xfs_trans.c

linux 内核源代码

/*
 * Copyright (c) 2000-2003,2005 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_dir2.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
#include "xfs_error.h"
#include "xfs_da_btree.h"
#include "xfs_bmap_btree.h"
#include "xfs_alloc_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_dir2_sf.h"
#include "xfs_attr_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_alloc.h"
#include "xfs_bmap.h"
#include "xfs_quota.h"
#include "xfs_trans_priv.h"
#include "xfs_trans_space.h"


STATIC void	xfs_trans_apply_sb_deltas(xfs_trans_t *);
STATIC uint	xfs_trans_count_vecs(xfs_trans_t *);
STATIC void	xfs_trans_fill_vecs(xfs_trans_t *, xfs_log_iovec_t *);
STATIC void	xfs_trans_uncommit(xfs_trans_t *, uint);
STATIC void	xfs_trans_committed(xfs_trans_t *, int);
STATIC void	xfs_trans_chunk_committed(xfs_log_item_chunk_t *, xfs_lsn_t, int);
STATIC void	xfs_trans_free(xfs_trans_t *);

kmem_zone_t	*xfs_trans_zone;


/*
 * Reservation functions here avoid a huge stack in xfs_trans_init
 * due to register overflow from temporaries in the calculations.
 */

STATIC uint
xfs_calc_write_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_WRITE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_itruncate_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_ITRUNCATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_rename_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_RENAME_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_link_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_LINK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_remove_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_REMOVE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_symlink_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_SYMLINK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_create_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_CREATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_mkdir_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_MKDIR_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_ifree_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_IFREE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_ichange_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_ICHANGE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_growdata_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_GROWDATA_LOG_RES(mp);
}

STATIC uint
xfs_calc_growrtalloc_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_GROWRTALLOC_LOG_RES(mp);
}

STATIC uint
xfs_calc_growrtzero_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_GROWRTZERO_LOG_RES(mp);
}

STATIC uint
xfs_calc_growrtfree_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_GROWRTFREE_LOG_RES(mp);
}

STATIC uint
xfs_calc_swrite_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_SWRITE_LOG_RES(mp);
}

STATIC uint
xfs_calc_writeid_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_WRITEID_LOG_RES(mp);
}

STATIC uint
xfs_calc_addafork_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_ADDAFORK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_attrinval_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_ATTRINVAL_LOG_RES(mp);
}

STATIC uint
xfs_calc_attrset_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_ATTRSET_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_attrrm_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_ATTRRM_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
}

STATIC uint
xfs_calc_clear_agi_bucket_reservation(xfs_mount_t *mp)
{
	return XFS_CALC_CLEAR_AGI_BUCKET_LOG_RES(mp);
}

/*
 * Initialize the precomputed transaction reservation values
 * in the mount structure.
 */
void
xfs_trans_init(
	xfs_mount_t	*mp)
{
	xfs_trans_reservations_t	*resp;

	resp = &(mp->m_reservations);
	resp->tr_write = xfs_calc_write_reservation(mp);
	resp->tr_itruncate = xfs_calc_itruncate_reservation(mp);
	resp->tr_rename = xfs_calc_rename_reservation(mp);
	resp->tr_link = xfs_calc_link_reservation(mp);
	resp->tr_remove = xfs_calc_remove_reservation(mp);
	resp->tr_symlink = xfs_calc_symlink_reservation(mp);
	resp->tr_create = xfs_calc_create_reservation(mp);
	resp->tr_mkdir = xfs_calc_mkdir_reservation(mp);
	resp->tr_ifree = xfs_calc_ifree_reservation(mp);
	resp->tr_ichange = xfs_calc_ichange_reservation(mp);
	resp->tr_growdata = xfs_calc_growdata_reservation(mp);
	resp->tr_swrite = xfs_calc_swrite_reservation(mp);
	resp->tr_writeid = xfs_calc_writeid_reservation(mp);
	resp->tr_addafork = xfs_calc_addafork_reservation(mp);
	resp->tr_attrinval = xfs_calc_attrinval_reservation(mp);
	resp->tr_attrset = xfs_calc_attrset_reservation(mp);
	resp->tr_attrrm = xfs_calc_attrrm_reservation(mp);
	resp->tr_clearagi = xfs_calc_clear_agi_bucket_reservation(mp);
	resp->tr_growrtalloc = xfs_calc_growrtalloc_reservation(mp);
	resp->tr_growrtzero = xfs_calc_growrtzero_reservation(mp);
	resp->tr_growrtfree = xfs_calc_growrtfree_reservation(mp);
}

/*
 * This routine is called to allocate a transaction structure.
 * The type parameter indicates the type of the transaction.  These
 * are enumerated in xfs_trans.h.
 *
 * Dynamically allocate the transaction structure from the transaction
 * zone, initialize it, and return it to the caller.
 */
xfs_trans_t *
xfs_trans_alloc(
	xfs_mount_t	*mp,
	uint		type)
{
	xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
	return _xfs_trans_alloc(mp, type);
}

xfs_trans_t *
_xfs_trans_alloc(
	xfs_mount_t	*mp,
	uint		type)
{
	xfs_trans_t	*tp;

	atomic_inc(&mp->m_active_trans);

	tp = kmem_zone_zalloc(xfs_trans_zone, KM_SLEEP);
	tp->t_magic = XFS_TRANS_MAGIC;
	tp->t_type = type;
	tp->t_mountp = mp;
	tp->t_items_free = XFS_LIC_NUM_SLOTS;
	tp->t_busy_free = XFS_LBC_NUM_SLOTS;
	XFS_LIC_INIT(&(tp->t_items));
	XFS_LBC_INIT(&(tp->t_busy));
	return tp;
}

/*
 * This is called to create a new transaction which will share the
 * permanent log reservation of the given transaction.  The remaining
 * unused block and rt extent reservations are also inherited.  This
 * implies that the original transaction is no longer allowed to allocate
 * blocks.  Locks and log items, however, are no inherited.  They must
 * be added to the new transaction explicitly.
 */
xfs_trans_t *
xfs_trans_dup(
	xfs_trans_t	*tp)
{
	xfs_trans_t	*ntp;

	ntp = kmem_zone_zalloc(xfs_trans_zone, KM_SLEEP);

	/*
	 * Initialize the new transaction structure.
	 */
	ntp->t_magic = XFS_TRANS_MAGIC;
	ntp->t_type = tp->t_type;
	ntp->t_mountp = tp->t_mountp;
	ntp->t_items_free = XFS_LIC_NUM_SLOTS;
	ntp->t_busy_free = XFS_LBC_NUM_SLOTS;
	XFS_LIC_INIT(&(ntp->t_items));
	XFS_LBC_INIT(&(ntp->t_busy));

	ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
	ASSERT(tp->t_ticket != NULL);

	ntp->t_flags = XFS_TRANS_PERM_LOG_RES | (tp->t_flags & XFS_TRANS_RESERVE);
	ntp->t_ticket = tp->t_ticket;
	ntp->t_blk_res = tp->t_blk_res - tp->t_blk_res_used;
	tp->t_blk_res = tp->t_blk_res_used;
	ntp->t_rtx_res = tp->t_rtx_res - tp->t_rtx_res_used;
	tp->t_rtx_res = tp->t_rtx_res_used;
	ntp->t_pflags = tp->t_pflags;

	XFS_TRANS_DUP_DQINFO(tp->t_mountp, tp, ntp);

	atomic_inc(&tp->t_mountp->m_active_trans);
	return ntp;
}

/*
 * This is called to reserve free disk blocks and log space for the
 * given transaction.  This must be done before allocating any resources
 * within the transaction.
 *
 * This will return ENOSPC if there are not enough blocks available.
 * It will sleep waiting for available log space.
 * The only valid value for the flags parameter is XFS_RES_LOG_PERM, which
 * is used by long running transactions.  If any one of the reservations
 * fails then they will all be backed out.
 *
 * This does not do quota reservations. That typically is done by the
 * caller afterwards.
 */
int
xfs_trans_reserve(
	xfs_trans_t	*tp,
	uint		blocks,
	uint		logspace,
	uint		rtextents,
	uint		flags,
	uint		logcount)
{
	int		log_flags;
	int		error = 0;
	int		rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0;

	/* Mark this thread as being in a transaction */
	current_set_flags_nested(&tp->t_pflags, PF_FSTRANS);

	/*
	 * Attempt to reserve the needed disk blocks by decrementing
	 * the number needed from the number available.  This will
	 * fail if the count would go below zero.
	 */
	if (blocks > 0) {
		error = xfs_mod_incore_sb(tp->t_mountp, XFS_SBS_FDBLOCKS,
					  -((int64_t)blocks), rsvd);
		if (error != 0) {
			current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
			return (XFS_ERROR(ENOSPC));
		}
		tp->t_blk_res += blocks;
	}

	/*
	 * Reserve the log space needed for this transaction.
	 */
	if (logspace > 0) {
		ASSERT((tp->t_log_res == 0) || (tp->t_log_res == logspace));
		ASSERT((tp->t_log_count == 0) ||
			(tp->t_log_count == logcount));
		if (flags & XFS_TRANS_PERM_LOG_RES) {
			log_flags = XFS_LOG_PERM_RESERV;
			tp->t_flags |= XFS_TRANS_PERM_LOG_RES;
		} else {
			ASSERT(tp->t_ticket == NULL);
			ASSERT(!(tp->t_flags & XFS_TRANS_PERM_LOG_RES));
			log_flags = 0;
		}

		error = xfs_log_reserve(tp->t_mountp, logspace, logcount,
					&tp->t_ticket,
					XFS_TRANSACTION, log_flags, tp->t_type);
		if (error) {
			goto undo_blocks;
		}
		tp->t_log_res = logspace;
		tp->t_log_count = logcount;
	}

	/*
	 * Attempt to reserve the needed realtime extents by decrementing
	 * the number needed from the number available.  This will
	 * fail if the count would go below zero.
	 */
	if (rtextents > 0) {
		error = xfs_mod_incore_sb(tp->t_mountp, XFS_SBS_FREXTENTS,
					  -((int64_t)rtextents), rsvd);
		if (error) {
			error = XFS_ERROR(ENOSPC);
			goto undo_log;
		}
		tp->t_rtx_res += rtextents;
	}

	return 0;

	/*
	 * Error cases jump to one of these labels to undo any
	 * reservations which have already been performed.
	 */
undo_log:
	if (logspace > 0) {
		if (flags & XFS_TRANS_PERM_LOG_RES) {
			log_flags = XFS_LOG_REL_PERM_RESERV;
		} else {
			log_flags = 0;
		}
		xfs_log_done(tp->t_mountp, tp->t_ticket, NULL, log_flags);
		tp->t_ticket = NULL;
		tp->t_log_res = 0;
		tp->t_flags &= ~XFS_TRANS_PERM_LOG_RES;
	}

undo_blocks:
	if (blocks > 0) {
		(void) xfs_mod_incore_sb(tp->t_mountp, XFS_SBS_FDBLOCKS,
					 (int64_t)blocks, rsvd);
		tp->t_blk_res = 0;
	}

	current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);

	return error;
}


/*
 * Record the indicated change to the given field for application
 * to the file system's superblock when the transaction commits.
 * For now, just store the change in the transaction structure.
 *
 * Mark the transaction structure to indicate that the superblock
 * needs to be updated before committing.
 *
 * Because we may not be keeping track of allocated/free inodes and
 * used filesystem blocks in the superblock, we do not mark the
 * superblock dirty in this transaction if we modify these fields.
 * We still need to update the transaction deltas so that they get
 * applied to the incore superblock, but we don't want them to
 * cause the superblock to get locked and logged if these are the
 * only fields in the superblock that the transaction modifies.
 */
void
xfs_trans_mod_sb(
	xfs_trans_t	*tp,
	uint		field,
	int64_t		delta)
{
	uint32_t	flags = (XFS_TRANS_DIRTY|XFS_TRANS_SB_DIRTY);
	xfs_mount_t	*mp = tp->t_mountp;

	switch (field) {
	case XFS_TRANS_SB_ICOUNT:
		tp->t_icount_delta += delta;
		if (xfs_sb_version_haslazysbcount(&mp->m_sb))
			flags &= ~XFS_TRANS_SB_DIRTY;
		break;
	case XFS_TRANS_SB_IFREE:
		tp->t_ifree_delta += delta;
		if (xfs_sb_version_haslazysbcount(&mp->m_sb))
			flags &= ~XFS_TRANS_SB_DIRTY;
		break;
	case XFS_TRANS_SB_FDBLOCKS:
		/*
		 * Track the number of blocks allocated in the
		 * transaction.  Make sure it does not exceed the
		 * number reserved.
		 */