linux/fs/cifs/cifs_dfs_ref.c

375 lines
9.6 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/*
* Contains the CIFS DFS referral mounting routines used for handling
* traversal via DFS junction point
*
* Copyright (c) 2007 Igor Mammedov
* Copyright (C) International Business Machines Corp., 2008
* Author(s): Igor Mammedov (niallain@gmail.com)
* Steve French (sfrench@us.ibm.com)
*/
#include <linux/dcache.h>
#include <linux/mount.h>
#include <linux/namei.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <linux/vfs.h>
#include <linux/fs.h>
#include <linux/inet.h>
#include "cifsglob.h"
#include "cifsproto.h"
#include "cifsfs.h"
#include "dns_resolve.h"
#include "cifs_debug.h"
#include "cifs_unicode.h"
#include "dfs_cache.h"
#include "fs_context.h"
static LIST_HEAD(cifs_dfs_automount_list);
static void cifs_dfs_expire_automounts(struct work_struct *work);
static DECLARE_DELAYED_WORK(cifs_dfs_automount_task,
cifs_dfs_expire_automounts);
static int cifs_dfs_mountpoint_expiry_timeout = 500 * HZ;
static void cifs_dfs_expire_automounts(struct work_struct *work)
{
struct list_head *list = &cifs_dfs_automount_list;
mark_mounts_for_expiry(list);
if (!list_empty(list))
schedule_delayed_work(&cifs_dfs_automount_task,
cifs_dfs_mountpoint_expiry_timeout);
}
void cifs_dfs_release_automount_timer(void)
{
BUG_ON(!list_empty(&cifs_dfs_automount_list));
cancel_delayed_work_sync(&cifs_dfs_automount_task);
}
/**
* cifs_build_devname - build a devicename from a UNC and optional prepath
* @nodename: pointer to UNC string
* @prepath: pointer to prefixpath (or NULL if there isn't one)
*
* Build a new cifs devicename after chasing a DFS referral. Allocate a buffer
* big enough to hold the final thing. Copy the UNC from the nodename, and
* concatenate the prepath onto the end of it if there is one.
*
* Returns pointer to the built string, or a ERR_PTR. Caller is responsible
* for freeing the returned string.
*/
static char *
cifs_build_devname(char *nodename, const char *prepath)
{
size_t pplen;
size_t unclen;
char *dev;
char *pos;
/* skip over any preceding delimiters */
nodename += strspn(nodename, "\\");
if (!*nodename)
return ERR_PTR(-EINVAL);
/* get length of UNC and set pos to last char */
unclen = strlen(nodename);
pos = nodename + unclen - 1;
/* trim off any trailing delimiters */
while (*pos == '\\') {
--pos;
--unclen;
}
/* allocate a buffer:
* +2 for preceding "//"
* +1 for delimiter between UNC and prepath
* +1 for trailing NULL
*/
pplen = prepath ? strlen(prepath) : 0;
dev = kmalloc(2 + unclen + 1 + pplen + 1, GFP_KERNEL);
if (!dev)
return ERR_PTR(-ENOMEM);
pos = dev;
/* add the initial "//" */
*pos = '/';
++pos;
*pos = '/';
++pos;
/* copy in the UNC portion from referral */
memcpy(pos, nodename, unclen);
pos += unclen;
/* copy the prefixpath remainder (if there is one) */
if (pplen) {
*pos = '/';
++pos;
memcpy(pos, prepath, pplen);
pos += pplen;
}
/* NULL terminator */
*pos = '\0';
convert_delimiter(dev, '/');
return dev;
}
/**
* cifs_compose_mount_options - creates mount options for referral
* @sb_mountdata: parent/root DFS mount options (template)
* @fullpath: full path in UNC format
* @ref: optional server's referral
* @devname: return the built cifs device name if passed pointer not NULL
* creates mount options for submount based on template options sb_mountdata
* and replacing unc,ip,prefixpath options with ones we've got form ref_unc.
*
* Returns: pointer to new mount options or ERR_PTR.
* Caller is responsible for freeing returned value if it is not error.
*/
char *cifs_compose_mount_options(const char *sb_mountdata,
const char *fullpath,
const struct dfs_info3_param *ref,
char **devname)
{
int rc;
char *name;
char *mountdata = NULL;
const char *prepath = NULL;
int md_len;
char *tkn_e;
char *srvIP = NULL;
char sep = ',';
int off, noff;
if (sb_mountdata == NULL)
return ERR_PTR(-EINVAL);
if (ref) {
if (WARN_ON_ONCE(!ref->node_name || ref->path_consumed < 0))
return ERR_PTR(-EINVAL);
if (strlen(fullpath) - ref->path_consumed) {
prepath = fullpath + ref->path_consumed;
/* skip initial delimiter */
if (*prepath == '/' || *prepath == '\\')
prepath++;
}
name = cifs_build_devname(ref->node_name, prepath);
if (IS_ERR(name)) {
rc = PTR_ERR(name);
name = NULL;
goto compose_mount_options_err;
}
} else {
name = cifs_build_devname((char *)fullpath, NULL);
if (IS_ERR(name)) {
rc = PTR_ERR(name);
name = NULL;
goto compose_mount_options_err;
}
}
rc = dns_resolve_server_name_to_ip(name, &srvIP, NULL);
if (rc < 0) {
cifs_dbg(FYI, "%s: Failed to resolve server part of %s to IP: %d\n",
__func__, name, rc);
goto compose_mount_options_err;
}
/*
* In most cases, we'll be building a shorter string than the original,
* but we do have to assume that the address in the ip= option may be
* much longer than the original. Add the max length of an address
* string to the length of the original string to allow for worst case.
*/
md_len = strlen(sb_mountdata) + INET6_ADDRSTRLEN;
mountdata = kzalloc(md_len + sizeof("ip=") + 1, GFP_KERNEL);
if (mountdata == NULL) {
rc = -ENOMEM;
goto compose_mount_options_err;
}
/* copy all options except of unc,ip,prefixpath */
off = 0;
if (strncmp(sb_mountdata, "sep=", 4) == 0) {
sep = sb_mountdata[4];
strncpy(mountdata, sb_mountdata, 5);
off += 5;
}
do {
tkn_e = strchr(sb_mountdata + off, sep);
if (tkn_e == NULL)
noff = strlen(sb_mountdata + off);
else
noff = tkn_e - (sb_mountdata + off) + 1;
if (strncasecmp(sb_mountdata + off, "cruid=", 6) == 0) {
off += noff;
continue;
}
if (strncasecmp(sb_mountdata + off, "unc=", 4) == 0) {
off += noff;
continue;
}
if (strncasecmp(sb_mountdata + off, "ip=", 3) == 0) {
off += noff;
continue;
}
if (strncasecmp(sb_mountdata + off, "prefixpath=", 11) == 0) {
off += noff;
continue;
}
strncat(mountdata, sb_mountdata + off, noff);
off += noff;
} while (tkn_e);
strcat(mountdata, sb_mountdata + off);
mountdata[md_len] = '\0';
/* copy new IP and ref share name */
if (mountdata[strlen(mountdata) - 1] != sep)
strncat(mountdata, &sep, 1);
strcat(mountdata, "ip=");
strcat(mountdata, srvIP);
if (devname)
*devname = name;
else
kfree(name);
/*cifs_dbg(FYI, "%s: parent mountdata: %s\n", __func__, sb_mountdata);*/
/*cifs_dbg(FYI, "%s: submount mountdata: %s\n", __func__, mountdata );*/
compose_mount_options_out:
kfree(srvIP);
return mountdata;
compose_mount_options_err:
kfree(mountdata);
mountdata = ERR_PTR(rc);
kfree(name);
goto compose_mount_options_out;
}
/**
* cifs_dfs_do_mount - mounts specified path using DFS full path
*
* Always pass down @fullpath to smb3_do_mount() so we can use the root server
* to perform failover in case we failed to connect to the first target in the
* referral.
*
* @mntpt: directory entry for the path we are trying to automount
* @cifs_sb: parent/root superblock
* @fullpath: full path in UNC format
*/
static struct vfsmount *cifs_dfs_do_mount(struct dentry *mntpt,
struct cifs_sb_info *cifs_sb,
const char *fullpath)
{
struct vfsmount *mnt;
char *mountdata;
char *devname;
devname = kstrdup(fullpath, GFP_KERNEL);
if (!devname)
return ERR_PTR(-ENOMEM);
convert_delimiter(devname, '/');
/* TODO: change to call fs_context_for_mount(), fill in context directly, call fc_mount */
/* See afs_mntpt_do_automount in fs/afs/mntpt.c for an example */
/* strip first '\' from fullpath */
mountdata = cifs_compose_mount_options(cifs_sb->ctx->mount_options,
fullpath + 1, NULL, NULL);
if (IS_ERR(mountdata)) {
kfree(devname);
return (struct vfsmount *)mountdata;
}
fs: Better permission checking for submounts To support unprivileged users mounting filesystems two permission checks have to be performed: a test to see if the user allowed to create a mount in the mount namespace, and a test to see if the user is allowed to access the specified filesystem. The automount case is special in that mounting the original filesystem grants permission to mount the sub-filesystems, to any user who happens to stumble across the their mountpoint and satisfies the ordinary filesystem permission checks. Attempting to handle the automount case by using override_creds almost works. It preserves the idea that permission to mount the original filesystem is permission to mount the sub-filesystem. Unfortunately using override_creds messes up the filesystems ordinary permission checks. Solve this by being explicit that a mount is a submount by introducing vfs_submount, and using it where appropriate. vfs_submount uses a new mount internal mount flags MS_SUBMOUNT, to let sget and friends know that a mount is a submount so they can take appropriate action. sget and sget_userns are modified to not perform any permission checks on submounts. follow_automount is modified to stop using override_creds as that has proven problemantic. do_mount is modified to always remove the new MS_SUBMOUNT flag so that we know userspace will never by able to specify it. autofs4 is modified to stop using current_real_cred that was put in there to handle the previous version of submount permission checking. cifs is modified to pass the mountpoint all of the way down to vfs_submount. debugfs is modified to pass the mountpoint all of the way down to trace_automount by adding a new parameter. To make this change easier a new typedef debugfs_automount_t is introduced to capture the type of the debugfs automount function. Cc: stable@vger.kernel.org Fixes: 069d5ac9ae0d ("autofs: Fix automounts by using current_real_cred()->uid") Fixes: aeaa4a79ff6a ("fs: Call d_automount with the filesystems creds") Reviewed-by: Trond Myklebust <trond.myklebust@primarydata.com> Reviewed-by: Seth Forshee <seth.forshee@canonical.com> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
2017-01-31 17:06:16 +00:00
mnt = vfs_submount(mntpt, &cifs_fs_type, devname, mountdata);
kfree(mountdata);
kfree(devname);
return mnt;
}
/*
* Create a vfsmount that we can automount
*/
static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt)
{
struct cifs_sb_info *cifs_sb;
cifs: allocate buffer in the caller of build_path_from_dentry() build_path_from_dentry() open-codes dentry_path_raw(). The reason we can't use dentry_path_raw() in there (and postprocess the result as needed) is that the callers of build_path_from_dentry() expect that the object to be freed on cleanup and the string to be used are at the same address. That's painful, since the path is naturally built end-to-beginning - we start at the leaf and go through the ancestors, accumulating the pathname. Life would be easier if we left the buffer allocation to callers. It wouldn't be exact-sized buffer, but none of the callers keep the result for long - it's always freed before the caller returns. So there's no need to do exact-sized allocation; better use __getname()/__putname(), same as we do for pathname arguments of syscalls. What's more, there's no need to do allocation under spinlocks, so GFP_ATOMIC is not needed. Next patch will replace the open-coded dentry_path_raw() (in build_path_from_dentry_optional_prefix()) with calling the real thing. This patch only introduces wrappers for allocating/freeing the buffers and switches to new calling conventions: build_path_from_dentry(dentry, buf) expects buf to be address of a page-sized object or NULL, return value is a pathname built inside that buffer on success, ERR_PTR(-ENOMEM) if buf is NULL and ERR_PTR(-ENAMETOOLONG) if the pathname won't fit into page. Note that we don't need to check for failure when allocating the buffer in the caller - build_path_from_dentry() will do the right thing. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-05 22:36:04 +00:00
void *page;
char *full_path;
struct vfsmount *mnt;
cifs_dbg(FYI, "in %s\n", __func__);
BUG_ON(IS_ROOT(mntpt));
/*
* The MSDFS spec states that paths in DFS referral requests and
* responses must be prefixed by a single '\' character instead of
* the double backslashes usually used in the UNC. This function
* gives us the latter, so we must adjust the result.
*/
cifs_sb = CIFS_SB(mntpt->d_sb);
if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS) {
mnt = ERR_PTR(-EREMOTE);
goto cdda_exit;
}
cifs: allocate buffer in the caller of build_path_from_dentry() build_path_from_dentry() open-codes dentry_path_raw(). The reason we can't use dentry_path_raw() in there (and postprocess the result as needed) is that the callers of build_path_from_dentry() expect that the object to be freed on cleanup and the string to be used are at the same address. That's painful, since the path is naturally built end-to-beginning - we start at the leaf and go through the ancestors, accumulating the pathname. Life would be easier if we left the buffer allocation to callers. It wouldn't be exact-sized buffer, but none of the callers keep the result for long - it's always freed before the caller returns. So there's no need to do exact-sized allocation; better use __getname()/__putname(), same as we do for pathname arguments of syscalls. What's more, there's no need to do allocation under spinlocks, so GFP_ATOMIC is not needed. Next patch will replace the open-coded dentry_path_raw() (in build_path_from_dentry_optional_prefix()) with calling the real thing. This patch only introduces wrappers for allocating/freeing the buffers and switches to new calling conventions: build_path_from_dentry(dentry, buf) expects buf to be address of a page-sized object or NULL, return value is a pathname built inside that buffer on success, ERR_PTR(-ENOMEM) if buf is NULL and ERR_PTR(-ENAMETOOLONG) if the pathname won't fit into page. Note that we don't need to check for failure when allocating the buffer in the caller - build_path_from_dentry() will do the right thing. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-05 22:36:04 +00:00
page = alloc_dentry_path();
/* always use tree name prefix */
cifs: allocate buffer in the caller of build_path_from_dentry() build_path_from_dentry() open-codes dentry_path_raw(). The reason we can't use dentry_path_raw() in there (and postprocess the result as needed) is that the callers of build_path_from_dentry() expect that the object to be freed on cleanup and the string to be used are at the same address. That's painful, since the path is naturally built end-to-beginning - we start at the leaf and go through the ancestors, accumulating the pathname. Life would be easier if we left the buffer allocation to callers. It wouldn't be exact-sized buffer, but none of the callers keep the result for long - it's always freed before the caller returns. So there's no need to do exact-sized allocation; better use __getname()/__putname(), same as we do for pathname arguments of syscalls. What's more, there's no need to do allocation under spinlocks, so GFP_ATOMIC is not needed. Next patch will replace the open-coded dentry_path_raw() (in build_path_from_dentry_optional_prefix()) with calling the real thing. This patch only introduces wrappers for allocating/freeing the buffers and switches to new calling conventions: build_path_from_dentry(dentry, buf) expects buf to be address of a page-sized object or NULL, return value is a pathname built inside that buffer on success, ERR_PTR(-ENOMEM) if buf is NULL and ERR_PTR(-ENAMETOOLONG) if the pathname won't fit into page. Note that we don't need to check for failure when allocating the buffer in the caller - build_path_from_dentry() will do the right thing. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-05 22:36:04 +00:00
full_path = build_path_from_dentry_optional_prefix(mntpt, page, true);
if (IS_ERR(full_path)) {
mnt = ERR_CAST(full_path);
goto free_full_path;
}
convert_delimiter(full_path, '\\');
cifs_dbg(FYI, "%s: full_path: %s\n", __func__, full_path);
mnt = cifs_dfs_do_mount(mntpt, cifs_sb, full_path);
cifs_dbg(FYI, "%s: cifs_dfs_do_mount:%s , mnt:%p\n", __func__, full_path + 1, mnt);
free_full_path:
cifs: allocate buffer in the caller of build_path_from_dentry() build_path_from_dentry() open-codes dentry_path_raw(). The reason we can't use dentry_path_raw() in there (and postprocess the result as needed) is that the callers of build_path_from_dentry() expect that the object to be freed on cleanup and the string to be used are at the same address. That's painful, since the path is naturally built end-to-beginning - we start at the leaf and go through the ancestors, accumulating the pathname. Life would be easier if we left the buffer allocation to callers. It wouldn't be exact-sized buffer, but none of the callers keep the result for long - it's always freed before the caller returns. So there's no need to do exact-sized allocation; better use __getname()/__putname(), same as we do for pathname arguments of syscalls. What's more, there's no need to do allocation under spinlocks, so GFP_ATOMIC is not needed. Next patch will replace the open-coded dentry_path_raw() (in build_path_from_dentry_optional_prefix()) with calling the real thing. This patch only introduces wrappers for allocating/freeing the buffers and switches to new calling conventions: build_path_from_dentry(dentry, buf) expects buf to be address of a page-sized object or NULL, return value is a pathname built inside that buffer on success, ERR_PTR(-ENOMEM) if buf is NULL and ERR_PTR(-ENAMETOOLONG) if the pathname won't fit into page. Note that we don't need to check for failure when allocating the buffer in the caller - build_path_from_dentry() will do the right thing. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-05 22:36:04 +00:00
free_dentry_path(page);
cdda_exit:
cifs_dbg(FYI, "leaving %s\n" , __func__);
return mnt;
}
/*
* Attempt to automount the referral
*/
struct vfsmount *cifs_dfs_d_automount(struct path *path)
{
struct vfsmount *newmnt;
cifs_dbg(FYI, "in %s\n", __func__);
newmnt = cifs_dfs_do_automount(path->dentry);
if (IS_ERR(newmnt)) {
cifs_dbg(FYI, "leaving %s [automount failed]\n" , __func__);
return newmnt;
}
mntget(newmnt); /* prevent immediate expiration */
mnt_set_expiry(newmnt, &cifs_dfs_automount_list);
schedule_delayed_work(&cifs_dfs_automount_task,
cifs_dfs_mountpoint_expiry_timeout);
cifs_dbg(FYI, "leaving %s [ok]\n" , __func__);
return newmnt;
}
const struct inode_operations cifs_dfs_referral_inode_operations = {
};