vfs.c

linux 内核源代码

out_nfserr:
	err = nfserrno(host_err);
	goto out;
}

/*
 * Create a hardlink
 * N.B. After this call _both_ ffhp and tfhp need an fh_put
 */
__be32
nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
				char *name, int len, struct svc_fh *tfhp)
{
	struct dentry	*ddir, *dnew, *dold;
	struct inode	*dirp, *dest;
	__be32		err;
	int		host_err;

	err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_CREATE);
	if (err)
		goto out;
	err = fh_verify(rqstp, tfhp, -S_IFDIR, MAY_NOP);
	if (err)
		goto out;

	err = nfserr_perm;
	if (!len)
		goto out;
	err = nfserr_exist;
	if (isdotent(name, len))
		goto out;

	fh_lock_nested(ffhp, I_MUTEX_PARENT);
	ddir = ffhp->fh_dentry;
	dirp = ddir->d_inode;

	dnew = lookup_one_len(name, ddir, len);
	host_err = PTR_ERR(dnew);
	if (IS_ERR(dnew))
		goto out_nfserr;

	dold = tfhp->fh_dentry;
	dest = dold->d_inode;

	host_err = vfs_link(dold, dirp, dnew);
	if (!host_err) {
		if (EX_ISSYNC(ffhp->fh_export)) {
			err = nfserrno(nfsd_sync_dir(ddir));
			write_inode_now(dest, 1);
		}
		err = 0;
	} else {
		if (host_err == -EXDEV && rqstp->rq_vers == 2)
			err = nfserr_acces;
		else
			err = nfserrno(host_err);
	}

	dput(dnew);
out_unlock:
	fh_unlock(ffhp);
out:
	return err;

out_nfserr:
	err = nfserrno(host_err);
	goto out_unlock;
}

/*
 * Rename a file
 * N.B. After this call _both_ ffhp and tfhp need an fh_put
 */
__be32
nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
			    struct svc_fh *tfhp, char *tname, int tlen)
{
	struct dentry	*fdentry, *tdentry, *odentry, *ndentry, *trap;
	struct inode	*fdir, *tdir;
	__be32		err;
	int		host_err;

	err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_REMOVE);
	if (err)
		goto out;
	err = fh_verify(rqstp, tfhp, S_IFDIR, MAY_CREATE);
	if (err)
		goto out;

	fdentry = ffhp->fh_dentry;
	fdir = fdentry->d_inode;

	tdentry = tfhp->fh_dentry;
	tdir = tdentry->d_inode;

	err = (rqstp->rq_vers == 2) ? nfserr_acces : nfserr_xdev;
	if (ffhp->fh_export != tfhp->fh_export)
		goto out;

	err = nfserr_perm;
	if (!flen || isdotent(fname, flen) || !tlen || isdotent(tname, tlen))
		goto out;

	/* cannot use fh_lock as we need deadlock protective ordering
	 * so do it by hand */
	trap = lock_rename(tdentry, fdentry);
	ffhp->fh_locked = tfhp->fh_locked = 1;
	fill_pre_wcc(ffhp);
	fill_pre_wcc(tfhp);

	odentry = lookup_one_len(fname, fdentry, flen);
	host_err = PTR_ERR(odentry);
	if (IS_ERR(odentry))
		goto out_nfserr;

	host_err = -ENOENT;
	if (!odentry->d_inode)
		goto out_dput_old;
	host_err = -EINVAL;
	if (odentry == trap)
		goto out_dput_old;

	ndentry = lookup_one_len(tname, tdentry, tlen);
	host_err = PTR_ERR(ndentry);
	if (IS_ERR(ndentry))
		goto out_dput_old;
	host_err = -ENOTEMPTY;
	if (ndentry == trap)
		goto out_dput_new;

#ifdef MSNFS
	if ((ffhp->fh_export->ex_flags & NFSEXP_MSNFS) &&
		((atomic_read(&odentry->d_count) > 1)
		 || (atomic_read(&ndentry->d_count) > 1))) {
			host_err = -EPERM;
	} else
#endif
	host_err = vfs_rename(fdir, odentry, tdir, ndentry);
	if (!host_err && EX_ISSYNC(tfhp->fh_export)) {
		host_err = nfsd_sync_dir(tdentry);
		if (!host_err)
			host_err = nfsd_sync_dir(fdentry);
	}

 out_dput_new:
	dput(ndentry);
 out_dput_old:
	dput(odentry);
 out_nfserr:
	err = nfserrno(host_err);

	/* we cannot reply on fh_unlock on the two filehandles,
	 * as that would do the wrong thing if the two directories
	 * were the same, so again we do it by hand
	 */
	fill_post_wcc(ffhp);
	fill_post_wcc(tfhp);
	unlock_rename(tdentry, fdentry);
	ffhp->fh_locked = tfhp->fh_locked = 0;

out:
	return err;
}

/*
 * Unlink a file or directory
 * N.B. After this call fhp needs an fh_put
 */
__be32
nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
				char *fname, int flen)
{
	struct dentry	*dentry, *rdentry;
	struct inode	*dirp;
	__be32		err;
	int		host_err;

	err = nfserr_acces;
	if (!flen || isdotent(fname, flen))
		goto out;
	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_REMOVE);
	if (err)
		goto out;

	fh_lock_nested(fhp, I_MUTEX_PARENT);
	dentry = fhp->fh_dentry;
	dirp = dentry->d_inode;

	rdentry = lookup_one_len(fname, dentry, flen);
	host_err = PTR_ERR(rdentry);
	if (IS_ERR(rdentry))
		goto out_nfserr;

	if (!rdentry->d_inode) {
		dput(rdentry);
		err = nfserr_noent;
		goto out;
	}

	if (!type)
		type = rdentry->d_inode->i_mode & S_IFMT;

	if (type != S_IFDIR) { /* It's UNLINK */
#ifdef MSNFS
		if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) &&
			(atomic_read(&rdentry->d_count) > 1)) {
			host_err = -EPERM;
		} else
#endif
		host_err = vfs_unlink(dirp, rdentry);
	} else { /* It's RMDIR */
		host_err = vfs_rmdir(dirp, rdentry);
	}

	dput(rdentry);

	if (host_err)
		goto out_nfserr;
	if (EX_ISSYNC(fhp->fh_export))
		host_err = nfsd_sync_dir(dentry);

out_nfserr:
	err = nfserrno(host_err);
out:
	return err;
}

/*
 * Read entries from a directory.
 * The  NFSv3/4 verifier we ignore for now.
 */
__be32
nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, 
	     struct readdir_cd *cdp, filldir_t func)
{
	__be32		err;
	int 		host_err;
	struct file	*file;
	loff_t		offset = *offsetp;

	err = nfsd_open(rqstp, fhp, S_IFDIR, MAY_READ, &file);
	if (err)
		goto out;

	offset = vfs_llseek(file, offset, 0);
	if (offset < 0) {
		err = nfserrno((int)offset);
		goto out_close;
	}

	/*
	 * Read the directory entries. This silly loop is necessary because
	 * readdir() is not guaranteed to fill up the entire buffer, but
	 * may choose to do less.
	 */

	do {
		cdp->err = nfserr_eof; /* will be cleared on successful read */
		host_err = vfs_readdir(file, func, cdp);
	} while (host_err >=0 && cdp->err == nfs_ok);
	if (host_err)
		err = nfserrno(host_err);
	else
		err = cdp->err;
	*offsetp = vfs_llseek(file, 0, 1);

	if (err == nfserr_eof || err == nfserr_toosmall)
		err = nfs_ok; /* can still be found in ->err */
out_close:
	nfsd_close(file);
out:
	return err;
}

/*
 * Get file system stats
 * N.B. After this call fhp needs an fh_put
 */
__be32
nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat)
{
	__be32 err = fh_verify(rqstp, fhp, 0, MAY_NOP);
	if (!err && vfs_statfs(fhp->fh_dentry,stat))
		err = nfserr_io;
	return err;
}

static int exp_rdonly(struct svc_rqst *rqstp, struct svc_export *exp)
{
	return nfsexp_flags(rqstp, exp) & NFSEXP_READONLY;
}

/*
 * Check for a user's access permissions to this inode.
 */
__be32
nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
					struct dentry *dentry, int acc)
{
	struct inode	*inode = dentry->d_inode;
	int		err;

	if (acc == MAY_NOP)
		return 0;
#if 0
	dprintk("nfsd: permission 0x%x%s%s%s%s%s%s%s mode 0%o%s%s%s\n",
		acc,
		(acc & MAY_READ)?	" read"  : "",
		(acc & MAY_WRITE)?	" write" : "",
		(acc & MAY_EXEC)?	" exec"  : "",
		(acc & MAY_SATTR)?	" sattr" : "",
		(acc & MAY_TRUNC)?	" trunc" : "",
		(acc & MAY_LOCK)?	" lock"  : "",
		(acc & MAY_OWNER_OVERRIDE)? " owneroverride" : "",
		inode->i_mode,
		IS_IMMUTABLE(inode)?	" immut" : "",
		IS_APPEND(inode)?	" append" : "",
		IS_RDONLY(inode)?	" ro" : "");
	dprintk("      owner %d/%d user %d/%d\n",
		inode->i_uid, inode->i_gid, current->fsuid, current->fsgid);
#endif

	/* Normally we reject any write/sattr etc access on a read-only file
	 * system.  But if it is IRIX doing check on write-access for a 
	 * device special file, we ignore rofs.
	 */
	if (!(acc & MAY_LOCAL_ACCESS))
		if (acc & (MAY_WRITE | MAY_SATTR | MAY_TRUNC)) {
			if (exp_rdonly(rqstp, exp) || IS_RDONLY(inode))
				return nfserr_rofs;
			if (/* (acc & MAY_WRITE) && */ IS_IMMUTABLE(inode))
				return nfserr_perm;
		}
	if ((acc & MAY_TRUNC) && IS_APPEND(inode))
		return nfserr_perm;

	if (acc & MAY_LOCK) {
		/* If we cannot rely on authentication in NLM requests,
		 * just allow locks, otherwise require read permission, or
		 * ownership
		 */
		if (exp->ex_flags & NFSEXP_NOAUTHNLM)
			return 0;
		else
			acc = MAY_READ | MAY_OWNER_OVERRIDE;
	}
	/*
	 * The file owner always gets access permission for accesses that
	 * would normally be checked at open time. This is to make
	 * file access work even when the client has done a fchmod(fd, 0).
	 *
	 * However, `cp foo bar' should fail nevertheless when bar is
	 * readonly. A sensible way to do this might be to reject all
	 * attempts to truncate a read-only file, because a creat() call
	 * always implies file truncation.
	 * ... but this isn't really fair.  A process may reasonably call
	 * ftruncate on an open file descriptor on a file with perm 000.
	 * We must trust the client to do permission checking - using "ACCESS"
	 * with NFSv3.
	 */
	if ((acc & MAY_OWNER_OVERRIDE) &&
	    inode->i_uid == current->fsuid)
		return 0;

	err = permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC), NULL);

	/* Allow read access to binaries even when mode 111 */
	if (err == -EACCES && S_ISREG(inode->i_mode) &&
	    acc == (MAY_READ | MAY_OWNER_OVERRIDE))
		err = permission(inode, MAY_EXEC, NULL);

	return err? nfserrno(err) : 0;
}

void
nfsd_racache_shutdown(void)
{
	if (!raparml)
		return;
	dprintk("nfsd: freeing readahead buffers.\n");
	kfree(raparml);
	raparml = NULL;
}
/*
 * Initialize readahead param cache
 */
int
nfsd_racache_init(int cache_size)
{
	int	i;
	int	j = 0;
	int	nperbucket;


	if (raparml)
		return 0;
	if (cache_size < 2*RAPARM_HASH_SIZE)
		cache_size = 2*RAPARM_HASH_SIZE;
	raparml = kcalloc(cache_size, sizeof(struct raparms), GFP_KERNEL);

	if (!raparml) {
		printk(KERN_WARNING
			"nfsd: Could not allocate memory read-ahead cache.\n");
		return -ENOMEM;
	}

	dprintk("nfsd: allocating %d readahead buffers.\n", cache_size);
	for (i = 0 ; i < RAPARM_HASH_SIZE ; i++) {
		raparm_hash[i].pb_head = NULL;
		spin_lock_init(&raparm_hash[i].pb_lock);
	}
	nperbucket = DIV_ROUND_UP(cache_size, RAPARM_HASH_SIZE);
	for (i = 0; i < cache_size - 1; i++) {
		if (i % nperbucket == 0)
			raparm_hash[j++].pb_head = raparml + i;
		if (i % nperbucket < nperbucket-1)
			raparml[i].p_next = raparml + i + 1;
	}

	nfsdstats.ra_size = cache_size;
	return 0;
}

#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
struct posix_acl *
nfsd_get_posix_acl(struct svc_fh *fhp, int type)
{
	struct inode *inode = fhp->fh_dentry->d_inode;
	char *name;
	void *value = NULL;
	ssize_t size;
	struct posix_acl *acl;

	if (!IS_POSIXACL(inode))
		return ERR_PTR(-EOPNOTSUPP);

	switch (type) {
	case ACL_TYPE_ACCESS:
		name = POSIX_ACL_XATTR_ACCESS;
		break;
	case ACL_TYPE_DEFAULT:
		name = POSIX_ACL_XATTR_DEFAULT;
		break;
	default:
		return ERR_PTR(-EOPNOTSUPP);
	}

	size = nfsd_getxattr(fhp->fh_dentry, name, &value);
	if (size < 0)
		return ERR_PTR(size);

	acl = posix_acl_from_xattr(value, size);
	kfree(value);
	return acl;
}

int
nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl)
{
	struct inode *inode = fhp->fh_dentry->d_inode;
	char *name;
	void *value = NULL;
	size_t size;
	int error;

	if (!IS_POSIXACL(inode) || !inode->i_op ||
	    !inode->i_op->setxattr || !inode->i_op->removexattr)
		return -EOPNOTSUPP;
	switch(type) {
		case ACL_TYPE_ACCESS:
			name = POSIX_ACL_XATTR_ACCESS;
			break;
		case ACL_TYPE_DEFAULT:
			name = POSIX_ACL_XATTR_DEFAULT;
			break;
		default:
			return -EOPNOTSUPP;
	}

	if (acl && acl->a_count) {
		size = posix_acl_xattr_size(acl->a_count);
		value = kmalloc(size, GFP_KERNEL);
		if (!value)
			return -ENOMEM;
		error = posix_acl_to_xattr(acl, value, size);
		if (error < 0)
			goto getout;
		size = error;
	} else
		size = 0;

	if (size)
		error = vfs_setxattr(fhp->fh_dentry, name, value, size, 0);
	else {
		if (!S_ISDIR(inode->i_mode) && type == ACL_TYPE_DEFAULT)
			error = 0;
		else {
			error = vfs_removexattr(fhp->fh_dentry, name);
			if (error == -ENODATA)
				error = 0;
		}
	}

getout:
	kfree(value);
	return error;
}
#endif  /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */