xfs_mount.c

来自「Linux Kernel 2.6.9 for OMAP1710」· C语言 代码 · 共 1,582 行 · 第 1/3 页

/*
 * Copyright (c) 2000-2004 Silicon Graphics, Inc.  All Rights Reserved.
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of version 2 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.
 *
 * Further, this software is distributed without any warranty that it is
 * free of the rightful claim of any third person regarding infringement
 * or the like.  Any license provided herein, whether implied or
 * otherwise, applies only to this software file.  Patent licenses, if
 * any, provided herein do not apply to combinations of this program with
 * other software, or any other product whatsoever.
 *
 * You should have received a copy of the GNU General Public License along
 * with this program; if not, write the Free Software Foundation, Inc., 59
 * Temple Place - Suite 330, Boston MA 02111-1307, USA.
 *
 * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy,
 * Mountain View, CA  94043, or:
 *
 * http://www.sgi.com
 *
 * For further information regarding this notice, see:
 *
 * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/
 */

#include "xfs.h"
#include "xfs_macros.h"
#include "xfs_types.h"
#include "xfs_inum.h"
#include "xfs_log.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_dir.h"
#include "xfs_dir2.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
#include "xfs_alloc_btree.h"
#include "xfs_bmap_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_attr_sf.h"
#include "xfs_dir_sf.h"
#include "xfs_dir2_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_alloc.h"
#include "xfs_rtalloc.h"
#include "xfs_bmap.h"
#include "xfs_error.h"
#include "xfs_bit.h"
#include "xfs_rw.h"
#include "xfs_quota.h"
#include "xfs_fsops.h"

STATIC void	xfs_mount_log_sbunit(xfs_mount_t *, __int64_t);
STATIC int	xfs_uuid_mount(xfs_mount_t *);
STATIC void	xfs_uuid_unmount(xfs_mount_t *mp);

void xfs_xlatesb(void *, xfs_sb_t *, int, xfs_arch_t, __int64_t);

static struct {
    short offset;
    short type;     /* 0 = integer
		* 1 = binary / string (no translation)
		*/
} xfs_sb_info[] = {
    { offsetof(xfs_sb_t, sb_magicnum),   0 },
    { offsetof(xfs_sb_t, sb_blocksize),  0 },
    { offsetof(xfs_sb_t, sb_dblocks),    0 },
    { offsetof(xfs_sb_t, sb_rblocks),    0 },
    { offsetof(xfs_sb_t, sb_rextents),   0 },
    { offsetof(xfs_sb_t, sb_uuid),       1 },
    { offsetof(xfs_sb_t, sb_logstart),   0 },
    { offsetof(xfs_sb_t, sb_rootino),    0 },
    { offsetof(xfs_sb_t, sb_rbmino),     0 },
    { offsetof(xfs_sb_t, sb_rsumino),    0 },
    { offsetof(xfs_sb_t, sb_rextsize),   0 },
    { offsetof(xfs_sb_t, sb_agblocks),   0 },
    { offsetof(xfs_sb_t, sb_agcount),    0 },
    { offsetof(xfs_sb_t, sb_rbmblocks),  0 },
    { offsetof(xfs_sb_t, sb_logblocks),  0 },
    { offsetof(xfs_sb_t, sb_versionnum), 0 },
    { offsetof(xfs_sb_t, sb_sectsize),   0 },
    { offsetof(xfs_sb_t, sb_inodesize),  0 },
    { offsetof(xfs_sb_t, sb_inopblock),  0 },
    { offsetof(xfs_sb_t, sb_fname[0]),   1 },
    { offsetof(xfs_sb_t, sb_blocklog),   0 },
    { offsetof(xfs_sb_t, sb_sectlog),    0 },
    { offsetof(xfs_sb_t, sb_inodelog),   0 },
    { offsetof(xfs_sb_t, sb_inopblog),   0 },
    { offsetof(xfs_sb_t, sb_agblklog),   0 },
    { offsetof(xfs_sb_t, sb_rextslog),   0 },
    { offsetof(xfs_sb_t, sb_inprogress), 0 },
    { offsetof(xfs_sb_t, sb_imax_pct),   0 },
    { offsetof(xfs_sb_t, sb_icount),     0 },
    { offsetof(xfs_sb_t, sb_ifree),      0 },
    { offsetof(xfs_sb_t, sb_fdblocks),   0 },
    { offsetof(xfs_sb_t, sb_frextents),  0 },
    { offsetof(xfs_sb_t, sb_uquotino),   0 },
    { offsetof(xfs_sb_t, sb_gquotino),   0 },
    { offsetof(xfs_sb_t, sb_qflags),     0 },
    { offsetof(xfs_sb_t, sb_flags),      0 },
    { offsetof(xfs_sb_t, sb_shared_vn),  0 },
    { offsetof(xfs_sb_t, sb_inoalignmt), 0 },
    { offsetof(xfs_sb_t, sb_unit),	 0 },
    { offsetof(xfs_sb_t, sb_width),	 0 },
    { offsetof(xfs_sb_t, sb_dirblklog),	 0 },
    { offsetof(xfs_sb_t, sb_logsectlog), 0 },
    { offsetof(xfs_sb_t, sb_logsectsize),0 },
    { offsetof(xfs_sb_t, sb_logsunit),	 0 },
    { offsetof(xfs_sb_t, sb_features2),	 0 },
    { sizeof(xfs_sb_t),			 0 }
};

/*
 * Return a pointer to an initialized xfs_mount structure.
 */
xfs_mount_t *
xfs_mount_init(void)
{
	xfs_mount_t *mp;

	mp = kmem_zalloc(sizeof(*mp), KM_SLEEP);

	AIL_LOCKINIT(&mp->m_ail_lock, "xfs_ail");
	spinlock_init(&mp->m_sb_lock, "xfs_sb");
	mutex_init(&mp->m_ilock, MUTEX_DEFAULT, "xfs_ilock");
	initnsema(&mp->m_growlock, 1, "xfs_grow");
	/*
	 * Initialize the AIL.
	 */
	xfs_trans_ail_init(mp);

	atomic_set(&mp->m_active_trans, 0);

	return mp;
}

/*
 * Free up the resources associated with a mount structure.  Assume that
 * the structure was initially zeroed, so we can tell which fields got
 * initialized.
 */
void
xfs_mount_free(
	xfs_mount_t *mp,
	int	    remove_bhv)
{
	if (mp->m_ihash)
		xfs_ihash_free(mp);
	if (mp->m_chash)
		xfs_chash_free(mp);

	if (mp->m_perag) {
		int	agno;

		for (agno = 0; agno < mp->m_maxagi; agno++)
			if (mp->m_perag[agno].pagb_list)
				kmem_free(mp->m_perag[agno].pagb_list,
						sizeof(xfs_perag_busy_t) *
							XFS_PAGB_NUM_SLOTS);
		kmem_free(mp->m_perag,
			  sizeof(xfs_perag_t) * mp->m_sb.sb_agcount);
	}

	AIL_LOCK_DESTROY(&mp->m_ail_lock);
	spinlock_destroy(&mp->m_sb_lock);
	mutex_destroy(&mp->m_ilock);
	freesema(&mp->m_growlock);
	if (mp->m_quotainfo)
		XFS_QM_DONE(mp);

	if (mp->m_fsname != NULL)
		kmem_free(mp->m_fsname, mp->m_fsname_len);

	if (remove_bhv) {
		struct vfs	*vfsp = XFS_MTOVFS(mp);

		bhv_remove_all_vfsops(vfsp, 0);
		VFS_REMOVEBHV(vfsp, &mp->m_bhv);
	}

	kmem_free(mp, sizeof(xfs_mount_t));
}


/*
 * Check the validity of the SB found.
 */
STATIC int
xfs_mount_validate_sb(
	xfs_mount_t	*mp,
	xfs_sb_t	*sbp)
{
	/*
	 * If the log device and data device have the
	 * same device number, the log is internal.
	 * Consequently, the sb_logstart should be non-zero.  If
	 * we have a zero sb_logstart in this case, we may be trying to mount
	 * a volume filesystem in a non-volume manner.
	 */
	if (sbp->sb_magicnum != XFS_SB_MAGIC) {
		cmn_err(CE_WARN, "XFS: bad magic number");
		return XFS_ERROR(EWRONGFS);
	}

	if (!XFS_SB_GOOD_VERSION(sbp)) {
		cmn_err(CE_WARN, "XFS: bad version");
		return XFS_ERROR(EWRONGFS);
	}

	if (unlikely(
	    sbp->sb_logstart == 0 && mp->m_logdev_targp == mp->m_ddev_targp)) {
		cmn_err(CE_WARN,
	"XFS: filesystem is marked as having an external log; "
	"specify logdev on the\nmount command line.");
		XFS_CORRUPTION_ERROR("xfs_mount_validate_sb(1)",
				     XFS_ERRLEVEL_HIGH, mp, sbp);
		return XFS_ERROR(EFSCORRUPTED);
	}

	if (unlikely(
	    sbp->sb_logstart != 0 && mp->m_logdev_targp != mp->m_ddev_targp)) {
		cmn_err(CE_WARN,
	"XFS: filesystem is marked as having an internal log; "
	"don't specify logdev on\nthe mount command line.");
		XFS_CORRUPTION_ERROR("xfs_mount_validate_sb(2)",
				     XFS_ERRLEVEL_HIGH, mp, sbp);
		return XFS_ERROR(EFSCORRUPTED);
	}

	/*
	 * More sanity checking. These were stolen directly from
	 * xfs_repair.
	 */
	if (unlikely(
	    sbp->sb_agcount <= 0					||
	    sbp->sb_sectsize < XFS_MIN_SECTORSIZE			||
	    sbp->sb_sectsize > XFS_MAX_SECTORSIZE			||
	    sbp->sb_sectlog < XFS_MIN_SECTORSIZE_LOG			||
	    sbp->sb_sectlog > XFS_MAX_SECTORSIZE_LOG			||
	    sbp->sb_blocksize < XFS_MIN_BLOCKSIZE			||
	    sbp->sb_blocksize > XFS_MAX_BLOCKSIZE			||
	    sbp->sb_blocklog < XFS_MIN_BLOCKSIZE_LOG			||
	    sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG			||
	    sbp->sb_inodesize < XFS_DINODE_MIN_SIZE			||
	    sbp->sb_inodesize > XFS_DINODE_MAX_SIZE			||
	    (sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE)	||
	    (sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE)	||
	    sbp->sb_imax_pct > 100)) {
		cmn_err(CE_WARN, "XFS: SB sanity check 1 failed");
		XFS_CORRUPTION_ERROR("xfs_mount_validate_sb(3)",
				     XFS_ERRLEVEL_LOW, mp, sbp);
		return XFS_ERROR(EFSCORRUPTED);
	}

	/*
	 * Sanity check AG count, size fields against data size field
	 */
	if (unlikely(
	    sbp->sb_dblocks == 0 ||
	    sbp->sb_dblocks >
	     (xfs_drfsbno_t)sbp->sb_agcount * sbp->sb_agblocks ||
	    sbp->sb_dblocks < (xfs_drfsbno_t)(sbp->sb_agcount - 1) *
			      sbp->sb_agblocks + XFS_MIN_AG_BLOCKS)) {
		cmn_err(CE_WARN, "XFS: SB sanity check 2 failed");
		XFS_ERROR_REPORT("xfs_mount_validate_sb(4)",
				 XFS_ERRLEVEL_LOW, mp);
		return XFS_ERROR(EFSCORRUPTED);
	}

	ASSERT(PAGE_SHIFT >= sbp->sb_blocklog);
	ASSERT(sbp->sb_blocklog >= BBSHIFT);

#if XFS_BIG_BLKNOS     /* Limited by ULONG_MAX of page cache index */
	if (unlikely(
	    (sbp->sb_dblocks >> (PAGE_SHIFT - sbp->sb_blocklog)) > ULONG_MAX ||
	    (sbp->sb_rblocks >> (PAGE_SHIFT - sbp->sb_blocklog)) > ULONG_MAX)) {
#else                  /* Limited by UINT_MAX of sectors */
	if (unlikely(
	    (sbp->sb_dblocks << (sbp->sb_blocklog - BBSHIFT)) > UINT_MAX ||
	    (sbp->sb_rblocks << (sbp->sb_blocklog - BBSHIFT)) > UINT_MAX)) {
#endif
		cmn_err(CE_WARN,
	"XFS: File system is too large to be mounted on this system.");
		return XFS_ERROR(E2BIG);
	}

	if (unlikely(sbp->sb_inprogress)) {
		cmn_err(CE_WARN, "XFS: file system busy");
		XFS_ERROR_REPORT("xfs_mount_validate_sb(5)",
				 XFS_ERRLEVEL_LOW, mp);
		return XFS_ERROR(EFSCORRUPTED);
	}

	/*
	 * Until this is fixed only page-sized or smaller data blocks work.
	 */
	if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) {
		cmn_err(CE_WARN,
		"XFS: Attempted to mount file system with blocksize %d bytes",
			sbp->sb_blocksize);
		cmn_err(CE_WARN,
		"XFS: Only page-sized (%d) or less blocksizes currently work.",
			PAGE_SIZE);
		return XFS_ERROR(ENOSYS);
	}

	return 0;
}

xfs_agnumber_t
xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount)
{
	xfs_agnumber_t	index, max_metadata;
	xfs_perag_t	*pag;
	xfs_agino_t	agino;
	xfs_ino_t	ino;
	xfs_sb_t	*sbp = &mp->m_sb;
	xfs_ino_t	max_inum = XFS_MAXINUMBER_32;

	/* Check to see if the filesystem can overflow 32 bit inodes */
	agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
	ino = XFS_AGINO_TO_INO(mp, agcount - 1, agino);

	/* Clear the mount flag if no inode can overflow 32 bits
	 * on this filesystem, or if specifically requested..
	 */
	if ((mp->m_flags & XFS_MOUNT_32BITINOOPT) && ino > max_inum) {
		mp->m_flags |= XFS_MOUNT_32BITINODES;
	} else {
		mp->m_flags &= ~XFS_MOUNT_32BITINODES;
	}

	/* If we can overflow then setup the ag headers accordingly */
	if (mp->m_flags & XFS_MOUNT_32BITINODES) {
		/* Calculate how much should be reserved for inodes to
		 * meet the max inode percentage.
		 */
		if (mp->m_maxicount) {
			__uint64_t	icount;

			icount = sbp->sb_dblocks * sbp->sb_imax_pct;
			do_div(icount, 100);
			icount += sbp->sb_agblocks - 1;
			do_div(icount, mp->m_ialloc_blks);
			max_metadata = icount;
		} else {
			max_metadata = agcount;
		}
		for (index = 0; index < agcount; index++) {
			ino = XFS_AGINO_TO_INO(mp, index, agino);
			if (ino > max_inum) {
				index++;
				break;
			}

			/* This ag is prefered for inodes */
			pag = &mp->m_perag[index];
			pag->pagi_inodeok = 1;
			if (index < max_metadata)
				pag->pagf_metadata = 1;
		}
	} else {
		/* Setup default behavior for smaller filesystems */
		for (index = 0; index < agcount; index++) {
			pag = &mp->m_perag[index];
			pag->pagi_inodeok = 1;
		}
	}
	return index;
}

/*
 * xfs_xlatesb
 *
 *     data       - on disk version of sb
 *     sb         - a superblock
 *     dir        - conversion direction: <0 - convert sb to buf
 *                                        >0 - convert buf to sb
 *     arch       - architecture to read/write from/to buf
 *     fields     - which fields to copy (bitmask)
 */
void
xfs_xlatesb(
	void		*data,
	xfs_sb_t	*sb,
	int		dir,
	xfs_arch_t	arch,
	__int64_t	fields)
{
	xfs_caddr_t	buf_ptr;
	xfs_caddr_t	mem_ptr;
	xfs_sb_field_t	f;
	int		first;
	int		size;

	ASSERT(dir);
	ASSERT(fields);

	if (!fields)
		return;

	buf_ptr = (xfs_caddr_t)data;
	mem_ptr = (xfs_caddr_t)sb;

	while (fields) {
		f = (xfs_sb_field_t)xfs_lowbit64((__uint64_t)fields);
		first = xfs_sb_info[f].offset;
		size = xfs_sb_info[f + 1].offset - first;

		ASSERT(xfs_sb_info[f].type == 0 || xfs_sb_info[f].type == 1);

		if (arch == ARCH_NOCONVERT ||
		    size == 1 ||
		    xfs_sb_info[f].type == 1) {
			if (dir > 0) {
				memcpy(mem_ptr + first, buf_ptr + first, size);
			} else {
				memcpy(buf_ptr + first, mem_ptr + first, size);
			}
		} else {
			switch (size) {
			case 2:
				INT_XLATE(*(__uint16_t*)(buf_ptr+first),
					  *(__uint16_t*)(mem_ptr+first),
					  dir, arch);
				break;
			case 4:
				INT_XLATE(*(__uint32_t*)(buf_ptr+first),
					  *(__uint32_t*)(mem_ptr+first),
					  dir, arch);
				break;
			case 8:
				INT_XLATE(*(__uint64_t*)(buf_ptr+first),
					  *(__uint64_t*)(mem_ptr+first), dir, arch);
				break;
			default:
				ASSERT(0);
			}
		}

		fields &= ~(1LL << f);
	}
}

/*
 * xfs_readsb
 *
 * Does the initial read of the superblock.
 */
int
xfs_readsb(xfs_mount_t *mp)
{
	unsigned int	sector_size;
	unsigned int	extra_flags;
	xfs_buf_t	*bp;
	xfs_sb_t	*sbp;
	int		error;

	ASSERT(mp->m_sb_bp == NULL);
	ASSERT(mp->m_ddev_targp != NULL);

	/*
	 * Allocate a (locked) buffer to hold the superblock.
	 * This will be kept around at all times to optimize
	 * access to the superblock.
	 */
	sector_size = xfs_getsize_buftarg(mp->m_ddev_targp);
	extra_flags = XFS_BUF_LOCK | XFS_BUF_MANAGE | XFS_BUF_MAPPED;

	bp = xfs_buf_read_flags(mp->m_ddev_targp, XFS_SB_DADDR,
				BTOBB(sector_size), extra_flags);
	if (!bp || XFS_BUF_ISERROR(bp)) {
		cmn_err(CE_WARN, "XFS: SB read failed");
		error = bp ? XFS_BUF_GETERROR(bp) : ENOMEM;
		goto fail;
	}
	ASSERT(XFS_BUF_ISBUSY(bp));
	ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);

	/*
	 * Initialize the mount structure from the superblock.
	 * But first do some basic consistency checking.
	 */
	sbp = XFS_BUF_TO_SBP(bp);
	xfs_xlatesb(XFS_BUF_PTR(bp), &(mp->m_sb), 1,
				ARCH_CONVERT, XFS_SB_ALL_BITS);

	error = xfs_mount_validate_sb(mp, &(mp->m_sb));
	if (error) {
		cmn_err(CE_WARN, "XFS: SB validate failed");
		goto fail;
	}

	/*
	 * We must be able to do sector-sized and sector-aligned IO.
	 */
	if (sector_size > mp->m_sb.sb_sectsize) {
		cmn_err(CE_WARN,
			"XFS: device supports only %u byte sectors (not %u)",
			sector_size, mp->m_sb.sb_sectsize);
		error = ENOSYS;
		goto fail;
	}

	/*
	 * If device sector size is smaller than the superblock size,
	 * re-read the superblock so the buffer is correctly sized.
	 */
	if (sector_size < mp->m_sb.sb_sectsize) {
		XFS_BUF_UNMANAGE(bp);
		xfs_buf_relse(bp);
		sector_size = mp->m_sb.sb_sectsize;
		bp = xfs_buf_read_flags(mp->m_ddev_targp, XFS_SB_DADDR,
					BTOBB(sector_size), extra_flags);
		if (!bp || XFS_BUF_ISERROR(bp)) {
			cmn_err(CE_WARN, "XFS: SB re-read failed");
			error = bp ? XFS_BUF_GETERROR(bp) : ENOMEM;