2002-07-30 02:04:05 +00:00
|
|
|
/*-
|
Rework MAC Framework synchronization in a number of ways in order to
improve performance:
- Eliminate custom reference count and condition variable to monitor
threads entering the framework, as this had both significant overhead
and behaved badly in the face of contention.
- Replace reference count with two locks: an rwlock and an sx lock,
which will be read-acquired by threads entering the framework
depending on whether a give policy entry point is permitted to sleep
or not.
- Replace previous mutex locking of the reference count for exclusive
access with write acquiring of both the policy list sx and rw locks,
which occurs only when policies are attached or detached.
- Do a lockless read of the dynamic policy list head before acquiring
any locks in order to reduce overhead when no dynamic policies are
loaded; this a race we can afford to lose.
- For every policy entry point invocation, decide whether sleeping is
permitted, and if not, use a _NOSLEEP() variant of the composition
macros, which will use the rwlock instead of the sxlock. In some
cases, we decide which to use based on allocation flags passed to the
MAC Framework entry point.
As with the move to rwlocks/rmlocks in pfil, this may trigger witness
warnings, but these should (generally) be false positives as all
acquisition of the locks is for read with two very narrow exceptions
for policy load/unload, and those code blocks should never acquire
other locks.
Sponsored by: Google, Inc.
Obtained from: TrustedBSD Project
Discussed with: csjp (idea, not specific patch)
2009-03-14 16:06:06 +00:00
|
|
|
* Copyright (c) 1999-2002, 2006, 2009 Robert N. M. Watson
|
2002-07-30 02:04:05 +00:00
|
|
|
* Copyright (c) 2001 Ilmar S. Habibulin
|
2005-04-16 18:33:13 +00:00
|
|
|
* Copyright (c) 2001-2005 Networks Associates Technology, Inc.
|
2006-10-22 11:52:19 +00:00
|
|
|
* Copyright (c) 2005-2006 SPARTA, Inc.
|
Introduce two related changes to the TrustedBSD MAC Framework:
(1) Abstract interpreter vnode labeling in execve(2) and mac_execve(2)
so that the general exec code isn't aware of the details of
allocating, copying, and freeing labels, rather, simply passes in
a void pointer to start and stop functions that will be used by
the framework. This change will be MFC'd.
(2) Introduce a new flags field to the MAC_POLICY_SET(9) interface
allowing policies to declare which types of objects require label
allocation, initialization, and destruction, and define a set of
flags covering various supported object types (MPC_OBJECT_PROC,
MPC_OBJECT_VNODE, MPC_OBJECT_INPCB, ...). This change reduces the
overhead of compiling the MAC Framework into the kernel if policies
aren't loaded, or if policies require labels on only a small number
or even no object types. Each time a policy is loaded or unloaded,
we recalculate a mask of labeled object types across all policies
present in the system. Eliminate MAC_ALWAYS_LABEL_MBUF option as it
is no longer required.
MFC after: 1 week ((1) only)
Reviewed by: csjp
Obtained from: TrustedBSD Project
Sponsored by: Apple, Inc.
2008-08-23 15:26:36 +00:00
|
|
|
* Copyright (c) 2008 Apple Inc.
|
2002-07-30 02:04:05 +00:00
|
|
|
* All rights reserved.
|
|
|
|
*
|
|
|
|
* This software was developed by Robert Watson and Ilmar Habibulin for the
|
|
|
|
* TrustedBSD Project.
|
|
|
|
*
|
2002-11-04 01:42:39 +00:00
|
|
|
* This software was developed for the FreeBSD Project in part by Network
|
|
|
|
* Associates Laboratories, the Security Research Division of Network
|
|
|
|
* Associates, Inc. under DARPA/SPAWAR contract N66001-01-C-8035 ("CBOSS"),
|
|
|
|
* as part of the DARPA CHATS research program.
|
2002-07-30 02:04:05 +00:00
|
|
|
*
|
2005-07-14 10:46:03 +00:00
|
|
|
* This software was enhanced by SPARTA ISSO under SPAWAR contract
|
|
|
|
* N66001-04-C-6019 ("SEFOS").
|
|
|
|
*
|
Rework MAC Framework synchronization in a number of ways in order to
improve performance:
- Eliminate custom reference count and condition variable to monitor
threads entering the framework, as this had both significant overhead
and behaved badly in the face of contention.
- Replace reference count with two locks: an rwlock and an sx lock,
which will be read-acquired by threads entering the framework
depending on whether a give policy entry point is permitted to sleep
or not.
- Replace previous mutex locking of the reference count for exclusive
access with write acquiring of both the policy list sx and rw locks,
which occurs only when policies are attached or detached.
- Do a lockless read of the dynamic policy list head before acquiring
any locks in order to reduce overhead when no dynamic policies are
loaded; this a race we can afford to lose.
- For every policy entry point invocation, decide whether sleeping is
permitted, and if not, use a _NOSLEEP() variant of the composition
macros, which will use the rwlock instead of the sxlock. In some
cases, we decide which to use based on allocation flags passed to the
MAC Framework entry point.
As with the move to rwlocks/rmlocks in pfil, this may trigger witness
warnings, but these should (generally) be false positives as all
acquisition of the locks is for read with two very narrow exceptions
for policy load/unload, and those code blocks should never acquire
other locks.
Sponsored by: Google, Inc.
Obtained from: TrustedBSD Project
Discussed with: csjp (idea, not specific patch)
2009-03-14 16:06:06 +00:00
|
|
|
* This software was developed at the University of Cambridge Computer
|
|
|
|
* Laboratory with support from a grant from Google, Inc.
|
|
|
|
*
|
2002-07-30 02:04:05 +00:00
|
|
|
* 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 AUTHOR 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 AUTHOR 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.
|
|
|
|
*/
|
2003-06-11 00:56:59 +00:00
|
|
|
|
|
|
|
#include <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
2002-07-30 02:04:05 +00:00
|
|
|
#include "opt_mac.h"
|
2002-08-01 17:47:56 +00:00
|
|
|
|
2002-07-30 02:04:05 +00:00
|
|
|
#include <sys/param.h>
|
2014-03-16 10:55:57 +00:00
|
|
|
#include <sys/capsicum.h>
|
2008-03-31 12:01:21 +00:00
|
|
|
#include <sys/fcntl.h>
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
#include <sys/kernel.h>
|
|
|
|
#include <sys/lock.h>
|
2002-09-05 07:02:43 +00:00
|
|
|
#include <sys/malloc.h>
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
#include <sys/mutex.h>
|
|
|
|
#include <sys/mac.h>
|
|
|
|
#include <sys/proc.h>
|
|
|
|
#include <sys/systm.h>
|
2011-02-25 10:11:01 +00:00
|
|
|
#include <sys/sysctl.h>
|
2002-07-30 02:04:05 +00:00
|
|
|
#include <sys/sysproto.h>
|
|
|
|
#include <sys/sysent.h>
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
#include <sys/vnode.h>
|
|
|
|
#include <sys/mount.h>
|
|
|
|
#include <sys/file.h>
|
|
|
|
#include <sys/namei.h>
|
|
|
|
#include <sys/socket.h>
|
|
|
|
#include <sys/pipe.h>
|
|
|
|
#include <sys/socketvar.h>
|
|
|
|
|
2006-10-22 11:52:19 +00:00
|
|
|
#include <security/mac/mac_framework.h>
|
mac_Finish break-out of kern_mac.c into parts:
Include src/sys/security/mac/mac_internal.h in kern_mac.c.
Remove redundant defines from the include: SYSCTL_DECL(), debug macros,
composition macros.
Unstaticize various bits now exposed to the remainder of the kernel:
mac_init_label(), mac_destroy_label().
Remove all the functions now implemented in mac_process/mac_vfs/mac_net/
mac_pipe. Also remove debug counters, sysctls exporting debug
counters, enforcement flags, sysctls exporting enforcement flags.
Leave module declaration, sysctl nodes, mactemp malloc type, system
calls.
This should conclude MAC/LINT/NOTES breakage from the break-out process,
but I'm running builds now to make sure I caught everything.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-10-22 20:59:31 +00:00
|
|
|
#include <security/mac/mac_internal.h>
|
2006-12-22 23:34:47 +00:00
|
|
|
#include <security/mac/mac_policy.h>
|
mac_Finish break-out of kern_mac.c into parts:
Include src/sys/security/mac/mac_internal.h in kern_mac.c.
Remove redundant defines from the include: SYSCTL_DECL(), debug macros,
composition macros.
Unstaticize various bits now exposed to the remainder of the kernel:
mac_init_label(), mac_destroy_label().
Remove all the functions now implemented in mac_process/mac_vfs/mac_net/
mac_pipe. Also remove debug counters, sysctls exporting debug
counters, enforcement flags, sysctls exporting enforcement flags.
Leave module declaration, sysctl nodes, mactemp malloc type, system
calls.
This should conclude MAC/LINT/NOTES breakage from the break-out process,
but I'm running builds now to make sure I caught everything.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-10-22 20:59:31 +00:00
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
#ifdef MAC
|
|
|
|
|
2011-03-04 09:03:54 +00:00
|
|
|
FEATURE(security_mac, "Mandatory Access Control Framework support");
|
2011-02-25 10:11:01 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_pid(struct thread *td, struct __mac_get_pid_args *uap)
|
2002-10-22 14:29:47 +00:00
|
|
|
{
|
|
|
|
char *elements, *buffer;
|
|
|
|
struct mac mac;
|
|
|
|
struct proc *tproc;
|
|
|
|
struct ucred *tcred;
|
|
|
|
int error;
|
|
|
|
|
2002-12-14 01:56:26 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
2002-10-22 14:29:47 +00:00
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
error = mac_check_structmac_consistent(&mac);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
tproc = pfind(uap->pid);
|
|
|
|
if (tproc == NULL)
|
|
|
|
return (ESRCH);
|
|
|
|
|
|
|
|
tcred = NULL; /* Satisfy gcc. */
|
|
|
|
error = p_cansee(td, tproc);
|
|
|
|
if (error == 0)
|
|
|
|
tcred = crhold(tproc->p_ucred);
|
|
|
|
PROC_UNLOCK(tproc);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
elements = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, elements, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
crfree(tcred);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK | M_ZERO);
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_cred_externalize_label(tcred->cr_label, elements,
|
2003-11-06 03:42:43 +00:00
|
|
|
buffer, mac.m_buflen);
|
2002-10-22 14:29:47 +00:00
|
|
|
if (error == 0)
|
|
|
|
error = copyout(buffer, mac.m_string, strlen(buffer)+1);
|
|
|
|
|
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
crfree(tcred);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_proc(struct thread *td, struct __mac_get_proc_args *uap)
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
{
|
2002-10-22 14:29:47 +00:00
|
|
|
char *elements, *buffer;
|
|
|
|
struct mac mac;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
int error;
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
error = mac_check_structmac_consistent(&mac);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
elements = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, elements, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK | M_ZERO);
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_cred_externalize_label(td->td_ucred->cr_label,
|
2003-11-06 03:42:43 +00:00
|
|
|
elements, buffer, mac.m_buflen);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error == 0)
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyout(buffer, mac.m_string, strlen(buffer)+1);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
free(elements, M_MACTEMP);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_proc(struct thread *td, struct __mac_set_proc_args *uap)
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
{
|
|
|
|
struct ucred *newcred, *oldcred;
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *intlabel;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct proc *p;
|
|
|
|
struct mac mac;
|
|
|
|
char *buffer;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
int error;
|
|
|
|
|
Introduce two related changes to the TrustedBSD MAC Framework:
(1) Abstract interpreter vnode labeling in execve(2) and mac_execve(2)
so that the general exec code isn't aware of the details of
allocating, copying, and freeing labels, rather, simply passes in
a void pointer to start and stop functions that will be used by
the framework. This change will be MFC'd.
(2) Introduce a new flags field to the MAC_POLICY_SET(9) interface
allowing policies to declare which types of objects require label
allocation, initialization, and destruction, and define a set of
flags covering various supported object types (MPC_OBJECT_PROC,
MPC_OBJECT_VNODE, MPC_OBJECT_INPCB, ...). This change reduces the
overhead of compiling the MAC Framework into the kernel if policies
aren't loaded, or if policies require labels on only a small number
or even no object types. Each time a policy is loaded or unloaded,
we recalculate a mask of labeled object types across all policies
present in the system. Eliminate MAC_ALWAYS_LABEL_MBUF option as it
is no longer required.
MFC after: 1 week ((1) only)
Reviewed by: csjp
Obtained from: TrustedBSD Project
Sponsored by: Apple, Inc.
2008-08-23 15:26:36 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_CRED))
|
|
|
|
return (EINVAL);
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = mac_check_structmac_consistent(&mac);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, buffer, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_cred_label_alloc();
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_cred_internalize_label(intlabel, buffer);
|
2002-10-22 14:29:47 +00:00
|
|
|
free(buffer, M_MACTEMP);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
if (error)
|
|
|
|
goto out;
|
2002-10-22 14:29:47 +00:00
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
newcred = crget();
|
|
|
|
|
|
|
|
p = td->td_proc;
|
|
|
|
PROC_LOCK(p);
|
|
|
|
oldcred = p->p_ucred;
|
|
|
|
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_cred_check_relabel(oldcred, intlabel);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error) {
|
|
|
|
PROC_UNLOCK(p);
|
|
|
|
crfree(newcred);
|
2002-10-22 14:29:47 +00:00
|
|
|
goto out;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
setsugid(p);
|
|
|
|
crcopy(newcred, oldcred);
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_cred_relabel(newcred, intlabel);
|
2015-03-16 00:10:03 +00:00
|
|
|
proc_set_cred(p, newcred);
|
2002-08-19 20:26:32 +00:00
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
PROC_UNLOCK(p);
|
|
|
|
crfree(oldcred);
|
2008-10-28 12:49:07 +00:00
|
|
|
mac_proc_vm_revoke(td);
|
2002-10-22 14:29:47 +00:00
|
|
|
|
|
|
|
out:
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_cred_label_free(intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_fd(struct thread *td, struct __mac_get_fd_args *uap)
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
{
|
2002-10-22 14:29:47 +00:00
|
|
|
char *elements, *buffer;
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *intlabel;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
struct file *fp;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct mac mac;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
struct vnode *vp;
|
|
|
|
struct pipe *pipe;
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
struct socket *so;
|
Change the cap_rights_t type from uint64_t to a structure that we can extend
in the future in a backward compatible (API and ABI) way.
The cap_rights_t represents capability rights. We used to use one bit to
represent one right, but we are running out of spare bits. Currently the new
structure provides place for 114 rights (so 50 more than the previous
cap_rights_t), but it is possible to grow the structure to hold at least 285
rights, although we can make it even larger if 285 rights won't be enough.
The structure definition looks like this:
struct cap_rights {
uint64_t cr_rights[CAP_RIGHTS_VERSION + 2];
};
The initial CAP_RIGHTS_VERSION is 0.
The top two bits in the first element of the cr_rights[] array contain total
number of elements in the array - 2. This means if those two bits are equal to
0, we have 2 array elements.
The top two bits in all remaining array elements should be 0.
The next five bits in all array elements contain array index. Only one bit is
used and bit position in this five-bits range defines array index. This means
there can be at most five array elements in the future.
To define new right the CAPRIGHT() macro must be used. The macro takes two
arguments - an array index and a bit to set, eg.
#define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL)
We still support aliases that combine few rights, but the rights have to belong
to the same array element, eg:
#define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL)
#define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL)
#define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP)
There is new API to manage the new cap_rights_t structure:
cap_rights_t *cap_rights_init(cap_rights_t *rights, ...);
void cap_rights_set(cap_rights_t *rights, ...);
void cap_rights_clear(cap_rights_t *rights, ...);
bool cap_rights_is_set(const cap_rights_t *rights, ...);
bool cap_rights_is_valid(const cap_rights_t *rights);
void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src);
void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src);
bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little);
Capability rights to the cap_rights_init(), cap_rights_set(),
cap_rights_clear() and cap_rights_is_set() functions are provided by
separating them with commas, eg:
cap_rights_t rights;
cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT);
There is no need to terminate the list of rights, as those functions are
actually macros that take care of the termination, eg:
#define cap_rights_set(rights, ...) \
__cap_rights_set((rights), __VA_ARGS__, 0ULL)
void __cap_rights_set(cap_rights_t *rights, ...);
Thanks to using one bit as an array index we can assert in those functions that
there are no two rights belonging to different array elements provided
together. For example this is illegal and will be detected, because CAP_LOOKUP
belongs to element 0 and CAP_PDKILL to element 1:
cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL);
Providing several rights that belongs to the same array's element this way is
correct, but is not advised. It should only be used for aliases definition.
This commit also breaks compatibility with some existing Capsicum system calls,
but I see no other way to do that. This should be fine as Capsicum is still
experimental and this change is not going to 9.x.
Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
|
|
|
cap_rights_t rights;
|
2002-10-22 14:29:47 +00:00
|
|
|
short label_type;
|
2012-10-22 17:50:54 +00:00
|
|
|
int error;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
error = mac_check_structmac_consistent(&mac);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
elements = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, elements, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK | M_ZERO);
|
Change the cap_rights_t type from uint64_t to a structure that we can extend
in the future in a backward compatible (API and ABI) way.
The cap_rights_t represents capability rights. We used to use one bit to
represent one right, but we are running out of spare bits. Currently the new
structure provides place for 114 rights (so 50 more than the previous
cap_rights_t), but it is possible to grow the structure to hold at least 285
rights, although we can make it even larger if 285 rights won't be enough.
The structure definition looks like this:
struct cap_rights {
uint64_t cr_rights[CAP_RIGHTS_VERSION + 2];
};
The initial CAP_RIGHTS_VERSION is 0.
The top two bits in the first element of the cr_rights[] array contain total
number of elements in the array - 2. This means if those two bits are equal to
0, we have 2 array elements.
The top two bits in all remaining array elements should be 0.
The next five bits in all array elements contain array index. Only one bit is
used and bit position in this five-bits range defines array index. This means
there can be at most five array elements in the future.
To define new right the CAPRIGHT() macro must be used. The macro takes two
arguments - an array index and a bit to set, eg.
#define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL)
We still support aliases that combine few rights, but the rights have to belong
to the same array element, eg:
#define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL)
#define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL)
#define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP)
There is new API to manage the new cap_rights_t structure:
cap_rights_t *cap_rights_init(cap_rights_t *rights, ...);
void cap_rights_set(cap_rights_t *rights, ...);
void cap_rights_clear(cap_rights_t *rights, ...);
bool cap_rights_is_set(const cap_rights_t *rights, ...);
bool cap_rights_is_valid(const cap_rights_t *rights);
void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src);
void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src);
bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little);
Capability rights to the cap_rights_init(), cap_rights_set(),
cap_rights_clear() and cap_rights_is_set() functions are provided by
separating them with commas, eg:
cap_rights_t rights;
cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT);
There is no need to terminate the list of rights, as those functions are
actually macros that take care of the termination, eg:
#define cap_rights_set(rights, ...) \
__cap_rights_set((rights), __VA_ARGS__, 0ULL)
void __cap_rights_set(cap_rights_t *rights, ...);
Thanks to using one bit as an array index we can assert in those functions that
there are no two rights belonging to different array elements provided
together. For example this is illegal and will be detected, because CAP_LOOKUP
belongs to element 0 and CAP_PDKILL to element 1:
cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL);
Providing several rights that belongs to the same array's element this way is
correct, but is not advised. It should only be used for aliases definition.
This commit also breaks compatibility with some existing Capsicum system calls,
but I see no other way to do that. This should be fine as Capsicum is still
experimental and this change is not going to 9.x.
Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
|
|
|
error = fget(td, uap->fd, cap_rights_init(&rights, CAP_MAC_GET), &fp);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
|
|
|
goto out;
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
label_type = fp->f_type;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
switch (fp->f_type) {
|
|
|
|
case DTYPE_FIFO:
|
|
|
|
case DTYPE_VNODE:
|
2012-04-08 11:01:49 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_VNODE)) {
|
|
|
|
error = EINVAL;
|
|
|
|
goto out_fdrop;
|
|
|
|
}
|
2003-06-22 08:41:43 +00:00
|
|
|
vp = fp->f_vnode;
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_vnode_label_alloc();
|
2008-01-10 01:10:58 +00:00
|
|
|
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_vnode_copy_label(vp->v_label, intlabel);
|
2008-01-13 14:44:15 +00:00
|
|
|
VOP_UNLOCK(vp, 0);
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_vnode_externalize_label(intlabel, elements,
|
2003-11-12 22:19:15 +00:00
|
|
|
buffer, mac.m_buflen);
|
|
|
|
mac_vnode_label_free(intlabel);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
break;
|
2003-11-12 22:19:15 +00:00
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
case DTYPE_PIPE:
|
2012-04-08 11:01:49 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_PIPE)) {
|
|
|
|
error = EINVAL;
|
|
|
|
goto out_fdrop;
|
|
|
|
}
|
2003-01-13 00:33:17 +00:00
|
|
|
pipe = fp->f_data;
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_pipe_label_alloc();
|
2002-10-22 14:29:47 +00:00
|
|
|
PIPE_LOCK(pipe);
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_pipe_copy_label(pipe->pipe_pair->pp_label, intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
PIPE_UNLOCK(pipe);
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_pipe_externalize_label(intlabel, elements,
|
2003-11-06 03:42:43 +00:00
|
|
|
buffer, mac.m_buflen);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_pipe_label_free(intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
break;
|
2003-11-12 22:19:15 +00:00
|
|
|
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
case DTYPE_SOCKET:
|
2012-04-08 11:01:49 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_SOCKET)) {
|
|
|
|
error = EINVAL;
|
|
|
|
goto out_fdrop;
|
|
|
|
}
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
so = fp->f_data;
|
|
|
|
intlabel = mac_socket_label_alloc(M_WAITOK);
|
2005-04-16 18:33:13 +00:00
|
|
|
SOCK_LOCK(so);
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_socket_copy_label(so->so_label, intlabel);
|
2005-04-16 18:33:13 +00:00
|
|
|
SOCK_UNLOCK(so);
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_socket_externalize_label(intlabel, elements,
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
buffer, mac.m_buflen);
|
|
|
|
mac_socket_label_free(intlabel);
|
|
|
|
break;
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
default:
|
2003-11-12 22:19:15 +00:00
|
|
|
error = EINVAL;
|
2002-10-22 14:29:47 +00:00
|
|
|
}
|
|
|
|
if (error == 0)
|
|
|
|
error = copyout(buffer, mac.m_string, strlen(buffer)+1);
|
2012-04-08 11:01:49 +00:00
|
|
|
out_fdrop:
|
|
|
|
fdrop(fp, td);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
out:
|
2002-10-22 14:29:47 +00:00
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
free(elements, M_MACTEMP);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_file(struct thread *td, struct __mac_get_file_args *uap)
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
{
|
2002-10-22 14:29:47 +00:00
|
|
|
char *elements, *buffer;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
struct nameidata nd;
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *intlabel;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct mac mac;
|
2012-10-22 17:50:54 +00:00
|
|
|
int error;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
Introduce two related changes to the TrustedBSD MAC Framework:
(1) Abstract interpreter vnode labeling in execve(2) and mac_execve(2)
so that the general exec code isn't aware of the details of
allocating, copying, and freeing labels, rather, simply passes in
a void pointer to start and stop functions that will be used by
the framework. This change will be MFC'd.
(2) Introduce a new flags field to the MAC_POLICY_SET(9) interface
allowing policies to declare which types of objects require label
allocation, initialization, and destruction, and define a set of
flags covering various supported object types (MPC_OBJECT_PROC,
MPC_OBJECT_VNODE, MPC_OBJECT_INPCB, ...). This change reduces the
overhead of compiling the MAC Framework into the kernel if policies
aren't loaded, or if policies require labels on only a small number
or even no object types. Each time a policy is loaded or unloaded,
we recalculate a mask of labeled object types across all policies
present in the system. Eliminate MAC_ALWAYS_LABEL_MBUF option as it
is no longer required.
MFC after: 1 week ((1) only)
Reviewed by: csjp
Obtained from: TrustedBSD Project
Sponsored by: Apple, Inc.
2008-08-23 15:26:36 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_VNODE))
|
|
|
|
return (EINVAL);
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
error = mac_check_structmac_consistent(&mac);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
elements = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, elements, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK | M_ZERO);
|
2012-10-22 17:50:54 +00:00
|
|
|
NDINIT(&nd, LOOKUP, LOCKLEAF | FOLLOW, UIO_USERSPACE,
|
2005-10-04 14:32:58 +00:00
|
|
|
uap->path_p, td);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
error = namei(&nd);
|
|
|
|
if (error)
|
|
|
|
goto out;
|
|
|
|
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_vnode_label_alloc();
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_vnode_copy_label(nd.ni_vp->v_label, intlabel);
|
|
|
|
error = mac_vnode_externalize_label(intlabel, elements, buffer,
|
2003-11-06 03:42:43 +00:00
|
|
|
mac.m_buflen);
|
2002-10-22 14:29:47 +00:00
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
NDFREE(&nd, 0);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_vnode_label_free(intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
if (error == 0)
|
|
|
|
error = copyout(buffer, mac.m_string, strlen(buffer)+1);
|
|
|
|
|
|
|
|
out:
|
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_link(struct thread *td, struct __mac_get_link_args *uap)
|
2002-10-22 14:29:47 +00:00
|
|
|
{
|
|
|
|
char *elements, *buffer;
|
|
|
|
struct nameidata nd;
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *intlabel;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct mac mac;
|
2012-10-22 17:50:54 +00:00
|
|
|
int error;
|
2002-10-22 14:29:47 +00:00
|
|
|
|
Introduce two related changes to the TrustedBSD MAC Framework:
(1) Abstract interpreter vnode labeling in execve(2) and mac_execve(2)
so that the general exec code isn't aware of the details of
allocating, copying, and freeing labels, rather, simply passes in
a void pointer to start and stop functions that will be used by
the framework. This change will be MFC'd.
(2) Introduce a new flags field to the MAC_POLICY_SET(9) interface
allowing policies to declare which types of objects require label
allocation, initialization, and destruction, and define a set of
flags covering various supported object types (MPC_OBJECT_PROC,
MPC_OBJECT_VNODE, MPC_OBJECT_INPCB, ...). This change reduces the
overhead of compiling the MAC Framework into the kernel if policies
aren't loaded, or if policies require labels on only a small number
or even no object types. Each time a policy is loaded or unloaded,
we recalculate a mask of labeled object types across all policies
present in the system. Eliminate MAC_ALWAYS_LABEL_MBUF option as it
is no longer required.
MFC after: 1 week ((1) only)
Reviewed by: csjp
Obtained from: TrustedBSD Project
Sponsored by: Apple, Inc.
2008-08-23 15:26:36 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_VNODE))
|
|
|
|
return (EINVAL);
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
error = mac_check_structmac_consistent(&mac);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
elements = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, elements, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK | M_ZERO);
|
2012-10-22 17:50:54 +00:00
|
|
|
NDINIT(&nd, LOOKUP, LOCKLEAF | NOFOLLOW, UIO_USERSPACE,
|
2005-10-04 14:32:58 +00:00
|
|
|
uap->path_p, td);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = namei(&nd);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
|
|
|
goto out;
|
|
|
|
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_vnode_label_alloc();
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_vnode_copy_label(nd.ni_vp->v_label, intlabel);
|
|
|
|
error = mac_vnode_externalize_label(intlabel, elements, buffer,
|
2003-11-06 03:42:43 +00:00
|
|
|
mac.m_buflen);
|
2002-10-22 14:29:47 +00:00
|
|
|
NDFREE(&nd, 0);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_vnode_label_free(intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
|
|
|
|
if (error == 0)
|
|
|
|
error = copyout(buffer, mac.m_string, strlen(buffer)+1);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
|
|
|
out:
|
2002-10-22 14:29:47 +00:00
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
free(elements, M_MACTEMP);
|
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_fd(struct thread *td, struct __mac_set_fd_args *uap)
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
{
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *intlabel;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct pipe *pipe;
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
struct socket *so;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct file *fp;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
struct mount *mp;
|
|
|
|
struct vnode *vp;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct mac mac;
|
Change the cap_rights_t type from uint64_t to a structure that we can extend
in the future in a backward compatible (API and ABI) way.
The cap_rights_t represents capability rights. We used to use one bit to
represent one right, but we are running out of spare bits. Currently the new
structure provides place for 114 rights (so 50 more than the previous
cap_rights_t), but it is possible to grow the structure to hold at least 285
rights, although we can make it even larger if 285 rights won't be enough.
The structure definition looks like this:
struct cap_rights {
uint64_t cr_rights[CAP_RIGHTS_VERSION + 2];
};
The initial CAP_RIGHTS_VERSION is 0.
The top two bits in the first element of the cr_rights[] array contain total
number of elements in the array - 2. This means if those two bits are equal to
0, we have 2 array elements.
The top two bits in all remaining array elements should be 0.
The next five bits in all array elements contain array index. Only one bit is
used and bit position in this five-bits range defines array index. This means
there can be at most five array elements in the future.
To define new right the CAPRIGHT() macro must be used. The macro takes two
arguments - an array index and a bit to set, eg.
#define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL)
We still support aliases that combine few rights, but the rights have to belong
to the same array element, eg:
#define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL)
#define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL)
#define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP)
There is new API to manage the new cap_rights_t structure:
cap_rights_t *cap_rights_init(cap_rights_t *rights, ...);
void cap_rights_set(cap_rights_t *rights, ...);
void cap_rights_clear(cap_rights_t *rights, ...);
bool cap_rights_is_set(const cap_rights_t *rights, ...);
bool cap_rights_is_valid(const cap_rights_t *rights);
void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src);
void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src);
bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little);
Capability rights to the cap_rights_init(), cap_rights_set(),
cap_rights_clear() and cap_rights_is_set() functions are provided by
separating them with commas, eg:
cap_rights_t rights;
cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT);
There is no need to terminate the list of rights, as those functions are
actually macros that take care of the termination, eg:
#define cap_rights_set(rights, ...) \
__cap_rights_set((rights), __VA_ARGS__, 0ULL)
void __cap_rights_set(cap_rights_t *rights, ...);
Thanks to using one bit as an array index we can assert in those functions that
there are no two rights belonging to different array elements provided
together. For example this is illegal and will be detected, because CAP_LOOKUP
belongs to element 0 and CAP_PDKILL to element 1:
cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL);
Providing several rights that belongs to the same array's element this way is
correct, but is not advised. It should only be used for aliases definition.
This commit also breaks compatibility with some existing Capsicum system calls,
but I see no other way to do that. This should be fine as Capsicum is still
experimental and this change is not going to 9.x.
Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
|
|
|
cap_rights_t rights;
|
2002-10-22 14:29:47 +00:00
|
|
|
char *buffer;
|
2012-10-22 17:50:54 +00:00
|
|
|
int error;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = mac_check_structmac_consistent(&mac);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, buffer, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
Change the cap_rights_t type from uint64_t to a structure that we can extend
in the future in a backward compatible (API and ABI) way.
The cap_rights_t represents capability rights. We used to use one bit to
represent one right, but we are running out of spare bits. Currently the new
structure provides place for 114 rights (so 50 more than the previous
cap_rights_t), but it is possible to grow the structure to hold at least 285
rights, although we can make it even larger if 285 rights won't be enough.
The structure definition looks like this:
struct cap_rights {
uint64_t cr_rights[CAP_RIGHTS_VERSION + 2];
};
The initial CAP_RIGHTS_VERSION is 0.
The top two bits in the first element of the cr_rights[] array contain total
number of elements in the array - 2. This means if those two bits are equal to
0, we have 2 array elements.
The top two bits in all remaining array elements should be 0.
The next five bits in all array elements contain array index. Only one bit is
used and bit position in this five-bits range defines array index. This means
there can be at most five array elements in the future.
To define new right the CAPRIGHT() macro must be used. The macro takes two
arguments - an array index and a bit to set, eg.
#define CAP_PDKILL CAPRIGHT(1, 0x0000000000000800ULL)
We still support aliases that combine few rights, but the rights have to belong
to the same array element, eg:
#define CAP_LOOKUP CAPRIGHT(0, 0x0000000000000400ULL)
#define CAP_FCHMOD CAPRIGHT(0, 0x0000000000002000ULL)
#define CAP_FCHMODAT (CAP_FCHMOD | CAP_LOOKUP)
There is new API to manage the new cap_rights_t structure:
cap_rights_t *cap_rights_init(cap_rights_t *rights, ...);
void cap_rights_set(cap_rights_t *rights, ...);
void cap_rights_clear(cap_rights_t *rights, ...);
bool cap_rights_is_set(const cap_rights_t *rights, ...);
bool cap_rights_is_valid(const cap_rights_t *rights);
void cap_rights_merge(cap_rights_t *dst, const cap_rights_t *src);
void cap_rights_remove(cap_rights_t *dst, const cap_rights_t *src);
bool cap_rights_contains(const cap_rights_t *big, const cap_rights_t *little);
Capability rights to the cap_rights_init(), cap_rights_set(),
cap_rights_clear() and cap_rights_is_set() functions are provided by
separating them with commas, eg:
cap_rights_t rights;
cap_rights_init(&rights, CAP_READ, CAP_WRITE, CAP_FSTAT);
There is no need to terminate the list of rights, as those functions are
actually macros that take care of the termination, eg:
#define cap_rights_set(rights, ...) \
__cap_rights_set((rights), __VA_ARGS__, 0ULL)
void __cap_rights_set(cap_rights_t *rights, ...);
Thanks to using one bit as an array index we can assert in those functions that
there are no two rights belonging to different array elements provided
together. For example this is illegal and will be detected, because CAP_LOOKUP
belongs to element 0 and CAP_PDKILL to element 1:
cap_rights_init(&rights, CAP_LOOKUP | CAP_PDKILL);
Providing several rights that belongs to the same array's element this way is
correct, but is not advised. It should only be used for aliases definition.
This commit also breaks compatibility with some existing Capsicum system calls,
but I see no other way to do that. This should be fine as Capsicum is still
experimental and this change is not going to 9.x.
Sponsored by: The FreeBSD Foundation
2013-09-05 00:09:56 +00:00
|
|
|
error = fget(td, uap->fd, cap_rights_init(&rights, CAP_MAC_SET), &fp);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
2002-10-22 14:29:47 +00:00
|
|
|
goto out;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
|
|
|
switch (fp->f_type) {
|
|
|
|
case DTYPE_FIFO:
|
|
|
|
case DTYPE_VNODE:
|
2012-04-08 11:01:49 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_VNODE)) {
|
|
|
|
error = EINVAL;
|
|
|
|
goto out_fdrop;
|
|
|
|
}
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_vnode_label_alloc();
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_vnode_internalize_label(intlabel, buffer);
|
2002-10-22 14:29:47 +00:00
|
|
|
if (error) {
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_vnode_label_free(intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
break;
|
|
|
|
}
|
2003-06-22 08:41:43 +00:00
|
|
|
vp = fp->f_vnode;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
error = vn_start_write(vp, &mp, V_WAIT | PCATCH);
|
2002-10-22 14:29:47 +00:00
|
|
|
if (error != 0) {
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_vnode_label_free(intlabel);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
break;
|
2002-10-22 14:29:47 +00:00
|
|
|
}
|
2008-01-10 01:10:58 +00:00
|
|
|
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
error = vn_setlabel(vp, intlabel, td->td_ucred);
|
2008-01-13 14:44:15 +00:00
|
|
|
VOP_UNLOCK(vp, 0);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
vn_finished_write(mp);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_vnode_label_free(intlabel);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
break;
|
2002-10-22 14:29:47 +00:00
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
case DTYPE_PIPE:
|
2012-04-08 11:01:49 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_PIPE)) {
|
|
|
|
error = EINVAL;
|
|
|
|
goto out_fdrop;
|
|
|
|
}
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_pipe_label_alloc();
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_pipe_internalize_label(intlabel, buffer);
|
2002-10-22 14:29:47 +00:00
|
|
|
if (error == 0) {
|
2003-01-13 00:33:17 +00:00
|
|
|
pipe = fp->f_data;
|
2002-10-22 14:29:47 +00:00
|
|
|
PIPE_LOCK(pipe);
|
Coalesce pipe allocations and frees. Previously, the pipe code
would allocate two 'struct pipe's from the pipe zone, and malloc a
mutex.
- Create a new "struct pipepair" object holding the two 'struct
pipe' instances, struct mutex, and struct label reference. Pipe
structures now have a back-pointer to the pipe pair, and a
'pipe_present' flag to indicate whether the half has been
closed.
- Perform mutex init/destroy in zone init/destroy, avoiding
reallocating the mutex for each pipe. Perform most pipe structure
setup in zone constructor.
- VM memory mappings for pageable buffers are still done outside of
the UMA zone.
- Change MAC API to speak 'struct pipepair' instead of 'struct pipe',
update many policies. MAC labels are also handled outside of the
UMA zone for now. Label-only policy modules don't have to be
recompiled, but if a module is recompiled, its pipe entry points
will need to be updated. If a module actually reached into the
pipe structures (unlikely), that would also need to be modified.
These changes substantially simplify failure handling in the pipe
code as there are many fewer possible failure modes.
On half-close, pipes no longer free the 'struct pipe' for the closed
half until a full-close takes place. However, VM mapped buffers
are still released on half-close.
Some code refactoring is now possible to clean up some of the back
references, etc; this patch attempts not to change the structure
of most of the pipe implementation, only allocation/free code
paths, so as to avoid introducing bugs (hopefully).
This cuts about 8%-9% off the cost of sequential pipe allocation
and free in system call tests on UP and SMP in my micro-benchmarks.
May or may not make a difference in macro-benchmarks, but doing
less work is good.
Reviewed by: juli, tjr
Testing help: dwhite, fenestro, scottl, et al
2004-02-01 05:56:51 +00:00
|
|
|
error = mac_pipe_label_set(td->td_ucred,
|
|
|
|
pipe->pipe_pair, intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
PIPE_UNLOCK(pipe);
|
|
|
|
}
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
mac_pipe_label_free(intlabel);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
break;
|
2002-10-22 14:29:47 +00:00
|
|
|
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
case DTYPE_SOCKET:
|
2012-04-08 11:01:49 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_SOCKET)) {
|
|
|
|
error = EINVAL;
|
|
|
|
goto out_fdrop;
|
|
|
|
}
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
intlabel = mac_socket_label_alloc(M_WAITOK);
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_socket_internalize_label(intlabel, buffer);
|
Implement sockets support for __mac_get_fd() and __mac_set_fd()
system calls, and prefer these calls over getsockopt()/setsockopt()
for ABI reasons. When addressing UNIX domain sockets, these calls
retrieve and modify the socket label, not the label of the
rendezvous vnode.
- Create mac_copy_socket_label() entry point based on
mac_copy_pipe_label() entry point, intended to copy the socket
label into temporary storage that doesn't require a socket lock
to be held (currently Giant).
- Implement mac_copy_socket_label() for various policies.
- Expose socket label allocation, free, internalize, externalize
entry points as non-static from mac_net.c.
- Use mac_socket_label_set() in __mac_set_fd().
MAC-aware applications may now use mac_get_fd(), mac_set_fd(), and
mac_get_peer() to retrieve and set various socket labels without
directly invoking the getsockopt() interface.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-16 23:31:45 +00:00
|
|
|
if (error == 0) {
|
|
|
|
so = fp->f_data;
|
|
|
|
error = mac_socket_label_set(td->td_ucred, so,
|
|
|
|
intlabel);
|
|
|
|
}
|
|
|
|
mac_socket_label_free(intlabel);
|
|
|
|
break;
|
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
default:
|
|
|
|
error = EINVAL;
|
|
|
|
}
|
2012-04-08 11:01:49 +00:00
|
|
|
out_fdrop:
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
fdrop(fp, td);
|
2002-10-22 14:29:47 +00:00
|
|
|
out:
|
|
|
|
free(buffer, M_MACTEMP);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_file(struct thread *td, struct __mac_set_file_args *uap)
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
{
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *intlabel;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct nameidata nd;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
struct mount *mp;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct mac mac;
|
|
|
|
char *buffer;
|
2012-10-22 17:50:54 +00:00
|
|
|
int error;
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
Introduce two related changes to the TrustedBSD MAC Framework:
(1) Abstract interpreter vnode labeling in execve(2) and mac_execve(2)
so that the general exec code isn't aware of the details of
allocating, copying, and freeing labels, rather, simply passes in
a void pointer to start and stop functions that will be used by
the framework. This change will be MFC'd.
(2) Introduce a new flags field to the MAC_POLICY_SET(9) interface
allowing policies to declare which types of objects require label
allocation, initialization, and destruction, and define a set of
flags covering various supported object types (MPC_OBJECT_PROC,
MPC_OBJECT_VNODE, MPC_OBJECT_INPCB, ...). This change reduces the
overhead of compiling the MAC Framework into the kernel if policies
aren't loaded, or if policies require labels on only a small number
or even no object types. Each time a policy is loaded or unloaded,
we recalculate a mask of labeled object types across all policies
present in the system. Eliminate MAC_ALWAYS_LABEL_MBUF option as it
is no longer required.
MFC after: 1 week ((1) only)
Reviewed by: csjp
Obtained from: TrustedBSD Project
Sponsored by: Apple, Inc.
2008-08-23 15:26:36 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_VNODE))
|
|
|
|
return (EINVAL);
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = mac_check_structmac_consistent(&mac);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, buffer, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_vnode_label_alloc();
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_vnode_internalize_label(intlabel, buffer);
|
2002-10-22 14:29:47 +00:00
|
|
|
free(buffer, M_MACTEMP);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
if (error)
|
|
|
|
goto out;
|
2002-10-22 14:29:47 +00:00
|
|
|
|
2012-10-22 17:50:54 +00:00
|
|
|
NDINIT(&nd, LOOKUP, LOCKLEAF | FOLLOW, UIO_USERSPACE,
|
2005-10-04 14:32:58 +00:00
|
|
|
uap->path_p, td);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
error = namei(&nd);
|
2002-10-22 14:29:47 +00:00
|
|
|
if (error == 0) {
|
|
|
|
error = vn_start_write(nd.ni_vp, &mp, V_WAIT | PCATCH);
|
2006-03-19 20:43:07 +00:00
|
|
|
if (error == 0) {
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
error = vn_setlabel(nd.ni_vp, intlabel,
|
2002-10-22 14:29:47 +00:00
|
|
|
td->td_ucred);
|
2006-03-19 20:43:07 +00:00
|
|
|
vn_finished_write(mp);
|
|
|
|
}
|
2002-10-22 14:29:47 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
NDFREE(&nd, 0);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
out:
|
|
|
|
mac_vnode_label_free(intlabel);
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_link(struct thread *td, struct __mac_set_link_args *uap)
|
2002-10-22 14:29:47 +00:00
|
|
|
{
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *intlabel;
|
2002-10-22 14:29:47 +00:00
|
|
|
struct nameidata nd;
|
|
|
|
struct mount *mp;
|
|
|
|
struct mac mac;
|
|
|
|
char *buffer;
|
2012-10-22 17:50:54 +00:00
|
|
|
int error;
|
2002-10-22 14:29:47 +00:00
|
|
|
|
Introduce two related changes to the TrustedBSD MAC Framework:
(1) Abstract interpreter vnode labeling in execve(2) and mac_execve(2)
so that the general exec code isn't aware of the details of
allocating, copying, and freeing labels, rather, simply passes in
a void pointer to start and stop functions that will be used by
the framework. This change will be MFC'd.
(2) Introduce a new flags field to the MAC_POLICY_SET(9) interface
allowing policies to declare which types of objects require label
allocation, initialization, and destruction, and define a set of
flags covering various supported object types (MPC_OBJECT_PROC,
MPC_OBJECT_VNODE, MPC_OBJECT_INPCB, ...). This change reduces the
overhead of compiling the MAC Framework into the kernel if policies
aren't loaded, or if policies require labels on only a small number
or even no object types. Each time a policy is loaded or unloaded,
we recalculate a mask of labeled object types across all policies
present in the system. Eliminate MAC_ALWAYS_LABEL_MBUF option as it
is no longer required.
MFC after: 1 week ((1) only)
Reviewed by: csjp
Obtained from: TrustedBSD Project
Sponsored by: Apple, Inc.
2008-08-23 15:26:36 +00:00
|
|
|
if (!(mac_labeled & MPC_OBJECT_VNODE))
|
|
|
|
return (EINVAL);
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyin(uap->mac_p, &mac, sizeof(mac));
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
|
|
|
|
|
|
|
error = mac_check_structmac_consistent(&mac);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
if (error)
|
2002-10-22 14:29:47 +00:00
|
|
|
return (error);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
buffer = malloc(mac.m_buflen, M_MACTEMP, M_WAITOK);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = copyinstr(mac.m_string, buffer, mac.m_buflen, NULL);
|
|
|
|
if (error) {
|
|
|
|
free(buffer, M_MACTEMP);
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
intlabel = mac_vnode_label_alloc();
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_vnode_internalize_label(intlabel, buffer);
|
2002-10-22 14:29:47 +00:00
|
|
|
free(buffer, M_MACTEMP);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
if (error)
|
|
|
|
goto out;
|
2002-10-22 14:29:47 +00:00
|
|
|
|
2012-10-22 17:50:54 +00:00
|
|
|
NDINIT(&nd, LOOKUP, LOCKLEAF | NOFOLLOW, UIO_USERSPACE,
|
2005-10-04 14:32:58 +00:00
|
|
|
uap->path_p, td);
|
2002-10-22 14:29:47 +00:00
|
|
|
error = namei(&nd);
|
|
|
|
if (error == 0) {
|
|
|
|
error = vn_start_write(nd.ni_vp, &mp, V_WAIT | PCATCH);
|
2006-03-19 20:43:07 +00:00
|
|
|
if (error == 0) {
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
error = vn_setlabel(nd.ni_vp, intlabel,
|
2002-10-22 14:29:47 +00:00
|
|
|
td->td_ucred);
|
2006-03-19 20:43:07 +00:00
|
|
|
vn_finished_write(mp);
|
|
|
|
}
|
2002-10-22 14:29:47 +00:00
|
|
|
}
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
|
|
|
NDFREE(&nd, 0);
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
out:
|
|
|
|
mac_vnode_label_free(intlabel);
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
2002-08-19 17:59:48 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys_mac_syscall(struct thread *td, struct mac_syscall_args *uap)
|
2002-08-19 17:59:48 +00:00
|
|
|
{
|
|
|
|
struct mac_policy_conf *mpc;
|
|
|
|
char target[MAC_MAX_POLICY_NAME];
|
Rework MAC Framework synchronization in a number of ways in order to
improve performance:
- Eliminate custom reference count and condition variable to monitor
threads entering the framework, as this had both significant overhead
and behaved badly in the face of contention.
- Replace reference count with two locks: an rwlock and an sx lock,
which will be read-acquired by threads entering the framework
depending on whether a give policy entry point is permitted to sleep
or not.
- Replace previous mutex locking of the reference count for exclusive
access with write acquiring of both the policy list sx and rw locks,
which occurs only when policies are attached or detached.
- Do a lockless read of the dynamic policy list head before acquiring
any locks in order to reduce overhead when no dynamic policies are
loaded; this a race we can afford to lose.
- For every policy entry point invocation, decide whether sleeping is
permitted, and if not, use a _NOSLEEP() variant of the composition
macros, which will use the rwlock instead of the sxlock. In some
cases, we decide which to use based on allocation flags passed to the
MAC Framework entry point.
As with the move to rwlocks/rmlocks in pfil, this may trigger witness
warnings, but these should (generally) be false positives as all
acquisition of the locks is for read with two very narrow exceptions
for policy load/unload, and those code blocks should never acquire
other locks.
Sponsored by: Google, Inc.
Obtained from: TrustedBSD Project
Discussed with: csjp (idea, not specific patch)
2009-03-14 16:06:06 +00:00
|
|
|
int error;
|
2002-08-19 17:59:48 +00:00
|
|
|
|
2002-12-14 01:56:26 +00:00
|
|
|
error = copyinstr(uap->policy, target, sizeof(target), NULL);
|
2002-08-19 17:59:48 +00:00
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
error = ENOSYS;
|
2003-08-26 17:29:02 +00:00
|
|
|
LIST_FOREACH(mpc, &mac_static_policy_list, mpc_list) {
|
2002-08-19 17:59:48 +00:00
|
|
|
if (strcmp(mpc->mpc_name, target) == 0 &&
|
|
|
|
mpc->mpc_ops->mpo_syscall != NULL) {
|
|
|
|
error = mpc->mpc_ops->mpo_syscall(td,
|
2002-12-14 01:56:26 +00:00
|
|
|
uap->call, uap->arg);
|
2002-08-19 17:59:48 +00:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Rework MAC Framework synchronization in a number of ways in order to
improve performance:
- Eliminate custom reference count and condition variable to monitor
threads entering the framework, as this had both significant overhead
and behaved badly in the face of contention.
- Replace reference count with two locks: an rwlock and an sx lock,
which will be read-acquired by threads entering the framework
depending on whether a give policy entry point is permitted to sleep
or not.
- Replace previous mutex locking of the reference count for exclusive
access with write acquiring of both the policy list sx and rw locks,
which occurs only when policies are attached or detached.
- Do a lockless read of the dynamic policy list head before acquiring
any locks in order to reduce overhead when no dynamic policies are
loaded; this a race we can afford to lose.
- For every policy entry point invocation, decide whether sleeping is
permitted, and if not, use a _NOSLEEP() variant of the composition
macros, which will use the rwlock instead of the sxlock. In some
cases, we decide which to use based on allocation flags passed to the
MAC Framework entry point.
As with the move to rwlocks/rmlocks in pfil, this may trigger witness
warnings, but these should (generally) be false positives as all
acquisition of the locks is for read with two very narrow exceptions
for policy load/unload, and those code blocks should never acquire
other locks.
Sponsored by: Google, Inc.
Obtained from: TrustedBSD Project
Discussed with: csjp (idea, not specific patch)
2009-03-14 16:06:06 +00:00
|
|
|
if (!LIST_EMPTY(&mac_policy_list)) {
|
|
|
|
mac_policy_slock_sleep();
|
Clean up locking for the MAC Framework:
(1) Accept that we're now going to use mutexes, so don't attempt
to avoid treating them as mutexes. This cleans up locking
accessor function names some.
(2) Rename variables to _mtx, _cv, _count, simplifying the naming.
(3) Add a new form of the _busy() primitive that conditionally
makes the list busy: if there are entries on the list, bump
the busy count. If there are no entries, don't bump the busy
count. Return a boolean indicating whether or not the busy
count was bumped.
(4) Break mac_policy_list into two lists: one with the same name
holding dynamic policies, and a new list, mac_static_policy_list,
which holds policies loaded before mac_late and without the
unload flag set. The static list may be accessed without
holding the busy count, since it can't change at run-time.
(5) In general, prefer making the list busy conditionally, meaning
we pay only one mutex lock per entry point if all modules are
on the static list, rather than two (since we don't have to
lower the busy count when we're done with the framework). For
systems running just Biba or MLS, this will halve the mutex
accesses in the network stack, and may offer a substantial
performance benefits.
(6) Lay the groundwork for a dynamic-free kernel option which
eliminates all locking associated with dynamically loaded or
unloaded policies, for pre-configured systems requiring
maximum performance but less run-time flexibility.
These changes have been running for a few weeks on MAC development
branch systems.
Approved by: re (jhb)
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-05-07 17:49:24 +00:00
|
|
|
LIST_FOREACH(mpc, &mac_policy_list, mpc_list) {
|
|
|
|
if (strcmp(mpc->mpc_name, target) == 0 &&
|
|
|
|
mpc->mpc_ops->mpo_syscall != NULL) {
|
|
|
|
error = mpc->mpc_ops->mpo_syscall(td,
|
|
|
|
uap->call, uap->arg);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
Rework MAC Framework synchronization in a number of ways in order to
improve performance:
- Eliminate custom reference count and condition variable to monitor
threads entering the framework, as this had both significant overhead
and behaved badly in the face of contention.
- Replace reference count with two locks: an rwlock and an sx lock,
which will be read-acquired by threads entering the framework
depending on whether a give policy entry point is permitted to sleep
or not.
- Replace previous mutex locking of the reference count for exclusive
access with write acquiring of both the policy list sx and rw locks,
which occurs only when policies are attached or detached.
- Do a lockless read of the dynamic policy list head before acquiring
any locks in order to reduce overhead when no dynamic policies are
loaded; this a race we can afford to lose.
- For every policy entry point invocation, decide whether sleeping is
permitted, and if not, use a _NOSLEEP() variant of the composition
macros, which will use the rwlock instead of the sxlock. In some
cases, we decide which to use based on allocation flags passed to the
MAC Framework entry point.
As with the move to rwlocks/rmlocks in pfil, this may trigger witness
warnings, but these should (generally) be false positives as all
acquisition of the locks is for read with two very narrow exceptions
for policy load/unload, and those code blocks should never acquire
other locks.
Sponsored by: Google, Inc.
Obtained from: TrustedBSD Project
Discussed with: csjp (idea, not specific patch)
2009-03-14 16:06:06 +00:00
|
|
|
mac_policy_sunlock_sleep();
|
Clean up locking for the MAC Framework:
(1) Accept that we're now going to use mutexes, so don't attempt
to avoid treating them as mutexes. This cleans up locking
accessor function names some.
(2) Rename variables to _mtx, _cv, _count, simplifying the naming.
(3) Add a new form of the _busy() primitive that conditionally
makes the list busy: if there are entries on the list, bump
the busy count. If there are no entries, don't bump the busy
count. Return a boolean indicating whether or not the busy
count was bumped.
(4) Break mac_policy_list into two lists: one with the same name
holding dynamic policies, and a new list, mac_static_policy_list,
which holds policies loaded before mac_late and without the
unload flag set. The static list may be accessed without
holding the busy count, since it can't change at run-time.
(5) In general, prefer making the list busy conditionally, meaning
we pay only one mutex lock per entry point if all modules are
on the static list, rather than two (since we don't have to
lower the busy count when we're done with the framework). For
systems running just Biba or MLS, this will halve the mutex
accesses in the network stack, and may offer a substantial
performance benefits.
(6) Lay the groundwork for a dynamic-free kernel option which
eliminates all locking associated with dynamically loaded or
unloaded policies, for pre-configured systems requiring
maximum performance but less run-time flexibility.
These changes have been running for a few weeks on MAC development
branch systems.
Approved by: re (jhb)
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-05-07 17:49:24 +00:00
|
|
|
}
|
2002-08-19 17:59:48 +00:00
|
|
|
out:
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
#else /* !MAC */
|
2002-07-30 02:04:05 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_pid(struct thread *td, struct __mac_get_pid_args *uap)
|
2002-10-22 14:29:47 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
2002-07-30 02:04:05 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_proc(struct thread *td, struct __mac_get_proc_args *uap)
|
2002-07-30 02:04:05 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_proc(struct thread *td, struct __mac_set_proc_args *uap)
|
2002-07-30 02:04:05 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_fd(struct thread *td, struct __mac_get_fd_args *uap)
|
2002-07-30 02:04:05 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_file(struct thread *td, struct __mac_get_file_args *uap)
|
2002-07-30 02:04:05 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_get_link(struct thread *td, struct __mac_get_link_args *uap)
|
2002-10-22 14:29:47 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
2002-07-30 02:04:05 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_fd(struct thread *td, struct __mac_set_fd_args *uap)
|
2002-07-30 02:04:05 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_file(struct thread *td, struct __mac_set_file_args *uap)
|
2002-07-30 02:04:05 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
Begin committing support for Mandatory Access Control and extensible
kernel access control. The MAC framework permits loadable kernel
modules to link to the kernel at compile-time, boot-time, or run-time,
and augment the system security policy. This commit includes the
initial kernel implementation, although the interface with the userland
components of the oeprating system is still under work, and not all
kernel subsystems are supported. Later in this commit sequence,
documentation of which kernel subsystems will not work correctly with
a kernel compiled with MAC support will be added.
kern_mac.c contains the body of the MAC framework. Kernel and
user APIs defined in mac.h are implemented here, providing a front end
to loaded security modules. This code implements a module registration
service, state (label) management, security configuration and policy
composition.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
2002-07-30 21:36:05 +00:00
|
|
|
|
2002-10-22 14:29:47 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys___mac_set_link(struct thread *td, struct __mac_set_link_args *uap)
|
2002-10-22 14:29:47 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
2002-08-19 17:59:48 +00:00
|
|
|
int
|
2011-09-16 13:58:51 +00:00
|
|
|
sys_mac_syscall(struct thread *td, struct mac_syscall_args *uap)
|
2002-08-19 17:59:48 +00:00
|
|
|
{
|
|
|
|
|
|
|
|
return (ENOSYS);
|
|
|
|
}
|
|
|
|
|
2004-05-03 22:54:46 +00:00
|
|
|
#endif /* !MAC */
|