/*

 * Copyright (C) 2013-2014 OpenSIPS Solutions

 * Copyright (C) 2001-2003 FhG Fokus

 *

 * This file is part of opensips, a free SIP server.

 *

 * opensips 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

 *

 * opensips 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301  USA

 *

 * History:

 * --------

 * 2015-02-05 imported from former tpc_read.c (bogdan)

 */

#include "../pt_load.h"

#include "../ipc.h"

#include "../timer.h"

#include "../reactor.h"

#include "../async.h"

#include "../cfg_reload.h"

#include "tcp_conn.h"

#include "tcp_passfd.h"

#include "net_tcp_report.h"

#include "trans.h"

#include "net_tcp_dbg.h"

/*!< the FD currently used by the process to communicate with TCP MAIN*/

static int _my_fd_to_tcp_main = -1;

/*!< list of tcp connections handled by this process */

static struct tcp_connection* tcp_conn_lst=0;

static int _tcp_done_reading_marker = 0;

static int tcpmain_sock=-1;

extern int unix_tcp_sock;

extern struct struct_hist_list *con_hist;

#define tcpconn_release_error(_conn, _writer, _reason) \

  do { \

    tcp_trigger_report( _conn, TCP_REPORT_CLOSE, _reason);\

    tcpconn_release( _conn, CONN_ERROR_TCPW, _writer, 1/*as TCP worker*/);\

  }while(0)

static void tcpconn_release(struct tcp_connection* c, long state, int writer,

                              int as_tcp_worker)

{

  long response[2];

  LM_DBG(" releasing con %p, state %ld, fd=%d, id=%d\n",

      c, state, c->fd, c->id);

  LM_DBG(" extra_data %p\n", c->extra_data);

  /* release req & signal the parent */

  if (!writer)

    c->proc_id = -1;

  /* errno==EINTR, EWOULDBLOCK a.s.o todo */

  response[0]=(long)c;

  response[1]=state;

  if (send_all( as_tcp_worker?tcpmain_sock:unix_tcp_sock, response,

  sizeof(response))<=0)

    LM_ERR("send_all failed state=%ld con=%p\n", state, c);

}

/* wrapper around internal tcpconn_release() - to be called by functions which

 * used tcp_conn_get(), in order to release the connection;

 * It does the unref and pushes back (if needed) some update to TCP main;

 * right now, it used only from the xxx_send() functions

 */

void tcp_conn_release(struct tcp_connection* c, int pending_data)

{

  if (c->state==S_CONN_BAD) {

    /* do more or less nothing, let the TCP READER owning the conn

     * to trash it based on the S_CONN_BAD marker */

    c->lifetime=0;

    /* but be sure we unref the conn */

    tcpconn_put(c);

    return;

  }

  if (pending_data) {

    tcpconn_release(c, ASYNC_WRITE_GENW, 1, 0 /*not TCP, but GEN worker*/);

    return;

  }

  tcpconn_put(c);

}

int tcp_done_reading(struct tcp_connection* con)

{

  if (_tcp_done_reading_marker==0) {

    reactor_del_all( con->fd, -1, IO_FD_CLOSING );

    tcpconn_check_del(con);

    tcpconn_listrm(tcp_conn_lst, con, c_next, c_prev);

    if (con->fd!=-1) { close(con->fd); con->fd = -1; }

    sh_log(con->hist, TCP_SEND2MAIN,

      "parallel read OK - releasing, ref: %d", con->refcnt);

    tcpconn_release(con, CONN_RELEASE, 0, 1 /*as TCP proc*/);

    _tcp_done_reading_marker = 1;

  }

  return 0;

}

/*! \brief  releases expired connections and cleans up bad ones (state<0) */

static void tcp_receive_timeout(void)

{

  struct tcp_connection* con;

  struct tcp_connection* next;

  unsigned int ticks;

  ticks=get_ticks();

  for (con=tcp_conn_lst; con; con=next) {

    next=con->c_next; /* safe for removing */

    if (con->state<0){   /* kill bad connections */

      /* S_CONN_BAD or S_CONN_ERROR, remove it */

      /* fd will be closed in tcpconn_release */

      LM_ERR("TCP_DBG - conn %p / %u found as bad, relasing back "

        "to main\n", con, con->id);

      reactor_del_reader(con->fd, -1/*idx*/, IO_FD_CLOSING/*io_flags*/ );

      tcpconn_check_del(con);

      tcpconn_listrm(tcp_conn_lst, con, c_next, c_prev);

      con->proc_id = -1;

      con->state=S_CONN_BAD;

      if (con->fd!=-1) { close(con->fd); con->fd = -1; }

      sh_log(con->hist, TCP_SEND2MAIN, "state: %d, att: %d, ref: %d",

             con->state, con->msg_attempts, con->refcnt);

      tcpconn_release_error(con, 0, "Unknown reason");

      continue;

    }

    /* pass back to Main connections that are inactive (expired) or

     * if we are in termination mode (this worker is doing graceful 

     * shutdown) and there is no pending data on the conn. */

    if (con->timeout<=ticks ||

    (_termination_in_progress && !con->msg_attempts) ){

      LM_DBG("%p expired - (%d, %d) lt=%d\n",

          con, con->timeout, ticks,con->lifetime);

      /* fd will be closed in tcpconn_release */

      reactor_del_reader(con->fd, -1/*idx*/, IO_FD_CLOSING/*io_flags*/ );

      tcpconn_check_del(con);

      tcpconn_listrm(tcp_conn_lst, con, c_next, c_prev);

      /* connection is going to main */

      con->proc_id = -1;

      if (con->fd!=-1) { close(con->fd); con->fd = -1; }

      sh_log(con->hist, TCP_SEND2MAIN, "timeout: %d, att: %d",

             con->timeout, con->msg_attempts);

      if (con->msg_attempts)

        tcpconn_release_error(con, 0, "Read timeout with"

          "incomplete SIP message");

      else

        tcpconn_release(con, CONN_RELEASE, 0,  1 /*as TCP proc*/);

    }

  }

}

/*! \brief

 *  handle io routine, based on the fd_map type

 * (it will be called from reactor_main_loop )

 * params:  fm  - pointer to a fd hash entry

 *          idx - index in the fd_array (or -1 if not known)

 * return: -1 on error, or when we are not interested any more on reads

 *            from this fd (e.g.: we are closing it )

 *          0 on EAGAIN or when by some other way it is known that no more

 *            io events are queued on the fd (the receive buffer is empty).

 *            Usefull to detect when there are no more io events queued for

 *            sigio_rt, epoll_et, kqueue.

 *         >0 on successful read from the fd (when there might be more io

 *            queued -- the receive buffer might still be non-empty)

 */

inline static int handle_io(struct fd_map* fm, int idx,int event_type)

{

  int ret=0;

  int n;

  struct tcp_connection* con;

  int s,rw;

  long resp;

  long response[2];

  pt_become_active();

  pre_run_handle_script_reload(fm->app_flags);

  switch(fm->type){

    case F_TIMER_JOB:

      handle_timer_job();

      break;

    case F_SCRIPT_ASYNC:

      async_script_resume_f( fm->fd, fm->data,

        (event_type==IO_WATCH_TIMEOUT)?1:0 );

      break;

    case F_FD_ASYNC:

      async_fd_resume( fm->fd, fm->data);

      break;

    case F_LAUNCH_ASYNC:

      async_launch_resume( fm->fd, fm->data);

      break;

    case F_IPC:

      ipc_handle_job(fm->fd);

      break;

    case F_TCPMAIN:

again:

      ret=n=receive_fd(fm->fd, response, sizeof(response), &s, 0);

      if (n<0){

        if (errno == EWOULDBLOCK || errno == EAGAIN){

          ret=0;

          break;

        }else if (errno == EINTR) goto again;

        else{

          LM_CRIT("read_fd: %s \n", strerror(errno));

            abort(); /* big error*/

        }

      }

      if (n==0){

        LM_WARN("0 bytes read\n");

        break;

      }

      con = (struct tcp_connection *)response[0];

      rw = (int)response[1];

      if (con==0){

          LM_CRIT("null pointer\n");

          break;

      }

      if (s==-1) {

        LM_BUG("read_fd:no fd read for conn %p, rw %d\n", con, rw);

        /* FIXME? */

        goto error;

      }

      if (!(con->flags & F_CONN_INIT)) {

        if (protos[con->type].net.stream.conn.init &&

            protos[con->type].net.stream.conn.init(con) < 0) {

          LM_ERR("failed to do proto %d specific init for conn %p\n",

              con->type, con);

          goto con_error;

        }

        con->flags |= F_CONN_INIT;

      }

      LM_DBG("We have received conn %p with rw %d on fd %d\n",con,rw,s);

      if (rw & IO_WATCH_READ) {

        if (tcpconn_list_find(con, tcp_conn_lst)) {

          LM_CRIT("duplicate connection received: %p, id %d, fd %d, "

                  "refcnt %d state %d (n=%d)\n", con, con->id,

                  con->fd, con->refcnt, con->state, n);

          tcpconn_release_error(con, 0, "Internal duplicate");

          break; /* try to recover */

        }

        /* 0 attempts so far for this SIP MSG */

        con->msg_attempts = 0;

        /* must be before reactor_add, as the add might catch some

         * already existing events => might call handle_io and

         * handle_io might decide to del. the new connection =>

         * must be in the list */

        tcpconn_check_add(con);

        tcpconn_listadd(tcp_conn_lst, con, c_next, c_prev);

        /* pending event on a connection -> prevent premature expiry */

        tcp_conn_reset_lifetime(con);

        con->timeout = con->lifetime;

        if (reactor_add_reader( s, F_TCPCONN, RCT_PRIO_NET, con )<0) {

          LM_CRIT("failed to add new socket to the fd list\n");

          tcpconn_check_del(con);

          tcpconn_listrm(tcp_conn_lst, con, c_next, c_prev);

          goto con_error;

        }

        sh_log(con->hist, TCP_ADD_READER, "add reader fd %d, ref: %d",

               s, con->refcnt);

        /* mark that the connection is currently in our process

        future writes to this con won't have to acquire FD */

        con->proc_id = process_no;

        /* save FD which is valid in context of this TCP worker */

        con->fd=s;

      } else if (rw & IO_WATCH_WRITE) {

        LM_DBG("Received con for async write %p ref = %d\n",

          con, con->refcnt);

        lock_get(&con->write_lock);

        resp = protos[con->type].net.stream.write( con, s );

        lock_release(&con->write_lock);

        if (resp<0) {

          ret=-1; /* some error occurred */

          con->state=S_CONN_BAD;

          sh_log(con->hist, TCP_SEND2MAIN,

            "handle write, err, state: %d, att: %d",

            con->state, con->msg_attempts);

          tcpconn_release_error(con, 1,"Write error");

          close(s); /* we always close the socket received for writing */

          break;

        } else if (resp==1) {

          sh_log(con->hist, TCP_SEND2MAIN,

            "handle write, async, state: %d, att: %d",

            con->state, con->msg_attempts);

          tcpconn_release(con, ASYNC_WRITE_TCPW, 1,

            1 /*as TCP proc*/);

        } else {

          sh_log(con->hist, TCP_SEND2MAIN,

            "handle write, ok, state: %d, att: %d",

            con->state, con->msg_attempts);

          tcpconn_release(con, CONN_RELEASE_WRITE, 1,

            1/*as TCP proc*/);

        }

        ret = 0;

        /* we always close the socket received for writing */

        close(s);

      }

      break;

    case F_TCPCONN:

      if (event_type & IO_WATCH_READ) {

        con=(struct tcp_connection*)fm->data;

        _tcp_done_reading_marker = 0;

        resp = protos[con->type].net.stream.read( con, &ret );

        if (resp<0) {

          ret=-1; /* some error occurred */

          con->state=S_CONN_BAD;

          reactor_del_all( con->fd, idx, IO_FD_CLOSING );

          tcpconn_check_del(con);

          tcpconn_listrm(tcp_conn_lst, con, c_next, c_prev);

          con->proc_id = -1;

          if (con->fd!=-1) { close(con->fd); con->fd = -1; }

          sh_log(con->hist, TCP_SEND2MAIN,

            "handle read, err, resp: %d, att: %d",

            resp, con->msg_attempts);

          tcpconn_release_error(con, 0, "Read error");

        } else if (resp == 1) {

          /* the connection is already released */

          break;

        } else if (con->state==S_CONN_EOF) {

          reactor_del_all( con->fd, idx, IO_FD_CLOSING );

          tcpconn_check_del(con);

          tcpconn_listrm(tcp_conn_lst, con, c_next, c_prev);

          con->proc_id = -1;

          if (con->fd!=-1) { close(con->fd); con->fd = -1; }

          tcp_trigger_report( con, TCP_REPORT_CLOSE,

            "EOF received");

          sh_log(con->hist, TCP_SEND2MAIN,

            "handle read, EOF, resp: %d, att: %d",

            resp, con->msg_attempts);

          tcpconn_release(con, CONN_EOF, 0, 1 /*as TCP proc*/);

        } else {

          if (con->profile.parallel_read)

            /* return the connection if not already */

            tcp_done_reading( con );

          break;

        }

      }

      break;

    case F_NONE:

      LM_CRIT("empty fd map %p: "

            "{%d, %d, %p}\n", fm,

            fm->fd, fm->type, fm->data);

      goto error;

    default:

      LM_CRIT("unknown fd type %d\n", fm->type);

      goto error;

  }

  if (_termination_in_progress==1) {

    /* force (again) passing back all the active conns */

    tcp_receive_timeout();

    /* check if anything is still left */

    if (reactor_is_empty()) {

      LM_WARN("reactor got empty while termination in progress\n");

      ipc_handle_all_pending_jobs(IPC_FD_READ_SELF);

      if (reactor_is_empty())

        dynamic_process_final_exit();

    }

  }

  post_run_handle_script_reload();

  pt_become_idle();

  return ret;

con_error:

  con->state=S_CONN_BAD;

  tcpconn_release_error(con, 0, "Internal error");

  pt_become_idle();

  return ret;

error:

  pt_become_idle();

  return -1;

}

int tcp_worker_proc_reactor_init( int unix_sock)

{

  /* init reactor for TCP worker */

  tcpmain_sock=unix_sock; /* init com. socket */

  if ( init_worker_reactor( "TCP_worker", RCT_PRIO_MAX)<0 ) {

    goto error;

  }

  /* start watching for the timer jobs */

  if (reactor_add_reader( timer_fd_out, F_TIMER_JOB, RCT_PRIO_TIMER,NULL)<0){

    LM_CRIT("failed to add timer pipe_out to reactor\n");

    goto error;

  }

  /* init: start watching for the IPC jobs */

  if (reactor_add_reader(IPC_FD_READ_SELF, F_IPC, RCT_PRIO_ASYNC, NULL)<0){

    LM_CRIT("failed to add IPC pipe to reactor\n");

    goto error;

  }

  /* init: start watching for IPC "dispatched" jobs */

  if (reactor_add_reader(IPC_FD_READ_SHARED, F_IPC, RCT_PRIO_ASYNC, NULL)<0){

    LM_CRIT("failed to add IPC shared pipe to reactor\n");

    return -1;

  }

  /* add the unix socket */

  if (reactor_add_reader( tcpmain_sock, F_TCPMAIN, RCT_PRIO_PROC, NULL)<0) {

    LM_CRIT("failed to add socket to the fd list\n");

    goto error;

  }

  _my_fd_to_tcp_main = tcpmain_sock;

  return 0;

error:

  destroy_worker_reactor();

  return -1;

}

void tcp_worker_proc_loop(void)

{

  /* main loop */

  reactor_main_loop( TCP_CHILD_SELECT_TIMEOUT, error, tcp_receive_timeout());

  LM_CRIT("exiting...");

  exit(-1);

error:

  destroy_worker_reactor();

}

void tcp_terminate_worker(void)

{

  /*remove from reactor all the shared fds, so we stop reading from them */

  /*remove timer jobs pipe */

  reactor_del_reader( timer_fd_out, -1, 0);

  /*remove IPC dispatcher pipe */

  reactor_del_reader( IPC_FD_READ_SHARED, -1, 0);

  /*remove private IPC pipe */

  reactor_del_reader( IPC_FD_READ_SELF, -1, 0);

  /*remove unix sock to TCP main */

  reactor_del_reader( _my_fd_to_tcp_main, -1, 0);

  _termination_in_progress = 1;

  /* let's drain the private IPC */

  ipc_handle_all_pending_jobs(IPC_FD_READ_SELF);

  /* force passing back all the active conns */

  tcp_receive_timeout();

  /* what is left now is the reactor are async fd's, so we need to 

   * wait to complete all of them */

  if (reactor_is_empty())

    dynamic_process_final_exit();

  /* the exit will be triggered by the reactor, when empty */

  LM_INFO("reactor not empty, waiting for pending async/conns\n");

}