/*

   * Copyright (c) 2000-2005 Silicon Graphics, Inc.
   * All Rights Reserved.

   *

   * 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.
   *

   * This program is distributed in the hope that it would 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 the Free Software Foundation,
   * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA

   */

  #include "xfs.h"

  #include "xfs_fs.h"

  #include "xfs_types.h"

  #include "xfs_bit.h"

  #include "xfs_log.h"

  #include "xfs_inum.h"

  #include "xfs_trans.h"

  #include "xfs_sb.h"
  #include "xfs_ag.h"

  #include "xfs_mount.h"
  #include "xfs_error.h"
  #include "xfs_log_priv.h"
  #include "xfs_buf_item.h"

  #include "xfs_bmap_btree.h"

  #include "xfs_alloc_btree.h"

  #include "xfs_ialloc_btree.h"

  #include "xfs_log_recover.h"

  #include "xfs_trans_priv.h"

  #include "xfs_dinode.h"
  #include "xfs_inode.h"
  #include "xfs_rw.h"

  #include "xfs_trace.h"

  kmem_zone_t	*xfs_log_ticket_zone;

  /* Local miscellaneous function prototypes */

  STATIC int	 xlog_commit_record(struct log *log, struct xlog_ticket *ticket,

  				    xlog_in_core_t **, xfs_lsn_t *);
  STATIC xlog_t *  xlog_alloc_log(xfs_mount_t	*mp,
  				xfs_buftarg_t	*log_target,
  				xfs_daddr_t	blk_offset,
  				int		num_bblks);

  STATIC int	 xlog_space_left(struct log *log, atomic64_t *head);

  STATIC int	 xlog_sync(xlog_t *log, xlog_in_core_t *iclog);

  STATIC void	 xlog_dealloc_log(xlog_t *log);

  
  /* local state machine functions */
  STATIC void xlog_state_done_syncing(xlog_in_core_t *iclog, int);
  STATIC void xlog_state_do_callback(xlog_t *log,int aborted, xlog_in_core_t *iclog);
  STATIC int  xlog_state_get_iclog_space(xlog_t		*log,
  				       int		len,
  				       xlog_in_core_t	**iclog,
  				       xlog_ticket_t	*ticket,
  				       int		*continued_write,
  				       int		*logoffsetp);

  STATIC int  xlog_state_release_iclog(xlog_t		*log,
  				     xlog_in_core_t	*iclog);
  STATIC void xlog_state_switch_iclogs(xlog_t		*log,
  				     xlog_in_core_t *iclog,
  				     int		eventual_size);

  STATIC void xlog_state_want_sync(xlog_t	*log, xlog_in_core_t *iclog);
  
  /* local functions to manipulate grant head */
  STATIC int  xlog_grant_log_space(xlog_t		*log,
  				 xlog_ticket_t	*xtic);

  STATIC void xlog_grant_push_ail(struct log	*log,

  				int		need_bytes);
  STATIC void xlog_regrant_reserve_log_space(xlog_t	 *log,
  					   xlog_ticket_t *ticket);
  STATIC int xlog_regrant_write_log_space(xlog_t		*log,
  					 xlog_ticket_t  *ticket);
  STATIC void xlog_ungrant_log_space(xlog_t	 *log,
  				   xlog_ticket_t *ticket);

  #if defined(DEBUG)

  STATIC void	xlog_verify_dest_ptr(xlog_t *log, char *ptr);

  STATIC void	xlog_verify_grant_tail(struct log *log);

  STATIC void	xlog_verify_iclog(xlog_t *log, xlog_in_core_t *iclog,
  				  int count, boolean_t syncing);
  STATIC void	xlog_verify_tail_lsn(xlog_t *log, xlog_in_core_t *iclog,
  				     xfs_lsn_t tail_lsn);
  #else
  #define xlog_verify_dest_ptr(a,b)

  #define xlog_verify_grant_tail(a)

  #define xlog_verify_iclog(a,b,c,d)
  #define xlog_verify_tail_lsn(a,b,c)
  #endif

  STATIC int	xlog_iclogs_empty(xlog_t *log);

  static void

  xlog_grant_sub_space(
  	struct log	*log,

  	atomic64_t	*head,

  	int		bytes)

  {

  	int64_t	head_val = atomic64_read(head);
  	int64_t new, old;

  	do {
  		int	cycle, space;

  		xlog_crack_grant_head_val(head_val, &cycle, &space);

  		space -= bytes;
  		if (space < 0) {
  			space += log->l_logsize;
  			cycle--;
  		}
  
  		old = head_val;
  		new = xlog_assign_grant_head_val(cycle, space);
  		head_val = atomic64_cmpxchg(head, old, new);
  	} while (head_val != old);

  }
  
  static void

  xlog_grant_add_space(
  	struct log	*log,

  	atomic64_t	*head,

  	int		bytes)

  {

  	int64_t	head_val = atomic64_read(head);
  	int64_t new, old;

  	do {
  		int		tmp;
  		int		cycle, space;

  		xlog_crack_grant_head_val(head_val, &cycle, &space);

  		tmp = log->l_logsize - space;
  		if (tmp > bytes)
  			space += bytes;
  		else {
  			space = bytes - tmp;
  			cycle++;
  		}
  
  		old = head_val;
  		new = xlog_assign_grant_head_val(cycle, space);
  		head_val = atomic64_cmpxchg(head, old, new);
  	} while (head_val != old);

  }

  STATIC bool
  xlog_reserveq_wake(
  	struct log		*log,
  	int			*free_bytes)
  {
  	struct xlog_ticket	*tic;
  	int			need_bytes;
  
  	list_for_each_entry(tic, &log->l_reserveq, t_queue) {
  		if (tic->t_flags & XLOG_TIC_PERM_RESERV)
  			need_bytes = tic->t_unit_res * tic->t_cnt;
  		else
  			need_bytes = tic->t_unit_res;
  
  		if (*free_bytes < need_bytes)
  			return false;
  		*free_bytes -= need_bytes;
  
  		trace_xfs_log_grant_wake_up(log, tic);
  		wake_up(&tic->t_wait);
  	}
  
  	return true;
  }
  
  STATIC bool
  xlog_writeq_wake(
  	struct log		*log,
  	int			*free_bytes)
  {
  	struct xlog_ticket	*tic;
  	int			need_bytes;
  
  	list_for_each_entry(tic, &log->l_writeq, t_queue) {
  		ASSERT(tic->t_flags & XLOG_TIC_PERM_RESERV);
  
  		need_bytes = tic->t_unit_res;
  
  		if (*free_bytes < need_bytes)
  			return false;
  		*free_bytes -= need_bytes;
  
  		trace_xfs_log_regrant_write_wake_up(log, tic);
  		wake_up(&tic->t_wait);
  	}
  
  	return true;
  }
  
  STATIC int
  xlog_reserveq_wait(
  	struct log		*log,
  	struct xlog_ticket	*tic,
  	int			need_bytes)
  {
  	list_add_tail(&tic->t_queue, &log->l_reserveq);
  
  	do {
  		if (XLOG_FORCED_SHUTDOWN(log))
  			goto shutdown;
  		xlog_grant_push_ail(log, need_bytes);
  
  		XFS_STATS_INC(xs_sleep_logspace);
  		trace_xfs_log_grant_sleep(log, tic);
  
  		xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock);
  		trace_xfs_log_grant_wake(log, tic);
  
  		spin_lock(&log->l_grant_reserve_lock);
  		if (XLOG_FORCED_SHUTDOWN(log))
  			goto shutdown;
  	} while (xlog_space_left(log, &log->l_grant_reserve_head) < need_bytes);
  
  	list_del_init(&tic->t_queue);
  	return 0;
  shutdown:
  	list_del_init(&tic->t_queue);
  	return XFS_ERROR(EIO);
  }
  
  STATIC int
  xlog_writeq_wait(
  	struct log		*log,
  	struct xlog_ticket	*tic,
  	int			need_bytes)
  {
  	list_add_tail(&tic->t_queue, &log->l_writeq);
  
  	do {
  		if (XLOG_FORCED_SHUTDOWN(log))
  			goto shutdown;
  		xlog_grant_push_ail(log, need_bytes);
  
  		XFS_STATS_INC(xs_sleep_logspace);
  		trace_xfs_log_regrant_write_sleep(log, tic);
  
  		xlog_wait(&tic->t_wait, &log->l_grant_write_lock);
  		trace_xfs_log_regrant_write_wake(log, tic);
  
  		spin_lock(&log->l_grant_write_lock);
  		if (XLOG_FORCED_SHUTDOWN(log))
  			goto shutdown;
  	} while (xlog_space_left(log, &log->l_grant_write_head) < need_bytes);
  
  	list_del_init(&tic->t_queue);
  	return 0;
  shutdown:
  	list_del_init(&tic->t_queue);
  	return XFS_ERROR(EIO);
  }

  static void
  xlog_tic_reset_res(xlog_ticket_t *tic)
  {
  	tic->t_res_num = 0;
  	tic->t_res_arr_sum = 0;
  	tic->t_res_num_ophdrs = 0;
  }
  
  static void
  xlog_tic_add_region(xlog_ticket_t *tic, uint len, uint type)
  {
  	if (tic->t_res_num == XLOG_TIC_LEN_MAX) {
  		/* add to overflow and start again */
  		tic->t_res_o_flow += tic->t_res_arr_sum;
  		tic->t_res_num = 0;
  		tic->t_res_arr_sum = 0;
  	}
  
  	tic->t_res_arr[tic->t_res_num].r_len = len;
  	tic->t_res_arr[tic->t_res_num].r_type = type;
  	tic->t_res_arr_sum += len;
  	tic->t_res_num++;
  }

  /*
   * NOTES:
   *
   *	1. currblock field gets updated at startup and after in-core logs
   *		marked as with WANT_SYNC.
   */
  
  /*
   * This routine is called when a user of a log manager ticket is done with
   * the reservation.  If the ticket was ever used, then a commit record for
   * the associated transaction is written out as a log operation header with
   * no data.  The flag XLOG_TIC_INITED is set when the first write occurs with
   * a given ticket.  If the ticket was one with a permanent reservation, then
   * a few operations are done differently.  Permanent reservation tickets by
   * default don't release the reservation.  They just commit the current
   * transaction with the belief that the reservation is still needed.  A flag
   * must be passed in before permanent reservations are actually released.
   * When these type of tickets are not released, they need to be set into
   * the inited state again.  By doing this, a start record will be written
   * out when the next write occurs.
   */
  xfs_lsn_t

  xfs_log_done(
  	struct xfs_mount	*mp,
  	struct xlog_ticket	*ticket,
  	struct xlog_in_core	**iclog,
  	uint			flags)

  {

  	struct log		*log = mp->m_log;
  	xfs_lsn_t		lsn = 0;

  	if (XLOG_FORCED_SHUTDOWN(log) ||
  	    /*
  	     * If nothing was ever written, don't write out commit record.
  	     * If we get an error, just continue and give back the log ticket.
  	     */
  	    (((ticket->t_flags & XLOG_TIC_INITED) == 0) &&

  	     (xlog_commit_record(log, ticket, iclog, &lsn)))) {

  		lsn = (xfs_lsn_t) -1;
  		if (ticket->t_flags & XLOG_TIC_PERM_RESERV) {
  			flags |= XFS_LOG_REL_PERM_RESERV;
  		}
  	}
  
  
  	if ((ticket->t_flags & XLOG_TIC_PERM_RESERV) == 0 ||
  	    (flags & XFS_LOG_REL_PERM_RESERV)) {

  		trace_xfs_log_done_nonperm(log, ticket);

  		/*

  		 * Release ticket if not permanent reservation or a specific

  		 * request has been made to release a permanent reservation.
  		 */
  		xlog_ungrant_log_space(log, ticket);

  		xfs_log_ticket_put(ticket);

  	} else {

  		trace_xfs_log_done_perm(log, ticket);

  		xlog_regrant_reserve_log_space(log, ticket);

  		/* If this ticket was a permanent reservation and we aren't
  		 * trying to release it, reset the inited flags; so next time
  		 * we write, a start record will be written out.
  		 */

  		ticket->t_flags |= XLOG_TIC_INITED;

  	}

  
  	return lsn;

  }

  /*
   * Attaches a new iclog I/O completion callback routine during
   * transaction commit.  If the log is in error state, a non-zero
   * return code is handed back and the caller is responsible for
   * executing the callback at an appropriate time.
   */
  int

  xfs_log_notify(
  	struct xfs_mount	*mp,
  	struct xlog_in_core	*iclog,
  	xfs_log_callback_t	*cb)

  {

  	int	abortflg;

  	spin_lock(&iclog->ic_callback_lock);

  	abortflg = (iclog->ic_state & XLOG_STATE_IOERROR);
  	if (!abortflg) {
  		ASSERT_ALWAYS((iclog->ic_state == XLOG_STATE_ACTIVE) ||
  			      (iclog->ic_state == XLOG_STATE_WANT_SYNC));
  		cb->cb_next = NULL;
  		*(iclog->ic_callback_tail) = cb;
  		iclog->ic_callback_tail = &(cb->cb_next);
  	}

  	spin_unlock(&iclog->ic_callback_lock);

  	return abortflg;

  }

  
  int

  xfs_log_release_iclog(
  	struct xfs_mount	*mp,
  	struct xlog_in_core	*iclog)

  {

  	if (xlog_state_release_iclog(mp->m_log, iclog)) {

  		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);

  		return EIO;

  	}
  
  	return 0;
  }
  
  /*
   *  1. Reserve an amount of on-disk log space and return a ticket corresponding
   *	to the reservation.
   *  2. Potentially, push buffers at tail of log to disk.
   *
   * Each reservation is going to reserve extra space for a log record header.
   * When writes happen to the on-disk log, we don't subtract the length of the
   * log record header from any reservation.  By wasting space in each
   * reservation, we prevent over allocation problems.
   */
  int

  xfs_log_reserve(
  	struct xfs_mount	*mp,
  	int		 	unit_bytes,
  	int		 	cnt,
  	struct xlog_ticket	**ticket,
  	__uint8_t	 	client,
  	uint		 	flags,
  	uint		 	t_type)

  {

  	struct log		*log = mp->m_log;
  	struct xlog_ticket	*internal_ticket;
  	int			retval = 0;

  	ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);

  
  	if (XLOG_FORCED_SHUTDOWN(log))
  		return XFS_ERROR(EIO);
  
  	XFS_STATS_INC(xs_try_logspace);

  	if (*ticket != NULL) {
  		ASSERT(flags & XFS_LOG_PERM_RESERV);

  		internal_ticket = *ticket;

  		/*
  		 * this is a new transaction on the ticket, so we need to
  		 * change the transaction ID so that the next transaction has a
  		 * different TID in the log. Just add one to the existing tid
  		 * so that we can see chains of rolling transactions in the log
  		 * easily.
  		 */
  		internal_ticket->t_tid++;

  		trace_xfs_log_reserve(log, internal_ticket);

  		xlog_grant_push_ail(log, internal_ticket->t_unit_res);

  		retval = xlog_regrant_write_log_space(log, internal_ticket);
  	} else {
  		/* may sleep if need to allocate more tickets */

  		internal_ticket = xlog_ticket_alloc(log, unit_bytes, cnt,

  						  client, flags,
  						  KM_SLEEP|KM_MAYFAIL);

  		if (!internal_ticket)
  			return XFS_ERROR(ENOMEM);

  		internal_ticket->t_trans_type = t_type;

  		*ticket = internal_ticket;

  
  		trace_xfs_log_reserve(log, internal_ticket);

  		xlog_grant_push_ail(log,

  				    (internal_ticket->t_unit_res *
  				     internal_ticket->t_cnt));
  		retval = xlog_grant_log_space(log, internal_ticket);
  	}

  	if (unlikely(retval)) {
  		/*
  		 * If we are failing, make sure the ticket doesn't have any
  		 * current reservations.  We don't want to add this back
  		 * when the ticket/ transaction gets cancelled.
  		 */
  		internal_ticket->t_curr_res = 0;
  		/* ungrant will give back unit_res * t_cnt. */
  		internal_ticket->t_cnt = 0;
  	}

  	return retval;

  }

  
  
  /*
   * Mount a log filesystem
   *
   * mp		- ubiquitous xfs mount point structure
   * log_target	- buftarg of on-disk log device
   * blk_offset	- Start block # where block size is 512 bytes (BBSIZE)
   * num_bblocks	- Number of BBSIZE blocks in on-disk log
   *
   * Return error or zero.
   */
  int

  xfs_log_mount(
  	xfs_mount_t	*mp,
  	xfs_buftarg_t	*log_target,
  	xfs_daddr_t	blk_offset,
  	int		num_bblks)

  {

  	int		error;

  	if (!(mp->m_flags & XFS_MOUNT_NORECOVERY))

  		xfs_notice(mp, "Mounting Filesystem");

  	else {

  		xfs_notice(mp,
  "Mounting filesystem in no-recovery mode.  Filesystem will be inconsistent.");

  		ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);

  	}
  
  	mp->m_log = xlog_alloc_log(mp, log_target, blk_offset, num_bblks);

  	if (IS_ERR(mp->m_log)) {
  		error = -PTR_ERR(mp->m_log);

  		goto out;
  	}

  	/*

  	 * Initialize the AIL now we have a log.
  	 */

  	error = xfs_trans_ail_init(mp);
  	if (error) {

  		xfs_warn(mp, "AIL initialisation failed: error %d", error);

  		goto out_free_log;

  	}

  	mp->m_log->l_ailp = mp->m_ail;

  
  	/*

  	 * skip log recovery on a norecovery mount.  pretend it all
  	 * just worked.
  	 */
  	if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) {

  		int	readonly = (mp->m_flags & XFS_MOUNT_RDONLY);

  
  		if (readonly)

  			mp->m_flags &= ~XFS_MOUNT_RDONLY;

  		error = xlog_recover(mp->m_log);

  
  		if (readonly)

  			mp->m_flags |= XFS_MOUNT_RDONLY;

  		if (error) {

  			xfs_warn(mp, "log mount/recovery failed: error %d",
  				error);

  			goto out_destroy_ail;

  		}
  	}
  
  	/* Normal transactions can now occur */
  	mp->m_log->l_flags &= ~XLOG_ACTIVE_RECOVERY;

  	/*
  	 * Now the log has been fully initialised and we know were our
  	 * space grant counters are, we can initialise the permanent ticket
  	 * needed for delayed logging to work.
  	 */
  	xlog_cil_init_post_recovery(mp->m_log);

  	return 0;

  
  out_destroy_ail:
  	xfs_trans_ail_destroy(mp);
  out_free_log:
  	xlog_dealloc_log(mp->m_log);

  out:

  	return error;

  }

  
  /*
   * Finish the recovery of the file system.  This is separate from
   * the xfs_log_mount() call, because it depends on the code in
   * xfs_mountfs() to read in the root and real-time bitmap inodes
   * between calling xfs_log_mount() and here.
   *
   * mp		- ubiquitous xfs mount point structure
   */
  int

  xfs_log_mount_finish(xfs_mount_t *mp)

  {
  	int	error;
  
  	if (!(mp->m_flags & XFS_MOUNT_NORECOVERY))

  		error = xlog_recover_finish(mp->m_log);

  	else {
  		error = 0;

  		ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);

  	}
  
  	return error;
  }
  
  /*

   * Final log writes as part of unmount.
   *
   * Mark the filesystem clean as unmount happens.  Note that during relocation
   * this routine needs to be executed as part of source-bag while the
   * deallocation must not be done until source-end.
   */
  
  /*
   * Unmount record used to have a string "Unmount filesystem--" in the
   * data section where the "Un" was really a magic number (XLOG_UNMOUNT_TYPE).
   * We just write the magic number now since that particular field isn't
   * currently architecture converted and "nUmount" is a bit foo.
   * As far as I know, there weren't any dependencies on the old behaviour.
   */
  
  int
  xfs_log_unmount_write(xfs_mount_t *mp)
  {
  	xlog_t		 *log = mp->m_log;
  	xlog_in_core_t	 *iclog;
  #ifdef DEBUG
  	xlog_in_core_t	 *first_iclog;
  #endif

  	xlog_ticket_t	*tic = NULL;

  	xfs_lsn_t	 lsn;
  	int		 error;

  	/*
  	 * Don't write out unmount record on read-only mounts.
  	 * Or, if we are doing a forced umount (typically because of IO errors).
  	 */

  	if (mp->m_flags & XFS_MOUNT_RDONLY)

  		return 0;

  	error = _xfs_log_force(mp, XFS_LOG_SYNC, NULL);

  	ASSERT(error || !(XLOG_FORCED_SHUTDOWN(log)));

  
  #ifdef DEBUG
  	first_iclog = iclog = log->l_iclog;
  	do {
  		if (!(iclog->ic_state & XLOG_STATE_IOERROR)) {
  			ASSERT(iclog->ic_state & XLOG_STATE_ACTIVE);
  			ASSERT(iclog->ic_offset == 0);
  		}
  		iclog = iclog->ic_next;
  	} while (iclog != first_iclog);
  #endif
  	if (! (XLOG_FORCED_SHUTDOWN(log))) {

  		error = xfs_log_reserve(mp, 600, 1, &tic,
  					XFS_LOG, 0, XLOG_UNMOUNT_REC_TYPE);

  		if (!error) {

  			/* the data section must be 32 bit size aligned */
  			struct {
  			    __uint16_t magic;
  			    __uint16_t pad1;
  			    __uint32_t pad2; /* may as well make it 64 bits */
  			} magic = {
  				.magic = XLOG_UNMOUNT_TYPE,
  			};
  			struct xfs_log_iovec reg = {

  				.i_addr = &magic,

  				.i_len = sizeof(magic),
  				.i_type = XLOG_REG_TYPE_UNMOUNT,
  			};
  			struct xfs_log_vec vec = {
  				.lv_niovecs = 1,
  				.lv_iovecp = &reg,
  			};

  			/* remove inited flag */

  			tic->t_flags = 0;
  			error = xlog_write(log, &vec, tic, &lsn,

  					   NULL, XLOG_UNMOUNT_TRANS);
  			/*
  			 * At this point, we're umounting anyway,
  			 * so there's no point in transitioning log state
  			 * to IOERROR. Just continue...
  			 */
  		}

  		if (error)
  			xfs_alert(mp, "%s: unmount record failed", __func__);

  		spin_lock(&log->l_icloglock);

  		iclog = log->l_iclog;

  		atomic_inc(&iclog->ic_refcnt);

  		xlog_state_want_sync(log, iclog);

  		spin_unlock(&log->l_icloglock);

  		error = xlog_state_release_iclog(log, iclog);

  		spin_lock(&log->l_icloglock);

  		if (!(iclog->ic_state == XLOG_STATE_ACTIVE ||
  		      iclog->ic_state == XLOG_STATE_DIRTY)) {
  			if (!XLOG_FORCED_SHUTDOWN(log)) {

  				xlog_wait(&iclog->ic_force_wait,
  							&log->l_icloglock);

  			} else {

  				spin_unlock(&log->l_icloglock);

  			}
  		} else {

  			spin_unlock(&log->l_icloglock);

  		}

  		if (tic) {

  			trace_xfs_log_umount_write(log, tic);

  			xlog_ungrant_log_space(log, tic);

  			xfs_log_ticket_put(tic);

  		}

  	} else {
  		/*
  		 * We're already in forced_shutdown mode, couldn't
  		 * even attempt to write out the unmount transaction.
  		 *
  		 * Go through the motions of sync'ing and releasing
  		 * the iclog, even though no I/O will actually happen,

  		 * we need to wait for other log I/Os that may already

  		 * be in progress.  Do this as a separate section of
  		 * code so we'll know if we ever get stuck here that
  		 * we're in this odd situation of trying to unmount
  		 * a file system that went into forced_shutdown as
  		 * the result of an unmount..
  		 */

  		spin_lock(&log->l_icloglock);

  		iclog = log->l_iclog;

  		atomic_inc(&iclog->ic_refcnt);

  
  		xlog_state_want_sync(log, iclog);

  		spin_unlock(&log->l_icloglock);

  		error =  xlog_state_release_iclog(log, iclog);

  		spin_lock(&log->l_icloglock);

  
  		if ( ! (   iclog->ic_state == XLOG_STATE_ACTIVE
  			|| iclog->ic_state == XLOG_STATE_DIRTY
  			|| iclog->ic_state == XLOG_STATE_IOERROR) ) {

  				xlog_wait(&iclog->ic_force_wait,
  							&log->l_icloglock);

  		} else {

  			spin_unlock(&log->l_icloglock);

  		}
  	}

  	return error;

  }	/* xfs_log_unmount_write */
  
  /*
   * Deallocate log structures for unmount/relocation.

   *
   * We need to stop the aild from running before we destroy
   * and deallocate the log as the aild references the log.

   */
  void

  xfs_log_unmount(xfs_mount_t *mp)

  {

  	xfs_trans_ail_destroy(mp);

  	xlog_dealloc_log(mp->m_log);

  }

  void
  xfs_log_item_init(
  	struct xfs_mount	*mp,
  	struct xfs_log_item	*item,
  	int			type,

  	const struct xfs_item_ops *ops)

  {
  	item->li_mountp = mp;
  	item->li_ailp = mp->m_ail;
  	item->li_type = type;
  	item->li_ops = ops;

  	item->li_lv = NULL;
  
  	INIT_LIST_HEAD(&item->li_ail);
  	INIT_LIST_HEAD(&item->li_cil);

  }

  void
  xfs_log_move_tail(xfs_mount_t	*mp,
  		  xfs_lsn_t	tail_lsn)
  {
  	xlog_ticket_t	*tic;
  	xlog_t		*log = mp->m_log;

  	int		need_bytes, free_bytes;

  	if (XLOG_FORCED_SHUTDOWN(log))
  		return;

  	if (tail_lsn == 0)
  		tail_lsn = atomic64_read(&log->l_last_sync_lsn);

  	/* tail_lsn == 1 implies that we weren't passed a valid value.  */
  	if (tail_lsn != 1)
  		atomic64_set(&log->l_tail_lsn, tail_lsn);

  	if (!list_empty_careful(&log->l_writeq)) {

  #ifdef DEBUG
  		if (log->l_flags & XLOG_ACTIVE_RECOVERY)
  			panic("Recovery problem");
  #endif

  		spin_lock(&log->l_grant_write_lock);

  		free_bytes = xlog_space_left(log, &log->l_grant_write_head);

  		list_for_each_entry(tic, &log->l_writeq, t_queue) {

  			ASSERT(tic->t_flags & XLOG_TIC_PERM_RESERV);
  
  			if (free_bytes < tic->t_unit_res && tail_lsn != 1)
  				break;
  			tail_lsn = 0;
  			free_bytes -= tic->t_unit_res;

  			trace_xfs_log_regrant_write_wake_up(log, tic);

  			wake_up(&tic->t_wait);

  		}

  		spin_unlock(&log->l_grant_write_lock);

  	}

  	if (!list_empty_careful(&log->l_reserveq)) {

  #ifdef DEBUG
  		if (log->l_flags & XLOG_ACTIVE_RECOVERY)
  			panic("Recovery problem");
  #endif

  		spin_lock(&log->l_grant_reserve_lock);

  		free_bytes = xlog_space_left(log, &log->l_grant_reserve_head);

  		list_for_each_entry(tic, &log->l_reserveq, t_queue) {

  			if (tic->t_flags & XLOG_TIC_PERM_RESERV)
  				need_bytes = tic->t_unit_res*tic->t_cnt;
  			else
  				need_bytes = tic->t_unit_res;
  			if (free_bytes < need_bytes && tail_lsn != 1)
  				break;
  			tail_lsn = 0;
  			free_bytes -= need_bytes;

  			trace_xfs_log_grant_wake_up(log, tic);

  			wake_up(&tic->t_wait);

  		}

  		spin_unlock(&log->l_grant_reserve_lock);

  	}

  }

  
  /*
   * Determine if we have a transaction that has gone to disk

   * that needs to be covered. To begin the transition to the idle state
   * firstly the log needs to be idle (no AIL and nothing in the iclogs).
   * If we are then in a state where covering is needed, the caller is informed
   * that dummy transactions are required to move the log into the idle state.
   *
   * Because this is called as part of the sync process, we should also indicate
   * that dummy transactions should be issued in anything but the covered or
   * idle states. This ensures that the log tail is accurately reflected in
   * the log at the end of the sync, hence if a crash occurrs avoids replay
   * of transactions where the metadata is already on disk.

   */
  int
  xfs_log_need_covered(xfs_mount_t *mp)
  {

  	int		needed = 0;

  	xlog_t		*log = mp->m_log;

  	if (!xfs_fs_writable(mp))

  		return 0;

  	spin_lock(&log->l_icloglock);

  	switch (log->l_covered_state) {
  	case XLOG_STATE_COVER_DONE:
  	case XLOG_STATE_COVER_DONE2:
  	case XLOG_STATE_COVER_IDLE:
  		break;
  	case XLOG_STATE_COVER_NEED:
  	case XLOG_STATE_COVER_NEED2:

  		if (!xfs_ail_min_lsn(log->l_ailp) &&

  		    xlog_iclogs_empty(log)) {
  			if (log->l_covered_state == XLOG_STATE_COVER_NEED)
  				log->l_covered_state = XLOG_STATE_COVER_DONE;
  			else
  				log->l_covered_state = XLOG_STATE_COVER_DONE2;

  		}

  		/* FALLTHRU */
  	default:

  		needed = 1;

  		break;

  	}

  	spin_unlock(&log->l_icloglock);

  	return needed;

  }
  
  /******************************************************************************
   *
   *	local routines
   *
   ******************************************************************************
   */
  
  /* xfs_trans_tail_ail returns 0 when there is nothing in the list.
   * The log manager must keep track of the last LR which was committed
   * to disk.  The lsn of this LR will become the new tail_lsn whenever
   * xfs_trans_tail_ail returns 0.  If we don't do this, we run into
   * the situation where stuff could be written into the log but nothing
   * was ever in the AIL when asked.  Eventually, we panic since the
   * tail hits the head.
   *
   * We may be holding the log iclog lock upon entering this routine.
   */
  xfs_lsn_t

  xlog_assign_tail_lsn(
  	struct xfs_mount	*mp)

  {

  	xfs_lsn_t		tail_lsn;
  	struct log		*log = mp->m_log;

  	tail_lsn = xfs_ail_min_lsn(mp->m_ail);

  	if (!tail_lsn)
  		tail_lsn = atomic64_read(&log->l_last_sync_lsn);

  	atomic64_set(&log->l_tail_lsn, tail_lsn);

  	return tail_lsn;

  }

  
  /*
   * Return the space in the log between the tail and the head.  The head
   * is passed in the cycle/bytes formal parms.  In the special case where
   * the reserve head has wrapped passed the tail, this calculation is no
   * longer valid.  In this case, just return 0 which means there is no space
   * in the log.  This works for all places where this function is called
   * with the reserve head.  Of course, if the write head were to ever
   * wrap the tail, we should blow up.  Rather than catch this case here,
   * we depend on other ASSERTions in other parts of the code.   XXXmiken
   *
   * This code also handles the case where the reservation head is behind
   * the tail.  The details of this case are described below, but the end
   * result is that we return the size of the log as the amount of space left.
   */

  STATIC int

  xlog_space_left(
  	struct log	*log,

  	atomic64_t	*head)

  {

  	int		free_bytes;
  	int		tail_bytes;
  	int		tail_cycle;
  	int		head_cycle;
  	int		head_bytes;

  	xlog_crack_grant_head(head, &head_cycle, &head_bytes);

  	xlog_crack_atomic_lsn(&log->l_tail_lsn, &tail_cycle, &tail_bytes);
  	tail_bytes = BBTOB(tail_bytes);

  	if (tail_cycle == head_cycle && head_bytes >= tail_bytes)
  		free_bytes = log->l_logsize - (head_bytes - tail_bytes);
  	else if (tail_cycle + 1 < head_cycle)

  		return 0;

  	else if (tail_cycle < head_cycle) {
  		ASSERT(tail_cycle == (head_cycle - 1));
  		free_bytes = tail_bytes - head_bytes;

  	} else {
  		/*
  		 * The reservation head is behind the tail.
  		 * In this case we just want to return the size of the
  		 * log as the amount of space left.
  		 */

  		xfs_alert(log->l_mp,

  			"xlog_space_left: head behind tail
  "
  			"  tail_cycle = %d, tail_bytes = %d
  "
  			"  GH   cycle = %d, GH   bytes = %d",

  			tail_cycle, tail_bytes, head_cycle, head_bytes);

  		ASSERT(0);
  		free_bytes = log->l_logsize;
  	}
  	return free_bytes;

  }

  
  
  /*
   * Log function which is called when an io completes.
   *
   * The log manager needs its own routine, in order to control what
   * happens with the buffer after the write completes.
   */
  void
  xlog_iodone(xfs_buf_t *bp)
  {

  	xlog_in_core_t	*iclog = bp->b_fspriv;
  	xlog_t		*l = iclog->ic_log;
  	int		aborted = 0;

  
  	/*
  	 * Race to shutdown the filesystem if we see an error.
  	 */

  	if (XFS_TEST_ERROR((xfs_buf_geterror(bp)), l->l_mp,

  			XFS_ERRTAG_IODONE_IOERR, XFS_RANDOM_IODONE_IOERR)) {

  		xfs_buf_ioerror_alert(bp, __func__);

  		xfs_buf_stale(bp);

  		xfs_force_shutdown(l->l_mp, SHUTDOWN_LOG_IO_ERROR);

  		/*
  		 * This flag will be propagated to the trans-committed
  		 * callback routines to let them know that the log-commit
  		 * didn't succeed.
  		 */
  		aborted = XFS_LI_ABORTED;
  	} else if (iclog->ic_state & XLOG_STATE_IOERROR) {
  		aborted = XFS_LI_ABORTED;
  	}

  
  	/* log I/O is always issued ASYNC */
  	ASSERT(XFS_BUF_ISASYNC(bp));

  	xlog_state_done_syncing(iclog, aborted);

  	/*
  	 * do not reference the buffer (bp) here as we could race
  	 * with it being freed after writing the unmount record to the
  	 * log.
  	 */

  }	/* xlog_iodone */
  
  /*

   * Return size of each in-core log record buffer.
   *

   * All machines get 8 x 32kB buffers by default, unless tuned otherwise.

   *
   * If the filesystem blocksize is too large, we may need to choose a
   * larger size since the directory code currently logs entire blocks.
   */
  
  STATIC void
  xlog_get_iclog_buffer_size(xfs_mount_t	*mp,
  			   xlog_t	*log)
  {
  	int size;
  	int xhdrs;

  	if (mp->m_logbufs <= 0)
  		log->l_iclog_bufs = XLOG_MAX_ICLOGS;
  	else

  		log->l_iclog_bufs = mp->m_logbufs;

  
  	/*
  	 * Buffer size passed in from mount system call.
  	 */

  	if (mp->m_logbsize > 0) {

  		size = log->l_iclog_size = mp->m_logbsize;
  		log->l_iclog_size_log = 0;
  		while (size != 1) {
  			log->l_iclog_size_log++;
  			size >>= 1;
  		}

  		if (xfs_sb_version_haslogv2(&mp->m_sb)) {

  			/* # headers = size / 32k
  			 * one header holds cycles from 32k of data

  			 */
  
  			xhdrs = mp->m_logbsize / XLOG_HEADER_CYCLE_SIZE;
  			if (mp->m_logbsize % XLOG_HEADER_CYCLE_SIZE)
  				xhdrs++;
  			log->l_iclog_hsize = xhdrs << BBSHIFT;
  			log->l_iclog_heads = xhdrs;
  		} else {
  			ASSERT(mp->m_logbsize <= XLOG_BIG_RECORD_BSIZE);
  			log->l_iclog_hsize = BBSIZE;
  			log->l_iclog_heads = 1;
  		}

  		goto done;

  	}

  	/* All machines use 32kB buffers by default. */

  	log->l_iclog_size = XLOG_BIG_RECORD_BSIZE;
  	log->l_iclog_size_log = XLOG_BIG_RECORD_BSHIFT;

  
  	/* the default log size is 16k or 32k which is one header sector */
  	log->l_iclog_hsize = BBSIZE;
  	log->l_iclog_heads = 1;

  done:
  	/* are we being asked to make the sizes selected above visible? */

  	if (mp->m_logbufs == 0)
  		mp->m_logbufs = log->l_iclog_bufs;
  	if (mp->m_logbsize == 0)
  		mp->m_logbsize = log->l_iclog_size;

  }	/* xlog_get_iclog_buffer_size */
  
  
  /*
   * This routine initializes some of the log structure for a given mount point.
   * Its primary purpose is to fill in enough, so recovery can occur.  However,
   * some other stuff may be filled in too.
   */
  STATIC xlog_t *
  xlog_alloc_log(xfs_mount_t	*mp,
  	       xfs_buftarg_t	*log_target,
  	       xfs_daddr_t	blk_offset,
  	       int		num_bblks)
  {
  	xlog_t			*log;
  	xlog_rec_header_t	*head;
  	xlog_in_core_t		**iclogp;
  	xlog_in_core_t		*iclog, *prev_iclog=NULL;
  	xfs_buf_t		*bp;
  	int			i;

  	int			error = ENOMEM;

  	uint			log2_size = 0;

  	log = kmem_zalloc(sizeof(xlog_t), KM_MAYFAIL);

  	if (!log) {

  		xfs_warn(mp, "Log allocation failed: No memory!");

  		goto out;
  	}

  
  	log->l_mp	   = mp;
  	log->l_targ	   = log_target;
  	log->l_logsize     = BBTOB(num_bblks);
  	log->l_logBBstart  = blk_offset;
  	log->l_logBBsize   = num_bblks;
  	log->l_covered_state = XLOG_STATE_COVER_IDLE;
  	log->l_flags	   |= XLOG_ACTIVE_RECOVERY;
  
  	log->l_prev_block  = -1;

  	/* log->l_tail_lsn = 0x100000000LL; cycle = 1; current block = 0 */

  	xlog_assign_atomic_lsn(&log->l_tail_lsn, 1, 0);
  	xlog_assign_atomic_lsn(&log->l_last_sync_lsn, 1, 0);

  	log->l_curr_cycle  = 1;	    /* 0 is bad since this is initial value */

  	xlog_assign_grant_head(&log->l_grant_reserve_head, 1, 0);
  	xlog_assign_grant_head(&log->l_grant_write_head, 1, 0);

  	INIT_LIST_HEAD(&log->l_reserveq);
  	INIT_LIST_HEAD(&log->l_writeq);

  	spin_lock_init(&log->l_grant_reserve_lock);
  	spin_lock_init(&log->l_grant_write_lock);

  	error = EFSCORRUPTED;

  	if (xfs_sb_version_hassector(&mp->m_sb)) {

  	        log2_size = mp->m_sb.sb_logsectlog;
  		if (log2_size < BBSHIFT) {

  			xfs_warn(mp, "Log sector size too small (0x%x < 0x%x)",
  				log2_size, BBSHIFT);

  			goto out_free_log;
  		}

  	        log2_size -= BBSHIFT;
  		if (log2_size > mp->m_sectbb_log) {

  			xfs_warn(mp, "Log sector size too large (0x%x > 0x%x)",
  				log2_size, mp->m_sectbb_log);

  			goto out_free_log;
  		}

  
  		/* for larger sector sizes, must have v2 or external log */
  		if (log2_size && log->l_logBBstart > 0 &&
  			    !xfs_sb_version_haslogv2(&mp->m_sb)) {

  			xfs_warn(mp,
  		"log sector size (0x%x) invalid for configuration.",
  				log2_size);

  			goto out_free_log;
  		}

  	}

  	log->l_sectBBsize = 1 << log2_size;

  
  	xlog_get_iclog_buffer_size(mp, log);

  	error = ENOMEM;

  	bp = xfs_buf_alloc(mp->m_logdev_targp, 0, log->l_iclog_size, 0);

  	if (!bp)
  		goto out_free_log;

  	bp->b_iodone = xlog_iodone;

  	ASSERT(xfs_buf_islocked(bp));

  	log->l_xbuf = bp;

  	spin_lock_init(&log->l_icloglock);

  	init_waitqueue_head(&log->l_flush_wait);

  
  	/* log record size must be multiple of BBSIZE; see xlog_rec_header_t */
  	ASSERT((XFS_BUF_SIZE(bp) & BBMASK) == 0);
  
  	iclogp = &log->l_iclog;
  	/*
  	 * The amount of memory to allocate for the iclog structure is
  	 * rather funky due to the way the structure is defined.  It is
  	 * done this way so that we can use different sizes for machines
  	 * with different amounts of memory.  See the definition of
  	 * xlog_in_core_t in xfs_log_priv.h for details.
  	 */

  	ASSERT(log->l_iclog_size >= 4096);
  	for (i=0; i < log->l_iclog_bufs; i++) {

  		*iclogp = kmem_zalloc(sizeof(xlog_in_core_t), KM_MAYFAIL);
  		if (!*iclogp)
  			goto out_free_iclog;

  		iclog = *iclogp;

  		iclog->ic_prev = prev_iclog;
  		prev_iclog = iclog;

  		bp = xfs_buf_get_uncached(mp->m_logdev_targp,
  						log->l_iclog_size, 0);

  		if (!bp)
  			goto out_free_iclog;

  		bp->b_iodone = xlog_iodone;

  		iclog->ic_bp = bp;

  		iclog->ic_data = bp->b_addr;

  #ifdef DEBUG

  		log->l_iclog_bak[i] = (xfs_caddr_t)&(iclog->ic_header);

  #endif

  		head = &iclog->ic_header;
  		memset(head, 0, sizeof(xlog_rec_header_t));

  		head->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM);
  		head->h_version = cpu_to_be32(

  			xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1);

  		head->h_size = cpu_to_be32(log->l_iclog_size);

  		/* new fields */

  		head->h_fmt = cpu_to_be32(XLOG_FMT);

  		memcpy(&head->h_fs_uuid, &mp->m_sb.sb_uuid, sizeof(uuid_t));

  		iclog->ic_size = XFS_BUF_SIZE(bp) - log->l_iclog_hsize;
  		iclog->ic_state = XLOG_STATE_ACTIVE;
  		iclog->ic_log = log;

  		atomic_set(&iclog->ic_refcnt, 0);
  		spin_lock_init(&iclog->ic_callback_lock);

  		iclog->ic_callback_tail = &(iclog->ic_callback);

  		iclog->ic_datap = (char *)iclog->ic_data + log->l_iclog_hsize;

  		ASSERT(xfs_buf_islocked(iclog->ic_bp));

  		init_waitqueue_head(&iclog->ic_force_wait);
  		init_waitqueue_head(&iclog->ic_write_wait);

  
  		iclogp = &iclog->ic_next;
  	}
  	*iclogp = log->l_iclog;			/* complete ring */
  	log->l_iclog->ic_prev = prev_iclog;	/* re-write 1st prev ptr */

  	error = xlog_cil_init(log);
  	if (error)
  		goto out_free_iclog;

  	return log;

  
  out_free_iclog:
  	for (iclog = log->l_iclog; iclog; iclog = prev_iclog) {
  		prev_iclog = iclog->ic_next;

  		if (iclog->ic_bp)

  			xfs_buf_free(iclog->ic_bp);

  		kmem_free(iclog);
  	}
  	spinlock_destroy(&log->l_icloglock);

  	xfs_buf_free(log->l_xbuf);
  out_free_log:
  	kmem_free(log);

  out:
  	return ERR_PTR(-error);

  }	/* xlog_alloc_log */
  
  
  /*
   * Write out the commit record of a transaction associated with the given
   * ticket.  Return the lsn of the commit record.
   */
  STATIC int

  xlog_commit_record(
  	struct log		*log,
  	struct xlog_ticket	*ticket,
  	struct xlog_in_core	**iclog,
  	xfs_lsn_t		*commitlsnp)

  {

  	struct xfs_mount *mp = log->l_mp;
  	int	error;
  	struct xfs_log_iovec reg = {
  		.i_addr = NULL,
  		.i_len = 0,
  		.i_type = XLOG_REG_TYPE_COMMIT,
  	};
  	struct xfs_log_vec vec = {
  		.lv_niovecs = 1,
  		.lv_iovecp = &reg,
  	};

  
  	ASSERT_ALWAYS(iclog);

  	error = xlog_write(log, &vec, ticket, commitlsnp, iclog,
  					XLOG_COMMIT_TRANS);
  	if (error)

  		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);

  	return error;

  }

  
  /*
   * Push on the buffer cache code if we ever use more than 75% of the on-disk
   * log space.  This code pushes on the lsn which would supposedly free up
   * the 25% which we want to leave free.  We may need to adopt a policy which
   * pushes on an lsn which is further along in the log once we reach the high
   * water mark.  In this manner, we would be creating a low water mark.
   */

  STATIC void

  xlog_grant_push_ail(
  	struct log	*log,
  	int		need_bytes)

  {

  	xfs_lsn_t	threshold_lsn = 0;

  	xfs_lsn_t	last_sync_lsn;

  	int		free_blocks;
  	int		free_bytes;
  	int		threshold_block;
  	int		threshold_cycle;
  	int		free_threshold;
  
  	ASSERT(BTOBB(need_bytes) < log->l_logBBsize);

  	free_bytes = xlog_space_left(log, &log->l_grant_reserve_head);
  	free_blocks = BTOBBT(free_bytes);
  
  	/*
  	 * Set the threshold for the minimum number of free blocks in the
  	 * log to the maximum of what the caller needs, one quarter of the
  	 * log, and 256 blocks.
  	 */
  	free_threshold = BTOBB(need_bytes);
  	free_threshold = MAX(free_threshold, (log->l_logBBsize >> 2));
  	free_threshold = MAX(free_threshold, 256);
  	if (free_blocks >= free_threshold)
  		return;

  	xlog_crack_atomic_lsn(&log->l_tail_lsn, &threshold_cycle,
  						&threshold_block);
  	threshold_block += free_threshold;

  	if (threshold_block >= log->l_logBBsize) {

  		threshold_block -= log->l_logBBsize;
  		threshold_cycle += 1;

  	}

  	threshold_lsn = xlog_assign_lsn(threshold_cycle,
  					threshold_block);
  	/*
  	 * Don't pass in an lsn greater than the lsn of the last

  	 * log record known to be on disk. Use a snapshot of the last sync lsn
  	 * so that it doesn't change between the compare and the set.

  	 */

  	last_sync_lsn = atomic64_read(&log->l_last_sync_lsn);
  	if (XFS_LSN_CMP(threshold_lsn, last_sync_lsn) > 0)
  		threshold_lsn = last_sync_lsn;

  
  	/*
  	 * Get the transaction layer to kick the dirty buffers out to
  	 * disk asynchronously. No point in trying to do this if
  	 * the filesystem is shutting down.
  	 */
  	if (!XLOG_FORCED_SHUTDOWN(log))

  		xfs_ail_push(log->l_ailp, threshold_lsn);

  }

  /*
   * The bdstrat callback function for log bufs. This gives us a central
   * place to trap bufs in case we get hit by a log I/O error and need to
   * shutdown. Actually, in practice, even when we didn't get a log error,
   * we transition the iclogs to IOERROR state *after* flushing all existing
   * iclogs to disk. This is because we don't want anymore new transactions to be
   * started or completed afterwards.
   */
  STATIC int
  xlog_bdstrat(
  	struct xfs_buf		*bp)
  {

  	struct xlog_in_core	*iclog = bp->b_fspriv;

  	if (iclog->ic_state & XLOG_STATE_IOERROR) {

  		xfs_buf_ioerror(bp, EIO);

  		xfs_buf_stale(bp);

  		xfs_buf_ioend(bp, 0);

  		/*
  		 * It would seem logical to return EIO here, but we rely on
  		 * the log state machine to propagate I/O errors instead of
  		 * doing it here.
  		 */
  		return 0;
  	}

  	xfs_buf_iorequest(bp);
  	return 0;
  }

  
  /*
   * Flush out the in-core log (iclog) to the on-disk log in an asynchronous 
   * fashion.  Previously, we should have moved the current iclog
   * ptr in the log to point to the next available iclog.  This allows further
   * write to continue while this code syncs out an iclog ready to go.
   * Before an in-core log can be written out, the data section must be scanned
   * to save away the 1st word of each BBSIZE block into the header.  We replace
   * it with the current cycle count.  Each BBSIZE block is tagged with the
   * cycle count because there in an implicit assumption that drives will
   * guarantee that entire 512 byte blocks get written at once.  In other words,
   * we can't have part of a 512 byte block written and part not written.  By
   * tagging each block, we will know which blocks are valid when recovering
   * after an unclean shutdown.
   *
   * This routine is single threaded on the iclog.  No other thread can be in
   * this routine with the same iclog.  Changing contents of iclog can there-
   * fore be done without grabbing the state machine lock.  Updating the global
   * log will require grabbing the lock though.
   *
   * The entire log manager uses a logical block numbering scheme.  Only
   * log_sync (and then only bwrite()) know about the fact that the log may
   * not start with block zero on a given device.  The log block start offset
   * is added immediately before calling bwrite().
   */

  STATIC int

  xlog_sync(xlog_t		*log,
  	  xlog_in_core_t	*iclog)
  {
  	xfs_caddr_t	dptr;		/* pointer to byte sized element */
  	xfs_buf_t	*bp;

  	int		i;

  	uint		count;		/* byte count of bwrite */
  	uint		count_init;	/* initial count before roundup */
  	int		roundoff;       /* roundoff to BB or stripe */
  	int		split = 0;	/* split write into two regions */
  	int		error;

  	int		v2 = xfs_sb_version_haslogv2(&log->l_mp->m_sb);

  
  	XFS_STATS_INC(xs_log_writes);

  	ASSERT(atomic_read(&iclog->ic_refcnt) == 0);

  
  	/* Add for LR header */
  	count_init = log->l_iclog_hsize + iclog->ic_offset;
  
  	/* Round out the log write size */
  	if (v2 && log->l_mp->m_sb.sb_logsunit > 1) {
  		/* we have a v2 stripe unit to use */
  		count = XLOG_LSUNITTOB(log, XLOG_BTOLSUNIT(log, count_init));
  	} else {
  		count = BBTOB(BTOBB(count_init));
  	}
  	roundoff = count - count_init;
  	ASSERT(roundoff >= 0);
  	ASSERT((v2 && log->l_mp->m_sb.sb_logsunit > 1 && 
                  roundoff < log->l_mp->m_sb.sb_logsunit)
  		|| 
  		(log->l_mp->m_sb.sb_logsunit <= 1 && 
  		 roundoff < BBTOB(1)));
  
  	/* move grant heads by roundoff in sync */

  	xlog_grant_add_space(log, &log->l_grant_reserve_head, roundoff);
  	xlog_grant_add_space(log, &log->l_grant_write_head, roundoff);

  
  	/* put cycle number in every block */
  	xlog_pack_data(log, iclog, roundoff); 
  
  	/* real byte length */
  	if (v2) {

  		iclog->ic_header.h_len =
  			cpu_to_be32(iclog->ic_offset + roundoff);

  	} else {

  		iclog->ic_header.h_len =
  			cpu_to_be32(iclog->ic_offset);

  	}

  	bp = iclog->ic_bp;

  	XFS_BUF_SET_ADDR(bp, BLOCK_LSN(be64_to_cpu(iclog->ic_header.h_lsn)));

  
  	XFS_STATS_ADD(xs_log_blocks, BTOBB(count));
  
  	/* Do we need to split this write into 2 parts? */
  	if (XFS_BUF_ADDR(bp) + BTOBB(count) > log->l_logBBsize) {
  		split = count - (BBTOB(log->l_logBBsize - XFS_BUF_ADDR(bp)));
  		count = BBTOB(log->l_logBBsize - XFS_BUF_ADDR(bp));
  		iclog->ic_bwritecnt = 2;	/* split into 2 writes */
  	} else {
  		iclog->ic_bwritecnt = 1;
  	}

  	XFS_BUF_SET_COUNT(bp, count);

  	bp->b_fspriv = iclog;

  	XFS_BUF_ZEROFLAGS(bp);

  	XFS_BUF_ASYNC(bp);

  	bp->b_flags |= XBF_SYNCIO;

  	if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) {

  		bp->b_flags |= XBF_FUA;

  		/*

  		 * Flush the data device before flushing the log to make
  		 * sure all meta data written back from the AIL actually made
  		 * it to disk before stamping the new log tail LSN into the
  		 * log buffer.  For an external log we need to issue the
  		 * flush explicitly, and unfortunately synchronously here;
  		 * for an internal log we can simply use the block layer
  		 * state machine for preflushes.

  		 */
  		if (log->l_mp->m_logdev_targp != log->l_mp->m_ddev_targp)
  			xfs_blkdev_issue_flush(log->l_mp->m_ddev_targp);

  		else
  			bp->b_flags |= XBF_FLUSH;

  	}

  
  	ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1);
  	ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize);
  
  	xlog_verify_iclog(log, iclog, count, B_TRUE);
  
  	/* account for log which doesn't start at block #0 */
  	XFS_BUF_SET_ADDR(bp, XFS_BUF_ADDR(bp) + log->l_logBBstart);
  	/*
  	 * Don't call xfs_bwrite here. We do log-syncs even when the filesystem
  	 * is shutting down.
  	 */
  	XFS_BUF_WRITE(bp);

  	error = xlog_bdstrat(bp);
  	if (error) {
  		xfs_buf_ioerror_alert(bp, "xlog_sync");

  		return error;

  	}
  	if (split) {

  		bp = iclog->ic_log->l_xbuf;

  		XFS_BUF_SET_ADDR(bp, 0);	     /* logical 0 */

  		xfs_buf_associate_memory(bp,
  				(char *)&iclog->ic_header + count, split);

  		bp->b_fspriv = iclog;

  		XFS_BUF_ZEROFLAGS(bp);

  		XFS_BUF_ASYNC(bp);

  		bp->b_flags |= XBF_SYNCIO;

  		if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)

  			bp->b_flags |= XBF_FUA;

  		dptr = bp->b_addr;

  		/*
  		 * Bump the cycle numbers at the start of each block
  		 * since this part of the buffer is at the start of
  		 * a new cycle.  Watch out for the header magic number
  		 * case, though.
  		 */

  		for (i = 0; i < split; i += BBSIZE) {

  			be32_add_cpu((__be32 *)dptr, 1);

  			if (be32_to_cpu(*(__be32 *)dptr) == XLOG_HEADER_MAGIC_NUM)

  				be32_add_cpu((__be32 *)dptr, 1);

  			dptr += BBSIZE;
  		}
  
  		ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1);
  		ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize);

  		/* account for internal log which doesn't start at block #0 */

  		XFS_BUF_SET_ADDR(bp, XFS_BUF_ADDR(bp) + log->l_logBBstart);
  		XFS_BUF_WRITE(bp);

  		error = xlog_bdstrat(bp);
  		if (error) {
  			xfs_buf_ioerror_alert(bp, "xlog_sync (split)");

  			return error;

  		}
  	}

  	return 0;

  }	/* xlog_sync */
  
  
  /*

   * Deallocate a log structure

   */

  STATIC void

  xlog_dealloc_log(xlog_t *log)

  {
  	xlog_in_core_t	*iclog, *next_iclog;

  	int		i;

  	xlog_cil_destroy(log);

  	/*
  	 * always need to ensure that the extra buffer does not point to memory
  	 * owned by another log buffer before we free it.
  	 */
  	xfs_buf_set_empty(log->l_xbuf, log->l_iclog_size);
  	xfs_buf_free(log->l_xbuf);

  	iclog = log->l_iclog;
  	for (i=0; i<log->l_iclog_bufs; i++) {

  		xfs_buf_free(iclog->ic_bp);

  		next_iclog = iclog->ic_next;

  		kmem_free(iclog);

  		iclog = next_iclog;
  	}

  	spinlock_destroy(&log->l_icloglock);

  	log->l_mp->m_log = NULL;

  	kmem_free(log);

  }	/* xlog_dealloc_log */

  
  /*
   * Update counters atomically now that memcpy is done.
   */
  /* ARGSUSED */
  static inline void
  xlog_state_finish_copy(xlog_t		*log,
  		       xlog_in_core_t	*iclog,
  		       int		record_cnt,
  		       int		copy_bytes)
  {

  	spin_lock(&log->l_icloglock);

  	be32_add_cpu(&iclog->ic_header.h_num_logops, record_cnt);

  	iclog->ic_offset += copy_bytes;

  	spin_unlock(&log->l_icloglock);

  }	/* xlog_state_finish_copy */
  
  
  
  
  /*

   * print out info relating to regions written which consume
   * the reservation
   */

  void
  xlog_print_tic_res(
  	struct xfs_mount	*mp,
  	struct xlog_ticket	*ticket)

  {
  	uint i;
  	uint ophdr_spc = ticket->t_res_num_ophdrs * (uint)sizeof(xlog_op_header_t);
  
  	/* match with XLOG_REG_TYPE_* in xfs_log.h */
  	static char *res_type_str[XLOG_REG_TYPE_MAX] = {
  	    "bformat",
  	    "bchunk",
  	    "efi_format",
  	    "efd_format",
  	    "iformat",
  	    "icore",
  	    "iext",
  	    "ibroot",
  	    "ilocal",
  	    "iattr_ext",
  	    "iattr_broot",
  	    "iattr_local",
  	    "qformat",
  	    "dquot",
  	    "quotaoff",
  	    "LR header",
  	    "unmount",
  	    "commit",
  	    "trans header"
  	};
  	static char *trans_type_str[XFS_TRANS_TYPE_MAX] = {
  	    "SETATTR_NOT_SIZE",
  	    "SETATTR_SIZE",
  	    "INACTIVE",
  	    "CREATE",
  	    "CREATE_TRUNC",
  	    "TRUNCATE_FILE",
  	    "REMOVE",
  	    "LINK",
  	    "RENAME",
  	    "MKDIR",
  	    "RMDIR",
  	    "SYMLINK",
  	    "SET_DMATTRS",
  	    "GROWFS",
  	    "STRAT_WRITE",
  	    "DIOSTRAT",
  	    "WRITE_SYNC",
  	    "WRITEID",
  	    "ADDAFORK",
  	    "ATTRINVAL",
  	    "ATRUNCATE",
  	    "ATTR_SET",
  	    "ATTR_RM",
  	    "ATTR_FLAG",
  	    "CLEAR_AGI_BUCKET",
  	    "QM_SBCHANGE",
  	    "DUMMY1",
  	    "DUMMY2",
  	    "QM_QUOTAOFF",
  	    "QM_DQALLOC",
  	    "QM_SETQLIM",
  	    "QM_DQCLUSTER",
  	    "QM_QINOCREATE",
  	    "QM_QUOTAOFF_END",
  	    "SB_UNIT",
  	    "FSYNC_TS",
  	    "GROWFSRT_ALLOC",
  	    "GROWFSRT_ZERO",
  	    "GROWFSRT_FREE",
  	    "SWAPEXT"
  	};

  	xfs_warn(mp,

  		"xlog_write: reservation summary:
  "

  		"  trans type  = %s (%u)
  "
  		"  unit res    = %d bytes
  "
  		"  current res = %d bytes
  "
  		"  total reg   = %u bytes (o/flow = %u bytes)
  "
  		"  ophdrs      = %u (ophdr space = %u bytes)
  "
  		"  ophdr + reg = %u bytes
  "
  		"  num regions = %u
  ",
  		((ticket->t_trans_type <= 0 ||
  		  ticket->t_trans_type > XFS_TRANS_TYPE_MAX) ?
  		  "bad-trans-type" : trans_type_str[ticket->t_trans_type-1]),
  		ticket->t_trans_type,
  		ticket->t_unit_res,
  		ticket->t_curr_res,
  		ticket->t_res_arr_sum, ticket->t_res_o_flow,
  		ticket->t_res_num_ophdrs, ophdr_spc,
  		ticket->t_res_arr_sum +
  		ticket->t_res_o_flow + ophdr_spc,
  		ticket->t_res_num);

  
  	for (i = 0; i < ticket->t_res_num; i++) {

  		uint r_type = ticket->t_res_arr[i].r_type;
  		xfs_warn(mp, "region[%u]: %s - %u bytes
  ", i,

  			    ((r_type <= 0 || r_type > XLOG_REG_TYPE_MAX) ?
  			    "bad-rtype" : res_type_str[r_type-1]),
  			    ticket->t_res_arr[i].r_len);
  	}

  	xfs_alert_tag(mp, XFS_PTAG_LOGRES,

  		"xlog_write: reservation ran out. Need to up reservation");

  	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);

  }

  
  /*

   * Calculate the potential space needed by the log vector.  Each region gets
   * its own xlog_op_header_t and may need to be double word aligned.
   */
  static int
  xlog_write_calc_vec_length(
  	struct xlog_ticket	*ticket,

  	struct xfs_log_vec	*log_vector)

  {

  	struct xfs_log_vec	*lv;

  	int			headers = 0;
  	int			len = 0;
  	int			i;
  
  	/* acct for start rec of xact */
  	if (ticket->t_flags & XLOG_TIC_INITED)
  		headers++;

  	for (lv = log_vector; lv; lv = lv->lv_next) {
  		headers += lv->lv_niovecs;
  
  		for (i = 0; i < lv->lv_niovecs; i++) {
  			struct xfs_log_iovec	*vecp = &lv->lv_iovecp[i];

  			len += vecp->i_len;
  			xlog_tic_add_region(ticket, vecp->i_len, vecp->i_type);
  		}

  	}
  
  	ticket->t_res_num_ophdrs += headers;
  	len += headers * sizeof(struct xlog_op_header);
  
  	return len;
  }
  
  /*
   * If first write for transaction, insert start record  We can't be trying to
   * commit if we are inited.  We can't have any "partial_copy" if we are inited.
   */
  static int
  xlog_write_start_rec(

  	struct xlog_op_header	*ophdr,

  	struct xlog_ticket	*ticket)
  {

  	if (!(ticket->t_flags & XLOG_TIC_INITED))
  		return 0;
  
  	ophdr->oh_tid	= cpu_to_be32(ticket->t_tid);
  	ophdr->oh_clientid = ticket->t_clientid;
  	ophdr->oh_len = 0;
  	ophdr->oh_flags = XLOG_START_TRANS;
  	ophdr->oh_res2 = 0;
  
  	ticket->t_flags &= ~XLOG_TIC_INITED;
  
  	return sizeof(struct xlog_op_header);
  }
  
  static xlog_op_header_t *
  xlog_write_setup_ophdr(
  	struct log		*log,

  	struct xlog_op_header	*ophdr,

  	struct xlog_ticket	*ticket,
  	uint			flags)
  {

  	ophdr->oh_tid = cpu_to_be32(ticket->t_tid);
  	ophdr->oh_clientid = ticket->t_clientid;
  	ophdr->oh_res2 = 0;
  
  	/* are we copying a commit or unmount record? */
  	ophdr->oh_flags = flags;
  
  	/*
  	 * We've seen logs corrupted with bad transaction client ids.  This
  	 * makes sure that XFS doesn't generate them on.  Turn this into an EIO
  	 * and shut down the filesystem.
  	 */
  	switch (ophdr->oh_clientid)  {
  	case XFS_TRANSACTION:
  	case XFS_VOLUME:
  	case XFS_LOG:
  		break;
  	default:

  		xfs_warn(log->l_mp,

  			"Bad XFS transaction clientid 0x%x in ticket 0x%p",
  			ophdr->oh_clientid, ticket);
  		return NULL;
  	}
  
  	return ophdr;
  }
  
  /*
   * Set up the parameters of the region copy into the log. This has
   * to handle region write split across multiple log buffers - this
   * state is kept external to this function so that this code can
   * can be written in an obvious, self documenting manner.
   */
  static int
  xlog_write_setup_copy(
  	struct xlog_ticket	*ticket,
  	struct xlog_op_header	*ophdr,
  	int			space_available,
  	int			space_required,
  	int			*copy_off,
  	int			*copy_len,
  	int			*last_was_partial_copy,
  	int			*bytes_consumed)
  {
  	int			still_to_copy;
  
  	still_to_copy = space_required - *bytes_consumed;
  	*copy_off = *bytes_consumed;
  
  	if (still_to_copy <= space_available) {
  		/* write of region completes here */
  		*copy_len = still_to_copy;
  		ophdr->oh_len = cpu_to_be32(*copy_len);
  		if (*last_was_partial_copy)
  			ophdr->oh_flags |= (XLOG_END_TRANS|XLOG_WAS_CONT_TRANS);
  		*last_was_partial_copy = 0;
  		*bytes_consumed = 0;
  		return 0;
  	}
  
  	/* partial write of region, needs extra log op header reservation */
  	*copy_len = space_available;
  	ophdr->oh_len = cpu_to_be32(*copy_len);
  	ophdr->oh_flags |= XLOG_CONTINUE_TRANS;
  	if (*last_was_partial_copy)
  		ophdr->oh_flags |= XLOG_WAS_CONT_TRANS;
  	*bytes_consumed += *copy_len;
  	(*last_was_partial_copy)++;
  
  	/* account for new log op header */
  	ticket->t_curr_res -= sizeof(struct xlog_op_header);
  	ticket->t_res_num_ophdrs++;
  
  	return sizeof(struct xlog_op_header);
  }
  
  static int
  xlog_write_copy_finish(
  	struct log		*log,
  	struct xlog_in_core	*iclog,
  	uint			flags,
  	int			*record_cnt,
  	int			*data_cnt,
  	int			*partial_copy,
  	int			*partial_copy_len,
  	int			log_offset,
  	struct xlog_in_core	**commit_iclog)
  {
  	if (*partial_copy) {
  		/*
  		 * This iclog has already been marked WANT_SYNC by
  		 * xlog_state_get_iclog_space.
  		 */
  		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
  		*record_cnt = 0;
  		*data_cnt = 0;
  		return xlog_state_release_iclog(log, iclog);
  	}
  
  	*partial_copy = 0;
  	*partial_copy_len = 0;
  
  	if (iclog->ic_size - log_offset <= sizeof(xlog_op_header_t)) {
  		/* no more space in this iclog - push it. */
  		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
  		*record_cnt = 0;
  		*data_cnt = 0;
  
  		spin_lock(&log->l_icloglock);
  		xlog_state_want_sync(log, iclog);
  		spin_unlock(&log->l_icloglock);
  
  		if (!commit_iclog)
  			return xlog_state_release_iclog(log, iclog);
  		ASSERT(flags & XLOG_COMMIT_TRANS);
  		*commit_iclog = iclog;
  	}
  
  	return 0;
  }
  
  /*

   * Write some region out to in-core log
   *
   * This will be called when writing externally provided regions or when
   * writing out a commit record for a given transaction.
   *
   * General algorithm:
   *	1. Find total length of this write.  This may include adding to the
   *		lengths passed in.
   *	2. Check whether we violate the tickets reservation.
   *	3. While writing to this iclog
   *	    A. Reserve as much space in this iclog as can get
   *	    B. If this is first write, save away start lsn
   *	    C. While writing this region:
   *		1. If first write of transaction, write start record
   *		2. Write log operation header (header per region)
   *		3. Find out if we can fit entire region into this iclog
   *		4. Potentially, verify destination memcpy ptr
   *		5. Memcpy (partial) region
   *		6. If partial copy, release iclog; otherwise, continue
   *			copying more regions into current iclog
   *	4. Mark want sync bit (in simulation mode)
   *	5. Release iclog for potential flush to on-disk log.
   *
   * ERRORS:
   * 1.	Panic if reservation is overrun.  This should never happen since
   *	reservation amounts are generated internal to the filesystem.
   * NOTES:
   * 1. Tickets are single threaded data structures.
   * 2. The XLOG_END_TRANS & XLOG_CONTINUE_TRANS flags are passed down to the
   *	syncing routine.  When a single log_write region needs to span
   *	multiple in-core logs, the XLOG_CONTINUE_TRANS bit should be set
   *	on all log operation writes which don't contain the end of the
   *	region.  The XLOG_END_TRANS bit is used for the in-core log
   *	operation which contains the end of the continued log_write region.
   * 3. When xlog_state_get_iclog_space() grabs the rest of the current iclog,
   *	we don't really know exactly how much space will be used.  As a result,
   *	we don't update ic_offset until the end when we know exactly how many
   *	bytes have been written out.
   */

  int

  xlog_write(

  	struct log		*log,
  	struct xfs_log_vec	*log_vector,

  	struct xlog_ticket	*ticket,
  	xfs_lsn_t		*start_lsn,
  	struct xlog_in_core	**commit_iclog,
  	uint			flags)

  {

  	struct xlog_in_core	*iclog = NULL;

  	struct xfs_log_iovec	*vecp;
  	struct xfs_log_vec	*lv;

  	int			len;
  	int			index;
  	int			partial_copy = 0;
  	int			partial_copy_len = 0;
  	int			contwr = 0;
  	int			record_cnt = 0;
  	int			data_cnt = 0;
  	int			error;
  
  	*start_lsn = 0;

  	len = xlog_write_calc_vec_length(ticket, log_vector);

  	/*
  	 * Region headers and bytes are already accounted for.
  	 * We only need to take into account start records and
  	 * split regions in this function.
  	 */
  	if (ticket->t_flags & XLOG_TIC_INITED)
  		ticket->t_curr_res -= sizeof(xlog_op_header_t);
  
  	/*
  	 * Commit record headers need to be accounted for. These
  	 * come in as separate writes so are easy to detect.
  	 */
  	if (flags & (XLOG_COMMIT_TRANS | XLOG_UNMOUNT_TRANS))
  		ticket->t_curr_res -= sizeof(xlog_op_header_t);

  
  	if (ticket->t_curr_res < 0)

  		xlog_print_tic_res(log->l_mp, ticket);

  	index = 0;
  	lv = log_vector;
  	vecp = lv->lv_iovecp;
  	while (lv && index < lv->lv_niovecs) {

  		void		*ptr;

  		int		log_offset;

  		error = xlog_state_get_iclog_space(log, len, &iclog, ticket,
  						   &contwr, &log_offset);
  		if (error)
  			return error;

  		ASSERT(log_offset <= iclog->ic_size - 1);

  		ptr = iclog->ic_datap + log_offset;

  		/* start_lsn is the first lsn written to. That's all we need. */
  		if (!*start_lsn)
  			*start_lsn = be64_to_cpu(iclog->ic_header.h_lsn);

  		/*
  		 * This loop writes out as many regions as can fit in the amount
  		 * of space which was allocated by xlog_state_get_iclog_space().
  		 */

  		while (lv && index < lv->lv_niovecs) {
  			struct xfs_log_iovec	*reg = &vecp[index];

  			struct xlog_op_header	*ophdr;
  			int			start_rec_copy;
  			int			copy_len;
  			int			copy_off;

  			ASSERT(reg->i_len % sizeof(__int32_t) == 0);

  			ASSERT((unsigned long)ptr % sizeof(__int32_t) == 0);

  
  			start_rec_copy = xlog_write_start_rec(ptr, ticket);
  			if (start_rec_copy) {
  				record_cnt++;

  				xlog_write_adv_cnt(&ptr, &len, &log_offset,

  						   start_rec_copy);
  			}

  			ophdr = xlog_write_setup_ophdr(log, ptr, ticket, flags);
  			if (!ophdr)
  				return XFS_ERROR(EIO);

  			xlog_write_adv_cnt(&ptr, &len, &log_offset,

  					   sizeof(struct xlog_op_header));
  
  			len += xlog_write_setup_copy(ticket, ophdr,
  						     iclog->ic_size-log_offset,

  						     reg->i_len,

  						     &copy_off, &copy_len,
  						     &partial_copy,
  						     &partial_copy_len);
  			xlog_verify_dest_ptr(log, ptr);
  
  			/* copy region */
  			ASSERT(copy_len >= 0);

  			memcpy(ptr, reg->i_addr + copy_off, copy_len);
  			xlog_write_adv_cnt(&ptr, &len, &log_offset, copy_len);

  
  			copy_len += start_rec_copy + sizeof(xlog_op_header_t);
  			record_cnt++;
  			data_cnt += contwr ? copy_len : 0;
  
  			error = xlog_write_copy_finish(log, iclog, flags,
  						       &record_cnt, &data_cnt,
  						       &partial_copy,
  						       &partial_copy_len,
  						       log_offset,
  						       commit_iclog);
  			if (error)
  				return error;
  
  			/*
  			 * if we had a partial copy, we need to get more iclog
  			 * space but we don't want to increment the region
  			 * index because there is still more is this region to
  			 * write.
  			 *
  			 * If we completed writing this region, and we flushed
  			 * the iclog (indicated by resetting of the record
  			 * count), then we also need to get more log space. If
  			 * this was the last record, though, we are done and
  			 * can just return.
  			 */
  			if (partial_copy)
  				break;

  			if (++index == lv->lv_niovecs) {
  				lv = lv->lv_next;
  				index = 0;
  				if (lv)
  					vecp = lv->lv_iovecp;
  			}

  			if (record_cnt == 0) {

  				if (!lv)

  					return 0;
  				break;
  			}
  		}
  	}
  
  	ASSERT(len == 0);
  
  	xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
  	if (!commit_iclog)
  		return xlog_state_release_iclog(log, iclog);

  	ASSERT(flags & XLOG_COMMIT_TRANS);
  	*commit_iclog = iclog;
  	return 0;

  }

  
  
  /*****************************************************************************
   *
   *		State Machine functions
   *
   *****************************************************************************
   */
  
  /* Clean iclogs starting from the head.  This ordering must be
   * maintained, so an iclog doesn't become ACTIVE beyond one that
   * is SYNCING.  This is also required to maintain the notion that we use

   * a ordered wait queue to hold off would be writers to the log when every

   * iclog is trying to sync to disk.
   *
   * State Change: DIRTY -> ACTIVE
   */

  STATIC void

  xlog_state_clean_log(xlog_t *log)
  {
  	xlog_in_core_t	*iclog;
  	int changed = 0;
  
  	iclog = log->l_iclog;
  	do {
  		if (iclog->ic_state == XLOG_STATE_DIRTY) {
  			iclog->ic_state	= XLOG_STATE_ACTIVE;
  			iclog->ic_offset       = 0;

  			ASSERT(iclog->ic_callback == NULL);

  			/*
  			 * If the number of ops in this iclog indicate it just
  			 * contains the dummy transaction, we can
  			 * change state into IDLE (the second time around).
  			 * Otherwise we should change the state into
  			 * NEED a dummy.
  			 * We don't need to cover the dummy.
  			 */
  			if (!changed &&

  			   (be32_to_cpu(iclog->ic_header.h_num_logops) ==
  			   		XLOG_COVER_OPS)) {

  				changed = 1;
  			} else {
  				/*
  				 * We have two dirty iclogs so start over
  				 * This could also be num of ops indicates
  				 * this is not the dummy going out.
  				 */
  				changed = 2;
  			}
  			iclog->ic_header.h_num_logops = 0;
  			memset(iclog->ic_header.h_cycle_data, 0,
  			      sizeof(iclog->ic_header.h_cycle_data));
  			iclog->ic_header.h_lsn = 0;
  		} else if (iclog->ic_state == XLOG_STATE_ACTIVE)
  			/* do nothing */;
  		else
  			break;	/* stop cleaning */
  		iclog = iclog->ic_next;
  	} while (iclog != log->l_iclog);
  
  	/* log is locked when we are called */
  	/*
  	 * Change state for the dummy log recording.
  	 * We usually go to NEED. But we go to NEED2 if the changed indicates
  	 * we are done writing the dummy record.
  	 * If we are done with the second dummy recored (DONE2), then
  	 * we go to IDLE.
  	 */
  	if (changed) {
  		switch (log->l_covered_state) {
  		case XLOG_STATE_COVER_IDLE:
  		case XLOG_STATE_COVER_NEED:
  		case XLOG_STATE_COVER_NEED2:
  			log->l_covered_state = XLOG_STATE_COVER_NEED;
  			break;
  
  		case XLOG_STATE_COVER_DONE:
  			if (changed == 1)
  				log->l_covered_state = XLOG_STATE_COVER_NEED2;
  			else
  				log->l_covered_state = XLOG_STATE_COVER_NEED;
  			break;
  
  		case XLOG_STATE_COVER_DONE2:
  			if (changed == 1)
  				log->l_covered_state = XLOG_STATE_COVER_IDLE;
  			else
  				log->l_covered_state = XLOG_STATE_COVER_NEED;
  			break;
  
  		default:
  			ASSERT(0);
  		}
  	}
  }	/* xlog_state_clean_log */
  
  STATIC xfs_lsn_t
  xlog_get_lowest_lsn(
  	xlog_t		*log)
  {
  	xlog_in_core_t  *lsn_log;
  	xfs_lsn_t	lowest_lsn, lsn;
  
  	lsn_log = log->l_iclog;
  	lowest_lsn = 0;
  	do {
  	    if (!(lsn_log->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_DIRTY))) {

  		lsn = be64_to_cpu(lsn_log->ic_header.h_lsn);

  		if ((lsn && !lowest_lsn) ||
  		    (XFS_LSN_CMP(lsn, lowest_lsn) < 0)) {
  			lowest_lsn = lsn;
  		}
  	    }
  	    lsn_log = lsn_log->ic_next;
  	} while (lsn_log != log->l_iclog);

  	return lowest_lsn;

  }
  
  
  STATIC void
  xlog_state_do_callback(
  	xlog_t		*log,
  	int		aborted,
  	xlog_in_core_t	*ciclog)
  {
  	xlog_in_core_t	   *iclog;
  	xlog_in_core_t	   *first_iclog;	/* used to know when we've
  						 * processed all iclogs once */
  	xfs_log_callback_t *cb, *cb_next;
  	int		   flushcnt = 0;
  	xfs_lsn_t	   lowest_lsn;
  	int		   ioerrors;	/* counter: iclogs with errors */
  	int		   loopdidcallbacks; /* flag: inner loop did callbacks*/
  	int		   funcdidcallbacks; /* flag: function did callbacks */
  	int		   repeats;	/* for issuing console warnings if
  					 * looping too many times */

  	int		   wake = 0;

  	spin_lock(&log->l_icloglock);

  	first_iclog = iclog = log->l_iclog;
  	ioerrors = 0;
  	funcdidcallbacks = 0;
  	repeats = 0;
  
  	do {
  		/*
  		 * Scan all iclogs starting with the one pointed to by the
  		 * log.  Reset this starting point each time the log is
  		 * unlocked (during callbacks).
  		 *
  		 * Keep looping through iclogs until one full pass is made
  		 * without running any callbacks.
  		 */
  		first_iclog = log->l_iclog;
  		iclog = log->l_iclog;
  		loopdidcallbacks = 0;
  		repeats++;
  
  		do {
  
  			/* skip all iclogs in the ACTIVE & DIRTY states */
  			if (iclog->ic_state &
  			    (XLOG_STATE_ACTIVE|XLOG_STATE_DIRTY)) {
  				iclog = iclog->ic_next;
  				continue;
  			}
  
  			/*
  			 * Between marking a filesystem SHUTDOWN and stopping
  			 * the log, we do flush all iclogs to disk (if there
  			 * wasn't a log I/O error). So, we do want things to
  			 * go smoothly in case of just a SHUTDOWN  w/o a
  			 * LOG_IO_ERROR.
  			 */
  			if (!(iclog->ic_state & XLOG_STATE_IOERROR)) {
  				/*
  				 * Can only perform callbacks in order.  Since
  				 * this iclog is not in the DONE_SYNC/
  				 * DO_CALLBACK state, we skip the rest and
  				 * just try to clean up.  If we set our iclog
  				 * to DO_CALLBACK, we will not process it when
  				 * we retry since a previous iclog is in the
  				 * CALLBACK and the state cannot change since