECT_TOTAL_CONFLICT_HARD_REGS (obj), forbidden_regs); if (! flag_caller_saves && ALLOCNO_CALLS_CROSSED_NUM (a) != 0) IOR_HARD_REG_SET (OBJECT_TOTAL_CONFLICT_HARD_REGS (obj), call_used_reg_set); } ALLOCNO_ASSIGNED_P (a) = false; aclass = ALLOCNO_CLASS (a); update_curr_costs (a); assign_hard_reg (a, true); hard_regno = ALLOCNO_HARD_REGNO (a); reg_renumber[regno] = hard_regno; if (hard_regno < 0) ALLOCNO_HARD_REGNO (a) = -1; else { ira_assert (ira_class_hard_reg_index[aclass][hard_regno] >= 0); ira_overall_cost -= (ALLOCNO_MEMORY_COST (a) - (ALLOCNO_HARD_REG_COSTS (a) == NULL ? ALLOCNO_CLASS_COST (a) : ALLOCNO_HARD_REG_COSTS (a)[ira_class_hard_reg_index [aclass][hard_regno]])); if (ALLOCNO_CALLS_CROSSED_NUM (a) != 0 && ira_hard_reg_set_intersection_p (hard_regno, ALLOCNO_MODE (a), call_used_reg_set)) { ira_assert (flag_caller_saves); caller_save_needed = 1; } } /* If we found a hard register, modify the RTL for the pseudo register to show the hard register, and mark the pseudo register live. */ if (reg_renumber[regno] >= 0) { if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, ": reassign to %d\n", reg_renumber[regno]); SET_REGNO (regno_reg_rtx[regno], reg_renumber[regno]); mark_home_live (regno); } else if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, "\n"); for (i = 0; i < n; i++) { ira_object_t obj = ALLOCNO_OBJECT (a, i); COPY_HARD_REG_SET (OBJECT_TOTAL_CONFLICT_HARD_REGS (obj), saved[i]); } return reg_renumber[regno] >= 0; } /* Sort pseudos according their usage frequencies (putting most frequently ones first). */ static int pseudo_reg_compare (const void *v1p, const void *v2p) { int regno1 = *(const int *) v1p; int regno2 = *(const int *) v2p; int diff; if ((diff = REG_FREQ (regno2) - REG_FREQ (regno1)) != 0) return diff; return regno1 - regno2; } /* Try to allocate hard registers to SPILLED_PSEUDO_REGS (there are NUM of them) or spilled pseudos conflicting with pseudos in SPILLED_PSEUDO_REGS. Return TRUE and update SPILLED, if the allocation has been changed. The function doesn't use BAD_SPILL_REGS and hard registers in PSEUDO_FORBIDDEN_REGS and PSEUDO_PREVIOUS_REGS for the corresponding pseudos. The function is called by the reload pass at the end of each reload iteration. */ bool ira_reassign_pseudos (int *spilled_pseudo_regs, int num, HARD_REG_SET bad_spill_regs, HARD_REG_SET *pseudo_forbidden_regs, HARD_REG_SET *pseudo_previous_regs, bitmap spilled) { int i, n, regno; bool changed_p; ira_allocno_t a; HARD_REG_SET forbidden_regs; bitmap temp = BITMAP_ALLOC (NULL); /* Add pseudos which conflict with pseudos already in SPILLED_PSEUDO_REGS to SPILLED_PSEUDO_REGS. This is preferable to allocating in two steps as some of the conflicts might have a higher priority than the pseudos passed in SPILLED_PSEUDO_REGS. */ for (i = 0; i < num; i++) bitmap_set_bit (temp, spilled_pseudo_regs[i]); for (i = 0, n = num; i < n; i++) { int nr, j; int regno = spilled_pseudo_regs[i]; bitmap_set_bit (temp, regno); a = ira_regno_allocno_map[regno]; nr = ALLOCNO_NUM_OBJECTS (a); for (j = 0; j < nr; j++) { ira_object_t conflict_obj; ira_object_t obj = ALLOCNO_OBJECT (a, j); ira_object_conflict_iterator oci; FOR_EACH_OBJECT_CONFLICT (obj, conflict_obj, oci) { ira_allocno_t conflict_a = OBJECT_ALLOCNO (conflict_obj); if (ALLOCNO_HARD_REGNO (conflict_a) < 0 && ! ALLOCNO_DONT_REASSIGN_P (conflict_a) && bitmap_set_bit (temp, ALLOCNO_REGNO (conflict_a))) { spilled_pseudo_regs[num++] = ALLOCNO_REGNO (conflict_a); /* ?!? This seems wrong. */ bitmap_set_bit (consideration_allocno_bitmap, ALLOCNO_NUM (conflict_a)); } } } } if (num > 1) qsort (spilled_pseudo_regs, num, sizeof (int), pseudo_reg_compare); changed_p = false; /* Try to assign hard registers to pseudos from SPILLED_PSEUDO_REGS. */ for (i = 0; i < num; i++) { regno = spilled_pseudo_regs[i]; COPY_HARD_REG_SET (forbidden_regs, bad_spill_regs); IOR_HARD_REG_SET (forbidden_regs, pseudo_forbidden_regs[regno]); IOR_HARD_REG_SET (forbidden_regs, pseudo_previous_regs[regno]); gcc_assert (reg_renumber[regno] < 0); a = ira_regno_allocno_map[regno]; ira_mark_allocation_change (regno); ira_assert (reg_renumber[regno] < 0); if (internal_flag_ira_verbose > 3 && ira_dump_file != NULL) fprintf (ira_dump_file, " Try Assign %d(a%d), cost=%d", regno, ALLOCNO_NUM (a), ALLOCNO_MEMORY_COST (a) - ALLOCNO_CLASS_COST (a)); allocno_reload_assign (a, forbidden_regs); if (reg_renumber[regno] >= 0) { CLEAR_REGNO_REG_SET (spilled, regno); changed_p = true; } } BITMAP_FREE (temp); return changed_p; } /* The function is called by reload and returns already allocated stack slot (if any) for REGNO with given INHERENT_SIZE and TOTAL_SIZE. In the case of failure to find a slot which can be used for REGNO, the function returns NULL. */ rtx ira_reuse_stack_slot (int regno, unsigned int inherent_size, unsigned int total_size) { unsigned int i; int slot_num, best_slot_num; int cost, best_cost; ira_copy_t cp, next_cp; ira_allocno_t another_allocno, allocno = ira_regno_allocno_map[regno]; rtx x; bitmap_iterator bi; struct ira_spilled_reg_stack_slot *slot = NULL; ira_assert (inherent_size == PSEUDO_REGNO_BYTES (regno) && inherent_size <= total_size && ALLOCNO_HARD_REGNO (allocno) < 0); if (! flag_ira_share_spill_slots) return NULL_RTX; slot_num = -ALLOCNO_HARD_REGNO (allocno) - 2; if (slot_num != -1) { slot = &ira_spilled_reg_stack_slots[slot_num]; x = slot->mem; } else { best_cost = best_slot_num = -1; x = NULL_RTX; /* It means that the pseudo was spilled in the reload pass, try to reuse a slot. */ for (slot_num = 0; slot_num < ira_spilled_reg_stack_slots_num; slot_num++) { slot = &ira_spilled_reg_stack_slots[slot_num]; if (slot->mem == NULL_RTX) continue; if (slot->width < total_size || GET_MODE_SIZE (GET_MODE (slot->mem)) < inherent_size) continue; EXECUTE_IF_SET_IN_BITMAP (&slot->spilled_regs, FIRST_PSEUDO_REGISTER, i, bi) { another_allocno = ira_regno_allocno_map[i]; if (allocnos_conflict_by_live_ranges_p (allocno, another_allocno)) goto cont; } for (cost = 0, cp = ALLOCNO_COPIES (allocno); cp != NULL; cp = next_cp) { if (cp->first == allocno) { next_cp = cp->next_first_allocno_copy; another_allocno = cp->second; } else if (cp->second == allocno) { next_cp = cp->next_second_allocno_copy; another_allocno = cp->first; } else gcc_unreachable (); if (cp->insn == NULL_RTX) continue; if (bitmap_bit_p (&slot->spilled_regs, ALLOCNO_REGNO (another_allocno))) cost += cp->freq; } if (cost > best_cost) { best_cost = cost; best_slot_num = slot_num; } cont: ; } if (best_cost >= 0) { slot_num = best_slot_num; slot = &ira_spilled_reg_stack_slots[slot_num]; SET_REGNO_REG_SET (&slot->spilled_regs, regno); x = slot->mem; ALLOCNO_HARD_REGNO (allocno) = -slot_num - 2; } } if (x != NULL_RTX) { ira_assert (slot->width >= total_size); #ifdef ENABLE_IRA_CHECKING EXECUTE_IF_SET_IN_BITMAP (&slot->spilled_regs, FIRST_PSEUDO_REGISTER, i, bi) { ira_assert (! conflict_by_live_ranges_p (regno, i)); } #endif SET_REGNO_REG_SET (&slot->spilled_regs, regno); if (internal_flag_ira_verbose > 3 && ira_dump_file) { fprintf (ira_dump_file, " Assigning %d(freq=%d) slot %d of", regno, REG_FREQ (regno), slot_num); EXECUTE_IF_SET_IN_BITMAP (&slot->spilled_regs, FIRST_PSEUDO_REGISTER, i, bi) { if ((unsigned) regno != i) fprintf (ira_dump_file, " %d", i); } fprintf (ira_dump_file, "\n"); } } return x; } /* This is called by reload every time a new stack slot X with TOTAL_SIZE was allocated for REGNO. We store this info for subsequent ira_reuse_stack_slot calls. */ void ira_mark_new_stack_slot (rtx x, int regno, unsigned int total_size) { struct ira_spilled_reg_stack_slot *slot; int slot_num; ira_allocno_t allocno; ira_assert (PSEUDO_REGNO_BYTES (regno) <= total_size); allocno = ira_regno_allocno_map[regno]; slot_num = -ALLOCNO_HARD_REGNO (allocno) - 2; if (slot_num == -1) { slot_num = ira_spilled_reg_stack_slots_num++; ALLOCNO_HARD_REGNO (allocno) = -slot_num - 2; } slot = &ira_spilled_reg_stack_slots[slot_num]; INIT_REG_SET (&slot->spilled_regs); SET_REGNO_REG_SET (&slot->spilled_regs, regno); slot->mem = x; slot->width = total_size; if (internal_flag_ira_verbose > 3 && ira_dump_file) fprintf (ira_dump_file, " Assigning %d(freq=%d) a new slot %d\n", regno, REG_FREQ (regno), slot_num); } /* Return spill cost for pseudo-registers whose numbers are in array REGNOS (with a negative number as an end marker) for reload with given IN and OUT for INSN. Return also number points (through EXCESS_PRESSURE_LIVE_LENGTH) where the pseudo-register lives and the register pressure is high, number of references of the pseudo-registers (through NREFS), number of callee-clobbered hard-registers occupied by the pseudo-registers (through CALL_USED_COUNT), and the first hard regno occupied by the pseudo-registers (through FIRST_HARD_REGNO). */ static int calculate_spill_cost (int *regnos, rtx in, rtx out, rtx insn, int *excess_pressure_live_length, int *nrefs, int *call_used_count, int *first_hard_regno) { int i, cost, regno, hard_regno, j, count, saved_cost, nregs; bool in_p, out_p; int length; ira_allocno_t a; *nrefs = 0; for (length = count = cost = i = 0;; i++) { regno = regnos[i]; if (regno < 0) break; *nrefs += REG_N_REFS (regno); hard_regno = reg_renumber[regno]; ira_assert (hard_regno >= 0); a = ira_regno_allocno_map[regno]; length += ALLOCNO_EXCESS_PRESSURE_POINTS_NUM (a) / ALLOCNO_NUM_OBJECTS (a); cost += ALLOCNO_MEMORY_COST (a) - ALLOCNO_CLASS_COST (a); nregs = hard_regno_nregs[hard_regno][ALLOCNO_MODE (a)]; for (j = 0; j < nregs; j++) if (! TEST_HARD_REG_BIT (call_used_reg_set, hard_regno + j)) break; if (j == nregs) count++; in_p = in && REG_P (in) && (int) REGNO (in) == hard_regno; out_p = out && REG_P (out) && (int) REGNO (out) == hard_regno; if ((in_p || out_p) && find_regno_note (insn, REG_DEAD, hard_regno) != NULL_RTX) { saved_cost = 0; if (in_p) saved_cost += ira_memory_move_cost [ALLOCNO_MODE (a)][ALLOCNO_CLASS (a)][1]; if (out_p) saved_cost += ira_memory_move_cost [ALLOCNO_MODE (a)][ALLOCNO_CLASS (a)][0]; cost -= REG_FREQ_FROM_BB (BLOCK_FOR_INSN (insn)) * saved_cost; } } *excess_pressure_live_length = length; *call_used_count = count; hard_regno = -1; if (regnos[0] >= 0) { hard_regno = reg_renumber[regnos[0]]; } *first_hard_regno = hard_regno; return cost; } /* Return TRUE if spilling pseudo-registers whose numbers are in array REGNOS is better than spilling pseudo-registers with numbers in OTHER_REGNOS for reload with given IN and OUT for INSN. The function used by the reload pass to make better register spilling decisions. */ bool ira_better_spill_reload_regno_p (int *regnos, int *other_regnos, rtx in, rtx out, rtx insn) { int cost, other_cost; int length, other_length; int nrefs, other_nrefs; int call_used_count, other_call_used_count; int hard_regno, other_hard_regno; cost = calculate_spill_cost (regnos, in, out, insn, &length, &nrefs, &call_used_count, &hard_regno); other_cost = calculate_spill_cost (other_regnos, in, out, insn, &other_length, &other_nrefs, &other_call_used_count, &other_hard_regno); if (nrefs == 0 && other_nrefs != 0) return true; if (nrefs != 0 && other_nrefs == 0) return false; if (cost != other_cost) return cost < other_cost; if (length != other_length) return length > other_length; #ifdef REG_ALLOC_ORDER if (hard_regno >= 0 && other_hard_regno >= 0) return (inv_reg_alloc_order[hard_regno] < inv_reg_alloc_order[other_hard_regno]); #else if (call_used_count != other_call_used_count) return call_used_count > other_call_used_count; #endif return false; } /* Allocate and initialize data necessary for assign_hard_reg. */ void ira_initiate_assign (void) { sorted_allocnos = (ira_allocno_t *) ira_allocate (sizeof (ira_allocno_t) * ira_allocnos_num); consideration_allocno_bitmap = ira_allocate_bitmap (); initiate_cost_update (); allocno_priorities = (int *) ira_allocate (sizeof (int) * ira_allocnos_num); sorted_copies = (ira_copy_t *) ira_allocate (ira_copies_num * sizeof (ira_copy_t)); } /* Deallocate data used by assign_hard_reg. */ void ira_finish_assign (void) { ira_free (sorted_allocnos); ira_free_bitmap (consideration_allocno_bitmap); finish_cost_update (); ira_free (allocno_priorities); ira_free (sorted_copies); } /* Entry function doing color-based register allocation. */ static void color (void) { allocno_stack_vec.create (ira_allocnos_num); memset (allocated_hardreg_p, 0, sizeof (allocated_hardreg_p)); ira_initiate_assign (); do_coloring (); ira_finish_assign (); allocno_stack_vec.release (); move_spill_restore (); } /* This page contains a simple register allocator without usage of allocno conflicts. This is used for fast allocation for -O0. */ /* Do register allocation by not using allocno conflicts. It uses only allocno live ranges. The algorithm is close to Chow's priority coloring. */ static void fast_allocation (void) { int i, j, k, num, class_size, hard_regno; #ifdef STACK_REGS bool no_stack_reg_p; #endif enum reg_class aclass; enum machine_mode mode; ira_allocno_t a; ira_allocno_iterator ai; live_range_t r; HARD_REG_SET conflict_hard_regs, *used_hard_regs; sorted_allocnos = (ira_allocno_t *) ira_allocate (sizeof (ira_allocno_t) * ira_allocnos_num); num = 0; FOR_EACH_ALLOCNO (a, ai) sorted_allocnos[num++] = a; allocno_priorities = (int *) ira_allocate (sizeof (int) * ira_allocnos_num); setup_allocno_priorities (sorted_allocnos, num); used_hard_regs = (HARD_REG_SET *) ira_allocate (sizeof (HARD_REG_SET) * ira_max_point); for (i = 0; i < ira_max_point; i++) CLEAR_HARD_REG_SET (used_hard_regs[i]); qsort (sorted_allocnos, num, sizeof (ira_allocno_t), allocno_priority_compare_func); for (i = 0; i < num; i++) { int nr, l; a = sorted_allocnos[i]; nr = ALLOCNO_NUM_OBJECTS (a); CLEAR_HARD_REG_SET (conflict_hard_regs); for (l = 0; l < nr; l++) { ira_object_t obj = ALLOCNO_OBJECT (a, l); IOR_HARD_REG_SET (conflict_hard_regs, OBJECT_CONFLICT_HARD_REGS (obj)); for (r = OBJECT_LIVE_RANGES (obj); r != NULL; r = r->next) for (j = r->start; j <= r->finish; j++) IOR_HARD_REG_SET (conflict_hard_regs, used_hard_regs[j]); } aclass = ALLOCNO_CLASS (a); ALLOCNO_ASSIGNED_P (a) = true; ALLOCNO_HARD_REGNO (a) = -1; if (hard_reg_set_subset_p (reg_class_contents[aclass], conflict_hard_regs)) continue; mode = ALLOCNO_MODE (a); #ifdef STACK_REGS no_stack_reg_p = ALLOCNO_NO_STACK_REG_P (a); #endif class_size = ira_class_hard_regs_num[aclass]; for (j = 0; j < class_size; j++) { hard_regno = ira_class_hard_regs[aclass][j]; #ifdef STACK_REGS if (no_stack_reg_p && FIRST_STACK_REG <= hard_regno && hard_regno <= LAST_STACK_REG) continue; #endif if (ira_hard_reg_set_intersection_p (hard_regno, mode, conflict_hard_regs) || (TEST_HARD_REG_BIT (ira_prohibited_class_mode_regs[aclass][mode], hard_regno))) continue; ALLOCNO_HARD_REGNO (a) = hard_regno; for (l = 0; l < nr; l++) { ira_object_t obj = ALLOCNO_OBJECT (a, l); for (r = OBJECT_LIVE_RANGES (obj); r != NULL; r = r->next) for (k = r->start; k <= r->finish; k++) IOR_HARD_REG_SET (used_hard_regs[k], ira_reg_mode_hard_regset[hard_regno][mode]); } break; } } ira_free (sorted_allocnos); ira_free (used_hard_regs); ira_free (allocno_priorities); if (internal_flag_ira_verbose > 1 && ira_dump_file != NULL) ira_print_disposition (ira_dump_file); } /* Entry function doing coloring. */ void ira_color (void) { ira_allocno_t a; ira_allocno_iterator ai; /* Setup updated costs. */ FOR_EACH_ALLOCNO (a, ai) { ALLOCNO_UPDATED_MEMORY_COST (a) = ALLOCNO_MEMORY_COST (a); ALLOCNO_UPDATED_CLASS_COST (a) = ALLOCNO_CLASS_COST (a); } if (ira_conflicts_p) color (); else fast_allocation (); }