示例#1
0
STATIC struct xfs_btree_cur *
xfs_inobt_dup_cursor(
	struct xfs_btree_cur	*cur)
{
	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
			cur->bc_private.a.agbp, cur->bc_private.a.agno);
}
/*
 * Allocate new inodes in the allocation group specified by agbp.
 * Return 0 for success, else error code.
 */
STATIC int				/* error code or 0 */
xfs_ialloc_ag_alloc(
	xfs_trans_t	*tp,		/* transaction pointer */
	xfs_buf_t	*agbp,		/* alloc group buffer */
	int		*alloc)
{
	xfs_agi_t	*agi;		/* allocation group header */
	xfs_alloc_arg_t	args;		/* allocation argument structure */
	xfs_btree_cur_t	*cur;		/* inode btree cursor */
	xfs_agnumber_t	agno;
	int		error;
	int		i;
	xfs_agino_t	newino;		/* new first inode's number */
	xfs_agino_t	newlen;		/* new number of inodes */
	xfs_agino_t	thisino;	/* current inode number, for loop */
	int		isaligned = 0;	/* inode allocation at stripe unit */
					/* boundary */
	struct xfs_perag *pag;

	args.tp = tp;
	args.mp = tp->t_mountp;

	/*
	 * Locking will ensure that we don't have two callers in here
	 * at one time.
	 */
	newlen = XFS_IALLOC_INODES(args.mp);
	if (args.mp->m_maxicount &&
	    args.mp->m_sb.sb_icount + newlen > args.mp->m_maxicount)
		return XFS_ERROR(ENOSPC);
	args.minlen = args.maxlen = XFS_IALLOC_BLOCKS(args.mp);
	/*
	 * First try to allocate inodes contiguous with the last-allocated
	 * chunk of inodes.  If the filesystem is striped, this will fill
	 * an entire stripe unit with inodes.
 	 */
	agi = XFS_BUF_TO_AGI(agbp);
	newino = be32_to_cpu(agi->agi_newino);
	agno = be32_to_cpu(agi->agi_seqno);
	args.agbno = XFS_AGINO_TO_AGBNO(args.mp, newino) +
			XFS_IALLOC_BLOCKS(args.mp);
	if (likely(newino != NULLAGINO &&
		  (args.agbno < be32_to_cpu(agi->agi_length)))) {
		args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
		args.type = XFS_ALLOCTYPE_THIS_BNO;
		args.mod = args.total = args.wasdel = args.isfl =
			args.userdata = args.minalignslop = 0;
		args.prod = 1;

		/*
		 * We need to take into account alignment here to ensure that
		 * we don't modify the free list if we fail to have an exact
		 * block. If we don't have an exact match, and every oher
		 * attempt allocation attempt fails, we'll end up cancelling
		 * a dirty transaction and shutting down.
		 *
		 * For an exact allocation, alignment must be 1,
		 * however we need to take cluster alignment into account when
		 * fixing up the freelist. Use the minalignslop field to
		 * indicate that extra blocks might be required for alignment,
		 * but not to use them in the actual exact allocation.
		 */
		args.alignment = 1;
		args.minalignslop = xfs_ialloc_cluster_alignment(&args) - 1;

		/* Allow space for the inode btree to split. */
		args.minleft = args.mp->m_in_maxlevels - 1;
		if ((error = xfs_alloc_vextent(&args)))
			return error;
	} else
		args.fsbno = NULLFSBLOCK;

	if (unlikely(args.fsbno == NULLFSBLOCK)) {
		/*
		 * Set the alignment for the allocation.
		 * If stripe alignment is turned on then align at stripe unit
		 * boundary.
		 * If the cluster size is smaller than a filesystem block
		 * then we're doing I/O for inodes in filesystem block size
		 * pieces, so don't need alignment anyway.
		 */
		isaligned = 0;
		if (args.mp->m_sinoalign) {
			ASSERT(!(args.mp->m_flags & XFS_MOUNT_NOALIGN));
			args.alignment = args.mp->m_dalign;
			isaligned = 1;
		} else
			args.alignment = xfs_ialloc_cluster_alignment(&args);
		/*
		 * Need to figure out where to allocate the inode blocks.
		 * Ideally they should be spaced out through the a.g.
		 * For now, just allocate blocks up front.
		 */
		args.agbno = be32_to_cpu(agi->agi_root);
		args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
		/*
		 * Allocate a fixed-size extent of inodes.
		 */
		args.type = XFS_ALLOCTYPE_NEAR_BNO;
		args.mod = args.total = args.wasdel = args.isfl =
			args.userdata = args.minalignslop = 0;
		args.prod = 1;
		/*
		 * Allow space for the inode btree to split.
		 */
		args.minleft = args.mp->m_in_maxlevels - 1;
		if ((error = xfs_alloc_vextent(&args)))
			return error;
	}

	/*
	 * If stripe alignment is turned on, then try again with cluster
	 * alignment.
	 */
	if (isaligned && args.fsbno == NULLFSBLOCK) {
		args.type = XFS_ALLOCTYPE_NEAR_BNO;
		args.agbno = be32_to_cpu(agi->agi_root);
		args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
		args.alignment = xfs_ialloc_cluster_alignment(&args);
		if ((error = xfs_alloc_vextent(&args)))
			return error;
	}

	if (args.fsbno == NULLFSBLOCK) {
		*alloc = 0;
		return 0;
	}
	ASSERT(args.len == args.minlen);

	/*
	 * Stamp and write the inode buffers.
	 *
	 * Seed the new inode cluster with a random generation number. This
	 * prevents short-term reuse of generation numbers if a chunk is
	 * freed and then immediately reallocated. We use random numbers
	 * rather than a linear progression to prevent the next generation
	 * number from being easily guessable.
	 */
	error = xfs_ialloc_inode_init(args.mp, tp, agno, args.agbno,
			args.len, prandom_u32());

	if (error)
		return error;
	/*
	 * Convert the results.
	 */
	newino = XFS_OFFBNO_TO_AGINO(args.mp, args.agbno, 0);
	be32_add_cpu(&agi->agi_count, newlen);
	be32_add_cpu(&agi->agi_freecount, newlen);
	pag = xfs_perag_get(args.mp, agno);
	pag->pagi_freecount += newlen;
	xfs_perag_put(pag);
	agi->agi_newino = cpu_to_be32(newino);

	/*
	 * Insert records describing the new inode chunk into the btree.
	 */
	cur = xfs_inobt_init_cursor(args.mp, tp, agbp, agno);
	for (thisino = newino;
	     thisino < newino + newlen;
	     thisino += XFS_INODES_PER_CHUNK) {
		cur->bc_rec.i.ir_startino = thisino;
		cur->bc_rec.i.ir_freecount = XFS_INODES_PER_CHUNK;
		cur->bc_rec.i.ir_free = XFS_INOBT_ALL_FREE;
		error = xfs_btree_lookup(cur, XFS_LOOKUP_EQ, &i);
		if (error) {
			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
			return error;
		}
		ASSERT(i == 0);
		error = xfs_btree_insert(cur, &i);
		if (error) {
			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
			return error;
		}
		ASSERT(i == 1);
	}
	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
	/*
	 * Log allocation group header fields
	 */
	xfs_ialloc_log_agi(tp, agbp,
		XFS_AGI_COUNT | XFS_AGI_FREECOUNT | XFS_AGI_NEWINO);
	/*
	 * Modify/log superblock values for inode count and inode free count.
	 */
	xfs_trans_mod_sb(tp, XFS_TRANS_SB_ICOUNT, (long)newlen);
	xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, (long)newlen);
	*alloc = 1;
	return 0;
}
/*
 * Allocate an inode on disk.
 * Mode is used to tell whether the new inode will need space, and whether
 * it is a directory.
 *
 * The arguments IO_agbp and alloc_done are defined to work within
 * the constraint of one allocation per transaction.
 * xfs_dialloc() is designed to be called twice if it has to do an
 * allocation to make more free inodes.  On the first call,
 * IO_agbp should be set to NULL. If an inode is available,
 * i.e., xfs_dialloc() did not need to do an allocation, an inode
 * number is returned.  In this case, IO_agbp would be set to the
 * current ag_buf and alloc_done set to false.
 * If an allocation needed to be done, xfs_dialloc would return
 * the current ag_buf in IO_agbp and set alloc_done to true.
 * The caller should then commit the current transaction, allocate a new
 * transaction, and call xfs_dialloc() again, passing in the previous
 * value of IO_agbp.  IO_agbp should be held across the transactions.
 * Since the agbp is locked across the two calls, the second call is
 * guaranteed to have a free inode available.
 *
 * Once we successfully pick an inode its number is returned and the
 * on-disk data structures are updated.  The inode itself is not read
 * in, since doing so would break ordering constraints with xfs_reclaim.
 */
int
xfs_dialloc(
	xfs_trans_t	*tp,		/* transaction pointer */
	xfs_ino_t	parent,		/* parent inode (directory) */
	umode_t		mode,		/* mode bits for new inode */
	int		okalloc,	/* ok to allocate more space */
	xfs_buf_t	**IO_agbp,	/* in/out ag header's buffer */
	boolean_t	*alloc_done,	/* true if we needed to replenish
					   inode freelist */
	xfs_ino_t	*inop)		/* inode number allocated */
{
	xfs_agnumber_t	agcount;	/* number of allocation groups */
	xfs_buf_t	*agbp;		/* allocation group header's buffer */
	xfs_agnumber_t	agno;		/* allocation group number */
	xfs_agi_t	*agi;		/* allocation group header structure */
	xfs_btree_cur_t	*cur;		/* inode allocation btree cursor */
	int		error;		/* error return value */
	int		i;		/* result code */
	int		ialloced;	/* inode allocation status */
	int		noroom = 0;	/* no space for inode blk allocation */
	xfs_ino_t	ino;		/* fs-relative inode to be returned */
	/* REFERENCED */
	int		j;		/* result code */
	xfs_mount_t	*mp;		/* file system mount structure */
	int		offset;		/* index of inode in chunk */
	xfs_agino_t	pagino;		/* parent's AG relative inode # */
	xfs_agnumber_t	pagno;		/* parent's AG number */
	xfs_inobt_rec_incore_t rec;	/* inode allocation record */
	xfs_agnumber_t	tagno;		/* testing allocation group number */
	xfs_btree_cur_t	*tcur;		/* temp cursor */
	xfs_inobt_rec_incore_t trec;	/* temp inode allocation record */
	struct xfs_perag *pag;


	if (*IO_agbp == NULL) {
		/*
		 * We do not have an agbp, so select an initial allocation
		 * group for inode allocation.
		 */
		agbp = xfs_ialloc_ag_select(tp, parent, mode, okalloc);
		/*
		 * Couldn't find an allocation group satisfying the
		 * criteria, give up.
		 */
		if (!agbp) {
			*inop = NULLFSINO;
			return 0;
		}
		agi = XFS_BUF_TO_AGI(agbp);
		ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
	} else {
		/*
		 * Continue where we left off before.  In this case, we
		 * know that the allocation group has free inodes.
		 */
		agbp = *IO_agbp;
		agi = XFS_BUF_TO_AGI(agbp);
		ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
		ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
	}
	mp = tp->t_mountp;
	agcount = mp->m_sb.sb_agcount;
	agno = be32_to_cpu(agi->agi_seqno);
	tagno = agno;
	pagno = XFS_INO_TO_AGNO(mp, parent);
	pagino = XFS_INO_TO_AGINO(mp, parent);

	/*
	 * If we have already hit the ceiling of inode blocks then clear
	 * okalloc so we scan all available agi structures for a free
	 * inode.
	 */

	if (mp->m_maxicount &&
	    mp->m_sb.sb_icount + XFS_IALLOC_INODES(mp) > mp->m_maxicount) {
		noroom = 1;
		okalloc = 0;
	}

	/*
	 * Loop until we find an allocation group that either has free inodes
	 * or in which we can allocate some inodes.  Iterate through the
	 * allocation groups upward, wrapping at the end.
	 */
	*alloc_done = B_FALSE;
	while (!agi->agi_freecount) {
		/*
		 * Don't do anything if we're not supposed to allocate
		 * any blocks, just go on to the next ag.
		 */
		if (okalloc) {
			/*
			 * Try to allocate some new inodes in the allocation
			 * group.
			 */
			if ((error = xfs_ialloc_ag_alloc(tp, agbp, &ialloced))) {
				xfs_trans_brelse(tp, agbp);
				if (error == ENOSPC) {
					*inop = NULLFSINO;
					return 0;
				} else
					return error;
			}
			if (ialloced) {
				/*
				 * We successfully allocated some inodes, return
				 * the current context to the caller so that it
				 * can commit the current transaction and call
				 * us again where we left off.
				 */
				ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
				*alloc_done = B_TRUE;
				*IO_agbp = agbp;
				*inop = NULLFSINO;
				return 0;
			}
		}
		/*
		 * If it failed, give up on this ag.
		 */
		xfs_trans_brelse(tp, agbp);
		/*
		 * Go on to the next ag: get its ag header.
		 */
nextag:
		if (++tagno == agcount)
			tagno = 0;
		if (tagno == agno) {
			*inop = NULLFSINO;
			return noroom ? ENOSPC : 0;
		}
		pag = xfs_perag_get(mp, tagno);
		if (pag->pagi_inodeok == 0) {
			xfs_perag_put(pag);
			goto nextag;
		}
		error = xfs_ialloc_read_agi(mp, tp, tagno, &agbp);
		xfs_perag_put(pag);
		if (error)
			goto nextag;
		agi = XFS_BUF_TO_AGI(agbp);
		ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
	}
	/*
	 * Here with an allocation group that has a free inode.
	 * Reset agno since we may have chosen a new ag in the
	 * loop above.
	 */
	agno = tagno;
	*IO_agbp = NULL;
	pag = xfs_perag_get(mp, agno);

 restart_pagno:
	cur = xfs_inobt_init_cursor(mp, tp, agbp, be32_to_cpu(agi->agi_seqno));
	/*
	 * If pagino is 0 (this is the root inode allocation) use newino.
	 * This must work because we've just allocated some.
	 */
	if (!pagino)
		pagino = be32_to_cpu(agi->agi_newino);

	error = xfs_check_agi_freecount(cur, agi);
	if (error)
		goto error0;

	/*
	 * If in the same AG as the parent, try to get near the parent.
	 */
	if (pagno == agno) {
		int		doneleft;	/* done, to the left */
		int		doneright;	/* done, to the right */
		int		searchdistance = 10;

		error = xfs_inobt_lookup(cur, pagino, XFS_LOOKUP_LE, &i);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);

		error = xfs_inobt_get_rec(cur, &rec, &j);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);

		if (rec.ir_freecount > 0) {
			/*
			 * Found a free inode in the same chunk
			 * as the parent, done.
			 */
			goto alloc_inode;
		}


		/*
		 * In the same AG as parent, but parent's chunk is full.
		 */

		/* duplicate the cursor, search left & right simultaneously */
		error = xfs_btree_dup_cursor(cur, &tcur);
		if (error)
			goto error0;

		/*
		 * Skip to last blocks looked up if same parent inode.
		 */
		if (pagino != NULLAGINO &&
		    pag->pagl_pagino == pagino &&
		    pag->pagl_leftrec != NULLAGINO &&
		    pag->pagl_rightrec != NULLAGINO) {
			error = xfs_ialloc_get_rec(tcur, pag->pagl_leftrec,
						   &trec, &doneleft, 1);
			if (error)
				goto error1;

			error = xfs_ialloc_get_rec(cur, pag->pagl_rightrec,
						   &rec, &doneright, 0);
			if (error)
				goto error1;
		} else {
			/* search left with tcur, back up 1 record */
			error = xfs_ialloc_next_rec(tcur, &trec, &doneleft, 1);
			if (error)
				goto error1;

			/* search right with cur, go forward 1 record. */
			error = xfs_ialloc_next_rec(cur, &rec, &doneright, 0);
			if (error)
				goto error1;
		}

		/*
		 * Loop until we find an inode chunk with a free inode.
		 */
		while (!doneleft || !doneright) {
			int	useleft;  /* using left inode chunk this time */

			if (!--searchdistance) {
				/*
				 * Not in range - save last search
				 * location and allocate a new inode
				 */
				xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
				pag->pagl_leftrec = trec.ir_startino;
				pag->pagl_rightrec = rec.ir_startino;
				pag->pagl_pagino = pagino;
				goto newino;
			}

			/* figure out the closer block if both are valid. */
			if (!doneleft && !doneright) {
				useleft = pagino -
				 (trec.ir_startino + XFS_INODES_PER_CHUNK - 1) <
				  rec.ir_startino - pagino;
			} else {
				useleft = !doneleft;
			}

			/* free inodes to the left? */
			if (useleft && trec.ir_freecount) {
				rec = trec;
				xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
				cur = tcur;

				pag->pagl_leftrec = trec.ir_startino;
				pag->pagl_rightrec = rec.ir_startino;
				pag->pagl_pagino = pagino;
				goto alloc_inode;
			}

			/* free inodes to the right? */
			if (!useleft && rec.ir_freecount) {
				xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);

				pag->pagl_leftrec = trec.ir_startino;
				pag->pagl_rightrec = rec.ir_startino;
				pag->pagl_pagino = pagino;
				goto alloc_inode;
			}

			/* get next record to check */
			if (useleft) {
				error = xfs_ialloc_next_rec(tcur, &trec,
								 &doneleft, 1);
			} else {
				error = xfs_ialloc_next_rec(cur, &rec,
								 &doneright, 0);
			}
			if (error)
				goto error1;
		}

		/*
		 * We've reached the end of the btree. because
		 * we are only searching a small chunk of the
		 * btree each search, there is obviously free
		 * inodes closer to the parent inode than we
		 * are now. restart the search again.
		 */
		pag->pagl_pagino = NULLAGINO;
		pag->pagl_leftrec = NULLAGINO;
		pag->pagl_rightrec = NULLAGINO;
		xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
		xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
		goto restart_pagno;
	}

	/*
	 * In a different AG from the parent.
	 * See if the most recently allocated block has any free.
	 */
newino:
	if (agi->agi_newino != cpu_to_be32(NULLAGINO)) {
		error = xfs_inobt_lookup(cur, be32_to_cpu(agi->agi_newino),
					 XFS_LOOKUP_EQ, &i);
		if (error)
			goto error0;

		if (i == 1) {
			error = xfs_inobt_get_rec(cur, &rec, &j);
			if (error)
				goto error0;

			if (j == 1 && rec.ir_freecount > 0) {
				/*
				 * The last chunk allocated in the group
				 * still has a free inode.
				 */
				goto alloc_inode;
			}
		}
	}

	/*
	 * None left in the last group, search the whole AG
	 */
	error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
	if (error)
		goto error0;
	XFS_WANT_CORRUPTED_GOTO(i == 1, error0);

	for (;;) {
		error = xfs_inobt_get_rec(cur, &rec, &i);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
		if (rec.ir_freecount > 0)
			break;
		error = xfs_btree_increment(cur, 0, &i);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
	}

alloc_inode:
	offset = xfs_ialloc_find_free(&rec.ir_free);
	ASSERT(offset >= 0);
	ASSERT(offset < XFS_INODES_PER_CHUNK);
	ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
				   XFS_INODES_PER_CHUNK) == 0);
	ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
	rec.ir_free &= ~XFS_INOBT_MASK(offset);
	rec.ir_freecount--;
	error = xfs_inobt_update(cur, &rec);
	if (error)
		goto error0;
	be32_add_cpu(&agi->agi_freecount, -1);
	xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
	pag->pagi_freecount--;

	error = xfs_check_agi_freecount(cur, agi);
	if (error)
		goto error0;

	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
	xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
	xfs_perag_put(pag);
	*inop = ino;
	return 0;
error1:
	xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR);
error0:
	xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
	xfs_perag_put(pag);
	return error;
}
示例#4
0
/*
 * Return inode number table for the filesystem.
 */
int					/* error status */
xfs_inumbers(
	struct xfs_mount	*mp,/* mount point for filesystem */
	xfs_ino_t		*lastino,/* last inode returned */
	int			*count,/* size of buffer/count returned */
	void			__user *ubuffer,/* buffer with inode descriptions */
	inumbers_fmt_pf		formatter)
{
	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, *lastino);
	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, *lastino);
	struct xfs_btree_cur	*cur = NULL;
	struct xfs_buf		*agbp = NULL;
	struct xfs_inogrp	*buffer;
	int			bcount;
	int			left = *count;
	int			bufidx = 0;
	int			error = 0;

	*count = 0;
	if (agno >= mp->m_sb.sb_agcount ||
	    *lastino != XFS_AGINO_TO_INO(mp, agno, agino))
		return error;

	bcount = MIN(left, (int)(PAGE_SIZE / sizeof(*buffer)));
	buffer = kmem_alloc(bcount * sizeof(*buffer), KM_SLEEP);
	do {
		struct xfs_inobt_rec_incore	r;
		int				stat;

		if (!agbp) {
			error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
			if (error)
				break;

			cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno,
						    XFS_BTNUM_INO);
			error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_GE,
						 &stat);
			if (error)
				break;
			if (!stat)
				goto next_ag;
		}

		error = xfs_inobt_get_rec(cur, &r, &stat);
		if (error)
			break;
		if (!stat)
			goto next_ag;

		agino = r.ir_startino + XFS_INODES_PER_CHUNK - 1;
		buffer[bufidx].xi_startino =
			XFS_AGINO_TO_INO(mp, agno, r.ir_startino);
		buffer[bufidx].xi_alloccount = r.ir_count - r.ir_freecount;
		buffer[bufidx].xi_allocmask = ~r.ir_free;
		if (++bufidx == bcount) {
			long	written;

			error = formatter(ubuffer, buffer, bufidx, &written);
			if (error)
				break;
			ubuffer += written;
			*count += bufidx;
			bufidx = 0;
		}
		if (!--left)
			break;

		error = xfs_btree_increment(cur, 0, &stat);
		if (error)
			break;
		if (stat)
			continue;

next_ag:
		xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
		cur = NULL;
		xfs_buf_relse(agbp);
		agbp = NULL;
		agino = 0;
		agno++;
	} while (agno < mp->m_sb.sb_agcount);

	if (!error) {
		if (bufidx) {
			long	written;

			error = formatter(ubuffer, buffer, bufidx, &written);
			if (!error)
				*count += bufidx;
		}
		*lastino = XFS_AGINO_TO_INO(mp, agno, agino);
	}

	kmem_free(buffer);
	if (cur)
		xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR :
					   XFS_BTREE_NOERROR));
	if (agbp)
		xfs_buf_relse(agbp);

	return error;
}
示例#5
0
/*
 * Return stat information in bulk (by-inode) for the filesystem.
 */
int					/* error status */
xfs_bulkstat(
	xfs_mount_t		*mp,	/* mount point for filesystem */
	xfs_ino_t		*lastinop, /* last inode returned */
	int			*ubcountp, /* size of buffer/count returned */
	bulkstat_one_pf		formatter, /* func that'd fill a single buf */
	size_t			statstruct_size, /* sizeof struct filling */
	char			__user *ubuffer, /* buffer with inode stats */
	int			*done)	/* 1 if there are more stats to get */
{
	xfs_buf_t		*agbp;	/* agi header buffer */
	xfs_agino_t		agino;	/* inode # in allocation group */
	xfs_agnumber_t		agno;	/* allocation group number */
	xfs_btree_cur_t		*cur;	/* btree cursor for ialloc btree */
	size_t			irbsize; /* size of irec buffer in bytes */
	xfs_inobt_rec_incore_t	*irbuf;	/* start of irec buffer */
	int			nirbuf;	/* size of irbuf */
	int			ubcount; /* size of user's buffer */
	struct xfs_bulkstat_agichunk ac;
	int			error = 0;

	/*
	 * Get the last inode value, see if there's nothing to do.
	 */
	agno = XFS_INO_TO_AGNO(mp, *lastinop);
	agino = XFS_INO_TO_AGINO(mp, *lastinop);
	if (agno >= mp->m_sb.sb_agcount ||
	    *lastinop != XFS_AGINO_TO_INO(mp, agno, agino)) {
		*done = 1;
		*ubcountp = 0;
		return 0;
	}

	ubcount = *ubcountp; /* statstruct's */
	ac.ac_ubuffer = &ubuffer;
	ac.ac_ubleft = ubcount * statstruct_size; /* bytes */;
	ac.ac_ubelem = 0;

	*ubcountp = 0;
	*done = 0;

	irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4);
	if (!irbuf)
		return -ENOMEM;

	nirbuf = irbsize / sizeof(*irbuf);

	/*
	 * Loop over the allocation groups, starting from the last
	 * inode returned; 0 means start of the allocation group.
	 */
	while (agno < mp->m_sb.sb_agcount) {
		struct xfs_inobt_rec_incore	*irbp = irbuf;
		struct xfs_inobt_rec_incore	*irbufend = irbuf + nirbuf;
		bool				end_of_ag = false;
		int				icount = 0;
		int				stat;

		error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
		if (error)
			break;
		/*
		 * Allocate and initialize a btree cursor for ialloc btree.
		 */
		cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno,
					    XFS_BTNUM_INO);
		if (agino > 0) {
			/*
			 * In the middle of an allocation group, we need to get
			 * the remainder of the chunk we're in.
			 */
			struct xfs_inobt_rec_incore	r;

			error = xfs_bulkstat_grab_ichunk(cur, agino, &icount, &r);
			if (error)
				goto del_cursor;
			if (icount) {
				irbp->ir_startino = r.ir_startino;
				irbp->ir_holemask = r.ir_holemask;
				irbp->ir_count = r.ir_count;
				irbp->ir_freecount = r.ir_freecount;
				irbp->ir_free = r.ir_free;
				irbp++;
			}
			/* Increment to the next record */
			error = xfs_btree_increment(cur, 0, &stat);
		} else {
			/* Start of ag.  Lookup the first inode chunk */
			error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &stat);
		}
		if (error || stat == 0) {
			end_of_ag = true;
			goto del_cursor;
		}

		/*
		 * Loop through inode btree records in this ag,
		 * until we run out of inodes or space in the buffer.
		 */
		while (irbp < irbufend && icount < ubcount) {
			struct xfs_inobt_rec_incore	r;

			error = xfs_inobt_get_rec(cur, &r, &stat);
			if (error || stat == 0) {
				end_of_ag = true;
				goto del_cursor;
			}

			/*
			 * If this chunk has any allocated inodes, save it.
			 * Also start read-ahead now for this chunk.
			 */
			if (r.ir_freecount < r.ir_count) {
				xfs_bulkstat_ichunk_ra(mp, agno, &r);
				irbp->ir_startino = r.ir_startino;
				irbp->ir_holemask = r.ir_holemask;
				irbp->ir_count = r.ir_count;
				irbp->ir_freecount = r.ir_freecount;
				irbp->ir_free = r.ir_free;
				irbp++;
				icount += r.ir_count - r.ir_freecount;
			}
			error = xfs_btree_increment(cur, 0, &stat);
			if (error || stat == 0) {
				end_of_ag = true;
				goto del_cursor;
			}
			cond_resched();
		}

		/*
		 * Drop the btree buffers and the agi buffer as we can't hold any
		 * of the locks these represent when calling iget. If there is a
		 * pending error, then we are done.
		 */
del_cursor:
		xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
		xfs_buf_relse(agbp);
		if (error)
			break;
		/*
		 * Now format all the good inodes into the user's buffer. The
		 * call to xfs_bulkstat_ag_ichunk() sets up the agino pointer
		 * for the next loop iteration.
		 */
		irbufend = irbp;
		for (irbp = irbuf;
		     irbp < irbufend && ac.ac_ubleft >= statstruct_size;
		     irbp++) {
			error = xfs_bulkstat_ag_ichunk(mp, agno, irbp,
					formatter, statstruct_size, &ac,
					&agino);
			if (error)
				break;

			cond_resched();
		}

		/*
		 * If we've run out of space or had a formatting error, we
		 * are now done
		 */
		if (ac.ac_ubleft < statstruct_size || error)
			break;

		if (end_of_ag) {
			agno++;
			agino = 0;
		}
	}
	/*
	 * Done, we're either out of filesystem or space to put the data.
	 */
	kmem_free(irbuf);
	*ubcountp = ac.ac_ubelem;

	/*
	 * We found some inodes, so clear the error status and return them.
	 * The lastino pointer will point directly at the inode that triggered
	 * any error that occurred, so on the next call the error will be
	 * triggered again and propagated to userspace as there will be no
	 * formatted inodes in the buffer.
	 */
	if (ac.ac_ubelem)
		error = 0;

	/*
	 * If we ran out of filesystem, lastino will point off the end of
	 * the filesystem so the next call will return immediately.
	 */
	*lastinop = XFS_AGINO_TO_INO(mp, agno, agino);
	if (agno >= mp->m_sb.sb_agcount)
		*done = 1;

	return error;
}
示例#6
0
/*
 * Allocate an inode.
 *
 * The caller selected an AG for us, and made sure that free inodes are
 * available.
 */
STATIC int
xfs_dialloc_ag(
	struct xfs_trans	*tp,
	struct xfs_buf		*agbp,
	xfs_ino_t		parent,
	xfs_ino_t		*inop)
{
	struct xfs_mount	*mp = tp->t_mountp;
	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
	xfs_agnumber_t		agno = be32_to_cpu(agi->agi_seqno);
	xfs_agnumber_t		pagno = XFS_INO_TO_AGNO(mp, parent);
	xfs_agino_t		pagino = XFS_INO_TO_AGINO(mp, parent);
	struct xfs_perag	*pag;
	struct xfs_btree_cur	*cur, *tcur;
	struct xfs_inobt_rec_incore rec, trec;
	xfs_ino_t		ino;
	int			error;
	int			offset;
	int			i, j;

	pag = xfs_perag_get(mp, agno);

	ASSERT(pag->pagi_init);
	ASSERT(pag->pagi_inodeok);
	ASSERT(pag->pagi_freecount > 0);

 restart_pagno:
	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
	/*
	 * If pagino is 0 (this is the root inode allocation) use newino.
	 * This must work because we've just allocated some.
	 */
	if (!pagino)
		pagino = be32_to_cpu(agi->agi_newino);

	error = xfs_check_agi_freecount(cur, agi);
	if (error)
		goto error0;

	/*
	 * If in the same AG as the parent, try to get near the parent.
	 */
	if (pagno == agno) {
		int		doneleft;	/* done, to the left */
		int		doneright;	/* done, to the right */
		int		searchdistance = 10;

		error = xfs_inobt_lookup(cur, pagino, XFS_LOOKUP_LE, &i);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);

		error = xfs_inobt_get_rec(cur, &rec, &j);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);

		if (rec.ir_freecount > 0) {
			/*
			 * Found a free inode in the same chunk
			 * as the parent, done.
			 */
			goto alloc_inode;
		}


		/*
		 * In the same AG as parent, but parent's chunk is full.
		 */

		/* duplicate the cursor, search left & right simultaneously */
		error = xfs_btree_dup_cursor(cur, &tcur);
		if (error)
			goto error0;

		/*
		 * Skip to last blocks looked up if same parent inode.
		 */
		if (pagino != NULLAGINO &&
		    pag->pagl_pagino == pagino &&
		    pag->pagl_leftrec != NULLAGINO &&
		    pag->pagl_rightrec != NULLAGINO) {
			error = xfs_ialloc_get_rec(tcur, pag->pagl_leftrec,
						   &trec, &doneleft);
			if (error)
				goto error1;

			error = xfs_ialloc_get_rec(cur, pag->pagl_rightrec,
						   &rec, &doneright);
			if (error)
				goto error1;
		} else {
			/* search left with tcur, back up 1 record */
			error = xfs_ialloc_next_rec(tcur, &trec, &doneleft, 1);
			if (error)
				goto error1;

			/* search right with cur, go forward 1 record. */
			error = xfs_ialloc_next_rec(cur, &rec, &doneright, 0);
			if (error)
				goto error1;
		}

		/*
		 * Loop until we find an inode chunk with a free inode.
		 */
		while (!doneleft || !doneright) {
			int	useleft;  /* using left inode chunk this time */

			if (!--searchdistance) {
				/*
				 * Not in range - save last search
				 * location and allocate a new inode
				 */
				xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
				pag->pagl_leftrec = trec.ir_startino;
				pag->pagl_rightrec = rec.ir_startino;
				pag->pagl_pagino = pagino;
				goto newino;
			}

			/* figure out the closer block if both are valid. */
			if (!doneleft && !doneright) {
				useleft = pagino -
				 (trec.ir_startino + XFS_INODES_PER_CHUNK - 1) <
				  rec.ir_startino - pagino;
			} else {
				useleft = !doneleft;
			}

			/* free inodes to the left? */
			if (useleft && trec.ir_freecount) {
				rec = trec;
				xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
				cur = tcur;

				pag->pagl_leftrec = trec.ir_startino;
				pag->pagl_rightrec = rec.ir_startino;
				pag->pagl_pagino = pagino;
				goto alloc_inode;
			}

			/* free inodes to the right? */
			if (!useleft && rec.ir_freecount) {
				xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);

				pag->pagl_leftrec = trec.ir_startino;
				pag->pagl_rightrec = rec.ir_startino;
				pag->pagl_pagino = pagino;
				goto alloc_inode;
			}

			/* get next record to check */
			if (useleft) {
				error = xfs_ialloc_next_rec(tcur, &trec,
								 &doneleft, 1);
			} else {
				error = xfs_ialloc_next_rec(cur, &rec,
								 &doneright, 0);
			}
			if (error)
				goto error1;
		}

		/*
		 * We've reached the end of the btree. because
		 * we are only searching a small chunk of the
		 * btree each search, there is obviously free
		 * inodes closer to the parent inode than we
		 * are now. restart the search again.
		 */
		pag->pagl_pagino = NULLAGINO;
		pag->pagl_leftrec = NULLAGINO;
		pag->pagl_rightrec = NULLAGINO;
		xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
		xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
		goto restart_pagno;
	}

	/*
	 * In a different AG from the parent.
	 * See if the most recently allocated block has any free.
	 */
newino:
	if (agi->agi_newino != cpu_to_be32(NULLAGINO)) {
		error = xfs_inobt_lookup(cur, be32_to_cpu(agi->agi_newino),
					 XFS_LOOKUP_EQ, &i);
		if (error)
			goto error0;

		if (i == 1) {
			error = xfs_inobt_get_rec(cur, &rec, &j);
			if (error)
				goto error0;

			if (j == 1 && rec.ir_freecount > 0) {
				/*
				 * The last chunk allocated in the group
				 * still has a free inode.
				 */
				goto alloc_inode;
			}
		}
	}

	/*
	 * None left in the last group, search the whole AG
	 */
	error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
	if (error)
		goto error0;
	XFS_WANT_CORRUPTED_GOTO(i == 1, error0);

	for (;;) {
		error = xfs_inobt_get_rec(cur, &rec, &i);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
		if (rec.ir_freecount > 0)
			break;
		error = xfs_btree_increment(cur, 0, &i);
		if (error)
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
	}

alloc_inode:
	offset = xfs_lowbit64(rec.ir_free);
	ASSERT(offset >= 0);
	ASSERT(offset < XFS_INODES_PER_CHUNK);
	ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
				   XFS_INODES_PER_CHUNK) == 0);
	ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
	rec.ir_free &= ~XFS_INOBT_MASK(offset);
	rec.ir_freecount--;
	error = xfs_inobt_update(cur, &rec);
	if (error)
		goto error0;
	be32_add_cpu(&agi->agi_freecount, -1);
	xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
	pag->pagi_freecount--;

	error = xfs_check_agi_freecount(cur, agi);
	if (error)
		goto error0;

	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
	xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
	xfs_perag_put(pag);
	*inop = ino;
	return 0;
error1:
	xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR);
error0:
	xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
	xfs_perag_put(pag);
	return error;
}
/*
 * Allocate an inode on disk.
 * Mode is used to tell whether the new inode will need space, and whether
 * it is a directory.
 *
 * The arguments IO_agbp and alloc_done are defined to work within
 * the constraint of one allocation per transaction.
 * xfs_dialloc() is designed to be called twice if it has to do an
 * allocation to make more free inodes.  On the first call,
 * IO_agbp should be set to NULL. If an inode is available,
 * i.e., xfs_dialloc() did not need to do an allocation, an inode
 * number is returned.  In this case, IO_agbp would be set to the
 * current ag_buf and alloc_done set to false.
 * If an allocation needed to be done, xfs_dialloc would return
 * the current ag_buf in IO_agbp and set alloc_done to true.
 * The caller should then commit the current transaction, allocate a new
 * transaction, and call xfs_dialloc() again, passing in the previous
 * value of IO_agbp.  IO_agbp should be held across the transactions.
 * Since the agbp is locked across the two calls, the second call is
 * guaranteed to have a free inode available.
 *
 * Once we successfully pick an inode its number is returned and the
 * on-disk data structures are updated.  The inode itself is not read
 * in, since doing so would break ordering constraints with xfs_reclaim.
 */
int
xfs_dialloc(
	xfs_trans_t	*tp,		/* transaction pointer */
	xfs_ino_t	parent,		/* parent inode (directory) */
	mode_t		mode,		/* mode bits for new inode */
	int		okalloc,	/* ok to allocate more space */
	xfs_buf_t	**IO_agbp,	/* in/out ag header's buffer */
	boolean_t	*alloc_done,	/* true if we needed to replenish
					   inode freelist */
	xfs_ino_t	*inop)		/* inode number allocated */
{
	xfs_agnumber_t	agcount;	/* number of allocation groups */
	xfs_buf_t	*agbp;		/* allocation group header's buffer */
	xfs_agnumber_t	agno;		/* allocation group number */
	xfs_agi_t	*agi;		/* allocation group header structure */
	xfs_btree_cur_t	*cur;		/* inode allocation btree cursor */
	int		error;		/* error return value */
	int		i;		/* result code */
	int		ialloced;	/* inode allocation status */
	int		noroom = 0;	/* no space for inode blk allocation */
	xfs_ino_t	ino;		/* fs-relative inode to be returned */
	/* REFERENCED */
	int		j;		/* result code */
	xfs_mount_t	*mp;		/* file system mount structure */
	int		offset;		/* index of inode in chunk */
	xfs_agino_t	pagino;		/* parent's a.g. relative inode # */
	xfs_agnumber_t	pagno;		/* parent's allocation group number */
	xfs_inobt_rec_incore_t rec;	/* inode allocation record */
	xfs_agnumber_t	tagno;		/* testing allocation group number */
	xfs_btree_cur_t	*tcur;		/* temp cursor */
	xfs_inobt_rec_incore_t trec;	/* temp inode allocation record */


	if (*IO_agbp == NULL) {
		/*
		 * We do not have an agbp, so select an initial allocation
		 * group for inode allocation.
		 */
		agbp = xfs_ialloc_ag_select(tp, parent, mode, okalloc);
		/*
		 * Couldn't find an allocation group satisfying the
		 * criteria, give up.
		 */
		if (!agbp) {
			*inop = NULLFSINO;
			return 0;
		}
		agi = XFS_BUF_TO_AGI(agbp);
		ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
	} else {
		/*
		 * Continue where we left off before.  In this case, we
		 * know that the allocation group has free inodes.
		 */
		agbp = *IO_agbp;
		agi = XFS_BUF_TO_AGI(agbp);
		ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
		ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
	}
	mp = tp->t_mountp;
	agcount = mp->m_sb.sb_agcount;
	agno = be32_to_cpu(agi->agi_seqno);
	tagno = agno;
	pagno = XFS_INO_TO_AGNO(mp, parent);
	pagino = XFS_INO_TO_AGINO(mp, parent);

	/*
	 * If we have already hit the ceiling of inode blocks then clear
	 * okalloc so we scan all available agi structures for a free
	 * inode.
	 */

	if (mp->m_maxicount &&
	    mp->m_sb.sb_icount + XFS_IALLOC_INODES(mp) > mp->m_maxicount) {
		noroom = 1;
		okalloc = 0;
	}

	/*
	 * Loop until we find an allocation group that either has free inodes
	 * or in which we can allocate some inodes.  Iterate through the
	 * allocation groups upward, wrapping at the end.
	 */
	*alloc_done = B_FALSE;
	while (!agi->agi_freecount) {
		/*
		 * Don't do anything if we're not supposed to allocate
		 * any blocks, just go on to the next ag.
		 */
		if (okalloc) {
			/*
			 * Try to allocate some new inodes in the allocation
			 * group.
			 */
			if ((error = xfs_ialloc_ag_alloc(tp, agbp, &ialloced))) {
				xfs_trans_brelse(tp, agbp);
				if (error == ENOSPC) {
					*inop = NULLFSINO;
					return 0;
				} else
					return error;
			}
			if (ialloced) {
				/*
				 * We successfully allocated some inodes, return
				 * the current context to the caller so that it
				 * can commit the current transaction and call
				 * us again where we left off.
				 */
				ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
				*alloc_done = B_TRUE;
				*IO_agbp = agbp;
				*inop = NULLFSINO;
				return 0;
			}
		}
		/*
		 * If it failed, give up on this ag.
		 */
		xfs_trans_brelse(tp, agbp);
		/*
		 * Go on to the next ag: get its ag header.
		 */
nextag:
		if (++tagno == agcount)
			tagno = 0;
		if (tagno == agno) {
			*inop = NULLFSINO;
			return noroom ? ENOSPC : 0;
		}
		down_read(&mp->m_peraglock);
		if (mp->m_perag[tagno].pagi_inodeok == 0) {
			up_read(&mp->m_peraglock);
			goto nextag;
		}
		error = xfs_ialloc_read_agi(mp, tp, tagno, &agbp);
		up_read(&mp->m_peraglock);
		if (error)
			goto nextag;
		agi = XFS_BUF_TO_AGI(agbp);
		ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
	}
	/*
	 * Here with an allocation group that has a free inode.
	 * Reset agno since we may have chosen a new ag in the
	 * loop above.
	 */
	agno = tagno;
	*IO_agbp = NULL;
	cur = xfs_inobt_init_cursor(mp, tp, agbp, be32_to_cpu(agi->agi_seqno));
	/*
	 * If pagino is 0 (this is the root inode allocation) use newino.
	 * This must work because we've just allocated some.
	 */
	if (!pagino)
		pagino = be32_to_cpu(agi->agi_newino);
#ifdef DEBUG
	if (cur->bc_nlevels == 1) {
		int	freecount = 0;

		if ((error = xfs_inobt_lookup_ge(cur, 0, 0, 0, &i)))
			goto error0;
		XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
		do {
			if ((error = xfs_inobt_get_rec(cur, &rec.ir_startino,
					&rec.ir_freecount, &rec.ir_free, &i)))
				goto error0;
			XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
			freecount += rec.ir_freecount;
			if ((error = xfs_btree_increment(cur, 0, &i)))
				goto error0;
		} while (i == 1);

		ASSERT(freecount == be32_to_cpu(agi->agi_freecount) ||
		       XFS_FORCED_SHUTDOWN(mp));
	}
#endif
	/*
	 * If in the same a.g. as the parent, try to get near the parent.
	 */
	if (pagno == agno) {
		if ((error = xfs_inobt_lookup_le(cur, pagino, 0, 0, &i)))
			goto error0;
		if (i != 0 &&
		    (error = xfs_inobt_get_rec(cur, &rec.ir_startino,
			    &rec.ir_freecount, &rec.ir_free, &j)) == 0 &&
		    j == 1 &&
		    rec.ir_freecount > 0) {
			/*
			 * Found a free inode in the same chunk
			 * as parent, done.
			 */
		}
		/*
		 * In the same a.g. as parent, but parent's chunk is full.
		 */
		else {
			int	doneleft;	/* done, to the left */
			int	doneright;	/* done, to the right */

			if (error)
				goto error0;
			ASSERT(i == 1);
			ASSERT(j == 1);
			/*
			 * Duplicate the cursor, search left & right
			 * simultaneously.
			 */
			if ((error = xfs_btree_dup_cursor(cur, &tcur)))
				goto error0;
			/*
			 * Search left with tcur, back up 1 record.
			 */
			if ((error = xfs_btree_decrement(tcur, 0, &i)))
				goto error1;
			doneleft = !i;
			if (!doneleft) {
				if ((error = xfs_inobt_get_rec(tcur,
						&trec.ir_startino,
						&trec.ir_freecount,
						&trec.ir_free, &i)))
					goto error1;
				XFS_WANT_CORRUPTED_GOTO(i == 1, error1);
			}
			/*
			 * Search right with cur, go forward 1 record.
			 */
			if ((error = xfs_btree_increment(cur, 0, &i)))
				goto error1;
			doneright = !i;
			if (!doneright) {
				if ((error = xfs_inobt_get_rec(cur,
						&rec.ir_startino,
						&rec.ir_freecount,
						&rec.ir_free, &i)))
					goto error1;
				XFS_WANT_CORRUPTED_GOTO(i == 1, error1);
			}
			/*
			 * Loop until we find the closest inode chunk
			 * with a free one.
			 */
			while (!doneleft || !doneright) {
				int	useleft;  /* using left inode
						     chunk this time */

				/*
				 * Figure out which block is closer,
				 * if both are valid.
				 */
				if (!doneleft && !doneright)
					useleft =
						pagino -
						(trec.ir_startino +
						 XFS_INODES_PER_CHUNK - 1) <
						 rec.ir_startino - pagino;
				else
					useleft = !doneleft;
				/*
				 * If checking the left, does it have
				 * free inodes?
				 */
				if (useleft && trec.ir_freecount) {
					/*
					 * Yes, set it up as the chunk to use.
					 */
					rec = trec;
					xfs_btree_del_cursor(cur,
						XFS_BTREE_NOERROR);
					cur = tcur;
					break;
				}
				/*
				 * If checking the right, does it have
				 * free inodes?
				 */
				if (!useleft && rec.ir_freecount) {
					/*
					 * Yes, it's already set up.
					 */
					xfs_btree_del_cursor(tcur,
						XFS_BTREE_NOERROR);
					break;
				}
				/*
				 * If used the left, get another one
				 * further left.
				 */
				if (useleft) {
					if ((error = xfs_btree_decrement(tcur, 0,
							&i)))
						goto error1;
					doneleft = !i;
					if (!doneleft) {
						if ((error = xfs_inobt_get_rec(
							    tcur,
							    &trec.ir_startino,
							    &trec.ir_freecount,
							    &trec.ir_free, &i)))
							goto error1;
						XFS_WANT_CORRUPTED_GOTO(i == 1,
							error1);
					}
				}
				/*
				 * If used the right, get another one
				 * further right.
				 */
				else {
					if ((error = xfs_btree_increment(cur, 0,
							&i)))
						goto error1;
					doneright = !i;
					if (!doneright) {
						if ((error = xfs_inobt_get_rec(
							    cur,
							    &rec.ir_startino,
							    &rec.ir_freecount,
							    &rec.ir_free, &i)))
							goto error1;
						XFS_WANT_CORRUPTED_GOTO(i == 1,
							error1);
					}
				}
			}
			ASSERT(!doneleft || !doneright);
		}
	}
	/*
	 * In a different a.g. from the parent.
	 * See if the most recently allocated block has any free.
	 */
	else if (be32_to_cpu(agi->agi_newino) != NULLAGINO) {
		if ((error = xfs_inobt_lookup_eq(cur,
				be32_to_cpu(agi->agi_newino), 0, 0, &i)))
			goto error0;
		if (i == 1 &&
		    (error = xfs_inobt_get_rec(cur, &rec.ir_startino,
			    &rec.ir_freecount, &rec.ir_free, &j)) == 0 &&
		    j == 1 &&
		    rec.ir_freecount > 0) {
			/*
			 * The last chunk allocated in the group still has
			 * a free inode.
			 */
		}
		/*
		 * None left in the last group, search the whole a.g.
		 */
		else {
			if (error)
				goto error0;
			if ((error = xfs_inobt_lookup_ge(cur, 0, 0, 0, &i)))
				goto error0;
			ASSERT(i == 1);
			for (;;) {
				if ((error = xfs_inobt_get_rec(cur,
						&rec.ir_startino,
						&rec.ir_freecount, &rec.ir_free,
						&i)))
					goto error0;
				XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
				if (rec.ir_freecount > 0)
					break;
				if ((error = xfs_btree_increment(cur, 0, &i)))
					goto error0;
				XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
			}
		}
	}
	offset = xfs_ialloc_find_free(&rec.ir_free);
	ASSERT(offset >= 0);
	ASSERT(offset < XFS_INODES_PER_CHUNK);
	ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
				   XFS_INODES_PER_CHUNK) == 0);
	ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
	XFS_INOBT_CLR_FREE(&rec, offset);
	rec.ir_freecount--;
	if ((error = xfs_inobt_update(cur, rec.ir_startino, rec.ir_freecount,
			rec.ir_free)))
		goto error0;
	be32_add_cpu(&agi->agi_freecount, -1);
	xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
	down_read(&mp->m_peraglock);
	mp->m_perag[tagno].pagi_freecount--;
	up_read(&mp->m_peraglock);
#ifdef DEBUG
	if (cur->bc_nlevels == 1) {
		int	freecount = 0;

		if ((error = xfs_inobt_lookup_ge(cur, 0, 0, 0, &i)))
			goto error0;
		do {
			if ((error = xfs_inobt_get_rec(cur, &rec.ir_startino,
					&rec.ir_freecount, &rec.ir_free, &i)))
				goto error0;
			XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
			freecount += rec.ir_freecount;
			if ((error = xfs_btree_increment(cur, 0, &i)))
				goto error0;
		} while (i == 1);
		ASSERT(freecount == be32_to_cpu(agi->agi_freecount) ||
		       XFS_FORCED_SHUTDOWN(mp));
	}
#endif
	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
	xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
	*inop = ino;
	return 0;
error1:
	xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR);
error0:
	xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
	return error;
}
/*
 * Allocate new inodes in the allocation group specified by agbp.
 * Return 0 for success, else error code.
 */
STATIC int				/* error code or 0 */
xfs_ialloc_ag_alloc(
	xfs_trans_t	*tp,		/* transaction pointer */
	xfs_buf_t	*agbp,		/* alloc group buffer */
	int		*alloc)
{
	xfs_agi_t	*agi;		/* allocation group header */
	xfs_alloc_arg_t	args;		/* allocation argument structure */
	int		blks_per_cluster;  /* fs blocks per inode cluster */
	xfs_btree_cur_t	*cur;		/* inode btree cursor */
	xfs_daddr_t	d;		/* disk addr of buffer */
	xfs_agnumber_t	agno;
	int		error;
	xfs_buf_t	*fbuf;		/* new free inodes' buffer */
	xfs_dinode_t	*free;		/* new free inode structure */
	int		i;		/* inode counter */
	int		j;		/* block counter */
	int		nbufs;		/* num bufs of new inodes */
	xfs_agino_t	newino;		/* new first inode's number */
	xfs_agino_t	newlen;		/* new number of inodes */
	int		ninodes;	/* num inodes per buf */
	xfs_agino_t	thisino;	/* current inode number, for loop */
	int		version;	/* inode version number to use */
	int		isaligned = 0;	/* inode allocation at stripe unit */
					/* boundary */
	unsigned int	gen;

	args.tp = tp;
	args.mp = tp->t_mountp;

	/*
	 * Locking will ensure that we don't have two callers in here
	 * at one time.
	 */
	newlen = XFS_IALLOC_INODES(args.mp);
	if (args.mp->m_maxicount &&
	    args.mp->m_sb.sb_icount + newlen > args.mp->m_maxicount)
		return XFS_ERROR(ENOSPC);
	args.minlen = args.maxlen = XFS_IALLOC_BLOCKS(args.mp);
	/*
	 * First try to allocate inodes contiguous with the last-allocated
	 * chunk of inodes.  If the filesystem is striped, this will fill
	 * an entire stripe unit with inodes.
 	 */
	agi = XFS_BUF_TO_AGI(agbp);
	newino = be32_to_cpu(agi->agi_newino);
	args.agbno = XFS_AGINO_TO_AGBNO(args.mp, newino) +
			XFS_IALLOC_BLOCKS(args.mp);
	if (likely(newino != NULLAGINO &&
		  (args.agbno < be32_to_cpu(agi->agi_length)))) {
		args.fsbno = XFS_AGB_TO_FSB(args.mp,
				be32_to_cpu(agi->agi_seqno), args.agbno);
		args.type = XFS_ALLOCTYPE_THIS_BNO;
		args.mod = args.total = args.wasdel = args.isfl =
			args.userdata = args.minalignslop = 0;
		args.prod = 1;

		/*
		 * We need to take into account alignment here to ensure that
		 * we don't modify the free list if we fail to have an exact
		 * block. If we don't have an exact match, and every oher
		 * attempt allocation attempt fails, we'll end up cancelling
		 * a dirty transaction and shutting down.
		 *
		 * For an exact allocation, alignment must be 1,
		 * however we need to take cluster alignment into account when
		 * fixing up the freelist. Use the minalignslop field to
		 * indicate that extra blocks might be required for alignment,
		 * but not to use them in the actual exact allocation.
		 */
		args.alignment = 1;
		args.minalignslop = xfs_ialloc_cluster_alignment(&args) - 1;

		/* Allow space for the inode btree to split. */
		args.minleft = XFS_IN_MAXLEVELS(args.mp) - 1;
		if ((error = xfs_alloc_vextent(&args)))
			return error;
	} else
		args.fsbno = NULLFSBLOCK;

	if (unlikely(args.fsbno == NULLFSBLOCK)) {
		/*
		 * Set the alignment for the allocation.
		 * If stripe alignment is turned on then align at stripe unit
		 * boundary.
		 * If the cluster size is smaller than a filesystem block
		 * then we're doing I/O for inodes in filesystem block size
		 * pieces, so don't need alignment anyway.
		 */
		isaligned = 0;
		if (args.mp->m_sinoalign) {
			ASSERT(!(args.mp->m_flags & XFS_MOUNT_NOALIGN));
			args.alignment = args.mp->m_dalign;
			isaligned = 1;
		} else
			args.alignment = xfs_ialloc_cluster_alignment(&args);
		/*
		 * Need to figure out where to allocate the inode blocks.
		 * Ideally they should be spaced out through the a.g.
		 * For now, just allocate blocks up front.
		 */
		args.agbno = be32_to_cpu(agi->agi_root);
		args.fsbno = XFS_AGB_TO_FSB(args.mp,
				be32_to_cpu(agi->agi_seqno), args.agbno);
		/*
		 * Allocate a fixed-size extent of inodes.
		 */
		args.type = XFS_ALLOCTYPE_NEAR_BNO;
		args.mod = args.total = args.wasdel = args.isfl =
			args.userdata = args.minalignslop = 0;
		args.prod = 1;
		/*
		 * Allow space for the inode btree to split.
		 */
		args.minleft = XFS_IN_MAXLEVELS(args.mp) - 1;
		if ((error = xfs_alloc_vextent(&args)))
			return error;
	}

	/*
	 * If stripe alignment is turned on, then try again with cluster
	 * alignment.
	 */
	if (isaligned && args.fsbno == NULLFSBLOCK) {
		args.type = XFS_ALLOCTYPE_NEAR_BNO;
		args.agbno = be32_to_cpu(agi->agi_root);
		args.fsbno = XFS_AGB_TO_FSB(args.mp,
				be32_to_cpu(agi->agi_seqno), args.agbno);
		args.alignment = xfs_ialloc_cluster_alignment(&args);
		if ((error = xfs_alloc_vextent(&args)))
			return error;
	}

	if (args.fsbno == NULLFSBLOCK) {
		*alloc = 0;
		return 0;
	}
	ASSERT(args.len == args.minlen);
	/*
	 * Convert the results.
	 */
	newino = XFS_OFFBNO_TO_AGINO(args.mp, args.agbno, 0);
	/*
	 * Loop over the new block(s), filling in the inodes.
	 * For small block sizes, manipulate the inodes in buffers
	 * which are multiples of the blocks size.
	 */
	if (args.mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(args.mp)) {
		blks_per_cluster = 1;
		nbufs = (int)args.len;
		ninodes = args.mp->m_sb.sb_inopblock;
	} else {
		blks_per_cluster = XFS_INODE_CLUSTER_SIZE(args.mp) /
				   args.mp->m_sb.sb_blocksize;
		nbufs = (int)args.len / blks_per_cluster;
		ninodes = blks_per_cluster * args.mp->m_sb.sb_inopblock;
	}
	/*
	 * Figure out what version number to use in the inodes we create.
	 * If the superblock version has caught up to the one that supports
	 * the new inode format, then use the new inode version.  Otherwise
	 * use the old version so that old kernels will continue to be
	 * able to use the file system.
	 */
	if (xfs_sb_version_hasnlink(&args.mp->m_sb))
		version = 2;
	else
		version = 1;

	/*
	 * Seed the new inode cluster with a random generation number. This
	 * prevents short-term reuse of generation numbers if a chunk is
	 * freed and then immediately reallocated. We use random numbers
	 * rather than a linear progression to prevent the next generation
	 * number from being easily guessable.
	 */
	gen = random32();
	for (j = 0; j < nbufs; j++) {
		/*
		 * Get the block.
		 */
		d = XFS_AGB_TO_DADDR(args.mp, be32_to_cpu(agi->agi_seqno),
				     args.agbno + (j * blks_per_cluster));
		fbuf = xfs_trans_get_buf(tp, args.mp->m_ddev_targp, d,
					 args.mp->m_bsize * blks_per_cluster,
					 XFS_BUF_LOCK);
		ASSERT(fbuf);
		ASSERT(!XFS_BUF_GETERROR(fbuf));

		/*
		 * Initialize all inodes in this buffer and then log them.
		 *
		 * XXX: It would be much better if we had just one transaction to
		 *      log a whole cluster of inodes instead of all the indivdual
		 *      transactions causing a lot of log traffic.
		 */
		xfs_biozero(fbuf, 0, ninodes << args.mp->m_sb.sb_inodelog);
		for (i = 0; i < ninodes; i++) {
			int	ioffset = i << args.mp->m_sb.sb_inodelog;
			uint	isize = sizeof(struct xfs_dinode);

			free = xfs_make_iptr(args.mp, fbuf, i);
			free->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
			free->di_version = version;
			free->di_gen = cpu_to_be32(gen);
			free->di_next_unlinked = cpu_to_be32(NULLAGINO);
			xfs_trans_log_buf(tp, fbuf, ioffset, ioffset + isize - 1);
		}
		xfs_trans_inode_alloc_buf(tp, fbuf);
	}
	be32_add_cpu(&agi->agi_count, newlen);
	be32_add_cpu(&agi->agi_freecount, newlen);
	agno = be32_to_cpu(agi->agi_seqno);
	down_read(&args.mp->m_peraglock);
	args.mp->m_perag[agno].pagi_freecount += newlen;
	up_read(&args.mp->m_peraglock);
	agi->agi_newino = cpu_to_be32(newino);
	/*
	 * Insert records describing the new inode chunk into the btree.
	 */
	cur = xfs_inobt_init_cursor(args.mp, tp, agbp, agno);
	for (thisino = newino;
	     thisino < newino + newlen;
	     thisino += XFS_INODES_PER_CHUNK) {
		if ((error = xfs_inobt_lookup_eq(cur, thisino,
				XFS_INODES_PER_CHUNK, XFS_INOBT_ALL_FREE, &i))) {
			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
			return error;
		}
		ASSERT(i == 0);
		if ((error = xfs_btree_insert(cur, &i))) {
			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
			return error;
		}
		ASSERT(i == 1);
	}
	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
	/*
	 * Log allocation group header fields
	 */
	xfs_ialloc_log_agi(tp, agbp,
		XFS_AGI_COUNT | XFS_AGI_FREECOUNT | XFS_AGI_NEWINO);
	/*
	 * Modify/log superblock values for inode count and inode free count.
	 */
	xfs_trans_mod_sb(tp, XFS_TRANS_SB_ICOUNT, (long)newlen);
	xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, (long)newlen);
	*alloc = 1;
	return 0;
}