xfs_inode.c

linux 内核源代码

		XFS_BMAP_INIT(&free_list, &first_block);
		error = XFS_BUNMAPI(mp, ntp, &ip->i_iocore,
				    first_unmap_block, unmap_len,
				    XFS_BMAPI_AFLAG(fork) |
				      (sync ? 0 : XFS_BMAPI_ASYNC),
				    XFS_ITRUNC_MAX_EXTENTS,
				    &first_block, &free_list,
				    NULL, &done);
		if (error) {
			/*
			 * If the bunmapi call encounters an error,
			 * return to the caller where the transaction
			 * can be properly aborted.  We just need to
			 * make sure we're not holding any resources
			 * that we were not when we came in.
			 */
			xfs_bmap_cancel(&free_list);
			return error;
		}

		/*
		 * Duplicate the transaction that has the permanent
		 * reservation and commit the old transaction.
		 */
		error = xfs_bmap_finish(tp, &free_list, &committed);
		ntp = *tp;
		if (error) {
			/*
			 * If the bmap finish call encounters an error,
			 * return to the caller where the transaction
			 * can be properly aborted.  We just need to
			 * make sure we're not holding any resources
			 * that we were not when we came in.
			 *
			 * Aborting from this point might lose some
			 * blocks in the file system, but oh well.
			 */
			xfs_bmap_cancel(&free_list);
			if (committed) {
				/*
				 * If the passed in transaction committed
				 * in xfs_bmap_finish(), then we want to
				 * add the inode to this one before returning.
				 * This keeps things simple for the higher
				 * level code, because it always knows that
				 * the inode is locked and held in the
				 * transaction that returns to it whether
				 * errors occur or not.  We don't mark the
				 * inode dirty so that this transaction can
				 * be easily aborted if possible.
				 */
				xfs_trans_ijoin(ntp, ip,
					XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
				xfs_trans_ihold(ntp, ip);
			}
			return error;
		}

		if (committed) {
			/*
			 * The first xact was committed,
			 * so add the inode to the new one.
			 * Mark it dirty so it will be logged
			 * and moved forward in the log as
			 * part of every commit.
			 */
			xfs_trans_ijoin(ntp, ip,
					XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
			xfs_trans_ihold(ntp, ip);
			xfs_trans_log_inode(ntp, ip, XFS_ILOG_CORE);
		}
		ntp = xfs_trans_dup(ntp);
		(void) xfs_trans_commit(*tp, 0);
		*tp = ntp;
		error = xfs_trans_reserve(ntp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0,
					  XFS_TRANS_PERM_LOG_RES,
					  XFS_ITRUNCATE_LOG_COUNT);
		/*
		 * Add the inode being truncated to the next chained
		 * transaction.
		 */
		xfs_trans_ijoin(ntp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
		xfs_trans_ihold(ntp, ip);
		if (error)
			return (error);
	}
	/*
	 * Only update the size in the case of the data fork, but
	 * always re-log the inode so that our permanent transaction
	 * can keep on rolling it forward in the log.
	 */
	if (fork == XFS_DATA_FORK) {
		xfs_isize_check(mp, ip, new_size);
		/*
		 * If we are not changing the file size then do
		 * not update the on-disk file size - we may be
		 * called from xfs_inactive_free_eofblocks().  If we
		 * update the on-disk file size and then the system
		 * crashes before the contents of the file are
		 * flushed to disk then the files may be full of
		 * holes (ie NULL files bug).
		 */
		if (ip->i_size != new_size) {
			ip->i_d.di_size = new_size;
			ip->i_size = new_size;
		}
	}
	xfs_trans_log_inode(ntp, ip, XFS_ILOG_CORE);
	ASSERT((new_size != 0) ||
	       (fork == XFS_ATTR_FORK) ||
	       (ip->i_delayed_blks == 0));
	ASSERT((new_size != 0) ||
	       (fork == XFS_ATTR_FORK) ||
	       (ip->i_d.di_nextents == 0));
	xfs_itrunc_trace(XFS_ITRUNC_FINISH2, ip, 0, new_size, 0, 0);
	return 0;
}


/*
 * xfs_igrow_start
 *
 * Do the first part of growing a file: zero any data in the last
 * block that is beyond the old EOF.  We need to do this before
 * the inode is joined to the transaction to modify the i_size.
 * That way we can drop the inode lock and call into the buffer
 * cache to get the buffer mapping the EOF.
 */
int
xfs_igrow_start(
	xfs_inode_t	*ip,
	xfs_fsize_t	new_size,
	cred_t		*credp)
{
	int		error;

	ASSERT(ismrlocked(&(ip->i_lock), MR_UPDATE) != 0);
	ASSERT(ismrlocked(&(ip->i_iolock), MR_UPDATE) != 0);
	ASSERT(new_size > ip->i_size);

	/*
	 * Zero any pages that may have been created by
	 * xfs_write_file() beyond the end of the file
	 * and any blocks between the old and new file sizes.
	 */
	error = xfs_zero_eof(XFS_ITOV(ip), &ip->i_iocore, new_size,
			     ip->i_size);
	return error;
}

/*
 * xfs_igrow_finish
 *
 * This routine is called to extend the size of a file.
 * The inode must have both the iolock and the ilock locked
 * for update and it must be a part of the current transaction.
 * The xfs_igrow_start() function must have been called previously.
 * If the change_flag is not zero, the inode change timestamp will
 * be updated.
 */
void
xfs_igrow_finish(
	xfs_trans_t	*tp,
	xfs_inode_t	*ip,
	xfs_fsize_t	new_size,
	int		change_flag)
{
	ASSERT(ismrlocked(&(ip->i_lock), MR_UPDATE) != 0);
	ASSERT(ismrlocked(&(ip->i_iolock), MR_UPDATE) != 0);
	ASSERT(ip->i_transp == tp);
	ASSERT(new_size > ip->i_size);

	/*
	 * Update the file size.  Update the inode change timestamp
	 * if change_flag set.
	 */
	ip->i_d.di_size = new_size;
	ip->i_size = new_size;
	if (change_flag)
		xfs_ichgtime(ip, XFS_ICHGTIME_CHG);
	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);

}


/*
 * This is called when the inode's link count goes to 0.
 * We place the on-disk inode on a list in the AGI.  It
 * will be pulled from this list when the inode is freed.
 */
int
xfs_iunlink(
	xfs_trans_t	*tp,
	xfs_inode_t	*ip)
{
	xfs_mount_t	*mp;
	xfs_agi_t	*agi;
	xfs_dinode_t	*dip;
	xfs_buf_t	*agibp;
	xfs_buf_t	*ibp;
	xfs_agnumber_t	agno;
	xfs_daddr_t	agdaddr;
	xfs_agino_t	agino;
	short		bucket_index;
	int		offset;
	int		error;
	int		agi_ok;

	ASSERT(ip->i_d.di_nlink == 0);
	ASSERT(ip->i_d.di_mode != 0);
	ASSERT(ip->i_transp == tp);

	mp = tp->t_mountp;

	agno = XFS_INO_TO_AGNO(mp, ip->i_ino);
	agdaddr = XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp));

	/*
	 * Get the agi buffer first.  It ensures lock ordering
	 * on the list.
	 */
	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, agdaddr,
				   XFS_FSS_TO_BB(mp, 1), 0, &agibp);
	if (error)
		return error;

	/*
	 * Validate the magic number of the agi block.
	 */
	agi = XFS_BUF_TO_AGI(agibp);
	agi_ok =
		be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC &&
		XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum));
	if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IUNLINK,
			XFS_RANDOM_IUNLINK))) {
		XFS_CORRUPTION_ERROR("xfs_iunlink", XFS_ERRLEVEL_LOW, mp, agi);
		xfs_trans_brelse(tp, agibp);
		return XFS_ERROR(EFSCORRUPTED);
	}
	/*
	 * Get the index into the agi hash table for the
	 * list this inode will go on.
	 */
	agino = XFS_INO_TO_AGINO(mp, ip->i_ino);
	ASSERT(agino != 0);
	bucket_index = agino % XFS_AGI_UNLINKED_BUCKETS;
	ASSERT(agi->agi_unlinked[bucket_index]);
	ASSERT(be32_to_cpu(agi->agi_unlinked[bucket_index]) != agino);

	error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
	if (error)
		return error;

	/*
	 * Clear the on-disk di_nlink. This is to prevent xfs_bulkstat
	 * from picking up this inode when it is reclaimed (its incore state
	 * initialzed but not flushed to disk yet). The in-core di_nlink is
	 * already cleared in xfs_droplink() and a corresponding transaction
	 * logged. The hack here just synchronizes the in-core to on-disk
	 * di_nlink value in advance before the actual inode sync to disk.
	 * This is OK because the inode is already unlinked and would never
	 * change its di_nlink again for this inode generation.
	 * This is a temporary hack that would require a proper fix
	 * in the future.
	 */
	dip->di_core.di_nlink = 0;

	if (be32_to_cpu(agi->agi_unlinked[bucket_index]) != NULLAGINO) {
		/*
		 * There is already another inode in the bucket we need
		 * to add ourselves to.  Add us at the front of the list.
		 * Here we put the head pointer into our next pointer,
		 * and then we fall through to point the head at us.
		 */
		ASSERT(be32_to_cpu(dip->di_next_unlinked) == NULLAGINO);
		/* both on-disk, don't endian flip twice */
		dip->di_next_unlinked = agi->agi_unlinked[bucket_index];
		offset = ip->i_boffset +
			offsetof(xfs_dinode_t, di_next_unlinked);
		xfs_trans_inode_buf(tp, ibp);
		xfs_trans_log_buf(tp, ibp, offset,
				  (offset + sizeof(xfs_agino_t) - 1));
		xfs_inobp_check(mp, ibp);
	}

	/*
	 * Point the bucket head pointer at the inode being inserted.
	 */
	ASSERT(agino != 0);
	agi->agi_unlinked[bucket_index] = cpu_to_be32(agino);
	offset = offsetof(xfs_agi_t, agi_unlinked) +
		(sizeof(xfs_agino_t) * bucket_index);
	xfs_trans_log_buf(tp, agibp, offset,
			  (offset + sizeof(xfs_agino_t) - 1));
	return 0;
}

/*
 * Pull the on-disk inode from the AGI unlinked list.
 */
STATIC int
xfs_iunlink_remove(
	xfs_trans_t	*tp,
	xfs_inode_t	*ip)
{
	xfs_ino_t	next_ino;
	xfs_mount_t	*mp;
	xfs_agi_t	*agi;
	xfs_dinode_t	*dip;
	xfs_buf_t	*agibp;
	xfs_buf_t	*ibp;
	xfs_agnumber_t	agno;
	xfs_daddr_t	agdaddr;
	xfs_agino_t	agino;
	xfs_agino_t	next_agino;
	xfs_buf_t	*last_ibp;
	xfs_dinode_t	*last_dip = NULL;
	short		bucket_index;
	int		offset, last_offset = 0;
	int		error;
	int		agi_ok;

	/*
	 * First pull the on-disk inode from the AGI unlinked list.
	 */
	mp = tp->t_mountp;

	agno = XFS_INO_TO_AGNO(mp, ip->i_ino);
	agdaddr = XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp));

	/*
	 * Get the agi buffer first.  It ensures lock ordering
	 * on the list.
	 */
	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, agdaddr,
				   XFS_FSS_TO_BB(mp, 1), 0, &agibp);
	if (error) {
		cmn_err(CE_WARN,
			"xfs_iunlink_remove: xfs_trans_read_buf()  returned an error %d on %s.  Returning error.",
			error, mp->m_fsname);
		return error;
	}
	/*
	 * Validate the magic number of the agi block.
	 */
	agi = XFS_BUF_TO_AGI(agibp);
	agi_ok =
		be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC &&
		XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum));
	if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IUNLINK_REMOVE,
			XFS_RANDOM_IUNLINK_REMOVE))) {
		XFS_CORRUPTION_ERROR("xfs_iunlink_remove", XFS_ERRLEVEL_LOW,
				     mp, agi);
		xfs_trans_brelse(tp, agibp);
		cmn_err(CE_WARN,
			"xfs_iunlink_remove: XFS_TEST_ERROR()  returned an error on %s.  Returning EFSCORRUPTED.",
			 mp->m_fsname);
		return XFS_ERROR(EFSCORRUPTED);
	}
	/*
	 * Get the index into the agi hash table for the
	 * list this inode will go on.
	 */
	agino = XFS_INO_TO_AGINO(mp, ip->i_ino);
	ASSERT(agino != 0);
	bucket_index = agino % XFS_AGI_UNLINKED_BUCKETS;
	ASSERT(be32_to_cpu(agi->agi_unlinked[bucket_index]) != NULLAGINO);
	ASSERT(agi->agi_unlinked[bucket_index]);

	if (be32_to_cpu(agi->agi_unlinked[bucket_index]) == agino) {
		/*
		 * We're at the head of the list.  Get the inode's
		 * on-disk buffer to see if there is anyone after us
		 * on the list.  Only modify our next pointer if it
		 * is not already NULLAGINO.  This saves us the overhead
		 * of dealing with the buffer when there is no need to
		 * change it.
		 */
		error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
		if (error) {
			cmn_err(CE_WARN,
				"xfs_iunlink_remove: xfs_itobp()  returned an error %d on %s.  Returning error.",
				error, mp->m_fsname);
			return error;
		}
		next_agino = be32_to_cpu(dip->di_next_unlinked);
		ASSERT(next_agino != 0);
		if (next_agino != NULLAGINO) {
			dip->di_next_unlinked = cpu_to_be32(NULLAGINO);
			offset = ip->i_boffset +
				offsetof(xfs_dinode_t, di_next_unlinked);
			xfs_trans_inode_buf(tp, ibp);
			xfs_trans_log_buf(tp, ibp, offset,
					  (offset + sizeof(xfs_agino_t) - 1));
			xfs_inobp_check(mp, ibp);
		} else {
			xfs_trans_brelse(tp, ibp);
		}
		/*
		 * Point the bucket head pointer at the next inode.
		 */
		ASSERT(next_agino != 0);
		ASSERT(next_agino != agino);
		agi->agi_unlinked[bucket_index] = cpu_to_be32(next_agino);
		offset = offsetof(xfs_agi_t, agi_unlinked) +
			(sizeof(xfs_agino_t) * bucket_index);
		xfs_trans_log_buf(tp, agibp, offset,
				  (offset + sizeof(xfs_agino_t) - 1));
	} else {
		/*
		 * We need to search the list for the inode being freed.
		 */
		next_agino = be32_to_cpu(agi->agi_unlinked[bucket_index]);
		last_ibp = NULL;
		while (next_agino != agino) {
			/*
			 * If the last inode wasn't the one pointing to
			 * us, then release its buffer since we're not
			 * going to do anything with it.
			 */
			if (last_ibp != NULL) {
				xfs_trans_brelse(tp, last_ibp);
			}