static int
nfs_file_release(struct inode *inode, struct file *filp)
{
struct dentry *dentry = filp->f_path.dentry;
dprintk("NFS: release(%s/%s)\n",
dentry->d_parent->d_name.name,
dentry->d_name.name);
nfs_inc_stats(inode, NFSIOS_VFSRELEASE);
return nfs_release(inode, filp);
}
static int
nfs3_async_handle_jukebox(struct rpc_task *task, struct inode *inode)
{
if (task->tk_status != -EJUKEBOX)
return 0;
if (task->tk_status == -EJUKEBOX)
nfs_inc_stats(inode, NFSIOS_DELAY);
task->tk_status = 0;
rpc_restart_call(task);
rpc_delay(task, NFS_JUKEBOX_RETRY_TIME);
return 1;
}
/*
* Flush any dirty pages for this process, and check for write errors.
* The return status from this call provides a reliable indication of
* whether any write errors occurred for this process.
*/
static int
nfs_file_fsync(struct file *file, struct dentry *dentry, int datasync)
{
struct nfs_open_context *ctx = nfs_file_open_context(file);
struct inode *inode = dentry->d_inode;
dprintk("NFS: fsync file(%s/%s) datasync %d\n",
dentry->d_parent->d_name.name, dentry->d_name.name,
datasync);
nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
return nfs_do_fsync(ctx, inode);
}
/*
* Flush all dirty pages, and check for write errors.
*/
static int
nfs_file_flush(struct file *file, fl_owner_t id)
{
struct inode *inode = file_inode(file);
dprintk("NFS: flush(%pD2)\n", file);
nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
if ((file->f_mode & FMODE_WRITE) == 0)
return 0;
/* Flush writes to the server and return any errors */
return vfs_fsync(file, 0);
}
/*
* Open file
*/
static int
nfs_file_open(struct inode *inode, struct file *filp)
{
int res;
dprintk("NFS: open file(%pD2)\n", filp);
nfs_inc_stats(inode, NFSIOS_VFSOPEN);
res = nfs_check_flags(filp->f_flags);
if (res)
return res;
res = nfs_open(inode, filp);
return res;
}
static int
nfs_file_release(struct inode *inode, struct file *filp)
{
struct dentry *dentry = filp->f_path.dentry;
dprintk("NFS: release(%s/%s)\n",
dentry->d_parent->d_name.name,
dentry->d_name.name);
/* Ensure that dirty pages are flushed out with the right creds */
if (filp->f_mode & FMODE_WRITE)
nfs_wb_all(dentry->d_inode);
nfs_inc_stats(inode, NFSIOS_VFSRELEASE);
return nfs_release(inode, filp);
}
/*
* Open file
*/
static int
nfs_file_open(struct inode *inode, struct file *filp)
{
int res;
res = nfs_check_flags(filp->f_flags);
if (res)
return res;
nfs_inc_stats(inode, NFSIOS_VFSOPEN);
lock_kernel();
res = NFS_PROTO(inode)->file_open(inode, filp);
unlock_kernel();
return res;
}
/*
* Invalidate the local caches
*/
static void nfs_zap_caches_locked(struct inode *inode)
{
struct nfs_inode *nfsi = NFS_I(inode);
int mode = inode->i_mode;
nfs_inc_stats(inode, NFSIOS_ATTRINVALIDATE);
nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
nfsi->attrtimeo_timestamp = jiffies;
memset(NFS_I(inode)->cookieverf, 0, sizeof(NFS_I(inode)->cookieverf));
if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode))
nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
else
nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE;
}
/*
* Flush all dirty pages, and check for write errors.
*/
static int
nfs_file_flush(struct file *file, fl_owner_t id)
{
struct dentry *dentry = file->f_path.dentry;
struct inode *inode = dentry->d_inode;
dprintk("NFS: flush(%s/%s)\n",
dentry->d_parent->d_name.name,
dentry->d_name.name);
nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
if ((file->f_mode & FMODE_WRITE) == 0)
return 0;
/* Flush writes to the server and return any errors */
return vfs_fsync(file, 0);
}
/*
* Open file
*/
static int
nfs_file_open(struct inode *inode, struct file *filp)
{
int res;
dprintk("NFS: open file(%s/%s)\n",
filp->f_path.dentry->d_parent->d_name.name,
filp->f_path.dentry->d_name.name);
res = nfs_check_flags(filp->f_flags);
if (res)
return res;
nfs_inc_stats(inode, NFSIOS_VFSOPEN);
res = nfs_open(inode, filp);
return res;
}
/* Adjust the file length if we're writing beyond the end */
static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int count)
{
struct inode *inode = page->mapping->host;
loff_t end, i_size;
pgoff_t end_index;
spin_lock(&inode->i_lock);
i_size = i_size_read(inode);
end_index = (i_size - 1) >> PAGE_CACHE_SHIFT;
if (i_size > 0 && page->index < end_index)
goto out;
end = ((loff_t)page->index << PAGE_CACHE_SHIFT) + ((loff_t)offset+count);
if (i_size >= end)
goto out;
i_size_write(inode, end);
nfs_inc_stats(inode, NFSIOS_EXTENDWRITE);
out:
spin_unlock(&inode->i_lock);
}
/*
* Flush any dirty pages for this process, and check for write errors.
* The return status from this call provides a reliable indication of
* whether any write errors occurred for this process.
*/
static int
nfs_fsync(struct file *file, struct dentry *dentry, int datasync)
{
struct nfs_open_context *ctx = (struct nfs_open_context *)file->private_data;
struct inode *inode = dentry->d_inode;
int status;
dfprintk(VFS, "nfs: fsync(%s/%ld)\n", inode->i_sb->s_id, inode->i_ino);
nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
lock_kernel();
status = nfs_wb_all(inode);
if (!status) {
status = ctx->error;
ctx->error = 0;
}
unlock_kernel();
return status;
}
/*
* Flush all dirty pages, and check for write errors.
*
*/
static int
nfs_file_flush(struct file *file, fl_owner_t id)
{
struct nfs_open_context *ctx = nfs_file_open_context(file);
struct inode *inode = file->f_path.dentry->d_inode;
int status;
dfprintk(VFS, "nfs: flush(%s/%ld)\n", inode->i_sb->s_id, inode->i_ino);
if ((file->f_mode & FMODE_WRITE) == 0)
return 0;
nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
/* Ensure that data+attribute caches are up to date after close() */
status = nfs_do_fsync(ctx, inode);
if (!status)
nfs_revalidate_inode(NFS_SERVER(inode), inode);
return status;
}
/*
* Lock a (portion of) a file
*/
static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl)
{
struct inode * inode = filp->f_mapping->host;
dprintk("NFS: nfs_lock(f=%s/%ld, t=%x, fl=%x, r=%Ld:%Ld)\n",
inode->i_sb->s_id, inode->i_ino,
fl->fl_type, fl->fl_flags,
(long long)fl->fl_start, (long long)fl->fl_end);
nfs_inc_stats(inode, NFSIOS_VFSLOCK);
/* No mandatory locks over NFS */
if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK)
return -ENOLCK;
if (IS_GETLK(cmd))
return do_getlk(filp, cmd, fl);
if (fl->fl_type == F_UNLCK)
return do_unlk(filp, cmd, fl);
return do_setlk(filp, cmd, fl);
}
/*
* Flush all dirty pages, and check for write errors.
*/
static int
nfs4_file_flush(struct file *file, fl_owner_t id)
{
struct inode *inode = file_inode(file);
dprintk("NFS: flush(%pD2)\n", file);
nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
if ((file->f_mode & FMODE_WRITE) == 0)
return 0;
/*
* If we're holding a write delegation, then check if we're required
* to flush the i/o on close. If not, then just start the i/o now.
*/
if (!nfs4_delegation_flush_on_close(inode))
return filemap_fdatawrite(file->f_mapping);
/* Flush writes to the server and return any errors */
return vfs_fsync(file, 0);
}
/*
* Flush all dirty pages, and check for write errors.
*/
int
nfs_file_flush(struct file *file, fl_owner_t id)
{
struct inode *inode = file_inode(file);
dprintk("NFS: flush(%pD2)\n", file);
nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
if ((file->f_mode & FMODE_WRITE) == 0)
return 0;
/*
* If we're holding a write delegation, then just start the i/o
* but don't wait for completion (or send a commit).
*/
if (NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE))
return filemap_fdatawrite(file->f_mapping);
/* Flush writes to the server and return any errors */
return vfs_fsync(file, 0);
}
/*
* Flush any dirty pages for this process, and check for write errors.
* The return status from this call provides a reliable indication of
* whether any write errors occurred for this process.
*
* Notice that it clears the NFS_CONTEXT_ERROR_WRITE before synching to
* disk, but it retrieves and clears ctx->error after synching, despite
* the two being set at the same time in nfs_context_set_write_error().
* This is because the former is used to notify the _next_ call to
* nfs_file_write() that a write error occured, and hence cause it to
* fall back to doing a synchronous write.
*/
static int
nfs_file_fsync(struct file *file, int datasync)
{
struct dentry *dentry = file->f_path.dentry;
struct nfs_open_context *ctx = nfs_file_open_context(file);
struct inode *inode = dentry->d_inode;
int have_error, status;
int ret = 0;
dprintk("NFS: fsync file(%s/%s) datasync %d\n",
dentry->d_parent->d_name.name, dentry->d_name.name,
datasync);
nfs_inc_stats(inode, NFSIOS_VFSFSYNC);
have_error = test_and_clear_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags);
status = nfs_commit_inode(inode, FLUSH_SYNC);
have_error |= test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags);
if (have_error)
ret = xchg(&ctx->error, 0);
if (!ret && status < 0)
ret = status;
return ret;
}
/*
* Flush all dirty pages, and check for write errors.
*
*/
static int
nfs_file_flush(struct file *file, fl_owner_t id)
{
struct nfs_open_context *ctx = (struct nfs_open_context *)file->private_data;
struct inode *inode = file->f_dentry->d_inode;
int status;
dfprintk(VFS, "nfs: flush(%s/%ld)\n", inode->i_sb->s_id, inode->i_ino);
if ((file->f_mode & FMODE_WRITE) == 0)
return 0;
nfs_inc_stats(inode, NFSIOS_VFSFLUSH);
lock_kernel();
/* Ensure that data+attribute caches are up to date after close() */
status = nfs_wb_all(inode);
if (!status) {
status = ctx->error;
ctx->error = 0;
if (!status)
nfs_revalidate_inode(NFS_SERVER(inode), inode);
}
unlock_kernel();
return status;
}
/*
* Lock a (portion of) a file
*/
int nfs_lock(struct file *filp, int cmd, struct file_lock *fl)
{
struct inode *inode = filp->f_mapping->host;
int ret = -ENOLCK;
int is_local = 0;
dprintk("NFS: lock(%s/%s, t=%x, fl=%x, r=%lld:%lld)\n",
filp->f_path.dentry->d_parent->d_name.name,
filp->f_path.dentry->d_name.name,
fl->fl_type, fl->fl_flags,
(long long)fl->fl_start, (long long)fl->fl_end);
nfs_inc_stats(inode, NFSIOS_VFSLOCK);
/* No mandatory locks over NFS */
if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK)
goto out_err;
if (NFS_SERVER(inode)->flags & NFS_MOUNT_LOCAL_FCNTL)
is_local = 1;
if (NFS_PROTO(inode)->lock_check_bounds != NULL) {
ret = NFS_PROTO(inode)->lock_check_bounds(fl);
if (ret < 0)
goto out_err;
}
if (IS_GETLK(cmd))
ret = do_getlk(filp, cmd, fl, is_local);
else if (fl->fl_type == F_UNLCK)
ret = do_unlk(filp, cmd, fl, is_local);
else
ret = do_setlk(filp, cmd, fl, is_local);
out_err:
return ret;
}
/*
* Update and possibly write a cached page of an NFS file.
*
* XXX: Keep an eye on generic_file_read to make sure it doesn't do bad
* things with a page scheduled for an RPC call (e.g. invalidate it).
*/
int nfs_updatepage(struct file *file, struct page *page,
unsigned int offset, unsigned int count)
{
struct nfs_open_context *ctx = nfs_file_open_context(file);
struct inode *inode = page->mapping->host;
int status = 0;
nfs_inc_stats(inode, NFSIOS_VFSUPDATEPAGE);
dprintk("NFS: nfs_updatepage(%s/%s %d@%lld)\n",
file->f_path.dentry->d_parent->d_name.name,
file->f_path.dentry->d_name.name, count,
(long long)(page_offset(page) + offset));
/* If we're not using byte range locks, and we know the page
* is up to date, it may be more efficient to extend the write
* to cover the entire page in order to avoid fragmentation
* inefficiencies.
*/
if (nfs_write_pageuptodate(page, inode) &&
inode->i_flock == NULL &&
!(file->f_flags & O_DSYNC)) {
count = max(count + offset, nfs_page_length(page));
offset = 0;
}
status = nfs_writepage_setup(ctx, page, offset, count);
if (status < 0)
nfs_set_pageerror(page);
else
__set_page_dirty_nobuffers(page);
dprintk("NFS: nfs_updatepage returns %d (isize %lld)\n",
status, (long long)i_size_read(inode));
return status;
}
int
nfs_sillyrename(struct inode *dir, struct dentry *dentry)
{
static unsigned int sillycounter;
const int fileidsize = sizeof(NFS_FILEID(dentry->d_inode))*2;
const int countersize = sizeof(sillycounter)*2;
const int slen = sizeof(".nfs")+fileidsize+countersize-1;
char silly[slen+1];
struct dentry *sdentry;
struct rpc_task *task;
int error = -EIO;
dfprintk(VFS, "NFS: silly-rename(%s/%s, ct=%d)\n",
dentry->d_parent->d_name.name, dentry->d_name.name,
dentry->d_count);
nfs_inc_stats(dir, NFSIOS_SILLYRENAME);
error = -EBUSY;
if (dentry->d_flags & DCACHE_NFSFS_RENAMED)
goto out;
sprintf(silly, ".nfs%*.*Lx",
fileidsize, fileidsize,
(unsigned long long)NFS_FILEID(dentry->d_inode));
nfs_inode_return_delegation(dentry->d_inode);
sdentry = NULL;
do {
char *suffix = silly + slen - countersize;
dput(sdentry);
sillycounter++;
sprintf(suffix, "%*.*x", countersize, countersize, sillycounter);
dfprintk(VFS, "NFS: trying to rename %s to %s\n",
dentry->d_name.name, silly);
sdentry = lookup_one_len(silly, dentry->d_parent, slen);
if (IS_ERR(sdentry))
goto out;
} while (sdentry->d_inode != NULL);
error = nfs_async_unlink(dir, dentry);
if (error)
goto out_dput;
error = nfs_copy_dname(sdentry,
(struct nfs_unlinkdata *)dentry->d_fsdata);
if (error) {
nfs_cancel_async_unlink(dentry);
goto out_dput;
}
task = nfs_async_rename(dir, dir, dentry, sdentry);
if (IS_ERR(task)) {
error = -EBUSY;
nfs_cancel_async_unlink(dentry);
goto out_dput;
}
error = rpc_wait_for_completion_task(task);
if (error == 0)
error = task->tk_status;
switch (error) {
case 0:
/* The rename succeeded */
nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
d_move(dentry, sdentry);
break;
case -ERESTARTSYS:
/* The result of the rename is unknown. Play it safe by
* forcing a new lookup */
d_drop(dentry);
d_drop(sdentry);
}
rpc_put_task(task);
out_dput:
dput(sdentry);
out:
return error;
}
/*
* This function is called when the WRITE call is complete.
*/
int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
{
struct nfs_writeargs *argp = &data->args;
struct nfs_writeres *resp = &data->res;
struct nfs_server *server = NFS_SERVER(data->inode);
int status;
dprintk("NFS: %5u nfs_writeback_done (status %d)\n",
task->tk_pid, task->tk_status);
/*
* ->write_done will attempt to use post-op attributes to detect
* conflicting writes by other clients. A strict interpretation
* of close-to-open would allow us to continue caching even if
* another writer had changed the file, but some applications
* depend on tighter cache coherency when writing.
*/
status = NFS_PROTO(data->inode)->write_done(task, data);
if (status != 0)
return status;
nfs_add_stats(data->inode, NFSIOS_SERVERWRITTENBYTES, resp->count);
#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
if (resp->verf->committed < argp->stable && task->tk_status >= 0) {
/* We tried a write call, but the server did not
* commit data to stable storage even though we
* requested it.
* Note: There is a known bug in Tru64 < 5.0 in which
* the server reports NFS_DATA_SYNC, but performs
* NFS_FILE_SYNC. We therefore implement this checking
* as a dprintk() in order to avoid filling syslog.
*/
static unsigned long complain;
if (time_before(complain, jiffies)) {
dprintk("NFS: faulty NFS server %s:"
" (committed = %d) != (stable = %d)\n",
server->nfs_client->cl_hostname,
resp->verf->committed, argp->stable);
complain = jiffies + 300 * HZ;
}
}
#endif
/* Is this a short write? */
if (task->tk_status >= 0 && resp->count < argp->count) {
static unsigned long complain;
nfs_inc_stats(data->inode, NFSIOS_SHORTWRITE);
/* Has the server at least made some progress? */
if (resp->count != 0) {
/* Was this an NFSv2 write or an NFSv3 stable write? */
if (resp->verf->committed != NFS_UNSTABLE) {
/* Resend from where the server left off */
argp->offset += resp->count;
argp->pgbase += resp->count;
argp->count -= resp->count;
} else {
/* Resend as a stable write in order to avoid
* headaches in the case of a server crash.
*/
argp->stable = NFS_FILE_SYNC;
}
nfs4_restart_rpc(task, server->nfs_client);
return -EAGAIN;
}
if (time_before(complain, jiffies)) {
printk(KERN_WARNING
"NFS: Server wrote zero bytes, expected %u.\n",
argp->count);
complain = jiffies + 300 * HZ;
}
/* Can't do anything about it except throw an error. */
task->tk_status = -EIO;
}
nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res);
return 0;
}
/**
* nfs_sillyrename - Perform a silly-rename of a dentry
* @dir: inode of directory that contains dentry
* @dentry: dentry to be sillyrenamed
*
* NFSv2/3 is stateless and the server doesn't know when the client is
* holding a file open. To prevent application problems when a file is
* unlinked while it's still open, the client performs a "silly-rename".
* That is, it renames the file to a hidden file in the same directory,
* and only performs the unlink once the last reference to it is put.
*
* The final cleanup is done during dentry_iput.
*
* (Note: NFSv4 is stateful, and has opens, so in theory an NFSv4 server
* could take responsibility for keeping open files referenced. The server
* would also need to ensure that opened-but-deleted files were kept over
* reboots. However, we may not assume a server does so. (RFC 5661
* does provide an OPEN4_RESULT_PRESERVE_UNLINKED flag that a server can
* use to advertise that it does this; some day we may take advantage of
* it.))
*/
int
nfs_sillyrename(struct inode *dir, struct dentry *dentry)
{
static unsigned int sillycounter;
const int fileidsize = sizeof(NFS_FILEID(dentry->d_inode))*2;
const int countersize = sizeof(sillycounter)*2;
const int slen = sizeof(".nfs")+fileidsize+countersize-1;
char silly[slen+1];
struct dentry *sdentry;
struct rpc_task *task;
int error = -EIO;
dfprintk(VFS, "NFS: silly-rename(%s/%s, ct=%d)\n",
dentry->d_parent->d_name.name, dentry->d_name.name,
dentry->d_count);
nfs_inc_stats(dir, NFSIOS_SILLYRENAME);
/*
* We don't allow a dentry to be silly-renamed twice.
*/
error = -EBUSY;
if (dentry->d_flags & DCACHE_NFSFS_RENAMED)
goto out;
sprintf(silly, ".nfs%*.*Lx",
fileidsize, fileidsize,
(unsigned long long)NFS_FILEID(dentry->d_inode));
/* Return delegation in anticipation of the rename */
NFS_PROTO(dentry->d_inode)->return_delegation(dentry->d_inode);
sdentry = NULL;
do {
char *suffix = silly + slen - countersize;
dput(sdentry);
sillycounter++;
sprintf(suffix, "%*.*x", countersize, countersize, sillycounter);
dfprintk(VFS, "NFS: trying to rename %s to %s\n",
dentry->d_name.name, silly);
sdentry = lookup_one_len(silly, dentry->d_parent, slen);
/*
* N.B. Better to return EBUSY here ... it could be
* dangerous to delete the file while it's in use.
*/
if (IS_ERR(sdentry))
goto out;
} while (sdentry->d_inode != NULL); /* need negative lookup */
/* queue unlink first. Can't do this from rpc_release as it
* has to allocate memory
*/
error = nfs_async_unlink(dir, dentry);
if (error)
goto out_dput;
/* populate unlinkdata with the right dname */
error = nfs_copy_dname(sdentry,
(struct nfs_unlinkdata *)dentry->d_fsdata);
if (error) {
nfs_cancel_async_unlink(dentry);
goto out_dput;
}
/* run the rename task, undo unlink if it fails */
task = nfs_async_rename(dir, dir, dentry, sdentry);
if (IS_ERR(task)) {
error = -EBUSY;
nfs_cancel_async_unlink(dentry);
goto out_dput;
}
/* wait for the RPC task to complete, unless a SIGKILL intervenes */
error = rpc_wait_for_completion_task(task);
if (error == 0)
error = task->tk_status;
switch (error) {
case 0:
/* The rename succeeded */
nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
d_move(dentry, sdentry);
break;
case -ERESTARTSYS:
/* The result of the rename is unknown. Play it safe by
* forcing a new lookup */
d_drop(dentry);
d_drop(sdentry);
}
rpc_put_task(task);
out_dput:
dput(sdentry);
out:
return error;
}
