linux_dsm_epyc7002/fs/cifs/cifs_dfs_ref.c
David Howells cc53ce53c8 Add a dentry op to allow processes to be held during pathwalk transit
Add a dentry op (d_manage) to permit a filesystem to hold a process and make it
sleep when it tries to transit away from one of that filesystem's directories
during a pathwalk.  The operation is keyed off a new dentry flag
(DCACHE_MANAGE_TRANSIT).

The filesystem is allowed to be selective about which processes it holds and
which it permits to continue on or prohibits from transiting from each flagged
directory.  This will allow autofs to hold up client processes whilst letting
its userspace daemon through to maintain the directory or the stuff behind it
or mounted upon it.

The ->d_manage() dentry operation:

	int (*d_manage)(struct path *path, bool mounting_here);

takes a pointer to the directory about to be transited away from and a flag
indicating whether the transit is undertaken by do_add_mount() or
do_move_mount() skipping through a pile of filesystems mounted on a mountpoint.

It should return 0 if successful and to let the process continue on its way;
-EISDIR to prohibit the caller from skipping to overmounted filesystems or
automounting, and to use this directory; or some other error code to return to
the user.

->d_manage() is called with namespace_sem writelocked if mounting_here is true
and no other locks held, so it may sleep.  However, if mounting_here is true,
it may not initiate or wait for a mount or unmount upon the parameter
directory, even if the act is actually performed by userspace.

Within fs/namei.c, follow_managed() is extended to check with d_manage() first
on each managed directory, before transiting away from it or attempting to
automount upon it.

follow_down() is renamed follow_down_one() and should only be used where the
filesystem deliberately intends to avoid management steps (e.g. autofs).

A new follow_down() is added that incorporates the loop done by all other
callers of follow_down() (do_add/move_mount(), autofs and NFSD; whilst AFS, NFS
and CIFS do use it, their use is removed by converting them to use
d_automount()).  The new follow_down() calls d_manage() as appropriate.  It
also takes an extra parameter to indicate if it is being called from mount code
(with namespace_sem writelocked) which it passes to d_manage().  follow_down()
ignores automount points so that it can be used to mount on them.

__follow_mount_rcu() is made to abort rcu-walk mode if it hits a directory with
DCACHE_MANAGE_TRANSIT set on the basis that we're probably going to have to
sleep.  It would be possible to enter d_manage() in rcu-walk mode too, and have
that determine whether to abort or not itself.  That would allow the autofs
daemon to continue on in rcu-walk mode.

Note that DCACHE_MANAGE_TRANSIT on a directory should be cleared when it isn't
required as every tranist from that directory will cause d_manage() to be
invoked.  It can always be set again when necessary.

==========================
WHAT THIS MEANS FOR AUTOFS
==========================

Autofs currently uses the lookup() inode op and the d_revalidate() dentry op to
trigger the automounting of indirect mounts, and both of these can be called
with i_mutex held.

autofs knows that the i_mutex will be held by the caller in lookup(), and so
can drop it before invoking the daemon - but this isn't so for d_revalidate(),
since the lock is only held on _some_ of the code paths that call it.  This
means that autofs can't risk dropping i_mutex from its d_revalidate() function
before it calls the daemon.

The bug could manifest itself as, for example, a process that's trying to
validate an automount dentry that gets made to wait because that dentry is
expired and needs cleaning up:

	mkdir         S ffffffff8014e05a     0 32580  24956
	Call Trace:
	 [<ffffffff885371fd>] :autofs4:autofs4_wait+0x674/0x897
	 [<ffffffff80127f7d>] avc_has_perm+0x46/0x58
	 [<ffffffff8009fdcf>] autoremove_wake_function+0x0/0x2e
	 [<ffffffff88537be6>] :autofs4:autofs4_expire_wait+0x41/0x6b
	 [<ffffffff88535cfc>] :autofs4:autofs4_revalidate+0x91/0x149
	 [<ffffffff80036d96>] __lookup_hash+0xa0/0x12f
	 [<ffffffff80057a2f>] lookup_create+0x46/0x80
	 [<ffffffff800e6e31>] sys_mkdirat+0x56/0xe4

versus the automount daemon which wants to remove that dentry, but can't
because the normal process is holding the i_mutex lock:

	automount     D ffffffff8014e05a     0 32581      1              32561
	Call Trace:
	 [<ffffffff80063c3f>] __mutex_lock_slowpath+0x60/0x9b
	 [<ffffffff8000ccf1>] do_path_lookup+0x2ca/0x2f1
	 [<ffffffff80063c89>] .text.lock.mutex+0xf/0x14
	 [<ffffffff800e6d55>] do_rmdir+0x77/0xde
	 [<ffffffff8005d229>] tracesys+0x71/0xe0
	 [<ffffffff8005d28d>] tracesys+0xd5/0xe0

which means that the system is deadlocked.

This patch allows autofs to hold up normal processes whilst the daemon goes
ahead and does things to the dentry tree behind the automouter point without
risking a deadlock as almost no locks are held in d_manage() and none in
d_automount().

Signed-off-by: David Howells <dhowells@redhat.com>
Was-Acked-by: Ian Kent <raven@themaw.net>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2011-01-15 20:07:31 -05:00

384 lines
9.8 KiB
C

/*
* 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)
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*/
#include <linux/dcache.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/slab.h>
#include <linux/vfs.h>
#include <linux/fs.h>
#include "cifsglob.h"
#include "cifsproto.h"
#include "cifsfs.h"
#include "dns_resolve.h"
#include "cifs_debug.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_get_share_name - extracts share name from UNC
* @node_name: pointer to UNC string
*
* Extracts sharename form full UNC.
* i.e. strips from UNC trailing path that is not part of share
* name and fixup missing '\' in the begining of DFS node refferal
* if necessary.
* Returns pointer to share name on success or ERR_PTR on error.
* Caller is responsible for freeing returned string.
*/
static char *cifs_get_share_name(const char *node_name)
{
int len;
char *UNC;
char *pSep;
len = strlen(node_name);
UNC = kmalloc(len+2 /*for term null and additional \ if it's missed */,
GFP_KERNEL);
if (!UNC)
return ERR_PTR(-ENOMEM);
/* get share name and server name */
if (node_name[1] != '\\') {
UNC[0] = '\\';
strncpy(UNC+1, node_name, len);
len++;
UNC[len] = 0;
} else {
strncpy(UNC, node_name, len);
UNC[len] = 0;
}
/* find server name end */
pSep = memchr(UNC+2, '\\', len-2);
if (!pSep) {
cERROR(1, "%s: no server name end in node name: %s",
__func__, node_name);
kfree(UNC);
return ERR_PTR(-EINVAL);
}
/* find sharename end */
pSep++;
pSep = memchr(UNC+(pSep-UNC), '\\', len-(pSep-UNC));
if (pSep) {
/* trim path up to sharename end
* now we have share name in UNC */
*pSep = 0;
}
return UNC;
}
/**
* cifs_compose_mount_options - creates mount options for refferral
* @sb_mountdata: parent/root DFS mount options (template)
* @fullpath: full path in UNC format
* @ref: server's referral
* @devname: pointer for saving device name
*
* 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 responcible for freeing retunrned 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 *mountdata = NULL;
int md_len;
char *tkn_e;
char *srvIP = NULL;
char sep = ',';
int off, noff;
if (sb_mountdata == NULL)
return ERR_PTR(-EINVAL);
*devname = cifs_get_share_name(ref->node_name);
if (IS_ERR(*devname)) {
rc = PTR_ERR(*devname);
*devname = NULL;
goto compose_mount_options_err;
}
rc = dns_resolve_server_name_to_ip(*devname, &srvIP);
if (rc < 0) {
cERROR(1, "%s: Failed to resolve server part of %s to IP: %d",
__func__, *devname, rc);
goto compose_mount_options_err;
}
/* md_len = strlen(...) + 12 for 'sep+prefixpath='
* assuming that we have 'unc=' and 'ip=' in
* the original sb_mountdata
*/
md_len = strlen(sb_mountdata) + rc + strlen(ref->node_name) + 12;
mountdata = kzalloc(md_len+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 (strnicmp(sb_mountdata + off, "unc=", 4) == 0) {
off += noff;
continue;
}
if (strnicmp(sb_mountdata + off, "ip=", 3) == 0) {
off += noff;
continue;
}
if (strnicmp(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);
strncat(mountdata, &sep, 1);
strcat(mountdata, "unc=");
strcat(mountdata, *devname);
/* find & copy prefixpath */
tkn_e = strchr(ref->node_name + 2, '\\');
if (tkn_e == NULL) {
/* invalid unc, missing share name*/
rc = -EINVAL;
goto compose_mount_options_err;
}
tkn_e = strchr(tkn_e + 1, '\\');
if (tkn_e || (strlen(fullpath) - ref->path_consumed)) {
strncat(mountdata, &sep, 1);
strcat(mountdata, "prefixpath=");
if (tkn_e)
strcat(mountdata, tkn_e + 1);
strcat(mountdata, fullpath + ref->path_consumed);
}
/*cFYI(1, "%s: parent mountdata: %s", __func__,sb_mountdata);*/
/*cFYI(1, "%s: submount mountdata: %s", __func__, mountdata );*/
compose_mount_options_out:
kfree(srvIP);
return mountdata;
compose_mount_options_err:
kfree(mountdata);
mountdata = ERR_PTR(rc);
goto compose_mount_options_out;
}
/**
* cifs_dfs_do_refmount - mounts specified path using provided refferal
* @cifs_sb: parent/root superblock
* @fullpath: full path in UNC format
* @ref: server's referral
*/
static struct vfsmount *cifs_dfs_do_refmount(struct cifs_sb_info *cifs_sb,
const char *fullpath, const struct dfs_info3_param *ref)
{
struct vfsmount *mnt;
char *mountdata;
char *devname = NULL;
/* strip first '\' from fullpath */
mountdata = cifs_compose_mount_options(cifs_sb->mountdata,
fullpath + 1, ref, &devname);
if (IS_ERR(mountdata))
return (struct vfsmount *)mountdata;
mnt = vfs_kern_mount(&cifs_fs_type, 0, devname, mountdata);
kfree(mountdata);
kfree(devname);
return mnt;
}
static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd,
struct list_head *mntlist)
{
/* stolen from afs code */
int err;
mntget(newmnt);
err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags | MNT_SHRINKABLE, mntlist);
switch (err) {
case 0:
path_put(&nd->path);
nd->path.mnt = newmnt;
nd->path.dentry = dget(newmnt->mnt_root);
schedule_delayed_work(&cifs_dfs_automount_task,
cifs_dfs_mountpoint_expiry_timeout);
break;
case -EBUSY:
/* someone else made a mount here whilst we were busy */
err = follow_down(&nd->path, false);
default:
mntput(newmnt);
break;
}
return err;
}
static void dump_referral(const struct dfs_info3_param *ref)
{
cFYI(1, "DFS: ref path: %s", ref->path_name);
cFYI(1, "DFS: node path: %s", ref->node_name);
cFYI(1, "DFS: fl: %hd, srv_type: %hd", ref->flags, ref->server_type);
cFYI(1, "DFS: ref_flags: %hd, path_consumed: %hd", ref->ref_flag,
ref->path_consumed);
}
static void*
cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
{
struct dfs_info3_param *referrals = NULL;
unsigned int num_referrals = 0;
struct cifs_sb_info *cifs_sb;
struct cifsSesInfo *ses;
char *full_path = NULL;
int xid, i;
int rc = 0;
struct vfsmount *mnt = ERR_PTR(-ENOENT);
struct tcon_link *tlink;
cFYI(1, "in %s", __func__);
BUG_ON(IS_ROOT(dentry));
xid = GetXid();
dput(nd->path.dentry);
nd->path.dentry = dget(dentry);
/*
* 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.
*/
full_path = build_path_from_dentry(dentry);
if (full_path == NULL) {
rc = -ENOMEM;
goto out_err;
}
cifs_sb = CIFS_SB(dentry->d_inode->i_sb);
tlink = cifs_sb_tlink(cifs_sb);
if (IS_ERR(tlink)) {
rc = PTR_ERR(tlink);
goto out_err;
}
ses = tlink_tcon(tlink)->ses;
rc = get_dfs_path(xid, ses, full_path + 1, cifs_sb->local_nls,
&num_referrals, &referrals,
cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
cifs_put_tlink(tlink);
for (i = 0; i < num_referrals; i++) {
int len;
dump_referral(referrals+i);
/* connect to a node */
len = strlen(referrals[i].node_name);
if (len < 2) {
cERROR(1, "%s: Net Address path too short: %s",
__func__, referrals[i].node_name);
rc = -EINVAL;
goto out_err;
}
mnt = cifs_dfs_do_refmount(cifs_sb,
full_path, referrals + i);
cFYI(1, "%s: cifs_dfs_do_refmount:%s , mnt:%p", __func__,
referrals[i].node_name, mnt);
/* complete mount procedure if we accured submount */
if (!IS_ERR(mnt))
break;
}
/* we need it cause for() above could exit without valid submount */
rc = PTR_ERR(mnt);
if (IS_ERR(mnt))
goto out_err;
rc = add_mount_helper(mnt, nd, &cifs_dfs_automount_list);
out:
FreeXid(xid);
free_dfs_info_array(referrals, num_referrals);
kfree(full_path);
cFYI(1, "leaving %s" , __func__);
return ERR_PTR(rc);
out_err:
path_put(&nd->path);
goto out;
}
const struct inode_operations cifs_dfs_referral_inode_operations = {
.follow_link = cifs_dfs_follow_mountpoint,
};