xfs_lrw.c

linux 内核源代码

	/*
	 * First handle zeroing the block on which isize resides.
	 * We only zero a part of that block so it is handled specially.
	 */
	error = xfs_zero_last_block(ip, io, offset, isize);
	if (error) {
		ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
		ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));
		return error;
	}

	/*
	 * Calculate the range between the new size and the old
	 * where blocks needing to be zeroed may exist.  To get the
	 * block where the last byte in the file currently resides,
	 * we need to subtract one from the size and truncate back
	 * to a block boundary.  We subtract 1 in case the size is
	 * exactly on a block boundary.
	 */
	last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
	start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
	end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
	ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
	if (last_fsb == end_zero_fsb) {
		/*
		 * The size was only incremented on its last block.
		 * We took care of that above, so just return.
		 */
		return 0;
	}

	ASSERT(start_zero_fsb <= end_zero_fsb);
	while (start_zero_fsb <= end_zero_fsb) {
		nimaps = 1;
		zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
		error = XFS_BMAPI(mp, NULL, io, start_zero_fsb, zero_count_fsb,
				  0, NULL, 0, &imap, &nimaps, NULL, NULL);
		if (error) {
			ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
			ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));
			return error;
		}
		ASSERT(nimaps > 0);

		if (imap.br_state == XFS_EXT_UNWRITTEN ||
		    imap.br_startblock == HOLESTARTBLOCK) {
			/*
			 * This loop handles initializing pages that were
			 * partially initialized by the code below this
			 * loop. It basically zeroes the part of the page
			 * that sits on a hole and sets the page as P_HOLE
			 * and calls remapf if it is a mapped file.
			 */
			start_zero_fsb = imap.br_startoff + imap.br_blockcount;
			ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
			continue;
		}

		/*
		 * There are blocks we need to zero.
		 * Drop the inode lock while we're doing the I/O.
		 * We'll still have the iolock to protect us.
		 */
		XFS_IUNLOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);

		zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
		zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);

		if ((zero_off + zero_len) > offset)
			zero_len = offset - zero_off;

		error = xfs_iozero(ip, zero_off, zero_len);
		if (error) {
			goto out_lock;
		}

		start_zero_fsb = imap.br_startoff + imap.br_blockcount;
		ASSERT(start_zero_fsb <= (end_zero_fsb + 1));

		XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
	}

	return 0;

out_lock:

	XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
	ASSERT(error >= 0);
	return error;
}

ssize_t				/* bytes written, or (-) error */
xfs_write(
	struct xfs_inode	*xip,
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned int		nsegs,
	loff_t			*offset,
	int			ioflags)
{
	struct file		*file = iocb->ki_filp;
	struct address_space	*mapping = file->f_mapping;
	struct inode		*inode = mapping->host;
	bhv_vnode_t		*vp = XFS_ITOV(xip);
	unsigned long		segs = nsegs;
	xfs_mount_t		*mp;
	ssize_t			ret = 0, error = 0;
	xfs_fsize_t		isize, new_size;
	xfs_iocore_t		*io;
	int			iolock;
	int			eventsent = 0;
	bhv_vrwlock_t		locktype;
	size_t			ocount = 0, count;
	loff_t			pos;
	int			need_i_mutex;

	XFS_STATS_INC(xs_write_calls);

	error = generic_segment_checks(iovp, &segs, &ocount, VERIFY_READ);
	if (error)
		return error;

	count = ocount;
	pos = *offset;

	if (count == 0)
		return 0;

	io = &xip->i_iocore;
	mp = io->io_mount;

	xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);

	if (XFS_FORCED_SHUTDOWN(mp))
		return -EIO;

relock:
	if (ioflags & IO_ISDIRECT) {
		iolock = XFS_IOLOCK_SHARED;
		locktype = VRWLOCK_WRITE_DIRECT;
		need_i_mutex = 0;
	} else {
		iolock = XFS_IOLOCK_EXCL;
		locktype = VRWLOCK_WRITE;
		need_i_mutex = 1;
		mutex_lock(&inode->i_mutex);
	}

	xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);

start:
	error = -generic_write_checks(file, &pos, &count,
					S_ISBLK(inode->i_mode));
	if (error) {
		xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
		goto out_unlock_mutex;
	}

	if ((DM_EVENT_ENABLED(xip, DM_EVENT_WRITE) &&
	    !(ioflags & IO_INVIS) && !eventsent)) {
		int		dmflags = FILP_DELAY_FLAG(file);

		if (need_i_mutex)
			dmflags |= DM_FLAGS_IMUX;

		xfs_iunlock(xip, XFS_ILOCK_EXCL);
		error = XFS_SEND_DATA(xip->i_mount, DM_EVENT_WRITE, vp,
				      pos, count,
				      dmflags, &locktype);
		if (error) {
			goto out_unlock_internal;
		}
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		eventsent = 1;

		/*
		 * The iolock was dropped and reacquired in XFS_SEND_DATA
		 * so we have to recheck the size when appending.
		 * We will only "goto start;" once, since having sent the
		 * event prevents another call to XFS_SEND_DATA, which is
		 * what allows the size to change in the first place.
		 */
		if ((file->f_flags & O_APPEND) && pos != xip->i_size)
			goto start;
	}

	if (ioflags & IO_ISDIRECT) {
		xfs_buftarg_t	*target =
			(xip->i_d.di_flags & XFS_DIFLAG_REALTIME) ?
				mp->m_rtdev_targp : mp->m_ddev_targp;

		if ((pos & target->bt_smask) || (count & target->bt_smask)) {
			xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
			return XFS_ERROR(-EINVAL);
		}

		if (!need_i_mutex && (VN_CACHED(vp) || pos > xip->i_size)) {
			xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
			iolock = XFS_IOLOCK_EXCL;
			locktype = VRWLOCK_WRITE;
			need_i_mutex = 1;
			mutex_lock(&inode->i_mutex);
			xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);
			goto start;
		}
	}

	new_size = pos + count;
	if (new_size > xip->i_size)
		io->io_new_size = new_size;

	if (likely(!(ioflags & IO_INVIS))) {
		file_update_time(file);
		xfs_ichgtime_fast(xip, inode,
				  XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
	}

	/*
	 * If the offset is beyond the size of the file, we have a couple
	 * of things to do. First, if there is already space allocated
	 * we need to either create holes or zero the disk or ...
	 *
	 * If there is a page where the previous size lands, we need
	 * to zero it out up to the new size.
	 */

	if (pos > xip->i_size) {
		error = xfs_zero_eof(vp, io, pos, xip->i_size);
		if (error) {
			xfs_iunlock(xip, XFS_ILOCK_EXCL);
			goto out_unlock_internal;
		}
	}
	xfs_iunlock(xip, XFS_ILOCK_EXCL);

	/*
	 * If we're writing the file then make sure to clear the
	 * setuid and setgid bits if the process is not being run
	 * by root.  This keeps people from modifying setuid and
	 * setgid binaries.
	 */

	if (((xip->i_d.di_mode & S_ISUID) ||
	    ((xip->i_d.di_mode & (S_ISGID | S_IXGRP)) ==
		(S_ISGID | S_IXGRP))) &&
	     !capable(CAP_FSETID)) {
		error = xfs_write_clear_setuid(xip);
		if (likely(!error))
			error = -remove_suid(file->f_path.dentry);
		if (unlikely(error)) {
			goto out_unlock_internal;
		}
	}

retry:
	/* We can write back this queue in page reclaim */
	current->backing_dev_info = mapping->backing_dev_info;

	if ((ioflags & IO_ISDIRECT)) {
		if (VN_CACHED(vp)) {
			WARN_ON(need_i_mutex == 0);
			xfs_inval_cached_trace(io, pos, -1,
					ctooff(offtoct(pos)), -1);
			error = xfs_flushinval_pages(xip,
					ctooff(offtoct(pos)),
					-1, FI_REMAPF_LOCKED);
			if (error)
				goto out_unlock_internal;
		}

		if (need_i_mutex) {
			/* demote the lock now the cached pages are gone */
			XFS_ILOCK_DEMOTE(mp, io, XFS_IOLOCK_EXCL);
			mutex_unlock(&inode->i_mutex);

			iolock = XFS_IOLOCK_SHARED;
			locktype = VRWLOCK_WRITE_DIRECT;
			need_i_mutex = 0;
		}

 		xfs_rw_enter_trace(XFS_DIOWR_ENTER, io, (void *)iovp, segs,
				*offset, ioflags);
		ret = generic_file_direct_write(iocb, iovp,
				&segs, pos, offset, count, ocount);

		/*
		 * direct-io write to a hole: fall through to buffered I/O
		 * for completing the rest of the request.
		 */
		if (ret >= 0 && ret != count) {
			XFS_STATS_ADD(xs_write_bytes, ret);

			pos += ret;
			count -= ret;

			ioflags &= ~IO_ISDIRECT;
			xfs_iunlock(xip, iolock);
			goto relock;
		}
	} else {
		xfs_rw_enter_trace(XFS_WRITE_ENTER, io, (void *)iovp, segs,
				*offset, ioflags);
		ret = generic_file_buffered_write(iocb, iovp, segs,
				pos, offset, count, ret);
	}

	current->backing_dev_info = NULL;

	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
		ret = wait_on_sync_kiocb(iocb);

	if (ret == -ENOSPC &&
	    DM_EVENT_ENABLED(xip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
		xfs_rwunlock(xip, locktype);
		if (need_i_mutex)
			mutex_unlock(&inode->i_mutex);
		error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, vp,
				DM_RIGHT_NULL, vp, DM_RIGHT_NULL, NULL, NULL,
				0, 0, 0); /* Delay flag intentionally  unused */
		if (need_i_mutex)
			mutex_lock(&inode->i_mutex);
		xfs_rwlock(xip, locktype);
		if (error)
			goto out_unlock_internal;
		pos = xip->i_size;
		ret = 0;
		goto retry;
	}

	isize = i_size_read(inode);
	if (unlikely(ret < 0 && ret != -EFAULT && *offset > isize))
		*offset = isize;

	if (*offset > xip->i_size) {
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		if (*offset > xip->i_size)
			xip->i_size = *offset;
		xfs_iunlock(xip, XFS_ILOCK_EXCL);
	}

	error = -ret;
	if (ret <= 0)
		goto out_unlock_internal;

	XFS_STATS_ADD(xs_write_bytes, ret);

	/* Handle various SYNC-type writes */
	if ((file->f_flags & O_SYNC) || IS_SYNC(inode)) {
		int error2;
		xfs_rwunlock(xip, locktype);
		if (need_i_mutex)
			mutex_unlock(&inode->i_mutex);
		error2 = sync_page_range(inode, mapping, pos, ret);
		if (!error)
			error = error2;
		if (need_i_mutex)
			mutex_lock(&inode->i_mutex);
		xfs_rwlock(xip, locktype);
		error2 = xfs_write_sync_logforce(mp, xip);
		if (!error)
			error = error2;
	}

 out_unlock_internal:
	if (io->io_new_size) {
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		io->io_new_size = 0;
		/*
		 * If this was a direct or synchronous I/O that failed (such
		 * as ENOSPC) then part of the I/O may have been written to
		 * disk before the error occured.  In this case the on-disk
		 * file size may have been adjusted beyond the in-memory file
		 * size and now needs to be truncated back.
		 */
		if (xip->i_d.di_size > xip->i_size)
			xip->i_d.di_size = xip->i_size;
		xfs_iunlock(xip, XFS_ILOCK_EXCL);
	}
	xfs_rwunlock(xip, locktype);
 out_unlock_mutex:
	if (need_i_mutex)
		mutex_unlock(&inode->i_mutex);
	return -error;
}

/*
 * All xfs metadata buffers except log state machine buffers
 * get this attached as their b_bdstrat callback function.
 * This is so that we can catch a buffer
 * after prematurely unpinning it to forcibly shutdown the filesystem.
 */
int
xfs_bdstrat_cb(struct xfs_buf *bp)
{
	xfs_mount_t	*mp;

	mp = XFS_BUF_FSPRIVATE3(bp, xfs_mount_t *);
	if (!XFS_FORCED_SHUTDOWN(mp)) {
		xfs_buf_iorequest(bp);
		return 0;
	} else {
		xfs_buftrace("XFS__BDSTRAT IOERROR", bp);
		/*
		 * Metadata write that didn't get logged but
		 * written delayed anyway. These aren't associated
		 * with a transaction, and can be ignored.
		 */
		if (XFS_BUF_IODONE_FUNC(bp) == NULL &&
		    (XFS_BUF_ISREAD(bp)) == 0)
			return (xfs_bioerror_relse(bp));
		else
			return (xfs_bioerror(bp));
	}
}


int
xfs_bmap(
	xfs_inode_t	*ip,
	xfs_off_t	offset,
	ssize_t		count,
	int		flags,
	xfs_iomap_t	*iomapp,
	int		*niomaps)
{
	xfs_iocore_t	*io = &ip->i_iocore;

	ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
	ASSERT(((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) != 0) ==
	       ((ip->i_iocore.io_flags & XFS_IOCORE_RT) != 0));

	return xfs_iomap(io, offset, count, flags, iomapp, niomaps);
}

/*
 * Wrapper around bdstrat so that we can stop data
 * from going to disk in case we are shutting down the filesystem.
 * Typically user data goes thru this path; one of the exceptions
 * is the superblock.
 */
int
xfsbdstrat(
	struct xfs_mount	*mp,
	struct xfs_buf		*bp)
{
	ASSERT(mp);
	if (!XFS_FORCED_SHUTDOWN(mp)) {
		/* Grio redirection would go here
		 * if (XFS_BUF_IS_GRIO(bp)) {
		 */

		xfs_buf_iorequest(bp);
		return 0;
	}

	xfs_buftrace("XFSBDSTRAT IOERROR", bp);
	return (xfs_bioerror_relse(bp));
}

/*
 * If the underlying (data/log/rt) device is readonly, there are some
 * operations that cannot proceed.
 */
int
xfs_dev_is_read_only(
	xfs_mount_t		*mp,
	char			*message)
{
	if (xfs_readonly_buftarg(mp->m_ddev_targp) ||
	    xfs_readonly_buftarg(mp->m_logdev_targp) ||
	    (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
		cmn_err(CE_NOTE,
			"XFS: %s required on read-only device.", message);
		cmn_err(CE_NOTE,
			"XFS: write access unavailable, cannot proceed.");
		return EROFS;
	}
	return 0;
}