You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 

933 lines
29 KiB

#include <argp.h>
#include <inttypes.h>
#include <locale.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/time.h>
#ifdef HAVE_PROFILER
#include <gperftools/profiler.h>
#endif
#include <getrss.h>
#include <sylvan.h>
#include <sylvan_int.h>
/* Configuration (via argp) */
static int report_levels = 0; // report states at end of every level
static int report_table = 0; // report table size at end of every level
static int report_nodes = 0; // report number of nodes of BDDs
static int strategy = 2; // 0 = BFS, 1 = PAR, 2 = SAT, 3 = CHAINING
static int check_deadlocks = 0; // set to 1 to check for deadlocks on-the-fly (only bfs/par)
static int merge_relations = 0; // merge relations to 1 relation
static int print_transition_matrix = 0; // print transition relation matrix
static int workers = 0; // autodetect
static char* model_filename = NULL; // filename of model
#ifdef HAVE_PROFILER
static char* profile_filename = NULL; // filename for profiling
#endif
/* argp configuration */
static struct argp_option options[] =
{
{"workers", 'w', "<workers>", 0, "Number of workers (default=0: autodetect)", 0},
{"strategy", 's', "<bfs|par|sat|chaining>", 0, "Strategy for reachability (default=sat)", 0},
#ifdef HAVE_PROFILER
{"profiler", 'p', "<filename>", 0, "Filename for profiling", 0},
#endif
{"deadlocks", 3, 0, 0, "Check for deadlocks", 1},
{"count-nodes", 5, 0, 0, "Report #nodes for BDDs", 1},
{"count-states", 1, 0, 0, "Report #states at each level", 1},
{"count-table", 2, 0, 0, "Report table usage at each level", 1},
{"merge-relations", 6, 0, 0, "Merge transition relations into one transition relation", 1},
{"print-matrix", 4, 0, 0, "Print transition matrix", 1},
{0, 0, 0, 0, 0, 0}
};
static error_t
parse_opt(int key, char *arg, struct argp_state *state)
{
switch (key) {
case 'w':
workers = atoi(arg);
break;
case 's':
if (strcmp(arg, "bfs")==0) strategy = 0;
else if (strcmp(arg, "par")==0) strategy = 1;
else if (strcmp(arg, "sat")==0) strategy = 2;
else if (strcmp(arg, "chaining")==0) strategy = 3;
else argp_usage(state);
break;
case 4:
print_transition_matrix = 1;
break;
case 3:
check_deadlocks = 1;
break;
case 1:
report_levels = 1;
break;
case 2:
report_table = 1;
break;
case 5:
report_nodes = 1;
break;
case 6:
merge_relations = 1;
break;
#ifdef HAVE_PROFILER
case 'p':
profile_filename = arg;
break;
#endif
case ARGP_KEY_ARG:
if (state->arg_num >= 1) argp_usage(state);
model_filename = arg;
break;
case ARGP_KEY_END:
if (state->arg_num < 1) argp_usage(state);
break;
default:
return ARGP_ERR_UNKNOWN;
}
return 0;
}
static struct argp argp = { options, parse_opt, "<model>", 0, 0, 0, 0 };
/**
* Types (set and relation)
*/
typedef struct set
{
BDD bdd;
BDD variables; // all variables in the set (used by satcount)
} *set_t;
typedef struct relation
{
BDD bdd;
BDD variables; // all variables in the relation (used by relprod)
int r_k, w_k, *r_proj, *w_proj;
} *rel_t;
static int vectorsize; // size of vector in integers
static int *statebits; // number of bits for each state integer
static int actionbits; // number of bits for action label
static int totalbits; // total number of bits
static int next_count; // number of partitions of the transition relation
static rel_t *next; // each partition of the transition relation
/**
* Obtain current wallclock time
*/
static double
wctime()
{
struct timeval tv;
gettimeofday(&tv, NULL);
return (tv.tv_sec + 1E-6 * tv.tv_usec);
}
static double t_start;
#define INFO(s, ...) fprintf(stdout, "[% 8.2f] " s, wctime()-t_start, ##__VA_ARGS__)
#define Abort(...) { fprintf(stderr, __VA_ARGS__); fprintf(stderr, "Abort at line %d!\n", __LINE__); exit(-1); }
static char*
to_h(double size, char *buf)
{
const char* units[] = {"B", "KB", "MB", "GB", "TB", "PB", "EB", "ZB", "YB"};
int i = 0;
for (;size>1024;size/=1024) i++;
sprintf(buf, "%.*f %s", i, size, units[i]);
return buf;
}
static void
print_memory_usage(void)
{
char buf[32];
to_h(getCurrentRSS(), buf);
INFO("Memory usage: %s\n", buf);
}
/**
* Load a set from file
* The expected binary format:
* - int k : projection size, or -1 for full state
* - int[k] proj : k integers specifying the variables of the projection
* - MTBDD[1] BDD (mtbdd binary format)
*/
#define set_load(f) CALL(set_load, f)
TASK_1(set_t, set_load, FILE*, f)
{
// allocate set
set_t set = (set_t)malloc(sizeof(struct set));
set->bdd = sylvan_false;
set->variables = sylvan_true;
sylvan_protect(&set->bdd);
sylvan_protect(&set->variables);
// read k
int k;
if (fread(&k, sizeof(int), 1, f) != 1) Abort("Invalid input file!\n");
if (k == -1) {
// create variables for a full state vector
uint32_t vars[totalbits];
for (int i=0; i<totalbits; i++) vars[i] = 2*i;
set->variables = sylvan_set_fromarray(vars, totalbits);
} else {
// read proj
int proj[k];
if (fread(proj, sizeof(int), k, f) != (size_t)k) Abort("Invalid input file!\n");
// create variables for a short/projected state vector
uint32_t vars[totalbits];
uint32_t cv = 0;
int j = 0, n = 0;
for (int i=0; i<vectorsize && j<k; i++) {
if (i == proj[j]) {
for (int x=0; x<statebits[i]; x++) vars[n++] = (cv += 2) - 2;
j++;
} else {
cv += 2 * statebits[i];
}
}
set->variables = sylvan_set_fromarray(vars, n);
}
// read bdd
if (mtbdd_reader_frombinary(f, &set->bdd, 1) != 0) Abort("Invalid input file!\n");
return set;
}
/**
* Load a relation from file
* This part just reads the r_k, w_k, r_proj and w_proj variables.
*/
#define rel_load_proj(f) CALL(rel_load_proj, f)
TASK_1(rel_t, rel_load_proj, FILE*, f)
{
rel_t rel = (rel_t)malloc(sizeof(struct relation));
int r_k, w_k;
if (fread(&r_k, sizeof(int), 1, f) != 1) Abort("Invalid file format.");
if (fread(&w_k, sizeof(int), 1, f) != 1) Abort("Invalid file format.");
rel->r_k = r_k;
rel->w_k = w_k;
int *r_proj = (int*)malloc(sizeof(int[r_k]));
int *w_proj = (int*)malloc(sizeof(int[w_k]));
if (fread(r_proj, sizeof(int), r_k, f) != (size_t)r_k) Abort("Invalid file format.");
if (fread(w_proj, sizeof(int), w_k, f) != (size_t)w_k) Abort("Invalid file format.");
rel->r_proj = r_proj;
rel->w_proj = w_proj;
rel->bdd = sylvan_false;
sylvan_protect(&rel->bdd);
/* Compute a_proj the union of r_proj and w_proj, and a_k the length of a_proj */
int a_proj[r_k+w_k];
int r_i = 0, w_i = 0, a_i = 0;
for (;r_i < r_k || w_i < w_k;) {
if (r_i < r_k && w_i < w_k) {
if (r_proj[r_i] < w_proj[w_i]) {
a_proj[a_i++] = r_proj[r_i++];
} else if (r_proj[r_i] > w_proj[w_i]) {
a_proj[a_i++] = w_proj[w_i++];
} else /* r_proj[r_i] == w_proj[w_i] */ {
a_proj[a_i++] = w_proj[w_i++];
r_i++;
}
} else if (r_i < r_k) {
a_proj[a_i++] = r_proj[r_i++];
} else if (w_i < w_k) {
a_proj[a_i++] = w_proj[w_i++];
}
}
const int a_k = a_i;
/* Compute all_variables, which are all variables the transition relation is defined on */
uint32_t all_vars[totalbits * 2];
uint32_t curvar = 0; // start with variable 0
int i=0, j=0, n=0;
for (; i<vectorsize && j<a_k; i++) {
if (i == a_proj[j]) {
for (int k=0; k<statebits[i]; k++) {
all_vars[n++] = curvar;
all_vars[n++] = curvar + 1;
curvar += 2;
}
j++;
} else {
curvar += 2 * statebits[i];
}
}
rel->variables = sylvan_set_fromarray(all_vars, n);
sylvan_protect(&rel->variables);
return rel;
}
/**
* Load a relation from file
* This part just reads the bdd of the relation
*/
#define rel_load(rel, f) CALL(rel_load, rel, f)
VOID_TASK_2(rel_load, rel_t, rel, FILE*, f)
{
if (mtbdd_reader_frombinary(f, &rel->bdd, 1) != 0) Abort("Invalid file format!\n");
}
/**
* Print a single example of a set to stdout
* Assumption: the example is a full vector and variables contains all state variables...
*/
#define print_example(example, variables) CALL(print_example, example, variables)
VOID_TASK_2(print_example, BDD, example, BDDSET, variables)
{
uint8_t str[totalbits];
if (example != sylvan_false) {
sylvan_sat_one(example, variables, str);
int x=0;
printf("[");
for (int i=0; i<vectorsize; i++) {
uint32_t res = 0;
for (int j=0; j<statebits[i]; j++) {
if (str[x++] == 1) res++;
res <<= 1;
}
if (i>0) printf(",");
printf("%" PRIu32, res);
}
printf("]");
}
}
/**
* Implementation of (parallel) saturation
* (assumes relations are ordered on first variable)
*/
TASK_2(BDD, go_sat, BDD, set, int, idx)
{
/* Terminal cases */
if (set == sylvan_false) return sylvan_false;
if (idx == next_count) return set;
/* Consult the cache */
BDD result;
const BDD _set = set;
if (cache_get3(200LL<<40, _set, idx, 0, &result)) return result;
mtbdd_refs_pushptr(&_set);
/**
* Possible improvement: cache more things (like intermediate results?)
* and chain-apply more of the current level before going deeper?
*/
/* Check if the relation should be applied */
const uint32_t var = sylvan_var(next[idx]->variables);
if (set == sylvan_true || var <= sylvan_var(set)) {
/* Count the number of relations starting here */
int count = idx+1;
while (count < next_count && var == sylvan_var(next[count]->variables)) count++;
count -= idx;
/*
* Compute until fixpoint:
* - SAT deeper
* - chain-apply all current level once
*/
BDD prev = sylvan_false;
BDD step = sylvan_false;
mtbdd_refs_pushptr(&set);
mtbdd_refs_pushptr(&prev);
mtbdd_refs_pushptr(&step);
while (prev != set) {
prev = set;
// SAT deeper
set = CALL(go_sat, set, idx+count);
// chain-apply all current level once
for (int i=0;i<count;i++) {
step = sylvan_relnext(set, next[idx+i]->bdd, next[idx+i]->variables);
set = sylvan_or(set, step);
step = sylvan_false; // unset, for gc
}
}
mtbdd_refs_popptr(3);
result = set;
} else {
/* Recursive computation */
mtbdd_refs_spawn(SPAWN(go_sat, sylvan_low(set), idx));
BDD high = mtbdd_refs_push(CALL(go_sat, sylvan_high(set), idx));
BDD low = mtbdd_refs_sync(SYNC(go_sat));
mtbdd_refs_pop(1);
result = sylvan_makenode(sylvan_var(set), low, high);
}
/* Store in cache */
cache_put3(200LL<<40, _set, idx, 0, result);
mtbdd_refs_popptr(1);
return result;
}
/**
* Wrapper for the Saturation strategy
*/
VOID_TASK_1(sat, set_t, set)
{
set->bdd = CALL(go_sat, set->bdd, 0);
}
/**
* Implement parallel strategy (that performs the relnext operations in parallel)
* This function does one level...
*/
TASK_5(BDD, go_par, BDD, cur, BDD, visited, size_t, from, size_t, len, BDD*, deadlocks)
{
if (len == 1) {
// Calculate NEW successors (not in visited)
BDD succ = sylvan_relnext(cur, next[from]->bdd, next[from]->variables);
bdd_refs_push(succ);
if (deadlocks) {
// check which BDDs in deadlocks do not have a successor in this relation
BDD anc = sylvan_relprev(next[from]->bdd, succ, next[from]->variables);
bdd_refs_push(anc);
*deadlocks = sylvan_diff(*deadlocks, anc);
bdd_refs_pop(1);
}
BDD result = sylvan_diff(succ, visited);
bdd_refs_pop(1);
return result;
} else {
BDD deadlocks_left;
BDD deadlocks_right;
if (deadlocks) {
deadlocks_left = *deadlocks;
deadlocks_right = *deadlocks;
sylvan_protect(&deadlocks_left);
sylvan_protect(&deadlocks_right);
}
// Recursively calculate left+right
bdd_refs_spawn(SPAWN(go_par, cur, visited, from, (len+1)/2, deadlocks ? &deadlocks_left: NULL));
BDD right = bdd_refs_push(CALL(go_par, cur, visited, from+(len+1)/2, len/2, deadlocks ? &deadlocks_right : NULL));
BDD left = bdd_refs_push(bdd_refs_sync(SYNC(go_par)));
// Merge results of left+right
BDD result = sylvan_or(left, right);
bdd_refs_pop(2);
if (deadlocks) {
bdd_refs_push(result);
*deadlocks = sylvan_and(deadlocks_left, deadlocks_right);
sylvan_unprotect(&deadlocks_left);
sylvan_unprotect(&deadlocks_right);
bdd_refs_pop(1);
}
return result;
}
}
/**
* Implementation of the PAR strategy
*/
VOID_TASK_1(par, set_t, set)
{
BDD visited = set->bdd;
BDD next_level = visited;
BDD cur_level = sylvan_false;
BDD deadlocks = sylvan_false;
sylvan_protect(&visited);
sylvan_protect(&next_level);
sylvan_protect(&cur_level);
sylvan_protect(&deadlocks);
int iteration = 1;
do {
// calculate successors in parallel
cur_level = next_level;
deadlocks = cur_level;
next_level = CALL(go_par, cur_level, visited, 0, next_count, check_deadlocks ? &deadlocks : NULL);
if (check_deadlocks && deadlocks != sylvan_false) {
INFO("Found %'0.0f deadlock states... ", sylvan_satcount(deadlocks, set->variables));
if (deadlocks != sylvan_false) {
printf("example: ");
print_example(deadlocks, set->variables);
check_deadlocks = 0;
}
printf("\n");
}
// visited = visited + new
visited = sylvan_or(visited, next_level);
if (report_table && report_levels) {
size_t filled, total;
sylvan_table_usage(&filled, &total);
INFO("Level %d done, %'0.0f states explored, table: %0.1f%% full (%'zu nodes)\n",
iteration, sylvan_satcount(visited, set->variables),
100.0*(double)filled/total, filled);
} else if (report_table) {
size_t filled, total;
sylvan_table_usage(&filled, &total);
INFO("Level %d done, table: %0.1f%% full (%'zu nodes)\n",
iteration,
100.0*(double)filled/total, filled);
} else if (report_levels) {
INFO("Level %d done, %'0.0f states explored\n", iteration, sylvan_satcount(visited, set->variables));
} else {
INFO("Level %d done\n", iteration);
}
iteration++;
} while (next_level != sylvan_false);
set->bdd = visited;
sylvan_unprotect(&visited);
sylvan_unprotect(&next_level);
sylvan_unprotect(&cur_level);
sylvan_unprotect(&deadlocks);
}
/**
* Implement sequential strategy (that performs the relnext operations one by one)
* This function does one level...
*/
TASK_5(BDD, go_bfs, BDD, cur, BDD, visited, size_t, from, size_t, len, BDD*, deadlocks)
{
if (len == 1) {
// Calculate NEW successors (not in visited)
BDD succ = sylvan_relnext(cur, next[from]->bdd, next[from]->variables);
bdd_refs_push(succ);
if (deadlocks) {
// check which BDDs in deadlocks do not have a successor in this relation
BDD anc = sylvan_relprev(next[from]->bdd, succ, next[from]->variables);
bdd_refs_push(anc);
*deadlocks = sylvan_diff(*deadlocks, anc);
bdd_refs_pop(1);
}
BDD result = sylvan_diff(succ, visited);
bdd_refs_pop(1);
return result;
} else {
BDD deadlocks_left;
BDD deadlocks_right;
if (deadlocks) {
deadlocks_left = *deadlocks;
deadlocks_right = *deadlocks;
sylvan_protect(&deadlocks_left);
sylvan_protect(&deadlocks_right);
}
// Recursively calculate left+right
BDD left = CALL(go_bfs, cur, visited, from, (len+1)/2, deadlocks ? &deadlocks_left : NULL);
bdd_refs_push(left);
BDD right = CALL(go_bfs, cur, visited, from+(len+1)/2, len/2, deadlocks ? &deadlocks_right : NULL);
bdd_refs_push(right);
// Merge results of left+right
BDD result = sylvan_or(left, right);
bdd_refs_pop(2);
if (deadlocks) {
bdd_refs_push(result);
*deadlocks = sylvan_and(deadlocks_left, deadlocks_right);
sylvan_unprotect(&deadlocks_left);
sylvan_unprotect(&deadlocks_right);
bdd_refs_pop(1);
}
return result;
}
}
/**
* Implementation of the BFS strategy
*/
VOID_TASK_1(bfs, set_t, set)
{
BDD visited = set->bdd;
BDD next_level = visited;
BDD cur_level = sylvan_false;
BDD deadlocks = sylvan_false;
sylvan_protect(&visited);
sylvan_protect(&next_level);
sylvan_protect(&cur_level);
sylvan_protect(&deadlocks);
int iteration = 1;
do {
// calculate successors in parallel
cur_level = next_level;
deadlocks = cur_level;
next_level = CALL(go_bfs, cur_level, visited, 0, next_count, check_deadlocks ? &deadlocks : NULL);
if (check_deadlocks && deadlocks != sylvan_false) {
INFO("Found %'0.0f deadlock states... ", sylvan_satcount(deadlocks, set->variables));
if (deadlocks != sylvan_false) {
printf("example: ");
print_example(deadlocks, set->variables);
check_deadlocks = 0;
}
printf("\n");
}
// visited = visited + new
visited = sylvan_or(visited, next_level);
if (report_table && report_levels) {
size_t filled, total;
sylvan_table_usage(&filled, &total);
INFO("Level %d done, %'0.0f states explored, table: %0.1f%% full (%'zu nodes)\n",
iteration, sylvan_satcount(visited, set->variables),
100.0*(double)filled/total, filled);
} else if (report_table) {
size_t filled, total;
sylvan_table_usage(&filled, &total);
INFO("Level %d done, table: %0.1f%% full (%'zu nodes)\n",
iteration,
100.0*(double)filled/total, filled);
} else if (report_levels) {
INFO("Level %d done, %'0.0f states explored\n", iteration, sylvan_satcount(visited, set->variables));
} else {
INFO("Level %d done\n", iteration);
}
iteration++;
} while (next_level != sylvan_false);
set->bdd = visited;
sylvan_unprotect(&visited);
sylvan_unprotect(&next_level);
sylvan_unprotect(&cur_level);
sylvan_unprotect(&deadlocks);
}
/**
* Implementation of the Chaining strategy (does not support deadlock detection)
*/
VOID_TASK_1(chaining, set_t, set)
{
BDD visited = set->bdd;
BDD next_level = visited;
BDD succ = sylvan_false;
bdd_refs_pushptr(&visited);
bdd_refs_pushptr(&next_level);
bdd_refs_pushptr(&succ);
int iteration = 1;
do {
// calculate successors in parallel
for (int i=0; i<next_count; i++) {
succ = sylvan_relnext(next_level, next[i]->bdd, next[i]->variables);
next_level = sylvan_or(next_level, succ);
succ = sylvan_false; // reset, for gc
}
// new = new - visited
// visited = visited + new
next_level = sylvan_diff(next_level, visited);
visited = sylvan_or(visited, next_level);
if (report_table && report_levels) {
size_t filled, total;
sylvan_table_usage(&filled, &total);
INFO("Level %d done, %'0.0f states explored, table: %0.1f%% full (%'zu nodes)\n",
iteration, sylvan_satcount(visited, set->variables),
100.0*(double)filled/total, filled);
} else if (report_table) {
size_t filled, total;
sylvan_table_usage(&filled, &total);
INFO("Level %d done, table: %0.1f%% full (%'zu nodes)\n",
iteration,
100.0*(double)filled/total, filled);
} else if (report_levels) {
INFO("Level %d done, %'0.0f states explored\n", iteration, sylvan_satcount(visited, set->variables));
} else {
INFO("Level %d done\n", iteration);
}
iteration++;
} while (next_level != sylvan_false);
set->bdd = visited;
bdd_refs_popptr(3);
}
/**
* Extend a transition relation to a larger domain (using s=s')
*/
#define extend_relation(rel, vars) CALL(extend_relation, rel, vars)
TASK_2(BDD, extend_relation, MTBDD, relation, MTBDD, variables)
{
/* first determine which state BDD variables are in rel */
int has[totalbits];
for (int i=0; i<totalbits; i++) has[i] = 0;
MTBDD s = variables;
while (!sylvan_set_isempty(s)) {
uint32_t v = sylvan_set_first(s);
if (v/2 >= (unsigned)totalbits) break; // action labels
has[v/2] = 1;
s = sylvan_set_next(s);
}
/* create "s=s'" for all variables not in rel */
BDD eq = sylvan_true;
for (int i=totalbits-1; i>=0; i--) {
if (has[i]) continue;
BDD low = sylvan_makenode(2*i+1, eq, sylvan_false);
bdd_refs_push(low);
BDD high = sylvan_makenode(2*i+1, sylvan_false, eq);
bdd_refs_pop(1);
eq = sylvan_makenode(2*i, low, high);
}
bdd_refs_push(eq);
BDD result = sylvan_and(relation, eq);
bdd_refs_pop(1);
return result;
}
/**
* Compute \BigUnion ( sets[i] )
*/
#define big_union(first, count) CALL(big_union, first, count)
TASK_2(BDD, big_union, int, first, int, count)
{
if (count == 1) return next[first]->bdd;
bdd_refs_spawn(SPAWN(big_union, first, count/2));
BDD right = bdd_refs_push(CALL(big_union, first+count/2, count-count/2));
BDD left = bdd_refs_push(bdd_refs_sync(SYNC(big_union)));
BDD result = sylvan_or(left, right);
bdd_refs_pop(2);
return result;
}
/**
* Print one row of the transition matrix (for vars)
*/
static void
print_matrix_row(rel_t rel)
{
int r_i = 0, w_i = 0;
for (int i=0; i<vectorsize; i++) {
int s = 0;
if (r_i < rel->r_k && rel->r_proj[r_i] == i) {
s |= 1;
r_i++;
}
if (w_i < rel->w_k && rel->w_proj[w_i] == i) {
s |= 2;
w_i++;
}
if (s == 0) fprintf(stdout, "-");
else if (s == 1) fprintf(stdout, "r");
else if (s == 2) fprintf(stdout, "w");
else if (s == 3) fprintf(stdout, "+");
}
}
VOID_TASK_0(gc_start)
{
char buf[32];
to_h(getCurrentRSS(), buf);
INFO("(GC) Starting garbage collection... (rss: %s)\n", buf);
}
VOID_TASK_0(gc_end)
{
char buf[32];
to_h(getCurrentRSS(), buf);
INFO("(GC) Garbage collection done. (rss: %s)\n", buf);
}
int
main(int argc, char **argv)
{
/**
* Parse command line, set locale, set startup time for INFO messages.
*/
argp_parse(&argp, argc, argv, 0, 0, 0);
setlocale(LC_NUMERIC, "en_US.utf-8");
t_start = wctime();
/**
* Initialize Lace.
*
* First: setup with given number of workers (0 for autodetect) and some large size task queue.
* Second: start all worker threads with default settings.
* Third: setup local variables using the LACE_ME macro.
*/
lace_init(workers, 1000000);
lace_startup(0, NULL, NULL);
LACE_ME;
/**
* Initialize Sylvan.
*
* First: set memory limits
* - 2 GB memory, nodes table twice as big as cache, initial size halved 6x
* (that means it takes 6 garbage collections to get to the maximum nodes&cache size)
* Second: initialize package and subpackages
* Third: add hooks to report garbage collection
*/
sylvan_set_limits(2LL<<30, 1, 6);
sylvan_init_package();
sylvan_init_bdd();
sylvan_gc_hook_pregc(TASK(gc_start));
sylvan_gc_hook_postgc(TASK(gc_end));
/**
* Read the model from file
*/
/* Open the file */
FILE *f = fopen(model_filename, "r");
if (f == NULL) Abort("Cannot open file '%s'!\n", model_filename);
/* Read domain data */
if (fread(&vectorsize, sizeof(int), 1, f) != 1) Abort("Invalid input file!\n");
statebits = (int*)malloc(sizeof(int[vectorsize]));
if (fread(statebits, sizeof(int), vectorsize, f) != (size_t)vectorsize) Abort("Invalid input file!\n");
if (fread(&actionbits, sizeof(int), 1, f) != 1) Abort("Invalid input file!\n");
totalbits = 0;
for (int i=0; i<vectorsize; i++) totalbits += statebits[i];
/* Read initial state */
set_t states = set_load(f);
/* Read number of transition relations */
if (fread(&next_count, sizeof(int), 1, f) != 1) Abort("Invalid input file!\n");
next = (rel_t*)malloc(sizeof(rel_t) * next_count);
/* Read transition relations */
for (int i=0; i<next_count; i++) next[i] = rel_load_proj(f);
for (int i=0; i<next_count; i++) rel_load(next[i], f);
/* We ignore the reachable states and action labels that are stored after the relations */
/* Close the file */
fclose(f);
/**
* Pre-processing and some statistics reporting
*/
if (strategy == 2 || strategy == 3) {
// for SAT and CHAINING, sort the transition relations (gnome sort because I like gnomes)
int i = 1, j = 2;
rel_t t;
while (i < next_count) {
rel_t *p = &next[i], *q = p-1;
if (sylvan_var((*q)->variables) > sylvan_var((*p)->variables)) {
t = *q;
*q = *p;
*p = t;
if (--i) continue;
}
i = j++;
}
}
INFO("Read file '%s'\n", model_filename);
INFO("%d integers per state, %d bits per state, %d transition groups\n", vectorsize, totalbits, next_count);
/* if requested, print the transition matrix */
if (print_transition_matrix) {
for (int i=0; i<next_count; i++) {
INFO(""); // print time prefix
print_matrix_row(next[i]); // print row
fprintf(stdout, "\n"); // print newline
}
}
/* merge all relations to one big transition relation if requested */
if (merge_relations) {
BDD newvars = sylvan_set_empty();
bdd_refs_pushptr(&newvars);
for (int i=totalbits-1; i>=0; i--) {
newvars = sylvan_set_add(newvars, i*2+1);
newvars = sylvan_set_add(newvars, i*2);
}
INFO("Extending transition relations to full domain.\n");
for (int i=0; i<next_count; i++) {
next[i]->bdd = extend_relation(next[i]->bdd, next[i]->variables);
next[i]->variables = newvars;
}
bdd_refs_popptr(1);
INFO("Taking union of all transition relations.\n");
next[0]->bdd = big_union(0, next_count);
for (int i=1; i<next_count; i++) {
next[i]->bdd = sylvan_false;
next[i]->variables = sylvan_true;
}
next_count = 1;
}
if (report_nodes) {
INFO("BDD nodes:\n");
INFO("Initial states: %zu BDD nodes\n", sylvan_nodecount(states->bdd));
for (int i=0; i<next_count; i++) {
INFO("Transition %d: %zu BDD nodes\n", i, sylvan_nodecount(next[i]->bdd));
}
}
print_memory_usage();
#ifdef HAVE_PROFILER
if (profile_filename != NULL) ProfilerStart(profile_filename);
#endif
if (strategy == 0) {
double t1 = wctime();
CALL(bfs, states);
double t2 = wctime();
INFO("BFS Time: %f\n", t2-t1);
} else if (strategy == 1) {
double t1 = wctime();
CALL(par, states);
double t2 = wctime();
INFO("PAR Time: %f\n", t2-t1);
} else if (strategy == 2) {
double t1 = wctime();
CALL(sat, states);
double t2 = wctime();
INFO("SAT Time: %f\n", t2-t1);
} else if (strategy == 3) {
double t1 = wctime();
CALL(chaining, states);
double t2 = wctime();
INFO("CHAINING Time: %f\n", t2-t1);
} else {
Abort("Invalid strategy set?!\n");
}
#ifdef HAVE_PROFILER
if (profile_filename != NULL) ProfilerStop();
#endif
// Now we just have states
INFO("Final states: %'0.0f states\n", sylvan_satcount(states->bdd, states->variables));
if (report_nodes) {
INFO("Final states: %'zu BDD nodes\n", sylvan_nodecount(states->bdd));
}
print_memory_usage();
sylvan_stats_report(stdout);
return 0;
}