diff --git a/jni/pjproject-android/.svn/pristine/ff/ffa7c01b3f0ac2f8ab4b3ff7fda4798256f83b47.svn-base b/jni/pjproject-android/.svn/pristine/ff/ffa7c01b3f0ac2f8ab4b3ff7fda4798256f83b47.svn-base
new file mode 100644
index 0000000..7a3a3a3
--- /dev/null
+++ b/jni/pjproject-android/.svn/pristine/ff/ffa7c01b3f0ac2f8ab4b3ff7fda4798256f83b47.svn-base
@@ -0,0 +1,1309 @@
+/* $Id$ */
+/* 
+ * Copyright (C) 2008-2011 Teluu Inc. (http://www.teluu.com)
+ * Copyright (C) 2003-2008 Benny Prijono <benny@prijono.org>
+ *
+ * This program 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 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program 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 this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA 
+ */
+#include <pjsip/sip_endpoint.h>
+#include <pjsip/sip_transaction.h>
+#include <pjsip/sip_private.h>
+#include <pjsip/sip_event.h>
+#include <pjsip/sip_resolve.h>
+#include <pjsip/sip_module.h>
+#include <pjsip/sip_util.h>
+#include <pjsip/sip_errno.h>
+#include <pj/except.h>
+#include <pj/log.h>
+#include <pj/string.h>
+#include <pj/os.h>
+#include <pj/pool.h>
+#include <pj/hash.h>
+#include <pj/assert.h>
+#include <pj/errno.h>
+#include <pj/lock.h>
+
+#define PJSIP_EX_NO_MEMORY  pj_NO_MEMORY_EXCEPTION()
+#define THIS_FILE	    "sip_endpoint.c"
+
+#define MAX_METHODS   32
+
+
+/* List of SIP endpoint exit callback. */
+typedef struct exit_cb
+{
+    PJ_DECL_LIST_MEMBER		    (struct exit_cb);
+    pjsip_endpt_exit_callback	    func;
+} exit_cb;
+
+
+/**
+ * The SIP endpoint.
+ */
+struct pjsip_endpoint
+{
+    /** Pool to allocate memory for the endpoint. */
+    pj_pool_t		*pool;
+
+    /** Mutex for the pool, hash table, and event list/queue. */
+    pj_mutex_t		*mutex;
+
+    /** Pool factory. */
+    pj_pool_factory	*pf;
+
+    /** Name. */
+    pj_str_t		 name;
+
+    /** Timer heap. */
+    pj_timer_heap_t	*timer_heap;
+
+    /** Transport manager. */
+    pjsip_tpmgr		*transport_mgr;
+
+    /** Ioqueue. */
+    pj_ioqueue_t	*ioqueue;
+
+    /** Last ioqueue err */
+    pj_status_t		 ioq_last_err;
+
+    /** DNS Resolver. */
+    pjsip_resolver_t	*resolver;
+
+    /** Modules lock. */
+    pj_rwmutex_t	*mod_mutex;
+
+    /** Modules. */
+    pjsip_module        *modules[PJSIP_MAX_MODULE];
+
+    /** Module list, sorted by priority. */
+    pjsip_module	 module_list;
+
+    /** Capability header list. */
+    pjsip_hdr		 cap_hdr;
+
+    /** Additional request headers. */
+    pjsip_hdr		 req_hdr;
+
+    /** List of exit callback. */
+    exit_cb		 exit_cb_list;
+};
+
+
+#if defined(PJSIP_SAFE_MODULE) && PJSIP_SAFE_MODULE!=0
+#   define LOCK_MODULE_ACCESS(ept)	pj_rwmutex_lock_read(ept->mod_mutex)
+#   define UNLOCK_MODULE_ACCESS(ept)	pj_rwmutex_unlock_read(ept->mod_mutex)
+#else
+#   define LOCK_MODULE_ACCESS(endpt)
+#   define UNLOCK_MODULE_ACCESS(endpt)
+#endif
+
+
+
+/*
+ * Prototypes.
+ */
+static void endpt_on_rx_msg( pjsip_endpoint*, 
+			     pj_status_t, pjsip_rx_data*);
+static pj_status_t endpt_on_tx_msg( pjsip_endpoint *endpt,
+				    pjsip_tx_data *tdata );
+static pj_status_t unload_module(pjsip_endpoint *endpt,
+				 pjsip_module *mod);
+
+/* Defined in sip_parser.c */
+void init_sip_parser(void);
+void deinit_sip_parser(void);
+
+/* Defined in sip_tel_uri.c */
+pj_status_t pjsip_tel_uri_subsys_init(void);
+
+
+/*
+ * This is the global handler for memory allocation failure, for pools that
+ * are created by the endpoint (by default, all pools ARE allocated by 
+ * endpoint). The error is handled by throwing exception, and hopefully,
+ * the exception will be handled by the application (or this library).
+ */
+static void pool_callback( pj_pool_t *pool, pj_size_t size )
+{
+    PJ_UNUSED_ARG(pool);
+    PJ_UNUSED_ARG(size);
+
+    PJ_THROW(PJSIP_EX_NO_MEMORY);
+}
+
+
+/* Compare module name, used for searching module based on name. */
+static int cmp_mod_name(void *name, const void *mod)
+{
+    return pj_stricmp((const pj_str_t*)name, &((pjsip_module*)mod)->name);
+}
+
+/*
+ * Register new module to the endpoint.
+ * The endpoint will then call the load and start function in the module to 
+ * properly initialize the module, and assign a unique module ID for the 
+ * module.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_register_module( pjsip_endpoint *endpt,
+						 pjsip_module *mod )
+{
+    pj_status_t status = PJ_SUCCESS;
+    pjsip_module *m;
+    unsigned i;
+
+    pj_rwmutex_lock_write(endpt->mod_mutex);
+
+    /* Make sure that this module has not been registered. */
+    PJ_ASSERT_ON_FAIL(	pj_list_find_node(&endpt->module_list, mod) == NULL,
+			{status = PJ_EEXISTS; goto on_return;});
+
+    /* Make sure that no module with the same name has been registered. */
+    PJ_ASSERT_ON_FAIL(	pj_list_search(&endpt->module_list, &mod->name, 
+				       &cmp_mod_name)==NULL,
+			{status = PJ_EEXISTS; goto on_return; });
+
+    /* Find unused ID for this module. */
+    for (i=0; i<PJ_ARRAY_SIZE(endpt->modules); ++i) {
+	if (endpt->modules[i] == NULL)
+	    break;
+    }
+    if (i == PJ_ARRAY_SIZE(endpt->modules)) {
+	pj_assert(!"Too many modules registered!");
+	status = PJ_ETOOMANY;
+	goto on_return;
+    }
+
+    /* Assign the ID. */
+    mod->id = i;
+
+    /* Try to load the module. */
+    if (mod->load) {
+	status = (*mod->load)(endpt);
+	if (status != PJ_SUCCESS)
+	    goto on_return;
+    }
+
+    /* Try to start the module. */
+    if (mod->start) {
+	status = (*mod->start)();
+	if (status != PJ_SUCCESS)
+	    goto on_return;
+    }
+
+    /* Save the module. */
+    endpt->modules[i] = mod;
+
+    /* Put in the module list, sorted by priority. */
+    m = endpt->module_list.next;
+    while (m != &endpt->module_list) {
+	if (m->priority > mod->priority)
+	    break;
+	m = m->next;
+    }
+    pj_list_insert_before(m, mod);
+
+    /* Done. */
+
+    PJ_LOG(4,(THIS_FILE, "Module \"%.*s\" registered", 
+	      (int)mod->name.slen, mod->name.ptr));
+
+on_return:
+    pj_rwmutex_unlock_write(endpt->mod_mutex);
+    return status;
+}
+
+/*
+ * Unregister a module from the endpoint.
+ * The endpoint will then call the stop and unload function in the module to 
+ * properly shutdown the module.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_unregister_module( pjsip_endpoint *endpt,
+						   pjsip_module *mod )
+{
+    pj_status_t status;
+
+    pj_rwmutex_lock_write(endpt->mod_mutex);
+
+    /* Make sure the module exists in the list. */
+    PJ_ASSERT_ON_FAIL(	pj_list_find_node(&endpt->module_list, mod) == mod,
+			{status = PJ_ENOTFOUND;goto on_return;} );
+
+    /* Make sure the module exists in the array. */
+    PJ_ASSERT_ON_FAIL(	mod->id>=0 && 
+			mod->id<(int)PJ_ARRAY_SIZE(endpt->modules) &&
+			endpt->modules[mod->id] == mod,
+			{status = PJ_ENOTFOUND; goto on_return;});
+
+    /* Try to stop the module. */
+    if (mod->stop) {
+	status = (*mod->stop)();
+	if (status != PJ_SUCCESS) goto on_return;
+    }
+
+    /* Unload module */
+    status = unload_module(endpt, mod);
+
+on_return:
+    pj_rwmutex_unlock_write(endpt->mod_mutex);
+
+    if (status != PJ_SUCCESS) {
+	char errmsg[PJ_ERR_MSG_SIZE];
+
+	pj_strerror(status, errmsg, sizeof(errmsg));
+	PJ_LOG(3,(THIS_FILE, "Module \"%.*s\" can not be unregistered: %s",
+		  (int)mod->name.slen, mod->name.ptr, errmsg));
+    }
+
+    return status;
+}
+
+static pj_status_t unload_module(pjsip_endpoint *endpt,
+				 pjsip_module *mod)
+{
+    pj_status_t status;
+
+    /* Try to unload the module. */
+    if (mod->unload) {
+	status = (*mod->unload)();
+	if (status != PJ_SUCCESS) 
+	    return status;
+    }
+
+    /* Module MUST NOT set module ID to -1. */
+    pj_assert(mod->id >= 0);
+
+    /* Remove module from array. */
+    endpt->modules[mod->id] = NULL;
+
+    /* Remove module from list. */
+    pj_list_erase(mod);
+
+    /* Set module Id to -1. */
+    mod->id = -1;
+
+    /* Done. */
+    status = PJ_SUCCESS;
+
+    PJ_LOG(4,(THIS_FILE, "Module \"%.*s\" unregistered", 
+	      (int)mod->name.slen, mod->name.ptr));
+
+    return status;
+}
+
+
+/*
+ * Get the value of the specified capability header field.
+ */
+PJ_DEF(const pjsip_hdr*) pjsip_endpt_get_capability( pjsip_endpoint *endpt,
+						     int htype,
+						     const pj_str_t *hname)
+{
+    pjsip_hdr *hdr = endpt->cap_hdr.next;
+
+    /* Check arguments. */
+    PJ_ASSERT_RETURN(endpt != NULL, NULL);
+    PJ_ASSERT_RETURN(htype != PJSIP_H_OTHER || hname, NULL);
+
+    if (htype != PJSIP_H_OTHER) {
+	while (hdr != &endpt->cap_hdr) {
+	    if (hdr->type == htype)
+		return hdr;
+	    hdr = hdr->next;
+	}
+    }
+    return NULL;
+}
+
+
+/*
+ * Check if the specified capability is supported.
+ */
+PJ_DEF(pj_bool_t) pjsip_endpt_has_capability( pjsip_endpoint *endpt,
+					      int htype,
+					      const pj_str_t *hname,
+					      const pj_str_t *token)
+{
+    const pjsip_generic_array_hdr *hdr;
+    unsigned i;
+
+    hdr = (const pjsip_generic_array_hdr*) 
+	   pjsip_endpt_get_capability(endpt, htype, hname);
+    if (!hdr)
+	return PJ_FALSE;
+
+    PJ_ASSERT_RETURN(token != NULL, PJ_FALSE);
+
+    for (i=0; i<hdr->count; ++i) {
+	if (!pj_stricmp(&hdr->values[i], token))
+	    return PJ_TRUE;
+    }
+
+    return PJ_FALSE;
+}
+
+/*
+ * Add or register new capabilities as indicated by the tags to the
+ * appropriate header fields in the endpoint.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_add_capability( pjsip_endpoint *endpt,
+						pjsip_module *mod,
+						int htype,
+						const pj_str_t *hname,
+						unsigned count,
+						const pj_str_t tags[])
+{
+    pjsip_generic_array_hdr *hdr;
+    unsigned i;
+
+    PJ_UNUSED_ARG(mod);
+
+    /* Check arguments. */
+    PJ_ASSERT_RETURN(endpt!=NULL && count>0 && tags, PJ_EINVAL);
+    PJ_ASSERT_RETURN(htype==PJSIP_H_ACCEPT || 
+		     htype==PJSIP_H_ALLOW ||
+		     htype==PJSIP_H_SUPPORTED,
+		     PJ_EINVAL);
+
+    /* Find the header. */
+    hdr = (pjsip_generic_array_hdr*) pjsip_endpt_get_capability(endpt, 
+								htype, hname);
+
+    /* Create the header when it's not present */
+    if (hdr == NULL) {
+	switch (htype) {
+	case PJSIP_H_ACCEPT:
+	    hdr = pjsip_accept_hdr_create(endpt->pool);
+	    break;
+	case PJSIP_H_ALLOW:
+	    hdr = pjsip_allow_hdr_create(endpt->pool);
+	    break;
+	case PJSIP_H_SUPPORTED:
+	    hdr = pjsip_supported_hdr_create(endpt->pool);
+	    break;
+	default:
+	    return PJ_EINVAL;
+	}
+
+	if (hdr) {
+	    pj_list_push_back(&endpt->cap_hdr, hdr);
+	}
+    }
+
+    /* Add the tags to the header. */
+    for (i=0; i<count; ++i) {
+	pj_strdup(endpt->pool, &hdr->values[hdr->count], &tags[i]);
+	++hdr->count;
+    }
+
+    /* Done. */
+    return PJ_SUCCESS;
+}
+
+/*
+ * Get additional headers to be put in outgoing request message.
+ */
+PJ_DEF(const pjsip_hdr*) pjsip_endpt_get_request_headers(pjsip_endpoint *endpt)
+{
+    return &endpt->req_hdr;
+}
+
+
+/*
+ * Initialize endpoint.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_create(pj_pool_factory *pf,
+				       const char *name,
+                                       pjsip_endpoint **p_endpt)
+{
+    pj_status_t status;
+    pj_pool_t *pool;
+    pjsip_endpoint *endpt;
+    pjsip_max_fwd_hdr *mf_hdr;
+    pj_lock_t *lock = NULL;
+
+
+    status = pj_register_strerror(PJSIP_ERRNO_START, PJ_ERRNO_SPACE_SIZE,
+				  &pjsip_strerror);
+    pj_assert(status == PJ_SUCCESS);
+
+    PJ_LOG(5, (THIS_FILE, "Creating endpoint instance..."));
+
+    *p_endpt = NULL;
+
+    /* Create pool */
+    pool = pj_pool_create(pf, "pept%p", 
+			  PJSIP_POOL_LEN_ENDPT, PJSIP_POOL_INC_ENDPT,
+			  &pool_callback);
+    if (!pool)
+	return PJ_ENOMEM;
+
+    /* Create endpoint. */
+    endpt = PJ_POOL_ZALLOC_T(pool, pjsip_endpoint);
+    endpt->pool = pool;
+    endpt->pf = pf;
+
+    /* Init modules list. */
+    pj_list_init(&endpt->module_list);
+
+    /* Initialize exit callback list. */
+    pj_list_init(&endpt->exit_cb_list);
+
+    /* Create R/W mutex for module manipulation. */
+    status = pj_rwmutex_create(endpt->pool, "ept%p", &endpt->mod_mutex);
+    if (status != PJ_SUCCESS)
+	goto on_error;
+
+    /* Init parser. */
+    init_sip_parser();
+
+    /* Init tel: uri */
+    pjsip_tel_uri_subsys_init();
+
+    /* Get name. */
+    if (name != NULL) {
+	pj_str_t temp;
+	pj_strdup_with_null(endpt->pool, &endpt->name, pj_cstr(&temp, name));
+    } else {
+	pj_strdup_with_null(endpt->pool, &endpt->name, pj_gethostname());
+    }
+
+    /* Create mutex for the events, etc. */
+    status = pj_mutex_create_recursive( endpt->pool, "ept%p", &endpt->mutex );
+    if (status != PJ_SUCCESS) {
+	goto on_error;
+    }
+
+    /* Create timer heap to manage all timers within this endpoint. */
+    status = pj_timer_heap_create( endpt->pool, PJSIP_MAX_TIMER_COUNT, 
+                                   &endpt->timer_heap);
+    if (status != PJ_SUCCESS) {
+	goto on_error;
+    }
+
+    /* Set recursive lock for the timer heap. */
+    status = pj_lock_create_recursive_mutex( endpt->pool, "edpt%p", &lock);
+    if (status != PJ_SUCCESS) {
+	goto on_error;
+    }
+    pj_timer_heap_set_lock(endpt->timer_heap, lock, PJ_TRUE);
+
+    /* Set maximum timed out entries to process in a single poll. */
+    pj_timer_heap_set_max_timed_out_per_poll(endpt->timer_heap, 
+					     PJSIP_MAX_TIMED_OUT_ENTRIES);
+
+    /* Create ioqueue. */
+    status = pj_ioqueue_create( endpt->pool, PJSIP_MAX_TRANSPORTS, &endpt->ioqueue);
+    if (status != PJ_SUCCESS) {
+	goto on_error;
+    }
+
+    /* Create transport manager. */
+    status = pjsip_tpmgr_create( endpt->pool, endpt,
+			         &endpt_on_rx_msg,
+				 &endpt_on_tx_msg,
+				 &endpt->transport_mgr);
+    if (status != PJ_SUCCESS) {
+	goto on_error;
+    }
+
+    /* Create asynchronous DNS resolver. */
+    status = pjsip_resolver_create(endpt->pool, &endpt->resolver);
+    if (status != PJ_SUCCESS) {
+	PJ_LOG(4, (THIS_FILE, "Error creating resolver instance"));
+	goto on_error;
+    }
+
+    /* Initialize request headers. */
+    pj_list_init(&endpt->req_hdr);
+
+    /* Add "Max-Forwards" for request header. */
+    mf_hdr = pjsip_max_fwd_hdr_create(endpt->pool,
+				      PJSIP_MAX_FORWARDS_VALUE);
+    pj_list_insert_before( &endpt->req_hdr, mf_hdr);
+
+    /* Initialize capability header list. */
+    pj_list_init(&endpt->cap_hdr);
+
+
+    /* Done. */
+    *p_endpt = endpt;
+    return status;
+
+on_error:
+    if (endpt->transport_mgr) {
+	pjsip_tpmgr_destroy(endpt->transport_mgr);
+	endpt->transport_mgr = NULL;
+    }
+    if (endpt->ioqueue) {
+	pj_ioqueue_destroy(endpt->ioqueue);
+	endpt->ioqueue = NULL;
+    }
+    if (endpt->timer_heap) {
+	pj_timer_heap_destroy(endpt->timer_heap);
+	endpt->timer_heap = NULL;
+    }
+    if (endpt->mutex) {
+	pj_mutex_destroy(endpt->mutex);
+	endpt->mutex = NULL;
+    }
+    deinit_sip_parser();
+    if (endpt->mod_mutex) {
+	pj_rwmutex_destroy(endpt->mod_mutex);
+	endpt->mod_mutex = NULL;
+    }
+    pj_pool_release( endpt->pool );
+
+    PJ_LOG(4, (THIS_FILE, "Error creating endpoint"));
+    return status;
+}
+
+/*
+ * Destroy endpoint.
+ */
+PJ_DEF(void) pjsip_endpt_destroy(pjsip_endpoint *endpt)
+{
+    pjsip_module *mod;
+    exit_cb *ecb;
+
+    PJ_LOG(5, (THIS_FILE, "Destroying endpoing instance.."));
+
+    /* Phase 1: stop all modules */
+    mod = endpt->module_list.prev;
+    while (mod != &endpt->module_list) {
+	pjsip_module *prev = mod->prev;
+	if (mod->stop) {
+	    (*mod->stop)();
+	}
+	mod = prev;
+    }
+
+    /* Phase 2: unload modules. */
+    mod = endpt->module_list.prev;
+    while (mod != &endpt->module_list) {
+	pjsip_module *prev = mod->prev;
+	unload_module(endpt, mod);
+	mod = prev;
+    }
+
+    /* Destroy resolver */
+    pjsip_resolver_destroy(endpt->resolver);
+
+    /* Shutdown and destroy all transports. */
+    pjsip_tpmgr_destroy(endpt->transport_mgr);
+
+    /* Destroy ioqueue */
+    pj_ioqueue_destroy(endpt->ioqueue);
+
+    /* Destroy timer heap */
+#if PJ_TIMER_DEBUG
+    pj_timer_heap_dump(endpt->timer_heap);
+#endif
+    pj_timer_heap_destroy(endpt->timer_heap);
+
+    /* Call all registered exit callbacks */
+    ecb = endpt->exit_cb_list.next;
+    while (ecb != &endpt->exit_cb_list) {
+	(*ecb->func)(endpt);
+	ecb = ecb->next;
+    }
+
+    /* Delete endpoint mutex. */
+    pj_mutex_destroy(endpt->mutex);
+
+    /* Deinit parser */
+    deinit_sip_parser();
+
+    /* Delete module's mutex */
+    pj_rwmutex_destroy(endpt->mod_mutex);
+
+    /* Finally destroy pool. */
+    pj_pool_release(endpt->pool);
+
+    PJ_LOG(4, (THIS_FILE, "Endpoint %p destroyed", endpt));
+}
+
+/*
+ * Get endpoint name.
+ */
+PJ_DEF(const pj_str_t*) pjsip_endpt_name(const pjsip_endpoint *endpt)
+{
+    return &endpt->name;
+}
+
+
+/*
+ * Create new pool.
+ */
+PJ_DEF(pj_pool_t*) pjsip_endpt_create_pool( pjsip_endpoint *endpt,
+					       const char *pool_name,
+					       pj_size_t initial,
+					       pj_size_t increment )
+{
+    pj_pool_t *pool;
+
+    /* Lock endpoint mutex. */
+    /* No need to lock mutex. Factory is thread safe.
+    pj_mutex_lock(endpt->mutex);
+     */
+
+    /* Create pool */
+    pool = pj_pool_create( endpt->pf, pool_name,
+			   initial, increment, &pool_callback);
+
+    /* Unlock mutex. */
+    /* No need to lock mutex. Factory is thread safe.
+    pj_mutex_unlock(endpt->mutex);
+     */
+
+    if (!pool) {
+	PJ_LOG(4, (THIS_FILE, "Unable to create pool %s!", pool_name));
+    }
+
+    return pool;
+}
+
+/*
+ * Return back pool to endpoint's pool manager to be either destroyed or
+ * recycled.
+ */
+PJ_DEF(void) pjsip_endpt_release_pool( pjsip_endpoint *endpt, pj_pool_t *pool )
+{
+    PJ_LOG(6, (THIS_FILE, "Releasing pool %s", pj_pool_getobjname(pool)));
+
+    /* Don't need to acquire mutex since pool factory is thread safe
+       pj_mutex_lock(endpt->mutex);
+     */
+    pj_pool_release( pool );
+
+    PJ_UNUSED_ARG(endpt);
+    /*
+    pj_mutex_unlock(endpt->mutex);
+     */
+}
+
+
+PJ_DEF(pj_status_t) pjsip_endpt_handle_events2(pjsip_endpoint *endpt,
+					       const pj_time_val *max_timeout,
+					       unsigned *p_count)
+{
+    /* timeout is 'out' var. This just to make compiler happy. */
+    pj_time_val timeout = { 0, 0};
+    unsigned count = 0, net_event_count = 0;
+    int c;
+
+    PJ_LOG(6, (THIS_FILE, "pjsip_endpt_handle_events()"));
+
+    /* Poll the timer. The timer heap has its own mutex for better 
+     * granularity, so we don't need to lock end endpoint. 
+     */
+    timeout.sec = timeout.msec = 0;
+    c = pj_timer_heap_poll( endpt->timer_heap, &timeout );
+    if (c > 0)
+	count += c;
+
+    /* timer_heap_poll should never ever returns negative value, or otherwise
+     * ioqueue_poll() will block forever!
+     */
+    pj_assert(timeout.sec >= 0 && timeout.msec >= 0);
+    if (timeout.msec >= 1000) timeout.msec = 999;
+
+    /* If caller specifies maximum time to wait, then compare the value with
+     * the timeout to wait from timer, and use the minimum value.
+     */
+    if (max_timeout && PJ_TIME_VAL_GT(timeout, *max_timeout)) {
+	timeout = *max_timeout;
+    }
+
+    /* Poll ioqueue. 
+     * Repeat polling the ioqueue while we have immediate events, because
+     * timer heap may process more than one events, so if we only process
+     * one network events at a time (such as when IOCP backend is used),
+     * the ioqueue may have trouble keeping up with the request rate.
+     *
+     * For example, for each send() request, one network event will be
+     *   reported by ioqueue for the send() completion. If we don't poll
+     *   the ioqueue often enough, the send() completion will not be
+     *   reported in timely manner.
+     */
+    do {
+	c = pj_ioqueue_poll( endpt->ioqueue, &timeout);
+	if (c < 0) {
+	    pj_status_t err = pj_get_netos_error();
+	    pj_thread_sleep(PJ_TIME_VAL_MSEC(timeout));
+	    if (p_count)
+		*p_count = count;
+	    return err;
+	} else if (c == 0) {
+	    break;
+	} else {
+	    net_event_count += c;
+	    timeout.sec = timeout.msec = 0;
+	}
+    } while (c > 0 && net_event_count < PJSIP_MAX_NET_EVENTS);
+
+    count += net_event_count;
+    if (p_count)
+	*p_count = count;
+
+    return PJ_SUCCESS;
+}
+
+/*
+ * Handle events.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_handle_events(pjsip_endpoint *endpt,
+					      const pj_time_val *max_timeout)
+{
+    return pjsip_endpt_handle_events2(endpt, max_timeout, NULL);
+}
+
+/*
+ * Schedule timer.
+ */
+#if PJ_TIMER_DEBUG
+PJ_DEF(pj_status_t) pjsip_endpt_schedule_timer_dbg(pjsip_endpoint *endpt,
+						    pj_timer_entry *entry,
+						    const pj_time_val *delay,
+						    const char *src_file,
+						    int src_line)
+{
+    PJ_LOG(6, (THIS_FILE, "pjsip_endpt_schedule_timer(entry=%p, delay=%u.%u)",
+			 entry, delay->sec, delay->msec));
+    return pj_timer_heap_schedule_dbg(endpt->timer_heap, entry, delay,
+                                      src_file, src_line);
+}
+#else
+PJ_DEF(pj_status_t) pjsip_endpt_schedule_timer( pjsip_endpoint *endpt,
+						pj_timer_entry *entry,
+						const pj_time_val *delay )
+{
+    PJ_LOG(6, (THIS_FILE, "pjsip_endpt_schedule_timer(entry=%p, delay=%u.%u)",
+			 entry, delay->sec, delay->msec));
+    return pj_timer_heap_schedule( endpt->timer_heap, entry, delay );
+}
+#endif
+
+/*
+ * Cancel the previously registered timer.
+ */
+PJ_DEF(void) pjsip_endpt_cancel_timer( pjsip_endpoint *endpt, 
+				       pj_timer_entry *entry )
+{
+    PJ_LOG(6, (THIS_FILE, "pjsip_endpt_cancel_timer(entry=%p)", entry));
+    pj_timer_heap_cancel( endpt->timer_heap, entry );
+}
+
+/*
+ * Get the timer heap instance of the SIP endpoint.
+ */
+PJ_DEF(pj_timer_heap_t*) pjsip_endpt_get_timer_heap(pjsip_endpoint *endpt)
+{
+    return endpt->timer_heap;
+}
+
+/* Init with default */
+PJ_DEF(void) pjsip_process_rdata_param_default(pjsip_process_rdata_param *p)
+{
+    pj_bzero(p, sizeof(*p));
+}
+
+/* Distribute rdata */
+PJ_DEF(pj_status_t) pjsip_endpt_process_rx_data( pjsip_endpoint *endpt,
+                                                 pjsip_rx_data *rdata,
+                                                 pjsip_process_rdata_param *p,
+                                                 pj_bool_t *p_handled)
+{
+    pjsip_msg *msg;
+    pjsip_process_rdata_param def_prm;
+    pjsip_module *mod;
+    pj_bool_t handled = PJ_FALSE;
+    unsigned i;
+    pj_status_t status;
+
+    PJ_ASSERT_RETURN(endpt && rdata, PJ_EINVAL);
+
+    if (p==NULL) {
+	p = &def_prm;
+	pjsip_process_rdata_param_default(p);
+    }
+
+    msg = rdata->msg_info.msg;
+
+    if (p_handled)
+	*p_handled = PJ_FALSE;
+
+    if (!p->silent) {
+	PJ_LOG(5, (THIS_FILE, "Distributing rdata to modules: %s",
+		   pjsip_rx_data_get_info(rdata)));
+	pj_log_push_indent();
+    }
+
+    LOCK_MODULE_ACCESS(endpt);
+
+    /* Find start module */
+    if (p->start_mod) {
+	mod = (pjsip_module*)
+	      pj_list_find_node(&endpt->module_list, p->start_mod);
+	if (!mod) {
+	    status = PJ_ENOTFOUND;
+	    goto on_return;
+	}
+    } else {
+	mod = endpt->module_list.next;
+    }
+
+    /* Start after the specified index */
+    for (i=0; i < p->idx_after_start && mod != &endpt->module_list; ++i) {
+	mod = mod->next;
+    }
+
+    /* Start with the specified priority */
+    while (mod != &endpt->module_list && mod->priority < (int)p->start_prio) {
+	mod = mod->next;
+    }
+
+    if (mod == &endpt->module_list) {
+	status = PJ_ENOTFOUND;
+	goto on_return;
+    }
+
+    /* Distribute */
+    if (msg->type == PJSIP_REQUEST_MSG) {
+	do {
+	    if (mod->on_rx_request)
+		handled = (*mod->on_rx_request)(rdata);
+	    if (handled)
+		break;
+	    mod = mod->next;
+	} while (mod != &endpt->module_list);
+    } else {
+	do {
+	    if (mod->on_rx_response)
+		handled = (*mod->on_rx_response)(rdata);
+	    if (handled)
+		break;
+	    mod = mod->next;
+	} while (mod != &endpt->module_list);
+    }
+
+    status = PJ_SUCCESS;
+
+on_return:
+    if (p_handled)
+	*p_handled = handled;
+
+    UNLOCK_MODULE_ACCESS(endpt);
+    if (!p->silent) {
+	pj_log_pop_indent();
+    }
+    return status;
+}
+
+/*
+ * This is the callback that is called by the transport manager when it 
+ * receives a message from the network.
+ */
+static void endpt_on_rx_msg( pjsip_endpoint *endpt,
+			     pj_status_t status,
+			     pjsip_rx_data *rdata )
+{
+    pjsip_msg *msg = rdata->msg_info.msg;
+    pjsip_process_rdata_param proc_prm;
+    pj_bool_t handled = PJ_FALSE;
+
+    if (status != PJ_SUCCESS) {
+	char info[30];
+	char errmsg[PJ_ERR_MSG_SIZE];
+
+	info[0] = '\0';
+
+	if (status == PJSIP_EMISSINGHDR) {
+	    pj_str_t p;
+
+	    p.ptr = info; p.slen = 0;
+
+	    if (rdata->msg_info.cid == NULL || rdata->msg_info.cid->id.slen)
+		pj_strcpy2(&p, "Call-ID");
+	    if (rdata->msg_info.from == NULL)
+		pj_strcpy2(&p, " From");
+	    if (rdata->msg_info.to == NULL)
+		pj_strcpy2(&p, " To");
+	    if (rdata->msg_info.via == NULL)
+		pj_strcpy2(&p, " Via");
+	    if (rdata->msg_info.cseq == NULL) 
+		pj_strcpy2(&p, " CSeq");
+
+	    p.ptr[p.slen] = '\0';
+	}
+
+	pj_strerror(status, errmsg, sizeof(errmsg));
+
+	PJ_LOG(1, (THIS_FILE, 
+		  "Error processing packet from %s:%d: %s %s [code %d]:\n"
+		  "%.*s\n"
+		  "-- end of packet.",
+		  rdata->pkt_info.src_name, 
+		  rdata->pkt_info.src_port,
+		  errmsg,
+		  info,
+		  status,
+		  (int)rdata->msg_info.len,	
+		  rdata->msg_info.msg_buf));
+	return;
+    }
+
+    PJ_LOG(5, (THIS_FILE, "Processing incoming message: %s", 
+	       pjsip_rx_data_get_info(rdata)));
+    pj_log_push_indent();
+
+#if defined(PJSIP_CHECK_VIA_SENT_BY) && PJSIP_CHECK_VIA_SENT_BY != 0
+    /* For response, check that the value in Via sent-by match the transport.
+     * If not matched, silently drop the response.
+     * Ref: RFC3261 Section 18.1.2 Receiving Response
+     */
+    if (msg->type == PJSIP_RESPONSE_MSG) {
+	const pj_str_t *local_addr;
+	int port = rdata->msg_info.via->sent_by.port;
+	pj_bool_t mismatch = PJ_FALSE;
+	if (port == 0) {
+	    pjsip_transport_type_e type;
+	    type = (pjsip_transport_type_e)rdata->tp_info.transport->key.type;
+	    port = pjsip_transport_get_default_port_for_type(type);
+	}
+	local_addr = &rdata->tp_info.transport->local_name.host;
+
+	if (pj_strcmp(&rdata->msg_info.via->sent_by.host, local_addr) != 0) {
+
+	    /* The RFC says that we should drop response when sent-by
+	     * address mismatch. But it could happen (e.g. with SER) when
+	     * endpoint with private IP is sending request to public
+	     * server.
+
+	    mismatch = PJ_TRUE;
+
+	     */
+
+	} else if (port != rdata->tp_info.transport->local_name.port) {
+	    /* Port or address mismatch, we should discard response */
+	    /* But we saw one implementation (we don't want to name it to 
+	     * protect the innocence) which put wrong sent-by port although
+	     * the "rport" parameter is correct.
+	     * So we discard the response only if the port doesn't match
+	     * both the port in sent-by and rport. We try to be lenient here!
+	     */
+	    if (rdata->msg_info.via->rport_param != 
+		rdata->tp_info.transport->local_name.port)
+		mismatch = PJ_TRUE;
+	    else {
+		PJ_LOG(4,(THIS_FILE, "Message %s from %s has mismatch port in "
+				     "sent-by but the rport parameter is "
+				     "correct",
+				     pjsip_rx_data_get_info(rdata), 
+				     rdata->pkt_info.src_name));
+	    }
+	}
+
+	if (mismatch) {
+	    PJ_TODO(ENDPT_REPORT_WHEN_DROPPING_MESSAGE);
+	    PJ_LOG(4,(THIS_FILE, "Dropping response %s from %s:%d because "
+				 "sent-by is mismatch", 
+				 pjsip_rx_data_get_info(rdata),
+				 rdata->pkt_info.src_name, 
+				 rdata->pkt_info.src_port));
+	    pj_log_pop_indent();
+	    return;
+	}
+    }
+#endif
+
+    pjsip_process_rdata_param_default(&proc_prm);
+    proc_prm.silent = PJ_TRUE;
+
+    pjsip_endpt_process_rx_data(endpt, rdata, &proc_prm, &handled);
+
+    /* No module is able to handle the message */
+    if (!handled) {
+	PJ_LOG(4,(THIS_FILE, "%s from %s:%d was dropped/unhandled by"
+			     " any modules",
+			     pjsip_rx_data_get_info(rdata),
+			     rdata->pkt_info.src_name,
+			     rdata->pkt_info.src_port));
+    }
+
+    /* Must clear mod_data before returning rdata to transport, since
+     * rdata may be reused.
+     */
+    pj_bzero(&rdata->endpt_info, sizeof(rdata->endpt_info));
+
+    pj_log_pop_indent();
+}
+
+/*
+ * This callback is called by transport manager before message is sent.
+ * Modules may inspect the message before it's actually sent.
+ */
+static pj_status_t endpt_on_tx_msg( pjsip_endpoint *endpt,
+				    pjsip_tx_data *tdata )
+{
+    pj_status_t status = PJ_SUCCESS;
+    pjsip_module *mod;
+
+    /* Distribute to modules, starting from modules with LOWEST priority */
+    LOCK_MODULE_ACCESS(endpt);
+
+    mod = endpt->module_list.prev;
+    if (tdata->msg->type == PJSIP_REQUEST_MSG) {
+	while (mod != &endpt->module_list) {
+	    if (mod->on_tx_request)
+		status = (*mod->on_tx_request)(tdata);
+	    if (status != PJ_SUCCESS)
+		break;
+	    mod = mod->prev;
+	}
+
+    } else {
+	while (mod != &endpt->module_list) {
+	    if (mod->on_tx_response)
+		status = (*mod->on_tx_response)(tdata);
+	    if (status != PJ_SUCCESS)
+		break;
+	    mod = mod->prev;
+	}
+    }
+
+    UNLOCK_MODULE_ACCESS(endpt);
+
+    return status;
+}
+
+
+/*
+ * Create transmit data buffer.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_create_tdata(  pjsip_endpoint *endpt,
+					       pjsip_tx_data **p_tdata)
+{
+    return pjsip_tx_data_create(endpt->transport_mgr, p_tdata);
+}
+
+/*
+ * Create the DNS resolver instance. 
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_create_resolver(pjsip_endpoint *endpt,
+						pj_dns_resolver **p_resv)
+{
+#if PJSIP_HAS_RESOLVER
+    PJ_ASSERT_RETURN(endpt && p_resv, PJ_EINVAL);
+    return pj_dns_resolver_create( endpt->pf, NULL, 0, endpt->timer_heap,
+				   endpt->ioqueue, p_resv);
+#else
+    PJ_UNUSED_ARG(endpt);
+    PJ_UNUSED_ARG(p_resv);
+    pj_assert(!"Resolver is disabled (PJSIP_HAS_RESOLVER==0)");
+    return PJ_EINVALIDOP;
+#endif
+}
+
+/*
+ * Set DNS resolver to be used by the SIP resolver.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_set_resolver( pjsip_endpoint *endpt,
+					      pj_dns_resolver *resv)
+{
+    return pjsip_resolver_set_resolver(endpt->resolver, resv);
+}
+
+/*
+ * Get the DNS resolver being used by the SIP resolver.
+ */
+PJ_DEF(pj_dns_resolver*) pjsip_endpt_get_resolver(pjsip_endpoint *endpt)
+{
+    PJ_ASSERT_RETURN(endpt, NULL);
+    return pjsip_resolver_get_resolver(endpt->resolver);
+}
+
+/*
+ * Resolve
+ */
+PJ_DEF(void) pjsip_endpt_resolve( pjsip_endpoint *endpt,
+				  pj_pool_t *pool,
+				  pjsip_host_info *target,
+				  void *token,
+				  pjsip_resolver_callback *cb)
+{
+    pjsip_resolve( endpt->resolver, pool, target, token, cb);
+}
+
+/*
+ * Get transport manager.
+ */
+PJ_DEF(pjsip_tpmgr*) pjsip_endpt_get_tpmgr(pjsip_endpoint *endpt)
+{
+    return endpt->transport_mgr;
+}
+
+/*
+ * Get ioqueue instance.
+ */
+PJ_DEF(pj_ioqueue_t*) pjsip_endpt_get_ioqueue(pjsip_endpoint *endpt)
+{
+    return endpt->ioqueue;
+}
+
+/*
+ * Find/create transport.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_acquire_transport(pjsip_endpoint *endpt,
+						  pjsip_transport_type_e type,
+						  const pj_sockaddr_t *remote,
+						  int addr_len,
+						  const pjsip_tpselector *sel,
+						  pjsip_transport **transport)
+{
+    return pjsip_tpmgr_acquire_transport(endpt->transport_mgr, type, 
+					 remote, addr_len, sel, transport);
+}
+
+
+/*
+ * Find/create transport.
+ */
+PJ_DEF(pj_status_t) pjsip_endpt_acquire_transport2(pjsip_endpoint *endpt,
+						   pjsip_transport_type_e type,
+						   const pj_sockaddr_t *remote,
+						   int addr_len,
+						   const pjsip_tpselector *sel,
+						   pjsip_tx_data *tdata,
+						   pjsip_transport **transport)
+{
+    return pjsip_tpmgr_acquire_transport2(endpt->transport_mgr, type, remote, 
+					  addr_len, sel, tdata, transport);
+}
+
+
+/*
+ * Report error.
+ */
+PJ_DEF(void) pjsip_endpt_log_error(  pjsip_endpoint *endpt,
+				     const char *sender,
+                                     pj_status_t error_code,
+                                     const char *format,
+                                     ... )
+{
+#if PJ_LOG_MAX_LEVEL > 0
+    char newformat[256];
+    pj_size_t len;
+    va_list marker;
+
+    va_start(marker, format);
+
+    PJ_UNUSED_ARG(endpt);
+
+    len = pj_ansi_strlen(format);
+    if (len < (int)sizeof(newformat)-30) {
+	pj_str_t errstr;
+
+	pj_ansi_strcpy(newformat, format);
+	pj_ansi_snprintf(newformat+len, sizeof(newformat)-len-1,
+			 ": [err %d] ", error_code);
+	len += pj_ansi_strlen(newformat+len);
+
+	errstr = pj_strerror( error_code, newformat+len, 
+			      sizeof(newformat)-len-1);
+
+	len += errstr.slen;
+	newformat[len] = '\0';
+
+	pj_log(sender, 1, newformat, marker);
+    } else {
+	pj_log(sender, 1, format, marker);
+    }
+
+    va_end(marker);
+#else
+    PJ_UNUSED_ARG(format);
+    PJ_UNUSED_ARG(error_code);
+    PJ_UNUSED_ARG(sender);
+    PJ_UNUSED_ARG(endpt);
+#endif
+}
+
+
+/*
+ * Dump endpoint.
+ */
+PJ_DEF(void) pjsip_endpt_dump( pjsip_endpoint *endpt, pj_bool_t detail )
+{
+#if PJ_LOG_MAX_LEVEL >= 3
+    PJ_LOG(5, (THIS_FILE, "pjsip_endpt_dump()"));
+
+    /* Lock mutex. */
+    pj_mutex_lock(endpt->mutex);
+
+    PJ_LOG(3, (THIS_FILE, "Dumping endpoint %p:", endpt));
+    
+    /* Dumping pool factory. */
+    pj_pool_factory_dump(endpt->pf, detail);
+
+    /* Pool health. */
+    PJ_LOG(3, (THIS_FILE," Endpoint pool capacity=%u, used_size=%u",
+	       pj_pool_get_capacity(endpt->pool),
+	       pj_pool_get_used_size(endpt->pool)));
+
+    /* Resolver */
+#if PJSIP_HAS_RESOLVER
+    if (pjsip_endpt_get_resolver(endpt)) {
+	pj_dns_resolver_dump(pjsip_endpt_get_resolver(endpt), detail);
+    }
+#endif
+
+    /* Transports. 
+     */
+    pjsip_tpmgr_dump_transports( endpt->transport_mgr );
+
+    /* Timer. */
+#if PJ_TIMER_DEBUG
+    pj_timer_heap_dump(endpt->timer_heap);
+#else
+    PJ_LOG(3,(THIS_FILE, " Timer heap has %u entries", 
+			pj_timer_heap_count(endpt->timer_heap)));
+#endif
+
+    /* Unlock mutex. */
+    pj_mutex_unlock(endpt->mutex);
+#else
+    PJ_UNUSED_ARG(endpt);
+    PJ_UNUSED_ARG(detail);
+    PJ_LOG(3,(THIS_FILE, "pjsip_end_dump: can't dump because it's disabled."));
+#endif
+}
+
+
+PJ_DEF(pj_status_t) pjsip_endpt_atexit( pjsip_endpoint *endpt,
+					pjsip_endpt_exit_callback func)
+{
+    exit_cb *new_cb;
+
+    PJ_ASSERT_RETURN(endpt && func, PJ_EINVAL);
+
+    new_cb = PJ_POOL_ZALLOC_T(endpt->pool, exit_cb);
+    new_cb->func = func;
+
+    pj_mutex_lock(endpt->mutex);
+    pj_list_push_back(&endpt->exit_cb_list, new_cb);
+    pj_mutex_unlock(endpt->mutex);
+
+    return PJ_SUCCESS;
+}