Ejemplo n.º 1
0
STATIC ssize_t
xfs_file_buffered_aio_write(
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned long		nr_segs,
	loff_t			pos,
	size_t			ocount)
{
	struct file		*file = iocb->ki_filp;
	struct address_space	*mapping = file->f_mapping;
	struct inode		*inode = mapping->host;
	struct xfs_inode	*ip = XFS_I(inode);
	ssize_t			ret;
	int			enospc = 0;
	int			iolock = XFS_IOLOCK_EXCL;
	size_t			count = ocount;

	xfs_rw_ilock(ip, iolock);

	ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
	if (ret)
		goto out;

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

write_retry:
	trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0);
	ret = generic_file_buffered_write(iocb, iovp, nr_segs,
			pos, &iocb->ki_pos, count, 0);

	/*
	 * If we just got an ENOSPC, try to write back all dirty inodes to
	 * convert delalloc space to free up some of the excess reserved
	 * metadata space.
	 */
	if (ret == -ENOSPC && !enospc) {
		enospc = 1;
		xfs_flush_inodes(ip->i_mount);
		goto write_retry;
	}

	current->backing_dev_info = NULL;
out:
	xfs_rw_iunlock(ip, iolock);
	return ret;
}
Ejemplo n.º 2
0
STATIC ssize_t
xfs_file_buffered_aio_write(
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned long		nr_segs,
	loff_t			pos,
	size_t			ocount)
{
	struct file		*file = iocb->ki_filp;
	struct address_space	*mapping = file->f_mapping;
	struct inode		*inode = mapping->host;
	struct xfs_inode	*ip = XFS_I(inode);
	ssize_t			ret;
	int			enospc = 0;
	int			iolock = XFS_IOLOCK_EXCL;
	size_t			count = ocount;

	xfs_rw_ilock(ip, iolock);

	ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
	if (ret)
		goto out;

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

write_retry:
	trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0);
	ret = generic_file_buffered_write(iocb, iovp, nr_segs,
			pos, &iocb->ki_pos, count, ret);
	/*
	 * if we just got an ENOSPC, flush the inode now we aren't holding any
	 * page locks and retry *once*
	 */
	if (ret == -ENOSPC && !enospc) {
		enospc = 1;
		ret = -xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
		if (!ret)
			goto write_retry;
	}

	current->backing_dev_info = NULL;
out:
	xfs_rw_iunlock(ip, iolock);
	return ret;
}
Ejemplo n.º 3
0
STATIC ssize_t
xfs_file_aio_write(
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned long		nr_segs,
	loff_t			pos)
{
	struct file		*file = iocb->ki_filp;
	struct address_space	*mapping = file->f_mapping;
	struct inode		*inode = mapping->host;
	struct xfs_inode	*ip = XFS_I(inode);
	struct xfs_mount	*mp = ip->i_mount;
	ssize_t			ret = 0, error = 0;
	int			ioflags = 0;
	xfs_fsize_t		isize, new_size;
	int			iolock;
	int			eventsent = 0;
	size_t			ocount = 0, count;
	int			need_i_mutex;

	XFS_STATS_INC(xs_write_calls);

	BUG_ON(iocb->ki_pos != pos);

	if (unlikely(file->f_flags & O_DIRECT))
		ioflags |= IO_ISDIRECT;
	if (file->f_mode & FMODE_NOCMTIME)
		ioflags |= IO_INVIS;

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

	count = ocount;
	if (count == 0)
		return 0;

	xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);

	if (XFS_FORCED_SHUTDOWN(mp))
		return -EIO;

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

	xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);

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

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

		if (need_i_mutex)
			dmflags |= DM_FLAGS_IMUX;

		xfs_iunlock(ip, XFS_ILOCK_EXCL);
		error = XFS_SEND_DATA(ip->i_mount, DM_EVENT_WRITE, ip,
				      pos, count, dmflags, &iolock);
		if (error) {
			goto out_unlock_internal;
		}
		xfs_ilock(ip, 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 != ip->i_size)
			goto start;
	}

	if (ioflags & IO_ISDIRECT) {
		xfs_buftarg_t	*target =
			XFS_IS_REALTIME_INODE(ip) ?
				mp->m_rtdev_targp : mp->m_ddev_targp;

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

		if (!need_i_mutex && (mapping->nrpages || pos > ip->i_size)) {
			xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
			iolock = XFS_IOLOCK_EXCL;
			need_i_mutex = 1;
			mutex_lock(&inode->i_mutex);
			xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
			goto start;
		}
	}

	new_size = pos + count;
	if (new_size > ip->i_size)
		ip->i_new_size = new_size;

	if (likely(!(ioflags & IO_INVIS)))
		file_update_time(file);

	/*
	 * 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 > ip->i_size) {
		error = xfs_zero_eof(ip, pos, ip->i_size);
		if (error) {
			xfs_iunlock(ip, XFS_ILOCK_EXCL);
			goto out_unlock_internal;
		}
	}
	xfs_iunlock(ip, 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.
	 */
	error = -file_remove_suid(file);
	if (unlikely(error))
		goto out_unlock_internal;

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

	if ((ioflags & IO_ISDIRECT)) {
		if (mapping->nrpages) {
			WARN_ON(need_i_mutex == 0);
			error = xfs_flushinval_pages(ip,
					(pos & PAGE_CACHE_MASK),
					-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(ip, XFS_IOLOCK_EXCL);
			mutex_unlock(&inode->i_mutex);

			iolock = XFS_IOLOCK_SHARED;
			need_i_mutex = 0;
		}

		trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags);
		ret = generic_file_direct_write(iocb, iovp,
				&nr_segs, pos, &iocb->ki_pos, 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(ip, iolock);
			goto relock;
		}
	} else {
		int enospc = 0;
		ssize_t ret2 = 0;

write_retry:
		trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, ioflags);
		ret2 = generic_file_buffered_write(iocb, iovp, nr_segs,
				pos, &iocb->ki_pos, count, ret);
		/*
		 * if we just got an ENOSPC, flush the inode now we
		 * aren't holding any page locks and retry *once*
		 */
		if (ret2 == -ENOSPC && !enospc) {
			error = xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
			if (error)
				goto out_unlock_internal;
			enospc = 1;
			goto write_retry;
		}
		ret = ret2;
	}

	current->backing_dev_info = NULL;

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

	if (iocb->ki_pos > ip->i_size) {
		xfs_ilock(ip, XFS_ILOCK_EXCL);
		if (iocb->ki_pos > ip->i_size)
			ip->i_size = iocb->ki_pos;
		xfs_iunlock(ip, XFS_ILOCK_EXCL);
	}

	if (ret == -ENOSPC &&
	    DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
		xfs_iunlock(ip, iolock);
		if (need_i_mutex)
			mutex_unlock(&inode->i_mutex);
		error = XFS_SEND_NAMESP(ip->i_mount, DM_EVENT_NOSPACE, ip,
				DM_RIGHT_NULL, ip, DM_RIGHT_NULL, NULL, NULL,
				0, 0, 0); /* Delay flag intentionally  unused */
		if (need_i_mutex)
			mutex_lock(&inode->i_mutex);
		xfs_ilock(ip, iolock);
		if (error)
			goto out_unlock_internal;
		goto start;
	}

	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_DSYNC) || IS_SYNC(inode)) {
		loff_t end = pos + ret - 1;
		int error2;

		xfs_iunlock(ip, iolock);
		if (need_i_mutex)
			mutex_unlock(&inode->i_mutex);

		error2 = filemap_write_and_wait_range(mapping, pos, end);
		if (!error)
			error = error2;
		if (need_i_mutex)
			mutex_lock(&inode->i_mutex);
		xfs_ilock(ip, iolock);

		error2 = -xfs_file_fsync(file, file->f_path.dentry,
					 (file->f_flags & __O_SYNC) ? 0 : 1);
		if (!error)
			error = error2;
	}

 out_unlock_internal:
	if (ip->i_new_size) {
		xfs_ilock(ip, XFS_ILOCK_EXCL);
		ip->i_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 (ip->i_d.di_size > ip->i_size)
			ip->i_d.di_size = ip->i_size;
		xfs_iunlock(ip, XFS_ILOCK_EXCL);
	}
	xfs_iunlock(ip, iolock);
 out_unlock_mutex:
	if (need_i_mutex)
		mutex_unlock(&inode->i_mutex);
	return -error;
}
Ejemplo n.º 4
0
ssize_t				/* bytes written, or (-) error */
xfs_write(
	bhv_desc_t		*bdp,
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned int		nsegs,
	loff_t			*offset,
	int			ioflags,
	cred_t			*credp)
{
	struct file		*file = iocb->ki_filp;
	struct address_space	*mapping = file->f_mapping;
	struct inode		*inode = mapping->host;
	unsigned long		segs = nsegs;
	xfs_inode_t		*xip;
	xfs_mount_t		*mp;
	ssize_t			ret = 0, error = 0;
	xfs_fsize_t		isize, new_size;
	xfs_iocore_t		*io;
	vnode_t			*vp;
	unsigned long		seg;
	int			iolock;
	int			eventsent = 0;
	vrwlock_t		locktype;
	size_t			ocount = 0, count;
	loff_t			pos;
	int			need_isem = 1, need_flush = 0;

	XFS_STATS_INC(xs_write_calls);

	vp = BHV_TO_VNODE(bdp);
	xip = XFS_BHVTOI(bdp);

	for (seg = 0; seg < segs; seg++) {
		const struct iovec *iv = &iovp[seg];

		/*
		 * If any segment has a negative length, or the cumulative
		 * length ever wraps negative then return -EINVAL.
		 */
		ocount += iv->iov_len;
		if (unlikely((ssize_t)(ocount|iv->iov_len) < 0))
			return -EINVAL;
		if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len))
			continue;
		if (seg == 0)
			return -EFAULT;
		segs = seg;
		ocount -= iv->iov_len;  /* This segment is no good */
		break;
	}

	count = ocount;
	pos = *offset;

	if (count == 0)
		return 0;

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

	if (XFS_FORCED_SHUTDOWN(mp))
		return -EIO;

	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->pbr_smask) || (count & target->pbr_smask))
			return XFS_ERROR(-EINVAL);

		if (!VN_CACHED(vp) && pos < i_size_read(inode))
			need_isem = 0;

		if (VN_CACHED(vp))
			need_flush = 1;
	}

relock:
	if (need_isem) {
		iolock = XFS_IOLOCK_EXCL;
		locktype = VRWLOCK_WRITE;

		down(&inode->i_sem);
	} else {
		iolock = XFS_IOLOCK_SHARED;
		locktype = VRWLOCK_WRITE_DIRECT;
	}

	xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);

	isize = i_size_read(inode);

	if (file->f_flags & O_APPEND)
		*offset = isize;

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_isem;
	}

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

	if ((DM_EVENT_ENABLED(vp->v_vfsp, xip, DM_EVENT_WRITE) &&
	    !(ioflags & IO_INVIS) && !eventsent)) {
		loff_t		savedsize = pos;
		int		dmflags = FILP_DELAY_FLAG(file);

		if (need_isem)
			dmflags |= DM_FLAGS_ISEM;

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

		/*
		 * The iolock was dropped and reaquired 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) && savedsize != isize) {
			pos = isize = xip->i_d.di_size;
			goto start;
		}
	}

	/*
	 * On Linux, generic_file_write updates the times even if
	 * no data is copied in so long as the write had a size.
	 *
	 * We must update xfs' times since revalidate will overcopy xfs.
	 */
	if (!(ioflags & IO_INVIS)) {
		xfs_ichgtime(xip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
		inode_update_time(inode, 1);
	}

	/*
	 * 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 > isize) {
		error = xfs_zero_eof(BHV_TO_VNODE(bdp), io, pos,
					isize, pos + count);
		if (error) {
			xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
			goto out_unlock_isem;
		}
	}
	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_dentry);
		if (unlikely(error)) {
			xfs_iunlock(xip, iolock);
			goto out_unlock_isem;
		}
	}

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

	if ((ioflags & IO_ISDIRECT)) {
		if (need_flush) {
			xfs_inval_cached_trace(io, pos, -1,
					ctooff(offtoct(pos)), -1);
			VOP_FLUSHINVAL_PAGES(vp, ctooff(offtoct(pos)),
					-1, FI_REMAPF_LOCKED);
		}

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

			iolock = XFS_IOLOCK_SHARED;
			locktype = VRWLOCK_WRITE_DIRECT;
			need_isem = 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;

			need_isem = 1;
			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)
		ret = wait_on_sync_kiocb(iocb);

	if ((ret == -ENOSPC) &&
	    DM_EVENT_ENABLED(vp->v_vfsp, xip, DM_EVENT_NOSPACE) &&
	    !(ioflags & IO_INVIS)) {

		xfs_rwunlock(bdp, locktype);
		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 (error)
			goto out_unlock_isem;
		xfs_rwlock(bdp, locktype);
		pos = xip->i_d.di_size;
		goto retry;
	}

	if (*offset > xip->i_d.di_size) {
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		if (*offset > xip->i_d.di_size) {
			xip->i_d.di_size = *offset;
			i_size_write(inode, *offset);
			xip->i_update_core = 1;
			xip->i_update_size = 1;
		}
		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)) {
		/*
		 * If we're treating this as O_DSYNC and we have not updated the
		 * size, force the log.
		 */
		if (!(mp->m_flags & XFS_MOUNT_OSYNCISOSYNC) &&
		    !(xip->i_update_size)) {
			xfs_inode_log_item_t	*iip = xip->i_itemp;

			/*
			 * If an allocation transaction occurred
			 * without extending the size, then we have to force
			 * the log up the proper point to ensure that the
			 * allocation is permanent.  We can't count on
			 * the fact that buffered writes lock out direct I/O
			 * writes - the direct I/O write could have extended
			 * the size nontransactionally, then finished before
			 * we started.  xfs_write_file will think that the file
			 * didn't grow but the update isn't safe unless the
			 * size change is logged.
			 *
			 * Force the log if we've committed a transaction
			 * against the inode or if someone else has and
			 * the commit record hasn't gone to disk (e.g.
			 * the inode is pinned).  This guarantees that
			 * all changes affecting the inode are permanent
			 * when we return.
			 */
			if (iip && iip->ili_last_lsn) {
				xfs_log_force(mp, iip->ili_last_lsn,
						XFS_LOG_FORCE | XFS_LOG_SYNC);
			} else if (xfs_ipincount(xip) > 0) {
				xfs_log_force(mp, (xfs_lsn_t)0,
						XFS_LOG_FORCE | XFS_LOG_SYNC);
			}

		} else {
			xfs_trans_t	*tp;

			/*
			 * O_SYNC or O_DSYNC _with_ a size update are handled
			 * the same way.
			 *
			 * If the write was synchronous then we need to make
			 * sure that the inode modification time is permanent.
			 * We'll have updated the timestamp above, so here
			 * we use a synchronous transaction to log the inode.
			 * It's not fast, but it's necessary.
			 *
			 * If this a dsync write and the size got changed
			 * non-transactionally, then we need to ensure that
			 * the size change gets logged in a synchronous
			 * transaction.
			 */

			tp = xfs_trans_alloc(mp, XFS_TRANS_WRITE_SYNC);
			if ((error = xfs_trans_reserve(tp, 0,
						      XFS_SWRITE_LOG_RES(mp),
						      0, 0, 0))) {
				/* Transaction reserve failed */
				xfs_trans_cancel(tp, 0);
			} else {
				/* Transaction reserve successful */
				xfs_ilock(xip, XFS_ILOCK_EXCL);
				xfs_trans_ijoin(tp, xip, XFS_ILOCK_EXCL);
				xfs_trans_ihold(tp, xip);
				xfs_trans_log_inode(tp, xip, XFS_ILOG_CORE);
				xfs_trans_set_sync(tp);
				error = xfs_trans_commit(tp, 0, NULL);
				xfs_iunlock(xip, XFS_ILOCK_EXCL);
				if (error)
					goto out_unlock_internal;
			}
		}
	
		xfs_rwunlock(bdp, locktype);
		if (need_isem)
			up(&inode->i_sem);

		error = sync_page_range(inode, mapping, pos, ret);
		if (!error)
			error = ret;
		return error;
	}

 out_unlock_internal:
	xfs_rwunlock(bdp, locktype);
 out_unlock_isem:
	if (need_isem)
		up(&inode->i_sem);
	return -error;
}
Ejemplo n.º 5
0
ssize_t				/* bytes written, or (-) error */
xfs_write(
	bhv_desc_t		*bdp,
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned int		nsegs,
	loff_t			*offset,
	int			ioflags,
	cred_t			*credp)
{
	struct file		*file = iocb->ki_filp;
	struct address_space	*mapping = file->f_mapping;
	struct inode		*inode = mapping->host;
	unsigned long		segs = nsegs;
	xfs_inode_t		*xip;
	xfs_mount_t		*mp;
	ssize_t			ret = 0, error = 0;
	xfs_fsize_t		isize, new_size;
	xfs_iocore_t		*io;
	bhv_vnode_t		*vp;
	unsigned long		seg;
	int			iolock;
	int			eventsent = 0;
	bhv_vrwlock_t		locktype;
	size_t			ocount = 0, count;
	loff_t			pos;
	int			need_i_mutex = 1, need_flush = 0;

	XFS_STATS_INC(xs_write_calls);

	vp = BHV_TO_VNODE(bdp);
	xip = XFS_BHVTOI(bdp);

	for (seg = 0; seg < segs; seg++) {
		const struct iovec *iv = &iovp[seg];

		/*
		 * If any segment has a negative length, or the cumulative
		 * length ever wraps negative then return -EINVAL.
		 */
		ocount += iv->iov_len;
		if (unlikely((ssize_t)(ocount|iv->iov_len) < 0))
			return -EINVAL;
		if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len))
			continue;
		if (seg == 0)
			return -EFAULT;
		segs = seg;
		ocount -= iv->iov_len;  /* This segment is no good */
		break;
	}

	count = ocount;
	pos = *offset;

	if (count == 0)
		return 0;

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

	vfs_wait_for_freeze(vp->v_vfsp, SB_FREEZE_WRITE);

	if (XFS_FORCED_SHUTDOWN(mp))
		return -EIO;

	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))
			return XFS_ERROR(-EINVAL);

		if (!VN_CACHED(vp) && pos < i_size_read(inode))
			need_i_mutex = 0;

		if (VN_CACHED(vp))
			need_flush = 1;
	}

relock:
	if (need_i_mutex) {
		iolock = XFS_IOLOCK_EXCL;
		locktype = VRWLOCK_WRITE;

		mutex_lock(&inode->i_mutex);
	} else {
		iolock = XFS_IOLOCK_SHARED;
		locktype = VRWLOCK_WRITE_DIRECT;
	}

	xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);

	isize = i_size_read(inode);

	if (file->f_flags & O_APPEND)
		*offset = isize;

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;
	}

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

	if ((DM_EVENT_ENABLED(vp->v_vfsp, xip, DM_EVENT_WRITE) &&
	    !(ioflags & IO_INVIS) && !eventsent)) {
		loff_t		savedsize = pos;
		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) {
			xfs_iunlock(xip, iolock);
			goto out_unlock_mutex;
		}
		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) && savedsize != isize) {
			pos = isize = xip->i_d.di_size;
			goto start;
		}
	}

	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 > isize) {
		error = xfs_zero_eof(BHV_TO_VNODE(bdp), io, pos, isize);
		if (error) {
			xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
			goto out_unlock_mutex;
		}
	}
	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)) {
			xfs_iunlock(xip, iolock);
			goto out_unlock_mutex;
		}
	}

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

	if ((ioflags & IO_ISDIRECT)) {
		if (need_flush) {
			xfs_inval_cached_trace(io, pos, -1,
					ctooff(offtoct(pos)), -1);
			bhv_vop_flushinval_pages(vp, ctooff(offtoct(pos)),
					-1, FI_REMAPF_LOCKED);
		}

		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;

			need_i_mutex = 1;
			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(vp->v_vfsp, xip, DM_EVENT_NOSPACE) &&
	    !(ioflags & IO_INVIS)) {

		xfs_rwunlock(bdp, 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 (error)
			goto out_nounlocks;
		if (need_i_mutex)
			mutex_lock(&inode->i_mutex);
		xfs_rwlock(bdp, locktype);
		pos = xip->i_d.di_size;
		ret = 0;
		goto retry;
	}

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

	if (*offset > xip->i_d.di_size) {
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		if (*offset > xip->i_d.di_size) {
			xip->i_d.di_size = *offset;
			i_size_write(inode, *offset);
			xip->i_update_core = 1;
			xip->i_update_size = 1;
		}
		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)) {
		error = xfs_write_sync_logforce(mp, xip);
		if (error)
			goto out_unlock_internal;

		xfs_rwunlock(bdp, locktype);
		if (need_i_mutex)
			mutex_unlock(&inode->i_mutex);

		error = sync_page_range(inode, mapping, pos, ret);
		if (!error)
			error = ret;
		return error;
	}

 out_unlock_internal:
	xfs_rwunlock(bdp, locktype);
 out_unlock_mutex:
	if (need_i_mutex)
		mutex_unlock(&inode->i_mutex);
 out_nounlocks:
	return -error;
}
Ejemplo n.º 6
0
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;
	unsigned long		segs = nsegs;
	xfs_mount_t		*mp;
	ssize_t			ret = 0, error = 0;
	xfs_fsize_t		isize, new_size;
	int			iolock;
	int			eventsent = 0;
	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;

	mp = xip->i_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;
		need_i_mutex = 0;
	} else {
		iolock = XFS_IOLOCK_EXCL;
		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, xip,
				      pos, count, dmflags, &iolock);
		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 =
			XFS_IS_REALTIME_INODE(xip) ?
				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 && (mapping->nrpages || pos > xip->i_size)) {
			xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
			iolock = XFS_IOLOCK_EXCL;
			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)
		xip->i_new_size = new_size;

	if (likely(!(ioflags & IO_INVIS)))
		xfs_ichgtime(xip, 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(xip, 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 = -file_remove_suid(file);
		if (unlikely(error)) {
			goto out_unlock_internal;
		}
	}

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

	if ((ioflags & IO_ISDIRECT)) {
		if (mapping->nrpages) {
			WARN_ON(need_i_mutex == 0);
			xfs_inval_cached_trace(xip, pos, -1,
					(pos & PAGE_CACHE_MASK), -1);
			error = xfs_flushinval_pages(xip,
					(pos & PAGE_CACHE_MASK),
					-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(xip, XFS_IOLOCK_EXCL);
			mutex_unlock(&inode->i_mutex);

			iolock = XFS_IOLOCK_SHARED;
			need_i_mutex = 0;
		}

 		xfs_rw_enter_trace(XFS_DIOWR_ENTER, xip, (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, xip, (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);

	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);
	}

	if (ret == -ENOSPC &&
	    DM_EVENT_ENABLED(xip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
		xfs_iunlock(xip, iolock);
		if (need_i_mutex)
			mutex_unlock(&inode->i_mutex);
		error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, xip,
				DM_RIGHT_NULL, xip, DM_RIGHT_NULL, NULL, NULL,
				0, 0, 0); /* Delay flag intentionally  unused */
		if (need_i_mutex)
			mutex_lock(&inode->i_mutex);
		xfs_ilock(xip, iolock);
		if (error)
			goto out_unlock_internal;
		goto start;
	}

	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_iunlock(xip, iolock);
		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_ilock(xip, iolock);
		error2 = xfs_write_sync_logforce(mp, xip);
		if (!error)
			error = error2;
	}

 out_unlock_internal:
	if (xip->i_new_size) {
		xfs_ilock(xip, XFS_ILOCK_EXCL);
		xip->i_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_iunlock(xip, iolock);
 out_unlock_mutex:
	if (need_i_mutex)
		mutex_unlock(&inode->i_mutex);
	return -error;
}