/*
* Common pre-write limit and setup checks.
*
* Called with the iolocked held either shared and exclusive according to
* @iolock, and returns with it held. Might upgrade the iolock to exclusive
* if called for a direct write beyond i_size.
*/
STATIC ssize_t
xfs_file_aio_write_checks(
struct file *file,
loff_t *pos,
size_t *count,
int *iolock)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
int error = 0;
xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
restart:
error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
if (error) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
return error;
}
if (likely(!(file->f_mode & FMODE_NOCMTIME)))
file_update_time(file);
/*
* If the offset is beyond the size of the file, we need to zero any
* blocks that fall between the existing EOF and the start of this
* write. If zeroing is needed and we are currently holding the
* iolock shared, we need to update it to exclusive which involves
* dropping all locks and relocking to maintain correct locking order.
* If we do this, restart the function to ensure all checks and values
* are still valid.
*/
if (*pos > i_size_read(inode)) {
if (*iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
*iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
goto restart;
}
error = -xfs_zero_eof(ip, *pos, i_size_read(inode));
}
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
/*
* 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.
*/
return file_remove_suid(file);
}
STATIC ssize_t
xfs_file_buffered_aio_write(
struct kiocb *iocb,
struct iov_iter *from)
{
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;
loff_t pos = iocb->ki_pos;
size_t count = iov_iter_count(from);
xfs_rw_ilock(ip, iolock);
ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
if (ret)
goto out;
iov_iter_truncate(from, count);
/* 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_perform_write(file, from, pos);
if (likely(ret >= 0))
iocb->ki_pos = pos + ret;
/*
* If we hit a space limit, try to free up some lingering preallocated
* space before returning an error. In the case of ENOSPC, first try to
* write back all dirty inodes to free up some of the excess reserved
* metadata space. This reduces the chances that the eofblocks scan
* waits on dirty mappings. Since xfs_flush_inodes() is serialized, this
* also behaves as a filter to prevent too many eofblocks scans from
* running at the same time.
*/
if (ret == -EDQUOT && !enospc) {
enospc = xfs_inode_free_quota_eofblocks(ip);
if (enospc)
goto write_retry;
} else if (ret == -ENOSPC && !enospc) {
struct xfs_eofblocks eofb = {0};
enospc = 1;
xfs_flush_inodes(ip->i_mount);
eofb.eof_scan_owner = ip->i_ino; /* for locking */
eofb.eof_flags = XFS_EOF_FLAGS_SYNC;
xfs_icache_free_eofblocks(ip->i_mount, &eofb);
goto write_retry;
}
current->backing_dev_info = NULL;
out:
xfs_rw_iunlock(ip, iolock);
return ret;
}
STATIC ssize_t
xfs_file_splice_read(
struct file *infilp,
loff_t *ppos,
struct pipe_inode_info *pipe,
size_t count,
unsigned int flags)
{
struct xfs_inode *ip = XFS_I(infilp->f_mapping->host);
int ioflags = 0;
ssize_t ret;
XFS_STATS_INC(xs_read_calls);
if (infilp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
/*
* Common pre-write limit and setup checks.
*
* Returns with iolock held according to @iolock.
*/
STATIC ssize_t
xfs_file_aio_write_checks(
struct file *file,
loff_t *pos,
size_t *count,
int *iolock)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
xfs_fsize_t new_size;
int error = 0;
xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
if (error) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
*iolock = 0;
return error;
}
new_size = *pos + *count;
if (new_size > ip->i_size)
ip->i_new_size = new_size;
if (likely(!(file->f_mode & FMODE_NOCMTIME)))
file_update_time(file);
/*
* If the offset is beyond the size of the file, we need to zero any
* blocks that fall between the existing EOF and the start of this
* write.
*/
if (*pos > ip->i_size)
error = -xfs_zero_eof(ip, *pos, ip->i_size);
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
/*
* 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.
*/
return file_remove_suid(file);
}
/*
* Common pre-write limit and setup checks.
*
* Called with the iolocked held either shared and exclusive according to
* @iolock, and returns with it held. Might upgrade the iolock to exclusive
* if called for a direct write beyond i_size.
*/
STATIC ssize_t
xfs_file_aio_write_checks(
struct file *file,
loff_t *pos,
size_t *count,
int *iolock)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
int error = 0;
restart:
error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
if (error)
return error;
/*
* If the offset is beyond the size of the file, we need to zero any
* blocks that fall between the existing EOF and the start of this
* write. If zeroing is needed and we are currently holding the
* iolock shared, we need to update it to exclusive which implies
* having to redo all checks before.
*/
if (*pos > i_size_read(inode)) {
if (*iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, *iolock);
*iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, *iolock);
goto restart;
}
error = -xfs_zero_eof(ip, *pos, i_size_read(inode));
if (error)
return error;
}
/*
* Updating the timestamps will grab the ilock again from
* xfs_fs_dirty_inode, so we have to call it after dropping the
* lock above. Eventually we should look into a way to avoid
* the pointless lock roundtrip.
*/
if (likely(!(file->f_mode & FMODE_NOCMTIME))) {
error = file_update_time(file);
if (error)
return error;
}
/*
* 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.
*/
return file_remove_suid(file);
}
/*
* 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 occurred. 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.
*/
STATIC void
xfs_aio_write_newsize_update(
struct xfs_inode *ip)
{
if (ip->i_new_size) {
xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
ip->i_new_size = 0;
if (ip->i_d.di_size > ip->i_size)
ip->i_d.di_size = ip->i_size;
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
}
}
STATIC ssize_t
xfs_file_aio_write_checks(
struct file *file,
loff_t *pos,
size_t *count,
int *iolock)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
int error = 0;
xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
restart:
error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
if (error) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
return error;
}
if (*pos > i_size_read(inode)) {
if (*iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
*iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
goto restart;
}
error = -xfs_zero_eof(ip, *pos, i_size_read(inode));
}
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
if (likely(!(file->f_mode & FMODE_NOCMTIME)))
file_update_time(file);
return file_remove_suid(file);
}
STATIC ssize_t
xfs_file_buffered_aio_read(
struct kiocb *iocb,
struct iov_iter *to)
{
struct xfs_inode *ip = XFS_I(file_inode(iocb->ki_filp));
ssize_t ret;
trace_xfs_file_buffered_read(ip, iov_iter_count(to), iocb->ki_pos);
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
ret = generic_file_read_iter(iocb, to);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
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;
}
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;
}
STATIC ssize_t
xfs_file_splice_read(
struct file *infilp,
loff_t *ppos,
struct pipe_inode_info *pipe,
size_t count,
unsigned int flags)
{
struct xfs_inode *ip = XFS_I(infilp->f_mapping->host);
int ioflags = 0;
ssize_t ret;
XFS_STATS_INC(ip->i_mount, xs_read_calls);
if (infilp->f_mode & FMODE_NOCMTIME)
ioflags |= XFS_IO_INVIS;
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
/*
* DAX inodes cannot ues the page cache for splice, so we have to push
* them through the VFS IO path. This means it goes through
* ->read_iter, which for us takes the XFS_IOLOCK_SHARED. Hence we
* cannot lock the splice operation at this level for DAX inodes.
*/
if (IS_DAX(VFS_I(ip))) {
ret = default_file_splice_read(infilp, ppos, pipe, count,
flags);
goto out;
}
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
out:
if (ret > 0)
XFS_STATS_ADD(ip->i_mount, xs_read_bytes, ret);
return ret;
}
static noinline ssize_t
xfs_file_dax_read(
struct kiocb *iocb,
struct iov_iter *to)
{
struct xfs_inode *ip = XFS_I(iocb->ki_filp->f_mapping->host);
size_t count = iov_iter_count(to);
ssize_t ret = 0;
trace_xfs_file_dax_read(ip, count, iocb->ki_pos);
if (!count)
return 0; /* skip atime */
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
ret = iomap_dax_rw(iocb, to, &xfs_iomap_ops);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
file_accessed(iocb->ki_filp);
return ret;
}
STATIC void
xfs_aio_write_isize_update(
struct inode *inode,
loff_t *ppos,
ssize_t bytes_written)
{
struct xfs_inode *ip = XFS_I(inode);
xfs_fsize_t isize = i_size_read(inode);
if (bytes_written > 0)
XFS_STATS_ADD(xs_write_bytes, bytes_written);
if (unlikely(bytes_written < 0 && bytes_written != -EFAULT &&
*ppos > isize))
*ppos = isize;
if (*ppos > ip->i_size) {
xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
if (*ppos > ip->i_size)
ip->i_size = *ppos;
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
}
}
/*
* xfs_file_dio_aio_write - handle direct IO writes
*
* Lock the inode appropriately to prepare for and issue a direct IO write.
* By separating it from the buffered write path we remove all the tricky to
* follow locking changes and looping.
*
* If there are cached pages or we're extending the file, we need IOLOCK_EXCL
* until we're sure the bytes at the new EOF have been zeroed and/or the cached
* pages are flushed out.
*
* In most cases the direct IO writes will be done holding IOLOCK_SHARED
* allowing them to be done in parallel with reads and other direct IO writes.
* However, if the IO is not aligned to filesystem blocks, the direct IO layer
* needs to do sub-block zeroing and that requires serialisation against other
* direct IOs to the same block. In this case we need to serialise the
* submission of the unaligned IOs so that we don't get racing block zeroing in
* the dio layer. To avoid the problem with aio, we also need to wait for
* outstanding IOs to complete so that unwritten extent conversion is completed
* before we try to map the overlapping block. This is currently implemented by
* hitting it with a big hammer (i.e. inode_dio_wait()).
*
* Returns with locks held indicated by @iolock and errors indicated by
* negative return values.
*/
STATIC ssize_t
xfs_file_dio_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);
struct xfs_mount *mp = ip->i_mount;
ssize_t ret = 0;
size_t count = ocount;
int unaligned_io = 0;
int iolock;
struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
if ((pos & target->bt_smask) || (count & target->bt_smask))
return -XFS_ERROR(EINVAL);
if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
unaligned_io = 1;
/*
* We don't need to take an exclusive lock unless there page cache needs
* to be invalidated or unaligned IO is being executed. We don't need to
* consider the EOF extension case here because
* xfs_file_aio_write_checks() will relock the inode as necessary for
* EOF zeroing cases and fill out the new inode size as appropriate.
*/
if (unaligned_io || mapping->nrpages)
iolock = XFS_IOLOCK_EXCL;
else
iolock = XFS_IOLOCK_SHARED;
xfs_rw_ilock(ip, iolock);
/*
* Recheck if there are cached pages that need invalidate after we got
* the iolock to protect against other threads adding new pages while
* we were waiting for the iolock.
*/
if (mapping->nrpages && iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, iolock);
iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, iolock);
}
ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
if (ret)
goto out;
if (mapping->nrpages) {
ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1,
FI_REMAPF_LOCKED);
if (ret)
goto out;
}
/*
* If we are doing unaligned IO, wait for all other IO to drain,
* otherwise demote the lock if we had to flush cached pages
*/
if (unaligned_io)
inode_dio_wait(inode);
else if (iolock == XFS_IOLOCK_EXCL) {
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
iolock = XFS_IOLOCK_SHARED;
}
trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
ret = generic_file_direct_write(iocb, iovp,
&nr_segs, pos, &iocb->ki_pos, count, ocount);
out:
xfs_rw_iunlock(ip, iolock);
/* No fallback to buffered IO on errors for XFS. */
ASSERT(ret < 0 || ret == count);
return ret;
}
/*
* Common pre-write limit and setup checks.
*
* Returns with iolock held according to @iolock.
*/
STATIC ssize_t
xfs_file_aio_write_checks(
struct file *file,
loff_t *pos,
size_t *count,
xfs_fsize_t *new_sizep,
int *iolock)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
xfs_fsize_t new_size;
int error = 0;
xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
*new_sizep = 0;
restart:
error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
if (error) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
*iolock = 0;
return error;
}
if (likely(!(file->f_mode & FMODE_NOCMTIME)))
file_update_time(file);
/*
* If the offset is beyond the size of the file, we need to zero any
* blocks that fall between the existing EOF and the start of this
* write. There is no need to issue zeroing if another in-flght IO ends
* at or before this one If zeronig is needed and we are currently
* holding the iolock shared, we need to update it to exclusive which
* involves dropping all locks and relocking to maintain correct locking
* order. If we do this, restart the function to ensure all checks and
* values are still valid.
*/
if ((ip->i_new_size && *pos > ip->i_new_size) ||
(!ip->i_new_size && *pos > ip->i_size)) {
if (*iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
*iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
goto restart;
}
error = -xfs_zero_eof(ip, *pos, ip->i_size);
}
/*
* If this IO extends beyond EOF, we may need to update ip->i_new_size.
* We have already zeroed space beyond EOF (if necessary). Only update
* ip->i_new_size if this IO ends beyond any other in-flight writes.
*/
new_size = *pos + *count;
if (new_size > ip->i_size) {
if (new_size > ip->i_new_size)
ip->i_new_size = new_size;
*new_sizep = new_size;
}
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
/*
* 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.
*/
return file_remove_suid(file);
}
/*
* xfs_file_dio_aio_write - handle direct IO writes
*
* Lock the inode appropriately to prepare for and issue a direct IO write.
* By separating it from the buffered write path we remove all the tricky to
* follow locking changes and looping.
*
* If there are cached pages or we're extending the file, we need IOLOCK_EXCL
* until we're sure the bytes at the new EOF have been zeroed and/or the cached
* pages are flushed out.
*
* In most cases the direct IO writes will be done holding IOLOCK_SHARED
* allowing them to be done in parallel with reads and other direct IO writes.
* However, if the IO is not aligned to filesystem blocks, the direct IO layer
* needs to do sub-block zeroing and that requires serialisation against other
* direct IOs to the same block. In this case we need to serialise the
* submission of the unaligned IOs so that we don't get racing block zeroing in
* the dio layer. To avoid the problem with aio, we also need to wait for
* outstanding IOs to complete so that unwritten extent conversion is completed
* before we try to map the overlapping block. This is currently implemented by
* hitting it with a big hammer (i.e. inode_dio_wait()).
*
* Returns with locks held indicated by @iolock and errors indicated by
* negative return values.
*/
STATIC ssize_t
xfs_file_dio_aio_write(
struct kiocb *iocb,
struct iov_iter *from)
{
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;
int unaligned_io = 0;
int iolock;
size_t count = iov_iter_count(from);
loff_t pos = iocb->ki_pos;
struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
/* DIO must be aligned to device logical sector size */
if ((pos | count) & target->bt_logical_sectormask)
return -EINVAL;
/* "unaligned" here means not aligned to a filesystem block */
if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
unaligned_io = 1;
/*
* We don't need to take an exclusive lock unless there page cache needs
* to be invalidated or unaligned IO is being executed. We don't need to
* consider the EOF extension case here because
* xfs_file_aio_write_checks() will relock the inode as necessary for
* EOF zeroing cases and fill out the new inode size as appropriate.
*/
if (unaligned_io || mapping->nrpages)
iolock = XFS_IOLOCK_EXCL;
else
iolock = XFS_IOLOCK_SHARED;
xfs_rw_ilock(ip, iolock);
/*
* Recheck if there are cached pages that need invalidate after we got
* the iolock to protect against other threads adding new pages while
* we were waiting for the iolock.
*/
if (mapping->nrpages && iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, iolock);
iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, iolock);
}
ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
if (ret)
goto out;
iov_iter_truncate(from, count);
if (mapping->nrpages) {
ret = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
pos, -1);
if (ret)
goto out;
truncate_pagecache_range(VFS_I(ip), pos, -1);
}
/*
* If we are doing unaligned IO, wait for all other IO to drain,
* otherwise demote the lock if we had to flush cached pages
*/
if (unaligned_io)
inode_dio_wait(inode);
else if (iolock == XFS_IOLOCK_EXCL) {
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
iolock = XFS_IOLOCK_SHARED;
}
trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
ret = generic_file_direct_write(iocb, from, pos);
out:
xfs_rw_iunlock(ip, iolock);
/* No fallback to buffered IO on errors for XFS. */
ASSERT(ret < 0 || ret == count);
return ret;
}
STATIC ssize_t
xfs_file_aio_read(
struct kiocb *iocb,
const struct iovec *iovp,
unsigned long nr_segs,
loff_t pos)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
size_t size = 0;
ssize_t ret = 0;
int ioflags = 0;
xfs_fsize_t n;
XFS_STATS_INC(xs_read_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;
ret = generic_segment_checks(iovp, &nr_segs, &size, VERIFY_WRITE);
if (ret < 0)
return ret;
if (unlikely(ioflags & IO_ISDIRECT)) {
xfs_buftarg_t *target =
XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
if ((iocb->ki_pos & target->bt_smask) ||
(size & target->bt_smask)) {
if (iocb->ki_pos == i_size_read(inode))
return 0;
return -XFS_ERROR(EINVAL);
}
}
n = mp->m_super->s_maxbytes - iocb->ki_pos;
if (n <= 0 || size == 0)
return 0;
if (n < size)
size = n;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
/*
* Locking is a bit tricky here. If we take an exclusive lock
* for direct IO, we effectively serialise all new concurrent
* read IO to this file and block it behind IO that is currently in
* progress because IO in progress holds the IO lock shared. We only
* need to hold the lock exclusive to blow away the page cache, so
* only take lock exclusively if the page cache needs invalidation.
* This allows the normal direct IO case of no page cache pages to
* proceeed concurrently without serialisation.
*/
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
if ((ioflags & IO_ISDIRECT) && inode->i_mapping->nrpages) {
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
if (inode->i_mapping->nrpages) {
ret = -xfs_flushinval_pages(ip,
(iocb->ki_pos & PAGE_CACHE_MASK),
-1, FI_REMAPF_LOCKED);
if (ret) {
xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
}
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
}
trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags);
ret = generic_file_aio_read(iocb, iovp, nr_segs, iocb->ki_pos);
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
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);
ssize_t ret;
int iolock;
size_t ocount = 0;
XFS_STATS_INC(xs_write_calls);
BUG_ON(iocb->ki_pos != pos);
ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
if (ret)
return ret;
if (ocount == 0)
return 0;
xfs_wait_for_freeze(ip->i_mount, SB_FREEZE_WRITE);
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
if (unlikely(file->f_flags & O_DIRECT))
ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos,
ocount, &iolock);
else
ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos,
ocount, &iolock);
xfs_aio_write_isize_update(inode, &iocb->ki_pos, ret);
if (ret <= 0)
goto out_unlock;
/* Handle various SYNC-type writes */
if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
loff_t end = pos + ret - 1;
int error, error2;
xfs_rw_iunlock(ip, iolock);
error = filemap_write_and_wait_range(mapping, pos, end);
xfs_rw_ilock(ip, iolock);
error2 = -xfs_file_fsync(file,
(file->f_flags & __O_SYNC) ? 0 : 1);
if (error)
ret = error;
else if (error2)
ret = error2;
}
out_unlock:
xfs_aio_write_newsize_update(ip);
xfs_rw_iunlock(ip, iolock);
return ret;
}
STATIC ssize_t
xfs_file_read_iter(
struct kiocb *iocb,
struct iov_iter *to)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
size_t size = iov_iter_count(to);
ssize_t ret = 0;
int ioflags = 0;
xfs_fsize_t n;
loff_t pos = iocb->ki_pos;
XFS_STATS_INC(mp, xs_read_calls);
if (unlikely(iocb->ki_flags & IOCB_DIRECT))
ioflags |= XFS_IO_ISDIRECT;
if (file->f_mode & FMODE_NOCMTIME)
ioflags |= XFS_IO_INVIS;
if ((ioflags & XFS_IO_ISDIRECT) && !IS_DAX(inode)) {
xfs_buftarg_t *target =
XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
/* DIO must be aligned to device logical sector size */
if ((pos | size) & target->bt_logical_sectormask) {
if (pos == i_size_read(inode))
return 0;
return -EINVAL;
}
}
n = mp->m_super->s_maxbytes - pos;
if (n <= 0 || size == 0)
return 0;
if (n < size)
size = n;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
/*
* Locking is a bit tricky here. If we take an exclusive lock for direct
* IO, we effectively serialise all new concurrent read IO to this file
* and block it behind IO that is currently in progress because IO in
* progress holds the IO lock shared. We only need to hold the lock
* exclusive to blow away the page cache, so only take lock exclusively
* if the page cache needs invalidation. This allows the normal direct
* IO case of no page cache pages to proceeed concurrently without
* serialisation.
*/
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
if ((ioflags & XFS_IO_ISDIRECT) && inode->i_mapping->nrpages) {
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
/*
* The generic dio code only flushes the range of the particular
* I/O. Because we take an exclusive lock here, this whole
* sequence is considerably more expensive for us. This has a
* noticeable performance impact for any file with cached pages,
* even when outside of the range of the particular I/O.
*
* Hence, amortize the cost of the lock against a full file
* flush and reduce the chances of repeated iolock cycles going
* forward.
*/
if (inode->i_mapping->nrpages) {
ret = filemap_write_and_wait(VFS_I(ip)->i_mapping);
if (ret) {
xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
/*
* Invalidate whole pages. This can return an error if
* we fail to invalidate a page, but this should never
* happen on XFS. Warn if it does fail.
*/
ret = invalidate_inode_pages2(VFS_I(ip)->i_mapping);
WARN_ON_ONCE(ret);
ret = 0;
}
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
}
trace_xfs_file_read(ip, size, pos, ioflags);
ret = generic_file_read_iter(iocb, to);
if (ret > 0)
XFS_STATS_ADD(mp, xs_read_bytes, ret);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
STATIC ssize_t
xfs_file_aio_read(
struct kiocb *iocb,
const struct iovec *iovp,
unsigned long nr_segs,
loff_t pos)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
size_t size = 0;
ssize_t ret = 0;
int ioflags = 0;
xfs_fsize_t n;
unsigned long seg;
XFS_STATS_INC(xs_read_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;
/* START copy & waste from filemap.c */
for (seg = 0; seg < nr_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 =
XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
if ((iocb->ki_pos & target->bt_smask) ||
(size & target->bt_smask)) {
if (iocb->ki_pos == ip->i_size)
return 0;
return -XFS_ERROR(EINVAL);
}
}
n = XFS_MAXIOFFSET(mp) - iocb->ki_pos;
if (n <= 0 || size == 0)
return 0;
if (n < size)
size = n;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
if (unlikely(ioflags & IO_ISDIRECT)) {
xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
if (inode->i_mapping->nrpages) {
ret = -xfs_flushinval_pages(ip,
(iocb->ki_pos & PAGE_CACHE_MASK),
-1, FI_REMAPF_LOCKED);
if (ret) {
xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
}
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
} else
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags);
ret = generic_file_aio_read(iocb, iovp, nr_segs, iocb->ki_pos);
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
/*
* xfs_file_dio_aio_write - handle direct IO writes
*
* Lock the inode appropriately to prepare for and issue a direct IO write.
* By separating it from the buffered write path we remove all the tricky to
* follow locking changes and looping.
*
* If there are cached pages or we're extending the file, we need IOLOCK_EXCL
* until we're sure the bytes at the new EOF have been zeroed and/or the cached
* pages are flushed out.
*
* In most cases the direct IO writes will be done holding IOLOCK_SHARED
* allowing them to be done in parallel with reads and other direct IO writes.
* However, if the IO is not aligned to filesystem blocks, the direct IO layer
* needs to do sub-block zeroing and that requires serialisation against other
* direct IOs to the same block. In this case we need to serialise the
* submission of the unaligned IOs so that we don't get racing block zeroing in
* the dio layer. To avoid the problem with aio, we also need to wait for
* outstanding IOs to complete so that unwritten extent conversion is completed
* before we try to map the overlapping block. This is currently implemented by
* hitting it with a big hammer (i.e. xfs_ioend_wait()).
*
* Returns with locks held indicated by @iolock and errors indicated by
* negative return values.
*/
STATIC ssize_t
xfs_file_dio_aio_write(
struct kiocb *iocb,
const struct iovec *iovp,
unsigned long nr_segs,
loff_t pos,
size_t ocount,
int *iolock)
{
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;
size_t count = ocount;
int unaligned_io = 0;
struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
*iolock = 0;
if ((pos & target->bt_smask) || (count & target->bt_smask))
return -XFS_ERROR(EINVAL);
if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
unaligned_io = 1;
if (unaligned_io || mapping->nrpages || pos > ip->i_size)
*iolock = XFS_IOLOCK_EXCL;
else
*iolock = XFS_IOLOCK_SHARED;
xfs_rw_ilock(ip, *iolock);
ret = xfs_file_aio_write_checks(file, &pos, &count, iolock);
if (ret)
return ret;
/*
* Recheck if there are cached pages that need invalidate after we got
* the iolock to protect against other threads adding new pages while
* we were waiting for the iolock.
*/
if (mapping->nrpages && *iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, *iolock);
*iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, *iolock);
}
if (mapping->nrpages) {
ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1,
FI_REMAPF_LOCKED);
if (ret)
return ret;
}
/*
* If we are doing unaligned IO, wait for all other IO to drain,
* otherwise demote the lock if we had to flush cached pages
*/
if (unaligned_io)
xfs_ioend_wait(ip);
else if (*iolock == XFS_IOLOCK_EXCL) {
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
*iolock = XFS_IOLOCK_SHARED;
}
trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
ret = generic_file_direct_write(iocb, iovp,
&nr_segs, pos, &iocb->ki_pos, count, ocount);
/* No fallback to buffered IO on errors for XFS. */
ASSERT(ret < 0 || ret == count);
return ret;
}
/*
* xfs_file_dio_aio_write - handle direct IO writes
*
* Lock the inode appropriately to prepare for and issue a direct IO write.
* By separating it from the buffered write path we remove all the tricky to
* follow locking changes and looping.
*
* If there are cached pages or we're extending the file, we need IOLOCK_EXCL
* until we're sure the bytes at the new EOF have been zeroed and/or the cached
* pages are flushed out.
*
* In most cases the direct IO writes will be done holding IOLOCK_SHARED
* allowing them to be done in parallel with reads and other direct IO writes.
* However, if the IO is not aligned to filesystem blocks, the direct IO layer
* needs to do sub-block zeroing and that requires serialisation against other
* direct IOs to the same block. In this case we need to serialise the
* submission of the unaligned IOs so that we don't get racing block zeroing in
* the dio layer. To avoid the problem with aio, we also need to wait for
* outstanding IOs to complete so that unwritten extent conversion is completed
* before we try to map the overlapping block. This is currently implemented by
* hitting it with a big hammer (i.e. inode_dio_wait()).
*
* Returns with locks held indicated by @iolock and errors indicated by
* negative return values.
*/
STATIC ssize_t
xfs_file_dio_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);
struct xfs_mount *mp = ip->i_mount;
ssize_t ret = 0;
size_t count = ocount;
int unaligned_io = 0;
int iolock;
struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
if ((pos & target->bt_smask) || (count & target->bt_smask))
return -XFS_ERROR(EINVAL);
if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
unaligned_io = 1;
if (unaligned_io || mapping->nrpages)
iolock = XFS_IOLOCK_EXCL;
else
iolock = XFS_IOLOCK_SHARED;
xfs_rw_ilock(ip, iolock);
if (mapping->nrpages && iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, iolock);
iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, iolock);
}
ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
if (ret)
goto out;
if (mapping->nrpages) {
ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1,
FI_REMAPF_LOCKED);
if (ret)
goto out;
}
if (unaligned_io)
inode_dio_wait(inode);
else if (iolock == XFS_IOLOCK_EXCL) {
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
iolock = XFS_IOLOCK_SHARED;
}
trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
ret = generic_file_direct_write(iocb, iovp,
&nr_segs, pos, &iocb->ki_pos, count, ocount);
out:
xfs_rw_iunlock(ip, iolock);
ASSERT(ret < 0 || ret == count);
return ret;
}
/*
* Common pre-write limit and setup checks.
*
* Called with the iolocked held either shared and exclusive according to
* @iolock, and returns with it held. Might upgrade the iolock to exclusive
* if called for a direct write beyond i_size.
*/
STATIC ssize_t
xfs_file_aio_write_checks(
struct kiocb *iocb,
struct iov_iter *from,
int *iolock)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
ssize_t error = 0;
size_t count = iov_iter_count(from);
bool drained_dio = false;
restart:
error = generic_write_checks(iocb, from);
if (error <= 0)
return error;
error = xfs_break_layouts(inode, iolock, true);
if (error)
return error;
/* For changing security info in file_remove_privs() we need i_mutex */
if (*iolock == XFS_IOLOCK_SHARED && !IS_NOSEC(inode)) {
xfs_rw_iunlock(ip, *iolock);
*iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, *iolock);
goto restart;
}
/*
* If the offset is beyond the size of the file, we need to zero any
* blocks that fall between the existing EOF and the start of this
* write. If zeroing is needed and we are currently holding the
* iolock shared, we need to update it to exclusive which implies
* having to redo all checks before.
*
* We need to serialise against EOF updates that occur in IO
* completions here. We want to make sure that nobody is changing the
* size while we do this check until we have placed an IO barrier (i.e.
* hold the XFS_IOLOCK_EXCL) that prevents new IO from being dispatched.
* The spinlock effectively forms a memory barrier once we have the
* XFS_IOLOCK_EXCL so we are guaranteed to see the latest EOF value
* and hence be able to correctly determine if we need to run zeroing.
*/
spin_lock(&ip->i_flags_lock);
if (iocb->ki_pos > i_size_read(inode)) {
bool zero = false;
spin_unlock(&ip->i_flags_lock);
if (!drained_dio) {
if (*iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, *iolock);
*iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, *iolock);
iov_iter_reexpand(from, count);
}
/*
* We now have an IO submission barrier in place, but
* AIO can do EOF updates during IO completion and hence
* we now need to wait for all of them to drain. Non-AIO
* DIO will have drained before we are given the
* XFS_IOLOCK_EXCL, and so for most cases this wait is a
* no-op.
*/
inode_dio_wait(inode);
drained_dio = true;
goto restart;
}
error = xfs_zero_eof(ip, iocb->ki_pos, i_size_read(inode), &zero);
if (error)
return error;
} else
spin_unlock(&ip->i_flags_lock);
/*
* Updating the timestamps will grab the ilock again from
* xfs_fs_dirty_inode, so we have to call it after dropping the
* lock above. Eventually we should look into a way to avoid
* the pointless lock roundtrip.
*/
if (likely(!(file->f_mode & FMODE_NOCMTIME))) {
error = file_update_time(file);
if (error)
return error;
}
/*
* 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 (!IS_NOSEC(inode))
return file_remove_privs(file);
return 0;
}
/*
* xfs_file_dio_aio_write - handle direct IO writes
*
* Lock the inode appropriately to prepare for and issue a direct IO write.
* By separating it from the buffered write path we remove all the tricky to
* follow locking changes and looping.
*
* If there are cached pages or we're extending the file, we need IOLOCK_EXCL
* until we're sure the bytes at the new EOF have been zeroed and/or the cached
* pages are flushed out.
*
* In most cases the direct IO writes will be done holding IOLOCK_SHARED
* allowing them to be done in parallel with reads and other direct IO writes.
* However, if the IO is not aligned to filesystem blocks, the direct IO layer
* needs to do sub-block zeroing and that requires serialisation against other
* direct IOs to the same block. In this case we need to serialise the
* submission of the unaligned IOs so that we don't get racing block zeroing in
* the dio layer. To avoid the problem with aio, we also need to wait for
* outstanding IOs to complete so that unwritten extent conversion is completed
* before we try to map the overlapping block. This is currently implemented by
* hitting it with a big hammer (i.e. inode_dio_wait()).
*
* Returns with locks held indicated by @iolock and errors indicated by
* negative return values.
*/
STATIC ssize_t
xfs_file_dio_aio_write(
struct kiocb *iocb,
struct iov_iter *from)
{
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;
int unaligned_io = 0;
int iolock;
size_t count = iov_iter_count(from);
loff_t end;
struct iov_iter data;
struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
/* DIO must be aligned to device logical sector size */
if (!IS_DAX(inode) &&
((iocb->ki_pos | count) & target->bt_logical_sectormask))
return -EINVAL;
/* "unaligned" here means not aligned to a filesystem block */
if ((iocb->ki_pos & mp->m_blockmask) ||
((iocb->ki_pos + count) & mp->m_blockmask))
unaligned_io = 1;
/*
* We don't need to take an exclusive lock unless there page cache needs
* to be invalidated or unaligned IO is being executed. We don't need to
* consider the EOF extension case here because
* xfs_file_aio_write_checks() will relock the inode as necessary for
* EOF zeroing cases and fill out the new inode size as appropriate.
*/
if (unaligned_io || mapping->nrpages)
iolock = XFS_IOLOCK_EXCL;
else
iolock = XFS_IOLOCK_SHARED;
xfs_rw_ilock(ip, iolock);
/*
* Recheck if there are cached pages that need invalidate after we got
* the iolock to protect against other threads adding new pages while
* we were waiting for the iolock.
*/
if (mapping->nrpages && iolock == XFS_IOLOCK_SHARED) {
xfs_rw_iunlock(ip, iolock);
iolock = XFS_IOLOCK_EXCL;
xfs_rw_ilock(ip, iolock);
}
ret = xfs_file_aio_write_checks(iocb, from, &iolock);
if (ret)
goto out;
count = iov_iter_count(from);
end = iocb->ki_pos + count - 1;
/*
* See xfs_file_read_iter() for why we do a full-file flush here.
*/
if (mapping->nrpages) {
ret = filemap_write_and_wait(VFS_I(ip)->i_mapping);
if (ret)
goto out;
/*
* Invalidate whole pages. This can return an error if we fail
* to invalidate a page, but this should never happen on XFS.
* Warn if it does fail.
*/
ret = invalidate_inode_pages2(VFS_I(ip)->i_mapping);
WARN_ON_ONCE(ret);
ret = 0;
}
/*
* If we are doing unaligned IO, wait for all other IO to drain,
* otherwise demote the lock if we had to flush cached pages
*/
if (unaligned_io)
inode_dio_wait(inode);
else if (iolock == XFS_IOLOCK_EXCL) {
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
iolock = XFS_IOLOCK_SHARED;
}
trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
data = *from;
ret = mapping->a_ops->direct_IO(iocb, &data);
/* see generic_file_direct_write() for why this is necessary */
if (mapping->nrpages) {
invalidate_inode_pages2_range(mapping,
iocb->ki_pos >> PAGE_SHIFT,
end >> PAGE_SHIFT);
}
STATIC ssize_t
xfs_file_read_iter(
struct kiocb *iocb,
struct iov_iter *to)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
size_t size = iov_iter_count(to);
ssize_t ret = 0;
int ioflags = 0;
xfs_fsize_t n;
loff_t pos = iocb->ki_pos;
XFS_STATS_INC(xs_read_calls);
if (unlikely(file->f_flags & O_DIRECT))
ioflags |= XFS_IO_ISDIRECT;
if (file->f_mode & FMODE_NOCMTIME)
ioflags |= XFS_IO_INVIS;
if (unlikely(ioflags & XFS_IO_ISDIRECT)) {
xfs_buftarg_t *target =
XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
/* DIO must be aligned to device logical sector size */
if ((pos | size) & target->bt_logical_sectormask) {
if (pos == i_size_read(inode))
return 0;
return -EINVAL;
}
}
n = mp->m_super->s_maxbytes - pos;
if (n <= 0 || size == 0)
return 0;
if (n < size)
size = n;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
/*
* Locking is a bit tricky here. If we take an exclusive lock
* for direct IO, we effectively serialise all new concurrent
* read IO to this file and block it behind IO that is currently in
* progress because IO in progress holds the IO lock shared. We only
* need to hold the lock exclusive to blow away the page cache, so
* only take lock exclusively if the page cache needs invalidation.
* This allows the normal direct IO case of no page cache pages to
* proceeed concurrently without serialisation.
*/
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
if ((ioflags & XFS_IO_ISDIRECT) && inode->i_mapping->nrpages) {
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
if (inode->i_mapping->nrpages) {
ret = filemap_write_and_wait_range(
VFS_I(ip)->i_mapping,
pos, pos + size - 1);
if (ret) {
xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
/*
* Invalidate whole pages. This can return an error if
* we fail to invalidate a page, but this should never
* happen on XFS. Warn if it does fail.
*/
ret = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping,
pos >> PAGE_CACHE_SHIFT,
(pos + size - 1) >> PAGE_CACHE_SHIFT);
WARN_ON_ONCE(ret);
ret = 0;
}
STATIC ssize_t
xfs_file_dio_aio_read(
struct kiocb *iocb,
struct iov_iter *to)
{
struct address_space *mapping = iocb->ki_filp->f_mapping;
struct inode *inode = mapping->host;
struct xfs_inode *ip = XFS_I(inode);
loff_t isize = i_size_read(inode);
size_t count = iov_iter_count(to);
struct iov_iter data;
struct xfs_buftarg *target;
ssize_t ret = 0;
trace_xfs_file_direct_read(ip, count, iocb->ki_pos);
if (!count)
return 0; /* skip atime */
if (XFS_IS_REALTIME_INODE(ip))
target = ip->i_mount->m_rtdev_targp;
else
target = ip->i_mount->m_ddev_targp;
/* DIO must be aligned to device logical sector size */
if ((iocb->ki_pos | count) & target->bt_logical_sectormask) {
if (iocb->ki_pos == isize)
return 0;
return -EINVAL;
}
file_accessed(iocb->ki_filp);
/*
* Locking is a bit tricky here. If we take an exclusive lock for direct
* IO, we effectively serialise all new concurrent read IO to this file
* and block it behind IO that is currently in progress because IO in
* progress holds the IO lock shared. We only need to hold the lock
* exclusive to blow away the page cache, so only take lock exclusively
* if the page cache needs invalidation. This allows the normal direct
* IO case of no page cache pages to proceeed concurrently without
* serialisation.
*/
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
if (mapping->nrpages) {
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
/*
* The generic dio code only flushes the range of the particular
* I/O. Because we take an exclusive lock here, this whole
* sequence is considerably more expensive for us. This has a
* noticeable performance impact for any file with cached pages,
* even when outside of the range of the particular I/O.
*
* Hence, amortize the cost of the lock against a full file
* flush and reduce the chances of repeated iolock cycles going
* forward.
*/
if (mapping->nrpages) {
ret = filemap_write_and_wait(mapping);
if (ret) {
xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
/*
* Invalidate whole pages. This can return an error if
* we fail to invalidate a page, but this should never
* happen on XFS. Warn if it does fail.
*/
ret = invalidate_inode_pages2(mapping);
WARN_ON_ONCE(ret);
ret = 0;
}
xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
}
data = *to;
ret = __blockdev_direct_IO(iocb, inode, target->bt_bdev, &data,
xfs_get_blocks_direct, NULL, NULL, 0);
if (ret >= 0) {
iocb->ki_pos += ret;
iov_iter_advance(to, ret);
}
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
