linux/fs/posix_acl.c

956 lines
22 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2002,2003 by Andreas Gruenbacher <a.gruenbacher@computer.org>
*
* Fixes from William Schumacher incorporated on 15 March 2001.
* (Reported by Charles Bertsch, <CBertsch@microtest.com>).
*/
/*
* This file contains generic functions for manipulating
* POSIX 1003.1e draft standard 17 ACLs.
*/
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/atomic.h>
#include <linux/fs.h>
#include <linux/sched.h>
#include <linux/cred.h>
#include <linux/posix_acl.h>
#include <linux/posix_acl_xattr.h>
#include <linux/xattr.h>
#include <linux/export.h>
#include <linux/user_namespace.h>
static struct posix_acl **acl_by_type(struct inode *inode, int type)
{
switch (type) {
case ACL_TYPE_ACCESS:
return &inode->i_acl;
case ACL_TYPE_DEFAULT:
return &inode->i_default_acl;
default:
BUG();
}
}
struct posix_acl *get_cached_acl(struct inode *inode, int type)
{
struct posix_acl **p = acl_by_type(inode, type);
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
struct posix_acl *acl;
for (;;) {
rcu_read_lock();
acl = rcu_dereference(*p);
if (!acl || is_uncached_acl(acl) ||
posix_acl: convert posix_acl.a_refcount from atomic_t to refcount_t atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable posix_acl.a_refcount is used as pure reference counter. Convert it to refcount_t and fix up the operations. **Important note for maintainers: Some functions from refcount_t API defined in lib/refcount.c have different memory ordering guarantees than their atomic counterparts. The full comparison can be seen in https://lkml.org/lkml/2017/11/15/57 and it is hopefully soon in state to be merged to the documentation tree. Normally the differences should not matter since refcount_t provides enough guarantees to satisfy the refcounting use cases, but in some rare cases it might matter. Please double check that you don't have some undocumented memory guarantees for this variable usage. For the posix_acl.a_refcount it might make a difference in following places: - get_cached_acl(): increment in refcount_inc_not_zero() only guarantees control dependency on success vs. fully ordered atomic counterpart. However this operation is performed under rcu_read_lock(), so this should be fine. - posix_acl_release(): decrement in refcount_dec_and_test() only provides RELEASE ordering and control dependency on success vs. fully ordered atomic counterpart Suggested-by: Kees Cook <keescook@chromium.org> Reviewed-by: David Windsor <dwindsor@gmail.com> Reviewed-by: Hans Liljestrand <ishkamiel@gmail.com> Signed-off-by: Elena Reshetova <elena.reshetova@intel.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-11-29 19:19:31 +08:00
refcount_inc_not_zero(&acl->a_refcount))
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
break;
rcu_read_unlock();
cpu_relax();
}
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
rcu_read_unlock();
return acl;
}
EXPORT_SYMBOL(get_cached_acl);
struct posix_acl *get_cached_acl_rcu(struct inode *inode, int type)
{
return rcu_dereference(*acl_by_type(inode, type));
}
EXPORT_SYMBOL(get_cached_acl_rcu);
void set_cached_acl(struct inode *inode, int type, struct posix_acl *acl)
{
struct posix_acl **p = acl_by_type(inode, type);
struct posix_acl *old;
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
old = xchg(p, posix_acl_dup(acl));
if (!is_uncached_acl(old))
posix_acl_release(old);
}
EXPORT_SYMBOL(set_cached_acl);
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
static void __forget_cached_acl(struct posix_acl **p)
{
struct posix_acl *old;
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
old = xchg(p, ACL_NOT_CACHED);
if (!is_uncached_acl(old))
posix_acl_release(old);
}
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
void forget_cached_acl(struct inode *inode, int type)
{
__forget_cached_acl(acl_by_type(inode, type));
}
EXPORT_SYMBOL(forget_cached_acl);
void forget_all_cached_acls(struct inode *inode)
{
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
__forget_cached_acl(&inode->i_acl);
__forget_cached_acl(&inode->i_default_acl);
}
EXPORT_SYMBOL(forget_all_cached_acls);
struct posix_acl *get_acl(struct inode *inode, int type)
{
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
void *sentinel;
struct posix_acl **p;
struct posix_acl *acl;
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
/*
* The sentinel is used to detect when another operation like
* set_cached_acl() or forget_cached_acl() races with get_acl().
* It is guaranteed that is_uncached_acl(sentinel) is true.
*/
acl = get_cached_acl(inode, type);
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
if (!is_uncached_acl(acl))
return acl;
if (!IS_POSIXACL(inode))
return NULL;
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
sentinel = uncached_acl_sentinel(current);
p = acl_by_type(inode, type);
/*
* If the ACL isn't being read yet, set our sentinel. Otherwise, the
* current value of the ACL will not be ACL_NOT_CACHED and so our own
* sentinel will not be set; another task will update the cache. We
* could wait for that other task to complete its job, but it's easier
* to just call ->get_acl to fetch the ACL ourself. (This is going to
* be an unlikely race.)
*/
if (cmpxchg(p, ACL_NOT_CACHED, sentinel) != ACL_NOT_CACHED)
/* fall through */ ;
/*
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
* Normally, the ACL returned by ->get_acl will be cached.
* A filesystem can prevent that by calling
* forget_cached_acl(inode, type) in ->get_acl.
*
* If the filesystem doesn't have a get_acl() function at all, we'll
* just create the negative cache entry.
*/
if (!inode->i_op->get_acl) {
set_cached_acl(inode, type, NULL);
return NULL;
}
posix_acl: Inode acl caching fixes When get_acl() is called for an inode whose ACL is not cached yet, the get_acl inode operation is called to fetch the ACL from the filesystem. The inode operation is responsible for updating the cached acl with set_cached_acl(). This is done without locking at the VFS level, so another task can call set_cached_acl() or forget_cached_acl() before the get_acl inode operation gets to calling set_cached_acl(), and then get_acl's call to set_cached_acl() results in caching an outdate ACL. Prevent this from happening by setting the cached ACL pointer to a task-specific sentinel value before calling the get_acl inode operation. Move the responsibility for updating the cached ACL from the get_acl inode operations to get_acl(). There, only set the cached ACL if the sentinel value hasn't changed. The sentinel values are chosen to have odd values. Likewise, the value of ACL_NOT_CACHED is odd. In contrast, ACL object pointers always have an even value (ACLs are aligned in memory). This allows to distinguish uncached ACLs values from ACL objects. In addition, switch from guarding inode->i_acl and inode->i_default_acl upates by the inode->i_lock spinlock to using xchg() and cmpxchg(). Filesystems that do not want ACLs returned from their get_acl inode operations to be cached must call forget_cached_acl() to prevent the VFS from doing so. (Patch written by Al Viro and Andreas Gruenbacher.) Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2016-03-24 21:38:37 +08:00
acl = inode->i_op->get_acl(inode, type);
if (IS_ERR(acl)) {
/*
* Remove our sentinel so that we don't block future attempts
* to cache the ACL.
*/
cmpxchg(p, sentinel, ACL_NOT_CACHED);
return acl;
}
/*
* Cache the result, but only if our sentinel is still in place.
*/
posix_acl_dup(acl);
if (unlikely(cmpxchg(p, sentinel, acl) != sentinel))
posix_acl_release(acl);
return acl;
}
EXPORT_SYMBOL(get_acl);
/*
* Init a fresh posix_acl
*/
void
posix_acl_init(struct posix_acl *acl, int count)
{
posix_acl: convert posix_acl.a_refcount from atomic_t to refcount_t atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable posix_acl.a_refcount is used as pure reference counter. Convert it to refcount_t and fix up the operations. **Important note for maintainers: Some functions from refcount_t API defined in lib/refcount.c have different memory ordering guarantees than their atomic counterparts. The full comparison can be seen in https://lkml.org/lkml/2017/11/15/57 and it is hopefully soon in state to be merged to the documentation tree. Normally the differences should not matter since refcount_t provides enough guarantees to satisfy the refcounting use cases, but in some rare cases it might matter. Please double check that you don't have some undocumented memory guarantees for this variable usage. For the posix_acl.a_refcount it might make a difference in following places: - get_cached_acl(): increment in refcount_inc_not_zero() only guarantees control dependency on success vs. fully ordered atomic counterpart. However this operation is performed under rcu_read_lock(), so this should be fine. - posix_acl_release(): decrement in refcount_dec_and_test() only provides RELEASE ordering and control dependency on success vs. fully ordered atomic counterpart Suggested-by: Kees Cook <keescook@chromium.org> Reviewed-by: David Windsor <dwindsor@gmail.com> Reviewed-by: Hans Liljestrand <ishkamiel@gmail.com> Signed-off-by: Elena Reshetova <elena.reshetova@intel.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-11-29 19:19:31 +08:00
refcount_set(&acl->a_refcount, 1);
acl->a_count = count;
}
EXPORT_SYMBOL(posix_acl_init);
/*
* Allocate a new ACL with the specified number of entries.
*/
struct posix_acl *
posix_acl_alloc(int count, gfp_t flags)
{
const size_t size = sizeof(struct posix_acl) +
count * sizeof(struct posix_acl_entry);
struct posix_acl *acl = kmalloc(size, flags);
if (acl)
posix_acl_init(acl, count);
return acl;
}
EXPORT_SYMBOL(posix_acl_alloc);
/*
* Clone an ACL.
*/
static struct posix_acl *
posix_acl_clone(const struct posix_acl *acl, gfp_t flags)
{
struct posix_acl *clone = NULL;
if (acl) {
int size = sizeof(struct posix_acl) + acl->a_count *
sizeof(struct posix_acl_entry);
clone = kmemdup(acl, size, flags);
if (clone)
posix_acl: convert posix_acl.a_refcount from atomic_t to refcount_t atomic_t variables are currently used to implement reference counters with the following properties: - counter is initialized to 1 using atomic_set() - a resource is freed upon counter reaching zero - once counter reaches zero, its further increments aren't allowed - counter schema uses basic atomic operations (set, inc, inc_not_zero, dec_and_test, etc.) Such atomic variables should be converted to a newly provided refcount_t type and API that prevents accidental counter overflows and underflows. This is important since overflows and underflows can lead to use-after-free situation and be exploitable. The variable posix_acl.a_refcount is used as pure reference counter. Convert it to refcount_t and fix up the operations. **Important note for maintainers: Some functions from refcount_t API defined in lib/refcount.c have different memory ordering guarantees than their atomic counterparts. The full comparison can be seen in https://lkml.org/lkml/2017/11/15/57 and it is hopefully soon in state to be merged to the documentation tree. Normally the differences should not matter since refcount_t provides enough guarantees to satisfy the refcounting use cases, but in some rare cases it might matter. Please double check that you don't have some undocumented memory guarantees for this variable usage. For the posix_acl.a_refcount it might make a difference in following places: - get_cached_acl(): increment in refcount_inc_not_zero() only guarantees control dependency on success vs. fully ordered atomic counterpart. However this operation is performed under rcu_read_lock(), so this should be fine. - posix_acl_release(): decrement in refcount_dec_and_test() only provides RELEASE ordering and control dependency on success vs. fully ordered atomic counterpart Suggested-by: Kees Cook <keescook@chromium.org> Reviewed-by: David Windsor <dwindsor@gmail.com> Reviewed-by: Hans Liljestrand <ishkamiel@gmail.com> Signed-off-by: Elena Reshetova <elena.reshetova@intel.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-11-29 19:19:31 +08:00
refcount_set(&clone->a_refcount, 1);
}
return clone;
}
/*
* Check if an acl is valid. Returns 0 if it is, or -E... otherwise.
*/
int
posix_acl_valid(struct user_namespace *user_ns, const struct posix_acl *acl)
{
const struct posix_acl_entry *pa, *pe;
int state = ACL_USER_OBJ;
int needs_mask = 0;
FOREACH_ACL_ENTRY(pa, acl, pe) {
if (pa->e_perm & ~(ACL_READ|ACL_WRITE|ACL_EXECUTE))
return -EINVAL;
switch (pa->e_tag) {
case ACL_USER_OBJ:
if (state == ACL_USER_OBJ) {
state = ACL_USER;
break;
}
return -EINVAL;
case ACL_USER:
if (state != ACL_USER)
return -EINVAL;
if (!kuid_has_mapping(user_ns, pa->e_uid))
return -EINVAL;
needs_mask = 1;
break;
case ACL_GROUP_OBJ:
if (state == ACL_USER) {
state = ACL_GROUP;
break;
}
return -EINVAL;
case ACL_GROUP:
if (state != ACL_GROUP)
return -EINVAL;
if (!kgid_has_mapping(user_ns, pa->e_gid))
2012-02-08 10:52:57 +08:00
return -EINVAL;
needs_mask = 1;
break;
case ACL_MASK:
if (state != ACL_GROUP)
return -EINVAL;
state = ACL_OTHER;
break;
case ACL_OTHER:
if (state == ACL_OTHER ||
(state == ACL_GROUP && !needs_mask)) {
state = 0;
break;
}
return -EINVAL;
default:
return -EINVAL;
}
}
if (state == 0)
return 0;
return -EINVAL;
}
EXPORT_SYMBOL(posix_acl_valid);
/*
* Returns 0 if the acl can be exactly represented in the traditional
* file mode permission bits, or else 1. Returns -E... on error.
*/
int
posix_acl_equiv_mode(const struct posix_acl *acl, umode_t *mode_p)
{
const struct posix_acl_entry *pa, *pe;
umode_t mode = 0;
int not_equiv = 0;
/*
* A null ACL can always be presented as mode bits.
*/
if (!acl)
return 0;
FOREACH_ACL_ENTRY(pa, acl, pe) {
switch (pa->e_tag) {
case ACL_USER_OBJ:
mode |= (pa->e_perm & S_IRWXO) << 6;
break;
case ACL_GROUP_OBJ:
mode |= (pa->e_perm & S_IRWXO) << 3;
break;
case ACL_OTHER:
mode |= pa->e_perm & S_IRWXO;
break;
case ACL_MASK:
mode = (mode & ~S_IRWXG) |
((pa->e_perm & S_IRWXO) << 3);
not_equiv = 1;
break;
case ACL_USER:
case ACL_GROUP:
not_equiv = 1;
break;
default:
return -EINVAL;
}
}
if (mode_p)
*mode_p = (*mode_p & ~S_IRWXUGO) | mode;
return not_equiv;
}
EXPORT_SYMBOL(posix_acl_equiv_mode);
/*
* Create an ACL representing the file mode permission bits of an inode.
*/
struct posix_acl *
posix_acl_from_mode(umode_t mode, gfp_t flags)
{
struct posix_acl *acl = posix_acl_alloc(3, flags);
if (!acl)
return ERR_PTR(-ENOMEM);
acl->a_entries[0].e_tag = ACL_USER_OBJ;
acl->a_entries[0].e_perm = (mode & S_IRWXU) >> 6;
acl->a_entries[1].e_tag = ACL_GROUP_OBJ;
acl->a_entries[1].e_perm = (mode & S_IRWXG) >> 3;
acl->a_entries[2].e_tag = ACL_OTHER;
acl->a_entries[2].e_perm = (mode & S_IRWXO);
return acl;
}
EXPORT_SYMBOL(posix_acl_from_mode);
/*
* Return 0 if current is granted want access to the inode
* by the acl. Returns -E... otherwise.
*/
int
posix_acl_permission(struct inode *inode, const struct posix_acl *acl, int want)
{
const struct posix_acl_entry *pa, *pe, *mask_obj;
int found = 0;
want &= MAY_READ | MAY_WRITE | MAY_EXEC | MAY_NOT_BLOCK;
FOREACH_ACL_ENTRY(pa, acl, pe) {
switch(pa->e_tag) {
case ACL_USER_OBJ:
/* (May have been checked already) */
2012-02-08 10:52:57 +08:00
if (uid_eq(inode->i_uid, current_fsuid()))
goto check_perm;
break;
case ACL_USER:
2012-02-08 10:52:57 +08:00
if (uid_eq(pa->e_uid, current_fsuid()))
goto mask;
break;
case ACL_GROUP_OBJ:
if (in_group_p(inode->i_gid)) {
found = 1;
if ((pa->e_perm & want) == want)
goto mask;
}
break;
case ACL_GROUP:
2012-02-08 10:52:57 +08:00
if (in_group_p(pa->e_gid)) {
found = 1;
if ((pa->e_perm & want) == want)
goto mask;
}
break;
case ACL_MASK:
break;
case ACL_OTHER:
if (found)
return -EACCES;
else
goto check_perm;
default:
return -EIO;
}
}
return -EIO;
mask:
for (mask_obj = pa+1; mask_obj != pe; mask_obj++) {
if (mask_obj->e_tag == ACL_MASK) {
if ((pa->e_perm & mask_obj->e_perm & want) == want)
return 0;
return -EACCES;
}
}
check_perm:
if ((pa->e_perm & want) == want)
return 0;
return -EACCES;
}
/*
* Modify acl when creating a new inode. The caller must ensure the acl is
* only referenced once.
*
* mode_p initially must contain the mode parameter to the open() / creat()
* system calls. All permissions that are not granted by the acl are removed.
* The permissions in the acl are changed to reflect the mode_p parameter.
*/
static int posix_acl_create_masq(struct posix_acl *acl, umode_t *mode_p)
{
struct posix_acl_entry *pa, *pe;
struct posix_acl_entry *group_obj = NULL, *mask_obj = NULL;
umode_t mode = *mode_p;
int not_equiv = 0;
/* assert(atomic_read(acl->a_refcount) == 1); */
FOREACH_ACL_ENTRY(pa, acl, pe) {
switch(pa->e_tag) {
case ACL_USER_OBJ:
pa->e_perm &= (mode >> 6) | ~S_IRWXO;
mode &= (pa->e_perm << 6) | ~S_IRWXU;
break;
case ACL_USER:
case ACL_GROUP:
not_equiv = 1;
break;
case ACL_GROUP_OBJ:
group_obj = pa;
break;
case ACL_OTHER:
pa->e_perm &= mode | ~S_IRWXO;
mode &= pa->e_perm | ~S_IRWXO;
break;
case ACL_MASK:
mask_obj = pa;
not_equiv = 1;
break;
default:
return -EIO;
}
}
if (mask_obj) {
mask_obj->e_perm &= (mode >> 3) | ~S_IRWXO;
mode &= (mask_obj->e_perm << 3) | ~S_IRWXG;
} else {
if (!group_obj)
return -EIO;
group_obj->e_perm &= (mode >> 3) | ~S_IRWXO;
mode &= (group_obj->e_perm << 3) | ~S_IRWXG;
}
*mode_p = (*mode_p & ~S_IRWXUGO) | mode;
return not_equiv;
}
/*
* Modify the ACL for the chmod syscall.
*/
static int __posix_acl_chmod_masq(struct posix_acl *acl, umode_t mode)
{
struct posix_acl_entry *group_obj = NULL, *mask_obj = NULL;
struct posix_acl_entry *pa, *pe;
/* assert(atomic_read(acl->a_refcount) == 1); */
FOREACH_ACL_ENTRY(pa, acl, pe) {
switch(pa->e_tag) {
case ACL_USER_OBJ:
pa->e_perm = (mode & S_IRWXU) >> 6;
break;
case ACL_USER:
case ACL_GROUP:
break;
case ACL_GROUP_OBJ:
group_obj = pa;
break;
case ACL_MASK:
mask_obj = pa;
break;
case ACL_OTHER:
pa->e_perm = (mode & S_IRWXO);
break;
default:
return -EIO;
}
}
if (mask_obj) {
mask_obj->e_perm = (mode & S_IRWXG) >> 3;
} else {
if (!group_obj)
return -EIO;
group_obj->e_perm = (mode & S_IRWXG) >> 3;
}
return 0;
}
int
__posix_acl_create(struct posix_acl **acl, gfp_t gfp, umode_t *mode_p)
{
struct posix_acl *clone = posix_acl_clone(*acl, gfp);
int err = -ENOMEM;
if (clone) {
err = posix_acl_create_masq(clone, mode_p);
if (err < 0) {
posix_acl_release(clone);
clone = NULL;
}
}
posix_acl_release(*acl);
*acl = clone;
return err;
}
EXPORT_SYMBOL(__posix_acl_create);
int
__posix_acl_chmod(struct posix_acl **acl, gfp_t gfp, umode_t mode)
{
struct posix_acl *clone = posix_acl_clone(*acl, gfp);
int err = -ENOMEM;
if (clone) {
err = __posix_acl_chmod_masq(clone, mode);
if (err) {
posix_acl_release(clone);
clone = NULL;
}
}
posix_acl_release(*acl);
*acl = clone;
return err;
}
EXPORT_SYMBOL(__posix_acl_chmod);
int
posix_acl_chmod(struct inode *inode, umode_t mode)
{
struct posix_acl *acl;
int ret = 0;
if (!IS_POSIXACL(inode))
return 0;
if (!inode->i_op->set_acl)
return -EOPNOTSUPP;
acl = get_acl(inode, ACL_TYPE_ACCESS);
if (IS_ERR_OR_NULL(acl)) {
if (acl == ERR_PTR(-EOPNOTSUPP))
return 0;
return PTR_ERR(acl);
}
ret = __posix_acl_chmod(&acl, GFP_KERNEL, mode);
if (ret)
return ret;
ret = inode->i_op->set_acl(inode, acl, ACL_TYPE_ACCESS);
posix_acl_release(acl);
return ret;
}
EXPORT_SYMBOL(posix_acl_chmod);
int
posix_acl_create(struct inode *dir, umode_t *mode,
struct posix_acl **default_acl, struct posix_acl **acl)
{
struct posix_acl *p;
struct posix_acl *clone;
int ret;
*acl = NULL;
*default_acl = NULL;
if (S_ISLNK(*mode) || !IS_POSIXACL(dir))
return 0;
p = get_acl(dir, ACL_TYPE_DEFAULT);
if (!p || p == ERR_PTR(-EOPNOTSUPP)) {
*mode &= ~current_umask();
return 0;
}
if (IS_ERR(p))
return PTR_ERR(p);
ret = -ENOMEM;
clone = posix_acl_clone(p, GFP_NOFS);
if (!clone)
goto err_release;
ret = posix_acl_create_masq(clone, mode);
if (ret < 0)
goto err_release_clone;
if (ret == 0)
posix_acl_release(clone);
else
*acl = clone;
if (!S_ISDIR(*mode))
posix_acl_release(p);
else
*default_acl = p;
return 0;
err_release_clone:
posix_acl_release(clone);
err_release:
posix_acl_release(p);
return ret;
}
EXPORT_SYMBOL_GPL(posix_acl_create);
/**
* posix_acl_update_mode - update mode in set_acl
*
* Update the file mode when setting an ACL: compute the new file permission
* bits based on the ACL. In addition, if the ACL is equivalent to the new
* file mode, set *acl to NULL to indicate that no ACL should be set.
*
* As with chmod, clear the setgit bit if the caller is not in the owning group
* or capable of CAP_FSETID (see inode_change_ok).
*
* Called from set_acl inode operations.
*/
int posix_acl_update_mode(struct inode *inode, umode_t *mode_p,
struct posix_acl **acl)
{
umode_t mode = inode->i_mode;
int error;
error = posix_acl_equiv_mode(*acl, &mode);
if (error < 0)
return error;
if (error == 0)
*acl = NULL;
if (!in_group_p(inode->i_gid) &&
!capable_wrt_inode_uidgid(inode, CAP_FSETID))
mode &= ~S_ISGID;
*mode_p = mode;
return 0;
}
EXPORT_SYMBOL(posix_acl_update_mode);
/*
* Fix up the uids and gids in posix acl extended attributes in place.
*/
static void posix_acl_fix_xattr_userns(
struct user_namespace *to, struct user_namespace *from,
void *value, size_t size)
{
struct posix_acl_xattr_header *header = value;
struct posix_acl_xattr_entry *entry = (void *)(header + 1), *end;
int count;
kuid_t uid;
kgid_t gid;
if (!value)
return;
if (size < sizeof(struct posix_acl_xattr_header))
return;
if (header->a_version != cpu_to_le32(POSIX_ACL_XATTR_VERSION))
return;
count = posix_acl_xattr_count(size);
if (count < 0)
return;
if (count == 0)
return;
for (end = entry + count; entry != end; entry++) {
switch(le16_to_cpu(entry->e_tag)) {
case ACL_USER:
uid = make_kuid(from, le32_to_cpu(entry->e_id));
entry->e_id = cpu_to_le32(from_kuid(to, uid));
break;
case ACL_GROUP:
gid = make_kgid(from, le32_to_cpu(entry->e_id));
entry->e_id = cpu_to_le32(from_kgid(to, gid));
break;
default:
break;
}
}
}
void posix_acl_fix_xattr_from_user(void *value, size_t size)
{
struct user_namespace *user_ns = current_user_ns();
if (user_ns == &init_user_ns)
return;
posix_acl_fix_xattr_userns(&init_user_ns, user_ns, value, size);
}
void posix_acl_fix_xattr_to_user(void *value, size_t size)
{
struct user_namespace *user_ns = current_user_ns();
if (user_ns == &init_user_ns)
return;
posix_acl_fix_xattr_userns(user_ns, &init_user_ns, value, size);
}
/*
* Convert from extended attribute to in-memory representation.
*/
struct posix_acl *
posix_acl_from_xattr(struct user_namespace *user_ns,
const void *value, size_t size)
{
const struct posix_acl_xattr_header *header = value;
const struct posix_acl_xattr_entry *entry = (const void *)(header + 1), *end;
int count;
struct posix_acl *acl;
struct posix_acl_entry *acl_e;
if (!value)
return NULL;
if (size < sizeof(struct posix_acl_xattr_header))
return ERR_PTR(-EINVAL);
if (header->a_version != cpu_to_le32(POSIX_ACL_XATTR_VERSION))
return ERR_PTR(-EOPNOTSUPP);
count = posix_acl_xattr_count(size);
if (count < 0)
return ERR_PTR(-EINVAL);
if (count == 0)
return NULL;
acl = posix_acl_alloc(count, GFP_NOFS);
if (!acl)
return ERR_PTR(-ENOMEM);
acl_e = acl->a_entries;
for (end = entry + count; entry != end; acl_e++, entry++) {
acl_e->e_tag = le16_to_cpu(entry->e_tag);
acl_e->e_perm = le16_to_cpu(entry->e_perm);
switch(acl_e->e_tag) {
case ACL_USER_OBJ:
case ACL_GROUP_OBJ:
case ACL_MASK:
case ACL_OTHER:
break;
case ACL_USER:
acl_e->e_uid =
make_kuid(user_ns,
le32_to_cpu(entry->e_id));
if (!uid_valid(acl_e->e_uid))
goto fail;
break;
case ACL_GROUP:
acl_e->e_gid =
make_kgid(user_ns,
le32_to_cpu(entry->e_id));
if (!gid_valid(acl_e->e_gid))
goto fail;
break;
default:
goto fail;
}
}
return acl;
fail:
posix_acl_release(acl);
return ERR_PTR(-EINVAL);
}
EXPORT_SYMBOL (posix_acl_from_xattr);
/*
* Convert from in-memory to extended attribute representation.
*/
int
posix_acl_to_xattr(struct user_namespace *user_ns, const struct posix_acl *acl,
void *buffer, size_t size)
{
struct posix_acl_xattr_header *ext_acl = buffer;
struct posix_acl_xattr_entry *ext_entry;
int real_size, n;
real_size = posix_acl_xattr_size(acl->a_count);
if (!buffer)
return real_size;
if (real_size > size)
return -ERANGE;
ext_entry = (void *)(ext_acl + 1);
ext_acl->a_version = cpu_to_le32(POSIX_ACL_XATTR_VERSION);
for (n=0; n < acl->a_count; n++, ext_entry++) {
const struct posix_acl_entry *acl_e = &acl->a_entries[n];
ext_entry->e_tag = cpu_to_le16(acl_e->e_tag);
ext_entry->e_perm = cpu_to_le16(acl_e->e_perm);
switch(acl_e->e_tag) {
case ACL_USER:
ext_entry->e_id =
cpu_to_le32(from_kuid(user_ns, acl_e->e_uid));
break;
case ACL_GROUP:
ext_entry->e_id =
cpu_to_le32(from_kgid(user_ns, acl_e->e_gid));
break;
default:
ext_entry->e_id = cpu_to_le32(ACL_UNDEFINED_ID);
break;
}
}
return real_size;
}
EXPORT_SYMBOL (posix_acl_to_xattr);
static int
posix_acl_xattr_get(const struct xattr_handler *handler,
struct dentry *unused, struct inode *inode,
const char *name, void *value, size_t size)
{
struct posix_acl *acl;
int error;
if (!IS_POSIXACL(inode))
return -EOPNOTSUPP;
if (S_ISLNK(inode->i_mode))
return -EOPNOTSUPP;
acl = get_acl(inode, handler->flags);
if (IS_ERR(acl))
return PTR_ERR(acl);
if (acl == NULL)
return -ENODATA;
error = posix_acl_to_xattr(&init_user_ns, acl, value, size);
posix_acl_release(acl);
return error;
}
int
set_posix_acl(struct inode *inode, int type, struct posix_acl *acl)
{
if (!IS_POSIXACL(inode))
return -EOPNOTSUPP;
if (!inode->i_op->set_acl)
return -EOPNOTSUPP;
if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode))
return acl ? -EACCES : 0;
if (!inode_owner_or_capable(inode))
return -EPERM;
if (acl) {
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace Pull userns vfs updates from Eric Biederman: "This tree contains some very long awaited work on generalizing the user namespace support for mounting filesystems to include filesystems with a backing store. The real world target is fuse but the goal is to update the vfs to allow any filesystem to be supported. This patchset is based on a lot of code review and testing to approach that goal. While looking at what is needed to support the fuse filesystem it became clear that there were things like xattrs for security modules that needed special treatment. That the resolution of those concerns would not be fuse specific. That sorting out these general issues made most sense at the generic level, where the right people could be drawn into the conversation, and the issues could be solved for everyone. At a high level what this patchset does a couple of simple things: - Add a user namespace owner (s_user_ns) to struct super_block. - Teach the vfs to handle filesystem uids and gids not mapping into to kuids and kgids and being reported as INVALID_UID and INVALID_GID in vfs data structures. By assigning a user namespace owner filesystems that are mounted with only user namespace privilege can be detected. This allows security modules and the like to know which mounts may not be trusted. This also allows the set of uids and gids that are communicated to the filesystem to be capped at the set of kuids and kgids that are in the owning user namespace of the filesystem. One of the crazier corner casees this handles is the case of inodes whose i_uid or i_gid are not mapped into the vfs. Most of the code simply doesn't care but it is easy to confuse the inode writeback path so no operation that could cause an inode write-back is permitted for such inodes (aka only reads are allowed). This set of changes starts out by cleaning up the code paths involved in user namespace permirted mounts. Then when things are clean enough adds code that cleanly sets s_user_ns. Then additional restrictions are added that are possible now that the filesystem superblock contains owner information. These changes should not affect anyone in practice, but there are some parts of these restrictions that are changes in behavior. - Andy's restriction on suid executables that does not honor the suid bit when the path is from another mount namespace (think /proc/[pid]/fd/) or when the filesystem was mounted by a less privileged user. - The replacement of the user namespace implicit setting of MNT_NODEV with implicitly setting SB_I_NODEV on the filesystem superblock instead. Using SB_I_NODEV is a stronger form that happens to make this state user invisible. The user visibility can be managed but it caused problems when it was introduced from applications reasonably expecting mount flags to be what they were set to. There is a little bit of work remaining before it is safe to support mounting filesystems with backing store in user namespaces, beyond what is in this set of changes. - Verifying the mounter has permission to read/write the block device during mount. - Teaching the integrity modules IMA and EVM to handle filesystems mounted with only user namespace root and to reduce trust in their security xattrs accordingly. - Capturing the mounters credentials and using that for permission checks in d_automount and the like. (Given that overlayfs already does this, and we need the work in d_automount it make sense to generalize this case). Furthermore there are a few changes that are on the wishlist: - Get all filesystems supporting posix acls using the generic posix acls so that posix_acl_fix_xattr_from_user and posix_acl_fix_xattr_to_user may be removed. [Maintainability] - Reducing the permission checks in places such as remount to allow the superblock owner to perform them. - Allowing the superblock owner to chown files with unmapped uids and gids to something that is mapped so the files may be treated normally. I am not considering even obvious relaxations of permission checks until it is clear there are no more corner cases that need to be locked down and handled generically. Many thanks to Seth Forshee who kept this code alive, and putting up with me rewriting substantial portions of what he did to handle more corner cases, and for his diligent testing and reviewing of my changes" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (30 commits) fs: Call d_automount with the filesystems creds fs: Update i_[ug]id_(read|write) to translate relative to s_user_ns evm: Translate user/group ids relative to s_user_ns when computing HMAC dquot: For now explicitly don't support filesystems outside of init_user_ns quota: Handle quota data stored in s_user_ns in quota_setxquota quota: Ensure qids map to the filesystem vfs: Don't create inodes with a uid or gid unknown to the vfs vfs: Don't modify inodes with a uid or gid unknown to the vfs cred: Reject inodes with invalid ids in set_create_file_as() fs: Check for invalid i_uid in may_follow_link() vfs: Verify acls are valid within superblock's s_user_ns. userns: Handle -1 in k[ug]id_has_mapping when !CONFIG_USER_NS fs: Refuse uid/gid changes which don't map into s_user_ns selinux: Add support for unprivileged mounts from user namespaces Smack: Handle labels consistently in untrusted mounts Smack: Add support for unprivileged mounts from user namespaces fs: Treat foreign mounts as nosuid fs: Limit file caps to the user namespace of the super block userns: Remove the now unnecessary FS_USERNS_DEV_MOUNT flag userns: Remove implicit MNT_NODEV fragility. ...
2016-07-30 06:54:19 +08:00
int ret = posix_acl_valid(inode->i_sb->s_user_ns, acl);
if (ret)
return ret;
}
return inode->i_op->set_acl(inode, acl, type);
}
EXPORT_SYMBOL(set_posix_acl);
static int
posix_acl_xattr_set(const struct xattr_handler *handler,
struct dentry *unused, struct inode *inode,
const char *name, const void *value,
size_t size, int flags)
{
struct posix_acl *acl = NULL;
int ret;
if (value) {
acl = posix_acl_from_xattr(&init_user_ns, value, size);
if (IS_ERR(acl))
return PTR_ERR(acl);
}
ret = set_posix_acl(inode, handler->flags, acl);
posix_acl_release(acl);
return ret;
}
static bool
posix_acl_xattr_list(struct dentry *dentry)
{
return IS_POSIXACL(d_backing_inode(dentry));
}
const struct xattr_handler posix_acl_access_xattr_handler = {
.name = XATTR_NAME_POSIX_ACL_ACCESS,
.flags = ACL_TYPE_ACCESS,
.list = posix_acl_xattr_list,
.get = posix_acl_xattr_get,
.set = posix_acl_xattr_set,
};
EXPORT_SYMBOL_GPL(posix_acl_access_xattr_handler);
const struct xattr_handler posix_acl_default_xattr_handler = {
.name = XATTR_NAME_POSIX_ACL_DEFAULT,
.flags = ACL_TYPE_DEFAULT,
.list = posix_acl_xattr_list,
.get = posix_acl_xattr_get,
.set = posix_acl_xattr_set,
};
EXPORT_SYMBOL_GPL(posix_acl_default_xattr_handler);
int simple_set_acl(struct inode *inode, struct posix_acl *acl, int type)
{
int error;
if (type == ACL_TYPE_ACCESS) {
error = posix_acl_update_mode(inode,
&inode->i_mode, &acl);
if (error)
return error;
}
inode->i_ctime = current_time(inode);
set_cached_acl(inode, type, acl);
return 0;
}
int simple_acl_create(struct inode *dir, struct inode *inode)
{
struct posix_acl *default_acl, *acl;
int error;
error = posix_acl_create(dir, &inode->i_mode, &default_acl, &acl);
if (error)
return error;
set_cached_acl(inode, ACL_TYPE_DEFAULT, default_acl);
set_cached_acl(inode, ACL_TYPE_ACCESS, acl);
if (default_acl)
posix_acl_release(default_acl);
if (acl)
posix_acl_release(acl);
return 0;
}