api_msg.c

Go to the documentation of this file.

/*
 * Copyright (c) 2008-2014, Pedigree Developers
 *
 * Please see the CONTRIB file in the root of the source tree for a full
 * list of contributors.
 *
 * Permission to use, copy, modify, and distribute this software for any
 * purpose with or without fee is hereby granted, provided that the above
 * copyright notice and this permission notice appear in all copies.
 *
 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 */

/*
 * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without modification,
 * are permitted provided that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 * 3. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
 * SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
 * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
 * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
 * OF SUCH DAMAGE.
 *
 * This file is part of the lwIP TCP/IP stack.
 *
 * Author: Adam Dunkels <adam@sics.se>
 *
 */

#include "lwip/opt.h"

#if LWIP_NETCONN /* don't build if not configured for use in lwipopts.h */

#include "lwip/priv/api_msg.h"

#include "lwip/ip.h"
#include "lwip/ip_addr.h"
#include "lwip/udp.h"
#include "lwip/tcp.h"
#include "lwip/raw.h"

#include "lwip/memp.h"
#include "lwip/igmp.h"
#include "lwip/dns.h"
#include "lwip/mld6.h"
#include "lwip/priv/tcpip_priv.h"

#include <string.h>

/* netconns are polled once per second (e.g. continue write on memory error) */
#define NETCONN_TCP_POLL_INTERVAL 2

#define SET_NONBLOCKING_CONNECT(conn, val)  do { if (val) { \
  (conn)->flags |= NETCONN_FLAG_IN_NONBLOCKING_CONNECT; \
} else { \
  (conn)->flags &= ~ NETCONN_FLAG_IN_NONBLOCKING_CONNECT; }} while(0)
#define IN_NONBLOCKING_CONNECT(conn) (((conn)->flags & NETCONN_FLAG_IN_NONBLOCKING_CONNECT) != 0)

/* forward declarations */
#if LWIP_TCP
#if LWIP_TCPIP_CORE_LOCKING
#define WRITE_DELAYED         , 1
#define WRITE_DELAYED_PARAM   , u8_t delayed
#else /* LWIP_TCPIP_CORE_LOCKING */
#define WRITE_DELAYED
#define WRITE_DELAYED_PARAM
#endif /* LWIP_TCPIP_CORE_LOCKING */
static err_t lwip_netconn_do_writemore(struct netconn *conn  WRITE_DELAYED_PARAM);
static err_t lwip_netconn_do_close_internal(struct netconn *conn  WRITE_DELAYED_PARAM);
#endif

#if LWIP_TCPIP_CORE_LOCKING
#define TCPIP_APIMSG_ACK(m)   NETCONN_SET_SAFE_ERR((m)->conn, (m)->err)
#else /* LWIP_TCPIP_CORE_LOCKING */
#define TCPIP_APIMSG_ACK(m)   do { NETCONN_SET_SAFE_ERR((m)->conn, (m)->err); sys_sem_signal(LWIP_API_MSG_SEM(m)); } while(0)
#endif /* LWIP_TCPIP_CORE_LOCKING */

#if LWIP_TCP
u8_t netconn_aborted;
#endif /* LWIP_TCP */

#if LWIP_RAW

static u8_t
recv_raw(void *arg, struct raw_pcb *pcb, struct pbuf *p,
    const ip_addr_t *addr)
{
  struct pbuf *q;
  struct netbuf *buf;
  struct netconn *conn;

  LWIP_UNUSED_ARG(addr);
  conn = (struct netconn *)arg;

  if ((conn != NULL) && sys_mbox_valid(&conn->recvmbox)) {
#if LWIP_SO_RCVBUF
    int recv_avail;
    SYS_ARCH_GET(conn->recv_avail, recv_avail);
    if ((recv_avail + (int)(p->tot_len)) > conn->recv_bufsize) {
      return 0;
    }
#endif /* LWIP_SO_RCVBUF */
    /* copy the whole packet into new pbufs */
    q = pbuf_alloc(PBUF_RAW, p->tot_len, PBUF_RAM);
    if (q != NULL) {
      if (pbuf_copy(q, p) != ERR_OK) {
        pbuf_free(q);
        q = NULL;
      }
    }

    if (q != NULL) {
      u16_t len;
      buf = (struct netbuf *)memp_malloc(MEMP_NETBUF);
      if (buf == NULL) {
        pbuf_free(q);
        return 0;
      }

      buf->p = q;
      buf->ptr = q;
      ip_addr_copy(buf->addr, *ip_current_src_addr());
      buf->port = pcb->protocol;

      len = q->tot_len;
      if (sys_mbox_trypost(&conn->recvmbox, buf) != ERR_OK) {
        netbuf_delete(buf);
        return 0;
      } else {
#if LWIP_SO_RCVBUF
        SYS_ARCH_INC(conn->recv_avail, len);
#endif /* LWIP_SO_RCVBUF */
        /* Register event with callback */
        API_EVENT(conn, NETCONN_EVT_RCVPLUS, len);
      }
    }
  }

  return 0; /* do not eat the packet */
}
#endif /* LWIP_RAW*/

#if LWIP_UDP

static void
recv_udp(void *arg, struct udp_pcb *pcb, struct pbuf *p,
   const ip_addr_t *addr, u16_t port)
{
  struct netbuf *buf;
  struct netconn *conn;
  u16_t len;
#if LWIP_SO_RCVBUF
  int recv_avail;
#endif /* LWIP_SO_RCVBUF */

  LWIP_UNUSED_ARG(pcb); /* only used for asserts... */
  LWIP_ASSERT("recv_udp must have a pcb argument", pcb != NULL);
  LWIP_ASSERT("recv_udp must have an argument", arg != NULL);
  conn = (struct netconn *)arg;
  LWIP_ASSERT("recv_udp: recv for wrong pcb!", conn->pcb.udp == pcb);

#if LWIP_SO_RCVBUF
  SYS_ARCH_GET(conn->recv_avail, recv_avail);
  if ((conn == NULL) || !sys_mbox_valid(&conn->recvmbox) ||
      ((recv_avail + (int)(p->tot_len)) > conn->recv_bufsize)) {
#else  /* LWIP_SO_RCVBUF */
  if ((conn == NULL) || !sys_mbox_valid(&conn->recvmbox)) {
#endif /* LWIP_SO_RCVBUF */
    pbuf_free(p);
    return;
  }

  buf = (struct netbuf *)memp_malloc(MEMP_NETBUF);
  if (buf == NULL) {
    pbuf_free(p);
    return;
  } else {
    buf->p = p;
    buf->ptr = p;
    ip_addr_set(&buf->addr, addr);
    buf->port = port;
#if LWIP_NETBUF_RECVINFO
    {
      /* get the UDP header - always in the first pbuf, ensured by udp_input */
      const struct udp_hdr* udphdr = (const struct udp_hdr*)ip_next_header_ptr();
#if LWIP_CHECKSUM_ON_COPY
      buf->flags = NETBUF_FLAG_DESTADDR;
#endif /* LWIP_CHECKSUM_ON_COPY */
      ip_addr_set(&buf->toaddr, ip_current_dest_addr());
      buf->toport_chksum = udphdr->dest;
    }
#endif /* LWIP_NETBUF_RECVINFO */
  }

  len = p->tot_len;
  if (sys_mbox_trypost(&conn->recvmbox, buf) != ERR_OK) {
    netbuf_delete(buf);
    return;
  } else {
#if LWIP_SO_RCVBUF
    SYS_ARCH_INC(conn->recv_avail, len);
#endif /* LWIP_SO_RCVBUF */
    /* Register event with callback */
    API_EVENT(conn, NETCONN_EVT_RCVPLUS, len);
  }
}
#endif /* LWIP_UDP */

#if LWIP_TCP

static err_t
recv_tcp(void *arg, struct tcp_pcb *pcb, struct pbuf *p, err_t err)
{
  struct netconn *conn;
  u16_t len;

  LWIP_UNUSED_ARG(pcb);
  LWIP_ASSERT("recv_tcp must have a pcb argument", pcb != NULL);
  LWIP_ASSERT("recv_tcp must have an argument", arg != NULL);
  conn = (struct netconn *)arg;

  if (conn == NULL) {
    return ERR_VAL;
  }
  LWIP_ASSERT("recv_tcp: recv for wrong pcb!", conn->pcb.tcp == pcb);

  if (!sys_mbox_valid(&conn->recvmbox)) {
    /* recvmbox already deleted */
    if (p != NULL) {
      tcp_recved(pcb, p->tot_len);
      pbuf_free(p);
    }
    return ERR_OK;
  }
  /* Unlike for UDP or RAW pcbs, don't check for available space
     using recv_avail since that could break the connection
     (data is already ACKed) */

  /* don't overwrite fatal errors! */
  if (err != ERR_OK) {
    NETCONN_SET_SAFE_ERR(conn, err);
  }

  if (p != NULL) {
    len = p->tot_len;
  } else {
    len = 0;
  }

  if (sys_mbox_trypost(&conn->recvmbox, p) != ERR_OK) {
    /* don't deallocate p: it is presented to us later again from tcp_fasttmr! */
    return ERR_MEM;
  } else {
#if LWIP_SO_RCVBUF
    SYS_ARCH_INC(conn->recv_avail, len);
#endif /* LWIP_SO_RCVBUF */
    /* Register event with callback */
    API_EVENT(conn, NETCONN_EVT_RCVPLUS, len);
  }

  return ERR_OK;
}

static err_t
poll_tcp(void *arg, struct tcp_pcb *pcb)
{
  struct netconn *conn = (struct netconn *)arg;

  LWIP_UNUSED_ARG(pcb);
  LWIP_ASSERT("conn != NULL", (conn != NULL));

  if (conn->state == NETCONN_WRITE) {
    lwip_netconn_do_writemore(conn  WRITE_DELAYED);
  } else if (conn->state == NETCONN_CLOSE) {
#if !LWIP_SO_SNDTIMEO && !LWIP_SO_LINGER
    if (conn->current_msg && conn->current_msg->msg.sd.polls_left) {
      conn->current_msg->msg.sd.polls_left--;
    }
#endif /* !LWIP_SO_SNDTIMEO && !LWIP_SO_LINGER */
    lwip_netconn_do_close_internal(conn  WRITE_DELAYED);
  }
  /* @todo: implement connect timeout here? */

  /* Did a nonblocking write fail before? Then check available write-space. */
  if (conn->flags & NETCONN_FLAG_CHECK_WRITESPACE) {
    /* If the queued byte- or pbuf-count drops below the configured low-water limit,
       let select mark this pcb as writable again. */
    if ((conn->pcb.tcp != NULL) && (tcp_sndbuf(conn->pcb.tcp) > TCP_SNDLOWAT) &&
      (tcp_sndqueuelen(conn->pcb.tcp) < TCP_SNDQUEUELOWAT)) {
      conn->flags &= ~NETCONN_FLAG_CHECK_WRITESPACE;
      API_EVENT(conn, NETCONN_EVT_SENDPLUS, 0);
    }
  }

  return ERR_OK;
}

static err_t
sent_tcp(void *arg, struct tcp_pcb *pcb, u16_t len)
{
  struct netconn *conn = (struct netconn *)arg;

  LWIP_UNUSED_ARG(pcb);
  LWIP_ASSERT("conn != NULL", (conn != NULL));

  if (conn) {
    if (conn->state == NETCONN_WRITE) {
      lwip_netconn_do_writemore(conn  WRITE_DELAYED);
    } else if (conn->state == NETCONN_CLOSE) {
      lwip_netconn_do_close_internal(conn  WRITE_DELAYED);
    }

    /* If the queued byte- or pbuf-count drops below the configured low-water limit,
       let select mark this pcb as writable again. */
    if ((conn->pcb.tcp != NULL) && (tcp_sndbuf(conn->pcb.tcp) > TCP_SNDLOWAT) &&
      (tcp_sndqueuelen(conn->pcb.tcp) < TCP_SNDQUEUELOWAT)) {
      conn->flags &= ~NETCONN_FLAG_CHECK_WRITESPACE;
      API_EVENT(conn, NETCONN_EVT_SENDPLUS, len);
    }
  }

  return ERR_OK;
}

static void
err_tcp(void *arg, err_t err)
{
  struct netconn *conn;
  enum netconn_state old_state;

  conn = (struct netconn *)arg;
  LWIP_ASSERT("conn != NULL", (conn != NULL));

  conn->pcb.tcp = NULL;

  /* reset conn->state now before waking up other threads */
  old_state = conn->state;
  conn->state = NETCONN_NONE;

  if (old_state == NETCONN_CLOSE) {
    /* RST during close: let close return success & dealloc the netconn */
    err = ERR_OK;
    NETCONN_SET_SAFE_ERR(conn, ERR_OK);
  } else {
    /* no check since this is always fatal! */
    SYS_ARCH_SET(conn->last_err, err);
  }

  /* @todo: the type of NETCONN_EVT created should depend on 'old_state' */

  /* Notify the user layer about a connection error. Used to signal select. */
  API_EVENT(conn, NETCONN_EVT_ERROR, 0);
  /* Try to release selects pending on 'read' or 'write', too.
     They will get an error if they actually try to read or write. */
  API_EVENT(conn, NETCONN_EVT_RCVPLUS, 0);
  API_EVENT(conn, NETCONN_EVT_SENDPLUS, 0);

  /* pass NULL-message to recvmbox to wake up pending recv */
  if (sys_mbox_valid(&conn->recvmbox)) {
    /* use trypost to prevent deadlock */
    sys_mbox_trypost(&conn->recvmbox, NULL);
  }
  /* pass NULL-message to acceptmbox to wake up pending accept */
  if (sys_mbox_valid(&conn->acceptmbox)) {
    /* use trypost to preven deadlock */
    sys_mbox_trypost(&conn->acceptmbox, NULL);
  }

  if ((old_state == NETCONN_WRITE) || (old_state == NETCONN_CLOSE) ||
      (old_state == NETCONN_CONNECT)) {
    /* calling lwip_netconn_do_writemore/lwip_netconn_do_close_internal is not necessary
       since the pcb has already been deleted! */
    int was_nonblocking_connect = IN_NONBLOCKING_CONNECT(conn);
    SET_NONBLOCKING_CONNECT(conn, 0);

    if (!was_nonblocking_connect) {
      sys_sem_t* op_completed_sem;
      /* set error return code */
      LWIP_ASSERT("conn->current_msg != NULL", conn->current_msg != NULL);
      conn->current_msg->err = err;
      op_completed_sem = LWIP_API_MSG_SEM(conn->current_msg);
      LWIP_ASSERT("inavlid op_completed_sem", sys_sem_valid(op_completed_sem));
      conn->current_msg = NULL;
      /* wake up the waiting task */
      NETCONN_SET_SAFE_ERR(conn, err);
      sys_sem_signal(op_completed_sem);
    }
  } else {
    LWIP_ASSERT("conn->current_msg == NULL", conn->current_msg == NULL);
  }
}

static void
setup_tcp(struct netconn *conn)
{
  struct tcp_pcb *pcb;

  pcb = conn->pcb.tcp;
  tcp_arg(pcb, conn);
  tcp_recv(pcb, recv_tcp);
  tcp_sent(pcb, sent_tcp);
  tcp_poll(pcb, poll_tcp, NETCONN_TCP_POLL_INTERVAL);
  tcp_err(pcb, err_tcp);
}

static err_t
accept_function(void *arg, struct tcp_pcb *newpcb, err_t err)
{
  struct netconn *newconn;
  struct netconn *conn = (struct netconn *)arg;

  LWIP_DEBUGF(API_MSG_DEBUG, ("accept_function: newpcb->tate: %s\n", tcp_debug_state_str(newpcb->state)));

  if (conn == NULL) {
    return ERR_VAL;
  }
  if (!sys_mbox_valid(&conn->acceptmbox)) {
    LWIP_DEBUGF(API_MSG_DEBUG, ("accept_function: acceptmbox already deleted\n"));
    return ERR_VAL;
  }

  if (newpcb == NULL) {
    /* out-of-pcbs during connect: pass on this error to the application */
    if (sys_mbox_trypost(&conn->acceptmbox, &netconn_aborted) == ERR_OK) {
      /* Register event with callback */
      API_EVENT(conn, NETCONN_EVT_RCVPLUS, 0);
    }
    return ERR_VAL;
  }

  /* We have to set the callback here even though
   * the new socket is unknown. newconn->socket is marked as -1. */
  newconn = netconn_alloc(conn->type, conn->callback);
  if (newconn == NULL) {
    /* outof netconns: pass on this error to the application */
    if (sys_mbox_trypost(&conn->acceptmbox, &netconn_aborted) == ERR_OK) {
      /* Register event with callback */
      API_EVENT(conn, NETCONN_EVT_RCVPLUS, 0);
    }
    return ERR_MEM;
  }
  newconn->pcb.tcp = newpcb;
  setup_tcp(newconn);
  /* no protection: when creating the pcb, the netconn is not yet known
     to the application thread */
  newconn->last_err = err;

  /* handle backlog counter */
  tcp_backlog_delayed(newpcb);

  if (sys_mbox_trypost(&conn->acceptmbox, newconn) != ERR_OK) {
    /* When returning != ERR_OK, the pcb is aborted in tcp_process(),
       so do nothing here! */
    /* remove all references to this netconn from the pcb */
    struct tcp_pcb* pcb = newconn->pcb.tcp;
    tcp_arg(pcb, NULL);
    tcp_recv(pcb, NULL);
    tcp_sent(pcb, NULL);
    tcp_poll(pcb, NULL, 0);
    tcp_err(pcb, NULL);
    /* remove reference from to the pcb from this netconn */
    newconn->pcb.tcp = NULL;
    /* no need to drain since we know the recvmbox is empty. */
    sys_mbox_free(&newconn->recvmbox);
    sys_mbox_set_invalid(&newconn->recvmbox);
    netconn_free(newconn);
    return ERR_MEM;
  } else {
    /* Register event with callback */
    API_EVENT(conn, NETCONN_EVT_RCVPLUS, 0);
  }

  return ERR_OK;
}
#endif /* LWIP_TCP */

static void
pcb_new(struct api_msg *msg)
{
  enum lwip_ip_addr_type iptype = IPADDR_TYPE_V4;

  LWIP_ASSERT("pcb_new: pcb already allocated", msg->conn->pcb.tcp == NULL);
 
#if LWIP_IPV6 && LWIP_IPV4
  /* IPv6: Dual-stack by default, unless netconn_set_ipv6only() is called */
  if(NETCONNTYPE_ISIPV6(netconn_type(msg->conn))) {
    iptype = IPADDR_TYPE_ANY;
  }
#endif
  
  /* Allocate a PCB for this connection */
  switch(NETCONNTYPE_GROUP(msg->conn->type)) {
#if LWIP_RAW
  case NETCONN_RAW:
    msg->conn->pcb.raw = raw_new_ip_type(iptype, msg->msg.n.proto);
    if (msg->conn->pcb.raw != NULL) {
#if LWIP_IPV6
      /* ICMPv6 packets should always have checksum calculated by the stack as per RFC 3542 chapter 3.1 */
      if (NETCONNTYPE_ISIPV6(msg->conn->type) && msg->conn->pcb.raw->protocol == IP6_NEXTH_ICMP6) {
        msg->conn->pcb.raw->chksum_reqd = 1;
        msg->conn->pcb.raw->chksum_offset = 2;
      }
#endif /* LWIP_IPV6 */
      raw_recv(msg->conn->pcb.raw, recv_raw, msg->conn);
    }
    break;
#endif /* LWIP_RAW */
#if LWIP_UDP
  case NETCONN_UDP:
    msg->conn->pcb.udp = udp_new_ip_type(iptype);
    if (msg->conn->pcb.udp != NULL) {
#if LWIP_UDPLITE
      if (NETCONNTYPE_ISUDPLITE(msg->conn->type)) {
        udp_setflags(msg->conn->pcb.udp, UDP_FLAGS_UDPLITE);
      }
#endif /* LWIP_UDPLITE */
      if (NETCONNTYPE_ISUDPNOCHKSUM(msg->conn->type)) {
        udp_setflags(msg->conn->pcb.udp, UDP_FLAGS_NOCHKSUM);
      }
      udp_recv(msg->conn->pcb.udp, recv_udp, msg->conn);
    }
    break;
#endif /* LWIP_UDP */
#if LWIP_TCP
  case NETCONN_TCP:
    msg->conn->pcb.tcp = tcp_new_ip_type(iptype);
    if (msg->conn->pcb.tcp != NULL) {
      setup_tcp(msg->conn);
    }
    break;
#endif /* LWIP_TCP */
  default:
    /* Unsupported netconn type, e.g. protocol disabled */
    msg->err = ERR_VAL;
    return;
  }
  if (msg->conn->pcb.ip == NULL) {
    msg->err = ERR_MEM;
  }
}

void
lwip_netconn_do_newconn(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  msg->err = ERR_OK;
  if (msg->conn->pcb.tcp == NULL) {
    pcb_new(msg);
  }
  /* Else? This "new" connection already has a PCB allocated. */
  /* Is this an error condition? Should it be deleted? */
  /* We currently just are happy and return. */

  TCPIP_APIMSG_ACK(msg);
}

struct netconn*
netconn_alloc(enum netconn_type t, netconn_callback callback)
{
  struct netconn *conn;
  int size;

  conn = (struct netconn *)memp_malloc(MEMP_NETCONN);
  if (conn == NULL) {
    return NULL;
  }

  conn->last_err = ERR_OK;
  conn->type = t;
  conn->pcb.tcp = NULL;

  /* If all sizes are the same, every compiler should optimize this switch to nothing */
  switch(NETCONNTYPE_GROUP(t)) {
#if LWIP_RAW
  case NETCONN_RAW:
    size = DEFAULT_RAW_RECVMBOX_SIZE;
    break;
#endif /* LWIP_RAW */
#if LWIP_UDP
  case NETCONN_UDP:
    size = DEFAULT_UDP_RECVMBOX_SIZE;
    break;
#endif /* LWIP_UDP */
#if LWIP_TCP
  case NETCONN_TCP:
    size = DEFAULT_TCP_RECVMBOX_SIZE;
    break;
#endif /* LWIP_TCP */
  default:
    LWIP_ASSERT("netconn_alloc: undefined netconn_type", 0);
    goto free_and_return;
  }

  if (sys_mbox_new(&conn->recvmbox, size) != ERR_OK) {
    goto free_and_return;
  }
#if !LWIP_NETCONN_SEM_PER_THREAD
  if (sys_sem_new(&conn->op_completed, 0) != ERR_OK) {
    sys_mbox_free(&conn->recvmbox);
    goto free_and_return;
  }
#endif

#if LWIP_TCP
  sys_mbox_set_invalid(&conn->acceptmbox);
#endif
  conn->state        = NETCONN_NONE;
#if LWIP_SOCKET
  /* initialize socket to -1 since 0 is a valid socket */
  conn->socket       = -1;
#endif /* LWIP_SOCKET */
  conn->callback     = callback;
#if LWIP_TCP
  conn->current_msg  = NULL;
  conn->write_offset = 0;
#endif /* LWIP_TCP */
#if LWIP_SO_SNDTIMEO
  conn->send_timeout = 0;
#endif /* LWIP_SO_SNDTIMEO */
#if LWIP_SO_RCVTIMEO
  conn->recv_timeout = 0;
#endif /* LWIP_SO_RCVTIMEO */
#if LWIP_SO_RCVBUF
  conn->recv_bufsize = RECV_BUFSIZE_DEFAULT;
  conn->recv_avail   = 0;
#endif /* LWIP_SO_RCVBUF */
#if LWIP_SO_LINGER
  conn->linger = -1;
#endif /* LWIP_SO_LINGER */
  conn->flags = 0;
  return conn;
free_and_return:
  memp_free(MEMP_NETCONN, conn);
  return NULL;
}

void
netconn_free(struct netconn *conn)
{
  LWIP_ASSERT("PCB must be deallocated outside this function", conn->pcb.tcp == NULL);
  LWIP_ASSERT("recvmbox must be deallocated before calling this function",
    !sys_mbox_valid(&conn->recvmbox));
#if LWIP_TCP
  LWIP_ASSERT("acceptmbox must be deallocated before calling this function",
    !sys_mbox_valid(&conn->acceptmbox));
#endif /* LWIP_TCP */

#if !LWIP_NETCONN_SEM_PER_THREAD
  sys_sem_free(&conn->op_completed);
  sys_sem_set_invalid(&conn->op_completed);
#endif

  memp_free(MEMP_NETCONN, conn);
}

static void
netconn_drain(struct netconn *conn)
{
  void *mem;
#if LWIP_TCP
  struct pbuf *p;
#endif /* LWIP_TCP */

  /* This runs in tcpip_thread, so we don't need to lock against rx packets */

  /* Delete and drain the recvmbox. */
  if (sys_mbox_valid(&conn->recvmbox)) {
    while (sys_mbox_tryfetch(&conn->recvmbox, &mem) != SYS_MBOX_EMPTY) {
#if LWIP_TCP
      if (NETCONNTYPE_GROUP(conn->type) == NETCONN_TCP) {
        if (mem != NULL) {
          p = (struct pbuf*)mem;
          /* pcb might be set to NULL already by err_tcp() */
          if (conn->pcb.tcp != NULL) {
            tcp_recved(conn->pcb.tcp, p->tot_len);
          }
          pbuf_free(p);
        }
      } else
#endif /* LWIP_TCP */
      {
        netbuf_delete((struct netbuf *)mem);
      }
    }
    sys_mbox_free(&conn->recvmbox);
    sys_mbox_set_invalid(&conn->recvmbox);
  }

  /* Delete and drain the acceptmbox. */
#if LWIP_TCP
  if (sys_mbox_valid(&conn->acceptmbox)) {
    while (sys_mbox_tryfetch(&conn->acceptmbox, &mem) != SYS_MBOX_EMPTY) {
      if (mem != &netconn_aborted) {
        struct netconn *newconn = (struct netconn *)mem;
        /* Only tcp pcbs have an acceptmbox, so no need to check conn->type */
        /* pcb might be set to NULL already by err_tcp() */
        /* drain recvmbox */
        netconn_drain(newconn);
        if (newconn->pcb.tcp != NULL) {
          tcp_abort(newconn->pcb.tcp);
          newconn->pcb.tcp = NULL;
        }
        netconn_free(newconn);
      }
    }
    sys_mbox_free(&conn->acceptmbox);
    sys_mbox_set_invalid(&conn->acceptmbox);
  }
#endif /* LWIP_TCP */
}

#if LWIP_TCP

static err_t
lwip_netconn_do_close_internal(struct netconn *conn  WRITE_DELAYED_PARAM)
{
  err_t err;
  u8_t shut, shut_rx, shut_tx, close;
  u8_t close_finished = 0;
  struct tcp_pcb* tpcb;
#if LWIP_SO_LINGER
  u8_t linger_wait_required = 0;
#endif /* LWIP_SO_LINGER */

  LWIP_ASSERT("invalid conn", (conn != NULL));
  LWIP_ASSERT("this is for tcp netconns only", (NETCONNTYPE_GROUP(conn->type) == NETCONN_TCP));
  LWIP_ASSERT("conn must be in state NETCONN_CLOSE", (conn->state == NETCONN_CLOSE));
  LWIP_ASSERT("pcb already closed", (conn->pcb.tcp != NULL));
  LWIP_ASSERT("conn->current_msg != NULL", conn->current_msg != NULL);

  tpcb = conn->pcb.tcp;
  shut = conn->current_msg->msg.sd.shut;
  shut_rx = shut & NETCONN_SHUT_RD;
  shut_tx = shut & NETCONN_SHUT_WR;
  /* shutting down both ends is the same as closing
     (also if RD or WR side was shut down before already) */
  if (shut == NETCONN_SHUT_RDWR) {
    close = 1;
  } else if (shut_rx &&
             ((tpcb->state == FIN_WAIT_1) ||
              (tpcb->state == FIN_WAIT_2) ||
              (tpcb->state == CLOSING))) {
    close = 1;
  } else if (shut_tx && ((tpcb->flags & TF_RXCLOSED) != 0)) {
    close = 1;
  } else {
    close = 0;
  }

  /* Set back some callback pointers */
  if (close) {
    tcp_arg(tpcb, NULL);
  }
  if (tpcb->state == LISTEN) {
    tcp_accept(tpcb, NULL);
  } else {
    /* some callbacks have to be reset if tcp_close is not successful */
    if (shut_rx) {
      tcp_recv(tpcb, NULL);
      tcp_accept(tpcb, NULL);
    }
    if (shut_tx) {
      tcp_sent(tpcb, NULL);
    }
    if (close) {
      tcp_poll(tpcb, NULL, 0);
      tcp_err(tpcb, NULL);
    }
  }
  /* Try to close the connection */
  if (close) {
#if LWIP_SO_LINGER
    /* check linger possibilites before calling tcp_close */
    err = ERR_OK;
    /* linger enabled/required at all? (i.e. is there untransmitted data left?) */
    if ((conn->linger >= 0) && (conn->pcb.tcp->unsent || conn->pcb.tcp->unacked)) {
      if ((conn->linger == 0)) {
        /* data left but linger prevents waiting */
        tcp_abort(tpcb);
        tpcb = NULL;
      } else if (conn->linger > 0) {
        /* data left and linger says we should wait */
        if (netconn_is_nonblocking(conn)) {
          /* data left on a nonblocking netconn -> cannot linger */
          err = ERR_WOULDBLOCK;
        } else if ((s32_t)(sys_now() - conn->current_msg->msg.sd.time_started) >=
          (conn->linger * 1000)) {
          /* data left but linger timeout has expired (this happens on further
             calls to this function through poll_tcp */
          tcp_abort(tpcb);
          tpcb = NULL;
        } else {
          /* data left -> need to wait for ACK after successful close */
          linger_wait_required = 1;
        }
      }
    }
    if ((err == ERR_OK) && (tpcb != NULL))
#endif /* LWIP_SO_LINGER */
    {
      err = tcp_close(tpcb);
    }
  } else {
    err = tcp_shutdown(tpcb, shut_rx, shut_tx);
  }
  if (err == ERR_OK) {
    close_finished = 1;
#if LWIP_SO_LINGER
    if (linger_wait_required) {
      /* wait for ACK of all unsent/unacked data by just getting called again */
      close_finished = 0;
      err = ERR_INPROGRESS;
    }
#endif /* LWIP_SO_LINGER */
  } else {
    if (err == ERR_MEM) {
      /* Closing failed because of memory shortage, try again later. Even for
         nonblocking netconns, we have to wait since no standard socket application
         is prepared for close failing because of resource shortage.
         Check the timeout: this is kind of an lwip addition to the standard sockets:
         we wait for some time when failing to allocate a segment for the FIN */
#if LWIP_SO_SNDTIMEO || LWIP_SO_LINGER
      s32_t close_timeout = LWIP_TCP_CLOSE_TIMEOUT_MS_DEFAULT;
#if LWIP_SO_SNDTIMEO
      if (conn->send_timeout > 0) {
        close_timeout = conn->send_timeout;
      }
#endif /* LWIP_SO_SNDTIMEO */
#if LWIP_SO_LINGER
      if (conn->linger >= 0) {
        /* use linger timeout (seconds) */
        close_timeout = conn->linger * 1000U;
      }
#endif
      if ((s32_t)(sys_now() - conn->current_msg->msg.sd.time_started) >= close_timeout) {
#else /* LWIP_SO_SNDTIMEO || LWIP_SO_LINGER */
      if (conn->current_msg->msg.sd.polls_left == 0) {
#endif /* LWIP_SO_SNDTIMEO || LWIP_SO_LINGER */
        close_finished = 1;
        if (close) {
          /* in this case, we want to RST the connection */
          tcp_abort(tpcb);
          err = ERR_OK;
        }
      }
    } else {
      /* Closing failed for a non-memory error: give up */
      close_finished = 1;
    }
  }
  if (close_finished) {
    /* Closing done (succeeded, non-memory error, nonblocking error or timeout) */
    sys_sem_t* op_completed_sem = LWIP_API_MSG_SEM(conn->current_msg);
    conn->current_msg->err = err;
    conn->current_msg = NULL;
    conn->state = NETCONN_NONE;
    if (err == ERR_OK) {
      if (close) {
        /* Set back some callback pointers as conn is going away */
        conn->pcb.tcp = NULL;
        /* Trigger select() in socket layer. Make sure everybody notices activity
         on the connection, error first! */
        API_EVENT(conn, NETCONN_EVT_ERROR, 0);
      }
      if (shut_rx) {
        API_EVENT(conn, NETCONN_EVT_RCVPLUS, 0);
      }
      if (shut_tx) {
        API_EVENT(conn, NETCONN_EVT_SENDPLUS, 0);
      }
    }
    NETCONN_SET_SAFE_ERR(conn, err);
#if LWIP_TCPIP_CORE_LOCKING
    if (delayed)
#endif
    {
      /* wake up the application task */
      sys_sem_signal(op_completed_sem);
    }
    return ERR_OK;
  }
  if (!close_finished) {
    /* Closing failed and we want to wait: restore some of the callbacks */
    /* Closing of listen pcb will never fail! */
    LWIP_ASSERT("Closing a listen pcb may not fail!", (tpcb->state != LISTEN));
    if (shut_tx) {
      tcp_sent(tpcb, sent_tcp);
    }
    /* when waiting for close, set up poll interval to 500ms */
    tcp_poll(tpcb, poll_tcp, 1);
    tcp_err(tpcb, err_tcp);
    tcp_arg(tpcb, conn);
    /* don't restore recv callback: we don't want to receive any more data */
  }
  /* If closing didn't succeed, we get called again either
     from poll_tcp or from sent_tcp */
  LWIP_ASSERT("err != ERR_OK", err != ERR_OK);
  return err;
}
#endif /* LWIP_TCP */

void
lwip_netconn_do_delconn(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  enum netconn_state state = msg->conn->state;
  LWIP_ASSERT("netconn state error", /* this only happens for TCP netconns */
    (state == NETCONN_NONE) || (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_TCP));
#if LWIP_NETCONN_FULLDUPLEX
  /* In full duplex mode, blocking write/connect is aborted with ERR_CLSD */
  if (state != NETCONN_NONE) {
    if ((state == NETCONN_WRITE) ||
        ((state == NETCONN_CONNECT) && !IN_NONBLOCKING_CONNECT(msg->conn))) {
      /* close requested, abort running write/connect */
      sys_sem_t* op_completed_sem;
      LWIP_ASSERT("msg->conn->current_msg != NULL", msg->conn->current_msg != NULL);
      op_completed_sem = LWIP_API_MSG_SEM(msg->conn->current_msg);
      msg->conn->current_msg->err = ERR_CLSD;
      msg->conn->current_msg = NULL;
      msg->conn->write_offset = 0;
      msg->conn->state = NETCONN_NONE;
      NETCONN_SET_SAFE_ERR(msg->conn, ERR_CLSD);
      sys_sem_signal(op_completed_sem);
    }
  }
#else /* LWIP_NETCONN_FULLDUPLEX */
  if (((state != NETCONN_NONE) &&
       (state != NETCONN_LISTEN) &&
       (state != NETCONN_CONNECT)) ||
      ((state == NETCONN_CONNECT) && !IN_NONBLOCKING_CONNECT(msg->conn))) {
    /* This means either a blocking write or blocking connect is running
       (nonblocking write returns and sets state to NONE) */
    msg->err = ERR_INPROGRESS;
  } else
#endif /* LWIP_NETCONN_FULLDUPLEX */
  {
    LWIP_ASSERT("blocking connect in progress",
      (state != NETCONN_CONNECT) || IN_NONBLOCKING_CONNECT(msg->conn));
    msg->err = ERR_OK;
    /* Drain and delete mboxes */
    netconn_drain(msg->conn);

    if (msg->conn->pcb.tcp != NULL) {

      switch (NETCONNTYPE_GROUP(msg->conn->type)) {
#if LWIP_RAW
      case NETCONN_RAW:
        raw_remove(msg->conn->pcb.raw);
        break;
#endif /* LWIP_RAW */
#if LWIP_UDP
      case NETCONN_UDP:
        msg->conn->pcb.udp->recv_arg = NULL;
        udp_remove(msg->conn->pcb.udp);
        break;
#endif /* LWIP_UDP */
#if LWIP_TCP
      case NETCONN_TCP:
        LWIP_ASSERT("already writing or closing", msg->conn->current_msg == NULL &&
          msg->conn->write_offset == 0);
        msg->conn->state = NETCONN_CLOSE;
        msg->msg.sd.shut = NETCONN_SHUT_RDWR;
        msg->conn->current_msg = msg;
#if LWIP_TCPIP_CORE_LOCKING
        if (lwip_netconn_do_close_internal(msg->conn, 0) != ERR_OK) {
          LWIP_ASSERT("state!", msg->conn->state == NETCONN_CLOSE);
          UNLOCK_TCPIP_CORE();
          sys_arch_sem_wait(LWIP_API_MSG_SEM(msg), 0);
          LOCK_TCPIP_CORE();
          LWIP_ASSERT("state!", msg->conn->state == NETCONN_NONE);
        }
#else /* LWIP_TCPIP_CORE_LOCKING */
        lwip_netconn_do_close_internal(msg->conn);
#endif /* LWIP_TCPIP_CORE_LOCKING */
        /* API_EVENT is called inside lwip_netconn_do_close_internal, before releasing
           the application thread, so we can return at this point! */
        return;
#endif /* LWIP_TCP */
      default:
        break;
      }
      msg->conn->pcb.tcp = NULL;
    }
    /* tcp netconns don't come here! */

    /* @todo: this lets select make the socket readable and writable,
       which is wrong! errfd instead? */
    API_EVENT(msg->conn, NETCONN_EVT_RCVPLUS, 0);
    API_EVENT(msg->conn, NETCONN_EVT_SENDPLUS, 0);
  }
  if (sys_sem_valid(LWIP_API_MSG_SEM(msg))) {
    TCPIP_APIMSG_ACK(msg);
  }
}

void
lwip_netconn_do_bind(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  if (ERR_IS_FATAL(msg->conn->last_err)) {
    msg->err = msg->conn->last_err;
  } else {
    msg->err = ERR_VAL;
    if (msg->conn->pcb.tcp != NULL) {
      switch (NETCONNTYPE_GROUP(msg->conn->type)) {
#if LWIP_RAW
      case NETCONN_RAW:
        msg->err = raw_bind(msg->conn->pcb.raw, API_EXPR_REF(msg->msg.bc.ipaddr));
        break;
#endif /* LWIP_RAW */
#if LWIP_UDP
      case NETCONN_UDP:
        msg->err = udp_bind(msg->conn->pcb.udp, API_EXPR_REF(msg->msg.bc.ipaddr), msg->msg.bc.port);
        break;
#endif /* LWIP_UDP */
#if LWIP_TCP
      case NETCONN_TCP:
        msg->err = tcp_bind(msg->conn->pcb.tcp, API_EXPR_REF(msg->msg.bc.ipaddr), msg->msg.bc.port);
        break;
#endif /* LWIP_TCP */
      default:
        break;
      }
    }
  }
  TCPIP_APIMSG_ACK(msg);
}

#if LWIP_TCP

static err_t
lwip_netconn_do_connected(void *arg, struct tcp_pcb *pcb, err_t err)
{
  struct netconn *conn;
  int was_blocking;
  sys_sem_t* op_completed_sem = NULL;

  LWIP_UNUSED_ARG(pcb);

  conn = (struct netconn *)arg;

  if (conn == NULL) {
    return ERR_VAL;
  }

  LWIP_ASSERT("conn->state == NETCONN_CONNECT", conn->state == NETCONN_CONNECT);
  LWIP_ASSERT("(conn->current_msg != NULL) || conn->in_non_blocking_connect",
    (conn->current_msg != NULL) || IN_NONBLOCKING_CONNECT(conn));

  if (conn->current_msg != NULL) {
    conn->current_msg->err = err;
    op_completed_sem = LWIP_API_MSG_SEM(conn->current_msg);
  }
  if ((NETCONNTYPE_GROUP(conn->type) == NETCONN_TCP) && (err == ERR_OK)) {
    setup_tcp(conn);
  }
  was_blocking = !IN_NONBLOCKING_CONNECT(conn);
  SET_NONBLOCKING_CONNECT(conn, 0);
  LWIP_ASSERT("blocking connect state error",
    (was_blocking && op_completed_sem != NULL) ||
    (!was_blocking && op_completed_sem == NULL));
  conn->current_msg = NULL;
  conn->state = NETCONN_NONE;
  NETCONN_SET_SAFE_ERR(conn, ERR_OK);
  API_EVENT(conn, NETCONN_EVT_SENDPLUS, 0);

  if (was_blocking) {
    sys_sem_signal(op_completed_sem);
  }
  return ERR_OK;
}
#endif /* LWIP_TCP */

void
lwip_netconn_do_connect(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  if (msg->conn->pcb.tcp == NULL) {
    /* This may happen when calling netconn_connect() a second time */
    msg->err = ERR_CLSD;
  } else {
    switch (NETCONNTYPE_GROUP(msg->conn->type)) {
#if LWIP_RAW
    case NETCONN_RAW:
      msg->err = raw_connect(msg->conn->pcb.raw, API_EXPR_REF(msg->msg.bc.ipaddr));
      break;
#endif /* LWIP_RAW */
#if LWIP_UDP
    case NETCONN_UDP:
      msg->err = udp_connect(msg->conn->pcb.udp, API_EXPR_REF(msg->msg.bc.ipaddr), msg->msg.bc.port);
      break;
#endif /* LWIP_UDP */
#if LWIP_TCP
    case NETCONN_TCP:
      /* Prevent connect while doing any other action. */
      if (msg->conn->state == NETCONN_CONNECT) {
        msg->err = ERR_ALREADY;
      } else if (msg->conn->state != NETCONN_NONE) {
        msg->err = ERR_ISCONN;
      } else {
        setup_tcp(msg->conn);
        msg->err = tcp_connect(msg->conn->pcb.tcp, API_EXPR_REF(msg->msg.bc.ipaddr),
          msg->msg.bc.port, lwip_netconn_do_connected);
        if (msg->err == ERR_OK) {
          u8_t non_blocking = netconn_is_nonblocking(msg->conn);
          msg->conn->state = NETCONN_CONNECT;
          SET_NONBLOCKING_CONNECT(msg->conn, non_blocking);
          if (non_blocking) {
            msg->err = ERR_INPROGRESS;
          } else {
            msg->conn->current_msg = msg;
            /* sys_sem_signal() is called from lwip_netconn_do_connected (or err_tcp()),
               when the connection is established! */
#if LWIP_TCPIP_CORE_LOCKING
            LWIP_ASSERT("state!", msg->conn->state == NETCONN_CONNECT);
            UNLOCK_TCPIP_CORE();
            sys_arch_sem_wait(LWIP_API_MSG_SEM(msg), 0);
            LOCK_TCPIP_CORE();
            LWIP_ASSERT("state!", msg->conn->state != NETCONN_CONNECT);
#endif /* LWIP_TCPIP_CORE_LOCKING */
            return;
          }
        }
      }
      break;
#endif /* LWIP_TCP */
    default:
      LWIP_ERROR("Invalid netconn type", 0, do{ msg->err = ERR_VAL; }while(0));
      break;
    }
  }
  /* For all other protocols, netconn_connect() calls TCPIP_APIMSG(),
     so use TCPIP_APIMSG_ACK() here. */
  TCPIP_APIMSG_ACK(msg);
}

void
lwip_netconn_do_disconnect(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

#if LWIP_UDP
  if (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_UDP) {
    udp_disconnect(msg->conn->pcb.udp);
    msg->err = ERR_OK;
  } else
#endif /* LWIP_UDP */
  {
    msg->err = ERR_VAL;
  }
  TCPIP_APIMSG_ACK(msg);
}

#if LWIP_TCP

void
lwip_netconn_do_listen(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  if (ERR_IS_FATAL(msg->conn->last_err)) {
    msg->err = msg->conn->last_err;
  } else {
    msg->err = ERR_CONN;
    if (msg->conn->pcb.tcp != NULL) {
      if (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_TCP) {
        if (msg->conn->state == NETCONN_NONE) {
          struct tcp_pcb* lpcb;
          if (msg->conn->pcb.tcp->state != CLOSED) {
            /* connection is not closed, cannot listen */
            msg->err = ERR_VAL;
          } else {
            err_t err;
            u8_t backlog;
#if TCP_LISTEN_BACKLOG
            backlog = msg->msg.lb.backlog;
#else  /* TCP_LISTEN_BACKLOG */
            backlog = TCP_DEFAULT_LISTEN_BACKLOG;
#endif /* TCP_LISTEN_BACKLOG */
#if LWIP_IPV4 && LWIP_IPV6
            /* "Socket API like" dual-stack support: If IP to listen to is IP6_ADDR_ANY,
             * and NETCONN_FLAG_IPV6_V6ONLY is NOT set, use IP_ANY_TYPE to listen
             */
            if (ip_addr_cmp(&msg->conn->pcb.ip->local_ip, IP6_ADDR_ANY) &&
                (netconn_get_ipv6only(msg->conn) == 0)) {
              /* change PCB type to IPADDR_TYPE_ANY */
              IP_SET_TYPE_VAL(msg->conn->pcb.tcp->local_ip,  IPADDR_TYPE_ANY);
              IP_SET_TYPE_VAL(msg->conn->pcb.tcp->remote_ip, IPADDR_TYPE_ANY);
            }
#endif /* LWIP_IPV4 && LWIP_IPV6 */

            lpcb = tcp_listen_with_backlog_and_err(msg->conn->pcb.tcp, backlog, &err);

            if (lpcb == NULL) {
              /* in this case, the old pcb is still allocated */
              msg->err = err;
            } else {
              /* delete the recvmbox and allocate the acceptmbox */
              if (sys_mbox_valid(&msg->conn->recvmbox)) {
                sys_mbox_free(&msg->conn->recvmbox);
                sys_mbox_set_invalid(&msg->conn->recvmbox);
              }
              msg->err = ERR_OK;
              if (!sys_mbox_valid(&msg->conn->acceptmbox)) {
                msg->err = sys_mbox_new(&msg->conn->acceptmbox, DEFAULT_ACCEPTMBOX_SIZE);
              }
              if (msg->err == ERR_OK) {
                msg->conn->state = NETCONN_LISTEN;
                msg->conn->pcb.tcp = lpcb;
                tcp_arg(msg->conn->pcb.tcp, msg->conn);
                tcp_accept(msg->conn->pcb.tcp, accept_function);
              } else {
                /* since the old pcb is already deallocated, free lpcb now */
                tcp_close(lpcb);
                msg->conn->pcb.tcp = NULL;
              }
            }
          }
        } else if (msg->conn->state == NETCONN_LISTEN) {
          /* already listening, allow updating of the backlog */
          msg->err = ERR_OK;
          tcp_backlog_set(msg->conn->pcb.tcp, msg->msg.lb.backlog);
        }
      } else {
        msg->err = ERR_ARG;
      }
    }
  }
  TCPIP_APIMSG_ACK(msg);
}
#endif /* LWIP_TCP */

void
lwip_netconn_do_send(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  if (ERR_IS_FATAL(msg->conn->last_err)) {
    msg->err = msg->conn->last_err;
  } else {
    msg->err = ERR_CONN;
    if (msg->conn->pcb.tcp != NULL) {
      switch (NETCONNTYPE_GROUP(msg->conn->type)) {
#if LWIP_RAW
      case NETCONN_RAW:
        if (ip_addr_isany(&msg->msg.b->addr) || IP_IS_ANY_TYPE_VAL(msg->msg.b->addr)) {
          msg->err = raw_send(msg->conn->pcb.raw, msg->msg.b->p);
        } else {
          msg->err = raw_sendto(msg->conn->pcb.raw, msg->msg.b->p, &msg->msg.b->addr);
        }
        break;
#endif
#if LWIP_UDP
      case NETCONN_UDP:
#if LWIP_CHECKSUM_ON_COPY
        if (ip_addr_isany(&msg->msg.b->addr) || IP_IS_ANY_TYPE_VAL(msg->msg.b->addr)) {
          msg->err = udp_send_chksum(msg->conn->pcb.udp, msg->msg.b->p,
            msg->msg.b->flags & NETBUF_FLAG_CHKSUM, msg->msg.b->toport_chksum);
        } else {
          msg->err = udp_sendto_chksum(msg->conn->pcb.udp, msg->msg.b->p,
            &msg->msg.b->addr, msg->msg.b->port,
            msg->msg.b->flags & NETBUF_FLAG_CHKSUM, msg->msg.b->toport_chksum);
        }
#else /* LWIP_CHECKSUM_ON_COPY */
        if (ip_addr_isany_val(msg->msg.b->addr) || IP_IS_ANY_TYPE_VAL(msg->msg.b->addr)) {
          msg->err = udp_send(msg->conn->pcb.udp, msg->msg.b->p);
        } else {
          msg->err = udp_sendto(msg->conn->pcb.udp, msg->msg.b->p, &msg->msg.b->addr, msg->msg.b->port);
        }
#endif /* LWIP_CHECKSUM_ON_COPY */
        break;
#endif /* LWIP_UDP */
      default:
        break;
      }
    }
  }
  TCPIP_APIMSG_ACK(msg);
}

#if LWIP_TCP

void
lwip_netconn_do_recv(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  msg->err = ERR_OK;
  if (msg->conn->pcb.tcp != NULL) {
    if (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_TCP) {
      u32_t remaining = msg->msg.r.len;
      do {
        u16_t recved = (remaining > 0xffff) ? 0xffff : (u16_t)remaining;
        tcp_recved(msg->conn->pcb.tcp, recved);
        remaining -= recved;
      } while (remaining != 0);
    }
  }
  TCPIP_APIMSG_ACK(msg);
}

#if TCP_LISTEN_BACKLOG

void
lwip_netconn_do_accepted(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  msg->err = ERR_OK;
  if (msg->conn->pcb.tcp != NULL) {
    if (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_TCP) {
      tcp_backlog_accepted(msg->conn->pcb.tcp);
    }
  }
  TCPIP_APIMSG_ACK(msg);
}
#endif /* TCP_LISTEN_BACKLOG */

static err_t
lwip_netconn_do_writemore(struct netconn *conn  WRITE_DELAYED_PARAM)
{
  err_t err;
  const void *dataptr;
  u16_t len, available;
  u8_t write_finished = 0;
  size_t diff;
  u8_t dontblock;
  u8_t apiflags;

  LWIP_ASSERT("conn != NULL", conn != NULL);
  LWIP_ASSERT("conn->state == NETCONN_WRITE", (conn->state == NETCONN_WRITE));
  LWIP_ASSERT("conn->current_msg != NULL", conn->current_msg != NULL);
  LWIP_ASSERT("conn->pcb.tcp != NULL", conn->pcb.tcp != NULL);
  LWIP_ASSERT("conn->write_offset < conn->current_msg->msg.w.len",
    conn->write_offset < conn->current_msg->msg.w.len);

  apiflags = conn->current_msg->msg.w.apiflags;
  dontblock = netconn_is_nonblocking(conn) || (apiflags & NETCONN_DONTBLOCK);

#if LWIP_SO_SNDTIMEO
  if ((conn->send_timeout != 0) &&
      ((s32_t)(sys_now() - conn->current_msg->msg.w.time_started) >= conn->send_timeout)) {
    write_finished = 1;
    if (conn->write_offset == 0) {
      /* nothing has been written */
      err = ERR_WOULDBLOCK;
      conn->current_msg->msg.w.len = 0;
    } else {
      /* partial write */
      err = ERR_OK;
      conn->current_msg->msg.w.len = conn->write_offset;
      conn->write_offset = 0;
    }
  } else
#endif /* LWIP_SO_SNDTIMEO */
  {
    dataptr = (const u8_t*)conn->current_msg->msg.w.dataptr + conn->write_offset;
    diff = conn->current_msg->msg.w.len - conn->write_offset;
    if (diff > 0xffffUL) { /* max_u16_t */
      len = 0xffff;
      apiflags |= TCP_WRITE_FLAG_MORE;
    } else {
      len = (u16_t)diff;
    }
    available = tcp_sndbuf(conn->pcb.tcp);
    if (available < len) {
      /* don't try to write more than sendbuf */
      len = available;
      if (dontblock) {
        if (!len) {
          err = ERR_WOULDBLOCK;
          goto err_mem;
        }
      } else {
        apiflags |= TCP_WRITE_FLAG_MORE;
      }
    }
    LWIP_ASSERT("lwip_netconn_do_writemore: invalid length!", ((conn->write_offset + len) <= conn->current_msg->msg.w.len));
    err = tcp_write(conn->pcb.tcp, dataptr, len, apiflags);
    /* if OK or memory error, check available space */
    if ((err == ERR_OK) || (err == ERR_MEM)) {
err_mem:
      if (dontblock && (len < conn->current_msg->msg.w.len)) {
        /* non-blocking write did not write everything: mark the pcb non-writable
           and let poll_tcp check writable space to mark the pcb writable again */
        API_EVENT(conn, NETCONN_EVT_SENDMINUS, len);
        conn->flags |= NETCONN_FLAG_CHECK_WRITESPACE;
      } else if ((tcp_sndbuf(conn->pcb.tcp) <= TCP_SNDLOWAT) ||
                 (tcp_sndqueuelen(conn->pcb.tcp) >= TCP_SNDQUEUELOWAT)) {
        /* The queued byte- or pbuf-count exceeds the configured low-water limit,
           let select mark this pcb as non-writable. */
        API_EVENT(conn, NETCONN_EVT_SENDMINUS, len);
      }
    }

    if (err == ERR_OK) {
      err_t out_err;
      conn->write_offset += len;
      if ((conn->write_offset == conn->current_msg->msg.w.len) || dontblock) {
        /* return sent length */
        conn->current_msg->msg.w.len = conn->write_offset;
        /* everything was written */
        write_finished = 1;
      }
      out_err = tcp_output(conn->pcb.tcp);
      if (ERR_IS_FATAL(out_err) || (out_err == ERR_RTE)) {
        /* If tcp_output fails with fatal error or no route is found,
           don't try writing any more but return the error
           to the application thread. */
        err = out_err;
        write_finished = 1;
        conn->current_msg->msg.w.len = 0;
      }
    } else if (err == ERR_MEM) {
      /* If ERR_MEM, we wait for sent_tcp or poll_tcp to be called.
         For blocking sockets, we do NOT return to the application
         thread, since ERR_MEM is only a temporary error! Non-blocking
         will remain non-writable until sent_tcp/poll_tcp is called */

      /* tcp_write returned ERR_MEM, try tcp_output anyway */
      err_t out_err = tcp_output(conn->pcb.tcp);
      if (ERR_IS_FATAL(out_err) || (out_err == ERR_RTE)) {
        /* If tcp_output fails with fatal error or no route is found,
           don't try writing any more but return the error
           to the application thread. */
        err = out_err;
        write_finished = 1;
        conn->current_msg->msg.w.len = 0;
      } else if (dontblock) {
        /* non-blocking write is done on ERR_MEM */
        err = ERR_WOULDBLOCK;
        write_finished = 1;
        conn->current_msg->msg.w.len = 0;
      }
    } else {
      /* On errors != ERR_MEM, we don't try writing any more but return
         the error to the application thread. */
      write_finished = 1;
      conn->current_msg->msg.w.len = 0;
    }
  }
  if (write_finished) {
    /* everything was written: set back connection state
       and back to application task */
    sys_sem_t* op_completed_sem = LWIP_API_MSG_SEM(conn->current_msg);
    conn->current_msg->err = err;
    conn->current_msg = NULL;
    conn->write_offset = 0;
    conn->state = NETCONN_NONE;
    NETCONN_SET_SAFE_ERR(conn, err);
#if LWIP_TCPIP_CORE_LOCKING
    if (delayed)
#endif
    {
      sys_sem_signal(op_completed_sem);
    }
  }
#if LWIP_TCPIP_CORE_LOCKING
  else {
    return ERR_MEM;
  }
#endif
  return ERR_OK;
}
#endif /* LWIP_TCP */

void
lwip_netconn_do_write(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  if (ERR_IS_FATAL(msg->conn->last_err)) {
    msg->err = msg->conn->last_err;
  } else {
    if (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_TCP) {
#if LWIP_TCP
      if (msg->conn->state != NETCONN_NONE) {
        /* netconn is connecting, closing or in blocking write */
        msg->err = ERR_INPROGRESS;
      } else if (msg->conn->pcb.tcp != NULL) {
        msg->conn->state = NETCONN_WRITE;
        /* set all the variables used by lwip_netconn_do_writemore */
        LWIP_ASSERT("already writing or closing", msg->conn->current_msg == NULL &&
          msg->conn->write_offset == 0);
        LWIP_ASSERT("msg->msg.w.len != 0", msg->msg.w.len != 0);
        msg->conn->current_msg = msg;
        msg->conn->write_offset = 0;
#if LWIP_TCPIP_CORE_LOCKING
        if (lwip_netconn_do_writemore(msg->conn, 0) != ERR_OK) {
          LWIP_ASSERT("state!", msg->conn->state == NETCONN_WRITE);
          UNLOCK_TCPIP_CORE();
          sys_arch_sem_wait(LWIP_API_MSG_SEM(msg), 0);
          LOCK_TCPIP_CORE();
          LWIP_ASSERT("state!", msg->conn->state != NETCONN_WRITE);
        }
#else /* LWIP_TCPIP_CORE_LOCKING */
        lwip_netconn_do_writemore(msg->conn);
#endif /* LWIP_TCPIP_CORE_LOCKING */
        /* for both cases: if lwip_netconn_do_writemore was called, don't ACK the APIMSG
           since lwip_netconn_do_writemore ACKs it! */
        return;
      } else {
        msg->err = ERR_CONN;
      }
#else /* LWIP_TCP */
      msg->err = ERR_VAL;
#endif /* LWIP_TCP */
#if (LWIP_UDP || LWIP_RAW)
    } else {
      msg->err = ERR_VAL;
#endif /* (LWIP_UDP || LWIP_RAW) */
    }
  }
  TCPIP_APIMSG_ACK(msg);
}

void
lwip_netconn_do_getaddr(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  if (msg->conn->pcb.ip != NULL) {
    if (msg->msg.ad.local) {
      ip_addr_copy(API_EXPR_DEREF(msg->msg.ad.ipaddr),
        msg->conn->pcb.ip->local_ip);
    } else {
      ip_addr_copy(API_EXPR_DEREF(msg->msg.ad.ipaddr),
        msg->conn->pcb.ip->remote_ip);
    }

    msg->err = ERR_OK;
    switch (NETCONNTYPE_GROUP(msg->conn->type)) {
#if LWIP_RAW
    case NETCONN_RAW:
      if (msg->msg.ad.local) {
        API_EXPR_DEREF(msg->msg.ad.port) = msg->conn->pcb.raw->protocol;
      } else {
        /* return an error as connecting is only a helper for upper layers */
        msg->err = ERR_CONN;
      }
      break;
#endif /* LWIP_RAW */
#if LWIP_UDP
    case NETCONN_UDP:
      if (msg->msg.ad.local) {
        API_EXPR_DEREF(msg->msg.ad.port) = msg->conn->pcb.udp->local_port;
      } else {
        if ((msg->conn->pcb.udp->flags & UDP_FLAGS_CONNECTED) == 0) {
          msg->err = ERR_CONN;
        } else {
          API_EXPR_DEREF(msg->msg.ad.port) = msg->conn->pcb.udp->remote_port;
        }
      }
      break;
#endif /* LWIP_UDP */
#if LWIP_TCP
    case NETCONN_TCP:
      if ((msg->msg.ad.local == 0) &&
          ((msg->conn->pcb.tcp->state == CLOSED) || (msg->conn->pcb.tcp->state == LISTEN))) {
        /* pcb is not connected and remote name is requested */
        msg->err = ERR_CONN;
      } else {
        API_EXPR_DEREF(msg->msg.ad.port) = (msg->msg.ad.local ? msg->conn->pcb.tcp->local_port : msg->conn->pcb.tcp->remote_port);
      }
      break;
#endif /* LWIP_TCP */
    default:
      LWIP_ASSERT("invalid netconn_type", 0);
      break;
    }
  } else {
    msg->err = ERR_CONN;
  }
  TCPIP_APIMSG_ACK(msg);
}

void
lwip_netconn_do_close(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

#if LWIP_TCP
  enum netconn_state state = msg->conn->state;
  /* First check if this is a TCP netconn and if it is in a correct state
      (LISTEN doesn't support half shutdown) */
  if ((msg->conn->pcb.tcp != NULL) &&
      (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_TCP) &&
      ((msg->msg.sd.shut == NETCONN_SHUT_RDWR) || (state != NETCONN_LISTEN))) {
    /* Check if we are in a connected state */
    if (state == NETCONN_CONNECT) {
      /* TCP connect in progress: cannot shutdown */
      msg->err = ERR_CONN;
    } else if (state == NETCONN_WRITE) {
#if LWIP_NETCONN_FULLDUPLEX
      if (msg->msg.sd.shut & NETCONN_SHUT_WR) {
        /* close requested, abort running write */
        sys_sem_t* write_completed_sem;
        LWIP_ASSERT("msg->conn->current_msg != NULL", msg->conn->current_msg != NULL);
        write_completed_sem = LWIP_API_MSG_SEM(msg->conn->current_msg);
        msg->conn->current_msg->err = ERR_CLSD;
        msg->conn->current_msg = NULL;
        msg->conn->write_offset = 0;
        msg->conn->state = NETCONN_NONE;
        state = NETCONN_NONE;
        NETCONN_SET_SAFE_ERR(msg->conn, ERR_CLSD);
        sys_sem_signal(write_completed_sem);
      } else {
        LWIP_ASSERT("msg->msg.sd.shut == NETCONN_SHUT_RD", msg->msg.sd.shut == NETCONN_SHUT_RD);
        /* In this case, let the write continue and do not interfere with
           conn->current_msg or conn->state! */
        msg->err = tcp_shutdown(msg->conn->pcb.tcp, 1, 0);
      }
    }
    if (state == NETCONN_NONE) {
#else /* LWIP_NETCONN_FULLDUPLEX */
      msg->err = ERR_INPROGRESS;
    } else {
#endif /* LWIP_NETCONN_FULLDUPLEX */
      if (msg->msg.sd.shut & NETCONN_SHUT_RD) {
        /* Drain and delete mboxes */
        netconn_drain(msg->conn);
      }
      LWIP_ASSERT("already writing or closing", msg->conn->current_msg == NULL &&
        msg->conn->write_offset == 0);
      msg->conn->state = NETCONN_CLOSE;
      msg->conn->current_msg = msg;
#if LWIP_TCPIP_CORE_LOCKING
      if (lwip_netconn_do_close_internal(msg->conn, 0) != ERR_OK) {
        LWIP_ASSERT("state!", msg->conn->state == NETCONN_CLOSE);
        UNLOCK_TCPIP_CORE();
        sys_arch_sem_wait(LWIP_API_MSG_SEM(msg), 0);
        LOCK_TCPIP_CORE();
        LWIP_ASSERT("state!", msg->conn->state == NETCONN_NONE);
      }
#else /* LWIP_TCPIP_CORE_LOCKING */
      lwip_netconn_do_close_internal(msg->conn);
#endif /* LWIP_TCPIP_CORE_LOCKING */
      /* for tcp netconns, lwip_netconn_do_close_internal ACKs the message */
      return;
    }
  } else
#endif /* LWIP_TCP */
  {
    msg->err = ERR_CONN;
  }
  TCPIP_APIMSG_ACK(msg);
}

#if LWIP_IGMP || (LWIP_IPV6 && LWIP_IPV6_MLD)

void
lwip_netconn_do_join_leave_group(void *m)
{
  struct api_msg *msg = (struct api_msg*)m;

  if (ERR_IS_FATAL(msg->conn->last_err)) {
    msg->err = msg->conn->last_err;
  } else {
    if (msg->conn->pcb.tcp != NULL) {
      if (NETCONNTYPE_GROUP(msg->conn->type) == NETCONN_UDP) {
#if LWIP_UDP
#if LWIP_IPV6 && LWIP_IPV6_MLD
        if (NETCONNTYPE_ISIPV6(msg->conn->type)) {
          if (msg->msg.jl.join_or_leave == NETCONN_JOIN) {
            msg->err = mld6_joingroup(ip_2_ip6(API_EXPR_REF(msg->msg.jl.netif_addr)),
              ip_2_ip6(API_EXPR_REF(msg->msg.jl.multiaddr)));
          } else {
            msg->err = mld6_leavegroup(ip_2_ip6(API_EXPR_REF(msg->msg.jl.netif_addr)),
              ip_2_ip6(API_EXPR_REF(msg->msg.jl.multiaddr)));
          }
        }
        else
#endif /* LWIP_IPV6 && LWIP_IPV6_MLD */
        {
#if LWIP_IGMP
          if (msg->msg.jl.join_or_leave == NETCONN_JOIN) {
            msg->err = igmp_joingroup(ip_2_ip4(API_EXPR_REF(msg->msg.jl.netif_addr)),
              ip_2_ip4(API_EXPR_REF(msg->msg.jl.multiaddr)));
          } else {
            msg->err = igmp_leavegroup(ip_2_ip4(API_EXPR_REF(msg->msg.jl.netif_addr)),
              ip_2_ip4(API_EXPR_REF(msg->msg.jl.multiaddr)));
          }
#endif /* LWIP_IGMP */
        }
#endif /* LWIP_UDP */
#if (LWIP_TCP || LWIP_RAW)
      } else {
        msg->err = ERR_VAL;
#endif /* (LWIP_TCP || LWIP_RAW) */
      }
    } else {
      msg->err = ERR_CONN;
    }
  }
  TCPIP_APIMSG_ACK(msg);
}
#endif /* LWIP_IGMP || (LWIP_IPV6 && LWIP_IPV6_MLD) */

#if LWIP_DNS

static void
lwip_netconn_do_dns_found(const char *name, const ip_addr_t *ipaddr, void *arg)
{
  struct dns_api_msg *msg = (struct dns_api_msg*)arg;

  /* we trust the internal implementation to be correct :-) */
  LWIP_UNUSED_ARG(name);

  if (ipaddr == NULL) {
    /* timeout or memory error */
    API_EXPR_DEREF(msg->err) = ERR_VAL;
  } else {
    /* address was resolved */
    API_EXPR_DEREF(msg->err) = ERR_OK;
    API_EXPR_DEREF(msg->addr) = *ipaddr;
  }
  /* wake up the application task waiting in netconn_gethostbyname */
  sys_sem_signal(API_EXPR_REF_SEM(msg->sem));
}

void
lwip_netconn_do_gethostbyname(void *arg)
{
  struct dns_api_msg *msg = (struct dns_api_msg*)arg;
  u8_t addrtype =
#if LWIP_IPV4 && LWIP_IPV6
    msg->dns_addrtype;
#else
    LWIP_DNS_ADDRTYPE_DEFAULT;
#endif

  API_EXPR_DEREF(msg->err) = dns_gethostbyname_addrtype(msg->name,
    API_EXPR_REF(msg->addr), lwip_netconn_do_dns_found, msg, addrtype);
  if (API_EXPR_DEREF(msg->err) != ERR_INPROGRESS) {
    /* on error or immediate success, wake up the application
     * task waiting in netconn_gethostbyname */
    sys_sem_signal(API_EXPR_REF_SEM(msg->sem));
  }
}
#endif /* LWIP_DNS */

#endif /* LWIP_NETCONN */