NetBSD/sys/nfs/nfs_export.c

878 lines
22 KiB
C
Raw Normal View History

/* $NetBSD: nfs_export.c,v 1.49 2010/11/19 06:44:46 dholland Exp $ */
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*-
* Copyright (c) 1997, 1998, 2004, 2005, 2008 The NetBSD Foundation, Inc.
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
* NASA Ames Research Center.
* This code is derived from software contributed to The NetBSD Foundation
* by Charles M. Hannum.
* This code is derived from software contributed to The NetBSD Foundation
* by Julio M. Merino Vidal.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
/*
* Copyright (c) 1989, 1993
* The Regents of the University of California. All rights reserved.
* (c) UNIX System Laboratories, Inc.
* All or some portions of this file are derived from material licensed
* to the University of California by American Telephone and Telegraph
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
* the permission of UNIX System Laboratories, Inc.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)vfs_subr.c 8.13 (Berkeley) 4/18/94
*/
/*
* VFS exports list management.
*
* Lock order: vfs_busy -> mnt_updating -> netexport_lock.
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: nfs_export.c,v 1.49 2010/11/19 06:44:46 dholland Exp $");
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
#include <sys/queue.h>
#include <sys/proc.h>
#include <sys/mount.h>
#include <sys/vnode.h>
#include <sys/namei.h>
#include <sys/errno.h>
#include <sys/malloc.h>
#include <sys/domain.h>
#include <sys/mbuf.h>
#include <sys/dirent.h>
#include <sys/socket.h> /* XXX for AF_MAX */
2006-05-15 01:31:52 +04:00
#include <sys/kauth.h>
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
#include <net/radix.h>
#include <netinet/in.h>
#include <nfs/rpcv2.h>
#include <nfs/nfsproto.h>
#include <nfs/nfs.h>
#include <nfs/nfs_var.h>
/*
* Network address lookup element.
*/
struct netcred {
struct radix_node netc_rnodes[2];
int netc_refcnt;
int netc_exflags;
2006-05-15 01:31:52 +04:00
kauth_cred_t netc_anon;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
};
/*
* Network export information.
*/
struct netexport {
CIRCLEQ_ENTRY(netexport) ne_list;
struct mount *ne_mount;
struct netcred ne_defexported; /* Default export */
struct radix_node_head *ne_rtable[AF_MAX+1]; /* Individual exports */
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
};
CIRCLEQ_HEAD(, netexport) netexport_list =
CIRCLEQ_HEAD_INITIALIZER(netexport_list);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/* Publicly exported file system. */
struct nfs_public nfs_pub;
/*
* Local prototypes.
*/
static int init_exports(struct mount *, struct netexport **);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
static int hang_addrlist(struct mount *, struct netexport *,
const struct export_args *);
static int sacheck(struct sockaddr *);
static int free_netcred(struct radix_node *, void *);
static int export(struct netexport *, const struct export_args *);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
static int setpublicfs(struct mount *, struct netexport *,
const struct export_args *);
static struct netcred *netcred_lookup(struct netexport *, struct mbuf *);
static struct netexport *netexport_lookup(const struct mount *);
static struct netexport *netexport_lookup_byfsid(const fsid_t *);
static void netexport_clear(struct netexport *);
static void netexport_insert(struct netexport *);
static void netexport_remove(struct netexport *);
static void netexport_wrlock(void);
static void netexport_wrunlock(void);
static int nfs_export_update_30(struct mount *mp, const char *path, void *);
static krwlock_t netexport_lock;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*
* PUBLIC INTERFACE
*/
/*
* Declare and initialize the file system export hooks.
*/
static void netexport_unmount(struct mount *);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
struct vfs_hooks nfs_export_hooks = {
{ NULL, NULL },
.vh_unmount = netexport_unmount,
.vh_reexport = nfs_export_update_30,
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
};
/*
* VFS unmount hook for NFS exports.
*
* Releases NFS exports list resources if the given mount point has some.
* As allocation happens lazily, it may be that it doesn't has this
* information, although it theorically should.
*/
static void
netexport_unmount(struct mount *mp)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
{
struct netexport *ne;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
KASSERT(mp != NULL);
netexport_wrlock();
ne = netexport_lookup(mp);
if (ne == NULL) {
netexport_wrunlock();
return;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
netexport_clear(ne);
netexport_remove(ne);
netexport_wrunlock();
kmem_free(ne, sizeof(*ne));
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
void
netexport_init(void)
{
rw_init(&netexport_lock);
}
void
netexport_fini(void)
{
struct netexport *ne;
struct mount *mp;
int error;
while (!CIRCLEQ_EMPTY(&netexport_list)) {
netexport_wrlock();
ne = CIRCLEQ_FIRST(&netexport_list);
mp = ne->ne_mount;
error = vfs_busy(mp, NULL);
netexport_wrunlock();
if (error != 0) {
kpause("nfsfini", false, hz, NULL);
continue;
}
mutex_enter(&mp->mnt_updating); /* mnt_flag */
netexport_unmount(mp);
mutex_exit(&mp->mnt_updating); /* mnt_flag */
vfs_unbusy(mp, false, NULL);
}
rw_destroy(&netexport_lock);
}
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*
* Atomically set the NFS exports list of the given file system, replacing
* it with a new list of entries.
*
* Returns zero on success or an appropriate error code otherwise.
*
* Helper function for the nfssvc(2) system call (NFSSVC_SETEXPORTSLIST
* command).
*/
int
mountd_set_exports_list(const struct mountd_exports_list *mel, struct lwp *l,
struct mount *nmp)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
{
int error;
#ifdef notyet
/* XXX: See below to see the reason why this is disabled. */
size_t i;
#endif
struct mount *mp;
struct netexport *ne;
struct pathbuf *pb;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
struct nameidata nd;
struct vnode *vp;
size_t fid_size;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
if (kauth_authorize_network(l->l_cred, KAUTH_NETWORK_NFS,
KAUTH_REQ_NETWORK_NFS_EXPORT, NULL, NULL, NULL) != 0)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return EPERM;
/* Lookup the file system path. */
error = pathbuf_copyin(mel->mel_path, &pb);
if (error) {
return error;
}
NDINIT(&nd, LOOKUP, FOLLOW | LOCKLEAF, pb);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
error = namei(&nd);
if (error != 0) {
pathbuf_destroy(pb);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return error;
}
vp = nd.ni_vp;
mp = vp->v_mount;
KASSERT(nmp == NULL || nmp == mp);
pathbuf_destroy(pb);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*
* Make sure the file system can do vptofh. If the file system
* knows the handle's size, just trust it's able to do the
* actual translation also (otherwise we should check fhtovp
* also, and that's getting a wee bit ridiculous).
*/
fid_size = 0;
if ((error = VFS_VPTOFH(vp, NULL, &fid_size)) != E2BIG) {
2007-02-05 19:03:53 +03:00
vput(vp);
return EOPNOTSUPP;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
/* Mark the file system busy. */
error = vfs_busy(mp, NULL);
vput(vp);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
if (error != 0)
return error;
if (nmp == NULL)
mutex_enter(&mp->mnt_updating); /* mnt_flag */
netexport_wrlock();
ne = netexport_lookup(mp);
if (ne == NULL) {
error = init_exports(mp, &ne);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
if (error != 0) {
goto out;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
}
KASSERT(ne != NULL);
KASSERT(ne->ne_mount == mp);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*
* XXX: The part marked as 'notyet' works fine from the kernel's
* point of view, in the sense that it is able to atomically update
* the complete exports list for a file system. However, supporting
* this in mountd(8) requires a lot of work; so, for now, keep the
* old behavior of updating a single entry per call.
*
* When mountd(8) is fixed, just remove the second branch of this
* preprocessor conditional and enable the first one.
*/
#ifdef notyet
netexport_clear(ne);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
for (i = 0; error == 0 && i < mel->mel_nexports; i++)
error = export(ne, &mel->mel_exports[i]);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
#else
if (mel->mel_nexports == 0)
netexport_clear(ne);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
else if (mel->mel_nexports == 1)
error = export(ne, &mel->mel_exports[0]);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
else {
printf("%s: Cannot set more than one "
"entry at once (unimplemented)\n", __func__);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
error = EOPNOTSUPP;
}
#endif
out:
netexport_wrunlock();
if (nmp == NULL)
mutex_exit(&mp->mnt_updating); /* mnt_flag */
vfs_unbusy(mp, false, NULL);
return error;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
static void
netexport_insert(struct netexport *ne)
{
CIRCLEQ_INSERT_HEAD(&netexport_list, ne, ne_list);
}
static void
netexport_remove(struct netexport *ne)
{
CIRCLEQ_REMOVE(&netexport_list, ne, ne_list);
}
static struct netexport *
netexport_lookup(const struct mount *mp)
{
struct netexport *ne;
CIRCLEQ_FOREACH(ne, &netexport_list, ne_list) {
if (ne->ne_mount == mp) {
goto done;
}
}
ne = NULL;
done:
return ne;
}
static struct netexport *
netexport_lookup_byfsid(const fsid_t *fsid)
{
struct netexport *ne;
CIRCLEQ_FOREACH(ne, &netexport_list, ne_list) {
const struct mount *mp = ne->ne_mount;
if (mp->mnt_stat.f_fsidx.__fsid_val[0] == fsid->__fsid_val[0] &&
mp->mnt_stat.f_fsidx.__fsid_val[1] == fsid->__fsid_val[1]) {
goto done;
}
}
ne = NULL;
done:
return ne;
}
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*
* Check if the file system specified by the 'mp' mount structure is
* exported to a client with 'anon' anonymous credentials. The 'mb'
* argument is an mbuf containing the network address of the client.
* The return parameters for the export flags for the client are returned
* in the address specified by 'wh'.
*
* This function is used exclusively by the NFS server. It is generally
* invoked before VFS_FHTOVP to validate that client has access to the
* file system.
*/
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
int
netexport_check(const fsid_t *fsid, struct mbuf *mb, struct mount **mpp,
2006-05-15 01:31:52 +04:00
int *wh, kauth_cred_t *anon)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
{
struct netexport *ne;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
struct netcred *np;
ne = netexport_lookup_byfsid(fsid);
if (ne == NULL) {
return EACCES;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
np = netcred_lookup(ne, mb);
if (np == NULL) {
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return EACCES;
}
*mpp = ne->ne_mount;
*wh = np->netc_exflags;
2006-05-15 01:31:52 +04:00
*anon = np->netc_anon;
return 0;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
/*
* Handles legacy export requests. In this case, the export information
* is hardcoded in a specific place of the mount arguments structure (given
* in data); the request for an update is given through the fspec field
* (also in a known location), which must be a null pointer.
*
* Returns EJUSTRETURN if the given command was not a export request.
* Otherwise, returns 0 on success or an appropriate error code otherwise.
*/
static int
nfs_export_update_30(struct mount *mp, const char *path, void *data)
{
struct mountd_exports_list mel;
struct mnt_export_args30 *args;
args = data;
mel.mel_path = path;
if (args->fspec != NULL)
return EJUSTRETURN;
if (args->eargs.ex_flags & 0x00020000) {
/* Request to delete exports. The mask above holds the
* value that used to be in MNT_DELEXPORT. */
mel.mel_nexports = 0;
} else {
/* The following assumes export_args has not changed since
* export_args30 - typedef checks sizes. */
typedef char x[sizeof args->eargs == sizeof *mel.mel_exports ? 1 : -1];
mel.mel_nexports = 1;
mel.mel_exports = (void *)&args->eargs;
}
return mountd_set_exports_list(&mel, curlwp, mp);
}
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*
* INTERNAL FUNCTIONS
*/
/*
* Initializes NFS exports for the mountpoint given in 'mp'.
* If successful, returns 0 and sets *nep to the address of the new
* netexport item; otherwise returns an appropriate error code
* and *nep remains unmodified.
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
*/
static int
init_exports(struct mount *mp, struct netexport **nep)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
{
int error;
struct export_args ea;
struct netexport *ne;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
KASSERT(mp != NULL);
/* Ensure that we do not already have this mount point. */
KASSERT(netexport_lookup(mp) == NULL);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
ne = kmem_zalloc(sizeof(*ne), KM_SLEEP);
ne->ne_mount = mp;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/* Set the default export entry. Handled internally by export upon
* first call. */
memset(&ea, 0, sizeof(ea));
ea.ex_root = -2;
if (mp->mnt_flag & MNT_RDONLY)
ea.ex_flags |= MNT_EXRDONLY;
error = export(ne, &ea);
if (error != 0) {
kmem_free(ne, sizeof(*ne));
} else {
netexport_insert(ne);
*nep = ne;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
return error;
}
/*
* Build hash lists of net addresses and hang them off the mount point.
* Called by export() to set up a new entry in the lists of export
* addresses.
*/
static int
hang_addrlist(struct mount *mp, struct netexport *nep,
const struct export_args *argp)
{
int error, i;
struct netcred *np, *enp;
struct radix_node_head *rnh;
struct sockaddr *saddr, *smask;
struct domain *dom;
smask = NULL;
if (argp->ex_addrlen == 0) {
if (mp->mnt_flag & MNT_DEFEXPORTED)
return EPERM;
np = &nep->ne_defexported;
KASSERT(np->netc_anon == NULL);
np->netc_anon = kauth_cred_alloc();
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
np->netc_exflags = argp->ex_flags;
kauth_uucred_to_cred(np->netc_anon, &argp->ex_anon);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
mp->mnt_flag |= MNT_DEFEXPORTED;
return 0;
}
if (argp->ex_addrlen > MLEN || argp->ex_masklen > MLEN)
return EINVAL;
i = sizeof(struct netcred) + argp->ex_addrlen + argp->ex_masklen;
np = malloc(i, M_NETADDR, M_WAITOK | M_ZERO);
np->netc_anon = kauth_cred_alloc();
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
saddr = (struct sockaddr *)(np + 1);
error = copyin(argp->ex_addr, saddr, argp->ex_addrlen);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
if (error)
goto out;
if (saddr->sa_len > argp->ex_addrlen)
saddr->sa_len = argp->ex_addrlen;
if (sacheck(saddr) == -1)
return EINVAL;
if (argp->ex_masklen) {
smask = (struct sockaddr *)((char *)saddr + argp->ex_addrlen);
error = copyin(argp->ex_mask, smask, argp->ex_masklen);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
if (error)
goto out;
if (smask->sa_len > argp->ex_masklen)
smask->sa_len = argp->ex_masklen;
if (smask->sa_family != saddr->sa_family)
return EINVAL;
if (sacheck(smask) == -1)
return EINVAL;
}
i = saddr->sa_family;
if ((rnh = nep->ne_rtable[i]) == 0) {
/*
* Seems silly to initialize every AF when most are not
* used, do so on demand here
*/
DOMAIN_FOREACH(dom) {
if (dom->dom_family == i && dom->dom_rtattach) {
dom->dom_rtattach((void **)&nep->ne_rtable[i],
dom->dom_rtoffset);
break;
}
}
if ((rnh = nep->ne_rtable[i]) == 0) {
error = ENOBUFS;
goto out;
}
}
enp = (struct netcred *)(*rnh->rnh_addaddr)(saddr, smask, rnh,
np->netc_rnodes);
if (enp != np) {
if (enp == NULL) {
enp = (struct netcred *)(*rnh->rnh_lookup)(saddr,
smask, rnh);
if (enp == NULL) {
error = EPERM;
goto out;
}
} else
enp->netc_refcnt++;
goto check;
} else
enp->netc_refcnt = 1;
np->netc_exflags = argp->ex_flags;
kauth_uucred_to_cred(np->netc_anon, &argp->ex_anon);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return 0;
check:
if (enp->netc_exflags != argp->ex_flags ||
2006-05-15 01:31:52 +04:00
kauth_cred_uucmp(enp->netc_anon, &argp->ex_anon) != 0)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
error = EPERM;
else
error = 0;
out:
KASSERT(np->netc_anon != NULL);
kauth_cred_free(np->netc_anon);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
free(np, M_NETADDR);
return error;
}
/*
* Ensure that the address stored in 'sa' is valid.
* Returns zero on success, otherwise -1.
*/
static int
sacheck(struct sockaddr *sa)
{
switch (sa->sa_family) {
case AF_INET: {
struct sockaddr_in *sin = (struct sockaddr_in *)sa;
char *p = (char *)sin->sin_zero;
size_t i;
if (sin->sin_len != sizeof(*sin))
return -1;
if (sin->sin_port != 0)
return -1;
for (i = 0; i < sizeof(sin->sin_zero); i++)
if (*p++ != '\0')
return -1;
return 0;
}
case AF_INET6: {
struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sa;
if (sin6->sin6_len != sizeof(*sin6))
return -1;
if (sin6->sin6_port != 0)
return -1;
return 0;
}
default:
return -1;
}
}
/*
* Free the netcred object pointed to by the 'rn' radix node.
* 'w' holds a pointer to the radix tree head.
*/
static int
free_netcred(struct radix_node *rn, void *w)
{
struct radix_node_head *rnh = (struct radix_node_head *)w;
struct netcred *np = (struct netcred *)(void *)rn;
(*rnh->rnh_deladdr)(rn->rn_key, rn->rn_mask, rnh);
if (--(np->netc_refcnt) <= 0) {
KASSERT(np->netc_anon != NULL);
kauth_cred_free(np->netc_anon);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
free(np, M_NETADDR);
}
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return 0;
}
/*
* Clears the exports list for a given file system.
*/
static void
netexport_clear(struct netexport *ne)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
{
struct radix_node_head *rnh;
struct mount *mp = ne->ne_mount;
int i;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
if (mp->mnt_flag & MNT_EXPUBLIC) {
setpublicfs(NULL, NULL, NULL);
mp->mnt_flag &= ~MNT_EXPUBLIC;
}
for (i = 0; i <= AF_MAX; i++) {
if ((rnh = ne->ne_rtable[i]) != NULL) {
rn_walktree(rnh, free_netcred, rnh);
free(rnh, M_RTABLE);
ne->ne_rtable[i] = NULL;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
}
if ((mp->mnt_flag & MNT_DEFEXPORTED) != 0) {
struct netcred *np = &ne->ne_defexported;
KASSERT(np->netc_anon != NULL);
kauth_cred_free(np->netc_anon);
np->netc_anon = NULL;
} else {
KASSERT(ne->ne_defexported.netc_anon == NULL);
}
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
mp->mnt_flag &= ~(MNT_EXPORTED | MNT_DEFEXPORTED);
}
/*
* Add a new export entry (described by an export_args structure) to the
* given file system.
*/
static int
export(struct netexport *nep, const struct export_args *argp)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
{
struct mount *mp = nep->ne_mount;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
int error;
if (argp->ex_flags & MNT_EXPORTED) {
if (argp->ex_flags & MNT_EXPUBLIC) {
if ((error = setpublicfs(mp, nep, argp)) != 0)
return error;
mp->mnt_flag |= MNT_EXPUBLIC;
}
if ((error = hang_addrlist(mp, nep, argp)) != 0)
return error;
mp->mnt_flag |= MNT_EXPORTED;
}
return 0;
}
/*
* Set the publicly exported filesystem (WebNFS). Currently, only
* one public filesystem is possible in the spec (RFC 2054 and 2055)
*/
static int
setpublicfs(struct mount *mp, struct netexport *nep,
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
const struct export_args *argp)
{
char *cp;
int error;
struct vnode *rvp;
size_t fhsize;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
/*
* mp == NULL -> invalidate the current info, the FS is
* no longer exported. May be called from either export
* or unmount, so check if it hasn't already been done.
*/
if (mp == NULL) {
if (nfs_pub.np_valid) {
nfs_pub.np_valid = 0;
if (nfs_pub.np_handle != NULL) {
free(nfs_pub.np_handle, M_TEMP);
nfs_pub.np_handle = NULL;
}
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
if (nfs_pub.np_index != NULL) {
2008-12-17 23:51:31 +03:00
free(nfs_pub.np_index, M_TEMP);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
nfs_pub.np_index = NULL;
}
}
return 0;
}
/*
* Only one allowed at a time.
*/
if (nfs_pub.np_valid != 0 && mp != nfs_pub.np_mount)
return EBUSY;
/*
* Get real filehandle for root of exported FS.
*/
if ((error = VFS_ROOT(mp, &rvp)))
return error;
fhsize = 0;
error = vfs_composefh(rvp, NULL, &fhsize);
if (error != E2BIG)
return error;
nfs_pub.np_handle = malloc(fhsize, M_TEMP, M_NOWAIT);
if (nfs_pub.np_handle == NULL)
error = ENOMEM;
else
error = vfs_composefh(rvp, nfs_pub.np_handle, &fhsize);
if (error)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return error;
vput(rvp);
/*
* If an indexfile was specified, pull it in.
*/
if (argp->ex_indexfile != NULL) {
2008-12-17 23:51:31 +03:00
nfs_pub.np_index = malloc(MAXNAMLEN + 1, M_TEMP, M_WAITOK);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
error = copyinstr(argp->ex_indexfile, nfs_pub.np_index,
MAXNAMLEN, (size_t *)0);
if (!error) {
/*
* Check for illegal filenames.
*/
for (cp = nfs_pub.np_index; *cp; cp++) {
if (*cp == '/') {
error = EINVAL;
break;
}
}
}
if (error) {
2008-12-17 23:51:31 +03:00
free(nfs_pub.np_index, M_TEMP);
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return error;
}
}
nfs_pub.np_mount = mp;
nfs_pub.np_valid = 1;
return 0;
}
/*
* Lookup an export entry in the exports list that matches the address
* stored in 'nam'. If no entry is found, the default one is used instead
* (if available).
*/
static struct netcred *
netcred_lookup(struct netexport *ne, struct mbuf *nam)
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
{
struct netcred *np;
struct radix_node_head *rnh;
struct sockaddr *saddr;
if ((ne->ne_mount->mnt_flag & MNT_EXPORTED) == 0) {
return NULL;
}
/*
* Lookup in the export list first.
*/
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
np = NULL;
if (nam != NULL) {
saddr = mtod(nam, struct sockaddr *);
rnh = ne->ne_rtable[saddr->sa_family];
if (rnh != NULL) {
np = (struct netcred *)
(*rnh->rnh_matchaddr)((void *)saddr,
rnh);
if (np && np->netc_rnodes->rn_flags & RNF_ROOT)
np = NULL;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
}
}
/*
* If no address match, use the default if it exists.
*/
if (np == NULL && ne->ne_mount->mnt_flag & MNT_DEFEXPORTED)
np = &ne->ne_defexported;
Apply the NFS exports list rototill patch: - Remove all NFS related stuff from file system specific code. - Drop the vfs_checkexp hook and generalize it in the new nfs_check_export function, thus removing redundancy from all file systems. - Move all NFS export-related stuff from kern/vfs_subr.c to the new file sys/nfs/nfs_export.c. The former was becoming large and its code is always compiled, regardless of the build options. Using the latter, the code is only compiled in when NFSSERVER is enabled. While doing this, also make some functions in nfs_subs.c conditional to NFSSERVER. - Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a path and a set of export entries. At the moment it can only clear the exports list or append entries, one by one, but it is done in a way that allows setting the whole set of entries atomically in the future (see the comment in mountd_set_exports_list or in doc/TODO). - Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so that it becomes file system agnostic. In fact, all this whole thing was done to remove a 'XXX' block from this utility! - Change the mount*, newfs and fsck* userland utilities to not deal with NFS exports initialization; done internally by the kernel when initializing the NFS support for each file system. - Implement an interface for VFS (called VFS hooks) so that several kernel subsystems can run arbitrary code upon receipt of specific VFS events. At the moment, this only provides support for unmount and is used to destroy NFS exports lists from the file systems being unmounted, though it has room for extension. Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments and advice in the development of this patch.
2005-09-23 16:10:31 +04:00
return np;
}
void
netexport_rdlock(void)
{
rw_enter(&netexport_lock, RW_READER);
}
void
netexport_rdunlock(void)
{
rw_exit(&netexport_lock);
}
static void
netexport_wrlock(void)
{
rw_enter(&netexport_lock, RW_WRITER);
}
static void
netexport_wrunlock(void)
{
rw_exit(&netexport_lock);
}