xfs_lrw.c

linux-2.6.15.6

/*
 * Copyright (c) 2000-2003,2005 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_bit.h"
#include "xfs_log.h"
#include "xfs_inum.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_dir.h"
#include "xfs_dir2.h"
#include "xfs_alloc.h"
#include "xfs_dmapi.h"
#include "xfs_quota.h"
#include "xfs_mount.h"
#include "xfs_bmap_btree.h"
#include "xfs_alloc_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_dir_sf.h"
#include "xfs_dir2_sf.h"
#include "xfs_attr_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_bmap.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_rtalloc.h"
#include "xfs_error.h"
#include "xfs_itable.h"
#include "xfs_rw.h"
#include "xfs_acl.h"
#include "xfs_cap.h"
#include "xfs_mac.h"
#include "xfs_attr.h"
#include "xfs_inode_item.h"
#include "xfs_buf_item.h"
#include "xfs_utils.h"
#include "xfs_iomap.h"

#include <linux/capability.h>
#include <linux/writeback.h>


#if defined(XFS_RW_TRACE)
void
xfs_rw_enter_trace(
	int			tag,
	xfs_iocore_t		*io,
	void			*data,
	size_t			segs,
	loff_t			offset,
	int			ioflags)
{
	xfs_inode_t	*ip = XFS_IO_INODE(io);

	if (ip->i_rwtrace == NULL)
		return;
	ktrace_enter(ip->i_rwtrace,
		(void *)(unsigned long)tag,
		(void *)ip,
		(void *)((unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff)),
		(void *)((unsigned long)(ip->i_d.di_size & 0xffffffff)),
		(void *)data,
		(void *)((unsigned long)segs),
		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
		(void *)((unsigned long)(offset & 0xffffffff)),
		(void *)((unsigned long)ioflags),
		(void *)((unsigned long)((io->io_new_size >> 32) & 0xffffffff)),
		(void *)((unsigned long)(io->io_new_size & 0xffffffff)),
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL);
}

void
xfs_inval_cached_trace(
	xfs_iocore_t	*io,
	xfs_off_t	offset,
	xfs_off_t	len,
	xfs_off_t	first,
	xfs_off_t	last)
{
	xfs_inode_t	*ip = XFS_IO_INODE(io);

	if (ip->i_rwtrace == NULL)
		return;
	ktrace_enter(ip->i_rwtrace,
		(void *)(__psint_t)XFS_INVAL_CACHED,
		(void *)ip,
		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
		(void *)((unsigned long)(offset & 0xffffffff)),
		(void *)((unsigned long)((len >> 32) & 0xffffffff)),
		(void *)((unsigned long)(len & 0xffffffff)),
		(void *)((unsigned long)((first >> 32) & 0xffffffff)),
		(void *)((unsigned long)(first & 0xffffffff)),
		(void *)((unsigned long)((last >> 32) & 0xffffffff)),
		(void *)((unsigned long)(last & 0xffffffff)),
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL,
		(void *)NULL);
}
#endif

/*
 *	xfs_iozero
 *
 *	xfs_iozero clears the specified range of buffer supplied,
 *	and marks all the affected blocks as valid and modified.  If
 *	an affected block is not allocated, it will be allocated.  If
 *	an affected block is not completely overwritten, and is not
 *	valid before the operation, it will be read from disk before
 *	being partially zeroed.
 */
STATIC int
xfs_iozero(
	struct inode		*ip,	/* inode			*/
	loff_t			pos,	/* offset in file		*/
	size_t			count,	/* size of data to zero		*/
	loff_t			end_size)	/* max file size to set */
{
	unsigned		bytes;
	struct page		*page;
	struct address_space	*mapping;
	char			*kaddr;
	int			status;

	mapping = ip->i_mapping;
	do {
		unsigned long index, offset;

		offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
		index = pos >> PAGE_CACHE_SHIFT;
		bytes = PAGE_CACHE_SIZE - offset;
		if (bytes > count)
			bytes = count;

		status = -ENOMEM;
		page = grab_cache_page(mapping, index);
		if (!page)
			break;

		kaddr = kmap(page);
		status = mapping->a_ops->prepare_write(NULL, page, offset,
							offset + bytes);
		if (status) {
			goto unlock;
		}

		memset((void *) (kaddr + offset), 0, bytes);
		flush_dcache_page(page);
		status = mapping->a_ops->commit_write(NULL, page, offset,
							offset + bytes);
		if (!status) {
			pos += bytes;
			count -= bytes;
			if (pos > i_size_read(ip))
				i_size_write(ip, pos < end_size ? pos : end_size);
		}

unlock:
		kunmap(page);
		unlock_page(page);
		page_cache_release(page);
		if (status)
			break;
	} while (count);

	return (-status);
}

ssize_t			/* bytes read, or (-)  error */
xfs_read(
	bhv_desc_t		*bdp,
	struct kiocb		*iocb,
	const struct iovec	*iovp,
	unsigned int		segs,
	loff_t			*offset,
	int			ioflags,
	cred_t			*credp)
{
	struct file		*file = iocb->ki_filp;
	struct inode		*inode = file->f_mapping->host;
	size_t			size = 0;
	ssize_t			ret;
	xfs_fsize_t		n;
	xfs_inode_t		*ip;
	xfs_mount_t		*mp;
	vnode_t			*vp;
	unsigned long		seg;

	ip = XFS_BHVTOI(bdp);
	vp = BHV_TO_VNODE(bdp);
	mp = ip->i_mount;

	XFS_STATS_INC(xs_read_calls);

	/* START copy & waste from filemap.c */
	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.
		 */
		size += iv->iov_len;
		if (unlikely((ssize_t)(size|iv->iov_len) < 0))
			return XFS_ERROR(-EINVAL);
	}
	/* END copy & waste from filemap.c */

	if (unlikely(ioflags & IO_ISDIRECT)) {
		xfs_buftarg_t	*target =
			(ip->i_d.di_flags & XFS_DIFLAG_REALTIME) ?
				mp->m_rtdev_targp : mp->m_ddev_targp;
		if ((*offset & target->pbr_smask) ||
		    (size & target->pbr_smask)) {
			if (*offset == ip->i_d.di_size) {
				return (0);
			}
			return -XFS_ERROR(EINVAL);
		}
	}

	n = XFS_MAXIOFFSET(mp) - *offset;
	if ((n <= 0) || (size == 0))
		return 0;

	if (n < size)
		size = n;

	if (XFS_FORCED_SHUTDOWN(mp)) {
		return -EIO;
	}

	if (unlikely(ioflags & IO_ISDIRECT))
		down(&inode->i_sem);
	xfs_ilock(ip, XFS_IOLOCK_SHARED);

	if (DM_EVENT_ENABLED(vp->v_vfsp, ip, DM_EVENT_READ) &&
	    !(ioflags & IO_INVIS)) {
		vrwlock_t locktype = VRWLOCK_READ;
		int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags);

		ret = -XFS_SEND_DATA(mp, DM_EVENT_READ,
					BHV_TO_VNODE(bdp), *offset, size,
					dmflags, &locktype);
		if (ret) {
			xfs_iunlock(ip, XFS_IOLOCK_SHARED);
			goto unlock_isem;
		}
	}

	xfs_rw_enter_trace(XFS_READ_ENTER, &ip->i_iocore,
				(void *)iovp, segs, *offset, ioflags);
	ret = __generic_file_aio_read(iocb, iovp, segs, offset);
	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
		ret = wait_on_sync_kiocb(iocb);
	if (ret > 0)
		XFS_STATS_ADD(xs_read_bytes, ret);

	xfs_iunlock(ip, XFS_IOLOCK_SHARED);

	if (likely(!(ioflags & IO_INVIS)))
		xfs_ichgtime_fast(ip, inode, XFS_ICHGTIME_ACC);

unlock_isem:
	if (unlikely(ioflags & IO_ISDIRECT))
		up(&inode->i_sem);
	return ret;
}

ssize_t
xfs_sendfile(
	bhv_desc_t		*bdp,
	struct file		*filp,
	loff_t			*offset,
	int			ioflags,
	size_t			count,
	read_actor_t		actor,
	void			*target,
	cred_t			*credp)
{
	ssize_t			ret;
	xfs_fsize_t		n;
	xfs_inode_t		*ip;
	xfs_mount_t		*mp;
	vnode_t			*vp;

	ip = XFS_BHVTOI(bdp);
	vp = BHV_TO_VNODE(bdp);
	mp = ip->i_mount;

	XFS_STATS_INC(xs_read_calls);

	n = XFS_MAXIOFFSET(mp) - *offset;
	if ((n <= 0) || (count == 0))
		return 0;

	if (n < count)
		count = n;

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

	xfs_ilock(ip, XFS_IOLOCK_SHARED);

	if (DM_EVENT_ENABLED(vp->v_vfsp, ip, DM_EVENT_READ) &&
	    (!(ioflags & IO_INVIS))) {
		vrwlock_t locktype = VRWLOCK_READ;
		int error;

		error = XFS_SEND_DATA(mp, DM_EVENT_READ, BHV_TO_VNODE(bdp), *offset, count,
				      FILP_DELAY_FLAG(filp), &locktype);
		if (error) {
			xfs_iunlock(ip, XFS_IOLOCK_SHARED);
			return -error;
		}
	}
	xfs_rw_enter_trace(XFS_SENDFILE_ENTER, &ip->i_iocore,
		   (void *)(unsigned long)target, count, *offset, ioflags);
	ret = generic_file_sendfile(filp, offset, count, actor, target);

	xfs_iunlock(ip, XFS_IOLOCK_SHARED);

	if (ret > 0)
		XFS_STATS_ADD(xs_read_bytes, ret);

	if (likely(!(ioflags & IO_INVIS)))
		xfs_ichgtime_fast(ip, LINVFS_GET_IP(vp), XFS_ICHGTIME_ACC);

	return ret;
}

/*
 * This routine is called to handle zeroing any space in the last
 * block of the file that is beyond the EOF.  We do this since the
 * size is being increased without writing anything to that block
 * and we don't want anyone to read the garbage on the disk.
 */
STATIC int				/* error (positive) */
xfs_zero_last_block(
	struct inode	*ip,
	xfs_iocore_t	*io,
	xfs_off_t	offset,
	xfs_fsize_t	isize,
	xfs_fsize_t	end_size)
{
	xfs_fileoff_t	last_fsb;
	xfs_mount_t	*mp;
	int		nimaps;
	int		zero_offset;
	int		zero_len;
	int		isize_fsb_offset;
	int		error = 0;
	xfs_bmbt_irec_t	imap;
	loff_t		loff;
	size_t		lsize;

	ASSERT(ismrlocked(io->io_lock, MR_UPDATE) != 0);
	ASSERT(offset > isize);

	mp = io->io_mount;

	isize_fsb_offset = XFS_B_FSB_OFFSET(mp, isize);
	if (isize_fsb_offset == 0) {
		/*
		 * There are no extra bytes in the last block on disk to
		 * zero, so return.
		 */
		return 0;
	}

	last_fsb = XFS_B_TO_FSBT(mp, isize);
	nimaps = 1;
	error = XFS_BMAPI(mp, NULL, io, last_fsb, 1, 0, NULL, 0, &imap,
			  &nimaps, NULL);
	if (error) {
		return error;
	}
	ASSERT(nimaps > 0);
	/*
	 * If the block underlying isize is just a hole, then there
	 * is nothing to zero.
	 */
	if (imap.br_startblock == HOLESTARTBLOCK) {
		return 0;
	}
	/*
	 * Zero the part of the last block beyond the EOF, and write it
	 * out sync.  We need to drop the ilock while we do this so we
	 * don't deadlock when the buffer cache calls back to us.
	 */
	XFS_IUNLOCK(mp, io, XFS_ILOCK_EXCL| XFS_EXTSIZE_RD);
	loff = XFS_FSB_TO_B(mp, last_fsb);
	lsize = XFS_FSB_TO_B(mp, 1);

	zero_offset = isize_fsb_offset;
	zero_len = mp->m_sb.sb_blocksize - isize_fsb_offset;

	error = xfs_iozero(ip, loff + zero_offset, zero_len, end_size);

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

/*
 * Zero any on disk space between the current EOF and the new,
 * larger EOF.  This handles the normal case of zeroing the remainder
 * of the last block in the file and the unusual case of zeroing blocks
 * out beyond the size of the file.  This second case only happens
 * with fixed size extents and when the system crashes before the inode
 * size was updated but after blocks were allocated.  If fill is set,
 * then any holes in the range are filled and zeroed.  If not, the holes
 * are left alone as holes.
 */

int					/* error (positive) */
xfs_zero_eof(
	vnode_t		*vp,
	xfs_iocore_t	*io,
	xfs_off_t	offset,		/* starting I/O offset */
	xfs_fsize_t	isize,		/* current inode size */
	xfs_fsize_t	end_size)	/* terminal inode size */
{
	struct inode	*ip = LINVFS_GET_IP(vp);
	xfs_fileoff_t	start_zero_fsb;
	xfs_fileoff_t	end_zero_fsb;
	xfs_fileoff_t	prev_zero_fsb;
	xfs_fileoff_t	zero_count_fsb;
	xfs_fileoff_t	last_fsb;
	xfs_extlen_t	buf_len_fsb;
	xfs_extlen_t	prev_zero_count;
	xfs_mount_t	*mp;
	int		nimaps;
	int		error = 0;
	xfs_bmbt_irec_t	imap;
	loff_t		loff;
	size_t		lsize;

	ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
	ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));

	mp = io->io_mount;

	/*
	 * 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, end_size);
	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);
	prev_zero_fsb = NULLFILEOFF;
	prev_zero_count = 0;
	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);
		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.