xfs_inode.c

linux 内核源代码

	xfs_buf_t	*bp;
	xfs_dinode_t	*dip;
	xfs_inode_t	*ip;
	int		error;

	ASSERT(xfs_inode_zone != NULL);

	ip = kmem_zone_zalloc(xfs_inode_zone, KM_SLEEP);
	ip->i_ino = ino;
	ip->i_mount = mp;
	atomic_set(&ip->i_iocount, 0);
	spin_lock_init(&ip->i_flags_lock);

	/*
	 * Get pointer's to the on-disk inode and the buffer containing it.
	 * If the inode number refers to a block outside the file system
	 * then xfs_itobp() will return NULL.  In this case we should
	 * return NULL as well.  Set i_blkno to 0 so that xfs_itobp() will
	 * know that this is a new incore inode.
	 */
	error = xfs_itobp(mp, tp, ip, &dip, &bp, bno, imap_flags);
	if (error) {
		kmem_zone_free(xfs_inode_zone, ip);
		return error;
	}

	/*
	 * Initialize inode's trace buffers.
	 * Do this before xfs_iformat in case it adds entries.
	 */
#ifdef	XFS_VNODE_TRACE
	ip->i_trace = ktrace_alloc(VNODE_TRACE_SIZE, KM_SLEEP);
#endif
#ifdef XFS_BMAP_TRACE
	ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_SLEEP);
#endif
#ifdef XFS_BMBT_TRACE
	ip->i_btrace = ktrace_alloc(XFS_BMBT_KTRACE_SIZE, KM_SLEEP);
#endif
#ifdef XFS_RW_TRACE
	ip->i_rwtrace = ktrace_alloc(XFS_RW_KTRACE_SIZE, KM_SLEEP);
#endif
#ifdef XFS_ILOCK_TRACE
	ip->i_lock_trace = ktrace_alloc(XFS_ILOCK_KTRACE_SIZE, KM_SLEEP);
#endif
#ifdef XFS_DIR2_TRACE
	ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_SLEEP);
#endif

	/*
	 * If we got something that isn't an inode it means someone
	 * (nfs or dmi) has a stale handle.
	 */
	if (be16_to_cpu(dip->di_core.di_magic) != XFS_DINODE_MAGIC) {
		kmem_zone_free(xfs_inode_zone, ip);
		xfs_trans_brelse(tp, bp);
#ifdef DEBUG
		xfs_fs_cmn_err(CE_ALERT, mp, "xfs_iread: "
				"dip->di_core.di_magic (0x%x) != "
				"XFS_DINODE_MAGIC (0x%x)",
				be16_to_cpu(dip->di_core.di_magic),
				XFS_DINODE_MAGIC);
#endif /* DEBUG */
		return XFS_ERROR(EINVAL);
	}

	/*
	 * If the on-disk inode is already linked to a directory
	 * entry, copy all of the inode into the in-core inode.
	 * xfs_iformat() handles copying in the inode format
	 * specific information.
	 * Otherwise, just get the truly permanent information.
	 */
	if (dip->di_core.di_mode) {
		xfs_dinode_from_disk(&ip->i_d, &dip->di_core);
		error = xfs_iformat(ip, dip);
		if (error)  {
			kmem_zone_free(xfs_inode_zone, ip);
			xfs_trans_brelse(tp, bp);
#ifdef DEBUG
			xfs_fs_cmn_err(CE_ALERT, mp, "xfs_iread: "
					"xfs_iformat() returned error %d",
					error);
#endif /* DEBUG */
			return error;
		}
	} else {
		ip->i_d.di_magic = be16_to_cpu(dip->di_core.di_magic);
		ip->i_d.di_version = dip->di_core.di_version;
		ip->i_d.di_gen = be32_to_cpu(dip->di_core.di_gen);
		ip->i_d.di_flushiter = be16_to_cpu(dip->di_core.di_flushiter);
		/*
		 * Make sure to pull in the mode here as well in
		 * case the inode is released without being used.
		 * This ensures that xfs_inactive() will see that
		 * the inode is already free and not try to mess
		 * with the uninitialized part of it.
		 */
		ip->i_d.di_mode = 0;
		/*
		 * Initialize the per-fork minima and maxima for a new
		 * inode here.  xfs_iformat will do it for old inodes.
		 */
		ip->i_df.if_ext_max =
			XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
	}

	INIT_LIST_HEAD(&ip->i_reclaim);

	/*
	 * The inode format changed when we moved the link count and
	 * made it 32 bits long.  If this is an old format inode,
	 * convert it in memory to look like a new one.  If it gets
	 * flushed to disk we will convert back before flushing or
	 * logging it.  We zero out the new projid field and the old link
	 * count field.  We'll handle clearing the pad field (the remains
	 * of the old uuid field) when we actually convert the inode to
	 * the new format. We don't change the version number so that we
	 * can distinguish this from a real new format inode.
	 */
	if (ip->i_d.di_version == XFS_DINODE_VERSION_1) {
		ip->i_d.di_nlink = ip->i_d.di_onlink;
		ip->i_d.di_onlink = 0;
		ip->i_d.di_projid = 0;
	}

	ip->i_delayed_blks = 0;
	ip->i_size = ip->i_d.di_size;

	/*
	 * Mark the buffer containing the inode as something to keep
	 * around for a while.  This helps to keep recently accessed
	 * meta-data in-core longer.
	 */
	 XFS_BUF_SET_REF(bp, XFS_INO_REF);

	/*
	 * Use xfs_trans_brelse() to release the buffer containing the
	 * on-disk inode, because it was acquired with xfs_trans_read_buf()
	 * in xfs_itobp() above.  If tp is NULL, this is just a normal
	 * brelse().  If we're within a transaction, then xfs_trans_brelse()
	 * will only release the buffer if it is not dirty within the
	 * transaction.  It will be OK to release the buffer in this case,
	 * because inodes on disk are never destroyed and we will be
	 * locking the new in-core inode before putting it in the hash
	 * table where other processes can find it.  Thus we don't have
	 * to worry about the inode being changed just because we released
	 * the buffer.
	 */
	xfs_trans_brelse(tp, bp);
	*ipp = ip;
	return 0;
}

/*
 * Read in extents from a btree-format inode.
 * Allocate and fill in if_extents.  Real work is done in xfs_bmap.c.
 */
int
xfs_iread_extents(
	xfs_trans_t	*tp,
	xfs_inode_t	*ip,
	int		whichfork)
{
	int		error;
	xfs_ifork_t	*ifp;
	xfs_extnum_t	nextents;
	size_t		size;

	if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
		XFS_ERROR_REPORT("xfs_iread_extents", XFS_ERRLEVEL_LOW,
				 ip->i_mount);
		return XFS_ERROR(EFSCORRUPTED);
	}
	nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
	size = nextents * sizeof(xfs_bmbt_rec_t);
	ifp = XFS_IFORK_PTR(ip, whichfork);

	/*
	 * We know that the size is valid (it's checked in iformat_btree)
	 */
	ifp->if_lastex = NULLEXTNUM;
	ifp->if_bytes = ifp->if_real_bytes = 0;
	ifp->if_flags |= XFS_IFEXTENTS;
	xfs_iext_add(ifp, 0, nextents);
	error = xfs_bmap_read_extents(tp, ip, whichfork);
	if (error) {
		xfs_iext_destroy(ifp);
		ifp->if_flags &= ~XFS_IFEXTENTS;
		return error;
	}
	xfs_validate_extents(ifp, nextents, XFS_EXTFMT_INODE(ip));
	return 0;
}

/*
 * Allocate an inode on disk and return a copy of its in-core version.
 * The in-core inode is locked exclusively.  Set mode, nlink, and rdev
 * appropriately within the inode.  The uid and gid for the inode are
 * set according to the contents of the given cred structure.
 *
 * Use xfs_dialloc() to allocate the on-disk inode. If xfs_dialloc()
 * has a free inode available, call xfs_iget()
 * to obtain the in-core version of the allocated inode.  Finally,
 * fill in the inode and log its initial contents.  In this case,
 * ialloc_context would be set to NULL and call_again set to false.
 *
 * If xfs_dialloc() does not have an available inode,
 * it will replenish its supply by doing an allocation. Since we can
 * only do one allocation within a transaction without deadlocks, we
 * must commit the current transaction before returning the inode itself.
 * In this case, therefore, we will set call_again to true and return.
 * The caller should then commit the current transaction, start a new
 * transaction, and call xfs_ialloc() again to actually get the inode.
 *
 * To ensure that some other process does not grab the inode that
 * was allocated during the first call to xfs_ialloc(), this routine
 * also returns the [locked] bp pointing to the head of the freelist
 * as ialloc_context.  The caller should hold this buffer across
 * the commit and pass it back into this routine on the second call.
 *
 * If we are allocating quota inodes, we do not have a parent inode
 * to attach to or associate with (i.e. pip == NULL) because they
 * are not linked into the directory structure - they are attached
 * directly to the superblock - and so have no parent.
 */
int
xfs_ialloc(
	xfs_trans_t	*tp,
	xfs_inode_t	*pip,
	mode_t		mode,
	xfs_nlink_t	nlink,
	xfs_dev_t	rdev,
	cred_t		*cr,
	xfs_prid_t	prid,
	int		okalloc,
	xfs_buf_t	**ialloc_context,
	boolean_t	*call_again,
	xfs_inode_t	**ipp)
{
	xfs_ino_t	ino;
	xfs_inode_t	*ip;
	bhv_vnode_t	*vp;
	uint		flags;
	int		error;

	/*
	 * Call the space management code to pick
	 * the on-disk inode to be allocated.
	 */
	error = xfs_dialloc(tp, pip ? pip->i_ino : 0, mode, okalloc,
			    ialloc_context, call_again, &ino);
	if (error != 0) {
		return error;
	}
	if (*call_again || ino == NULLFSINO) {
		*ipp = NULL;
		return 0;
	}
	ASSERT(*ialloc_context == NULL);

	/*
	 * Get the in-core inode with the lock held exclusively.
	 * This is because we're setting fields here we need
	 * to prevent others from looking at until we're done.
	 */
	error = xfs_trans_iget(tp->t_mountp, tp, ino,
				XFS_IGET_CREATE, XFS_ILOCK_EXCL, &ip);
	if (error != 0) {
		return error;
	}
	ASSERT(ip != NULL);

	vp = XFS_ITOV(ip);
	ip->i_d.di_mode = (__uint16_t)mode;
	ip->i_d.di_onlink = 0;
	ip->i_d.di_nlink = nlink;
	ASSERT(ip->i_d.di_nlink == nlink);
	ip->i_d.di_uid = current_fsuid(cr);
	ip->i_d.di_gid = current_fsgid(cr);
	ip->i_d.di_projid = prid;
	memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));

	/*
	 * If the superblock version is up to where we support new format
	 * inodes and this is currently an old format inode, then change
	 * the inode version number now.  This way we only do the conversion
	 * here rather than here and in the flush/logging code.
	 */
	if (XFS_SB_VERSION_HASNLINK(&tp->t_mountp->m_sb) &&
	    ip->i_d.di_version == XFS_DINODE_VERSION_1) {
		ip->i_d.di_version = XFS_DINODE_VERSION_2;
		/*
		 * We've already zeroed the old link count, the projid field,
		 * and the pad field.
		 */
	}

	/*
	 * Project ids won't be stored on disk if we are using a version 1 inode.
	 */
	if ((prid != 0) && (ip->i_d.di_version == XFS_DINODE_VERSION_1))
		xfs_bump_ino_vers2(tp, ip);

	if (pip && XFS_INHERIT_GID(pip)) {
		ip->i_d.di_gid = pip->i_d.di_gid;
		if ((pip->i_d.di_mode & S_ISGID) && (mode & S_IFMT) == S_IFDIR) {
			ip->i_d.di_mode |= S_ISGID;
		}
	}

	/*
	 * If the group ID of the new file does not match the effective group
	 * ID or one of the supplementary group IDs, the S_ISGID bit is cleared
	 * (and only if the irix_sgid_inherit compatibility variable is set).
	 */
	if ((irix_sgid_inherit) &&
	    (ip->i_d.di_mode & S_ISGID) &&
	    (!in_group_p((gid_t)ip->i_d.di_gid))) {
		ip->i_d.di_mode &= ~S_ISGID;
	}

	ip->i_d.di_size = 0;
	ip->i_size = 0;
	ip->i_d.di_nextents = 0;
	ASSERT(ip->i_d.di_nblocks == 0);
	xfs_ichgtime(ip, XFS_ICHGTIME_CHG|XFS_ICHGTIME_ACC|XFS_ICHGTIME_MOD);
	/*
	 * di_gen will have been taken care of in xfs_iread.
	 */
	ip->i_d.di_extsize = 0;
	ip->i_d.di_dmevmask = 0;
	ip->i_d.di_dmstate = 0;
	ip->i_d.di_flags = 0;
	flags = XFS_ILOG_CORE;
	switch (mode & S_IFMT) {
	case S_IFIFO:
	case S_IFCHR:
	case S_IFBLK:
	case S_IFSOCK:
		ip->i_d.di_format = XFS_DINODE_FMT_DEV;
		ip->i_df.if_u2.if_rdev = rdev;
		ip->i_df.if_flags = 0;
		flags |= XFS_ILOG_DEV;
		break;
	case S_IFREG:
		if (pip && xfs_inode_is_filestream(pip)) {
			error = xfs_filestream_associate(pip, ip);
			if (error < 0)
				return -error;
			if (!error)
				xfs_iflags_set(ip, XFS_IFILESTREAM);
		}
		/* fall through */
	case S_IFDIR:
		if (pip && (pip->i_d.di_flags & XFS_DIFLAG_ANY)) {
			uint	di_flags = 0;

			if ((mode & S_IFMT) == S_IFDIR) {
				if (pip->i_d.di_flags & XFS_DIFLAG_RTINHERIT)
					di_flags |= XFS_DIFLAG_RTINHERIT;
				if (pip->i_d.di_flags & XFS_DIFLAG_EXTSZINHERIT) {
					di_flags |= XFS_DIFLAG_EXTSZINHERIT;
					ip->i_d.di_extsize = pip->i_d.di_extsize;
				}
			} else if ((mode & S_IFMT) == S_IFREG) {
				if (pip->i_d.di_flags & XFS_DIFLAG_RTINHERIT) {
					di_flags |= XFS_DIFLAG_REALTIME;
					ip->i_iocore.io_flags |= XFS_IOCORE_RT;
				}
				if (pip->i_d.di_flags & XFS_DIFLAG_EXTSZINHERIT) {
					di_flags |= XFS_DIFLAG_EXTSIZE;
					ip->i_d.di_extsize = pip->i_d.di_extsize;
				}
			}
			if ((pip->i_d.di_flags & XFS_DIFLAG_NOATIME) &&
			    xfs_inherit_noatime)
				di_flags |= XFS_DIFLAG_NOATIME;
			if ((pip->i_d.di_flags & XFS_DIFLAG_NODUMP) &&
			    xfs_inherit_nodump)
				di_flags |= XFS_DIFLAG_NODUMP;
			if ((pip->i_d.di_flags & XFS_DIFLAG_SYNC) &&
			    xfs_inherit_sync)
				di_flags |= XFS_DIFLAG_SYNC;
			if ((pip->i_d.di_flags & XFS_DIFLAG_NOSYMLINKS) &&
			    xfs_inherit_nosymlinks)
				di_flags |= XFS_DIFLAG_NOSYMLINKS;
			if (pip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
				di_flags |= XFS_DIFLAG_PROJINHERIT;
			if ((pip->i_d.di_flags & XFS_DIFLAG_NODEFRAG) &&
			    xfs_inherit_nodefrag)
				di_flags |= XFS_DIFLAG_NODEFRAG;
			if (pip->i_d.di_flags & XFS_DIFLAG_FILESTREAM)
				di_flags |= XFS_DIFLAG_FILESTREAM;
			ip->i_d.di_flags |= di_flags;
		}
		/* FALLTHROUGH */
	case S_IFLNK:
		ip->i_d.di_format = XFS_DINODE_FMT_EXTENTS;
		ip->i_df.if_flags = XFS_IFEXTENTS;
		ip->i_df.if_bytes = ip->i_df.if_real_bytes = 0;
		ip->i_df.if_u1.if_extents = NULL;
		break;
	default:
		ASSERT(0);
	}
	/*
	 * Attribute fork settings for new inode.
	 */
	ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
	ip->i_d.di_anextents = 0;

	/*
	 * Log the new values stuffed into the inode.
	 */
	xfs_trans_log_inode(tp, ip, flags);

	/* now that we have an i_mode we can setup inode ops and unlock */
	xfs_initialize_vnode(tp->t_mountp, vp, ip);

	*ipp = ip;
	return 0;
}

/*
 * Check to make sure that there are no blocks allocated to the
 * file beyond the size of the file.  We don't check this for
 * files with fixed size extents or real time extents, but we