ssize_t v9fs_fid_xattr_get(struct p9_fid *fid, const char *name,
void *buffer, size_t buffer_size)
{
ssize_t retval;
u64 attr_size;
struct p9_fid *attr_fid;
struct kvec kvec = {.iov_base = buffer, .iov_len = buffer_size};
struct iov_iter to;
int err;
iov_iter_kvec(&to, READ | ITER_KVEC, &kvec, 1, buffer_size);
attr_fid = p9_client_xattrwalk(fid, name, &attr_size);
if (IS_ERR(attr_fid)) {
retval = PTR_ERR(attr_fid);
p9_debug(P9_DEBUG_VFS, "p9_client_attrwalk failed %zd\n",
retval);
return retval;
}
if (attr_size > buffer_size) {
if (!buffer_size) /* request to get the attr_size */
retval = attr_size;
else
retval = -ERANGE;
} else {
iov_iter_truncate(&to, attr_size);
retval = p9_client_read(attr_fid, 0, &to, &err);
if (err)
retval = err;
}
p9_client_clunk(attr_fid);
return retval;
}
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;
}
/**
* v9fs_file_write - write to a file
* @filp: file pointer to write
* @data: data buffer to write data from
* @count: size of buffer
* @offset: offset at which to write data
*
*/
static ssize_t
v9fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
ssize_t retval = 0;
loff_t origin = iocb->ki_pos;
size_t count = iov_iter_count(from);
int err = 0;
retval = generic_write_checks(file, &origin, &count, 0);
if (retval)
return retval;
iov_iter_truncate(from, count);
if (!count)
return 0;
retval = p9_client_write(file->private_data, origin, from, &err);
if (retval > 0) {
struct inode *inode = file_inode(file);
loff_t i_size;
unsigned long pg_start, pg_end;
pg_start = origin >> PAGE_CACHE_SHIFT;
pg_end = (origin + retval - 1) >> PAGE_CACHE_SHIFT;
if (inode->i_mapping && inode->i_mapping->nrpages)
invalidate_inode_pages2_range(inode->i_mapping,
pg_start, pg_end);
origin += retval;
i_size = i_size_read(inode);
iocb->ki_pos = origin;
if (origin > i_size) {
inode_add_bytes(inode, origin - i_size);
i_size_write(inode, origin);
}
return retval;
}
return err;
}
static ssize_t
ncp_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file_inode(file);
size_t already_read = 0;
off_t pos = iocb->ki_pos;
size_t bufsize;
int error;
void *freepage;
size_t freelen;
ncp_dbg(1, "enter %pD2\n", file);
if (!iov_iter_count(to))
return 0;
if (pos > inode->i_sb->s_maxbytes)
return 0;
iov_iter_truncate(to, inode->i_sb->s_maxbytes - pos);
error = ncp_make_open(inode, O_RDONLY);
if (error) {
ncp_dbg(1, "open failed, error=%d\n", error);
return error;
}
bufsize = NCP_SERVER(inode)->buffer_size;
error = -EIO;
freelen = ncp_read_bounce_size(bufsize);
freepage = vmalloc(freelen);
if (!freepage)
goto outrel;
error = 0;
/* First read in as much as possible for each bufsize. */
while (iov_iter_count(to)) {
int read_this_time;
size_t to_read = min_t(size_t,
bufsize - (pos % bufsize),
iov_iter_count(to));
error = ncp_read_bounce(NCP_SERVER(inode),
NCP_FINFO(inode)->file_handle,
pos, to_read, to, &read_this_time,
freepage, freelen);
if (error) {
error = -EIO; /* NW errno -> Linux errno */
break;
}
pos += read_this_time;
already_read += read_this_time;
if (read_this_time != to_read)
break;
}
vfree(freepage);
iocb->ki_pos = pos;
file_accessed(file);
ncp_dbg(1, "exit %pD2\n", file);
outrel:
ncp_inode_close(inode);
return already_read ? already_read : error;
}
static int sg_io(struct request_queue *q, struct gendisk *bd_disk,
struct sg_io_hdr *hdr, fmode_t mode)
{
unsigned long start_time;
ssize_t ret = 0;
int writing = 0;
int at_head = 0;
struct request *rq;
char sense[SCSI_SENSE_BUFFERSIZE];
struct bio *bio;
if (hdr->interface_id != 'S')
return -EINVAL;
if (hdr->dxfer_len > (queue_max_hw_sectors(q) << 9))
return -EIO;
if (hdr->dxfer_len)
switch (hdr->dxfer_direction) {
default:
return -EINVAL;
case SG_DXFER_TO_DEV:
writing = 1;
break;
case SG_DXFER_TO_FROM_DEV:
case SG_DXFER_FROM_DEV:
break;
}
if (hdr->flags & SG_FLAG_Q_AT_HEAD)
at_head = 1;
ret = -ENOMEM;
rq = blk_get_request(q, writing ? WRITE : READ, GFP_KERNEL);
if (IS_ERR(rq))
return PTR_ERR(rq);
blk_rq_set_block_pc(rq);
if (hdr->cmd_len > BLK_MAX_CDB) {
rq->cmd = kzalloc(hdr->cmd_len, GFP_KERNEL);
if (!rq->cmd)
goto out_put_request;
}
ret = blk_fill_sghdr_rq(q, rq, hdr, mode);
if (ret < 0)
goto out_free_cdb;
ret = 0;
if (hdr->iovec_count) {
struct iov_iter i;
struct iovec *iov = NULL;
ret = import_iovec(rq_data_dir(rq),
hdr->dxferp, hdr->iovec_count,
0, &iov, &i);
if (ret < 0)
goto out_free_cdb;
/* SG_IO howto says that the shorter of the two wins */
iov_iter_truncate(&i, hdr->dxfer_len);
ret = blk_rq_map_user_iov(q, rq, NULL, &i, GFP_KERNEL);
kfree(iov);
} else if (hdr->dxfer_len)
ret = blk_rq_map_user(q, rq, NULL, hdr->dxferp, hdr->dxfer_len,
GFP_KERNEL);
if (ret)
goto out_free_cdb;
bio = rq->bio;
memset(sense, 0, sizeof(sense));
rq->sense = sense;
rq->sense_len = 0;
rq->retries = 0;
start_time = jiffies;
/* ignore return value. All information is passed back to caller
* (if he doesn't check that is his problem).
* N.B. a non-zero SCSI status is _not_ necessarily an error.
*/
blk_execute_rq(q, bd_disk, rq, at_head);
hdr->duration = jiffies_to_msecs(jiffies - start_time);
ret = blk_complete_sghdr_rq(rq, hdr, bio);
out_free_cdb:
if (rq->cmd != rq->__cmd)
kfree(rq->cmd);
out_put_request:
blk_put_request(rq);
return ret;
}
static ssize_t
ncp_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file_inode(file);
size_t already_written = 0;
loff_t pos = iocb->ki_pos;
size_t count = iov_iter_count(from);
size_t bufsize;
int errno;
void *bouncebuffer;
ncp_dbg(1, "enter %pD2\n", file);
errno = generic_write_checks(file, &pos, &count, 0);
if (errno)
return errno;
iov_iter_truncate(from, count);
if (!count)
return 0;
errno = ncp_make_open(inode, O_WRONLY);
if (errno) {
ncp_dbg(1, "open failed, error=%d\n", errno);
return errno;
}
bufsize = NCP_SERVER(inode)->buffer_size;
errno = file_update_time(file);
if (errno)
goto outrel;
bouncebuffer = vmalloc(bufsize);
if (!bouncebuffer) {
errno = -EIO; /* -ENOMEM */
goto outrel;
}
while (iov_iter_count(from)) {
int written_this_time;
size_t to_write = min_t(size_t,
bufsize - ((off_t)pos % bufsize),
iov_iter_count(from));
if (copy_from_iter(bouncebuffer, to_write, from) != to_write) {
errno = -EFAULT;
break;
}
if (ncp_write_kernel(NCP_SERVER(inode),
NCP_FINFO(inode)->file_handle,
pos, to_write, bouncebuffer, &written_this_time) != 0) {
errno = -EIO;
break;
}
pos += written_this_time;
already_written += written_this_time;
if (written_this_time != to_write)
break;
}
vfree(bouncebuffer);
iocb->ki_pos = pos;
if (pos > i_size_read(inode)) {
mutex_lock(&inode->i_mutex);
if (pos > i_size_read(inode))
i_size_write(inode, pos);
mutex_unlock(&inode->i_mutex);
}
ncp_dbg(1, "exit %pD2\n", file);
outrel:
ncp_inode_close(inode);
return already_written ? already_written : errno;
}
/*
* 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;
}
