xfs_ialloc.c

linux 内核源代码

/*
 * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc.
 * All Rights Reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it would be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write the Free Software Foundation,
 * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 */
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_types.h"
#include "xfs_bit.h"
#include "xfs_log.h"
#include "xfs_inum.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_dir2.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
#include "xfs_bmap_btree.h"
#include "xfs_alloc_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_dir2_sf.h"
#include "xfs_attr_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_alloc.h"
#include "xfs_rtalloc.h"
#include "xfs_error.h"
#include "xfs_bmap.h"

/*
 * Log specified fields for the inode given by bp and off.
 */
STATIC void
xfs_ialloc_log_di(
	xfs_trans_t	*tp,		/* transaction pointer */
	xfs_buf_t	*bp,		/* inode buffer */
	int		off,		/* index of inode in buffer */
	int		fields)		/* bitmask of fields to log */
{
	int			first;		/* first byte number */
	int			ioffset;	/* off in bytes */
	int			last;		/* last byte number */
	xfs_mount_t		*mp;		/* mount point structure */
	static const short	offsets[] = {	/* field offsets */
						/* keep in sync with bits */
		offsetof(xfs_dinode_core_t, di_magic),
		offsetof(xfs_dinode_core_t, di_mode),
		offsetof(xfs_dinode_core_t, di_version),
		offsetof(xfs_dinode_core_t, di_format),
		offsetof(xfs_dinode_core_t, di_onlink),
		offsetof(xfs_dinode_core_t, di_uid),
		offsetof(xfs_dinode_core_t, di_gid),
		offsetof(xfs_dinode_core_t, di_nlink),
		offsetof(xfs_dinode_core_t, di_projid),
		offsetof(xfs_dinode_core_t, di_pad),
		offsetof(xfs_dinode_core_t, di_atime),
		offsetof(xfs_dinode_core_t, di_mtime),
		offsetof(xfs_dinode_core_t, di_ctime),
		offsetof(xfs_dinode_core_t, di_size),
		offsetof(xfs_dinode_core_t, di_nblocks),
		offsetof(xfs_dinode_core_t, di_extsize),
		offsetof(xfs_dinode_core_t, di_nextents),
		offsetof(xfs_dinode_core_t, di_anextents),
		offsetof(xfs_dinode_core_t, di_forkoff),
		offsetof(xfs_dinode_core_t, di_aformat),
		offsetof(xfs_dinode_core_t, di_dmevmask),
		offsetof(xfs_dinode_core_t, di_dmstate),
		offsetof(xfs_dinode_core_t, di_flags),
		offsetof(xfs_dinode_core_t, di_gen),
		offsetof(xfs_dinode_t, di_next_unlinked),
		offsetof(xfs_dinode_t, di_u),
		offsetof(xfs_dinode_t, di_a),
		sizeof(xfs_dinode_t)
	};


	ASSERT(offsetof(xfs_dinode_t, di_core) == 0);
	ASSERT((fields & (XFS_DI_U|XFS_DI_A)) == 0);
	mp = tp->t_mountp;
	/*
	 * Get the inode-relative first and last bytes for these fields
	 */
	xfs_btree_offsets(fields, offsets, XFS_DI_NUM_BITS, &first, &last);
	/*
	 * Convert to buffer offsets and log it.
	 */
	ioffset = off << mp->m_sb.sb_inodelog;
	first += ioffset;
	last += ioffset;
	xfs_trans_log_buf(tp, bp, first, last);
}

/*
 * Allocation group level functions.
 */

/*
 * 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 */

	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;
		args.alignment = 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 if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) &&
			   args.mp->m_sb.sb_inoalignmt >=
			   XFS_B_TO_FSBT(args.mp,
			  	XFS_INODE_CLUSTER_SIZE(args.mp)))
				args.alignment = args.mp->m_sb.sb_inoalignmt;
		else
			args.alignment = 1;
		/*
		 * 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);
		if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) &&
			args.mp->m_sb.sb_inoalignmt >=
			XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp)))
				args.alignment = args.mp->m_sb.sb_inoalignmt;
		else
			args.alignment = 1;
		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 = XFS_DINODE_VERSION_2;
	else
		version = XFS_DINODE_VERSION_1;

	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));
		/*
		 * Set initial values for the inodes in this buffer.
		 */
		xfs_biozero(fbuf, 0, ninodes << args.mp->m_sb.sb_inodelog);
		for (i = 0; i < ninodes; i++) {
			free = XFS_MAKE_IPTR(args.mp, fbuf, i);
			free->di_core.di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
			free->di_core.di_version = version;
			free->di_next_unlinked = cpu_to_be32(NULLAGINO);
			xfs_ialloc_log_di(tp, fbuf, i,
				XFS_DI_CORE_BITS | XFS_DI_NEXT_UNLINKED);
		}
		xfs_trans_inode_alloc_buf(tp, fbuf);
	}
	be32_add(&agi->agi_count, newlen);
	be32_add(&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_btree_init_cursor(args.mp, tp, agbp, agno,
			XFS_BTNUM_INO, (xfs_inode_t *)0, 0);
	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_inobt_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;
}

STATIC_INLINE xfs_agnumber_t
xfs_ialloc_next_ag(
	xfs_mount_t	*mp)
{
	xfs_agnumber_t	agno;

	spin_lock(&mp->m_agirotor_lock);
	agno = mp->m_agirotor;
	if (++mp->m_agirotor == mp->m_maxagi)
		mp->m_agirotor = 0;
	spin_unlock(&mp->m_agirotor_lock);

	return agno;
}

/*
 * Select an allocation group to look for a free inode in, based on the parent
 * inode and then mode.  Return the allocation group buffer.
 */
STATIC xfs_buf_t *			/* allocation group buffer */
xfs_ialloc_ag_select(
	xfs_trans_t	*tp,		/* transaction pointer */
	xfs_ino_t	parent,		/* parent directory inode number */
	mode_t		mode,		/* bits set to indicate file type */
	int		okalloc)	/* ok to allocate more space */
{
	xfs_buf_t	*agbp;		/* allocation group header buffer */
	xfs_agnumber_t	agcount;	/* number of ag's in the filesystem */
	xfs_agnumber_t	agno;		/* current ag number */
	int		flags;		/* alloc buffer locking flags */
	xfs_extlen_t	ineed;		/* blocks needed for inode allocation */
	xfs_extlen_t	longest = 0;	/* longest extent available */
	xfs_mount_t	*mp;		/* mount point structure */
	int		needspace;	/* file mode implies space allocated */
	xfs_perag_t	*pag;		/* per allocation group data */
	xfs_agnumber_t	pagno;		/* parent (starting) ag number */

	/*
	 * Files of these types need at least one block if length > 0
	 * (and they won't fit in the inode, but that's hard to figure out).
	 */
	needspace = S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode);
	mp = tp->t_mountp;
	agcount = mp->m_maxagi;
	if (S_ISDIR(mode))
		pagno = xfs_ialloc_next_ag(mp);
	else {
		pagno = XFS_INO_TO_AGNO(mp, parent);
		if (pagno >= agcount)
			pagno = 0;
	}
	ASSERT(pagno < agcount);
	/*
	 * Loop through allocation groups, looking for one with a little
	 * free space in it.  Note we don't look for free inodes, exactly.
	 * Instead, we include whether there is a need to allocate inodes
	 * to mean that blocks must be allocated for them,
	 * if none are currently free.
	 */
	agno = pagno;
	flags = XFS_ALLOC_FLAG_TRYLOCK;
	down_read(&mp->m_peraglock);
	for (;;) {
		pag = &mp->m_perag[agno];
		if (!pag->pagi_init) {
			if (xfs_ialloc_read_agi(mp, tp, agno, &agbp)) {
				agbp = NULL;
				goto nextag;
			}
		} else
			agbp = NULL;

		if (!pag->pagi_inodeok) {
			xfs_ialloc_next_ag(mp);
			goto unlock_nextag;
		}

		/*
		 * Is there enough free space for the file plus a block
		 * of inodes (if we need to allocate some)?
		 */
		ineed = pag->pagi_freecount ? 0 : XFS_IALLOC_BLOCKS(mp);
		if (ineed && !pag->pagf_init) {
			if (agbp == NULL &&
			    xfs_ialloc_read_agi(mp, tp, agno, &agbp)) {
				agbp = NULL;
				goto nextag;
			}
			(void)xfs_alloc_pagf_init(mp, tp, agno, flags);
		}
		if (!ineed || pag->pagf_init) {
			if (ineed && !(longest = pag->pagf_longest))
				longest = pag->pagf_flcount > 0;
			if (!ineed ||
			    (pag->pagf_freeblks >= needspace + ineed &&
			     longest >= ineed &&
			     okalloc)) {
				if (agbp == NULL &&
				    xfs_ialloc_read_agi(mp, tp, agno, &agbp)) {
					agbp = NULL;
					goto nextag;
				}
				up_read(&mp->m_peraglock);
				return agbp;
			}
		}
unlock_nextag:
		if (agbp)
			xfs_trans_brelse(tp, agbp);
nextag:
		/*
		 * No point in iterating over the rest, if we're shutting
		 * down.
		 */
		if (XFS_FORCED_SHUTDOWN(mp)) {
			up_read(&mp->m_peraglock);
			return NULL;
		}
		agno++;
		if (agno >= agcount)
			agno = 0;
		if (agno == pagno) {
			if (flags == 0) {
				up_read(&mp->m_peraglock);
				return NULL;
			}
			flags = 0;
		}
	}
}

/*
 * Visible inode allocation functions.
 */