[deliverable/linux.git] / fs / xfs / libxfs / xfs_ialloc_btree.c

/*
 * Copyright (c) 2000-2001,2005 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_shared.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_trans_resv.h"
#include "xfs_bit.h"
#include "xfs_mount.h"
#include "xfs_inode.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_ialloc_btree.h"
#include "xfs_alloc.h"
#include "xfs_error.h"
#include "xfs_trace.h"
#include "xfs_cksum.h"
#include "xfs_trans.h"


STATIC int
xfs_inobt_get_minrecs(
	struct xfs_btree_cur	*cur,
	int			level)
{
	return cur->bc_mp->m_inobt_mnr[level != 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,
			cur->bc_btnum);
}

STATIC void
xfs_inobt_set_root(
	struct xfs_btree_cur	*cur,
	union xfs_btree_ptr	*nptr,
	int			inc)	/* level change */
{
	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);

	agi->agi_root = nptr->s;
	be32_add_cpu(&agi->agi_level, inc);
	xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL);
}

STATIC void
xfs_finobt_set_root(
	struct xfs_btree_cur	*cur,
	union xfs_btree_ptr	*nptr,
	int			inc)	/* level change */
{
	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);

	agi->agi_free_root = nptr->s;
	be32_add_cpu(&agi->agi_free_level, inc);
	xfs_ialloc_log_agi(cur->bc_tp, agbp,
			   XFS_AGI_FREE_ROOT | XFS_AGI_FREE_LEVEL);
}

STATIC int
xfs_inobt_alloc_block(
	struct xfs_btree_cur	*cur,
	union xfs_btree_ptr	*start,
	union xfs_btree_ptr	*new,
	int			*stat)
{
	xfs_alloc_arg_t		args;		/* block allocation args */
	int			error;		/* error return value */
	xfs_agblock_t		sbno = be32_to_cpu(start->s);

	XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY);

	memset(&args, 0, sizeof(args));
	args.tp = cur->bc_tp;
	args.mp = cur->bc_mp;
	args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno, sbno);
	args.minlen = 1;
	args.maxlen = 1;
	args.prod = 1;
	args.type = XFS_ALLOCTYPE_NEAR_BNO;

	error = xfs_alloc_vextent(&args);
	if (error) {
		XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR);
		return error;
	}
	if (args.fsbno == NULLFSBLOCK) {
		XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);
		*stat = 0;
		return 0;
	}
	ASSERT(args.len == 1);
	XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);

	new->s = cpu_to_be32(XFS_FSB_TO_AGBNO(args.mp, args.fsbno));
	*stat = 1;
	return 0;
}

STATIC int
xfs_inobt_free_block(
	struct xfs_btree_cur	*cur,
	struct xfs_buf		*bp)
{
	xfs_fsblock_t		fsbno;
	int			error;

	fsbno = XFS_DADDR_TO_FSB(cur->bc_mp, XFS_BUF_ADDR(bp));
	error = xfs_free_extent(cur->bc_tp, fsbno, 1);
	if (error)
		return error;

	xfs_trans_binval(cur->bc_tp, bp);
	return error;
}

STATIC int
xfs_inobt_get_maxrecs(
	struct xfs_btree_cur	*cur,
	int			level)
{
	return cur->bc_mp->m_inobt_mxr[level != 0];
}

STATIC void
xfs_inobt_init_key_from_rec(
	union xfs_btree_key	*key,
	union xfs_btree_rec	*rec)
{
	key->inobt.ir_startino = rec->inobt.ir_startino;
}

STATIC void
xfs_inobt_init_rec_from_key(
	union xfs_btree_key	*key,
	union xfs_btree_rec	*rec)
{
	rec->inobt.ir_startino = key->inobt.ir_startino;
}

STATIC void
xfs_inobt_init_rec_from_cur(
	struct xfs_btree_cur	*cur,
	union xfs_btree_rec	*rec)
{
	rec->inobt.ir_startino = cpu_to_be32(cur->bc_rec.i.ir_startino);
	if (xfs_sb_version_hassparseinodes(&cur->bc_mp->m_sb)) {
		rec->inobt.ir_u.sp.ir_holemask =
					cpu_to_be16(cur->bc_rec.i.ir_holemask);
		rec->inobt.ir_u.sp.ir_count = cur->bc_rec.i.ir_count;
		rec->inobt.ir_u.sp.ir_freecount = cur->bc_rec.i.ir_freecount;
	} else {
		/* ir_holemask/ir_count not supported on-disk */
		rec->inobt.ir_u.f.ir_freecount =
					cpu_to_be32(cur->bc_rec.i.ir_freecount);
	}
	rec->inobt.ir_free = cpu_to_be64(cur->bc_rec.i.ir_free);
}

/*
 * initial value of ptr for lookup
 */
STATIC void
xfs_inobt_init_ptr_from_cur(
	struct xfs_btree_cur	*cur,
	union xfs_btree_ptr	*ptr)
{
	struct xfs_agi		*agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);

	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno));

	ptr->s = agi->agi_root;
}

STATIC void
xfs_finobt_init_ptr_from_cur(
	struct xfs_btree_cur	*cur,
	union xfs_btree_ptr	*ptr)
{
	struct xfs_agi		*agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);

	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno));
	ptr->s = agi->agi_free_root;
}

STATIC __int64_t
xfs_inobt_key_diff(
	struct xfs_btree_cur	*cur,
	union xfs_btree_key	*key)
{
	return (__int64_t)be32_to_cpu(key->inobt.ir_startino) -
			  cur->bc_rec.i.ir_startino;
}

static int
xfs_inobt_verify(
	struct xfs_buf		*bp)
{
	struct xfs_mount	*mp = bp->b_target->bt_mount;
	struct xfs_btree_block	*block = XFS_BUF_TO_BLOCK(bp);
	struct xfs_perag	*pag = bp->b_pag;
	unsigned int		level;

	/*
	 * During growfs operations, we can't verify the exact owner as the
	 * perag is not fully initialised and hence not attached to the buffer.
	 *
	 * Similarly, during log recovery we will have a perag structure
	 * attached, but the agi information will not yet have been initialised
	 * from the on disk AGI. We don't currently use any of this information,
	 * but beware of the landmine (i.e. need to check pag->pagi_init) if we
	 * ever do.
	 */
	switch (block->bb_magic) {
	case cpu_to_be32(XFS_IBT_CRC_MAGIC):
	case cpu_to_be32(XFS_FIBT_CRC_MAGIC):
		if (!xfs_sb_version_hascrc(&mp->m_sb))
			return false;
		if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid))
			return false;
		if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
			return false;
		if (pag &&
		    be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
			return false;
		/* fall through */
	case cpu_to_be32(XFS_IBT_MAGIC):
	case cpu_to_be32(XFS_FIBT_MAGIC):
		break;
	default:
		return 0;
	}

	/* numrecs and level verification */
	level = be16_to_cpu(block->bb_level);
	if (level >= mp->m_in_maxlevels)
		return false;
	if (be16_to_cpu(block->bb_numrecs) > mp->m_inobt_mxr[level != 0])
		return false;

	/* sibling pointer verification */
	if (!block->bb_u.s.bb_leftsib ||
	    (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks &&
	     block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK)))
		return false;
	if (!block->bb_u.s.bb_rightsib ||
	    (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks &&
	     block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK)))
		return false;

	return true;
}

static void
xfs_inobt_read_verify(
	struct xfs_buf	*bp)
{
	if (!xfs_btree_sblock_verify_crc(bp))
		xfs_buf_ioerror(bp, -EFSBADCRC);
	else if (!xfs_inobt_verify(bp))
		xfs_buf_ioerror(bp, -EFSCORRUPTED);

	if (bp->b_error) {
		trace_xfs_btree_corrupt(bp, _RET_IP_);
		xfs_verifier_error(bp);
	}
}

static void
xfs_inobt_write_verify(
	struct xfs_buf	*bp)
{
	if (!xfs_inobt_verify(bp)) {
		trace_xfs_btree_corrupt(bp, _RET_IP_);
		xfs_buf_ioerror(bp, -EFSCORRUPTED);
		xfs_verifier_error(bp);
		return;
	}
	xfs_btree_sblock_calc_crc(bp);

}

const struct xfs_buf_ops xfs_inobt_buf_ops = {
	.name = "xfs_inobt",
	.verify_read = xfs_inobt_read_verify,
	.verify_write = xfs_inobt_write_verify,
};

#if defined(DEBUG) || defined(XFS_WARN)
STATIC int
xfs_inobt_keys_inorder(
	struct xfs_btree_cur	*cur,
	union xfs_btree_key	*k1,
	union xfs_btree_key	*k2)
{
	return be32_to_cpu(k1->inobt.ir_startino) <
		be32_to_cpu(k2->inobt.ir_startino);
}

STATIC int
xfs_inobt_recs_inorder(
	struct xfs_btree_cur	*cur,
	union xfs_btree_rec	*r1,
	union xfs_btree_rec	*r2)
{
	return be32_to_cpu(r1->inobt.ir_startino) + XFS_INODES_PER_CHUNK <=
		be32_to_cpu(r2->inobt.ir_startino);
}
#endif	/* DEBUG */

static const struct xfs_btree_ops xfs_inobt_ops = {
	.rec_len		= sizeof(xfs_inobt_rec_t),
	.key_len		= sizeof(xfs_inobt_key_t),

	.dup_cursor		= xfs_inobt_dup_cursor,
	.set_root		= xfs_inobt_set_root,
	.alloc_block		= xfs_inobt_alloc_block,
	.free_block		= xfs_inobt_free_block,
	.get_minrecs		= xfs_inobt_get_minrecs,
	.get_maxrecs		= xfs_inobt_get_maxrecs,
	.init_key_from_rec	= xfs_inobt_init_key_from_rec,
	.init_rec_from_key	= xfs_inobt_init_rec_from_key,
	.init_rec_from_cur	= xfs_inobt_init_rec_from_cur,
	.init_ptr_from_cur	= xfs_inobt_init_ptr_from_cur,
	.key_diff		= xfs_inobt_key_diff,
	.buf_ops		= &xfs_inobt_buf_ops,
#if defined(DEBUG) || defined(XFS_WARN)
	.keys_inorder		= xfs_inobt_keys_inorder,
	.recs_inorder		= xfs_inobt_recs_inorder,
#endif
};

static const struct xfs_btree_ops xfs_finobt_ops = {
	.rec_len		= sizeof(xfs_inobt_rec_t),
	.key_len		= sizeof(xfs_inobt_key_t),

	.dup_cursor		= xfs_inobt_dup_cursor,
	.set_root		= xfs_finobt_set_root,
	.alloc_block		= xfs_inobt_alloc_block,
	.free_block		= xfs_inobt_free_block,
	.get_minrecs		= xfs_inobt_get_minrecs,
	.get_maxrecs		= xfs_inobt_get_maxrecs,
	.init_key_from_rec	= xfs_inobt_init_key_from_rec,
	.init_rec_from_key	= xfs_inobt_init_rec_from_key,
	.init_rec_from_cur	= xfs_inobt_init_rec_from_cur,
	.init_ptr_from_cur	= xfs_finobt_init_ptr_from_cur,
	.key_diff		= xfs_inobt_key_diff,
	.buf_ops		= &xfs_inobt_buf_ops,
#if defined(DEBUG) || defined(XFS_WARN)
	.keys_inorder		= xfs_inobt_keys_inorder,
	.recs_inorder		= xfs_inobt_recs_inorder,
#endif
};

/*
 * Allocate a new inode btree cursor.
 */
struct xfs_btree_cur *				/* new inode btree cursor */
xfs_inobt_init_cursor(
	struct xfs_mount	*mp,		/* file system mount point */
	struct xfs_trans	*tp,		/* transaction pointer */
	struct xfs_buf		*agbp,		/* buffer for agi structure */
	xfs_agnumber_t		agno,		/* allocation group number */
	xfs_btnum_t		btnum)		/* ialloc or free ino btree */
{
	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
	struct xfs_btree_cur	*cur;

	cur = kmem_zone_zalloc(xfs_btree_cur_zone, KM_SLEEP);

	cur->bc_tp = tp;
	cur->bc_mp = mp;
	cur->bc_btnum = btnum;
	if (btnum == XFS_BTNUM_INO) {
		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
		cur->bc_ops = &xfs_inobt_ops;
	} else {
		cur->bc_nlevels = be32_to_cpu(agi->agi_free_level);
		cur->bc_ops = &xfs_finobt_ops;
	}

	cur->bc_blocklog = mp->m_sb.sb_blocklog;

	if (xfs_sb_version_hascrc(&mp->m_sb))
		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;

	cur->bc_private.a.agbp = agbp;
	cur->bc_private.a.agno = agno;

	return cur;
}

/*
 * Calculate number of records in an inobt btree block.
 */
int
xfs_inobt_maxrecs(
	struct xfs_mount	*mp,
	int			blocklen,
	int			leaf)
{
	blocklen -= XFS_INOBT_BLOCK_LEN(mp);

	if (leaf)
		return blocklen / sizeof(xfs_inobt_rec_t);
	return blocklen / (sizeof(xfs_inobt_key_t) + sizeof(xfs_inobt_ptr_t));
}

/*
 * Convert the inode record holemask to an inode allocation bitmap. The inode
 * allocation bitmap is inode granularity and specifies whether an inode is
 * physically allocated on disk (not whether the inode is considered allocated
 * or free by the fs).
 *
 * A bit value of 1 means the inode is allocated, a value of 0 means it is free.
 */
uint64_t
xfs_inobt_irec_to_allocmask(
	struct xfs_inobt_rec_incore	*rec)
{
	uint64_t			bitmap = 0;
	uint64_t			inodespbit;
	int				nextbit;
	uint				allocbitmap;

	/*
	 * The holemask has 16-bits for a 64 inode record. Therefore each
	 * holemask bit represents multiple inodes. Create a mask of bits to set
	 * in the allocmask for each holemask bit.
	 */
	inodespbit = (1 << XFS_INODES_PER_HOLEMASK_BIT) - 1;

	/*
	 * Allocated inodes are represented by 0 bits in holemask. Invert the 0
	 * bits to 1 and convert to a uint so we can use xfs_next_bit(). Mask
	 * anything beyond the 16 holemask bits since this casts to a larger
	 * type.
	 */
	allocbitmap = ~rec->ir_holemask & ((1 << XFS_INOBT_HOLEMASK_BITS) - 1);

	/*
	 * allocbitmap is the inverted holemask so every set bit represents
	 * allocated inodes. To expand from 16-bit holemask granularity to
	 * 64-bit (e.g., bit-per-inode), set inodespbit bits in the target
	 * bitmap for every holemask bit.
	 */
	nextbit = xfs_next_bit(&allocbitmap, 1, 0);
	while (nextbit != -1) {
		ASSERT(nextbit < (sizeof(rec->ir_holemask) * NBBY));

		bitmap |= (inodespbit <<
			   (nextbit * XFS_INODES_PER_HOLEMASK_BIT));

		nextbit = xfs_next_bit(&allocbitmap, 1, nextbit + 1);
	}

	return bitmap;
}

#if defined(DEBUG) || defined(XFS_WARN)
/*
 * Verify that an in-core inode record has a valid inode count.
 */
int
xfs_inobt_rec_check_count(
	struct xfs_mount		*mp,
	struct xfs_inobt_rec_incore	*rec)
{
	int				inocount = 0;
	int				nextbit = 0;
	uint64_t			allocbmap;
	int				wordsz;

	wordsz = sizeof(allocbmap) / sizeof(unsigned int);
	allocbmap = xfs_inobt_irec_to_allocmask(rec);

	nextbit = xfs_next_bit((uint *) &allocbmap, wordsz, nextbit);
	while (nextbit != -1) {
		inocount++;
		nextbit = xfs_next_bit((uint *) &allocbmap, wordsz,
				       nextbit + 1);
	}

	if (inocount != rec->ir_count)
		return -EFSCORRUPTED;

	return 0;
}
#endif	/* DEBUG */
Commit	Line	Data
1da177e4	1	/*
7b718769 NS	2	* Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
7b718769 NS	3	* All Rights Reserved.
1da177e4	4	*
7b718769 NS	5	* This program is free software; you can redistribute it and/or
7b718769 NS	6	* modify it under the terms of the GNU General Public License as
1da177e4 LT	7	* published by the Free Software Foundation.
1da177e4 LT	8	*
7b718769 NS	9	* This program is distributed in the hope that it would be useful,
	10	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	11	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	12	* GNU General Public License for more details.
1da177e4	13	*
7b718769 NS	14	* You should have received a copy of the GNU General Public License
	15	* along with this program; if not, write the Free Software Foundation,
	16	* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
1da177e4	17	*/
1da177e4	18	#include "xfs.h"
a844f451	19	#include "xfs_fs.h"
632b89e8	20	#include "xfs_shared.h"
a4fbe6ab	21	#include "xfs_format.h"
239880ef DC	22	#include "xfs_log_format.h"
239880ef DC	23	#include "xfs_trans_resv.h"
a844f451	24	#include "xfs_bit.h"
1da177e4	25	#include "xfs_mount.h"
a844f451	26	#include "xfs_inode.h"
1da177e4 LT	27	#include "xfs_btree.h"
1da177e4 LT	28	#include "xfs_ialloc.h"
a4fbe6ab	29	#include "xfs_ialloc_btree.h"
1da177e4 LT	30	#include "xfs_alloc.h"
1da177e4 LT	31	#include "xfs_error.h"
3d3e6f64	32	#include "xfs_trace.h"
ee1a47ab	33	#include "xfs_cksum.h"
239880ef	34	#include "xfs_trans.h"
1da177e4	35
1da177e4	36
91cca5df CH	37	STATIC int
	38	xfs_inobt_get_minrecs(
	39	struct xfs_btree_cur *cur,
	40	int level)
	41	{
	42	return cur->bc_mp->m_inobt_mnr[level != 0];
	43	}
1da177e4	44
561f7d17 CH	45	STATIC struct xfs_btree_cur *
	46	xfs_inobt_dup_cursor(
	47	struct xfs_btree_cur *cur)
	48	{
	49	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
57bd3dbe BF	50	cur->bc_private.a.agbp, cur->bc_private.a.agno,
57bd3dbe BF	51	cur->bc_btnum);
561f7d17 CH	52	}
561f7d17 CH	53
344207ce CH	54	STATIC void
	55	xfs_inobt_set_root(
	56	struct xfs_btree_cur *cur,
	57	union xfs_btree_ptr *nptr,
	58	int inc) /* level change */
	59	{
	60	struct xfs_buf *agbp = cur->bc_private.a.agbp;
	61	struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp);
	62
	63	agi->agi_root = nptr->s;
	64	be32_add_cpu(&agi->agi_level, inc);
	65	xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_ROOT \| XFS_AGI_LEVEL);
	66	}
	67
aafc3c24 BF	68	STATIC void
	69	xfs_finobt_set_root(
	70	struct xfs_btree_cur *cur,
	71	union xfs_btree_ptr *nptr,
	72	int inc) /* level change */
	73	{
	74	struct xfs_buf *agbp = cur->bc_private.a.agbp;
	75	struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp);
	76
	77	agi->agi_free_root = nptr->s;
	78	be32_add_cpu(&agi->agi_free_level, inc);
	79	xfs_ialloc_log_agi(cur->bc_tp, agbp,
	80	XFS_AGI_FREE_ROOT \| XFS_AGI_FREE_LEVEL);
	81	}
	82
f5eb8e7c CH	83	STATIC int
	84	xfs_inobt_alloc_block(
	85	struct xfs_btree_cur *cur,
	86	union xfs_btree_ptr *start,
	87	union xfs_btree_ptr *new,
f5eb8e7c CH	88	int *stat)
	89	{
	90	xfs_alloc_arg_t args; /* block allocation args */
	91	int error; /* error return value */
	92	xfs_agblock_t sbno = be32_to_cpu(start->s);
	93
	94	XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY);
	95
	96	memset(&args, 0, sizeof(args));
	97	args.tp = cur->bc_tp;
	98	args.mp = cur->bc_mp;
	99	args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno, sbno);
	100	args.minlen = 1;
	101	args.maxlen = 1;
	102	args.prod = 1;
	103	args.type = XFS_ALLOCTYPE_NEAR_BNO;
	104
	105	error = xfs_alloc_vextent(&args);
	106	if (error) {
	107	XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR);
	108	return error;
	109	}
	110	if (args.fsbno == NULLFSBLOCK) {
	111	XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);
	112	*stat = 0;
	113	return 0;
	114	}
	115	ASSERT(args.len == 1);
	116	XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);
	117
	118	new->s = cpu_to_be32(XFS_FSB_TO_AGBNO(args.mp, args.fsbno));
	119	*stat = 1;
	120	return 0;
	121	}
	122
d4b3a4b7 CH	123	STATIC int
	124	xfs_inobt_free_block(
	125	struct xfs_btree_cur *cur,
	126	struct xfs_buf *bp)
	127	{
	128	xfs_fsblock_t fsbno;
	129	int error;
	130
	131	fsbno = XFS_DADDR_TO_FSB(cur->bc_mp, XFS_BUF_ADDR(bp));
	132	error = xfs_free_extent(cur->bc_tp, fsbno, 1);
	133	if (error)
	134	return error;
	135
	136	xfs_trans_binval(cur->bc_tp, bp);
	137	return error;
	138	}
f5eb8e7c	139
ce5e42db CH	140	STATIC int
	141	xfs_inobt_get_maxrecs(
	142	struct xfs_btree_cur *cur,
	143	int level)
	144	{
	145	return cur->bc_mp->m_inobt_mxr[level != 0];
	146	}
	147
fe033cc8 CH	148	STATIC void
	149	xfs_inobt_init_key_from_rec(
	150	union xfs_btree_key *key,
	151	union xfs_btree_rec *rec)
	152	{
	153	key->inobt.ir_startino = rec->inobt.ir_startino;
	154	}
	155
4b22a571 CH	156	STATIC void
	157	xfs_inobt_init_rec_from_key(
	158	union xfs_btree_key *key,
	159	union xfs_btree_rec *rec)
	160	{
	161	rec->inobt.ir_startino = key->inobt.ir_startino;
	162	}
	163
	164	STATIC void
	165	xfs_inobt_init_rec_from_cur(
	166	struct xfs_btree_cur *cur,
	167	union xfs_btree_rec *rec)
	168	{
	169	rec->inobt.ir_startino = cpu_to_be32(cur->bc_rec.i.ir_startino);
5419040f BF	170	if (xfs_sb_version_hassparseinodes(&cur->bc_mp->m_sb)) {
	171	rec->inobt.ir_u.sp.ir_holemask =
	172	cpu_to_be16(cur->bc_rec.i.ir_holemask);
	173	rec->inobt.ir_u.sp.ir_count = cur->bc_rec.i.ir_count;
	174	rec->inobt.ir_u.sp.ir_freecount = cur->bc_rec.i.ir_freecount;
	175	} else {
	176	/* ir_holemask/ir_count not supported on-disk */
	177	rec->inobt.ir_u.f.ir_freecount =
	178	cpu_to_be32(cur->bc_rec.i.ir_freecount);
	179	}
4b22a571 CH	180	rec->inobt.ir_free = cpu_to_be64(cur->bc_rec.i.ir_free);
	181	}
	182
fe033cc8	183	/*
9da096fd	184	* initial value of ptr for lookup
fe033cc8 CH	185	*/
	186	STATIC void
	187	xfs_inobt_init_ptr_from_cur(
	188	struct xfs_btree_cur *cur,
	189	union xfs_btree_ptr *ptr)
	190	{
	191	struct xfs_agi *agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);
	192
	193	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno));
	194
	195	ptr->s = agi->agi_root;
	196	}
	197
aafc3c24 BF	198	STATIC void
	199	xfs_finobt_init_ptr_from_cur(
	200	struct xfs_btree_cur *cur,
	201	union xfs_btree_ptr *ptr)
	202	{
	203	struct xfs_agi *agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);
	204
	205	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno));
	206	ptr->s = agi->agi_free_root;
	207	}
	208
fe033cc8 CH	209	STATIC __int64_t
	210	xfs_inobt_key_diff(
	211	struct xfs_btree_cur *cur,
	212	union xfs_btree_key *key)
	213	{
	214	return (__int64_t)be32_to_cpu(key->inobt.ir_startino) -
	215	cur->bc_rec.i.ir_startino;
	216	}
	217
ee1a47ab	218	static int
612cfbfe	219	xfs_inobt_verify(
3d3e6f64 DC	220	struct xfs_buf *bp)
	221	{
	222	struct xfs_mount *mp = bp->b_target->bt_mount;
	223	struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
ee1a47ab	224	struct xfs_perag *pag = bp->b_pag;
3d3e6f64	225	unsigned int level;
3d3e6f64	226
ee1a47ab CH	227	/*
	228	* During growfs operations, we can't verify the exact owner as the
	229	* perag is not fully initialised and hence not attached to the buffer.
	230	*
	231	* Similarly, during log recovery we will have a perag structure
	232	* attached, but the agi information will not yet have been initialised
	233	* from the on disk AGI. We don't currently use any of this information,
	234	* but beware of the landmine (i.e. need to check pag->pagi_init) if we
	235	* ever do.
	236	*/
	237	switch (block->bb_magic) {
	238	case cpu_to_be32(XFS_IBT_CRC_MAGIC):
aafc3c24	239	case cpu_to_be32(XFS_FIBT_CRC_MAGIC):
ee1a47ab CH	240	if (!xfs_sb_version_hascrc(&mp->m_sb))
ee1a47ab CH	241	return false;
ce748eaa	242	if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid))
ee1a47ab CH	243	return false;
	244	if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
	245	return false;
	246	if (pag &&
	247	be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
	248	return false;
	249	/* fall through */
	250	case cpu_to_be32(XFS_IBT_MAGIC):
aafc3c24	251	case cpu_to_be32(XFS_FIBT_MAGIC):
ee1a47ab CH	252	break;
	253	default:
	254	return 0;
	255	}
3d3e6f64	256
ee1a47ab CH	257	/* numrecs and level verification */
	258	level = be16_to_cpu(block->bb_level);
	259	if (level >= mp->m_in_maxlevels)
	260	return false;
	261	if (be16_to_cpu(block->bb_numrecs) > mp->m_inobt_mxr[level != 0])
	262	return false;
3d3e6f64 DC	263
3d3e6f64 DC	264	/* sibling pointer verification */
ee1a47ab CH	265	if (!block->bb_u.s.bb_leftsib \|\|
	266	(be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks &&
	267	block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK)))
	268	return false;
	269	if (!block->bb_u.s.bb_rightsib \|\|
	270	(be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks &&
	271	block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK)))
	272	return false;
	273
	274	return true;
612cfbfe DC	275	}
	276
	277	static void
1813dd64	278	xfs_inobt_read_verify(
612cfbfe DC	279	struct xfs_buf *bp)
612cfbfe DC	280	{
ce5028cf	281	if (!xfs_btree_sblock_verify_crc(bp))
2451337d	282	xfs_buf_ioerror(bp, -EFSBADCRC);
ce5028cf	283	else if (!xfs_inobt_verify(bp))
2451337d	284	xfs_buf_ioerror(bp, -EFSCORRUPTED);
ce5028cf ES	285
	286	if (bp->b_error) {
	287	trace_xfs_btree_corrupt(bp, _RET_IP_);
	288	xfs_verifier_error(bp);
ee1a47ab	289	}
612cfbfe	290	}
3d3e6f64	291
1813dd64 DC	292	static void
1813dd64 DC	293	xfs_inobt_write_verify(
612cfbfe DC	294	struct xfs_buf *bp)
612cfbfe DC	295	{
ee1a47ab CH	296	if (!xfs_inobt_verify(bp)) {
ee1a47ab CH	297	trace_xfs_btree_corrupt(bp, _RET_IP_);
2451337d	298	xfs_buf_ioerror(bp, -EFSCORRUPTED);
ce5028cf	299	xfs_verifier_error(bp);
e0d2c23a	300	return;
ee1a47ab CH	301	}
	302	xfs_btree_sblock_calc_crc(bp);
	303
3d3e6f64 DC	304	}
3d3e6f64 DC	305
1813dd64	306	const struct xfs_buf_ops xfs_inobt_buf_ops = {
233135b7	307	.name = "xfs_inobt",
1813dd64 DC	308	.verify_read = xfs_inobt_read_verify,
	309	.verify_write = xfs_inobt_write_verify,
	310	};
	311
742ae1e3	312	#if defined(DEBUG) \|\| defined(XFS_WARN)
4a26e66e CH	313	STATIC int
	314	xfs_inobt_keys_inorder(
	315	struct xfs_btree_cur *cur,
	316	union xfs_btree_key *k1,
	317	union xfs_btree_key *k2)
	318	{
	319	return be32_to_cpu(k1->inobt.ir_startino) <
	320	be32_to_cpu(k2->inobt.ir_startino);
	321	}
	322
	323	STATIC int
	324	xfs_inobt_recs_inorder(
	325	struct xfs_btree_cur *cur,
	326	union xfs_btree_rec *r1,
	327	union xfs_btree_rec *r2)
	328	{
	329	return be32_to_cpu(r1->inobt.ir_startino) + XFS_INODES_PER_CHUNK <=
	330	be32_to_cpu(r2->inobt.ir_startino);
	331	}
	332	#endif /* DEBUG */
	333
561f7d17	334	static const struct xfs_btree_ops xfs_inobt_ops = {
65f1eaea CH	335	.rec_len = sizeof(xfs_inobt_rec_t),
	336	.key_len = sizeof(xfs_inobt_key_t),
	337
561f7d17	338	.dup_cursor = xfs_inobt_dup_cursor,
344207ce	339	.set_root = xfs_inobt_set_root,
f5eb8e7c	340	.alloc_block = xfs_inobt_alloc_block,
d4b3a4b7	341	.free_block = xfs_inobt_free_block,
91cca5df	342	.get_minrecs = xfs_inobt_get_minrecs,
ce5e42db	343	.get_maxrecs = xfs_inobt_get_maxrecs,
fe033cc8	344	.init_key_from_rec = xfs_inobt_init_key_from_rec,
4b22a571 CH	345	.init_rec_from_key = xfs_inobt_init_rec_from_key,
4b22a571 CH	346	.init_rec_from_cur = xfs_inobt_init_rec_from_cur,
fe033cc8 CH	347	.init_ptr_from_cur = xfs_inobt_init_ptr_from_cur,
fe033cc8 CH	348	.key_diff = xfs_inobt_key_diff,
1813dd64	349	.buf_ops = &xfs_inobt_buf_ops,
742ae1e3	350	#if defined(DEBUG) \|\| defined(XFS_WARN)
4a26e66e CH	351	.keys_inorder = xfs_inobt_keys_inorder,
	352	.recs_inorder = xfs_inobt_recs_inorder,
	353	#endif
561f7d17 CH	354	};
561f7d17 CH	355
aafc3c24 BF	356	static const struct xfs_btree_ops xfs_finobt_ops = {
	357	.rec_len = sizeof(xfs_inobt_rec_t),
	358	.key_len = sizeof(xfs_inobt_key_t),
	359
	360	.dup_cursor = xfs_inobt_dup_cursor,
	361	.set_root = xfs_finobt_set_root,
	362	.alloc_block = xfs_inobt_alloc_block,
	363	.free_block = xfs_inobt_free_block,
	364	.get_minrecs = xfs_inobt_get_minrecs,
	365	.get_maxrecs = xfs_inobt_get_maxrecs,
	366	.init_key_from_rec = xfs_inobt_init_key_from_rec,
	367	.init_rec_from_key = xfs_inobt_init_rec_from_key,
	368	.init_rec_from_cur = xfs_inobt_init_rec_from_cur,
	369	.init_ptr_from_cur = xfs_finobt_init_ptr_from_cur,
	370	.key_diff = xfs_inobt_key_diff,
	371	.buf_ops = &xfs_inobt_buf_ops,
	372	#if defined(DEBUG) \|\| defined(XFS_WARN)
	373	.keys_inorder = xfs_inobt_keys_inorder,
	374	.recs_inorder = xfs_inobt_recs_inorder,
	375	#endif
	376	};
	377
561f7d17 CH	378	/*
	379	* Allocate a new inode btree cursor.
	380	*/
	381	struct xfs_btree_cur * /* new inode btree cursor */
	382	xfs_inobt_init_cursor(
	383	struct xfs_mount mp, / file system mount point */
	384	struct xfs_trans tp, / transaction pointer */
	385	struct xfs_buf agbp, / buffer for agi structure */
57bd3dbe BF	386	xfs_agnumber_t agno, /* allocation group number */
57bd3dbe BF	387	xfs_btnum_t btnum) /* ialloc or free ino btree */
561f7d17 CH	388	{
	389	struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp);
	390	struct xfs_btree_cur *cur;
	391
	392	cur = kmem_zone_zalloc(xfs_btree_cur_zone, KM_SLEEP);
	393
	394	cur->bc_tp = tp;
	395	cur->bc_mp = mp;
57bd3dbe	396	cur->bc_btnum = btnum;
aafc3c24 BF	397	if (btnum == XFS_BTNUM_INO) {
	398	cur->bc_nlevels = be32_to_cpu(agi->agi_level);
	399	cur->bc_ops = &xfs_inobt_ops;
	400	} else {
	401	cur->bc_nlevels = be32_to_cpu(agi->agi_free_level);
	402	cur->bc_ops = &xfs_finobt_ops;
	403	}
	404
561f7d17 CH	405	cur->bc_blocklog = mp->m_sb.sb_blocklog;
561f7d17 CH	406
ee1a47ab CH	407	if (xfs_sb_version_hascrc(&mp->m_sb))
ee1a47ab CH	408	cur->bc_flags \|= XFS_BTREE_CRC_BLOCKS;
561f7d17 CH	409
	410	cur->bc_private.a.agbp = agbp;
	411	cur->bc_private.a.agno = agno;
	412
	413	return cur;
	414	}
60197e8d CH	415
	416	/*
	417	* Calculate number of records in an inobt btree block.
	418	*/
	419	int
	420	xfs_inobt_maxrecs(
	421	struct xfs_mount *mp,
	422	int blocklen,
	423	int leaf)
	424	{
7cc95a82	425	blocklen -= XFS_INOBT_BLOCK_LEN(mp);
60197e8d CH	426
	427	if (leaf)
	428	return blocklen / sizeof(xfs_inobt_rec_t);
	429	return blocklen / (sizeof(xfs_inobt_key_t) + sizeof(xfs_inobt_ptr_t));
	430	}
4148c347 BF	431
	432	/*
	433	* Convert the inode record holemask to an inode allocation bitmap. The inode
	434	* allocation bitmap is inode granularity and specifies whether an inode is
	435	* physically allocated on disk (not whether the inode is considered allocated
	436	* or free by the fs).
	437	*
	438	* A bit value of 1 means the inode is allocated, a value of 0 means it is free.
	439	*/
	440	uint64_t
	441	xfs_inobt_irec_to_allocmask(
	442	struct xfs_inobt_rec_incore *rec)
	443	{
	444	uint64_t bitmap = 0;
	445	uint64_t inodespbit;
	446	int nextbit;
	447	uint allocbitmap;
	448
	449	/*
	450	* The holemask has 16-bits for a 64 inode record. Therefore each
	451	* holemask bit represents multiple inodes. Create a mask of bits to set
	452	* in the allocmask for each holemask bit.
	453	*/
	454	inodespbit = (1 << XFS_INODES_PER_HOLEMASK_BIT) - 1;
	455
	456	/*
	457	* Allocated inodes are represented by 0 bits in holemask. Invert the 0
	458	* bits to 1 and convert to a uint so we can use xfs_next_bit(). Mask
	459	* anything beyond the 16 holemask bits since this casts to a larger
	460	* type.
	461	*/
	462	allocbitmap = ~rec->ir_holemask & ((1 << XFS_INOBT_HOLEMASK_BITS) - 1);
	463
	464	/*
	465	* allocbitmap is the inverted holemask so every set bit represents
	466	* allocated inodes. To expand from 16-bit holemask granularity to
	467	* 64-bit (e.g., bit-per-inode), set inodespbit bits in the target
	468	* bitmap for every holemask bit.
	469	*/
	470	nextbit = xfs_next_bit(&allocbitmap, 1, 0);
	471	while (nextbit != -1) {
	472	ASSERT(nextbit < (sizeof(rec->ir_holemask) * NBBY));
	473
	474	bitmap \|= (inodespbit <<
	475	(nextbit * XFS_INODES_PER_HOLEMASK_BIT));
	476
	477	nextbit = xfs_next_bit(&allocbitmap, 1, nextbit + 1);
	478	}
	479
	480	return bitmap;
	481	}
56d1115c BF	482
	483	#if defined(DEBUG) \|\| defined(XFS_WARN)
	484	/*
	485	* Verify that an in-core inode record has a valid inode count.
	486	*/
	487	int
	488	xfs_inobt_rec_check_count(
	489	struct xfs_mount *mp,
	490	struct xfs_inobt_rec_incore *rec)
	491	{
	492	int inocount = 0;
	493	int nextbit = 0;
	494	uint64_t allocbmap;
	495	int wordsz;
	496
	497	wordsz = sizeof(allocbmap) / sizeof(unsigned int);
	498	allocbmap = xfs_inobt_irec_to_allocmask(rec);
	499
	500	nextbit = xfs_next_bit((uint *) &allocbmap, wordsz, nextbit);
	501	while (nextbit != -1) {
	502	inocount++;
	503	nextbit = xfs_next_bit((uint *) &allocbmap, wordsz,
	504	nextbit + 1);
	505	}
	506
	507	if (inocount != rec->ir_count)
	508	return -EFSCORRUPTED;
	509
	510	return 0;
	511	}
	512	#endif /* DEBUG */