/**
* @file
- * @brief Loop peeling and unrolling
* @author Christian Helmer
+ * @brief loop inversion and loop unrolling, loop peeling
+ *
* @version $Id$
*/
#include "config.h"
+#include "iroptimize.h"
+#include "opt_init.h"
#include "irnode.h"
#include "debug.h"
#include "irouts.h"
#include "iredges.h"
#include "irtools.h"
-#include "array_t.h" /* automatic array */
-#include "beutil.h" /* get_block */
-
-// TODO during DBG
-//#include "irnode_t.h"
-#include "irdump.h"
+#include "array_t.h"
+#include "beutil.h"
+#include "irloop_t.h"
+#include "irpass.h"
DEBUG_ONLY(static firm_dbg_module_t *dbg);
/* current loop */
static ir_loop *cur_loop;
-/* The loop walker should be possible to abort if nothing can be done anymore */
+/* abortable walker function */
typedef unsigned irg_walk_func_abortable(ir_node *, void *);
-/* condition for breaking a copy_walk */
+/* condition for walking a node during a copy_walk */
typedef unsigned walker_condition(ir_node *);
-/* stores node and position of a predecessor */
+/* node and position of a predecessor */
typedef struct out_edges {
ir_node *node;
int pred_irn_n;
-} out_edges;
+} out_edge;
/* access complex values through the nodes links */
typedef struct node_info {
ir_node *copy;
ir_node *link; /* temporary links for ssa creation */
ir_node **ins; /* ins for phi nodes, during rewiring of blocks */
+ unsigned done;
struct node_info *freelistnext; /* linked list to free all node_infos */
} node_info;
static node_info *link_node_state_list; /* head of the linked list to free all node_infos */
-static out_edges *cur_loop_outs; /* A walker may start visiting the current loop with these nodes. */
-static out_edges *cur_head_outs; /* A walker may start visiting the cur head with these nodes. */
+static out_edge *cur_loop_outs; /* A walker may start visiting the current loop with these nodes. */
+static out_edge *cur_head_outs; /* A walker may start visiting the cur head with these nodes. */
static ir_node *loop_cf_head = NULL; /* Loop head node */
static unsigned loop_cf_head_valid = 1; /* A loop may have one head, otherwise we do not touch it. */
+ir_node **loops;
+
/* Inverted head */
static ir_node *loop_inv_head = NULL;
/* Peeled head */
/* Loop analysis informations */
typedef struct loop_info_t {
- unsigned calls;
- unsigned loads;
- unsigned invariant_loads; /* number of load nodes */
- unsigned stores; /* number of store nodes */
unsigned blocks; /* number of blocks in the loop */
- unsigned opnodes_n; /* nodes that should result in an instruction */
- unsigned opnodes_head;
+ unsigned calls; /* number of calls */
+ unsigned loads; /* number of load nodes */
+ unsigned outs; /* outs without keepalives */
+#if 0
+ unsigned invariant_loads;
+ unsigned stores; /* number of store nodes */
+ unsigned opnodes_n; /* nodes that probably result in an instruction */
+ unsigned do_invariant_opt;
+#endif
} loop_info_t;
/* Information about the current loop */
static loop_info_t loop_info;
/* A walker may start visiting a condition chain with these nodes. */
-static out_edges *cond_chain_entries;
+static out_edge *cond_chain_entries;
+
+/* Number of unrolling */
+int unroll_times;
static unsigned head_inversion_node_count;
-static unsigned head_inversion_node_limit;
-//static unsigned head_inversion_block_count;
+static unsigned inversion_head_node_limit;
+static unsigned head_inversion_block_count;
+
+static unsigned enable_peeling;
+static unsigned enable_inversion;
+static unsigned enable_unrolling;
/**
*
* ============= AUXILIARY FUNCTIONS =====================================
*/
+
/**
* Creates object on the heap, and adds it to a linked list to free it later.
*/
-static node_info *new_node_info(void) {
+static node_info *new_node_info(void)
+{
node_info *l = XMALLOCZ(node_info);
l->freelistnext = link_node_state_list;
link_node_state_list = l;
/* Allocates a node_info struct for the given node. For use with a walker. */
static void alloc_node_info(ir_node *node, void *env)
{
- node_info *state = new_node_info();
+ node_info *state;
(void) env;
+ state = new_node_info();
set_irn_link(node, (void *)state);
}
static void free_node_info(void)
{
- node_info *next;
- next = link_node_state_list;
- while(next->freelistnext) {
- node_info *cur = next;
- next = cur->freelistnext;
- xfree( cur );
+ int a = 0;
+ node_info *n;
+ n = link_node_state_list;
+ while (n) {
+ node_info *next = n->freelistnext;
+ ++a;
+ xfree(n);
+ n = next;
}
+ link_node_state_list = NULL;
}
/**
{
node_info *next;
next = link_node_state_list;
- while(next->freelistnext) {
+ while (next->freelistnext) {
node_info *cur = next;
next = cur->freelistnext;
cur->copy = NULL;
}
}
-/* Returns the */
+/* Returns the nodes node_info link. */
+static ir_node *get_link(ir_node *n)
+{
+ return ((node_info *)get_irn_link(n))->link;
+}
+
+/* Sets the nodes node_info link. */
+static void set_link(ir_node *n, ir_node *link)
+{
+ ((node_info *)get_irn_link(n))->link = link;
+}
+
+/* Returns a nodes copy. */
static ir_node *get_copy(ir_node *n)
{
return ((node_info *)get_irn_link(n))->copy;
}
-/* Links the node to its copy */
+/* Sets a nodes copy. */
static void set_copy(ir_node *n, ir_node *copy)
{
((node_info *)get_irn_link(n) )->copy = copy;
}
-/* Returns 0 if the node or block is not in cur_loop */
+/**
+ * Convenience macro for iterating over every copy in a linked list
+ * of copies.
+ */
+#define for_each_copy(node) \
+ for ( ; (node) ; (node) = get_copy(node))
+
+/**
+ * Convenience macro for iterating over every copy in 2 linked lists
+ * of copies in parallel.
+ */
+#define for_each_copy2(high1, low1, high2, low2) \
+ for ( ; (low1) && (low2); (high1) = (low1), (low1) = get_copy(low1), \
+ (high2) = (low2), (low2) = get_copy(low2))
+
+/*
+ * Returns 0 if the node or block is not in cur_loop.
+ */
static unsigned is_in_loop(ir_node *node)
{
return (get_irn_loop(get_block(node)) == cur_loop);
return(!is_in_loop(get_irn_n(n, i)));
}
-/* used for walker */
-static void unmark_block(ir_node *node, void * env)
+/* used for block walker */
+static void reset_block_mark(ir_node *node, void * env)
{
(void) env;
-// DB((dbg, LEVEL_1, "UNMARK %ld\n", node->node_nr));
- if(is_Block(node))
+
+ if (is_Block(node))
set_Block_mark(node, 0);
}
return (get_Block_mark(get_block(node)));
}
+/* Returns the number of blocks in a loop. */
+static int get_loop_n_blocks(ir_loop *loop)
+{
+ int elements, e;
+ int blocks = 0;
+ elements = get_loop_n_elements(loop);
+
+ for (e=0; e<elements; e++) {
+ loop_element elem = get_loop_element(loop, e);
+ if (is_ir_node(elem.kind) && is_Block(elem.node))
+ ++blocks;
+ }
+ return blocks;
+}
+
/**
* Add newpred at position pos to node and also add the corresponding value to the phis.
* Requires block phi list.
*/
-static void duplicate_preds(ir_node* node, unsigned pos, ir_node* newpred)
+static int duplicate_preds(ir_node* block, unsigned pos, ir_node* newpred)
{
- ir_node** ins;
+ ir_node **ins;
+ /*int *is_be;*/
ir_node *phi;
int block_arity;
int i;
- assert(is_Block(node) && "duplicate_preds is only allowed for blocks");
+ assert(is_Block(block) && "duplicate_preds may be called for blocks only");
+
+ DB((dbg, LEVEL_5, "duplicate_preds(node %N, pos %d, newpred %N)\n", block, pos, newpred));
- DB((dbg, LEVEL_5, "duplicate_preds(node %ld, pos %d, newpred %ld)\n", get_irn_node_nr(node), pos, get_irn_node_nr(newpred)));
+ block_arity = get_irn_arity(block);
- block_arity = get_irn_arity(node);
+ NEW_ARR_A(ir_node*, ins, block_arity + 1);
- NEW_ARR_A(ir_node*, ins, block_arity + 1 );
- for (i = 0; i < block_arity; ++i)
- ins[i] = get_irn_n(node, i);
+ for (i = 0; i < block_arity; ++i) {
+ ins[i] = get_irn_n(block, i);
+ }
ins[block_arity] = newpred;
- set_irn_in(node, block_arity + 1, ins);
+ set_irn_in(block, block_arity + 1, ins);
- for_each_phi(node, phi) {
+ /* LDBG 1 */
+#if 1
+ for_each_phi(block, phi) {
int phi_arity = get_irn_arity(phi);
- DB((dbg, LEVEL_5, "duplicate_preds: fixing phi %ld\n", get_irn_node_nr(phi)));
+ DB((dbg, LEVEL_5, "duplicate_preds: fixing phi %N\n", phi));
NEW_ARR_A(ir_node *, ins, block_arity + 1);
- for(i=0; i < phi_arity; ++i) {
- DB((dbg, LEVEL_5, "in %ld\n", get_irn_node_nr(get_irn_n(phi, i))));
+ for (i = 0; i < phi_arity; ++i) {
+ DB((dbg, LEVEL_5, "pos %N\n", get_irn_n(phi, i)));
ins[i] = get_irn_n(phi, i);
}
- ins[block_arity] = get_irn_n(phi, pos);
+ ins[block_arity] = get_copy(get_irn_n(phi, pos));
set_irn_in(phi, block_arity + 1, ins);
}
-}
-
-/* Walks through all nodes of cur_loop */
-static unsigned loop_walker_rec(ir_node *node,
- irg_walk_func_abortable *pre,
- irg_walk_func_abortable *post, void * env)
-{
- int i;
- unsigned stop = 0;
- ir_graph *irg = get_current_ir_graph();
-
- /* RETURN if we walked out of the loop*/
- if (!is_in_loop(node))
- return 0;
-
- if (pre) {
- unsigned stop = pre(node, env);
- if (stop)
- return stop;
- }
-
- set_irn_visited(node, get_irg_visited(irg));
-
- if (get_irn_op(node) != op_Block) {
- ir_node *pred = get_irn_n(node, -1);
- if (get_irn_visited(pred) < get_irg_visited(irg))
- {
- stop = loop_walker_rec(pred, pre, post, env);
- if (stop)
- return stop;
- }
- }
-
- for (i = get_irn_arity(node) - 1; i >= 0; --i) {
- ir_node *pred = get_irn_n(node, i);
- if (get_irn_visited(pred) < get_irg_visited(irg)) {
- stop = loop_walker_rec(pred, pre, post, env);
- if (stop)
- return stop;
- }
- }
-
- if (post)
- return post(node, env);
- return 0;
+#endif
+ return block_arity;
}
/**
- * Walks through loop nodes.
- * The entries of the loop (all edges pointing into the loop) have to be given.
+ * Finds loop head and loop_info.
*/
-static unsigned loop_walker(out_edges *entries,
- irg_walk_func_abortable *pre, irg_walk_func_abortable *post, void * env)
-{
- int i;
- int stop = 0;
-
- for (i=0; !stop && i<ARR_LEN(entries); i++) {
- ir_node *pred;
- out_edges entry;
- entry = entries[i];
-
- pred = get_irn_n( entry.node , entry.pred_irn_n);
-
- stop = loop_walker_rec( pred, pre, post, env);
- }
- return stop;
-}
-
-
-/* Adds all nodes pointing into the loop to loop_entries and also finds the loops head */
-static void get_loop_outs_and_info(ir_node *node, void *env)
+static void get_loop_info(ir_node *node, void *env)
{
unsigned node_in_loop, pred_in_loop;
int i, arity;
/* collect some loop information */
if (node_in_loop) {
- if ( !is_Store(node) )
- ++loop_info.stores;
- if ( !is_Load(node) )
- ++loop_info.loads;
- if ( !is_Call(node) )
+ if (is_Call(node))
++loop_info.calls;
- if ( !is_Block(node) && !is_Proj(node) && !is_Phi(node) )
- ++loop_info.opnodes_n;
}
- //Find the loops head/the blocks with cfpred outside of the loop
+ /* Find the loops head/the blocks with cfpred outside of the loop */
if (is_Block(node) && node_in_loop && !pred_in_loop && loop_cf_head_valid) {
ir_node *cfgpred = get_Block_cfgpred(node, i);
- if ( !is_in_loop(cfgpred) ) {
- //DB((dbg, LEVEL_1, "potential head %+F\n", node));
+
+ if (!is_in_loop(cfgpred)) {
+ DB((dbg, LEVEL_5, "potential head %+F because inloop and pred %+F not inloop\n", node, pred));
/* another head? We do not touch this. */
if (loop_cf_head && loop_cf_head != node) {
loop_cf_head_valid = 0;
}
}
}
-
- if ( pred_in_loop && !node_in_loop ) {
- out_edges entry;
- entry.node = node;
- entry.pred_irn_n = i;
- ARR_APP1(out_edges, cur_loop_outs, entry);
- }
}
}
-/**
- * Finds invariant loads and marks them as invariant.
- * (has to be post walk)
- */
-static unsigned get_invariants(ir_node *node, void *env)
+/* Adds all nodes pointing into the loop to loop_entries and also finds the loops head */
+static void get_loop_outs(ir_node *node, void *env)
{
- unsigned invar = 1;
- int arity = get_irn_arity(node);
+ unsigned node_in_loop, pred_in_loop;
+ int i, arity;
(void) env;
- /* RETURN, no preds to visit */
- if (arity == 0) return 0;
-
- if (is_Load(node)) {
- assert(arity>=2 && "expected load node to have in[1] (address)");
-
- ir_node *pred = get_irn_n(node, 1);
- if ( (get_Load_volatility(node) == volatility_non_volatile) &
- (!is_in_loop(pred)
- || is_Const(pred)
- || is_SymConst(pred)
- || get_node_info(node)->invariant ) ) {
- get_node_info(node)->invariant = 1;
- ++loop_info.invariant_loads;
- } else
- {
- get_node_info(node)->invariant = 0;
- }
- } else {
- int i;
- invar = 1;
- /* find loop variant preds */
- for(i = 0; i < arity; ++i) {
- ir_node *pred = get_irn_n(node, i);
-
- if ( is_in_loop(pred) /* outside loop is loop invariant */
- && !is_Const(pred) /* constants */
- && !is_SymConst(pred) /* SymConst */
- && !get_node_info(node)->invariant ) { /* pred is marked as invariant */
- invar = 0;
- }
- }
+ arity = get_irn_arity(node);
+ for (i = 0; i < arity; ++i) {
+ ir_node *pred = get_irn_n(node, i);
- if (invar) {
- get_node_info(node)->invariant = 1;
- } else {
- get_node_info(node)->invariant = 0;
+ pred_in_loop = is_in_loop(pred);
+ node_in_loop = is_in_loop(node);
+
+ if (pred_in_loop && !node_in_loop) {
+ out_edge entry;
+ entry.node = node;
+ entry.pred_irn_n = i;
+ ARR_APP1(out_edge, cur_loop_outs, entry);
}
}
- return 0;
}
-
static ir_node *ssa_second_def;
static ir_node *ssa_second_def_block;
/**
- * Walks the graph bottom up, searching for definitions and create phis.
- * (Does not handle the special case where the second definition is in the block of the user
- * of the original definition because it is not necessary here.)
+ * Walks the graph bottom up, searching for definitions and creates phis.
*/
static ir_node *search_def_and_create_phis(ir_node *block, ir_mode *mode)
{
ir_node *phi;
ir_node **in;
- DB((dbg, LEVEL_5, "ssa sdacp: block %ld\n", get_irn_node_nr(block)));
+ DB((dbg, LEVEL_5, "ssa search_def_and_create_phis: block %N\n", block));
/* Prevents creation of phi that would be bad anyway.
* Dead and bad blocks. */
return new_Bad();
if (block == ssa_second_def_block) {
- DB((dbg, LEVEL_5, "ssa found second definition: use second def %ld\n", get_irn_node_nr(ssa_second_def)));
+ DB((dbg, LEVEL_5, "ssa found second definition: use second def %N\n", ssa_second_def));
return ssa_second_def;
}
/* already processed this block? */
if (irn_visited(block)) {
- ir_node *value = get_node_info(block)->link;
- DB((dbg, LEVEL_5, "ssa already visited: use linked %ld\n", get_irn_node_nr(value)));
+ ir_node *value = get_link(block);
+ DB((dbg, LEVEL_5, "ssa already visited: use linked %N\n", value));
return value;
}
ir_node *pred_block = get_Block_cfgpred_block(block, 0);
ir_node *value;
- DB((dbg, LEVEL_5, "ssa 1 pred: walk pred %ld\n", get_irn_node_nr(pred_block)));
+ DB((dbg, LEVEL_5, "ssa 1 pred: walk pred %N\n", pred_block));
value = search_def_and_create_phis(pred_block, mode);
- get_node_info(block)->link = value;
+ set_link(block, value);
mark_irn_visited(block);
return value;
/* create a new Phi */
NEW_ARR_A(ir_node*, in, n_cfgpreds);
- for(i = 0; i < n_cfgpreds; ++i)
+ for (i = 0; i < n_cfgpreds; ++i)
in[i] = new_Unknown(mode);
phi = new_r_Phi(block, n_cfgpreds, in, mode);
/* EVERY node is assumed to have a node_info linked. */
alloc_node_info(phi, NULL);
- DB((dbg, LEVEL_5, "ssa phi creation: link new phi %ld to block %ld\n", get_irn_node_nr(phi), get_irn_node_nr(block)));
+ DB((dbg, LEVEL_5, "ssa phi creation: link new phi %N to block %N\n", phi, block));
- get_node_info(block)->link = phi;
+ set_link(block, phi);
mark_irn_visited(block);
/* set Phi predecessors */
- for(i = 0; i < n_cfgpreds; ++i) {
+ for (i = 0; i < n_cfgpreds; ++i) {
+ ir_node *pred_val;
ir_node *pred_block = get_Block_cfgpred_block(block, i);
- ir_node *pred_val = search_def_and_create_phis(pred_block, mode);
- DB((dbg, LEVEL_5, "ssa phi pred:phi %ld, pred %ld\n", get_irn_node_nr(phi), get_irn_node_nr(pred_val)));
+ assert(pred_block != NULL);
+ pred_val = search_def_and_create_phis(pred_block, mode);
+ assert(pred_val != NULL);
+
+ DB((dbg, LEVEL_5, "ssa phi pred:phi %N, pred %N\n", phi, pred_val));
set_irn_n(phi, i, pred_val);
}
-
return phi;
}
/**
* Given a set of values this function constructs SSA-form for the users of the
* first value (the users are determined through the out-edges of the value).
- * Uses the irn_visited flags. Works without using the dominance tree.
+ * Works without using the dominance tree.
*/
static void construct_ssa(ir_node *orig_block, ir_node *orig_val,
ir_node *second_block, ir_node *second_val)
inc_irg_visited(irg);
mode = get_irn_mode(orig_val);
- get_node_info(orig_block)->link = orig_val;
+ set_link(orig_block, orig_val);
mark_irn_visited(orig_block);
ssa_second_def_block = second_block;
if (is_End(user))
continue;
- DB((dbg, LEVEL_5, "original user %ld\n", get_irn_node_nr(user)));
+ DB((dbg, LEVEL_5, "original user %N\n", user));
if (is_Phi(user)) {
ir_node *pred_block = get_Block_cfgpred_block(user_block, j);
ir_free_resources(irg, IR_RESOURCE_IRN_VISITED);
}
+/*
+ * Construct SSA for def and all of its copies.
+ */
+static void construct_ssa_n(ir_node *def, ir_node *user)
+{
+ ir_graph *irg;
+ ir_mode *mode;
+ ir_node *iter = def;
+ const ir_edge_t *edge;
+ const ir_edge_t *next;
+ irg = get_irn_irg(def);
+
+ ir_reserve_resources(irg, IR_RESOURCE_IRN_VISITED);
+ inc_irg_visited(irg);
+
+ mode = get_irn_mode(def);
+
+ for_each_copy(iter) {
+ set_link(get_nodes_block(iter), iter);
+ mark_irn_visited(get_nodes_block(iter));
+
+ DB((dbg, LEVEL_5, "ssa_n: Link def %N to block %N\n",
+ iter, get_nodes_block(iter)));
+ }
+
+ /* Need to search the outs, because we need the in-pos on the user node. */
+ foreach_out_edge_safe(def, edge, next) {
+ ir_node *edge_user = get_edge_src_irn(edge);
+ int edge_src = get_edge_src_pos(edge);
+ ir_node *user_block = get_nodes_block(user);
+ ir_node *newval;
+
+ if (edge_user != user)
+ continue;
+
+ if (is_Phi(user)) {
+ ir_node *pred_block = get_Block_cfgpred_block(user_block, edge_src);
+ newval = search_def_and_create_phis(pred_block, mode);
+ } else {
+ newval = search_def_and_create_phis(user_block, mode);
+ }
+
+ if (newval != user && !is_Bad(newval))
+ set_irn_n(user, edge_src, newval);
+ }
+
+ ir_free_resources(irg, IR_RESOURCE_IRN_VISITED);
+}
+
+/**
+ * Construct SSA for all definitions in arr.
+ */
+static void construct_ssa_foreach(ir_node **arr, int arr_n)
+{
+ int i;
+ for (i = 0; i < arr_n ; ++i) {
+ ir_node *cppred, *block, *cpblock, *pred;
+
+ pred = arr[i];
+ cppred = get_copy(pred);
+ block = get_nodes_block(pred);
+ cpblock = get_nodes_block(cppred);
+ construct_ssa(block, pred, cpblock, cppred);
+ }
+}
+
/* get the number of backedges without alien bes */
static int get_backedge_n(ir_node *loophead, unsigned with_alien)
{
int arity = get_irn_arity(loophead);
for (i = 0; i < arity; ++i) {
ir_node *pred = get_irn_n(loophead, i);
- if (is_backedge(loophead, i) && ( with_alien || is_in_loop(pred)) )
+ if (is_backedge(loophead, i) && (with_alien || is_in_loop(pred)))
++be_n;
}
return be_n;
}
-/**
- * Sets the nodes backedges, according to its predecessors link.
- */
-static void fix_backedge_info(ir_node *node)
-{
- int i;
- for (i = 0; i < get_irn_arity(node); ++i)
- {
- ir_node *pred = get_irn_n(node, i);
- if (get_node_info(pred)->link != NULL)
- set_backedge(node, i);
- else
- set_not_backedge(node, i);
- }
-}
-
-/**
- *
- * ============= PEELING =====================================
- *
- */
-
/**
* Rewires the heads after peeling.
*/
*/
if (is_backedge(loophead, i) && !is_alien_edge(loophead, i)) {
loopheadnins[lheadin_c] = orgjmp;
- /* marks out edge as backedge */
- get_node_info(orgjmp)->link = orgjmp;
for_each_phi(loophead, phi) {
get_node_info( phi )->ins[lheadin_c] = get_irn_n( phi, i) ;
}
++lheadin_c;
- loopheadnins[lheadin_c] = copyjmp; /* former bes of the peeled code origin now from the loophead */
- /* marks out edge as normal edge */
- get_node_info(copyjmp)->link = NULL;
- /* get_irn_n( get_copy_of(phi), i) <!=> get_copy_of(get_irn_n( phi, i))
+ /* former bes of the peeled code origin now from the loophead */
+ loopheadnins[lheadin_c] = copyjmp;
+
+ /* get_irn_n( get_copy_of(phi), i ) <!=> get_copy_of( get_irn_n( phi, i) )
* Order is crucial! Predecessors outside of the loop are non existent.
* The copy (cloned with its ins!) has pred i,
* but phis pred i might not have a copy of itself.
*/
for_each_phi(loophead, phi) {
- //printf("normalbe phi %ld @ %d -> %ld\n", phi->node_nr, i, get_irn_n( get_copy_of(phi), i)->node_nr);
get_node_info( phi )->ins[lheadin_c] = get_irn_n( get_copy(phi), i) ;
}
++lheadin_c;
} else {
peelheadnins[pheadin_c] = orgjmp;
- /* marks out edge as normal edge */
- get_node_info(orgjmp)->link = NULL;
for_each_phi(peelhead, phi) {
get_node_info( phi )->ins[pheadin_c] = get_irn_n(phi, i);
}
}
}/* for */
- //DBG
assert(pheadin_c == ARR_LEN(peelheadnins) &&
lheadin_c == ARR_LEN(loopheadnins) &&
"the constructed head arities do not match the predefined arities");
- /**
- * assign the ins to the nodes
- */
+ /* assign the ins to the nodes */
set_irn_in(loophead, ARR_LEN(loopheadnins), loopheadnins);
set_irn_in(peelhead, ARR_LEN(peelheadnins), peelheadnins);
- /* Fixes the backedge information according to the link.
- * Following loop optimizations might depend on it. */
- fix_backedge_info(loophead);
- fix_backedge_info(peelhead);
-
for_each_phi(loophead, phi) {
ir_node **ins = get_node_info( phi )->ins;
set_irn_in(phi, lhead_arity, ins);
/**
* Create a raw copy (ins are still the old ones) of the given node.
+ * We rely on copies to be NOT visited.
*/
static ir_node *rawcopy_node(ir_node *node)
{
+ int i, arity;
ir_node *cp;
node_info *cpstate;
cp = exact_copy(node);
+
+ arity = get_irn_arity(node);
+
+ for (i = 0; i < arity; ++i) {
+ if (is_backedge(node, i))
+ set_backedge(cp, i);
+ }
+
set_copy(node, cp);
cpstate = new_node_info();
set_irn_link(cp, cpstate);
- mark_irn_visited(cp);
- return cp;
-}
-//int temp = 0;
-//
-///* This walker copies all walked nodes. The walk_condition determines the nodes to walk. */
-//static void keepalives_walk(ir_node *node, walker_condition *walk_condition)
-//{
-// int i;
-// int arity;
-// ir_graph *irg = current_ir_graph;
-//
-// /**
-// * break condition and cycle resolver, creating temporary node copies
-// */
-// if (get_irn_visited(node) >= get_irg_visited(irg)) {
-// return;
-// }
-//
-// /* Walk */
-// mark_irn_visited(node);
-//
-// if (!is_Block(node)) {
-// ir_node *pred = get_nodes_block(node);
-// if (walk_condition(pred))
-// keepalives_walk( pred, walk_condition );
-// }
-//
-// arity = get_irn_arity(node);
-//
-// for (i = get_irn_arity(node) - 1; i >= 0; --i) {
-// ir_node *pred = get_irn_n(node, i);
-//
-// if (walk_condition(pred))
-// keepalives_walk( pred, walk_condition );
-// }
-//
-// add_End_keepalive(get_irg_end(current_ir_graph), node);
-//}
+ if (is_Block(cp)) {
+ /* TODO
+ * exact_copy already sets Macroblock.
+ * Why should we do this anyway? */
+ set_Block_MacroBlock(cp, cp);
+ }
+
+ return cp;
+}
/**
* This walker copies all walked nodes.
* If the walk_condition is true for a node, it is walked.
- * All nodes node_info->copy attributes has to be NULL prior to every to every walk.
+ * All nodes node_info->copy attributes have to be NULL prior to every walk.
*/
-static void copy_walk(ir_node *node, walker_condition *walk_condition)
+static void copy_walk(ir_node *node, walker_condition *walk_condition, ir_loop *set_loop)
{
int i;
int arity;
*/
if (get_irn_visited(node) >= get_irg_visited(irg)) {
/* Here we rely on nodestate's copy being initialized with NULL */
- DB((dbg, LEVEL_5, "copy_walk: We have already visited %ld\n", get_irn_node_nr(node)));
+ DB((dbg, LEVEL_5, "copy_walk: We have already visited %N\n", node));
if (node_info->copy == NULL) {
- if (!is_Const(node) && !is_SymConst(node)) {
- cp = rawcopy_node(node);
- } else {
- cp = node;
- node_info->copy = cp;
- }
- DB((dbg, LEVEL_5, "The TEMP copy of %ld is created %ld\n", get_irn_node_nr(node), get_irn_node_nr(cp)));
+ cp = rawcopy_node(node);
+ DB((dbg, LEVEL_5, "The TEMP copy of %N is created %N\n", node, cp));
}
return;
}
-// add_End_keepalive(get_irg_end(current_ir_graph), node);
-
/* Walk */
mark_irn_visited(node);
if (!is_Block(node)) {
ir_node *pred = get_nodes_block(node);
if (walk_condition(pred))
- DB((dbg, LEVEL_5, "walk block %ld\n", get_irn_node_nr(pred)));
- copy_walk( pred, walk_condition );
+ DB((dbg, LEVEL_5, "walk block %N\n", pred));
+ copy_walk(pred, walk_condition, set_loop);
}
arity = get_irn_arity(node);
NEW_ARR_A(ir_node *, cpin, arity);
- for (i = get_irn_arity(node) - 1; i >= 0; --i) {
+ for (i = 0; i < arity; ++i) {
ir_node *pred = get_irn_n(node, i);
if (walk_condition(pred)) {
- DB((dbg, LEVEL_5, "walk node %ld\n", get_irn_node_nr(pred)));
- copy_walk( pred, walk_condition );
+ DB((dbg, LEVEL_5, "walk node %N\n", pred));
+ copy_walk(pred, walk_condition, set_loop);
cpin[i] = get_copy(pred);
- DB((dbg, LEVEL_5, "copy of %ld gets new in %ld which is copy of %ld\n",
- get_irn_node_nr(node), get_irn_node_nr(get_copy(pred)), get_irn_node_nr(pred)));
+ DB((dbg, LEVEL_5, "copy of %N gets new in %N which is copy of %N\n",
+ node, get_copy(pred), pred));
} else {
cpin[i] = pred;
}
/* copy node / finalize temp node */
if (node_info->copy == NULL) {
/* No temporary copy existent */
-
- /* Do not copy constants TODO right? */
- if (!is_Const(node) && !is_SymConst(node)) {
- cp = rawcopy_node(node);
- } else {
- cp = node;
- node_info->copy = cp;
- }
- DB((dbg, LEVEL_5, "The FINAL copy of %ld is CREATED %ld\n", get_irn_node_nr(node), get_irn_node_nr(cp)));
+ cp = rawcopy_node(node);
+ DB((dbg, LEVEL_5, "The FINAL copy of %N is CREATED %N\n", node, cp));
} else {
/* temporary copy is existent but without correct ins */
cp = get_copy(node);
- DB((dbg, LEVEL_5, "The FINAL copy of %ld is EXISTENT %ld\n", get_irn_node_nr(node), get_irn_node_nr(cp)));
+ DB((dbg, LEVEL_5, "The FINAL copy of %N is EXISTENT %N\n", node, cp));
}
if (!is_Block(node)) {
ir_node *cpblock = get_copy(get_nodes_block(node));
set_nodes_block(cp, cpblock );
- /* fix the phi information in attr.phis */
- if( is_Phi(cp) )
+ if (is_Phi(cp))
add_Block_phi(cpblock, cp);
- } else {
- /* macroblock info has not been copied */
- set_Block_MacroBlock(cp, cp);
}
+ set_irn_loop(cp, set_loop);
set_irn_in(cp, ARR_LEN(cpin), cpin);
}
-/* Loop peeling, and fix the cf for the loop entry nodes, which have now more preds */
-static void peel(out_edges *loop_outs)
+/**
+ * Loop peeling, and fix the cf for the loop entry nodes, which have now more preds
+ */
+static void peel(out_edge *loop_outs)
{
int i;
ir_node **entry_buffer;
NEW_ARR_A(ir_node *, entry_buffer, ARR_LEN(loop_outs));
/* duplicate loop walk */
-// cur_head = loop_cf_head;
inc_irg_visited(current_ir_graph);
- for(i = 0; i < ARR_LEN(loop_outs); i++) {
- out_edges entry = loop_outs[i];
+ for (i = 0; i < ARR_LEN(loop_outs); i++) {
+ out_edge entry = loop_outs[i];
ir_node *node = entry.node;
ir_node *pred = get_irn_n(entry.node, entry.pred_irn_n);
if (is_Block(node)) {
- copy_walk( pred, is_in_loop );
+ copy_walk(pred, is_in_loop, NULL);
duplicate_preds(node, entry.pred_irn_n, get_copy(pred) );
} else {
- copy_walk( pred, is_in_loop );
- if (!is_End(node)) /* leave out keepalives */
+ copy_walk(pred, is_in_loop, NULL);
+ /* leave out keepalives */
+ if (!is_End(node)) {
/* Node is user of a value defined inside the loop.
* We'll need a phi since we duplicated the loop. */
- /* cannot construct_ssa here, because it needs another walker */
- entry_buffer[entry_c++] = pred;
+ /* Cannot construct_ssa here, because it needs another walker. */
+ entry_buffer[entry_c] = pred;
+ ++entry_c;
+ }
}
}
peel_fix_heads();
/* Generate phis for values from peeled code and original loop */
- for(i = 0; i < entry_c; i++)
+ construct_ssa_foreach(entry_buffer, entry_c);
+ /*for (i = 0; i < entry_c; i++)
{
ir_node *cppred, *block, *cpblock, *pred;
- /* It is not possible to use
- * pred = get_irn_n(entry.node, entry.pred_irn_n);
- * because we might have changed the nodes predecessors in construct_ssa
- */
pred = entry_buffer[i];
cppred = get_copy(pred);
block = get_nodes_block(pred);
cpblock = get_nodes_block(cppred);
construct_ssa(block, pred, cpblock, cppred);
- }
+ }*/
}
-/*
+/**
* Populates head_entries with (node, pred_pos) tuple
* whereas the node's pred at pred_pos is in the head but not the node itself.
* Head and condition chain blocks must be marked.
*/
-static void get_head_entries(ir_node *node, void *env)
+static void get_head_outs(ir_node *node, void *env)
{
int i;
int arity = get_irn_arity(node);
(void) env;
- for(i = 0; i < arity; ++i) {
+ DB((dbg, LEVEL_5, "get head entries %N \n", node));
+
+ for (i = 0; i < arity; ++i) {
/* node is not in the head, but the predecessor is.
* (head or loop chain nodes are marked) */
+
+ DB((dbg, LEVEL_5, "... "));
+ DB((dbg, LEVEL_5, "node %N marked %d (0) pred %d marked %d (1) \n",
+ node->node_nr, is_nodesblock_marked(node),i, is_nodesblock_marked(get_irn_n(node, i))));
+
if (!is_nodesblock_marked(node) && is_nodesblock_marked(get_irn_n(node, i))) {
- out_edges entry;
+ out_edge entry;
entry.node = node;
entry.pred_irn_n = i;
DB((dbg, LEVEL_5,
- "Found head chain entry %ld @%d because !inloop %ld and inloop %ld\n",
- node->node_nr, i, node->node_nr, get_irn_n(node, i)->node_nr));
- ARR_APP1(out_edges, cur_head_outs, entry);
+ "Found head chain entry %N @%d because !inloop %N and inloop %N\n",
+ node, i, node, get_irn_n(node, i)));
+ ARR_APP1(out_edge, cur_head_outs, entry);
}
}
}
/**
* Find condition chains, and add them to be inverted, until the node count exceeds the limit.
* A block belongs to the chain if a condition branches out of the loop.
- * Returns if the given block belongs to the condition chain.
- * FIXME prevent collecting ALL loop blocks (may happen if all blocks jump out of the loop)
+ * Returns 1 if the given block belongs to the condition chain.
*/
-static unsigned condition_chains(ir_node *block) {
+static unsigned find_condition_chains(ir_node *block)
+{
const ir_edge_t *edge;
unsigned mark = 0;
- //unsigned over_limit = 0;
-
int nodes_n = 0;
- /* we need all outs, including keeps (TODO firm function for that??) */
+ DB((dbg, LEVEL_5, "condition_chains for block %N\n", block));
+
+ /* Collect all outs, including keeps.
+ * (TODO firm function for number of out edges?) */
foreach_out_edge_kind(block, edge, EDGE_KIND_NORMAL) {
++nodes_n;
}
- /* We do not want to collect more nodes from condition chains, than the limit allows us to. */
- if (head_inversion_node_count + nodes_n > head_inversion_node_limit) {
- //over_limit = 1;
+ /* We do not want to collect more nodes from condition chains, than the limit allows us to.
+ * Also, leave at least one block as body. */
+ if (head_inversion_node_count + nodes_n > inversion_head_node_limit
+ || head_inversion_block_count + 1 == loop_info.blocks) {
set_Block_mark(block, 0);
-// printf(" %ld over limit\n", block->node_nr);
+
return 0;
}
- ++loop_info.blocks;
-
/* First: check our successors, and add all succs that are outside of the loop to the list */
foreach_block_succ(block, edge) {
ir_node *src = get_edge_src_irn( edge );
int pos = get_edge_src_pos( edge );
if (!is_in_loop(src)) {
- //printf(" src %ld @ %d into block %ld \n", src->node_nr, pos, block->node_nr);
+ out_edge entry;
+
mark = 1;
- out_edges entry;
entry.node = src;
entry.pred_irn_n = pos;
- ARR_APP1(out_edges, cond_chain_entries, entry);
+ ARR_APP1(out_edge, cond_chain_entries, entry);
mark_irn_visited(src);
}
}
- /* this block is not part of the chain,
- * because the chain would become too big or we have no succ outside of the loop */
if (mark == 0) {
+ /* this block is not part of the chain,
+ * because the chain would become too long or we have no successor outside of the loop */
+
set_Block_mark(block, 0);
return 0;
} else {
set_Block_mark(block, 1);
- DB((dbg, LEVEL_5, "block %ld is part of condition chain\n", get_irn_node_nr(block)));
+ ++head_inversion_block_count;
+ DB((dbg, LEVEL_5, "block %N is part of condition chain\n", block));
head_inversion_node_count += nodes_n;
}
int pos = get_edge_src_pos( edge );
/* already done cases */
- if (!is_in_loop( src ) || (get_irn_visited(src) >= get_irg_visited(current_ir_graph)))
+ if (!is_in_loop( src ) || (get_irn_visited(src) >= get_irg_visited(current_ir_graph))) {
continue;
+ }
mark_irn_visited(src);
-
- inchain = condition_chains( src );
+ DB((dbg, LEVEL_5, "condition chain walk %N\n", src));
+ inchain = find_condition_chains(src);
/* if successor is not part of chain we need to collect its outs */
- if ( !inchain ) {
- out_edges entry;
+ if (!inchain) {
+ out_edge entry;
entry.node = src;
entry.pred_irn_n = pos;
- ARR_APP1(out_edges, cond_chain_entries, entry);
+ ARR_APP1(out_edge, cond_chain_entries, entry);
}
}
return mark;
}
/**
- *
+ * Rewire the loop head and inverted head for loop inversion.
*/
static void inversion_fix_heads(void)
{
int iheadin_c = 0;
int backedges_n = get_backedge_n(loophead, 0);
- int lhead_arity = headarity - backedges_n;
- int ihead_arity = backedges_n;
+ int lhead_arity = backedges_n;
+ int ihead_arity = headarity - backedges_n;
+
+ assert(lhead_arity != 0 && "Loophead has arity 0. Probably wrong backedge informations.");
+ assert(ihead_arity != 0 && "Inversionhead has arity 0. Probably wrong backedge informations.");
/* new in arrays for all phis in the head blocks */
NEW_ARR_A(ir_node *, loopheadnins, lhead_arity);
* Rewire the head blocks ins and their phi ins.
* Requires phi list per block.
*/
- if ( is_backedge(loophead, i) ) {
- invheadnins[iheadin_c] = pred;
- for_each_phi(invhead, phi) {
- get_node_info( phi )->ins[iheadin_c] = get_irn_n( phi, i) ;
- }
- ++iheadin_c;
- } else {
+ if (is_backedge(loophead, i) && !is_alien_edge(loophead, i)) {
/* just copy these edges */
loopheadnins[lheadin_c] = pred;
for_each_phi(loophead, phi) {
- get_node_info( phi )->ins[lheadin_c] = get_irn_n(phi, i);
+ get_node_info(phi)->ins[lheadin_c] = get_irn_n(phi, i);
}
++lheadin_c;
+ } else {
+ invheadnins[iheadin_c] = pred;
+ for_each_phi(invhead, phi) {
+ get_node_info(phi)->ins[iheadin_c] = get_irn_n(phi, i) ;
+ }
+ ++iheadin_c;
}
- }/* for */
+ }
/* assign the ins to the head blocks */
set_irn_in(loophead, ARR_LEN(loopheadnins), loopheadnins);
}
}
-
-static void loop_inversion_walk(out_edges *head_entries)
+static void inversion_walk(out_edge *head_entries)
{
int i;
ir_node *phi;
head_phi_assign = NEW_ARR_F(ir_node *, 0);
- /* Find assignments in the condition chain, to construct_ssa for them after the loop inversion. */
- for_each_phi( loop_cf_head , phi) {
- for(i=0; i<get_irn_arity(phi); ++i) {
+ /* Find assignments in the condition chain,
+ * to construct_ssa for them after the loop inversion. */
+ for_each_phi(loop_cf_head , phi) {
+ int arity = get_irn_arity(phi);
+ for (i = 0; i < arity; ++i) {
ir_node *def = get_irn_n(phi, i);
- if ( is_nodesblock_marked(def) ) {
+ if (is_nodesblock_marked(def)) {
ARR_APP1(ir_node *, head_phi_assign, def);
}
}
ir_reserve_resources(current_ir_graph, IR_RESOURCE_IRN_VISITED);
- /* duplicate condition chain */
+ /**
+ * duplicate condition chain
+ **/
inc_irg_visited(current_ir_graph);
- for(i = 0; i < ARR_LEN(head_entries); ++i) {
- out_edges entry = head_entries[i];
+ for (i = 0; i < ARR_LEN(head_entries); ++i) {
+ out_edge entry = head_entries[i];
ir_node *node = entry.node;
ir_node *pred = get_irn_n(entry.node, entry.pred_irn_n);
-// add_End_keepalive(get_irg_end(current_ir_graph), pred);
-
if (is_Block(node)) {
- DB((dbg, LEVEL_5, "\nINIT walk block %ld\n", get_irn_node_nr(pred)));
- copy_walk(pred, is_nodesblock_marked);
+ DB((dbg, LEVEL_5, "\nInit walk block %N\n", pred));
+
+ copy_walk(pred, is_nodesblock_marked, cur_loop);
duplicate_preds(node, entry.pred_irn_n, get_copy(pred) );
} else {
- DB((dbg, LEVEL_5, "\nInit walk node %ld\n", get_irn_node_nr(pred)));
- copy_walk( pred, is_nodesblock_marked );
+ DB((dbg, LEVEL_5, "\nInit walk node %N\n", pred));
+
+ copy_walk(pred, is_nodesblock_marked, cur_loop);
/* ignore keepalives */
- if (!is_End(node))
+ if (!is_End(node)) {
/* Node is user of a value assigned inside the loop.
- * We'll need a phi since we duplicated the head. */
- entry_buffer[entry_c++] = pred;
+ * We will need a phi since we duplicated the head. */
+ entry_buffer[entry_c] = pred;
+ ++entry_c;
+ }
}
}
inversion_fix_heads();
- DEBUG_ONLY(dump_ir_block_graph(current_ir_graph, "-headsfixed"));
+ /* Generate phis for users of values assigned in the condition chain
+ * and read in the loops body */
+ construct_ssa_foreach(entry_buffer, entry_c);
- /* Generate phis for users of values assigned in the condition chain and read in the loops body */
- for(i = 0; i < entry_c; i++) {
- ir_node *cppred, *block, *cpblock, *pred;
+ /* Generate phis for values that are assigned in the condition chain
+ * but not read in the loops body. */
+ construct_ssa_foreach(head_phi_assign, ARR_LEN(head_phi_assign));
- /* It is not possible to use
- * pred = get_irn_n(entry.node, entry.pred_irn_n);
- * because we might have changed the nodes predecessors in construct_ssa
- */
- pred = entry_buffer[i];
- cppred = get_copy(pred);
- block = get_nodes_block(pred);
- cpblock = get_nodes_block(cppred);
- DB((dbg, LEVEL_5,
- "construct_ssa (loop out value) original %ld and clone %ld\n",
- get_irn_node_nr(pred), get_irn_node_nr(cppred)));
- construct_ssa(block, pred, cpblock, cppred);
+ loop_cf_head = get_copy(loop_cf_head);
+}
+
+/* Loop peeling */
+static void loop_peeling(void)
+{
+ cur_loop_outs = NEW_ARR_F(out_edge, 0);
+ irg_walk_graph( current_ir_graph, get_loop_outs, NULL, NULL );
+
+ peel(cur_loop_outs);
+
+ /* clean up */
+ reset_node_infos();
+
+ set_irg_doms_inconsistent(current_ir_graph);
+ set_irg_loopinfo_inconsistent(current_ir_graph);
+ set_irg_outs_inconsistent(current_ir_graph);
+
+ DEL_ARR_F(cur_loop_outs);
+}
+
+/* Loop inversion */
+static void loop_inversion(void)
+{
+ unsigned do_inversion = 1;
+
+ inversion_head_node_limit = INT_MAX;
+ /* Search for condition chains. */
+ ir_reserve_resources(current_ir_graph, IR_RESOURCE_BLOCK_MARK);
+
+ irg_walk_graph(current_ir_graph, reset_block_mark, NULL, NULL);
+
+ loop_info.blocks = get_loop_n_blocks(cur_loop);
+ cond_chain_entries = NEW_ARR_F(out_edge, 0);
+
+ head_inversion_node_count = 0;
+ head_inversion_block_count = 0;
+
+ set_Block_mark(loop_cf_head, 1);
+ mark_irn_visited(loop_cf_head);
+ inc_irg_visited(current_ir_graph);
-// char *res;
-// char *s = "-SSA_";
-// char *n = strdup(" ");
-// n[0] = 'a' + (char)i;
-// res = strdup(s);
-// strcat(res, n);
-// dump_ir_block_graph(current_ir_graph, res );
+ find_condition_chains(loop_cf_head);
+
+ DB((dbg, LEVEL_3, "Loop contains %d blocks.\n", loop_info.blocks));
+ if (loop_info.blocks < 2) {
+ do_inversion = 0;
+ DB((dbg, LEVEL_3, "Loop contains %d (less than 2) blocks => No Inversion done.\n", loop_info.blocks));
}
- /* Generate phis for values that are assigned in the condition chain
- * but not read in the loops body.
- */
- for(i = 0; i < ARR_LEN(head_phi_assign); ++i) {
- ir_node *def_block, *inhead_phi_def, *inv_def_block, *inv_inhead_phi_def;
- /* Note: construct_ssa only fixes the FIRST nodes usage. */
- inhead_phi_def = head_phi_assign[i];
- inv_inhead_phi_def = get_copy(inhead_phi_def);
- def_block = get_nodes_block(inhead_phi_def);
- inv_def_block = get_nodes_block(inv_inhead_phi_def);
- DB((dbg, LEVEL_5,
- "construct_ssa (condition chain out values) original %ld and clone %ld\n",
- get_irn_node_nr(inv_inhead_phi_def), get_irn_node_nr(inhead_phi_def)));
- construct_ssa(inv_def_block, inv_inhead_phi_def, def_block, inhead_phi_def);
+ /* We also catch endless loops here,
+ * because they do not have a condition chain. */
+ if (head_inversion_block_count < 1) {
+ do_inversion = 0;
+ DB((dbg, LEVEL_3, "Loop contains %d (less than 1) invertible blocks => No Inversion done.\n", head_inversion_block_count));
}
- loop_cf_head = get_copy(loop_cf_head);
+
+ if (do_inversion) {
+ cur_head_outs = NEW_ARR_F(out_edge, 0);
+
+ /* Get all edges pointing into the head or condition chain (outs). */
+ irg_walk_graph(current_ir_graph, get_head_outs, NULL, NULL);
+ inversion_walk(cur_head_outs);
+
+ DEL_ARR_F(cur_head_outs);
+
+ set_irg_doms_inconsistent(current_ir_graph);
+ set_irg_loopinfo_inconsistent(current_ir_graph);
+ set_irg_outs_inconsistent(current_ir_graph);
+ }
+
+ /* free */
+ DEL_ARR_F(cond_chain_entries);
+ ir_free_resources(current_ir_graph, IR_RESOURCE_BLOCK_MARK);
}
/**
- * Decide if loop inversion, peeling or unrolling should be performed.
- * Inversion creates abnormal looking loops. Be careful with optimizations after that.
+ * Returns last element of linked list of copies by
+ * walking the linked list.
*/
-static void decision_maker(void)
+static ir_node *get_last_copy(ir_node *node)
{
- unsigned do_peel = 0;
- unsigned do_inversion = 1;
+ ir_node *copy, *cur;
+ cur = node;
+ while ((copy = get_copy(cur))) {
+ cur = copy;
+ }
+ return cur;
+}
- unsigned max_loop_opnodes = 200000;
+/**
+ * Rewire floating copies of the current loop.
+ */
+static void unrolling_fix_cf(void)
+{
+ ir_node *loophead = loop_cf_head;
+ int headarity = get_irn_arity(loophead);
+ ir_node *phi, *headnode;
+ /*ir_node *high, *low;*/
+ int i;
- head_inversion_node_limit = 99910;
+ int uhead_in_n = 0;
+ int backedges_n = get_backedge_n(loophead, 0);
+ int unroll_arity = backedges_n;
+
+ /* Create ins for all heads and their phis */
+ headnode = get_copy(loophead);
+ for_each_copy(headnode) {
+ NEW_ARR_A(ir_node *, get_node_info(headnode)->ins, unroll_arity);
+ for_each_phi(headnode, phi) {
+ NEW_ARR_A(ir_node *, get_node_info(phi)->ins, unroll_arity);
+ }
+ }
- cur_loop_outs = NEW_ARR_F(out_edges, 0);
+ /* Append the copies to the existing loop. */
+ for (i = 0; i < headarity; i++) {
+ ir_node *upper_head = loophead;
+ ir_node *lower_head = get_copy(loophead);
- /* Find loop entries walk, find head */
- inc_irg_visited( current_ir_graph );
- irg_walk_graph( current_ir_graph, get_loop_outs_and_info, NULL, NULL );
+ ir_node *upper_pred = get_irn_n(loophead, i);
+ ir_node *lower_pred = get_copy(get_irn_n(loophead, i));
- /* RETURN if there is no valid head */
- if (!loop_cf_head || !loop_cf_head_valid) {
- DB((dbg, LEVEL_1, "No valid loop head. Nothing done.\n"));
- return;
+ ir_node *last_pred;
+
+ /**
+ * Build unrolled loop top down
+ */
+ if (is_backedge(loophead, i) && !is_alien_edge(loophead, i)) {
+ for_each_copy2(upper_head, lower_head, upper_pred, lower_pred) {
+ get_node_info(lower_head)->ins[uhead_in_n] = upper_pred;
+
+ for_each_phi(upper_head, phi) {
+ ir_node *phi_copy = get_copy(phi);
+ get_node_info(phi_copy)->ins[uhead_in_n] = get_irn_n(phi, i);
+ }
+ }
+
+ last_pred = upper_pred;
+ ++uhead_in_n;
+
+ /* Fix the topmost loop heads backedges. */
+ set_irn_n(loophead, i, last_pred);
+ for_each_phi(loophead, phi) {
+ ir_node *last_phi = get_last_copy(phi);
+ ir_node *pred = get_irn_n(last_phi, i);
+ set_irn_n(phi, i, pred);
+ }
+ }
}
+
+ headnode = get_copy(loophead);
+ for_each_copy(headnode) {
+ set_irn_in(headnode, unroll_arity, get_node_info(headnode)->ins);
+ for_each_phi(headnode, phi) {
+ set_irn_in(phi, unroll_arity, get_node_info(phi)->ins);
+ }
+ }
+}
+
#if 0
- /* RETURN if there is a call in the loop */
- if (loop_info.calls)
- return;
+static ir_node *add_phi(ir_node *node, int phi_pos)
+{
+ ir_mode *mode;
+ ir_node *phi;
+ ir_node **in;
+ mode = get_irn_mode(get_irn_n(node, phi_pos));
+ ir_node *block = get_nodes_block(node);
+ int n_cfgpreds = get_irn_arity(block);
+ ir_node *pred = get_irn_n(node, phi_pos);
+ int i;
- /* Loop complexity too high */
- if (loop_info.opnodes_n > max_loop_opnodes)
- return;
+ /* create a new Phi */
+ NEW_ARR_A(ir_node*, in, n_cfgpreds);
+ for (i = 0; i < n_cfgpreds; ++i)
+ in[i] = new_Unknown(mode); /*pred;*/
-// foreach_out_edge(loop_cf_head, edge) {
-// ir_node *node = get_edge_src_irn(edge);
-// if ( !is_Block(node) && !is_Proj(node) && !is_Phi(node) )
-// ++loop_info.opnodes_head;
-// }
+ phi = new_r_Phi(block, n_cfgpreds, in, mode);
- inc_irg_visited(current_ir_graph);
- loop_walker( loop_outs, NULL, get_invariants, NULL );
+ assert(phi && "phi null");
+ assert(is_Bad(phi) && "phi bad");
- /* This could be improved with knowledge about variable range. */
- if (loop_info.stores == 0 && loop_info.invariant_loads > 0)
- do_peel = 1;
+ /* Important: always keep block phi list up to date. */
+ add_Block_phi(block, phi);
+ /* EVERY node is assumed to have a node_info linked. */
+ alloc_node_info(phi, NULL);
+ set_irn_n(node, phi_pos, phi);
+ return phi;
+}
#endif
- do_peel = 1;
- do_inversion = 1;
-
- /* Loop peeling */
- if (do_peel) {
- peel(cur_loop_outs);
- reset_node_infos();
- }
+/**
+ * Loop unrolling
+ * Could be improved with variable range informations.
+ */
+static void loop_unrolling(void)
+{
+ int i, j;
- DEBUG_ONLY(dump_ir_block_graph(current_ir_graph, "-peeled1"));
+ unroll_times = 7;
- DEL_ARR_F(cur_loop_outs);
+ cur_loop_outs = NEW_ARR_F(out_edge, 0);
+ irg_walk_graph( current_ir_graph, get_loop_outs, NULL, NULL );
- /* Loop inversion */
- /* Search for condition chains. We may not do this before peeling, as peeling changes things. */
- ir_reserve_resources(current_ir_graph, IR_RESOURCE_BLOCK_MARK);
- irg_walk_graph(current_ir_graph, unmark_block, NULL, NULL);
+ ir_reserve_resources(current_ir_graph, IR_RESOURCE_IRN_VISITED);
- cond_chain_entries = NEW_ARR_F(out_edges, 0);
- head_inversion_node_count = 0;
+ /* duplicate whole loop content */
inc_irg_visited(current_ir_graph);
- set_Block_mark(loop_cf_head, 1);
- mark_irn_visited(loop_cf_head);
- /* find condition chains */
- condition_chains(loop_cf_head);
- // TODO assume number of phis to be created. prevent inversion...
+ for (i = 0; i < ARR_LEN(cur_loop_outs); ++i) {
+ out_edge entry = cur_loop_outs[i];
+ ir_node *node = entry.node;
+ ir_node *pred = get_irn_n(entry.node, entry.pred_irn_n);
+ if (!is_Block(node)) {
+ for (j = 0; j < unroll_times - 1; ++j) {
+ copy_walk(pred, is_in_loop, cur_loop);
+ pred = get_copy(pred);
+ }
+ }
+ }
- /* Loop inversion */
- if (loop_info.blocks < 2) {
- do_inversion = 0;
- DB((dbg, LEVEL_2, "Loop contains %d (less than 2) blocks => No Inversion done.\n", loop_info.blocks));
+ for (i = 0; i < ARR_LEN(cur_loop_outs); ++i) {
+ out_edge entry = cur_loop_outs[i];
+ ir_node *node = entry.node;
+ ir_node *pred = get_irn_n(entry.node, entry.pred_irn_n);
+
+ if (is_Block(node)) {
+ for (j = 0; j < unroll_times - 1; ++j) {
+ copy_walk(pred, is_in_loop, cur_loop);
+ duplicate_preds(node, entry.pred_irn_n, get_copy(pred));
+
+ pred = get_copy(pred);
+ }
+ }
}
- if (do_inversion) {
- cur_head_outs = NEW_ARR_F(out_edges, 0);
+ ir_free_resources(current_ir_graph, IR_RESOURCE_IRN_VISITED);
- /* get all edges pointing into the head or condition chain */
- irg_walk_graph(current_ir_graph, get_head_entries, NULL, NULL);
+ /*dump_ir_graph(current_ir_graph, "-raw");*/
- loop_inversion_walk(cur_head_outs);
+ /* LDBG 2 */
+#if 1
+ /* Line up the floating copies. */
+ unrolling_fix_cf();
- DEL_ARR_F(cur_head_outs);
+ /* Generate phis for all loop outs */
+ for (i = 0; i < ARR_LEN(cur_loop_outs); ++i) {
+ out_edge entry = cur_loop_outs[i];
+ ir_node *node = entry.node;
+ ir_node *pred = get_irn_n(entry.node, entry.pred_irn_n);
+
+ if (!is_Block(node) && !is_End(node)) {
+ DB((dbg, LEVEL_5, " construct_ssa_n def %N node %N pos %d\n",
+ pred, node, entry.pred_irn_n));
+ construct_ssa_n(pred, node);
+ }
}
+#endif
- DEBUG_ONLY(dump_ir_block_graph(current_ir_graph, "-inversed2"));
+ DEL_ARR_F(cur_loop_outs);
- /* FREE */
- DEL_ARR_F(cond_chain_entries);
- ir_free_resources(current_ir_graph, IR_RESOURCE_BLOCK_MARK);
+ set_irg_doms_inconsistent(current_ir_graph);
+ set_irg_loopinfo_inconsistent(current_ir_graph);
+ set_irg_outs_inconsistent(current_ir_graph);
}
-/* */
-static void analyze_loop(ir_loop *loop)
+/* Initialization and */
+static void init_analyze(ir_loop *loop)
{
/* Init new for every loop */
cur_loop = loop;
loop_inv_head = NULL;
loop_peeled_head = NULL;
+ loop_info.outs = 0;
loop_info.calls = 0;
- loop_info.invariant_loads = 0;
loop_info.loads = 0;
- loop_info.stores = 0;
- loop_info.opnodes_n = 0;
loop_info.blocks = 0;
- DB((dbg, LEVEL_1, " >>>> current loop includes node %ld <<<\n", get_irn_node_nr(get_loop_node(loop, 0))));
+ DB((dbg, LEVEL_2, " >>>> current loop includes node %N <<<\n", get_loop_node(loop, 0)));
+
+ irg_walk_graph(current_ir_graph, get_loop_info, NULL, NULL);
+
+ /* RETURN if there is no valid head */
+ if (!loop_cf_head || !loop_cf_head_valid) {
+ DB((dbg, LEVEL_2, "No valid loop head. Nothing done.\n"));
+ return;
+ }
+
+ if (enable_peeling)
+ loop_peeling();
- decision_maker();
+ if (enable_inversion)
+ loop_inversion();
+ if (enable_unrolling)
+ loop_unrolling();
- DB((dbg, LEVEL_1, " <<<< end of loop with node %ld >>>>\n", get_irn_node_nr(get_loop_node(loop, 0))));
+#if 0
+ /* RETURN if there is a call in the loop */
+ if (loop_info.calls)
+ return;
+#endif
+
+ DB((dbg, LEVEL_2, " <<<< end of loop with node %N >>>>\n", get_loop_node(loop, 0)));
}
/* Find most inner loops and send them to analyze_loop */
-static void analyze_inner_loop(ir_loop *loop)
+static void find_most_inner_loop(ir_loop *loop)
{
/* descend into sons */
int sons = get_loop_n_sons(loop);
- if (sons==0) {
- analyze_loop(loop);
+ if (sons == 0) {
+ loop_element elem;
+ int el_n, i;
+
+ el_n = get_loop_n_elements(loop);
+
+ for (i=0; i < el_n; ++i) {
+ elem = get_loop_element(loop, i);
+ /* We can only rely on the blocks,
+ * as the loop attribute of the nodes seems not to be set. */
+ if (is_ir_node(elem.kind) && is_Block(elem.node)) {
+ ARR_APP1(ir_node *, loops, elem.node);
+ DB((dbg, LEVEL_5, "Found most inner loop (contains block %+F)\n", elem.node));
+ break;
+ }
+ }
} else {
int s;
- for(s=0; s<sons; s++) {
- analyze_inner_loop( get_loop_son(loop, s) );
+ for (s=0; s<sons; s++) {
+ find_most_inner_loop(get_loop_son(loop, s));
}
}
}
/**
- *
+ * Assure preconditions are met and go through all loops.
*/
void loop_optimization(ir_graph *irg)
{
ir_loop *loop;
- int sons, nr;
-
- DB((dbg, LEVEL_1, " >>> loop optimization (Startnode %ld) <<<\n", get_irn_node_nr(get_irg_start(irg))));
+ int i, sons, nr;
/* Init */
link_node_state_list = NULL;
+ set_current_ir_graph(irg);
/* preconditions */
+ edges_assure(irg);
+ assure_irg_outs(irg);
+
+ /* NOTE: sets only the loop attribute of blocks, not nodes */
+ /* NOTE: Kills links */
+ assure_cf_loop(irg);
+
ir_reserve_resources(irg, IR_RESOURCE_IRN_LINK|IR_RESOURCE_PHI_LIST);
collect_phiprojs(irg);
ir_free_resources(irg, IR_RESOURCE_IRN_LINK);
- set_current_ir_graph(irg);
- assure_cf_loop(irg);
-
/* allocate node_info for additional information on nodes */
ir_reserve_resources(irg, IR_RESOURCE_IRN_LINK);
irg_walk_graph(current_ir_graph, alloc_node_info, NULL, NULL);
loop = get_irg_loop(irg);
sons = get_loop_n_sons(loop);
- for (nr=0; nr<sons; nr++) {
- analyze_inner_loop(get_loop_son(loop, nr));
+ loops = NEW_ARR_F(ir_node *, 0);
+
+ for (nr = 0; nr < sons; ++nr) {
+ find_most_inner_loop(get_loop_son(loop, nr));
}
+/* TODO Keep backedges during optimization to avoid
+ * this ugly allocation and deallocation.
+ * (set_irn_in seems to destroy them)
+ */
+#if 0
+ for (i = 0; i < ARR_LEN(loops); ++i) {
+ ir_loop *loop;
+
+ loop = get_irn_loop(loops[i]);
+ init_analyze(loop);
+ }
+#else
+ /* This part is useful for testing
+ * or has to be used if the backedge information is destroyed.
+ * Which is the case at the moment, because the backedge information gets lost
+ * before inversion_fix_heads/unrolling_fix_cf, which results in bads.
+ * NOTE!: Testsuite runs successfully nevertheless...
+ */
+
+ /**
+ * assure_cf_loop() creates a completely new loop tree.
+ * Thus we cannot optimize a loop, assure_cf_loop() and continue with the next loop,
+ * as the next loop must be searched because it is not distinguishable from the
+ * already done loops.
+ * The links of the loops are also not available anymore (to store a "loop done" flag).
+ * Therefore we save a block per loop.
+ * NOTE: We rely on the loop optimizations not to remove any block from the loop.
+ * Later, we fetch the blocks loop attribute, as it is updated by assure_cf_loop.
+ */
+ for (i = 0; i < ARR_LEN(loops); ++i) {
+ ir_loop *loop;
+
+ free_node_info();
+ ir_free_resources(irg, IR_RESOURCE_IRN_LINK);
+
+ edges_assure(current_ir_graph);
+ assure_irg_outs(current_ir_graph);
+
+ /* NOTE: sets only the loop attribute of blocks */
+ /* NOTE: Kills links */
+ assure_cf_loop(current_ir_graph);
+
+ irg_walk_graph(current_ir_graph, alloc_node_info, NULL, NULL);
+ ir_reserve_resources(irg, IR_RESOURCE_IRN_LINK);
+
+ /* Get loop from block */
+ loop = get_irn_loop(loops[i]);
+ init_analyze(loop);
+ }
+#endif
+
/* Free */
+ DEL_ARR_F(loops);
+
free_node_info();
- ir_free_resources(irg, IR_RESOURCE_PHI_LIST|IR_RESOURCE_IRN_LINK);
+ ir_free_resources(irg, IR_RESOURCE_IRN_LINK);
+ ir_free_resources(irg, IR_RESOURCE_PHI_LIST);
+}
+
+void do_loop_unrolling(ir_graph *irg)
+{
+ enable_unrolling = 1;
+ enable_peeling = 0;
+ enable_inversion = 0;
+
+ DB((dbg, LEVEL_2, " >>> unrolling (Startnode %N) <<<\n",
+ get_irg_start(irg)));
+
+ loop_optimization(irg);
+
+ DB((dbg, LEVEL_2, " >>> unrolling done (Startnode %N) <<<\n",
+ get_irg_start(irg)));
+}
+
+void do_loop_inversion(ir_graph *irg)
+{
+ enable_unrolling = 0;
+ enable_peeling = 0;
+ enable_inversion = 1;
+
+ DB((dbg, LEVEL_2, " >>> inversion (Startnode %N) <<<\n",
+ get_irg_start(irg)));
+
+ loop_optimization(irg);
+
+ DB((dbg, LEVEL_2, " >>> inversion done (Startnode %N) <<<\n",
+ get_irg_start(irg)));
+}
+
+void do_loop_peeling(ir_graph *irg)
+{
+ enable_unrolling = 0;
+ enable_peeling = 1;
+ enable_inversion = 0;
+
+ DB((dbg, LEVEL_2, " >>> peeling (Startnode %N) <<<\n",
+ get_irg_start(irg)));
+
+ loop_optimization(irg);
+
+ DB((dbg, LEVEL_2, " >>> peeling done (Startnode %N) <<<\n",
+ get_irg_start(irg)));
- DB((dbg, LEVEL_1, " >>> loop optimization done (Startnode %ld)<<<\n", get_irn_node_nr(get_irg_start(irg))));
}
-void firm_init_loop(void) {
+ir_graph_pass_t *loop_inversion_pass(const char *name)
+{
+ return def_graph_pass(name ? name : "loop_inversion", do_loop_inversion);
+}
+
+ir_graph_pass_t *loop_unroll_pass(const char *name)
+{
+ return def_graph_pass(name ? name : "loop_unroll", do_loop_unrolling);
+}
+
+ir_graph_pass_t *loop_peeling_pass(const char *name)
+{
+ return def_graph_pass(name ? name : "loop_peeling", do_loop_peeling);
+}
+
+void firm_init_loop_opt(void)
+{
FIRM_DBG_REGISTER(dbg, "firm.opt.loop");
}