samba-mirror/source3/smbd/oplock.c

/* 
   Unix SMB/Netbios implementation.
   Version 1.9.
   oplock processing
   Copyright (C) Andrew Tridgell 1992-1998
   
   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., 675 Mass Ave, Cambridge, MA 02139, USA.
*/

#include "includes.h"

extern int DEBUGLEVEL;

/* Oplock ipc UDP socket. */
static int oplock_sock = -1;
uint16 global_oplock_port = 0;

/* Current number of oplocks we have outstanding. */
static int32 exclusive_oplocks_open = 0;
static int32 level_II_oplocks_open = 0;
BOOL global_client_failed_oplock_break = False;
BOOL global_oplock_break = False;

extern int smb_read_error;

static struct kernel_oplocks *koplocks;

static BOOL oplock_break(SMB_DEV_T dev, SMB_INO_T inode, struct timeval *tval, BOOL local);

/****************************************************************************
 Get the number of current exclusive oplocks.
****************************************************************************/

int32 get_number_of_exclusive_open_oplocks(void)
{
  return exclusive_oplocks_open;
}


/****************************************************************************
 Read an oplock break message from either the oplock UDP fd or the
 kernel (if kernel oplocks are supported).

 If timeout is zero then *fds contains the file descriptors that
 are ready to be read and acted upon. If timeout is non-zero then
 *fds contains the file descriptors to be selected on for read.
 The timeout is in milliseconds

****************************************************************************/

BOOL receive_local_message(fd_set *fds, char *buffer, int buffer_len, int timeout)
{
  struct sockaddr_in from;
  int fromlen = sizeof(from);
  int32 msg_len = 0;

  smb_read_error = 0;

  if(timeout != 0) {
	  struct timeval to;
	  int selrtn;
	  int maxfd = oplock_sock;

	  if (koplocks && koplocks->notification_fd != -1) {
		  FD_SET(koplocks->notification_fd, fds);
		  maxfd = MAX(maxfd, koplocks->notification_fd);
	  }

	  to.tv_sec = timeout / 1000;
	  to.tv_usec = (timeout % 1000) * 1000;

	  selrtn = sys_select(maxfd+1,fds,&to);

	  if (selrtn == -1 && errno == EINTR) {
		  /* could be a kernel oplock interrupt */
		  if (koplocks && koplocks->msg_waiting(fds)) {
			  return koplocks->receive_message(fds, buffer, buffer_len);
		  }
	  }

	  /* Check if error */
	  if(selrtn == -1) {
		  /* something is wrong. Maybe the socket is dead? */
		  smb_read_error = READ_ERROR;
		  return False;
	  }

	  /* Did we timeout ? */
	  if (selrtn == 0) {
		  smb_read_error = READ_TIMEOUT;
		  return False;
	  }
  }

  if (koplocks && koplocks->msg_waiting(fds)) {
	  return koplocks->receive_message(fds, buffer, buffer_len);
  }

  if (!FD_ISSET(oplock_sock, fds)) return False;

  /*
   * From here down we deal with the smbd <--> smbd
   * oplock break protocol only.
   */

  /*
   * Read a loopback udp message.
   */
  msg_len = recvfrom(oplock_sock, &buffer[OPBRK_CMD_HEADER_LEN],
                     buffer_len - OPBRK_CMD_HEADER_LEN, 0,
                     (struct sockaddr *)&from, &fromlen);

  if(msg_len < 0) {
    DEBUG(0,("receive_local_message. Error in recvfrom. (%s).\n",strerror(errno)));
    return False;
  }

  /* Validate message length. */
  if(msg_len > (buffer_len - OPBRK_CMD_HEADER_LEN)) {
    DEBUG(0,("receive_local_message: invalid msg_len (%d) max can be %d\n",
              msg_len,
              buffer_len  - OPBRK_CMD_HEADER_LEN));
    return False;
  }

  /* Validate message from address (must be localhost). */
  if(from.sin_addr.s_addr != htonl(INADDR_LOOPBACK)) {
    DEBUG(0,("receive_local_message: invalid 'from' address \
(was %lx should be 127.0.0.1\n", (long)from.sin_addr.s_addr));
   return False;
  }

  /* Setup the message header */
  SIVAL(buffer,OPBRK_CMD_LEN_OFFSET,msg_len);
  SSVAL(buffer,OPBRK_CMD_PORT_OFFSET,ntohs(from.sin_port));

  return True;
}

/****************************************************************************
 Attempt to set an oplock on a file. Always succeeds if kernel oplocks are
 disabled (just sets flags). Returns True if oplock set.
****************************************************************************/
BOOL set_file_oplock(files_struct *fsp, int oplock_type)
{
	if (koplocks && !koplocks->set_oplock(fsp, oplock_type))
		return False;

	fsp->oplock_type = oplock_type;
	fsp->sent_oplock_break = NO_BREAK_SENT;
	if (oplock_type == LEVEL_II_OPLOCK)
		level_II_oplocks_open++;
	else
		exclusive_oplocks_open++;

	DEBUG(5,("set_file_oplock: granted oplock on file %s, dev = %x, inode = %.0f, tv_sec = %x, tv_usec = %x\n",
		 fsp->fsp_name, (unsigned int)fsp->dev, (double)fsp->inode,
		 (int)fsp->open_time.tv_sec, (int)fsp->open_time.tv_usec ));

	return True;
}

/****************************************************************************
 Attempt to release an oplock on a file. Decrements oplock count.
****************************************************************************/

void release_file_oplock(files_struct *fsp)
{
	if (koplocks) koplocks->release_oplock(fsp);

	if (fsp->oplock_type == LEVEL_II_OPLOCK)
		level_II_oplocks_open--;
	else
		exclusive_oplocks_open--;
	
	fsp->oplock_type = NO_OPLOCK;
	fsp->sent_oplock_break = NO_BREAK_SENT;
	
	flush_write_cache(fsp, OPLOCK_RELEASE_FLUSH);
}

/****************************************************************************
 Attempt to downgrade an oplock on a file. Doesn't decrement oplock count.
****************************************************************************/

static void downgrade_file_oplock(files_struct *fsp)
{
	if (koplocks) koplocks->release_oplock(fsp);
	fsp->oplock_type = LEVEL_II_OPLOCK;
	exclusive_oplocks_open--;
	level_II_oplocks_open++;
	fsp->sent_oplock_break = NO_BREAK_SENT;
}

/****************************************************************************
 Remove a file oplock. Copes with level II and exclusive.
 Locks then unlocks the share mode lock. Client can decide to go directly
 to none even if a "break-to-level II" was sent.
****************************************************************************/

BOOL remove_oplock(files_struct *fsp, BOOL break_to_none)
{
	SMB_DEV_T dev = fsp->dev;
	SMB_INO_T inode = fsp->inode;
	BOOL ret = True;

	/* Remove the oplock flag from the sharemode. */
	if (lock_share_entry_fsp(fsp) == False) {
		DEBUG(0,("remove_oplock: failed to lock share entry for file %s\n",
			 fsp->fsp_name ));
		ret = False;
	}

	if (fsp->sent_oplock_break == EXCLUSIVE_BREAK_SENT || break_to_none) {
		/*
		 * Deal with a reply when a break-to-none was sent.
		 */

		if(remove_share_oplock(fsp)==False) {
			DEBUG(0,("remove_oplock: failed to remove share oplock for file %s fnum %d, \
dev = %x, inode = %.0f\n", fsp->fsp_name, fsp->fnum, (unsigned int)dev, (double)inode));
			ret = False;
		}

		release_file_oplock(fsp);
	} else {
		/*
		 * Deal with a reply when a break-to-level II was sent.
		 */
		if(downgrade_share_oplock(fsp)==False) {
			DEBUG(0,("remove_oplock: failed to downgrade share oplock for file %s fnum %d, \
dev = %x, inode = %.0f\n", fsp->fsp_name, fsp->fnum, (unsigned int)dev, (double)inode));
			ret = False;
		}
		
		downgrade_file_oplock(fsp);
	}

	unlock_share_entry_fsp(fsp);
	return ret;
}

/****************************************************************************
 Setup the listening set of file descriptors for an oplock break
 message either from the UDP socket or from the kernel. Returns the maximum
 fd used.
****************************************************************************/

int setup_oplock_select_set( fd_set *fds)
{
	int maxfd = oplock_sock;

	if(oplock_sock == -1)
		return 0;

	FD_SET(oplock_sock,fds);

	if (koplocks && koplocks->notification_fd != -1) {
		FD_SET(koplocks->notification_fd, fds);
		maxfd = MAX(maxfd, koplocks->notification_fd);
	}

	return maxfd;
}

/****************************************************************************
 Process an oplock break message - whether it came from the UDP socket
 or from the kernel.
****************************************************************************/
BOOL process_local_message(char *buffer, int buf_size)
{
  int32 msg_len;
  uint16 from_port;
  char *msg_start;
  SMB_DEV_T dev;
  SMB_INO_T inode;
  pid_t remotepid;
  struct timeval tval;
  struct timeval *ptval = NULL;
  uint16 break_cmd_type;

  msg_len = IVAL(buffer,OPBRK_CMD_LEN_OFFSET);
  from_port = SVAL(buffer,OPBRK_CMD_PORT_OFFSET);

  msg_start = &buffer[OPBRK_CMD_HEADER_LEN];

  DEBUG(5,("process_local_message: Got a message of length %d from port (%d)\n", 
            msg_len, from_port));

  /* 
   * Pull the info out of the requesting packet.
   */

  break_cmd_type = SVAL(msg_start,OPBRK_MESSAGE_CMD_OFFSET);

  switch(break_cmd_type)
  {
    case KERNEL_OPLOCK_BREAK_CMD:
	    if (!koplocks) {
		    DEBUG(0,("unexpected kernel oplock break!\n"));
		    break;
	    } 
	    if (!koplocks->parse_message(msg_start, msg_len, &inode, &dev)) {
		    DEBUG(0,("kernel oplock break parse failure!\n"));
	    }
	    break;

    case OPLOCK_BREAK_CMD:
    case LEVEL_II_OPLOCK_BREAK_CMD:

      /* Ensure that the msg length is correct. */
      if(msg_len != OPLOCK_BREAK_MSG_LEN)
      {
        DEBUG(0,("process_local_message: incorrect length for OPLOCK_BREAK_CMD (was %d, should be %d).\n",
              (int)msg_len, (int)OPLOCK_BREAK_MSG_LEN));
        return False;
      }
      {
        long usec;
        time_t sec;

        memcpy((char *)&inode, msg_start+OPLOCK_BREAK_INODE_OFFSET,sizeof(inode));
        memcpy((char *)&dev, msg_start+OPLOCK_BREAK_DEV_OFFSET,sizeof(dev));
        memcpy((char *)&sec, msg_start+OPLOCK_BREAK_SEC_OFFSET,sizeof(sec));
        tval.tv_sec = sec;
        memcpy((char *)&usec, msg_start+OPLOCK_BREAK_USEC_OFFSET, sizeof(usec));
        tval.tv_usec = usec;

        ptval = &tval;

        memcpy((char *)&remotepid, msg_start+OPLOCK_BREAK_PID_OFFSET,sizeof(remotepid));

        DEBUG(5,("process_local_message: (%s) oplock break request from \
pid %d, port %d, dev = %x, inode = %.0f\n",
             (break_cmd_type == OPLOCK_BREAK_CMD) ? "exclusive" : "level II",
             (int)remotepid, from_port, (unsigned int)dev, (double)inode));
      }
      break;

    /* 
     * Keep this as a debug case - eventually we can remove it.
     */
    case 0x8001:
      DEBUG(0,("process_local_message: Received unsolicited break \
reply - dumping info.\n"));

      if(msg_len != OPLOCK_BREAK_MSG_LEN)
      {
        DEBUG(0,("process_local_message: ubr: incorrect length for reply \
(was %d, should be %d).\n", (int)msg_len, (int)OPLOCK_BREAK_MSG_LEN));
        return False;
      }

      {
        memcpy((char *)&inode, msg_start+OPLOCK_BREAK_INODE_OFFSET,sizeof(inode));
        memcpy((char *)&remotepid, msg_start+OPLOCK_BREAK_PID_OFFSET,sizeof(remotepid));
        memcpy((char *)&dev, msg_start+OPLOCK_BREAK_DEV_OFFSET,sizeof(dev));

        DEBUG(0,("process_local_message: unsolicited oplock break reply from \
pid %d, port %d, dev = %x, inode = %.0f\n", (int)remotepid, from_port, (unsigned int)dev, (double)inode));

       }
       return False;

    default:
      DEBUG(0,("process_local_message: unknown UDP message command code (%x) - ignoring.\n",
                (unsigned int)SVAL(msg_start,0)));
      return False;
  }

  /*
   * Now actually process the break request.
   */

  if((exclusive_oplocks_open + level_II_oplocks_open) != 0)
  {
    if (oplock_break(dev, inode, ptval, False) == False)
    {
      DEBUG(0,("process_local_message: oplock break failed.\n"));
      return False;
    }
  }
  else
  {
    /*
     * If we have no record of any currently open oplocks,
     * it's not an error, as a close command may have
     * just been issued on the file that was oplocked.
     * Just log a message and return success in this case.
     */
    DEBUG(3,("process_local_message: oplock break requested with no outstanding \
oplocks. Returning success.\n"));
  }

  /* 
   * Do the appropriate reply - none in the kernel or level II case.
   */

  if(SVAL(msg_start,OPBRK_MESSAGE_CMD_OFFSET) == OPLOCK_BREAK_CMD)
  {
    struct sockaddr_in toaddr;

    /* Send the message back after OR'ing in the 'REPLY' bit. */
    SSVAL(msg_start,OPBRK_MESSAGE_CMD_OFFSET,OPLOCK_BREAK_CMD | CMD_REPLY);

    memset((char *)&toaddr,'\0',sizeof(toaddr));
    toaddr.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
    toaddr.sin_port = htons(from_port);
    toaddr.sin_family = AF_INET;

    if(sendto( oplock_sock, msg_start, OPLOCK_BREAK_MSG_LEN, 0,
            (struct sockaddr *)&toaddr, sizeof(toaddr)) < 0) 
    {
      DEBUG(0,("process_local_message: sendto process %d failed. Errno was %s\n",
                (int)remotepid, strerror(errno)));
      return False;
    }

    DEBUG(5,("process_local_message: oplock break reply sent to \
pid %d, port %d, for file dev = %x, inode = %.0f\n",
          (int)remotepid, from_port, (unsigned int)dev, (double)inode));
  }

  return True;
}

/****************************************************************************
 Set up an oplock break message.
****************************************************************************/

static void prepare_break_message(char *outbuf, files_struct *fsp, BOOL level2)
{
	memset(outbuf,'\0',smb_size);
	set_message(outbuf,8,0,True);

	SCVAL(outbuf,smb_com,SMBlockingX);
	SSVAL(outbuf,smb_tid,fsp->conn->cnum);
	SSVAL(outbuf,smb_pid,0xFFFF);
	SSVAL(outbuf,smb_uid,0);
	SSVAL(outbuf,smb_mid,0xFFFF);
	SCVAL(outbuf,smb_vwv0,0xFF);
	SSVAL(outbuf,smb_vwv2,fsp->fnum);
	SCVAL(outbuf,smb_vwv3,LOCKING_ANDX_OPLOCK_RELEASE);
	SCVAL(outbuf,smb_vwv3+1,level2 ? OPLOCKLEVEL_II : OPLOCKLEVEL_NONE);
}

/****************************************************************************
 Function to do the waiting before sending a local break.
****************************************************************************/

static void wait_before_sending_break(BOOL local_request)
{
  extern struct timeval smb_last_time;

  if(local_request) {
    struct timeval cur_tv;
    long wait_left = (long)lp_oplock_break_wait_time();

	if (wait_left == 0)
		return;

    GetTimeOfDay(&cur_tv);

    wait_left -= ((cur_tv.tv_sec - smb_last_time.tv_sec)*1000) +
                ((cur_tv.tv_usec - smb_last_time.tv_usec)/1000);

    if(wait_left > 0) {
      wait_left = MIN(wait_left, 1000);
      sys_usleep(wait_left * 1000);
    }
  }
}

/****************************************************************************
 Ensure that we have a valid oplock.
****************************************************************************/
static files_struct *initial_break_processing(SMB_DEV_T dev, SMB_INO_T inode, struct timeval *tval)
{
  files_struct *fsp = NULL;

  if( DEBUGLVL( 3 ) )
  {
    dbgtext( "initial_break_processing: called for dev = %x, inode = %.0f tv_sec = %x, tv_usec = %x.\n",
      (unsigned int)dev, (double)inode, tval ? (int)tval->tv_sec : 0,
      tval ? (int)tval->tv_usec : 0);
    dbgtext( "Current oplocks_open (exclusive = %d, levelII = %d)\n",
              exclusive_oplocks_open, level_II_oplocks_open );
  }

  /* We need to search the file open table for the
     entry containing this dev and inode, and ensure
     we have an oplock on it. */
  fsp = file_find_dit(dev, inode, tval);

  if(fsp == NULL)
  {
    /* The file could have been closed in the meantime - return success. */
    if( DEBUGLVL( 3 ) )
    {
      dbgtext( "initial_break_processing: cannot find open file with " );
      dbgtext( "dev = %x, inode = %.0f ", (unsigned int)dev, (double)inode);
      dbgtext( "allowing break to succeed.\n" );
    }
    return NULL;
  }

  /* Ensure we have an oplock on the file */

  /* There is a potential race condition in that an oplock could
     have been broken due to another udp request, and yet there are
     still oplock break messages being sent in the udp message
     queue for this file. So return true if we don't have an oplock,
     as we may have just freed it.
   */

  if(fsp->oplock_type == NO_OPLOCK)
  {
    if( DEBUGLVL( 3 ) )
    {
      dbgtext( "initial_break_processing: file %s ", fsp->fsp_name );
      dbgtext( "(dev = %x, inode = %.0f) has no oplock.\n", (unsigned int)dev, (double)inode );
      dbgtext( "Allowing break to succeed regardless.\n" );
    }
    return NULL;
  }

  return fsp;
}

/****************************************************************************
 Process a level II oplock break directly.
****************************************************************************/

BOOL oplock_break_level2(files_struct *fsp, BOOL local_request, int token)
{
  extern uint32 global_client_caps;
  char outbuf[128];
  BOOL got_lock = False;
  SMB_DEV_T dev = fsp->dev;
  SMB_INO_T inode = fsp->inode;

  /*
   * We can have a level II oplock even if the client is not
   * level II oplock aware. In this case just remove the
   * flags and don't send the break-to-none message to
   * the client.
   */

  if (global_client_caps & CAP_LEVEL_II_OPLOCKS) {
    /*
     * If we are sending an oplock break due to an SMB sent
     * by our own client we ensure that we wait at leat
     * lp_oplock_break_wait_time() milliseconds before sending
     * the packet. Sending the packet sooner can break Win9x
     * and has reported to cause problems on NT. JRA.
     */

    wait_before_sending_break(local_request);

    /* Prepare the SMBlockingX message. */

    prepare_break_message( outbuf, fsp, False);
    send_smb(smbd_server_fd(), outbuf);
  }

  /*
   * Now we must update the shared memory structure to tell
   * everyone else we no longer have a level II oplock on 
   * this open file. If local_request is true then token is
   * the existing lock on the shared memory area.
   */

  if(!local_request && lock_share_entry_fsp(fsp) == False) {
      DEBUG(0,("oplock_break_level2: unable to lock share entry for file %s\n", fsp->fsp_name ));
  } else {
    got_lock = True;
  }

  if(remove_share_oplock(fsp)==False) {
    DEBUG(0,("oplock_break_level2: unable to remove level II oplock for file %s\n", fsp->fsp_name ));
  }

  if (!local_request && got_lock)
    unlock_share_entry_fsp(fsp);

  fsp->oplock_type = NO_OPLOCK;
  level_II_oplocks_open--;

  if(level_II_oplocks_open < 0)
  {
    DEBUG(0,("oplock_break_level2: level_II_oplocks_open < 0 (%d). PANIC ERROR\n",
              level_II_oplocks_open));
    abort();
  }

  if( DEBUGLVL( 3 ) )
  {
    dbgtext( "oplock_break_level2: returning success for " );
    dbgtext( "dev = %x, inode = %.0f\n", (unsigned int)dev, (double)inode );
    dbgtext( "Current level II oplocks_open = %d\n", level_II_oplocks_open );
  }

  return True;
}

/****************************************************************************
 Process an oplock break directly.
****************************************************************************/

static BOOL oplock_break(SMB_DEV_T dev, SMB_INO_T inode, struct timeval *tval, BOOL local_request)
{
  extern uint32 global_client_caps;
  extern struct current_user current_user;
  char *inbuf = NULL;
  char *outbuf = NULL;
  files_struct *fsp = NULL;
  time_t start_time;
  BOOL shutdown_server = False;
  BOOL oplock_timeout = False;
  connection_struct *saved_conn;
  int saved_vuid;
  pstring saved_dir; 
  int timeout = (OPLOCK_BREAK_TIMEOUT * 1000);
  pstring file_name;
  BOOL using_levelII;

  if((fsp = initial_break_processing(dev, inode, tval)) == NULL)
    return True;

  /*
   * Deal with a level II oplock going break to none separately.
   */

  if (LEVEL_II_OPLOCK_TYPE(fsp->oplock_type))
    return oplock_break_level2(fsp, local_request, -1);

  /* Mark the oplock break as sent - we don't want to send twice! */
  if (fsp->sent_oplock_break)
  {
    if( DEBUGLVL( 0 ) )
    {
      dbgtext( "oplock_break: ERROR: oplock_break already sent for " );
      dbgtext( "file %s ", fsp->fsp_name);
      dbgtext( "(dev = %x, inode = %.0f)\n", (unsigned int)dev, (double)inode );
    }

    /* We have to fail the open here as we cannot send another oplock break on
       this file whilst we are awaiting a response from the client - neither
       can we allow another open to succeed while we are waiting for the
       client.
     */
    return False;
  }

  if(global_oplock_break) {
    DEBUG(0,("ABORT : ABORT : recursion in oplock_break !!!!!\n"));
    abort();
  }

  /* Now comes the horrid part. We must send an oplock break to the client,
     and then process incoming messages until we get a close or oplock release.
     At this point we know we need a new inbuf/outbuf buffer pair.
     We cannot use these staticaly as we may recurse into here due to
     messages crossing on the wire.
   */

  if((inbuf = (char *)malloc(BUFFER_SIZE + SAFETY_MARGIN))==NULL)
  {
    DEBUG(0,("oplock_break: malloc fail for input buffer.\n"));
    return False;
  }

  if((outbuf = (char *)malloc(BUFFER_SIZE + SAFETY_MARGIN))==NULL)
  {
    DEBUG(0,("oplock_break: malloc fail for output buffer.\n"));
    free(inbuf);
    inbuf = NULL;
    return False;
  }

  /*
   * If we are sending an oplock break due to an SMB sent
   * by our own client we ensure that we wait at leat
   * lp_oplock_break_wait_time() milliseconds before sending
   * the packet. Sending the packet sooner can break Win9x
   * and has reported to cause problems on NT. JRA.
   */

  wait_before_sending_break(local_request);

  /* Prepare the SMBlockingX message. */

  if ((global_client_caps & CAP_LEVEL_II_OPLOCKS) && 
      !koplocks && /* NOTE: we force levelII off for kernel oplocks -
                      this will change when it is supported */
      lp_level2_oplocks(SNUM(fsp->conn))) {
	  using_levelII = True;
  } else {
	  using_levelII = False;
  }

  prepare_break_message( outbuf, fsp, using_levelII);
  /* Remember if we just sent a break to level II on this file. */
  fsp->sent_oplock_break = using_levelII?
	  LEVEL_II_BREAK_SENT:EXCLUSIVE_BREAK_SENT;

  send_smb(smbd_server_fd(), outbuf);

  /* We need this in case a readraw crosses on the wire. */
  global_oplock_break = True;
 
  /* Process incoming messages. */

  /* JRA - If we don't get a break from the client in OPLOCK_BREAK_TIMEOUT
     seconds we should just die.... */

  start_time = time(NULL);

  /*
   * Save the information we need to re-become the
   * user, then unbecome the user whilst we're doing this.
   */
  saved_conn = fsp->conn;
  saved_vuid = current_user.vuid;
  vfs_GetWd(saved_conn,saved_dir);
  unbecome_user();
  /* Save the chain fnum. */
  file_chain_save();

  /*
   * From Charles Hoch <hoch@exemplary.com>. If the break processing
   * code closes the file (as it often does), then the fsp pointer here
   * points to free()'d memory. We *must* revalidate fsp each time
   * around the loop.
   */

  pstrcpy(file_name, fsp->fsp_name);

  while((fsp = initial_break_processing(dev, inode, tval)) &&
        OPEN_FSP(fsp) && EXCLUSIVE_OPLOCK_TYPE(fsp->oplock_type))
  {
    if(receive_smb(smbd_server_fd(),inbuf, timeout) == False)
    {
      /*
       * Die if we got an error.
       */

      if (smb_read_error == READ_EOF) {
        DEBUG( 0, ( "oplock_break: end of file from client\n" ) );
        shutdown_server = True;
      } else if (smb_read_error == READ_ERROR) {
        DEBUG( 0, ("oplock_break: receive_smb error (%s)\n", strerror(errno)) );
        shutdown_server = True;
      } else if (smb_read_error == READ_TIMEOUT) {
        DEBUG( 0, ( "oplock_break: receive_smb timed out after %d seconds.\n",
                     OPLOCK_BREAK_TIMEOUT ) );
        oplock_timeout = True;
      }

      DEBUGADD( 0, ( "oplock_break failed for file %s ", file_name ) );
      DEBUGADD( 0, ( "(dev = %x, inode = %.0f).\n", (unsigned int)dev, (double)inode));

      break;
    }

    /*
     * There are certain SMB requests that we shouldn't allow
     * to recurse. opens, renames and deletes are the obvious
     * ones. This is handled in the switch_message() function.
     * If global_oplock_break is set they will push the packet onto
     * the pending smb queue and return -1 (no reply).
     * JRA.
     */

    process_smb(inbuf, outbuf);

    /*
     * Die if we go over the time limit.
     */

    if((time(NULL) - start_time) > OPLOCK_BREAK_TIMEOUT)
    {
      if( DEBUGLVL( 0 ) )
      {
        dbgtext( "oplock_break: no break received from client " );
        dbgtext( "within %d seconds.\n", OPLOCK_BREAK_TIMEOUT );
        dbgtext( "oplock_break failed for file %s ", fsp->fsp_name );
        dbgtext( "(dev = %x, inode = %.0f).\n", (unsigned int)dev, (double)inode );
      }
      oplock_timeout = True;
      break;
    }
  }

  /*
   * Go back to being the user who requested the oplock
   * break.
   */
  if(!become_user(saved_conn, saved_vuid))
  {
    DEBUG( 0, ( "oplock_break: unable to re-become user!" ) );
    DEBUGADD( 0, ( "Shutting down server\n" ) );
    close(oplock_sock);
    exit_server("unable to re-become user");
  }
  /* Including the directory. */
  vfs_ChDir(saved_conn,saved_dir);

  /* Restore the chain fnum. */
  file_chain_restore();

  /* Free the buffers we've been using to recurse. */
  free(inbuf);
  free(outbuf);

  /* We need this in case a readraw crossed on the wire. */
  if(global_oplock_break)
    global_oplock_break = False;

  /*
   * If the client timed out then clear the oplock (or go to level II)
   * and continue. This seems to be what NT does and is better than dropping
   * the connection.
   */

  if(oplock_timeout && (fsp = initial_break_processing(dev, inode, tval)) &&
        OPEN_FSP(fsp) && EXCLUSIVE_OPLOCK_TYPE(fsp->oplock_type))
  {
    DEBUG(0,("oplock_break: client failure in oplock break in file %s\n", fsp->fsp_name));
    remove_oplock(fsp,True);
    global_client_failed_oplock_break = True; /* Never grant this client an oplock again. */
  }

  /*
   * If the client had an error we must die.
   */

  if(shutdown_server)
  {
    DEBUG( 0, ( "oplock_break: client failure in break - " ) );
    DEBUGADD( 0, ( "shutting down this smbd.\n" ) );
    close(oplock_sock);
    exit_server("oplock break failure");
  }

  /* Santity check - remove this later. JRA */
  if(exclusive_oplocks_open < 0)
  {
    DEBUG(0,("oplock_break: exclusive_oplocks_open < 0 (%d). PANIC ERROR\n",
              exclusive_oplocks_open));
    abort();
  }

  if( DEBUGLVL( 3 ) )
  {
    dbgtext( "oplock_break: returning success for " );
    dbgtext( "dev = %x, inode = %.0f\n", (unsigned int)dev, (double)inode );
    dbgtext( "Current exclusive_oplocks_open = %d\n", exclusive_oplocks_open );
  }

  return True;
}

/****************************************************************************
Send an oplock break message to another smbd process. If the oplock is held 
by the local smbd then call the oplock break function directly.
****************************************************************************/

BOOL request_oplock_break(share_mode_entry *share_entry, 
                          SMB_DEV_T dev, SMB_INO_T inode)
{
  char op_break_msg[OPLOCK_BREAK_MSG_LEN];
  struct sockaddr_in addr_out;
  pid_t pid = sys_getpid();
  time_t start_time;
  int time_left;
  long usec;
  time_t sec;

  if(pid == share_entry->pid)
  {
    /* We are breaking our own oplock, make sure it's us. */
    if(share_entry->op_port != global_oplock_port)
    {
      DEBUG(0,("request_oplock_break: corrupt share mode entry - pid = %d, port = %d \
should be %d\n", (int)pid, share_entry->op_port, global_oplock_port));
      return False;
    }

    DEBUG(5,("request_oplock_break: breaking our own oplock\n"));

    /* Call oplock break direct. */
    return oplock_break(dev, inode, &share_entry->time, True);
  }

  /* We need to send a OPLOCK_BREAK_CMD message to the
     port in the share mode entry. */

  if (LEVEL_II_OPLOCK_TYPE(share_entry->op_type)) {
    SSVAL(op_break_msg,OPBRK_MESSAGE_CMD_OFFSET,LEVEL_II_OPLOCK_BREAK_CMD);
  } else {
    SSVAL(op_break_msg,OPBRK_MESSAGE_CMD_OFFSET,OPLOCK_BREAK_CMD);
  }
  memcpy(op_break_msg+OPLOCK_BREAK_PID_OFFSET,(char *)&pid,sizeof(pid));
  sec = (time_t)share_entry->time.tv_sec;
  memcpy(op_break_msg+OPLOCK_BREAK_SEC_OFFSET,(char *)&sec,sizeof(sec));
  usec = (long)share_entry->time.tv_usec;
  memcpy(op_break_msg+OPLOCK_BREAK_USEC_OFFSET,(char *)&usec,sizeof(usec));
  memcpy(op_break_msg+OPLOCK_BREAK_DEV_OFFSET,(char *)&dev,sizeof(dev));
  memcpy(op_break_msg+OPLOCK_BREAK_INODE_OFFSET,(char *)&inode,sizeof(inode));

  /* set the address and port */
  memset((char *)&addr_out,'\0',sizeof(addr_out));
  addr_out.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
  addr_out.sin_port = htons( share_entry->op_port );
  addr_out.sin_family = AF_INET;
   
  if( DEBUGLVL( 3 ) )
  {
    dbgtext( "request_oplock_break: sending a oplock break message to " );
    dbgtext( "pid %d on port %d ", (int)share_entry->pid, share_entry->op_port );
    dbgtext( "for dev = %x, inode = %.0f, tv_sec = %x, tv_usec = %x\n",
            (unsigned int)dev, (double)inode, (int)share_entry->time.tv_sec,
            (int)share_entry->time.tv_usec );

  }

  if(sendto(oplock_sock,op_break_msg,OPLOCK_BREAK_MSG_LEN,0,
         (struct sockaddr *)&addr_out,sizeof(addr_out)) < 0)
  {
    if( DEBUGLVL( 0 ) )
    {
      dbgtext( "request_oplock_break: failed when sending a oplock " );
      dbgtext( "break message to pid %d ", (int)share_entry->pid );
      dbgtext( "on port %d ", share_entry->op_port );
      dbgtext( "for dev = %x, inode = %.0f, tv_sec = %x, tv_usec = %x\n",
          (unsigned int)dev, (double)inode, (int)share_entry->time.tv_sec,
          (int)share_entry->time.tv_usec );
      dbgtext( "Error was %s\n", strerror(errno) );
    }
    return False;
  }

  /*
   * If we just sent a message to a level II oplock share entry then
   * we are done and may return.
   */

  if (LEVEL_II_OPLOCK_TYPE(share_entry->op_type)) {
    DEBUG(3,("request_oplock_break: sent break message to level II entry.\n"));
    return True;
  }

  /*
   * Now we must await the oplock broken message coming back
   * from the target smbd process. Timeout if it fails to
   * return in (OPLOCK_BREAK_TIMEOUT + OPLOCK_BREAK_TIMEOUT_FUDGEFACTOR) seconds.
   * While we get messages that aren't ours, loop.
   */

  start_time = time(NULL);
  time_left = OPLOCK_BREAK_TIMEOUT+OPLOCK_BREAK_TIMEOUT_FUDGEFACTOR;

  while(time_left >= 0)
  {
    char op_break_reply[OPBRK_CMD_HEADER_LEN+OPLOCK_BREAK_MSG_LEN];
    uint16 reply_from_port;
    char *reply_msg_start;
    fd_set fds;

    FD_ZERO(&fds);
    FD_SET(oplock_sock,&fds);

    if (koplocks && koplocks->notification_fd != -1) {
	    FD_SET(koplocks->notification_fd, &fds);
    }

    if(receive_local_message(&fds, op_break_reply, sizeof(op_break_reply),
               time_left ? time_left * 1000 : 1) == False)
    {
      if(smb_read_error == READ_TIMEOUT)
      {
        if( DEBUGLVL( 0 ) )
        {
          dbgtext( "request_oplock_break: no response received to oplock " );
          dbgtext( "break request to pid %d ", (int)share_entry->pid );
          dbgtext( "on port %d ", share_entry->op_port );
          dbgtext( "for dev = %x, inode = %.0f\n", (unsigned int)dev, (double)inode );
          dbgtext( "for dev = %x, inode = %.0f, tv_sec = %x, tv_usec = %x\n",
             (unsigned int)dev, (double)inode, (int)share_entry->time.tv_sec,
             (int)share_entry->time.tv_usec );
        }

        /*
         * This is a hack to make handling of failing clients more robust.
         * If a oplock break response message is not received in the timeout
         * period we may assume that the smbd servicing that client holding
         * the oplock has died and the client changes were lost anyway, so
         * we should continue to try and open the file.
         */
        break;
      }
      else
        if( DEBUGLVL( 0 ) )
        {
          dbgtext( "request_oplock_break: error in response received " );
          dbgtext( "to oplock break request to pid %d ", (int)share_entry->pid );
          dbgtext( "on port %d ", share_entry->op_port );
          dbgtext( "for dev = %x, inode = %.0f, tv_sec = %x, tv_usec = %x\n",
               (unsigned int)dev, (double)inode, (int)share_entry->time.tv_sec,
               (int)share_entry->time.tv_usec );
          dbgtext( "Error was (%s).\n", strerror(errno) );
        }
      return False;
    }

    reply_from_port = SVAL(op_break_reply,OPBRK_CMD_PORT_OFFSET);

    reply_msg_start = &op_break_reply[OPBRK_CMD_HEADER_LEN];


    /*
     * Test to see if this is the reply we are awaiting.
     */
    if((SVAL(reply_msg_start,OPBRK_MESSAGE_CMD_OFFSET) & CMD_REPLY) &&
       ((SVAL(reply_msg_start,OPBRK_MESSAGE_CMD_OFFSET) & ~CMD_REPLY) == OPLOCK_BREAK_CMD) &&
       (reply_from_port == share_entry->op_port) && 
       (memcmp(&reply_msg_start[OPLOCK_BREAK_PID_OFFSET], 
               &op_break_msg[OPLOCK_BREAK_PID_OFFSET],
               OPLOCK_BREAK_MSG_LEN - OPLOCK_BREAK_PID_OFFSET) == 0))
    {
      /*
       * This is the reply we've been waiting for.
       */
      break;
    }
    else
    {
      /*
       * This is another message - a break request.
       * Note that both kernel oplock break requests
       * and UDP inter-smbd oplock break requests will
       * be processed here.
       *
       * Process it to prevent potential deadlock.
       * Note that the code in switch_message() prevents
       * us from recursing into here as any SMB requests
       * we might process that would cause another oplock
       * break request to be made will be queued.
       * JRA.
       */

      process_local_message(op_break_reply, sizeof(op_break_reply));
    }

    time_left -= (time(NULL) - start_time);
  }

  DEBUG(3,("request_oplock_break: broke oplock.\n"));

  return True;
}

/****************************************************************************
  Attempt to break an oplock on a file (if oplocked).
  Returns True if the file was closed as a result of
  the oplock break, False otherwise.
  Used as a last ditch attempt to free a space in the 
  file table when we have run out.
****************************************************************************/
BOOL attempt_close_oplocked_file(files_struct *fsp)
{

  DEBUG(5,("attempt_close_oplocked_file: checking file %s.\n", fsp->fsp_name));

  if (EXCLUSIVE_OPLOCK_TYPE(fsp->oplock_type) && !fsp->sent_oplock_break && (fsp->fd != -1)) {
	  /* Try and break the oplock. */
	  if (oplock_break(fsp->dev, fsp->inode, &fsp->open_time, True)) {
		  if(file_find_fsp(fsp) == NULL) /* Did the oplock break close the file ? */
			  return True;
	  }
  }

  return False;
}

/****************************************************************************
 This function is called on any file modification or lock request. If a file
 is level 2 oplocked then it must tell all other level 2 holders to break to none.
****************************************************************************/

void release_level_2_oplocks_on_change(files_struct *fsp)
{
	share_mode_entry *share_list = NULL;
	pid_t pid = sys_getpid();
	int token = -1;
	int num_share_modes = 0;
	int i;

	/*
	 * If this file is level II oplocked then we need
	 * to grab the shared memory lock and inform all
	 * other files with a level II lock that they need
	 * to flush their read caches. We keep the lock over
	 * the shared memory area whilst doing this.
	 */

	if (!LEVEL_II_OPLOCK_TYPE(fsp->oplock_type))
		return;

	if (lock_share_entry_fsp(fsp) == False) {
		DEBUG(0,("release_level_2_oplocks_on_change: failed to lock share mode entry for file %s.\n", fsp->fsp_name ));
	}

	num_share_modes = get_share_modes(fsp->conn, fsp->dev, fsp->inode, &share_list);

	for(i = 0; i < num_share_modes; i++) {
		share_mode_entry *share_entry = &share_list[i];

		/*
		 * As there could have been multiple writes waiting at the lock_share_entry
		 * gate we may not be the first to enter. Hence the state of the op_types
		 * in the share mode entries may be partly NO_OPLOCK and partly LEVEL_II
		 * oplock. It will do no harm to re-send break messages to those smbd's
		 * that are still waiting their turn to remove their LEVEL_II state, and
		 * also no harm to ignore existing NO_OPLOCK states. JRA.
		 */

		if (share_entry->op_type == NO_OPLOCK)
			continue;

		/* Paranoia .... */
		if (EXCLUSIVE_OPLOCK_TYPE(share_entry->op_type)) {
			DEBUG(0,("release_level_2_oplocks_on_change: PANIC. share mode entry %d is an exlusive oplock !\n", i ));
			unlock_share_entry(fsp->conn, fsp->dev, fsp->inode);
			abort();
		}

		/*
		 * Check if this is a file we have open (including the
		 * file we've been called to do write_file on. If so
		 * then break it directly without releasing the lock.
		 */

		if (pid == share_entry->pid) {
			files_struct *new_fsp = file_find_dit(fsp->dev, fsp->inode, &share_entry->time);

			/* Paranoia check... */
			if(new_fsp == NULL) {
				DEBUG(0,("release_level_2_oplocks_on_change: PANIC. share mode entry %d is not a local file !\n", i ));
				unlock_share_entry(fsp->conn, fsp->dev, fsp->inode);
				abort();
			}

			oplock_break_level2(new_fsp, True, token);

		} else {

			/*
			 * This is a remote file and so we send an asynchronous
			 * message.
			 */

			request_oplock_break(share_entry, fsp->dev, fsp->inode);
		}
	}

	free((char *)share_list);
	unlock_share_entry_fsp(fsp);

	/* Paranoia check... */
	if (LEVEL_II_OPLOCK_TYPE(fsp->oplock_type)) {
		DEBUG(0,("release_level_2_oplocks_on_change: PANIC. File %s still has a level II oplock.\n", fsp->fsp_name));
		abort();
	}
}

/****************************************************************************
setup oplocks for this process
****************************************************************************/
BOOL init_oplocks(void)
{
	struct sockaddr_in sock_name;
	int len = sizeof(sock_name);

	DEBUG(3,("open_oplock_ipc: opening loopback UDP socket.\n"));

	/* Open a lookback UDP socket on a random port. */
	oplock_sock = open_socket_in(SOCK_DGRAM, 0, 0, htonl(INADDR_LOOPBACK),False);
	if (oplock_sock == -1) {
		DEBUG(0,("open_oplock_ipc: Failed to get local UDP socket for \
address %lx. Error was %s\n", (long)htonl(INADDR_LOOPBACK), strerror(errno)));
		global_oplock_port = 0;
		return(False);
	}

	/* Find out the transient UDP port we have been allocated. */
	if(getsockname(oplock_sock, (struct sockaddr *)&sock_name, &len)<0) {
		DEBUG(0,("open_oplock_ipc: Failed to get local UDP port. Error was %s\n",
			 strerror(errno)));
		close(oplock_sock);
		oplock_sock = -1;
		global_oplock_port = 0;
		return False;
	}
	global_oplock_port = ntohs(sock_name.sin_port);

	if (lp_kernel_oplocks()) {
#if HAVE_KERNEL_OPLOCKS_IRIX
		koplocks = irix_init_kernel_oplocks();
#elif HAVE_KERNEL_OPLOCKS_LINUX
		koplocks = linux_init_kernel_oplocks();
#endif
	}

	DEBUG(3,("open_oplock ipc: pid = %d, global_oplock_port = %u\n", 
		 (int)sys_getpid(), global_oplock_port));

	return True;
}