xfs_vnodeops.c

linux 内核源代码

/*
 * Copyright (c) 2000-2006 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_da_btree.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_inode_item.h"
#include "xfs_itable.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_alloc.h"
#include "xfs_bmap.h"
#include "xfs_attr.h"
#include "xfs_rw.h"
#include "xfs_error.h"
#include "xfs_quota.h"
#include "xfs_utils.h"
#include "xfs_rtalloc.h"
#include "xfs_refcache.h"
#include "xfs_trans_space.h"
#include "xfs_log_priv.h"
#include "xfs_filestream.h"
#include "xfs_vnodeops.h"

int
xfs_open(
	xfs_inode_t	*ip)
{
	int		mode;

	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
		return XFS_ERROR(EIO);

	/*
	 * If it's a directory with any blocks, read-ahead block 0
	 * as we're almost certain to have the next operation be a read there.
	 */
	if (S_ISDIR(ip->i_d.di_mode) && ip->i_d.di_nextents > 0) {
		mode = xfs_ilock_map_shared(ip);
		if (ip->i_d.di_nextents > 0)
			(void)xfs_da_reada_buf(NULL, ip, 0, XFS_DATA_FORK);
		xfs_iunlock(ip, mode);
	}
	return 0;
}

/*
 * xfs_getattr
 */
int
xfs_getattr(
	xfs_inode_t	*ip,
	bhv_vattr_t	*vap,
	int		flags)
{
	bhv_vnode_t	*vp = XFS_ITOV(ip);
	xfs_mount_t	*mp = ip->i_mount;

	vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address);

	if (XFS_FORCED_SHUTDOWN(mp))
		return XFS_ERROR(EIO);

	if (!(flags & ATTR_LAZY))
		xfs_ilock(ip, XFS_ILOCK_SHARED);

	vap->va_size = XFS_ISIZE(ip);
	if (vap->va_mask == XFS_AT_SIZE)
		goto all_done;

	vap->va_nblocks =
		XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks);
	vap->va_nodeid = ip->i_ino;
#if XFS_BIG_INUMS
	vap->va_nodeid += mp->m_inoadd;
#endif
	vap->va_nlink = ip->i_d.di_nlink;

	/*
	 * Quick exit for non-stat callers
	 */
	if ((vap->va_mask &
	    ~(XFS_AT_SIZE|XFS_AT_FSID|XFS_AT_NODEID|
	      XFS_AT_NLINK|XFS_AT_BLKSIZE)) == 0)
		goto all_done;

	/*
	 * Copy from in-core inode.
	 */
	vap->va_mode = ip->i_d.di_mode;
	vap->va_uid = ip->i_d.di_uid;
	vap->va_gid = ip->i_d.di_gid;
	vap->va_projid = ip->i_d.di_projid;

	/*
	 * Check vnode type block/char vs. everything else.
	 */
	switch (ip->i_d.di_mode & S_IFMT) {
	case S_IFBLK:
	case S_IFCHR:
		vap->va_rdev = ip->i_df.if_u2.if_rdev;
		vap->va_blocksize = BLKDEV_IOSIZE;
		break;
	default:
		vap->va_rdev = 0;

		if (!(ip->i_d.di_flags & XFS_DIFLAG_REALTIME)) {
			vap->va_blocksize = xfs_preferred_iosize(mp);
		} else {

			/*
			 * If the file blocks are being allocated from a
			 * realtime partition, then return the inode's
			 * realtime extent size or the realtime volume's
			 * extent size.
			 */
			vap->va_blocksize =
				xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog;
		}
		break;
	}

	vn_atime_to_timespec(vp, &vap->va_atime);
	vap->va_mtime.tv_sec = ip->i_d.di_mtime.t_sec;
	vap->va_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
	vap->va_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
	vap->va_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;

	/*
	 * Exit for stat callers.  See if any of the rest of the fields
	 * to be filled in are needed.
	 */
	if ((vap->va_mask &
	     (XFS_AT_XFLAGS|XFS_AT_EXTSIZE|XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS|
	      XFS_AT_GENCOUNT|XFS_AT_VCODE)) == 0)
		goto all_done;

	/*
	 * Convert di_flags to xflags.
	 */
	vap->va_xflags = xfs_ip2xflags(ip);

	/*
	 * Exit for inode revalidate.  See if any of the rest of
	 * the fields to be filled in are needed.
	 */
	if ((vap->va_mask &
	     (XFS_AT_EXTSIZE|XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS|
	      XFS_AT_GENCOUNT|XFS_AT_VCODE)) == 0)
		goto all_done;

	vap->va_extsize = ip->i_d.di_extsize << mp->m_sb.sb_blocklog;
	vap->va_nextents =
		(ip->i_df.if_flags & XFS_IFEXTENTS) ?
			ip->i_df.if_bytes / sizeof(xfs_bmbt_rec_t) :
			ip->i_d.di_nextents;
	if (ip->i_afp)
		vap->va_anextents =
			(ip->i_afp->if_flags & XFS_IFEXTENTS) ?
				ip->i_afp->if_bytes / sizeof(xfs_bmbt_rec_t) :
				 ip->i_d.di_anextents;
	else
		vap->va_anextents = 0;
	vap->va_gen = ip->i_d.di_gen;

 all_done:
	if (!(flags & ATTR_LAZY))
		xfs_iunlock(ip, XFS_ILOCK_SHARED);
	return 0;
}


/*
 * xfs_setattr
 */
int
xfs_setattr(
	xfs_inode_t		*ip,
	bhv_vattr_t		*vap,
	int			flags,
	cred_t			*credp)
{
	bhv_vnode_t		*vp = XFS_ITOV(ip);
	xfs_mount_t		*mp = ip->i_mount;
	xfs_trans_t		*tp;
	int			mask;
	int			code;
	uint			lock_flags;
	uint			commit_flags=0;
	uid_t			uid=0, iuid=0;
	gid_t			gid=0, igid=0;
	int			timeflags = 0;
	xfs_prid_t		projid=0, iprojid=0;
	int			mandlock_before, mandlock_after;
	struct xfs_dquot	*udqp, *gdqp, *olddquot1, *olddquot2;
	int			file_owner;
	int			need_iolock = 1;

	vn_trace_entry(ip, __FUNCTION__, (inst_t *)__return_address);

	if (mp->m_flags & XFS_MOUNT_RDONLY)
		return XFS_ERROR(EROFS);

	/*
	 * Cannot set certain attributes.
	 */
	mask = vap->va_mask;
	if (mask & XFS_AT_NOSET) {
		return XFS_ERROR(EINVAL);
	}

	if (XFS_FORCED_SHUTDOWN(mp))
		return XFS_ERROR(EIO);

	/*
	 * Timestamps do not need to be logged and hence do not
	 * need to be done within a transaction.
	 */
	if (mask & XFS_AT_UPDTIMES) {
		ASSERT((mask & ~XFS_AT_UPDTIMES) == 0);
		timeflags = ((mask & XFS_AT_UPDATIME) ? XFS_ICHGTIME_ACC : 0) |
			    ((mask & XFS_AT_UPDCTIME) ? XFS_ICHGTIME_CHG : 0) |
			    ((mask & XFS_AT_UPDMTIME) ? XFS_ICHGTIME_MOD : 0);
		xfs_ichgtime(ip, timeflags);
		return 0;
	}

	olddquot1 = olddquot2 = NULL;
	udqp = gdqp = NULL;

	/*
	 * If disk quotas is on, we make sure that the dquots do exist on disk,
	 * before we start any other transactions. Trying to do this later
	 * is messy. We don't care to take a readlock to look at the ids
	 * in inode here, because we can't hold it across the trans_reserve.
	 * If the IDs do change before we take the ilock, we're covered
	 * because the i_*dquot fields will get updated anyway.
	 */
	if (XFS_IS_QUOTA_ON(mp) &&
	    (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID))) {
		uint	qflags = 0;

		if ((mask & XFS_AT_UID) && XFS_IS_UQUOTA_ON(mp)) {
			uid = vap->va_uid;
			qflags |= XFS_QMOPT_UQUOTA;
		} else {
			uid = ip->i_d.di_uid;
		}
		if ((mask & XFS_AT_GID) && XFS_IS_GQUOTA_ON(mp)) {
			gid = vap->va_gid;
			qflags |= XFS_QMOPT_GQUOTA;
		}  else {
			gid = ip->i_d.di_gid;
		}
		if ((mask & XFS_AT_PROJID) && XFS_IS_PQUOTA_ON(mp)) {
			projid = vap->va_projid;
			qflags |= XFS_QMOPT_PQUOTA;
		}  else {
			projid = ip->i_d.di_projid;
		}
		/*
		 * We take a reference when we initialize udqp and gdqp,
		 * so it is important that we never blindly double trip on
		 * the same variable. See xfs_create() for an example.
		 */
		ASSERT(udqp == NULL);
		ASSERT(gdqp == NULL);
		code = XFS_QM_DQVOPALLOC(mp, ip, uid, gid, projid, qflags,
					 &udqp, &gdqp);
		if (code)
			return code;
	}

	/*
	 * For the other attributes, we acquire the inode lock and
	 * first do an error checking pass.
	 */
	tp = NULL;
	lock_flags = XFS_ILOCK_EXCL;
	if (flags & ATTR_NOLOCK)
		need_iolock = 0;
	if (!(mask & XFS_AT_SIZE)) {
		if ((mask != (XFS_AT_CTIME|XFS_AT_ATIME|XFS_AT_MTIME)) ||
		    (mp->m_flags & XFS_MOUNT_WSYNC)) {
			tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
			commit_flags = 0;
			if ((code = xfs_trans_reserve(tp, 0,
						     XFS_ICHANGE_LOG_RES(mp), 0,
						     0, 0))) {
				lock_flags = 0;
				goto error_return;
			}
		}
	} else {
		if (DM_EVENT_ENABLED(ip, DM_EVENT_TRUNCATE) &&
		    !(flags & ATTR_DMI)) {
			int dmflags = AT_DELAY_FLAG(flags) | DM_SEM_FLAG_WR;
			code = XFS_SEND_DATA(mp, DM_EVENT_TRUNCATE, vp,
				vap->va_size, 0, dmflags, NULL);
			if (code) {
				lock_flags = 0;
				goto error_return;
			}
		}
		if (need_iolock)
			lock_flags |= XFS_IOLOCK_EXCL;
	}

	xfs_ilock(ip, lock_flags);

	/* boolean: are we the file owner? */
	file_owner = (current_fsuid(credp) == ip->i_d.di_uid);

	/*
	 * Change various properties of a file.
	 * Only the owner or users with CAP_FOWNER
	 * capability may do these things.
	 */
	if (mask &
	    (XFS_AT_MODE|XFS_AT_XFLAGS|XFS_AT_EXTSIZE|XFS_AT_UID|
	     XFS_AT_GID|XFS_AT_PROJID)) {
		/*
		 * CAP_FOWNER overrides the following restrictions:
		 *
		 * The user ID of the calling process must be equal
		 * to the file owner ID, except in cases where the
		 * CAP_FSETID capability is applicable.
		 */
		if (!file_owner && !capable(CAP_FOWNER)) {
			code = XFS_ERROR(EPERM);
			goto error_return;
		}

		/*
		 * CAP_FSETID overrides the following restrictions:
		 *
		 * The effective user ID of the calling process shall match
		 * the file owner when setting the set-user-ID and
		 * set-group-ID bits on that file.
		 *
		 * The effective group ID or one of the supplementary group
		 * IDs of the calling process shall match the group owner of
		 * the file when setting the set-group-ID bit on that file
		 */
		if (mask & XFS_AT_MODE) {
			mode_t m = 0;

			if ((vap->va_mode & S_ISUID) && !file_owner)
				m |= S_ISUID;
			if ((vap->va_mode & S_ISGID) &&
			    !in_group_p((gid_t)ip->i_d.di_gid))
				m |= S_ISGID;
#if 0
			/* Linux allows this, Irix doesn't. */
			if ((vap->va_mode & S_ISVTX) && !VN_ISDIR(vp))
				m |= S_ISVTX;
#endif
			if (m && !capable(CAP_FSETID))
				vap->va_mode &= ~m;
		}
	}

	/*
	 * Change file ownership.  Must be the owner or privileged.
	 * If the system was configured with the "restricted_chown"
	 * option, the owner is not permitted to give away the file,
	 * and can change the group id only to a group of which he
	 * or she is a member.
	 */
	if (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID)) {
		/*
		 * These IDs could have changed since we last looked at them.
		 * But, we're assured that if the ownership did change
		 * while we didn't have the inode locked, inode's dquot(s)
		 * would have changed also.
		 */
		iuid = ip->i_d.di_uid;
		iprojid = ip->i_d.di_projid;
		igid = ip->i_d.di_gid;
		gid = (mask & XFS_AT_GID) ? vap->va_gid : igid;
		uid = (mask & XFS_AT_UID) ? vap->va_uid : iuid;
		projid = (mask & XFS_AT_PROJID) ? (xfs_prid_t)vap->va_projid :
			 iprojid;

		/*
		 * CAP_CHOWN overrides the following restrictions:
		 *
		 * If _POSIX_CHOWN_RESTRICTED is defined, this capability
		 * shall override the restriction that a process cannot
		 * change the user ID of a file it owns and the restriction
		 * that the group ID supplied to the chown() function
		 * shall be equal to either the group ID or one of the
		 * supplementary group IDs of the calling process.
		 */
		if (restricted_chown &&
		    (iuid != uid || (igid != gid &&
				     !in_group_p((gid_t)gid))) &&
		    !capable(CAP_CHOWN)) {
			code = XFS_ERROR(EPERM);
			goto error_return;
		}
		/*
		 * Do a quota reservation only if uid/projid/gid is actually
		 * going to change.
		 */
		if ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) ||
		    (XFS_IS_PQUOTA_ON(mp) && iprojid != projid) ||
		    (XFS_IS_GQUOTA_ON(mp) && igid != gid)) {
			ASSERT(tp);
			code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp,
						capable(CAP_FOWNER) ?
						XFS_QMOPT_FORCE_RES : 0);
			if (code)	/* out of quota */
				goto error_return;
		}
	}

	/*
	 * Truncate file.  Must have write permission and not be a directory.
	 */
	if (mask & XFS_AT_SIZE) {
		/* Short circuit the truncate case for zero length files */
		if ((vap->va_size == 0) &&
		   (ip->i_size == 0) && (ip->i_d.di_nextents == 0)) {
			xfs_iunlock(ip, XFS_ILOCK_EXCL);
			lock_flags &= ~XFS_ILOCK_EXCL;
			if (mask & XFS_AT_CTIME)
				xfs_ichgtime(ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
			code = 0;
			goto error_return;
		}

		if (VN_ISDIR(vp)) {
			code = XFS_ERROR(EISDIR);
			goto error_return;
		} else if (!VN_ISREG(vp)) {
			code = XFS_ERROR(EINVAL);
			goto error_return;
		}
		/*
		 * Make sure that the dquots are attached to the inode.
		 */
		if ((code = XFS_QM_DQATTACH(mp, ip, XFS_QMOPT_ILOCKED)))
			goto error_return;
	}

	/*
	 * Change file access or modified times.
	 */
	if (mask & (XFS_AT_ATIME|XFS_AT_MTIME)) {
		if (!file_owner) {
			if ((flags & ATTR_UTIME) &&
			    !capable(CAP_FOWNER)) {
				code = XFS_ERROR(EPERM);
				goto error_return;
			}
		}
	}

	/*
	 * Change extent size or realtime flag.
	 */
	if (mask & (XFS_AT_EXTSIZE|XFS_AT_XFLAGS)) {
		/*
		 * Can't change extent size if any extents are allocated.
		 */
		if (ip->i_d.di_nextents && (mask & XFS_AT_EXTSIZE) &&
		    ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=