/*
* Write to a file (through the page cache).
*/
static ssize_t
nfs_file_write(struct file *file, const char *buf, size_t count, loff_t *ppos)
{
struct dentry * dentry = file->f_dentry;
struct inode * inode = dentry->d_inode;
ssize_t result;
dfprintk(VFS, "nfs: write(%s/%s(%ld), %lu@%lu)\n",
dentry->d_parent->d_name.name, dentry->d_name.name,
inode->i_ino, (unsigned long) count, (unsigned long) *ppos);
result = -EBUSY;
if (IS_SWAPFILE(inode))
goto out_swapfile;
result = nfs_revalidate_inode(NFS_DSERVER(dentry), dentry);
if (result)
goto out;
result = count;
if (!count)
goto out;
result = generic_file_write(file, buf, count, ppos);
out:
return result;
out_swapfile:
printk(KERN_INFO "NFS: attempt to write to active swap file!\n");
goto out;
}
static void *nfs_follow_link(struct dentry *dentry, struct nameidata *nd)
{
struct inode *inode = dentry->d_inode;
struct page *page;
void *err = ERR_PTR(nfs_revalidate_inode(NFS_SERVER(inode), inode));
if (err)
goto read_failed;
page = read_cache_page(&inode->i_data, 0,
(filler_t *)nfs_symlink_filler, inode);
if (IS_ERR(page)) {
err = page;
goto read_failed;
}
if (!PageUptodate(page)) {
err = ERR_PTR(-EIO);
goto getlink_read_error;
}
nd_set_link(nd, kmap(page));
return page;
getlink_read_error:
page_cache_release(page);
read_failed:
nd_set_link(nd, err);
return NULL;
}
static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
{
struct inode *inode = NULL;
int error = 0;
/* Check that we are indeed trying to open this file */
if (!is_atomic_open(dir, nd))
goto no_open;
if (dentry->d_name.len > NFS_SERVER(dir)->namelen) {
error = -ENAMETOOLONG;
goto out;
}
dentry->d_op = NFS_PROTO(dir)->dentry_ops;
/* Let vfs_create() deal with O_EXCL */
if (nd->intent.open.flags & O_EXCL)
goto no_entry;
/* Open the file on the server */
lock_kernel();
/* Revalidate parent directory attribute cache */
nfs_revalidate_inode(NFS_SERVER(dir), dir);
if (nd->intent.open.flags & O_CREAT) {
nfs_begin_data_update(dir);
inode = nfs4_atomic_open(dir, dentry, nd);
nfs_end_data_update(dir);
} else
inode = nfs4_atomic_open(dir, dentry, nd);
unlock_kernel();
if (IS_ERR(inode)) {
error = PTR_ERR(inode);
switch (error) {
/* Make a negative dentry */
case -ENOENT:
inode = NULL;
break;
/* This turned out not to be a regular file */
case -ELOOP:
if (!(nd->intent.open.flags & O_NOFOLLOW))
goto no_open;
/* case -EISDIR: */
/* case -EINVAL: */
default:
goto out;
}
}
no_entry:
d_add(dentry, inode);
nfs_renew_times(dentry);
nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
out:
BUG_ON(error > 0);
return ERR_PTR(error);
no_open:
return nfs_lookup(dir, dentry, nd);
}
/*
* A check for whether or not the parent directory has changed.
* In the case it has, we assume that the dentries are untrustworthy
* and may need to be looked up again.
*/
static inline
int nfs_check_verifier(struct inode *dir, struct dentry *dentry)
{
if (IS_ROOT(dentry))
return 1;
if (nfs_revalidate_inode(NFS_SERVER(dir), dir))
return 0;
return time_after(dentry->d_time, NFS_MTIME_UPDATE(dir));
}
static inline
int nfs_lookup_verify_inode(struct inode *inode, int isopen)
{
struct nfs_server *server = NFS_SERVER(inode);
if (isopen && !(server->flags & NFS_MOUNT_NOCTO))
return __nfs_revalidate_inode(server, inode);
return nfs_revalidate_inode(server, inode);
}
static inline
int nfs_lookup_verify_inode(struct inode *inode, int flags)
{
struct nfs_server *server = NFS_SERVER(inode);
/*
* If we're interested in close-to-open cache consistency,
* then we revalidate the inode upon lookup.
*/
if (!(server->flags & NFS_MOUNT_NOCTO) && !(flags & LOOKUP_CONTINUE))
NFS_CACHEINV(inode);
return nfs_revalidate_inode(server, inode);
}
static int
nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
{
struct dentry *dentry = file->f_dentry;
int status;
dfprintk(VFS, "nfs: mmap(%s/%s)\n",
dentry->d_parent->d_name.name, dentry->d_name.name);
status = nfs_revalidate_inode(NFS_DSERVER(dentry), dentry);
if (!status)
status = generic_file_mmap(file, vma);
return status;
}
static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd)
{
struct inode *inode = NULL;
int error;
struct nfs_fh fhandle;
struct nfs_fattr fattr;
dfprintk(VFS, "NFS: lookup(%s/%s)\n",
dentry->d_parent->d_name.name, dentry->d_name.name);
error = -ENAMETOOLONG;
if (dentry->d_name.len > NFS_SERVER(dir)->namelen)
goto out;
error = -ENOMEM;
dentry->d_op = NFS_PROTO(dir)->dentry_ops;
lock_kernel();
/* Revalidate parent directory attribute cache */
nfs_revalidate_inode(NFS_SERVER(dir), dir);
/* If we're doing an exclusive create, optimize away the lookup */
if (nfs_is_exclusive_create(dir, nd))
goto no_entry;
error = nfs_cached_lookup(dir, dentry, &fhandle, &fattr);
if (error != 0) {
error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name,
&fhandle, &fattr);
if (error == -ENOENT)
goto no_entry;
if (error != 0)
goto out_unlock;
}
error = -EACCES;
inode = nfs_fhget(dentry->d_sb, &fhandle, &fattr);
if (!inode)
goto out_unlock;
no_entry:
error = 0;
d_add(dentry, inode);
nfs_renew_times(dentry);
nfs_set_verifier(dentry, nfs_save_change_attribute(dir));
out_unlock:
unlock_kernel();
out:
BUG_ON(error > 0);
return ERR_PTR(error);
}
static ssize_t
nfs_file_read(struct file * file, char * buf, size_t count, loff_t *ppos)
{
struct dentry * dentry = file->f_dentry;
ssize_t result;
dfprintk(VFS, "nfs: read(%s/%s, %lu@%lu)\n",
dentry->d_parent->d_name.name, dentry->d_name.name,
(unsigned long) count, (unsigned long) *ppos);
result = nfs_revalidate_inode(NFS_DSERVER(dentry), dentry);
if (!result)
result = generic_file_read(file, buf, count, ppos);
return result;
}
/*
* 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;
}
/*
* Flush all dirty pages, and check for write errors.
*
*/
static int
nfs_file_flush(struct file *file)
{
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;
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;
}
int
nfs_permission(struct inode *inode, int mask)
{
struct nfs_server *server = NFS_SERVER(inode);
struct nfs_access_cache *cache = &NFS_I(inode)->cache_access;
struct rpc_cred *cred;
int mode = inode->i_mode;
int error;
if (mask & MAY_WRITE) {
/*
*
* Nobody gets write access to a read-only fs.
*
*/
if (IS_RDONLY(inode) &&
(S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)))
return -EROFS;
/*
*
* Nobody gets write access to an immutable file.
*
*/
if (IS_IMMUTABLE(inode))
return -EACCES;
}
if ((server->flags & NFS_MOUNT_NOACL) || !NFS_PROTO(inode)->access)
goto out_notsup;
cred = rpcauth_lookupcred(NFS_CLIENT(inode)->cl_auth, 0);
if (cache->cred == cred
&& time_before(jiffies, cache->jiffies + NFS_ATTRTIMEO(inode))) {
if (!cache->err) {
/* Is the mask a subset of an accepted mask? */
if ((cache->mask & mask) == mask)
goto out_cached;
} else {
/* ...or is it a superset of a rejected mask? */
if ((cache->mask & mask) == cache->mask)
goto out_cached;
}
}
error = NFS_PROTO(inode)->access(inode, cred, mask);
if (!error || error == -EACCES) {
cache->jiffies = jiffies;
if (cache->cred)
put_rpccred(cache->cred);
cache->cred = cred;
cache->mask = mask;
cache->err = error;
return error;
}
put_rpccred(cred);
out_notsup:
nfs_revalidate_inode(NFS_SERVER(inode), inode);
return vfs_permission(inode, mask);
out_cached:
put_rpccred(cred);
return cache->err;
}
struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
{
struct nfs_server *server = NFS_SERVER(inode);
struct nfs_fattr fattr;
struct page *pages[NFSACL_MAXPAGES] = { };
struct nfs3_getaclargs args = {
.fh = NFS_FH(inode),
/* The xdr layer may allocate pages here. */
.pages = pages,
};
struct nfs3_getaclres res = {
.fattr = &fattr,
};
struct rpc_message msg = {
.rpc_argp = &args,
.rpc_resp = &res,
};
struct posix_acl *acl;
int status, count;
if (!nfs_server_capable(inode, NFS_CAP_ACLS))
return ERR_PTR(-EOPNOTSUPP);
status = nfs_revalidate_inode(server, inode);
if (status < 0)
return ERR_PTR(status);
if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL)
nfs_zap_acl_cache(inode);
acl = nfs3_get_cached_acl(inode, type);
if (acl != ERR_PTR(-EAGAIN))
return acl;
acl = NULL;
/*
* Only get the access acl when explicitly requested: We don't
* need it for access decisions, and only some applications use
* it. Applications which request the access acl first are not
* penalized from this optimization.
*/
if (type == ACL_TYPE_ACCESS)
args.mask |= NFS_ACLCNT|NFS_ACL;
if (S_ISDIR(inode->i_mode))
args.mask |= NFS_DFACLCNT|NFS_DFACL;
if (args.mask == 0)
return NULL;
dprintk("NFS call getacl\n");
msg.rpc_proc = &server->client_acl->cl_procinfo[ACLPROC3_GETACL];
nfs_fattr_init(&fattr);
status = rpc_call_sync(server->client_acl, &msg, 0);
dprintk("NFS reply getacl: %d\n", status);
/* pages may have been allocated at the xdr layer. */
for (count = 0; count < NFSACL_MAXPAGES && args.pages[count]; count++)
__free_page(args.pages[count]);
switch (status) {
case 0:
status = nfs_refresh_inode(inode, &fattr);
break;
case -EPFNOSUPPORT:
case -EPROTONOSUPPORT:
dprintk("NFS_V3_ACL extension not supported; disabling\n");
server->caps &= ~NFS_CAP_ACLS;
case -ENOTSUPP:
status = -EOPNOTSUPP;
default:
goto getout;
}
if ((args.mask & res.mask) != args.mask) {
status = -EIO;
goto getout;
}
if (res.acl_access != NULL) {
if (posix_acl_equiv_mode(res.acl_access, NULL) == 0) {
posix_acl_release(res.acl_access);
res.acl_access = NULL;
}
}
nfs3_cache_acls(inode,
(res.mask & NFS_ACL) ? res.acl_access : ERR_PTR(-EINVAL),
(res.mask & NFS_DFACL) ? res.acl_default : ERR_PTR(-EINVAL));
switch(type) {
case ACL_TYPE_ACCESS:
acl = res.acl_access;
res.acl_access = NULL;
break;
case ACL_TYPE_DEFAULT:
acl = res.acl_default;
res.acl_default = NULL;
}
getout:
posix_acl_release(res.acl_access);
posix_acl_release(res.acl_default);
if (status != 0) {
posix_acl_release(acl);
acl = ERR_PTR(status);
}
return acl;
}
/*
* This is called every time the dcache has a lookup hit,
* and we should check whether we can really trust that
* lookup.
*
* NOTE! The hit can be a negative hit too, don't assume
* we have an inode!
*
* If the parent directory is seen to have changed, we throw out the
* cached dentry and do a new lookup.
*/
static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd)
{
struct inode *dir;
struct inode *inode;
struct dentry *parent;
int error;
struct nfs_fh fhandle;
struct nfs_fattr fattr;
unsigned long verifier;
int isopen = 0;
parent = dget_parent(dentry);
lock_kernel();
dir = parent->d_inode;
inode = dentry->d_inode;
if (nd && !(nd->flags & LOOKUP_CONTINUE) && (nd->flags & LOOKUP_OPEN))
isopen = 1;
if (!inode) {
if (nfs_neg_need_reval(dir, dentry, nd))
goto out_bad;
goto out_valid;
}
if (is_bad_inode(inode)) {
dfprintk(VFS, "nfs_lookup_validate: %s/%s has dud inode\n",
dentry->d_parent->d_name.name, dentry->d_name.name);
goto out_bad;
}
/* Revalidate parent directory attribute cache */
nfs_revalidate_inode(NFS_SERVER(dir), dir);
/* Force a full look up iff the parent directory has changed */
if (nfs_check_verifier(dir, dentry)) {
if (nfs_lookup_verify_inode(inode, isopen))
goto out_zap_parent;
goto out_valid;
}
/*
* Note: we're not holding inode->i_sem and so may be racing with
* operations that change the directory. We therefore save the
* change attribute *before* we do the RPC call.
*/
verifier = nfs_save_change_attribute(dir);
error = nfs_cached_lookup(dir, dentry, &fhandle, &fattr);
if (!error) {
if (memcmp(NFS_FH(inode), &fhandle, sizeof(struct nfs_fh))!= 0)
goto out_bad;
if (nfs_lookup_verify_inode(inode, isopen))
goto out_zap_parent;
goto out_valid_renew;
}
if (NFS_STALE(inode))
goto out_bad;
error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, &fhandle, &fattr);
if (error)
goto out_bad;
if (memcmp(NFS_FH(inode), &fhandle, sizeof(struct nfs_fh))!= 0)
goto out_bad;
if ((error = nfs_refresh_inode(inode, &fattr)) != 0)
goto out_bad;
out_valid_renew:
nfs_renew_times(dentry);
nfs_set_verifier(dentry, verifier);
out_valid:
unlock_kernel();
dput(parent);
return 1;
out_zap_parent:
nfs_zap_caches(dir);
out_bad:
NFS_CACHEINV(dir);
if (inode && S_ISDIR(inode->i_mode)) {
/* Purge readdir caches. */
nfs_zap_caches(inode);
/* If we have submounts, don't unhash ! */
if (have_submounts(dentry))
goto out_valid;
shrink_dcache_parent(dentry);
}
d_drop(dentry);
unlock_kernel();
dput(parent);
return 0;
}
/* The file offset position is now represented as a true offset into the
* page cache as is the case in most of the other filesystems.
*/
static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
{
struct dentry *dentry = filp->f_dentry;
struct inode *inode = dentry->d_inode;
nfs_readdir_descriptor_t my_desc,
*desc = &my_desc;
struct nfs_entry my_entry;
struct nfs_fh fh;
struct nfs_fattr fattr;
long res;
lock_kernel();
res = nfs_revalidate_inode(NFS_SERVER(inode), inode);
if (res < 0) {
unlock_kernel();
return res;
}
/*
* filp->f_pos points to the file offset in the page cache.
* but if the cache has meanwhile been zapped, we need to
* read from the last dirent to revalidate f_pos
* itself.
*/
memset(desc, 0, sizeof(*desc));
desc->file = filp;
desc->target = filp->f_pos;
desc->decode = NFS_PROTO(inode)->decode_dirent;
desc->plus = NFS_USE_READDIRPLUS(inode);
my_entry.cookie = my_entry.prev_cookie = 0;
my_entry.eof = 0;
my_entry.fh = &fh;
my_entry.fattr = &fattr;
desc->entry = &my_entry;
while(!desc->entry->eof) {
res = readdir_search_pagecache(desc);
if (res == -EBADCOOKIE) {
/* This means either end of directory */
if (desc->entry->cookie != desc->target) {
/* Or that the server has 'lost' a cookie */
res = uncached_readdir(desc, dirent, filldir);
if (res >= 0)
continue;
}
res = 0;
break;
}
if (res == -ETOOSMALL && desc->plus) {
NFS_FLAGS(inode) &= ~NFS_INO_ADVISE_RDPLUS;
nfs_zap_caches(inode);
desc->plus = 0;
desc->entry->eof = 0;
continue;
}
if (res < 0)
break;
res = nfs_do_filldir(desc, dirent, filldir);
if (res < 0) {
res = 0;
break;
}
}
unlock_kernel();
if (desc->error < 0)
return desc->error;
if (res < 0)
return res;
return 0;
}
int
nfs_permission(struct inode *inode, int mask, struct nameidata *nd)
{
struct nfs_access_cache *cache = &NFS_I(inode)->cache_access;
struct rpc_cred *cred;
int mode = inode->i_mode;
int res;
if (mask == 0)
return 0;
if (mask & MAY_WRITE) {
/*
*
* Nobody gets write access to a read-only fs.
*
*/
if (IS_RDONLY(inode) &&
(S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)))
return -EROFS;
/*
*
* Nobody gets write access to an immutable file.
*
*/
if (IS_IMMUTABLE(inode))
return -EACCES;
}
/* Are we checking permissions on anything other than lookup/execute? */
if ((mask & MAY_EXEC) == 0) {
/* We only need to check permissions on file open() and access() */
if (!nd || !(nd->flags & (LOOKUP_OPEN|LOOKUP_ACCESS)))
return 0;
/* NFSv4 has atomic_open... */
if (NFS_PROTO(inode)->version > 3 && (nd->flags & LOOKUP_OPEN))
return 0;
}
lock_kernel();
if (!NFS_PROTO(inode)->access)
goto out_notsup;
cred = rpcauth_lookupcred(NFS_CLIENT(inode)->cl_auth, 0);
if (cache->cred == cred
&& time_before(jiffies, cache->jiffies + NFS_ATTRTIMEO(inode))
&& !(NFS_FLAGS(inode) & NFS_INO_INVALID_ATTR)) {
if (!(res = cache->err)) {
/* Is the mask a subset of an accepted mask? */
if ((cache->mask & mask) == mask)
goto out;
} else {
/* ...or is it a superset of a rejected mask? */
if ((cache->mask & mask) == cache->mask)
goto out;
}
}
res = NFS_PROTO(inode)->access(inode, cred, mask);
if (!res || res == -EACCES)
goto add_cache;
out:
put_rpccred(cred);
unlock_kernel();
return res;
out_notsup:
nfs_revalidate_inode(NFS_SERVER(inode), inode);
res = vfs_permission(inode, mask);
unlock_kernel();
return res;
add_cache:
cache->jiffies = jiffies;
if (cache->cred)
put_rpccred(cache->cred);
cache->cred = cred;
cache->mask = mask;
cache->err = res;
unlock_kernel();
return res;
}
