d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
fea9ea0005
freebsd-dev/sys/security/mac_mls/mac_mls.c

3013 lines
70 KiB
C
Raw Normal View History

Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*-
Update my personal copyrights and NETA copyrights in the kernel to use the "year1-year3" format, as opposed to "year1, year2, year3". This seems to make lawyers more happy, but also prevents the lines from getting excessively long as the years start to add up. Suggested by: imp
2004-02-22 00:33:12 +00:00
* Copyright (c) 1999-2002 Robert N. M. Watson
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
* Copyright (c) 2001-2005 McAfee, Inc.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
* All rights reserved.
*
* This software was developed by Robert Watson for the TrustedBSD Project.
*
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
* This software was developed for the FreeBSD Project in part by McAfee
* Research, the Security Research Division of McAfee, Inc. under
* DARPA/SPAWAR contract N66001-01-C-8035 ("CBOSS"), as part of the DARPA
* CHATS research program.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +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.
*
* $FreeBSD$
*/
/*
* Developed by the TrustedBSD Project.
* MLS fixed label mandatory confidentiality policy.
*/
#include <sys/types.h>
#include <sys/param.h>
#include <sys/acl.h>
#include <sys/conf.h>
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
#include <sys/extattr.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <sys/kernel.h>
Merge posix4/* into normal kernel hierarchy. Reviewed by: glanced at by jhb Approved by: silence on -arch@ and -standards@
2006-11-11 16:26:58 +00:00
#include <sys/ksem.h>
Move MAC check_vnode_mmap entry point out from being exclusive to MAP_SHARED so that the entry point gets executed un-conditionally. This may be useful for security policies which want to perform access control checks around run-time linking. -add the mmap(2) flags argument to the check_vnode_mmap entry point so that we can make access control decisions based on the type of mapped object. -update any dependent API around this parameter addition such as function prototype modifications, entry point parameter additions and the inclusion of sys/mman.h header file. -Change the MLS, BIBA and LOMAC security policies so that subject domination routines are not executed unless the type of mapping is shared. This is done to maintain compatibility between the old vm_mmap_vnode(9) and these policies. Reviewed by: rwatson MFC after: 1 month
2005-04-14 16:03:30 +00:00
#include <sys/mman.h>
Include <sys/malloc.h> instead of depending on namespace pollution 2 layers deep in <sys/proc.h> or <sys/vnode.h>.
2002-09-10 12:45:02 +00:00
#include <sys/malloc.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <sys/mount.h>
#include <sys/proc.h>
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
#include <sys/sbuf.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <sys/systm.h>
#include <sys/sysproto.h>
#include <sys/sysent.h>
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
#include <sys/systm.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <sys/vnode.h>
#include <sys/file.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/pipe.h>
Add #include <sys/sx.h>, devfs is going to require this shortly.
2005-09-19 18:52:51 +00:00
#include <sys/sx.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <sys/sysctl.h>
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
#include <sys/msg.h>
#include <sys/sem.h>
#include <sys/shm.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <fs/devfs/devfs.h>
#include <net/bpfdesc.h>
#include <net/if.h>
#include <net/if_types.h>
#include <net/if_var.h>
#include <netinet/in.h>
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
#include <netinet/in_pcb.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <netinet/ip_var.h>
Use UMA zone allocator for Biba and MLS labels rather than MALLOC(9). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 04:11:52 +00:00
#include <vm/uma.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <vm/vm.h>
Move src/sys/sys/mac_policy.h, the kernel interface between the MAC Framework and security modules, to src/sys/security/mac/mac_policy.h, completing the removal of kernel-only MAC Framework include files from src/sys/sys. Update the MAC Framework and MAC policy modules. Delete the old mac_policy.h. Third party policy modules will need similar updating. Obtained from: TrustedBSD Project
2006-12-22 23:34:47 +00:00
#include <security/mac/mac_policy.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
#include <security/mac_mls/mac_mls.h>
SYSCTL_DECL(_security_mac);
SYSCTL_NODE(_security_mac, OID_AUTO, mls, CTLFLAG_RW, 0,
"TrustedBSD mac_mls policy controls");
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
static int mac_mls_label_size = sizeof(struct mac_mls);
SYSCTL_INT(_security_mac_mls, OID_AUTO, label_size, CTLFLAG_RD,
&mac_mls_label_size, 0, "Size of struct mac_mls");
Default policies to on: if you load them or compile them into your kernel, you should expect them to do something, so now they do. This doesn't affect users who don't load or explicitly compile in the policies. Approved by: re (jhb) Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-12-10 16:20:34 +00:00
static int mac_mls_enabled = 1;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
SYSCTL_INT(_security_mac_mls, OID_AUTO, enabled, CTLFLAG_RW,
&mac_mls_enabled, 0, "Enforce MAC/MLS policy");
Re-insert TUNABLE() export of MAC Biba and MAC MLS module 'enabled' sysctls. Apparently lost during some or another merge. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-09-05 18:52:52 +00:00
TUNABLE_INT("security.mac.mls.enabled", &mac_mls_enabled);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int destroyed_not_inited;
SYSCTL_INT(_security_mac_mls, OID_AUTO, destroyed_not_inited, CTLFLAG_RD,
&destroyed_not_inited, 0, "Count of labels destroyed but not inited");
Add a twiddle to create PTY's with a biba/equal or mls/equal label instead of the default biba/high, mls/low, making it easier to use ptys with these policies. This isn't the final solution, but does help. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 04:15:40 +00:00
static int ptys_equal = 0;
SYSCTL_INT(_security_mac_mls, OID_AUTO, ptys_equal, CTLFLAG_RW,
&ptys_equal, 0, "Label pty devices as mls/equal on create");
TUNABLE_INT("security.mac.mls.ptys_equal", &ptys_equal);
More in the way of minor consistency improvements: trim 'mac_mls_' from another variable to line this up with mac_biba.c Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:14:30 +00:00
static int revocation_enabled = 0;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
SYSCTL_INT(_security_mac_mls, OID_AUTO, revocation_enabled, CTLFLAG_RW,
More in the way of minor consistency improvements: trim 'mac_mls_' from another variable to line this up with mac_biba.c Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:14:30 +00:00
&revocation_enabled, 0, "Revoke access to objects on relabel");
TUNABLE_INT("security.mac.mls.revocation_enabled", &revocation_enabled);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
static int max_compartments = MAC_MLS_MAX_COMPARTMENTS;
SYSCTL_INT(_security_mac_mls, OID_AUTO, max_compartments, CTLFLAG_RD,
&max_compartments, 0, "Maximum compartments the policy supports");
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int mac_mls_slot;
Introduce accessor functions mac_label_get() and mac_label_set() to replace LABEL_TO_SLOT() macro used by policy modules to query and set label data in struct label. Instead of using a union, store an intptr_t, simplifying the API. Update policies: in most cases this required only small tweaks to current wrapper macros. In two cases, a single wrapper macros had to be split into separate get and set macros. Move struct label definition from _label.h to mac_internal.h and remove _label.h. With this change, policies may now treat struct label * as opaque, allowing us to change the layout of struct label without breaking the policy module ABI. For example, we could make the maximum number of policies with labels modifiable at boot-time rather than just at compile-time. Obtained from: TrustedBSD Project
2007-02-06 14:19:25 +00:00
#define SLOT(l) ((struct mac_mls *)mac_label_get((l), mac_mls_slot))
#define SLOT_SET(l, val) mac_label_set((l), mac_mls_slot, (uintptr_t)(val))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Use UMA zone allocator for Biba and MLS labels rather than MALLOC(9). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 04:11:52 +00:00
static uma_zone_t zone_mls;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
static __inline int
mls_bit_set_empty(u_char *set) {
int i;
for (i = 0; i < MAC_MLS_MAX_COMPARTMENTS >> 3; i++)
if (set[i] != 0)
return (0);
return (1);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static struct mac_mls *
Begin another merge from the TrustedBSD MAC branch: - Change mpo_init_foo(obj, label) and mpo_destroy_foo(obj, label) policy entry points to mpo_init_foo_label(label) and mpo_destroy_foo_label(label). This will permit the use of the same entry points for holding temporary type-specific label during internalization and externalization, as well as for caching purposes. - Because of this, break out mpo_{init,destroy}_socket() and mpo_{init,destroy}_mount() into seperate entry points for socket main/peer labels and mount main/fs labels. - Since the prototype for label initialization is the same across almost all entry points, implement these entry points using common implementations for Biba, MLS, and Test, reducing the number of almost identical looking functions. This simplifies policy implementation, as well as preparing us for the merge of the new flexible userland API for managing labels on objects. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 15:10:00 +00:00
mls_alloc(int flag)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Use UMA zone allocator for Biba and MLS labels rather than MALLOC(9). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 04:11:52 +00:00
return (uma_zalloc(zone_mls, flag | M_ZERO));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mls_free(struct mac_mls *mac_mls)
{
if (mac_mls != NULL)
Use UMA zone allocator for Biba and MLS labels rather than MALLOC(9). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 04:11:52 +00:00
uma_zfree(zone_mls, mac_mls);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
else
atomic_add_int(&destroyed_not_inited, 1);
}
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
static int
mls_atmostflags(struct mac_mls *mac_mls, int flags)
{
if ((mac_mls->mm_flags & flags) != mac_mls->mm_flags)
return (EINVAL);
return (0);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_dominate_element(struct mac_mls_element *a,
struct mac_mls_element *b)
{
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
int bit;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Style fix: space between 'switch' and '('.
2002-10-22 19:01:49 +00:00
switch (a->mme_type) {
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
case MAC_MLS_TYPE_EQUAL:
case MAC_MLS_TYPE_HIGH:
return (1);
case MAC_MLS_TYPE_LOW:
switch (b->mme_type) {
case MAC_MLS_TYPE_LEVEL:
case MAC_MLS_TYPE_HIGH:
return (0);
case MAC_MLS_TYPE_EQUAL:
case MAC_MLS_TYPE_LOW:
return (1);
default:
panic("mac_mls_dominate_element: b->mme_type invalid");
}
case MAC_MLS_TYPE_LEVEL:
switch (b->mme_type) {
case MAC_MLS_TYPE_EQUAL:
case MAC_MLS_TYPE_LOW:
return (1);
case MAC_MLS_TYPE_HIGH:
return (0);
case MAC_MLS_TYPE_LEVEL:
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
for (bit = 1; bit <= MAC_MLS_MAX_COMPARTMENTS; bit++)
if (!MAC_MLS_BIT_TEST(bit,
a->mme_compartments) &&
MAC_MLS_BIT_TEST(bit, b->mme_compartments))
return (0);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (a->mme_level >= b->mme_level);
default:
panic("mac_mls_dominate_element: b->mme_type invalid");
}
default:
panic("mac_mls_dominate_element: a->mme_type invalid");
}
return (0);
}
static int
mac_mls_range_in_range(struct mac_mls *rangea, struct mac_mls *rangeb)
{
return (mac_mls_dominate_element(&rangeb->mm_rangehigh,
&rangea->mm_rangehigh) &&
mac_mls_dominate_element(&rangea->mm_rangelow,
&rangeb->mm_rangelow));
}
static int
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_effective_in_range(struct mac_mls *effective, struct mac_mls *range)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
KASSERT((effective->mm_flags & MAC_MLS_FLAG_EFFECTIVE) != 0,
("mac_mls_effective_in_range: a not effective"));
As INVARIANTS isn't supported for code that loads only as a kernel module and is not linked into the base system, two KASSERT's rotted. Fix them by fixing variable names. It would be really nice if opt_global.h was used when building modules as part of a buildkernel. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-09-21 18:29:37 +00:00
KASSERT((range->mm_flags & MAC_MLS_FLAG_RANGE) != 0,
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
("mac_mls_effective_in_range: b not range"));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (mac_mls_dominate_element(&range->mm_rangehigh,
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
&effective->mm_effective) &&
mac_mls_dominate_element(&effective->mm_effective,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
&range->mm_rangelow));
return (1);
}
static int
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_dominate_effective(struct mac_mls *a, struct mac_mls *b)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
KASSERT((a->mm_flags & MAC_MLS_FLAG_EFFECTIVE) != 0,
("mac_mls_dominate_effective: a not effective"));
KASSERT((b->mm_flags & MAC_MLS_FLAG_EFFECTIVE) != 0,
("mac_mls_dominate_effective: b not effective"));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
return (mac_mls_dominate_element(&a->mm_effective, &b->mm_effective));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static int
mac_mls_equal_element(struct mac_mls_element *a, struct mac_mls_element *b)
{
if (a->mme_type == MAC_MLS_TYPE_EQUAL ||
b->mme_type == MAC_MLS_TYPE_EQUAL)
return (1);
return (a->mme_type == b->mme_type && a->mme_level == b->mme_level);
}
static int
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_equal_effective(struct mac_mls *a, struct mac_mls *b)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
KASSERT((a->mm_flags & MAC_MLS_FLAG_EFFECTIVE) != 0,
("mac_mls_equal_effective: a not effective"));
KASSERT((b->mm_flags & MAC_MLS_FLAG_EFFECTIVE) != 0,
("mac_mls_equal_effective: b not effective"));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
return (mac_mls_equal_element(&a->mm_effective, &b->mm_effective));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
static int
mac_mls_contains_equal(struct mac_mls *mac_mls)
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (mac_mls->mm_flags & MAC_MLS_FLAG_EFFECTIVE)
if (mac_mls->mm_effective.mme_type == MAC_MLS_TYPE_EQUAL)
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (1);
if (mac_mls->mm_flags & MAC_MLS_FLAG_RANGE) {
if (mac_mls->mm_rangelow.mme_type == MAC_MLS_TYPE_EQUAL)
return (1);
if (mac_mls->mm_rangehigh.mme_type == MAC_MLS_TYPE_EQUAL)
return (1);
}
return (0);
}
static int
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
mac_mls_subject_privileged(struct mac_mls *mac_mls)
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
{
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
KASSERT((mac_mls->mm_flags & MAC_MLS_FLAGS_BOTH) ==
MAC_MLS_FLAGS_BOTH,
("mac_mls_subject_privileged: subject doesn't have both labels"));
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
/* If the effective is EQUAL, it's ok. */
if (mac_mls->mm_effective.mme_type == MAC_MLS_TYPE_EQUAL)
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (0);
/* If either range endpoint is EQUAL, it's ok. */
if (mac_mls->mm_rangelow.mme_type == MAC_MLS_TYPE_EQUAL ||
mac_mls->mm_rangehigh.mme_type == MAC_MLS_TYPE_EQUAL)
return (0);
/* If the range is low-high, it's ok. */
if (mac_mls->mm_rangelow.mme_type == MAC_MLS_TYPE_LOW &&
mac_mls->mm_rangehigh.mme_type == MAC_MLS_TYPE_HIGH)
return (0);
/* It's not ok. */
return (EPERM);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_valid(struct mac_mls *mac_mls)
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (mac_mls->mm_flags & MAC_MLS_FLAG_EFFECTIVE) {
switch (mac_mls->mm_effective.mme_type) {
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
case MAC_MLS_TYPE_LEVEL:
break;
case MAC_MLS_TYPE_EQUAL:
case MAC_MLS_TYPE_HIGH:
case MAC_MLS_TYPE_LOW:
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (mac_mls->mm_effective.mme_level != 0 ||
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
!MAC_MLS_BIT_SET_EMPTY(
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls->mm_effective.mme_compartments))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EINVAL);
break;
default:
return (EINVAL);
}
} else {
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (mac_mls->mm_effective.mme_type != MAC_MLS_TYPE_UNDEF)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EINVAL);
}
if (mac_mls->mm_flags & MAC_MLS_FLAG_RANGE) {
switch (mac_mls->mm_rangelow.mme_type) {
case MAC_MLS_TYPE_LEVEL:
break;
case MAC_MLS_TYPE_EQUAL:
case MAC_MLS_TYPE_HIGH:
case MAC_MLS_TYPE_LOW:
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
if (mac_mls->mm_rangelow.mme_level != 0 ||
!MAC_MLS_BIT_SET_EMPTY(
mac_mls->mm_rangelow.mme_compartments))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EINVAL);
break;
default:
return (EINVAL);
}
switch (mac_mls->mm_rangehigh.mme_type) {
case MAC_MLS_TYPE_LEVEL:
break;
case MAC_MLS_TYPE_EQUAL:
case MAC_MLS_TYPE_HIGH:
case MAC_MLS_TYPE_LOW:
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
if (mac_mls->mm_rangehigh.mme_level != 0 ||
!MAC_MLS_BIT_SET_EMPTY(
mac_mls->mm_rangehigh.mme_compartments))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EINVAL);
break;
default:
return (EINVAL);
}
if (!mac_mls_dominate_element(&mac_mls->mm_rangehigh,
&mac_mls->mm_rangelow))
return (EINVAL);
} else {
if (mac_mls->mm_rangelow.mme_type != MAC_MLS_TYPE_UNDEF ||
mac_mls->mm_rangehigh.mme_type != MAC_MLS_TYPE_UNDEF)
return (EINVAL);
}
return (0);
}
static void
mac_mls_set_range(struct mac_mls *mac_mls, u_short typelow,
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
u_short levellow, u_char *compartmentslow, u_short typehigh,
u_short levelhigh, u_char *compartmentshigh)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
mac_mls->mm_rangelow.mme_type = typelow;
mac_mls->mm_rangelow.mme_level = levellow;
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
if (compartmentslow != NULL)
memcpy(mac_mls->mm_rangelow.mme_compartments,
compartmentslow,
sizeof(mac_mls->mm_rangelow.mme_compartments));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
mac_mls->mm_rangehigh.mme_type = typehigh;
mac_mls->mm_rangehigh.mme_level = levelhigh;
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
if (compartmentshigh != NULL)
memcpy(mac_mls->mm_rangehigh.mme_compartments,
compartmentshigh,
sizeof(mac_mls->mm_rangehigh.mme_compartments));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
mac_mls->mm_flags |= MAC_MLS_FLAG_RANGE;
}
static void
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_set_effective(struct mac_mls *mac_mls, u_short type, u_short level,
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
u_char *compartments)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls->mm_effective.mme_type = type;
mac_mls->mm_effective.mme_level = level;
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
if (compartments != NULL)
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
memcpy(mac_mls->mm_effective.mme_compartments, compartments,
sizeof(mac_mls->mm_effective.mme_compartments));
mac_mls->mm_flags |= MAC_MLS_FLAG_EFFECTIVE;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_copy_range(struct mac_mls *labelfrom, struct mac_mls *labelto)
{
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
KASSERT((labelfrom->mm_flags & MAC_MLS_FLAG_RANGE) != 0,
("mac_mls_copy_range: labelfrom not range"));
labelto->mm_rangelow = labelfrom->mm_rangelow;
labelto->mm_rangehigh = labelfrom->mm_rangehigh;
labelto->mm_flags |= MAC_MLS_FLAG_RANGE;
}
static void
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(struct mac_mls *labelfrom, struct mac_mls *labelto)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
KASSERT((labelfrom->mm_flags & MAC_MLS_FLAG_EFFECTIVE) != 0,
("mac_mls_copy_effective: labelfrom not effective"));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
labelto->mm_effective = labelfrom->mm_effective;
labelto->mm_flags |= MAC_MLS_FLAG_EFFECTIVE;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Introduce mac_biba_copy() and mac_mls_copy(), which conditionally copy elements of one Biba or MLS label to another based on the flags on the source label element. Use this instead of mac_{biba,mls}_{single,range}() to simplify the existing code, as well as support partial label updates (we don't update if none is requested). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 20:55:39 +00:00
static void
mac_mls_copy(struct mac_mls *source, struct mac_mls *dest)
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (source->mm_flags & MAC_MLS_FLAG_EFFECTIVE)
mac_mls_copy_effective(source, dest);
Introduce mac_biba_copy() and mac_mls_copy(), which conditionally copy elements of one Biba or MLS label to another based on the flags on the source label element. Use this instead of mac_{biba,mls}_{single,range}() to simplify the existing code, as well as support partial label updates (we don't update if none is requested). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 20:55:39 +00:00
if (source->mm_flags & MAC_MLS_FLAG_RANGE)
mac_mls_copy_range(source, dest);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
* Policy module operations.
*/
static void
mac_mls_init(struct mac_policy_conf *conf)
{
Use UMA zone allocator for Biba and MLS labels rather than MALLOC(9). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 04:11:52 +00:00
zone_mls = uma_zcreate("mac_mls", sizeof(struct mac_mls), NULL,
NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
/*
* Label operations.
*/
static void
Begin another merge from the TrustedBSD MAC branch: - Change mpo_init_foo(obj, label) and mpo_destroy_foo(obj, label) policy entry points to mpo_init_foo_label(label) and mpo_destroy_foo_label(label). This will permit the use of the same entry points for holding temporary type-specific label during internalization and externalization, as well as for caching purposes. - Because of this, break out mpo_{init,destroy}_socket() and mpo_{init,destroy}_mount() into seperate entry points for socket main/peer labels and mount main/fs labels. - Since the prototype for label initialization is the same across almost all entry points, implement these entry points using common implementations for Biba, MLS, and Test, reducing the number of almost identical looking functions. This simplifies policy implementation, as well as preparing us for the merge of the new flexible userland API for managing labels on objects. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 15:10:00 +00:00
mac_mls_init_label(struct label *label)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Introduce SLOT_SET macro and use it in place of casts as lvalues.
2004-07-28 07:01:33 +00:00
SLOT_SET(label, mls_alloc(M_WAITOK));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static int
Begin another merge from the TrustedBSD MAC branch: - Change mpo_init_foo(obj, label) and mpo_destroy_foo(obj, label) policy entry points to mpo_init_foo_label(label) and mpo_destroy_foo_label(label). This will permit the use of the same entry points for holding temporary type-specific label during internalization and externalization, as well as for caching purposes. - Because of this, break out mpo_{init,destroy}_socket() and mpo_{init,destroy}_mount() into seperate entry points for socket main/peer labels and mount main/fs labels. - Since the prototype for label initialization is the same across almost all entry points, implement these entry points using common implementations for Biba, MLS, and Test, reducing the number of almost identical looking functions. This simplifies policy implementation, as well as preparing us for the merge of the new flexible userland API for managing labels on objects. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 15:10:00 +00:00
mac_mls_init_label_waitcheck(struct label *label, int flag)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Introduce SLOT_SET macro and use it in place of casts as lvalues.
2004-07-28 07:01:33 +00:00
SLOT_SET(label, mls_alloc(flag));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
if (SLOT(label) == NULL)
return (ENOMEM);
return (0);
}
static void
Begin another merge from the TrustedBSD MAC branch: - Change mpo_init_foo(obj, label) and mpo_destroy_foo(obj, label) policy entry points to mpo_init_foo_label(label) and mpo_destroy_foo_label(label). This will permit the use of the same entry points for holding temporary type-specific label during internalization and externalization, as well as for caching purposes. - Because of this, break out mpo_{init,destroy}_socket() and mpo_{init,destroy}_mount() into seperate entry points for socket main/peer labels and mount main/fs labels. - Since the prototype for label initialization is the same across almost all entry points, implement these entry points using common implementations for Biba, MLS, and Test, reducing the number of almost identical looking functions. This simplifies policy implementation, as well as preparing us for the merge of the new flexible userland API for managing labels on objects. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 15:10:00 +00:00
mac_mls_destroy_label(struct label *label)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
mls_free(SLOT(label));
Introduce SLOT_SET macro and use it in place of casts as lvalues.
2004-07-28 07:01:33 +00:00
SLOT_SET(label, NULL);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
/*
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
* mac_mls_element_to_string() accepts an sbuf and MLS element. It
* converts the MLS element to a string and stores the result in the
* sbuf; if there isn't space in the sbuf, -1 is returned.
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
*/
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
static int
mac_mls_element_to_string(struct sbuf *sb, struct mac_mls_element *element)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
{
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
int i, first;
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
switch (element->mme_type) {
case MAC_MLS_TYPE_HIGH:
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
return (sbuf_printf(sb, "high"));
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
case MAC_MLS_TYPE_LOW:
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
return (sbuf_printf(sb, "low"));
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
case MAC_MLS_TYPE_EQUAL:
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
return (sbuf_printf(sb, "equal"));
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
case MAC_MLS_TYPE_LEVEL:
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
if (sbuf_printf(sb, "%d", element->mme_level) == -1)
return (-1);
first = 1;
for (i = 1; i <= MAC_MLS_MAX_COMPARTMENTS; i++) {
if (MAC_MLS_BIT_TEST(i, element->mme_compartments)) {
if (first) {
if (sbuf_putc(sb, ':') == -1)
return (-1);
if (sbuf_printf(sb, "%d", i) == -1)
return (-1);
first = 0;
} else {
if (sbuf_printf(sb, "+%d", i) == -1)
return (-1);
}
}
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
}
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
return (0);
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
default:
panic("mac_mls_element_to_string: invalid type (%d)",
element->mme_type);
}
}
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
/*
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
* mac_mls_to_string() converts an MLS label to a string, and places
* the results in the passed sbuf. It returns 0 on success, or EINVAL
* if there isn't room in the sbuf. Note: the sbuf will be modified
* even in a failure case, so the caller may need to revert the sbuf
* by restoring the offset if that's undesired.
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
*/
static int
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
mac_mls_to_string(struct sbuf *sb, struct mac_mls *mac_mls)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (mac_mls->mm_flags & MAC_MLS_FLAG_EFFECTIVE) {
if (mac_mls_element_to_string(sb, &mac_mls->mm_effective)
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
== -1)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
}
if (mac_mls->mm_flags & MAC_MLS_FLAG_RANGE) {
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
if (sbuf_putc(sb, '(') == -1)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
if (mac_mls_element_to_string(sb, &mac_mls->mm_rangelow)
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
== -1)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
if (sbuf_putc(sb, '-') == -1)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
if (mac_mls_element_to_string(sb, &mac_mls->mm_rangehigh)
Rewrite Biba and MLS label externalization code to use sbufs instead of C strings internally; C strings require a lot of return value checking that (a) takes a lot of space, and (b) is difficult to get right. Prior to the advent of compartment support, modeling APIs for helper functions on snprintf worked fine; with the additional complexity, the sbuf_printf() API makes a lot more sense. While doing this, break out the printing of sequential compartment lists into a helper function, mac_{biba,mls}_compartment_to_string(). This permits the main body of mac_{biba,mls}_element_to_string() to be concerned only with identifying sequential ranges rather than rendering. At a less disruptive moment, we'll push the move from snprintf()-like interface to sbuf()-like interface up into the MAC Framework layer. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-05-31 19:01:44 +00:00
== -1)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
if (sbuf_putc(sb, ')') == -1)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
}
return (0);
}
static int
mac_mls_externalize_label(struct label *label, char *element_name,
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
struct sbuf *sb, int *claimed)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
{
struct mac_mls *mac_mls;
if (strcmp(MAC_MLS_LABEL_NAME, element_name) != 0)
return (0);
(*claimed)++;
mac_mls = SLOT(label);
Redesign the externalization APIs from the MAC Framework to the MAC policy modules to improve robustness against C string bugs and vulnerabilities. Following these revisions, all string construction of labels for export to userspace (or elsewhere) is performed using the sbuf API, which prevents the consumer from having to perform laborious and intricate pointer and buffer checks. This substantially simplifies the externalization logic, both at the MAC Framework level, and in individual policies; this becomes especially useful when policies export more complex label data, such as with compartments in Biba and MLS. Bundled in here are some other minor fixes associated with externalization: including avoiding malloc while holding the process mutex in mac_lomac, and hence avoid a failure mode when printing labels during a downgrade operation due to the removal of the M_NOWAIT case. This has been running in the MAC development tree for about three weeks without problems. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-23 01:26:34 +00:00
return (mac_mls_to_string(sb, mac_mls));
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
mac_mls_parse_element(struct mac_mls_element *element, char *string)
{
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
char *compartment, *end, *level;
int value;
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
if (strcmp(string, "high") == 0 ||
strcmp(string, "hi") == 0) {
element->mme_type = MAC_MLS_TYPE_HIGH;
element->mme_level = MAC_MLS_TYPE_UNDEF;
} else if (strcmp(string, "low") == 0 ||
strcmp(string, "lo") == 0) {
element->mme_type = MAC_MLS_TYPE_LOW;
element->mme_level = MAC_MLS_TYPE_UNDEF;
} else if (strcmp(string, "equal") == 0 ||
strcmp(string, "eq") == 0) {
element->mme_type = MAC_MLS_TYPE_EQUAL;
element->mme_level = MAC_MLS_TYPE_UNDEF;
} else {
element->mme_type = MAC_MLS_TYPE_LEVEL;
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
/*
* Numeric level piece of the element.
*/
level = strsep(&string, ":");
value = strtol(level, &end, 10);
if (end == level || *end != '\0')
return (EINVAL);
if (value < 0 || value > 65535)
return (EINVAL);
element->mme_level = value;
/*
* Optional compartment piece of the element. If none
* are included, we assume that the label has no
* compartments.
*/
if (string == NULL)
return (0);
if (*string == '\0')
return (0);
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
while ((compartment = strsep(&string, "+")) != NULL) {
value = strtol(compartment, &end, 10);
if (compartment == end || *end != '\0')
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
if (value < 1 || value > MAC_MLS_MAX_COMPARTMENTS)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
MAC_MLS_BIT_SET(value, element->mme_compartments);
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
}
}
return (0);
}
/*
* Note: destructively consumes the string, make a local copy before
* calling if that's a problem.
*/
static int
mac_mls_parse(struct mac_mls *mac_mls, char *string)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
char *rangehigh, *rangelow, *effective;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
int error;
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
effective = strsep(&string, "(");
if (*effective == '\0')
effective = NULL;
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
if (string != NULL) {
rangelow = strsep(&string, "-");
if (string == NULL)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
rangehigh = strsep(&string, ")");
if (string == NULL)
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
if (*string != '\0')
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (EINVAL);
Make sure all character pointers are properly initialized; this was mismerged from the MAC tree, and didn't get picked up because warnings are not normally fatal in per-module builds, only when they are linked into a kernel (such as LINT). Reported by: des and the technicolor tinderbox Approved by: re (scottl)
2003-05-30 17:02:36 +00:00
} else {
rangelow = NULL;
rangehigh = NULL;
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
}
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
KASSERT((rangelow != NULL && rangehigh != NULL) ||
(rangelow == NULL && rangehigh == NULL),
Use strsep() in preference to manual string parsing for Biba and MLS label internalization. Use sensible variable names. Include comments. Doesn't fix any known bugs, but may fix unknown ones. Approved by: re (scottl)
2003-05-29 22:51:52 +00:00
("mac_mls_parse: range mismatch"));
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
bzero(mac_mls, sizeof(*mac_mls));
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (effective != NULL) {
error = mac_mls_parse_element(&mac_mls->mm_effective, effective);
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
if (error)
return (error);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls->mm_flags |= MAC_MLS_FLAG_EFFECTIVE;
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
}
if (rangelow != NULL) {
error = mac_mls_parse_element(&mac_mls->mm_rangelow,
rangelow);
if (error)
return (error);
error = mac_mls_parse_element(&mac_mls->mm_rangehigh,
rangehigh);
if (error)
return (error);
mac_mls->mm_flags |= MAC_MLS_FLAG_RANGE;
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
error = mac_mls_valid(mac_mls);
if (error)
return (error);
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
return (0);
}
static int
mac_mls_internalize_label(struct label *label, char *element_name,
char *element_data, int *claimed)
{
struct mac_mls *mac_mls, mac_mls_temp;
int error;
if (strcmp(MAC_MLS_LABEL_NAME, element_name) != 0)
return (0);
(*claimed)++;
error = mac_mls_parse(&mac_mls_temp, element_data);
if (error)
return (error);
mac_mls = SLOT(label);
*mac_mls = mac_mls_temp;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (0);
}
Adapt MAC policies for the new user API changes; teach policies how to parse their own label elements (some cleanup to occur here in the future to use the newly added kernel strsep()). Policies now entirely encapsulate their notion of label in the policy module. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-22 14:31:34 +00:00
static void
mac_mls_copy_label(struct label *src, struct label *dest)
{
*SLOT(dest) = *SLOT(src);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
* Labeling event operations: file system objects, and things that look
* a lot like file system objects.
*/
static void
When devfs cloning takes place, provide access to the credential of the process that caused the clone event to take place for the device driver creating the device. This allows cloned device drivers to adapt the device node based on security aspects of the process, such as the uid, gid, and MAC label. - Add a cred reference to struct cdev, so that when a device node is instantiated as a vnode, the cloning credential can be exposed to MAC. - Add make_dev_cred(), a version of make_dev() that additionally accepts the credential to stick in the struct cdev. Implement it and make_dev() in terms of a back-end make_dev_credv(). - Add a new event handler, dev_clone_cred, which can be registered to receive the credential instead of dev_clone, if desired. - Modify the MAC entry point mac_create_devfs_device() to accept an optional credential pointer (may be NULL), so that MAC policies can inspect and act on the label or other elements of the credential when initializing the skeleton device protections. - Modify tty_pty.c to register clone_dev_cred and invoke make_dev_cred(), so that the pty clone credential is exposed to the MAC Framework. While currently primarily focussed on MAC policies, this change is also a prerequisite for changes to allow ptys to be instantiated with the UID of the process looking up the pty. This requires further changes to the pty driver -- in particular, to immediately recycle pty nodes on last close so that the credential-related state can be recreated on next lookup. Submitted by: Andrew Reisse <andrew.reisse@sparta.com> Obtained from: TrustedBSD Project Sponsored by: SPAWAR, SPARTA MFC after: 1 week MFC note: Merge to 6.x, but not 5.x for ABI reasons
2005-07-14 10:22:09 +00:00
mac_mls_create_devfs_device(struct ucred *cred, struct mount *mp,
struct cdev *dev, struct devfs_dirent *devfs_dirent, struct label *label)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *mac_mls;
int mls_type;
mac_mls = SLOT(label);
if (strcmp(dev->si_name, "null") == 0 ||
strcmp(dev->si_name, "zero") == 0 ||
strcmp(dev->si_name, "random") == 0 ||
strncmp(dev->si_name, "fd/", strlen("fd/")) == 0)
mls_type = MAC_MLS_TYPE_EQUAL;
else if (strcmp(dev->si_name, "kmem") == 0 ||
strcmp(dev->si_name, "mem") == 0)
mls_type = MAC_MLS_TYPE_HIGH;
Add a twiddle to create PTY's with a biba/equal or mls/equal label instead of the default biba/high, mls/low, making it easier to use ptys with these policies. This isn't the final solution, but does help. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 04:15:40 +00:00
else if (ptys_equal &&
(strncmp(dev->si_name, "ttyp", strlen("ttyp")) == 0 ||
strncmp(dev->si_name, "ptyp", strlen("ptyp")) == 0))
mls_type = MAC_MLS_TYPE_EQUAL;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
else
mls_type = MAC_MLS_TYPE_LOW;
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_set_effective(mac_mls, mls_type, 0, NULL);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
Remove dm_root entry from struct devfs_mount. It's never set, and is unused. Replace it with a dm_mount back-pointer to the struct mount that the devfs_mount is associated with. Export that pointer to MAC Framework entry points, where all current policies don't use the pointer. This permits the SEBSD port of SELinux's FLASK/TE to compile out-of-the-box on 5.0-CURRENT with full file system labeling support. Approved by: re (murray) Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-12-09 03:44:28 +00:00
mac_mls_create_devfs_directory(struct mount *mp, char *dirname,
int dirnamelen, struct devfs_dirent *devfs_dirent, struct label *label)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *mac_mls;
mac_mls = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_set_effective(mac_mls, MAC_MLS_TYPE_LOW, 0, NULL);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Implement mac_create_devfs_symlink() for policies that interact with vnode labels. Sync from MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:56:25 +00:00
static void
Remove dm_root entry from struct devfs_mount. It's never set, and is unused. Replace it with a dm_mount back-pointer to the struct mount that the devfs_mount is associated with. Export that pointer to MAC Framework entry points, where all current policies don't use the pointer. This permits the SEBSD port of SELinux's FLASK/TE to compile out-of-the-box on 5.0-CURRENT with full file system labeling support. Approved by: re (murray) Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-12-09 03:44:28 +00:00
mac_mls_create_devfs_symlink(struct ucred *cred, struct mount *mp,
struct devfs_dirent *dd, struct label *ddlabel, struct devfs_dirent *de,
Remove extraneous fullpath variable, which broke lint build. This extra argument to the devfs MAC policy entry points was accidentally merged from the MAC branch during my earlier commit to these policies, and is not scheduled to be merged just yet.
2003-11-12 15:09:39 +00:00
struct label *delabel)
Implement mac_create_devfs_symlink() for policies that interact with vnode labels. Sync from MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:56:25 +00:00
{
struct mac_mls *source, *dest;
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
source = SLOT(cred->cr_label);
Implement mac_create_devfs_symlink() for policies that interact with vnode labels. Sync from MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:56:25 +00:00
dest = SLOT(delabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Implement mac_create_devfs_symlink() for policies that interact with vnode labels. Sync from MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:56:25 +00:00
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static void
mac_mls_create_mount(struct ucred *cred, struct mount *mp,
struct label *mntlabel, struct label *fslabel)
{
struct mac_mls *source, *dest;
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
source = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
dest = SLOT(mntlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
dest = SLOT(fslabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_relabel_vnode(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel, struct label *label)
{
struct mac_mls *source, *dest;
source = SLOT(label);
dest = SLOT(vnodelabel);
Introduce mac_biba_copy() and mac_mls_copy(), which conditionally copy elements of one Biba or MLS label to another based on the flags on the source label element. Use this instead of mac_{biba,mls}_{single,range}() to simplify the existing code, as well as support partial label updates (we don't update if none is requested). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 20:55:39 +00:00
mac_mls_copy(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
Remove dm_root entry from struct devfs_mount. It's never set, and is unused. Replace it with a dm_mount back-pointer to the struct mount that the devfs_mount is associated with. Export that pointer to MAC Framework entry points, where all current policies don't use the pointer. This permits the SEBSD port of SELinux's FLASK/TE to compile out-of-the-box on 5.0-CURRENT with full file system labeling support. Approved by: re (murray) Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-12-09 03:44:28 +00:00
mac_mls_update_devfsdirent(struct mount *mp,
struct devfs_dirent *devfs_dirent, struct label *direntlabel,
struct vnode *vp, struct label *vnodelabel)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *source, *dest;
source = SLOT(vnodelabel);
dest = SLOT(direntlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
mac_mls_associate_vnode_devfs(struct mount *mp, struct label *fslabel,
struct devfs_dirent *de, struct label *delabel, struct vnode *vp,
struct label *vlabel)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *source, *dest;
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
source = SLOT(delabel);
dest = SLOT(vlabel);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static int
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
mac_mls_associate_vnode_extattr(struct mount *mp, struct label *fslabel,
struct vnode *vp, struct label *vlabel)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
struct mac_mls temp, *source, *dest;
Fix some warnings on 64 bit architectures. The vn_extattr_get() function takes an int * parameter, not a size_t * parameter. Arguably, it should rather take a size_t *, but that would require changing the uio_resid field of struct uio to be a size_t instead of an int, which I don't want to do that close to 5.0-RELEASE. Reviewed by: rwatson
2002-11-02 20:16:35 +00:00
int buflen, error;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
source = SLOT(fslabel);
dest = SLOT(vlabel);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
buflen = sizeof(temp);
bzero(&temp, buflen);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
error = vn_extattr_get(vp, IO_NODELOCKED, MAC_MLS_EXTATTR_NAMESPACE,
MAC_MLS_EXTATTR_NAME, &buflen, (char *) &temp, curthread);
if (error == ENOATTR || error == EOPNOTSUPP) {
/* Fall back to the fslabel. */
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
return (0);
} else if (error)
return (error);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
if (buflen != sizeof(temp)) {
printf("mac_mls_associate_vnode_extattr: bad size %d\n",
buflen);
return (EPERM);
}
if (mac_mls_valid(&temp) != 0) {
printf("mac_mls_associate_vnode_extattr: invalid\n");
return (EPERM);
}
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if ((temp.mm_flags & MAC_MLS_FLAGS_BOTH) != MAC_MLS_FLAG_EFFECTIVE) {
printf("mac_mls_associated_vnode_extattr: not effective\n");
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
return (EPERM);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(&temp, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (0);
}
static void
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
mac_mls_associate_vnode_singlelabel(struct mount *mp,
struct label *fslabel, struct vnode *vp, struct label *vlabel)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *source, *dest;
source = SLOT(fslabel);
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
dest = SLOT(vlabel);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
static int
mac_mls_create_vnode_extattr(struct ucred *cred, struct mount *mp,
struct label *fslabel, struct vnode *dvp, struct label *dlabel,
struct vnode *vp, struct label *vlabel, struct componentname *cnp)
{
struct mac_mls *source, *dest, temp;
size_t buflen;
int error;
buflen = sizeof(temp);
bzero(&temp, 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
source = SLOT(cred->cr_label);
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
dest = SLOT(vlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, &temp);
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
error = vn_extattr_set(vp, IO_NODELOCKED, MAC_MLS_EXTATTR_NAMESPACE,
MAC_MLS_EXTATTR_NAME, buflen, (char *) &temp, curthread);
if (error == 0)
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
return (error);
}
static int
mac_mls_setlabel_vnode_extattr(struct ucred *cred, struct vnode *vp,
struct label *vlabel, struct label *intlabel)
{
struct mac_mls *source, temp;
size_t buflen;
int error;
buflen = sizeof(temp);
bzero(&temp, buflen);
source = SLOT(intlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if ((source->mm_flags & MAC_MLS_FLAG_EFFECTIVE) == 0)
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
return (0);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, &temp);
Slightly change the semantics of vnode labels for MAC: rather than "refreshing" the label on the vnode before use, just get the label right from inception. For single-label file systems, set the label in the generic VFS getnewvnode() code; for multi-label file systems, leave the labeling up to the file system. With UFS1/2, this means reading the extended attribute during vfs_vget() as the inode is pulled off disk, rather than hitting the extended attributes frequently during operations later, improving performance. This also corrects sematics for shared vnode locks, which were not previously present in the system. This chances the cache coherrency properties WRT out-of-band access to label data, but in an acceptable form. With UFS1, there is a small race condition during automatic extended attribute start -- this is not present with UFS2, and occurs because EAs aren't available at vnode inception. We'll introduce a work around for this shortly. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-26 14:38:24 +00:00
error = vn_extattr_set(vp, IO_NODELOCKED, MAC_MLS_EXTATTR_NAMESPACE,
MAC_MLS_EXTATTR_NAME, buflen, (char *) &temp, curthread);
return (error);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
* Labeling event operations: IPC object.
*/
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
static void
mac_mls_create_inpcb_from_socket(struct socket *so, struct label *solabel,
struct inpcb *inp, struct label *inplabel)
{
struct mac_mls *source, *dest;
source = SLOT(solabel);
dest = SLOT(inplabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static void
mac_mls_create_mbuf_from_socket(struct socket *so, struct label *socketlabel,
struct mbuf *m, struct label *mbuflabel)
{
struct mac_mls *source, *dest;
source = SLOT(socketlabel);
dest = SLOT(mbuflabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_socket(struct ucred *cred, struct socket *socket,
struct label *socketlabel)
{
struct mac_mls *source, *dest;
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
source = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
dest = SLOT(socketlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
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
mac_mls_create_pipe(struct ucred *cred, struct pipepair *pp,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
struct label *pipelabel)
{
struct mac_mls *source, *dest;
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
source = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
dest = SLOT(pipelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Introduce MAC Framework and MAC Policy entry points to label and control access to POSIX Semaphores: mac_init_posix_sem() Initialize label for POSIX semaphore mac_create_posix_sem() Create POSIX semaphore mac_destroy_posix_sem() Destroy POSIX semaphore mac_check_posix_sem_destroy() Check whether semaphore may be destroyed mac_check_posix_sem_getvalue() Check whether semaphore may be queried mac_check_possix_sem_open() Check whether semaphore may be opened mac_check_posix_sem_post() Check whether semaphore may be posted to mac_check_posix_sem_unlink() Check whether semaphore may be unlinked mac_check_posix_sem_wait() Check whether may wait on semaphore Update Biba, MLS, Stub, and Test policies to implement these entry points. For information flow policies, most semaphore operations are effectively read/write. Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Sponsored by: DARPA, McAfee, SPARTA Obtained from: TrustedBSD Project
2005-05-04 10:39:15 +00:00
static void
mac_mls_create_posix_sem(struct ucred *cred, struct ksem *ksemptr,
struct label *ks_label)
{
struct mac_mls *source, *dest;
source = SLOT(cred->cr_label);
dest = SLOT(ks_label);
mac_mls_copy_effective(source, dest);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static void
mac_mls_create_socket_from_socket(struct socket *oldsocket,
struct label *oldsocketlabel, struct socket *newsocket,
struct label *newsocketlabel)
{
struct mac_mls *source, *dest;
source = SLOT(oldsocketlabel);
dest = SLOT(newsocketlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_relabel_socket(struct ucred *cred, struct socket *socket,
struct label *socketlabel, struct label *newlabel)
{
struct mac_mls *source, *dest;
source = SLOT(newlabel);
dest = SLOT(socketlabel);
Introduce mac_biba_copy() and mac_mls_copy(), which conditionally copy elements of one Biba or MLS label to another based on the flags on the source label element. Use this instead of mac_{biba,mls}_{single,range}() to simplify the existing code, as well as support partial label updates (we don't update if none is requested). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 20:55:39 +00:00
mac_mls_copy(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
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
mac_mls_relabel_pipe(struct ucred *cred, struct pipepair *pp,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
struct label *pipelabel, struct label *newlabel)
{
struct mac_mls *source, *dest;
source = SLOT(newlabel);
dest = SLOT(pipelabel);
Introduce mac_biba_copy() and mac_mls_copy(), which conditionally copy elements of one Biba or MLS label to another based on the flags on the source label element. Use this instead of mac_{biba,mls}_{single,range}() to simplify the existing code, as well as support partial label updates (we don't update if none is requested). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 20:55:39 +00:00
mac_mls_copy(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_set_socket_peer_from_mbuf(struct mbuf *mbuf, struct label *mbuflabel,
struct socket *socket, struct label *socketpeerlabel)
{
struct mac_mls *source, *dest;
source = SLOT(mbuflabel);
dest = SLOT(socketpeerlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
/*
* Labeling event operations: System V IPC objects.
*/
static void
mac_mls_create_sysv_msgmsg(struct ucred *cred, struct msqid_kernel *msqkptr,
struct label *msqlabel, struct msg *msgptr, struct label *msglabel)
{
struct mac_mls *source, *dest;
/* Ignore the msgq label */
source = SLOT(cred->cr_label);
dest = SLOT(msglabel);
mac_mls_copy_effective(source, dest);
}
static void
mac_mls_create_sysv_msgqueue(struct ucred *cred, struct msqid_kernel *msqkptr,
struct label *msqlabel)
{
struct mac_mls *source, *dest;
source = SLOT(cred->cr_label);
dest = SLOT(msqlabel);
mac_mls_copy_effective(source, dest);
}
static void
Gratuitous renaming of four System V Semaphore MAC Framework entry points to convert _sema() to _sem() for consistency purposes with respect to the other semaphore-related entry points: mac_init_sysv_sema() -> mac_init_sysv_sem() mac_destroy_sysv_sem() -> mac_destroy_sysv_sem() mac_create_sysv_sema() -> mac_create_sysv_sem() mac_cleanup_sysv_sema() -> mac_cleanup_sysv_sem() Congruent changes are made to the policy interface to support this. Obtained from: TrustedBSD Project Sponsored by: SPAWAR, SPARTA
2005-06-07 05:03:28 +00:00
mac_mls_create_sysv_sem(struct ucred *cred, struct semid_kernel *semakptr,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
struct label *semalabel)
{
struct mac_mls *source, *dest;
source = SLOT(cred->cr_label);
dest = SLOT(semalabel);
mac_mls_copy_effective(source, dest);
}
static void
mac_mls_create_sysv_shm(struct ucred *cred, struct shmid_kernel *shmsegptr,
struct label *shmlabel)
{
struct mac_mls *source, *dest;
source = SLOT(cred->cr_label);
dest = SLOT(shmlabel);
mac_mls_copy_effective(source, dest);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
* Labeling event operations: network objects.
*/
static void
mac_mls_set_socket_peer_from_socket(struct socket *oldsocket,
struct label *oldsocketlabel, struct socket *newsocket,
struct label *newsocketpeerlabel)
{
struct mac_mls *source, *dest;
source = SLOT(oldsocketlabel);
dest = SLOT(newsocketpeerlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_bpfdesc(struct ucred *cred, struct bpf_d *bpf_d,
struct label *bpflabel)
{
struct mac_mls *source, *dest;
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
source = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
dest = SLOT(bpflabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_ifnet(struct ifnet *ifnet, struct label *ifnetlabel)
{
struct mac_mls *dest;
Rename the variable 'grade' to 'type' in interface parsing and labeling for Biba. Rename the variable 'level' to 'type' in interface parsing and labeling for MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:00:51 +00:00
int type;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
dest = SLOT(ifnetlabel);
if (ifnet->if_type == IFT_LOOP)
Rename the variable 'grade' to 'type' in interface parsing and labeling for Biba. Rename the variable 'level' to 'type' in interface parsing and labeling for MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:00:51 +00:00
type = MAC_MLS_TYPE_EQUAL;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
else
Rename the variable 'grade' to 'type' in interface parsing and labeling for Biba. Rename the variable 'level' to 'type' in interface parsing and labeling for MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:00:51 +00:00
type = MAC_MLS_TYPE_LOW;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_set_effective(dest, type, 0, NULL);
Rename the variable 'grade' to 'type' in interface parsing and labeling for Biba. Rename the variable 'level' to 'type' in interface parsing and labeling for MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:00:51 +00:00
mac_mls_set_range(dest, type, 0, NULL, type, 0, NULL);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_ipq(struct mbuf *fragment, struct label *fragmentlabel,
struct ipq *ipq, struct label *ipqlabel)
{
struct mac_mls *source, *dest;
source = SLOT(fragmentlabel);
dest = SLOT(ipqlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_datagram_from_ipq(struct ipq *ipq, struct label *ipqlabel,
struct mbuf *datagram, struct label *datagramlabel)
{
struct mac_mls *source, *dest;
source = SLOT(ipqlabel);
dest = SLOT(datagramlabel);
/* Just use the head, since we require them all to match. */
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_fragment(struct mbuf *datagram, struct label *datagramlabel,
struct mbuf *fragment, struct label *fragmentlabel)
{
struct mac_mls *source, *dest;
source = SLOT(datagramlabel);
dest = SLOT(fragmentlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Switch TCP over to using the inpcb label when responding in timed wait, rather than the socket label. This avoids reaching up to the socket layer during connection close, which requires locking changes. To do this, introduce MAC Framework entry point mac_create_mbuf_from_inpcb(), which is called from tcp_twrespond() instead of calling mac_create_mbuf_from_socket() or mac_create_mbuf_netlayer(). Introduce MAC Policy entry point mpo_create_mbuf_from_inpcb(), and implementations for various policies, which generally just copy label data from the inpcb to the mbuf. Assert the inpcb lock in the entry point since we require consistency for the inpcb label reference. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-12-17 14:55:11 +00:00
static void
mac_mls_create_mbuf_from_inpcb(struct inpcb *inp, struct label *inplabel,
struct mbuf *m, struct label *mlabel)
{
struct mac_mls *source, *dest;
source = SLOT(inplabel);
dest = SLOT(mlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Switch TCP over to using the inpcb label when responding in timed wait, rather than the socket label. This avoids reaching up to the socket layer during connection close, which requires locking changes. To do this, introduce MAC Framework entry point mac_create_mbuf_from_inpcb(), which is called from tcp_twrespond() instead of calling mac_create_mbuf_from_socket() or mac_create_mbuf_netlayer(). Introduce MAC Policy entry point mpo_create_mbuf_from_inpcb(), and implementations for various policies, which generally just copy label data from the inpcb to the mbuf. Assert the inpcb lock in the entry point since we require consistency for the inpcb label reference. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-12-17 14:55:11 +00:00
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static void
mac_mls_create_mbuf_linklayer(struct ifnet *ifnet, struct label *ifnetlabel,
struct mbuf *mbuf, struct label *mbuflabel)
{
struct mac_mls *dest;
dest = SLOT(mbuflabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_set_effective(dest, MAC_MLS_TYPE_EQUAL, 0, NULL);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_mbuf_from_bpfdesc(struct bpf_d *bpf_d, struct label *bpflabel,
struct mbuf *mbuf, struct label *mbuflabel)
{
struct mac_mls *source, *dest;
source = SLOT(bpflabel);
dest = SLOT(mbuflabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_mbuf_from_ifnet(struct ifnet *ifnet, struct label *ifnetlabel,
struct mbuf *m, struct label *mbuflabel)
{
struct mac_mls *source, *dest;
source = SLOT(ifnetlabel);
dest = SLOT(mbuflabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_mbuf_multicast_encap(struct mbuf *oldmbuf,
struct label *oldmbuflabel, struct ifnet *ifnet, struct label *ifnetlabel,
struct mbuf *newmbuf, struct label *newmbuflabel)
{
struct mac_mls *source, *dest;
source = SLOT(oldmbuflabel);
dest = SLOT(newmbuflabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_mbuf_netlayer(struct mbuf *oldmbuf, struct label *oldmbuflabel,
struct mbuf *newmbuf, struct label *newmbuflabel)
{
struct mac_mls *source, *dest;
source = SLOT(oldmbuflabel);
dest = SLOT(newmbuflabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_copy_effective(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static int
mac_mls_fragment_match(struct mbuf *fragment, struct label *fragmentlabel,
struct ipq *ipq, struct label *ipqlabel)
{
struct mac_mls *a, *b;
a = SLOT(ipqlabel);
b = SLOT(fragmentlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
return (mac_mls_equal_effective(a, b));
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_relabel_ifnet(struct ucred *cred, struct ifnet *ifnet,
struct label *ifnetlabel, struct label *newlabel)
{
struct mac_mls *source, *dest;
source = SLOT(newlabel);
dest = SLOT(ifnetlabel);
Introduce mac_biba_copy() and mac_mls_copy(), which conditionally copy elements of one Biba or MLS label to another based on the flags on the source label element. Use this instead of mac_{biba,mls}_{single,range}() to simplify the existing code, as well as support partial label updates (we don't update if none is requested). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 20:55:39 +00:00
mac_mls_copy(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_update_ipq(struct mbuf *fragment, struct label *fragmentlabel,
struct ipq *ipq, struct label *ipqlabel)
{
/* NOOP: we only accept matching labels, so no need to update */
}
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
static void
mac_mls_inpcb_sosetlabel(struct socket *so, struct label *solabel,
struct inpcb *inp, struct label *inplabel)
{
struct mac_mls *source, *dest;
source = SLOT(solabel);
dest = SLOT(inplabel);
mac_mls_copy(source, dest);
}
Introduce a new entry point, mac_create_mbuf_from_firewall. This entry point exists to allow the mandatory access control policy to properly initialize mbufs generated by the firewall. An example where this might happen is keep alive packets, or ICMP error packets in response to other packets. This takes care of kernel panics associated with un-initialize mbuf labels when the firewall generates packets. [1] I modified this patch from it's original version, the initial patch introduced a number of entry points which were programmatically equivalent. So I introduced only one. Instead, we should leverage mac_create_mbuf_netlayer() which is used for similar situations, an example being icmp_error() This will minimize the impact associated with the MFC Submitted by: mlaier [1] MFC after: 1 week This is a RELENG_6 candidate
2006-09-12 04:25:13 +00:00
static void
mac_mls_create_mbuf_from_firewall(struct mbuf *m, struct label *mbuflabel)
{
struct mac_mls *dest;
dest = SLOT(mbuflabel);
/* XXX: where is the label for the firewall really comming from? */
mac_mls_set_effective(dest, MAC_MLS_TYPE_EQUAL, 0, NULL);
}
Teach the MAC policies which utilize mbuf labeling the new syncache entry points. Properly initialize the mbuf label based on the label we copy from the PCB. This fixes an LOR between the PCB and syncache code.
2006-12-13 06:03:22 +00:00
static void
mac_mls_init_syncache_from_inpcb(struct label *label, struct inpcb *inp)
{
struct mac_mls *source, *dest;
source = SLOT(inp->inp_label);
dest = SLOT(label);
mac_mls_copy_effective(source, dest);
}
static void
mac_mls_create_mbuf_from_syncache(struct label *sc_label, struct mbuf *m,
struct label *mbuf_label)
{
struct mac_mls *source, *dest;
source = SLOT(sc_label);
dest = SLOT(mbuf_label);
mac_mls_copy_effective(source, dest);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
* Labeling event operations: processes.
*/
static void
mac_mls_create_proc0(struct ucred *cred)
{
struct mac_mls *dest;
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
dest = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_set_effective(dest, MAC_MLS_TYPE_EQUAL, 0, NULL);
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
mac_mls_set_range(dest, MAC_MLS_TYPE_LOW, 0, NULL, MAC_MLS_TYPE_HIGH,
0, NULL);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_create_proc1(struct ucred *cred)
{
struct mac_mls *dest;
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
dest = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
mac_mls_set_effective(dest, MAC_MLS_TYPE_LOW, 0, NULL);
Add compartment support to Biba and MLS policies. The logic of the policies remains the same: subjects and objects are labeled for integrity or sensitivity, and a dominance operator determines whether or not subject/object accesses are permitted to limit inappropriate information flow. Compartments are a non-hierarchal component to the label, so add a bitfield to the label element for each, and a set check as part of the dominance operator. This permits the implementation of "need to know" elements of MLS. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:42:01 +00:00
mac_mls_set_range(dest, MAC_MLS_TYPE_LOW, 0, NULL, MAC_MLS_TYPE_HIGH,
0, NULL);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static void
mac_mls_relabel_cred(struct ucred *cred, struct label *newlabel)
{
struct mac_mls *source, *dest;
source = SLOT(newlabel);
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
dest = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Introduce mac_biba_copy() and mac_mls_copy(), which conditionally copy elements of one Biba or MLS label to another based on the flags on the source label element. Use this instead of mac_{biba,mls}_{single,range}() to simplify the existing code, as well as support partial label updates (we don't update if none is requested). Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 20:55:39 +00:00
mac_mls_copy(source, dest);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
/*
* Label cleanup/flush operations.
*/
static void
mac_mls_cleanup_sysv_msgmsg(struct label *msglabel)
{
bzero(SLOT(msglabel), sizeof(struct mac_mls));
}
static void
mac_mls_cleanup_sysv_msgqueue(struct label *msqlabel)
{
bzero(SLOT(msqlabel), sizeof(struct mac_mls));
}
static void
Gratuitous renaming of four System V Semaphore MAC Framework entry points to convert _sema() to _sem() for consistency purposes with respect to the other semaphore-related entry points: mac_init_sysv_sema() -> mac_init_sysv_sem() mac_destroy_sysv_sem() -> mac_destroy_sysv_sem() mac_create_sysv_sema() -> mac_create_sysv_sem() mac_cleanup_sysv_sema() -> mac_cleanup_sysv_sem() Congruent changes are made to the policy interface to support this. Obtained from: TrustedBSD Project Sponsored by: SPAWAR, SPARTA
2005-06-07 05:03:28 +00:00
mac_mls_cleanup_sysv_sem(struct label *semalabel)
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
{
bzero(SLOT(semalabel), sizeof(struct mac_mls));
}
static void
mac_mls_cleanup_sysv_shm(struct label *shmlabel)
{
bzero(SLOT(shmlabel), sizeof(struct mac_mls));
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
* Access control checks.
*/
static int
mac_mls_check_bpfdesc_receive(struct bpf_d *bpf_d, struct label *bpflabel,
struct ifnet *ifnet, struct label *ifnetlabel)
{
struct mac_mls *a, *b;
if (!mac_mls_enabled)
return (0);
a = SLOT(bpflabel);
b = SLOT(ifnetlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (mac_mls_equal_effective(a, b))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (0);
return (EACCES);
}
static int
mac_mls_check_cred_relabel(struct ucred *cred, struct label *newlabel)
{
struct mac_mls *subj, *new;
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
int error;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
new = SLOT(newlabel);
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If there is an MLS label update for the credential, it may be
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* an update of effective, range, or both.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
error = mls_atmostflags(new, MAC_MLS_FLAGS_BOTH);
if (error)
return (error);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If the MLS label is to be changed, authorize as appropriate.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (new->mm_flags & MAC_MLS_FLAGS_BOTH) {
Place more stringent checks on process credential relabeling for the Biba and MLS policies: as we support both an effective (single) element and range (available) elements, require that the single be in the range if both the single and range are defined in the update. Remove comments suggesting that such a check might be a good idea. Don't introduce a similar check for network interfaces; due to different interpretations of the single and range elements, it's not clear that it's useful to do so. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:28:46 +00:00
/*
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* If the change request modifies both the MLS label effective
* and range, check that the new effective will be in the
Place more stringent checks on process credential relabeling for the Biba and MLS policies: as we support both an effective (single) element and range (available) elements, require that the single be in the range if both the single and range are defined in the update. Remove comments suggesting that such a check might be a good idea. Don't introduce a similar check for network interfaces; due to different interpretations of the single and range elements, it's not clear that it's useful to do so. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:28:46 +00:00
* new range.
*/
if ((new->mm_flags & MAC_MLS_FLAGS_BOTH) ==
MAC_MLS_FLAGS_BOTH &&
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
!mac_mls_effective_in_range(new, new))
Place more stringent checks on process credential relabeling for the Biba and MLS policies: as we support both an effective (single) element and range (available) elements, require that the single be in the range if both the single and range are defined in the update. Remove comments suggesting that such a check might be a good idea. Don't introduce a similar check for network interfaces; due to different interpretations of the single and range elements, it's not clear that it's useful to do so. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:28:46 +00:00
return (EINVAL);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
/*
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* To change the MLS effective label on a credential, the
* new effective label must be in the current range.
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (new->mm_flags & MAC_MLS_FLAG_EFFECTIVE &&
!mac_mls_effective_in_range(new, subj))
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (EPERM);
/*
* To change the MLS range label on a credential, the
Place more stringent checks on process credential relabeling for the Biba and MLS policies: as we support both an effective (single) element and range (available) elements, require that the single be in the range if both the single and range are defined in the update. Remove comments suggesting that such a check might be a good idea. Don't introduce a similar check for network interfaces; due to different interpretations of the single and range elements, it's not clear that it's useful to do so. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-02-04 21:28:46 +00:00
* new range must be in the current range.
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
*/
if (new->mm_flags & MAC_MLS_FLAG_RANGE &&
White space nit the crept in during merge.
2002-10-21 17:01:30 +00:00
!mac_mls_range_in_range(new, subj))
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (EPERM);
/*
* To have EQUAL in any component of the new credential
* MLS label, the subject must already have EQUAL in
* their label.
*/
if (mac_mls_contains_equal(new)) {
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
error = mac_mls_subject_privileged(subj);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (error)
return (error);
}
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (0);
}
static int
mac_mls_check_cred_visible(struct ucred *u1, struct ucred *u2)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(u1->cr_label);
obj = SLOT(u2->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/* XXX: range */
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (ESRCH);
return (0);
}
static int
mac_mls_check_ifnet_relabel(struct ucred *cred, struct ifnet *ifnet,
struct label *ifnetlabel, struct label *newlabel)
{
struct mac_mls *subj, *new;
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
int error;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
new = SLOT(newlabel);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
/*
* If there is an MLS label update for the interface, it may
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* be an update of effective, range, or both.
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
*/
error = mls_atmostflags(new, MAC_MLS_FLAGS_BOTH);
if (error)
return (error);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
/*
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
* Relabeling network interfaces requires MLS privilege.
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
*/
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
error = mac_mls_subject_privileged(subj);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (0);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static int
mac_mls_check_ifnet_transmit(struct ifnet *ifnet, struct label *ifnetlabel,
struct mbuf *m, struct label *mbuflabel)
{
struct mac_mls *p, *i;
if (!mac_mls_enabled)
return (0);
p = SLOT(mbuflabel);
i = SLOT(ifnetlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
return (mac_mls_effective_in_range(p, i) ? 0 : EACCES);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
static int
mac_mls_check_inpcb_deliver(struct inpcb *inp, struct label *inplabel,
struct mbuf *m, struct label *mlabel)
{
struct mac_mls *p, *i;
if (!mac_mls_enabled)
return (0);
p = SLOT(mlabel);
i = SLOT(inplabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
return (mac_mls_equal_effective(p, i) ? 0 : EACCES);
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
}
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
static int
mac_mls_check_sysv_msgrcv(struct ucred *cred, struct msg *msgptr,
struct label *msglabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(msglabel);
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_msgrmid(struct ucred *cred, struct msg *msgptr,
struct label *msglabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(msglabel);
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_msqget(struct ucred *cred, struct msqid_kernel *msqkptr,
struct label *msqklabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(msqklabel);
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_msqsnd(struct ucred *cred, struct msqid_kernel *msqkptr,
struct label *msqklabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(msqklabel);
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_msqrcv(struct ucred *cred, struct msqid_kernel *msqkptr,
struct label *msqklabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(msqklabel);
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_msqctl(struct ucred *cred, struct msqid_kernel *msqkptr,
struct label *msqklabel, int cmd)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(msqklabel);
switch(cmd) {
case IPC_RMID:
case IPC_SET:
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
break;
case IPC_STAT:
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
break;
default:
return (EACCES);
}
return (0);
}
static int
mac_mls_check_sysv_semctl(struct ucred *cred, struct semid_kernel *semakptr,
struct label *semaklabel, int cmd)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(semaklabel);
switch(cmd) {
case IPC_RMID:
case IPC_SET:
case SETVAL:
case SETALL:
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
break;
case IPC_STAT:
case GETVAL:
case GETPID:
case GETNCNT:
case GETZCNT:
case GETALL:
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
break;
default:
return (EACCES);
}
return (0);
}
static int
mac_mls_check_sysv_semget(struct ucred *cred, struct semid_kernel *semakptr,
struct label *semaklabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(semaklabel);
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_semop(struct ucred *cred, struct semid_kernel *semakptr,
struct label *semaklabel, size_t accesstype)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(semaklabel);
if( accesstype & SEM_R )
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
if( accesstype & SEM_A )
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_shmat(struct ucred *cred, struct shmid_kernel *shmsegptr,
struct label *shmseglabel, int shmflg)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(shmseglabel);
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
if ((shmflg & SHM_RDONLY) == 0)
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
return (0);
}
static int
mac_mls_check_sysv_shmctl(struct ucred *cred, struct shmid_kernel *shmsegptr,
struct label *shmseglabel, int cmd)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(shmseglabel);
switch(cmd) {
case IPC_RMID:
case IPC_SET:
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
break;
case IPC_STAT:
case SHM_STAT:
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
break;
default:
return (EACCES);
}
return (0);
}
static int
mac_mls_check_sysv_shmget(struct ucred *cred, struct shmid_kernel *shmsegptr,
struct label *shmseglabel, int shmflg)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(shmseglabel);
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
return (0);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_check_mount_stat(struct ucred *cred, struct mount *mp,
struct label *mntlabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(mntlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
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
mac_mls_check_pipe_ioctl(struct ucred *cred, struct pipepair *pp,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
struct label *pipelabel, unsigned long cmd, void /* caddr_t */ *data)
{
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
if(!mac_mls_enabled)
return (0);
/* XXX: This will be implemented soon... */
return (0);
}
static int
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
mac_mls_check_pipe_poll(struct ucred *cred, struct pipepair *pp,
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
struct label *pipelabel)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT((pipelabel));
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
return (EACCES);
return (0);
}
static int
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
mac_mls_check_pipe_read(struct ucred *cred, struct pipepair *pp,
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
struct label *pipelabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
obj = SLOT((pipelabel));
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
return (EACCES);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (0);
}
static int
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
mac_mls_check_pipe_relabel(struct ucred *cred, struct pipepair *pp,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
struct label *pipelabel, struct label *newlabel)
{
struct mac_mls *subj, *obj, *new;
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
int error;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
new = SLOT(newlabel);
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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(pipelabel);
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If there is an MLS label update for a pipe, it must be a
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* effective update.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
error = mls_atmostflags(new, MAC_MLS_FLAG_EFFECTIVE);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (error)
return (error);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* To perform a relabel of a pipe (MLS label or not), MLS must
* authorize the relabel.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_effective_in_range(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EPERM);
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If the MLS label is to be changed, authorize as appropriate.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (new->mm_flags & MAC_MLS_FLAG_EFFECTIVE) {
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
/*
* To change the MLS label on a pipe, the new pipe label
* must be in the subject range.
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_effective_in_range(new, subj))
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (EPERM);
/*
* To change the MLS label on a pipe to be EQUAL, the
* subject must have appropriate privilege.
*/
if (mac_mls_contains_equal(new)) {
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
error = mac_mls_subject_privileged(subj);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (error)
return (error);
}
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (0);
}
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
static int
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
mac_mls_check_pipe_stat(struct ucred *cred, struct pipepair *pp,
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
struct label *pipelabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
obj = SLOT((pipelabel));
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
return (EACCES);
return (0);
}
static int
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
mac_mls_check_pipe_write(struct ucred *cred, struct pipepair *pp,
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
struct label *pipelabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
obj = SLOT((pipelabel));
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Break out mac_check_pipe_op() into component check entry points: mac_check_pipe_poll(), mac_check_pipe_read(), mac_check_pipe_stat(), and mac_check_pipe_write(). This is improves consistency with other access control entry points and permits security modules to only control the object methods that they are interested in, avoiding switch statements. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:59:37 +00:00
return (EACCES);
return (0);
}
Introduce MAC Framework and MAC Policy entry points to label and control access to POSIX Semaphores: mac_init_posix_sem() Initialize label for POSIX semaphore mac_create_posix_sem() Create POSIX semaphore mac_destroy_posix_sem() Destroy POSIX semaphore mac_check_posix_sem_destroy() Check whether semaphore may be destroyed mac_check_posix_sem_getvalue() Check whether semaphore may be queried mac_check_possix_sem_open() Check whether semaphore may be opened mac_check_posix_sem_post() Check whether semaphore may be posted to mac_check_posix_sem_unlink() Check whether semaphore may be unlinked mac_check_posix_sem_wait() Check whether may wait on semaphore Update Biba, MLS, Stub, and Test policies to implement these entry points. For information flow policies, most semaphore operations are effectively read/write. Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Sponsored by: DARPA, McAfee, SPARTA Obtained from: TrustedBSD Project
2005-05-04 10:39:15 +00:00
static int
mac_mls_check_posix_sem_write(struct ucred *cred, struct ksem *ksemptr,
struct label *ks_label)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(ks_label);
if (!mac_mls_dominate_effective(obj, subj))
return (EACCES);
return (0);
}
static int
mac_mls_check_posix_sem_rdonly(struct ucred *cred, struct ksem *ksemptr,
struct label *ks_label)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (0);
subj = SLOT(cred->cr_label);
obj = SLOT(ks_label);
if (!mac_mls_dominate_effective(subj, obj))
return (EACCES);
return (0);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_check_proc_debug(struct ucred *cred, struct proc *proc)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
obj = SLOT(proc->p_ucred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/* XXX: range checks */
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (ESRCH);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_proc_sched(struct ucred *cred, struct proc *proc)
{
struct mac_mls *subj, *obj;
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
obj = SLOT(proc->p_ucred->cr_label);
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/* XXX: range checks */
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (ESRCH);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_proc_signal(struct ucred *cred, struct proc *proc, int signum)
{
struct mac_mls *subj, *obj;
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
obj = SLOT(proc->p_ucred->cr_label);
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/* XXX: range checks */
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (ESRCH);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
Rename mac_check_socket_receive() to mac_check_socket_deliver() so that we can use the names _receive() and _send() for the receive() and send() checks. Rename related constants, policy implementations, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 18:51:27 +00:00
mac_mls_check_socket_deliver(struct socket *so, struct label *socketlabel,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
struct mbuf *m, struct label *mbuflabel)
{
struct mac_mls *p, *s;
if (!mac_mls_enabled)
return (0);
p = SLOT(mbuflabel);
s = SLOT(socketlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
return (mac_mls_equal_effective(p, s) ? 0 : EACCES);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static int
mac_mls_check_socket_relabel(struct ucred *cred, struct socket *socket,
struct label *socketlabel, struct label *newlabel)
{
struct mac_mls *subj, *obj, *new;
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
int error;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
new = SLOT(newlabel);
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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(socketlabel);
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If there is an MLS label update for the socket, it may be
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* an update of effective.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
error = mls_atmostflags(new, MAC_MLS_FLAG_EFFECTIVE);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (error)
return (error);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* To relabel a socket, the old socket effective must be in the subject
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
* range.
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_effective_in_range(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EPERM);
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If the MLS label is to be changed, authorize as appropriate.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (new->mm_flags & MAC_MLS_FLAG_EFFECTIVE) {
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
/*
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* To relabel a socket, the new socket effective must be in
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* the subject range.
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_effective_in_range(new, subj))
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (EPERM);
/*
* To change the MLS label on the socket to contain EQUAL,
* the subject must have appropriate privilege.
*/
if (mac_mls_contains_equal(new)) {
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
error = mac_mls_subject_privileged(subj);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (error)
return (error);
}
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (0);
}
static int
mac_mls_check_socket_visible(struct ucred *cred, struct socket *socket,
struct label *socketlabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(socketlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (ENOENT);
return (0);
}
Expand scope of the MLS policy to include a new entry point available for enforcement: mac_mls_check_system_swapon() - Require that the subject and the swapfile target vnode labels dominate one another. An additional check is probably needed here to require that the swapfile target has a label of mls/high to prevent information leakage through swapfiles. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-03-25 01:16:10 +00:00
static int
mac_mls_check_system_swapon(struct ucred *cred, struct vnode *vp,
struct label *label)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Expand scope of the MLS policy to include a new entry point available for enforcement: mac_mls_check_system_swapon() - Require that the subject and the swapfile target vnode labels dominate one another. An additional check is probably needed here to require that the swapfile target has a label of mls/high to prevent information leakage through swapfiles. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-03-25 01:16:10 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj) ||
!mac_mls_dominate_effective(subj, obj))
Expand scope of the MLS policy to include a new entry point available for enforcement: mac_mls_check_system_swapon() - Require that the subject and the swapfile target vnode labels dominate one another. An additional check is probably needed here to require that the swapfile target has a label of mls/high to prevent information leakage through swapfiles. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-03-25 01:16:10 +00:00
return (EACCES);
return (0);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_check_vnode_chdir(struct ucred *cred, struct vnode *dvp,
struct label *dlabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_chroot(struct ucred *cred, struct vnode *dvp,
struct label *dlabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_create(struct ucred *cred, struct vnode *dvp,
struct label *dlabel, struct componentname *cnp, struct vattr *vap)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_delete(struct ucred *cred, struct vnode *dvp,
struct label *dlabel, struct vnode *vp, struct label *label,
struct componentname *cnp)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_deleteacl(struct ucred *cred, struct vnode *vp,
struct label *label, acl_type_t type)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
static int
mac_mls_check_vnode_deleteextattr(struct ucred *cred, struct vnode *vp,
struct label *label, int attrnamespace, const char *name)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
return (EACCES);
return (0);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_check_vnode_exec(struct ucred *cred, struct vnode *vp,
Update MAC modules for changes in arguments for exec MAC policy entry points to include an explicit execlabel. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-11-08 18:04:36 +00:00
struct label *label, struct image_params *imgp,
struct label *execlabel)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
Update MAC modules for changes in arguments for exec MAC policy entry points to include an explicit execlabel. Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-11-08 18:04:36 +00:00
struct mac_mls *subj, *obj, *exec;
int error;
if (execlabel != NULL) {
/*
* We currently don't permit labels to be changed at
* exec-time as part of MLS, so disallow non-NULL
* MLS label elements in the execlabel.
*/
exec = SLOT(execlabel);
error = mls_atmostflags(exec, 0);
if (error)
return (error);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_getacl(struct ucred *cred, struct vnode *vp,
struct label *label, acl_type_t type)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_getextattr(struct ucred *cred, struct vnode *vp,
struct label *label, int attrnamespace, const char *name, struct uio *uio)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
static int
Merge implementation of mpo_check_vnode_link() for various appropriate file-system aware MAC policies. Sync to MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:25:48 +00:00
mac_mls_check_vnode_link(struct ucred *cred, struct vnode *dvp,
struct label *dlabel, struct vnode *vp, struct label *label,
struct componentname *cnp)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Merge implementation of mpo_check_vnode_link() for various appropriate file-system aware MAC policies. Sync to MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:25:48 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Merge implementation of mpo_check_vnode_link() for various appropriate file-system aware MAC policies. Sync to MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:25:48 +00:00
return (EACCES);
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Merge implementation of mpo_check_vnode_link() for various appropriate file-system aware MAC policies. Sync to MAC tree. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-05 18:25:48 +00:00
return (EACCES);
return (0);
}
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
static int
mac_mls_check_vnode_listextattr(struct ucred *cred, struct vnode *vp,
struct label *label, int attrnamespace)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
return (EACCES);
return (0);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
mac_mls_check_vnode_lookup(struct ucred *cred, struct vnode *dvp,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
struct label *dlabel, struct componentname *cnp)
{
struct mac_mls *subj, *obj;
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
if (!mac_mls_enabled)
return (0);
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
Trim trailing whitespace from the ends of lines.
2002-09-21 19:26:59 +00:00
return (0);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
Sync from MAC tree: break out the single mmap entry point into seperate entry points for each occasion: mac_check_vnode_mmap() Check at initial mapping mac_check_vnode_mprotect() Check at mapping protection change mac_check_vnode_mmap_downgrade() Determine if a mapping downgrade should take place following subject relabel. Implement mmap() and mprotect() entry points for labeled vnode policies. These entry points are currently not hooked up to the VM system in the base tree. These changes improve the consistency of the access control interface and offer more flexibility regarding limiting access to vnode mmaping. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-06 02:46:26 +00:00
static int
mac_mls_check_vnode_mmap(struct ucred *cred, struct vnode *vp,
Move MAC check_vnode_mmap entry point out from being exclusive to MAP_SHARED so that the entry point gets executed un-conditionally. This may be useful for security policies which want to perform access control checks around run-time linking. -add the mmap(2) flags argument to the check_vnode_mmap entry point so that we can make access control decisions based on the type of mapped object. -update any dependent API around this parameter addition such as function prototype modifications, entry point parameter additions and the inclusion of sys/mman.h header file. -Change the MLS, BIBA and LOMAC security policies so that subject domination routines are not executed unless the type of mapping is shared. This is done to maintain compatibility between the old vm_mmap_vnode(9) and these policies. Reviewed by: rwatson MFC after: 1 month
2005-04-14 16:03:30 +00:00
struct label *label, int prot, int flags)
Sync from MAC tree: break out the single mmap entry point into seperate entry points for each occasion: mac_check_vnode_mmap() Check at initial mapping mac_check_vnode_mprotect() Check at mapping protection change mac_check_vnode_mmap_downgrade() Determine if a mapping downgrade should take place following subject relabel. Implement mmap() and mprotect() entry points for labeled vnode policies. These entry points are currently not hooked up to the VM system in the base tree. These changes improve the consistency of the access control interface and offer more flexibility regarding limiting access to vnode mmaping. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-06 02:46:26 +00:00
{
struct mac_mls *subj, *obj;
/*
* Rely on the use of open()-time protections to handle
* non-revocation cases.
*/
More in the way of minor consistency improvements: trim 'mac_mls_' from another variable to line this up with mac_biba.c Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:14:30 +00:00
if (!mac_mls_enabled || !revocation_enabled)
Sync from MAC tree: break out the single mmap entry point into seperate entry points for each occasion: mac_check_vnode_mmap() Check at initial mapping mac_check_vnode_mprotect() Check at mapping protection change mac_check_vnode_mmap_downgrade() Determine if a mapping downgrade should take place following subject relabel. Implement mmap() and mprotect() entry points for labeled vnode policies. These entry points are currently not hooked up to the VM system in the base tree. These changes improve the consistency of the access control interface and offer more flexibility regarding limiting access to vnode mmaping. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-06 02:46:26 +00:00
return (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
subj = SLOT(cred->cr_label);
Sync from MAC tree: break out the single mmap entry point into seperate entry points for each occasion: mac_check_vnode_mmap() Check at initial mapping mac_check_vnode_mprotect() Check at mapping protection change mac_check_vnode_mmap_downgrade() Determine if a mapping downgrade should take place following subject relabel. Implement mmap() and mprotect() entry points for labeled vnode policies. These entry points are currently not hooked up to the VM system in the base tree. These changes improve the consistency of the access control interface and offer more flexibility regarding limiting access to vnode mmaping. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-06 02:46:26 +00:00
obj = SLOT(label);
if (prot & (VM_PROT_READ | VM_PROT_EXECUTE)) {
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Sync from MAC tree: break out the single mmap entry point into seperate entry points for each occasion: mac_check_vnode_mmap() Check at initial mapping mac_check_vnode_mprotect() Check at mapping protection change mac_check_vnode_mmap_downgrade() Determine if a mapping downgrade should take place following subject relabel. Implement mmap() and mprotect() entry points for labeled vnode policies. These entry points are currently not hooked up to the VM system in the base tree. These changes improve the consistency of the access control interface and offer more flexibility regarding limiting access to vnode mmaping. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-06 02:46:26 +00:00
return (EACCES);
}
Move MAC check_vnode_mmap entry point out from being exclusive to MAP_SHARED so that the entry point gets executed un-conditionally. This may be useful for security policies which want to perform access control checks around run-time linking. -add the mmap(2) flags argument to the check_vnode_mmap entry point so that we can make access control decisions based on the type of mapped object. -update any dependent API around this parameter addition such as function prototype modifications, entry point parameter additions and the inclusion of sys/mman.h header file. -Change the MLS, BIBA and LOMAC security policies so that subject domination routines are not executed unless the type of mapping is shared. This is done to maintain compatibility between the old vm_mmap_vnode(9) and these policies. Reviewed by: rwatson MFC after: 1 month
2005-04-14 16:03:30 +00:00
if (((prot & VM_PROT_WRITE) != 0) && ((flags & MAP_SHARED) != 0)) {
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Sync from MAC tree: break out the single mmap entry point into seperate entry points for each occasion: mac_check_vnode_mmap() Check at initial mapping mac_check_vnode_mprotect() Check at mapping protection change mac_check_vnode_mmap_downgrade() Determine if a mapping downgrade should take place following subject relabel. Implement mmap() and mprotect() entry points for labeled vnode policies. These entry points are currently not hooked up to the VM system in the base tree. These changes improve the consistency of the access control interface and offer more flexibility regarding limiting access to vnode mmaping. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-06 02:46:26 +00:00
return (EACCES);
}
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (0);
Sync from MAC tree: break out the single mmap entry point into seperate entry points for each occasion: mac_check_vnode_mmap() Check at initial mapping mac_check_vnode_mprotect() Check at mapping protection change mac_check_vnode_mmap_downgrade() Determine if a mapping downgrade should take place following subject relabel. Implement mmap() and mprotect() entry points for labeled vnode policies. These entry points are currently not hooked up to the VM system in the base tree. These changes improve the consistency of the access control interface and offer more flexibility regarding limiting access to vnode mmaping. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-06 02:46:26 +00:00
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_check_vnode_open(struct ucred *cred, struct vnode *vp,
While 'mode_t' seemed like a good idea for the access mode argument for MAC access() and open() checks, the argument actually has an int type where it becomes available. Switch to using 'int' for the mode argument throughout the MAC Framework and policy modules. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 17:56:57 +00:00
struct label *vnodelabel, int acc_mode)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
/* XXX privilege override for admin? */
if (acc_mode & (VREAD | VEXEC | VSTAT)) {
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
}
if (acc_mode & (VWRITE | VAPPEND | VADMIN)) {
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
}
return (0);
}
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
static int
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
mac_mls_check_vnode_poll(struct ucred *active_cred, struct ucred *file_cred,
struct vnode *vp, struct label *label)
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
{
struct mac_mls *subj, *obj;
More in the way of minor consistency improvements: trim 'mac_mls_' from another variable to line this up with mac_biba.c Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:14:30 +00:00
if (!mac_mls_enabled || !revocation_enabled)
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
return (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
subj = SLOT(active_cred->cr_label);
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
return (EACCES);
return (0);
}
static int
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
mac_mls_check_vnode_read(struct ucred *active_cred, struct ucred *file_cred,
struct vnode *vp, struct label *label)
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
{
struct mac_mls *subj, *obj;
More in the way of minor consistency improvements: trim 'mac_mls_' from another variable to line this up with mac_biba.c Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:14:30 +00:00
if (!mac_mls_enabled || !revocation_enabled)
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
return (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
subj = SLOT(active_cred->cr_label);
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
return (EACCES);
return (0);
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
static int
mac_mls_check_vnode_readdir(struct ucred *cred, struct vnode *dvp,
struct label *dlabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_readlink(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_relabel(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel, struct label *newlabel)
{
struct mac_mls *old, *new, *subj;
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
int error;
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
old = SLOT(vnodelabel);
new = SLOT(newlabel);
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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If there is an MLS label update for the vnode, it must be a
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
* effective label.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
error = mls_atmostflags(new, MAC_MLS_FLAG_EFFECTIVE);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (error)
return (error);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* To perform a relabel of the vnode (MLS label or not), MLS must
* authorize the relabel.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_effective_in_range(old, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EPERM);
/*
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
* If the MLS label is to be changed, authorize as appropriate.
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (new->mm_flags & MAC_MLS_FLAG_EFFECTIVE) {
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
/*
* To change the MLS label on a vnode, the new vnode label
* must be in the subject range.
*/
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_effective_in_range(new, subj))
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (EPERM);
/*
* To change the MLS label on the vnode to be EQUAL,
* the subject must have appropriate privilege.
*/
if (mac_mls_contains_equal(new)) {
Improve consistency with the Biba policy -- rename mac_mls_subject_equal_ok() to mac_mls_subject_privileged(), which more consistently reflects the fact that this is really about our notion of privilege in the MLS policy. Since we don't use suser() for privilege in MLS, remove the suser check from the ifnet relabel ioctl, and replace it with an MLS privilege check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-31 20:00:06 +00:00
error = mac_mls_subject_privileged(subj);
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
if (error)
return (error);
}
}
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
Cleanup of relabel authorization checks -- almost identical logic, we just break out some of the tests better. Minor change in that we now better support incremental update of labels. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 16:35:54 +00:00
return (0);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
}
static int
mac_mls_check_vnode_rename_from(struct ucred *cred, struct vnode *dvp,
struct label *dlabel, struct vnode *vp, struct label *label,
struct componentname *cnp)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_rename_to(struct ucred *cred, struct vnode *dvp,
struct label *dlabel, struct vnode *vp, struct label *label, int samedir,
struct componentname *cnp)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(dlabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
if (vp != NULL) {
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
}
return (0);
}
static int
mac_mls_check_vnode_revoke(struct ucred *cred, struct vnode *vp,
struct label *label)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_setacl(struct ucred *cred, struct vnode *vp,
struct label *label, acl_type_t type, struct acl *acl)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_setextattr(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel, int attrnamespace, const char *name,
struct uio *uio)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
/* XXX: protect the MAC EA in a special way? */
return (0);
}
static int
mac_mls_check_vnode_setflags(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel, u_long flags)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_setmode(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel, mode_t mode)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_setowner(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel, uid_t uid, gid_t gid)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
mac_mls_check_vnode_setutimes(struct ucred *cred, struct vnode *vp,
struct label *vnodelabel, struct timespec atime, struct timespec mtime)
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
static int
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
mac_mls_check_vnode_stat(struct ucred *active_cred, struct ucred *file_cred,
struct vnode *vp, struct label *vnodelabel)
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
{
struct mac_mls *subj, *obj;
if (!mac_mls_enabled)
return (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
subj = SLOT(active_cred->cr_label);
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
obj = SLOT(vnodelabel);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(subj, obj))
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
return (EACCES);
return (0);
}
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
static int
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
mac_mls_check_vnode_write(struct ucred *active_cred, struct ucred *file_cred,
struct vnode *vp, struct label *label)
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
{
struct mac_mls *subj, *obj;
More in the way of minor consistency improvements: trim 'mac_mls_' from another variable to line this up with mac_biba.c Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-21 18:14:30 +00:00
if (!mac_mls_enabled || !revocation_enabled)
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
return (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
subj = SLOT(active_cred->cr_label);
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
obj = SLOT(label);
Rename Biba and MLS _single label elements to _effective, which more accurately represents the intention of the 'single' label element in Biba and MLS labels. It also approximates the use of 'effective' in traditional UNIX credentials, and avoids confusion with 'singlelabel' in the context of file systems. Inspired by: trhodes
2004-07-16 02:03:50 +00:00
if (!mac_mls_dominate_effective(obj, subj))
Break out mac_check_vnode_op() into three seperate checks: mac_check_vnode_poll(), mac_check_vnode_read(), mac_check_vnode_write(). This improves the consistency with other existing vnode checks, and allows policies to avoid implementing switch statements to determine what operations they do and do not want to authorize. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 16:43:25 +00:00
return (EACCES);
return (0);
}
Fix panic associated with file creation via RPC/NFS when the MLS policy is loaded. This problem stems from the fact that the policy is not properly initializing the mac label associated with the NFS daemon. Obtained from: TrustedBSD Project Discussed with: rwatson
2006-08-26 20:13:35 +00:00
static void
mac_mls_associate_nfsd_label(struct ucred *cred)
{
struct mac_mls *label;
label = SLOT(cred->cr_label);
mac_mls_set_effective(label, MAC_MLS_TYPE_LOW, 0, NULL);
mac_mls_set_range(label, MAC_MLS_TYPE_LOW, 0, NULL,
MAC_MLS_TYPE_HIGH, 0, NULL);
}
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
static struct mac_policy_ops mac_mls_ops =
{
.mpo_init = mac_mls_init,
.mpo_init_bpfdesc_label = mac_mls_init_label,
.mpo_init_cred_label = mac_mls_init_label,
.mpo_init_devfsdirent_label = mac_mls_init_label,
.mpo_init_ifnet_label = mac_mls_init_label,
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
.mpo_init_inpcb_label = mac_mls_init_label_waitcheck,
Teach the MAC policies which utilize mbuf labeling the new syncache entry points. Properly initialize the mbuf label based on the label we copy from the PCB. This fixes an LOR between the PCB and syncache code.
2006-12-13 06:03:22 +00:00
.mpo_init_syncache_label = mac_mls_init_label_waitcheck,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_init_sysv_msgmsg_label = mac_mls_init_label,
.mpo_init_sysv_msgqueue_label = mac_mls_init_label,
Gratuitous renaming of four System V Semaphore MAC Framework entry points to convert _sema() to _sem() for consistency purposes with respect to the other semaphore-related entry points: mac_init_sysv_sema() -> mac_init_sysv_sem() mac_destroy_sysv_sem() -> mac_destroy_sysv_sem() mac_create_sysv_sema() -> mac_create_sysv_sem() mac_cleanup_sysv_sema() -> mac_cleanup_sysv_sem() Congruent changes are made to the policy interface to support this. Obtained from: TrustedBSD Project Sponsored by: SPAWAR, SPARTA
2005-06-07 05:03:28 +00:00
.mpo_init_sysv_sem_label = mac_mls_init_label,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_init_sysv_shm_label = mac_mls_init_label,
Modify the mac_init_ipq() MAC Framework entry point to accept an additional flags argument to indicate blocking disposition, and pass in M_NOWAIT from the IP reassembly code to indicate that blocking is not OK when labeling a new IP fragment reassembly queue. This should eliminate some of the WITNESS warnings that have started popping up since fine-grained IP stack locking started going in; if memory allocation fails, the creation of the fragment queue will be aborted. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-03-26 15:12:03 +00:00
.mpo_init_ipq_label = mac_mls_init_label_waitcheck,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_init_mbuf_label = mac_mls_init_label_waitcheck,
.mpo_init_mount_label = mac_mls_init_label,
.mpo_init_mount_fs_label = mac_mls_init_label,
.mpo_init_pipe_label = mac_mls_init_label,
Introduce MAC Framework and MAC Policy entry points to label and control access to POSIX Semaphores: mac_init_posix_sem() Initialize label for POSIX semaphore mac_create_posix_sem() Create POSIX semaphore mac_destroy_posix_sem() Destroy POSIX semaphore mac_check_posix_sem_destroy() Check whether semaphore may be destroyed mac_check_posix_sem_getvalue() Check whether semaphore may be queried mac_check_possix_sem_open() Check whether semaphore may be opened mac_check_posix_sem_post() Check whether semaphore may be posted to mac_check_posix_sem_unlink() Check whether semaphore may be unlinked mac_check_posix_sem_wait() Check whether may wait on semaphore Update Biba, MLS, Stub, and Test policies to implement these entry points. For information flow policies, most semaphore operations are effectively read/write. Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Sponsored by: DARPA, McAfee, SPARTA Obtained from: TrustedBSD Project
2005-05-04 10:39:15 +00:00
.mpo_init_posix_sem_label = mac_mls_init_label,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_init_socket_label = mac_mls_init_label_waitcheck,
.mpo_init_socket_peer_label = mac_mls_init_label_waitcheck,
.mpo_init_vnode_label = mac_mls_init_label,
.mpo_destroy_bpfdesc_label = mac_mls_destroy_label,
.mpo_destroy_cred_label = mac_mls_destroy_label,
.mpo_destroy_devfsdirent_label = mac_mls_destroy_label,
.mpo_destroy_ifnet_label = mac_mls_destroy_label,
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
.mpo_destroy_inpcb_label = mac_mls_destroy_label,
Teach the MAC policies which utilize mbuf labeling the new syncache entry points. Properly initialize the mbuf label based on the label we copy from the PCB. This fixes an LOR between the PCB and syncache code.
2006-12-13 06:03:22 +00:00
.mpo_destroy_syncache_label = mac_mls_destroy_label,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_destroy_sysv_msgmsg_label = mac_mls_destroy_label,
.mpo_destroy_sysv_msgqueue_label = mac_mls_destroy_label,
Gratuitous renaming of four System V Semaphore MAC Framework entry points to convert _sema() to _sem() for consistency purposes with respect to the other semaphore-related entry points: mac_init_sysv_sema() -> mac_init_sysv_sem() mac_destroy_sysv_sem() -> mac_destroy_sysv_sem() mac_create_sysv_sema() -> mac_create_sysv_sem() mac_cleanup_sysv_sema() -> mac_cleanup_sysv_sem() Congruent changes are made to the policy interface to support this. Obtained from: TrustedBSD Project Sponsored by: SPAWAR, SPARTA
2005-06-07 05:03:28 +00:00
.mpo_destroy_sysv_sem_label = mac_mls_destroy_label,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_destroy_sysv_shm_label = mac_mls_destroy_label,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_destroy_ipq_label = mac_mls_destroy_label,
.mpo_destroy_mbuf_label = mac_mls_destroy_label,
.mpo_destroy_mount_label = mac_mls_destroy_label,
.mpo_destroy_mount_fs_label = mac_mls_destroy_label,
.mpo_destroy_pipe_label = mac_mls_destroy_label,
Introduce MAC Framework and MAC Policy entry points to label and control access to POSIX Semaphores: mac_init_posix_sem() Initialize label for POSIX semaphore mac_create_posix_sem() Create POSIX semaphore mac_destroy_posix_sem() Destroy POSIX semaphore mac_check_posix_sem_destroy() Check whether semaphore may be destroyed mac_check_posix_sem_getvalue() Check whether semaphore may be queried mac_check_possix_sem_open() Check whether semaphore may be opened mac_check_posix_sem_post() Check whether semaphore may be posted to mac_check_posix_sem_unlink() Check whether semaphore may be unlinked mac_check_posix_sem_wait() Check whether may wait on semaphore Update Biba, MLS, Stub, and Test policies to implement these entry points. For information flow policies, most semaphore operations are effectively read/write. Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Sponsored by: DARPA, McAfee, SPARTA Obtained from: TrustedBSD Project
2005-05-04 10:39:15 +00:00
.mpo_destroy_posix_sem_label = mac_mls_destroy_label,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_destroy_socket_label = mac_mls_destroy_label,
.mpo_destroy_socket_peer_label = mac_mls_destroy_label,
.mpo_destroy_vnode_label = mac_mls_destroy_label,
Rename mac_create_cred() MAC Framework entry point to mac_copy_cred(), and the mpo_create_cred() MAC policy entry point to mpo_copy_cred_label(). This is more consistent with similar entry points for creation and label copying, as mac_create_cred() was called from crdup() as opposed to during process creation. For a number of policies, this removes the requirement for special handling when copying credential labels, and improves consistency. Approved by: re (scottl) Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-12-06 21:48:03 +00:00
.mpo_copy_cred_label = mac_mls_copy_label,
Introduce a temporary mutex, mac_ifnet_mtx, to lock MAC labels on network interfaces. This global mutex will protect all ifnet labels. Acquire the mutex across various MAC activities on interfaces, such as security checks, propagating interface labels to mbufs generated from the interface, retrieving and setting the interface label. Introduce mpo_copy_ifnet_label MAC policy entry point to copy the value of an interface label from one label to another. Use this to avoid performing a label externalize while holding mac_ifnet_mtx; copy the label to a temporary ifnet label and then externalize that. Implement mpo_copy_ifnet_label for various MAC policies that implement interface labeling using generic label copying routines. Obtained from: TrustedBSD Project Sponsored by: DARPA, McAfee Research
2004-06-24 03:34:46 +00:00
.mpo_copy_ifnet_label = mac_mls_copy_label,
Use mac_biba_label_copy() and mac_mls_label_copy() to implement the mpo_copy_mbuf_label() entry point for Biba and MLS, respectively. Otherwise, labels in m_tags may not be properly propagated across some classes of mbuf operations. This problem caused these policies to fail-stop the system with a panic. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-02 17:21:38 +00:00
.mpo_copy_mbuf_label = mac_mls_copy_label,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_copy_pipe_label = mac_mls_copy_label,
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
.mpo_copy_socket_label = mac_mls_copy_label,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_copy_vnode_label = mac_mls_copy_label,
.mpo_externalize_cred_label = mac_mls_externalize_label,
.mpo_externalize_ifnet_label = mac_mls_externalize_label,
.mpo_externalize_pipe_label = mac_mls_externalize_label,
.mpo_externalize_socket_label = mac_mls_externalize_label,
.mpo_externalize_socket_peer_label = mac_mls_externalize_label,
.mpo_externalize_vnode_label = mac_mls_externalize_label,
.mpo_internalize_cred_label = mac_mls_internalize_label,
.mpo_internalize_ifnet_label = mac_mls_internalize_label,
.mpo_internalize_pipe_label = mac_mls_internalize_label,
.mpo_internalize_socket_label = mac_mls_internalize_label,
.mpo_internalize_vnode_label = mac_mls_internalize_label,
.mpo_create_devfs_device = mac_mls_create_devfs_device,
.mpo_create_devfs_directory = mac_mls_create_devfs_directory,
.mpo_create_devfs_symlink = mac_mls_create_devfs_symlink,
.mpo_create_mount = mac_mls_create_mount,
.mpo_relabel_vnode = mac_mls_relabel_vnode,
.mpo_update_devfsdirent = mac_mls_update_devfsdirent,
.mpo_associate_vnode_devfs = mac_mls_associate_vnode_devfs,
.mpo_associate_vnode_extattr = mac_mls_associate_vnode_extattr,
.mpo_associate_vnode_singlelabel = mac_mls_associate_vnode_singlelabel,
.mpo_create_vnode_extattr = mac_mls_create_vnode_extattr,
.mpo_setlabel_vnode_extattr = mac_mls_setlabel_vnode_extattr,
.mpo_create_mbuf_from_socket = mac_mls_create_mbuf_from_socket,
Teach the MAC policies which utilize mbuf labeling the new syncache entry points. Properly initialize the mbuf label based on the label we copy from the PCB. This fixes an LOR between the PCB and syncache code.
2006-12-13 06:03:22 +00:00
.mpo_create_mbuf_from_syncache = mac_mls_create_mbuf_from_syncache,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_create_pipe = mac_mls_create_pipe,
Introduce MAC Framework and MAC Policy entry points to label and control access to POSIX Semaphores: mac_init_posix_sem() Initialize label for POSIX semaphore mac_create_posix_sem() Create POSIX semaphore mac_destroy_posix_sem() Destroy POSIX semaphore mac_check_posix_sem_destroy() Check whether semaphore may be destroyed mac_check_posix_sem_getvalue() Check whether semaphore may be queried mac_check_possix_sem_open() Check whether semaphore may be opened mac_check_posix_sem_post() Check whether semaphore may be posted to mac_check_posix_sem_unlink() Check whether semaphore may be unlinked mac_check_posix_sem_wait() Check whether may wait on semaphore Update Biba, MLS, Stub, and Test policies to implement these entry points. For information flow policies, most semaphore operations are effectively read/write. Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Sponsored by: DARPA, McAfee, SPARTA Obtained from: TrustedBSD Project
2005-05-04 10:39:15 +00:00
.mpo_create_posix_sem = mac_mls_create_posix_sem,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_create_socket = mac_mls_create_socket,
.mpo_create_socket_from_socket = mac_mls_create_socket_from_socket,
.mpo_relabel_pipe = mac_mls_relabel_pipe,
.mpo_relabel_socket = mac_mls_relabel_socket,
.mpo_set_socket_peer_from_mbuf = mac_mls_set_socket_peer_from_mbuf,
.mpo_set_socket_peer_from_socket = mac_mls_set_socket_peer_from_socket,
.mpo_create_bpfdesc = mac_mls_create_bpfdesc,
.mpo_create_datagram_from_ipq = mac_mls_create_datagram_from_ipq,
.mpo_create_fragment = mac_mls_create_fragment,
.mpo_create_ifnet = mac_mls_create_ifnet,
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
.mpo_create_inpcb_from_socket = mac_mls_create_inpcb_from_socket,
Teach the MAC policies which utilize mbuf labeling the new syncache entry points. Properly initialize the mbuf label based on the label we copy from the PCB. This fixes an LOR between the PCB and syncache code.
2006-12-13 06:03:22 +00:00
.mpo_init_syncache_from_inpcb = mac_mls_init_syncache_from_inpcb,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_create_ipq = mac_mls_create_ipq,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_create_sysv_msgmsg = mac_mls_create_sysv_msgmsg,
.mpo_create_sysv_msgqueue = mac_mls_create_sysv_msgqueue,
Gratuitous renaming of four System V Semaphore MAC Framework entry points to convert _sema() to _sem() for consistency purposes with respect to the other semaphore-related entry points: mac_init_sysv_sema() -> mac_init_sysv_sem() mac_destroy_sysv_sem() -> mac_destroy_sysv_sem() mac_create_sysv_sema() -> mac_create_sysv_sem() mac_cleanup_sysv_sema() -> mac_cleanup_sysv_sem() Congruent changes are made to the policy interface to support this. Obtained from: TrustedBSD Project Sponsored by: SPAWAR, SPARTA
2005-06-07 05:03:28 +00:00
.mpo_create_sysv_sem = mac_mls_create_sysv_sem,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_create_sysv_shm = mac_mls_create_sysv_shm,
Switch TCP over to using the inpcb label when responding in timed wait, rather than the socket label. This avoids reaching up to the socket layer during connection close, which requires locking changes. To do this, introduce MAC Framework entry point mac_create_mbuf_from_inpcb(), which is called from tcp_twrespond() instead of calling mac_create_mbuf_from_socket() or mac_create_mbuf_netlayer(). Introduce MAC Policy entry point mpo_create_mbuf_from_inpcb(), and implementations for various policies, which generally just copy label data from the inpcb to the mbuf. Assert the inpcb lock in the entry point since we require consistency for the inpcb label reference. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-12-17 14:55:11 +00:00
.mpo_create_mbuf_from_inpcb = mac_mls_create_mbuf_from_inpcb,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_create_mbuf_linklayer = mac_mls_create_mbuf_linklayer,
.mpo_create_mbuf_from_bpfdesc = mac_mls_create_mbuf_from_bpfdesc,
.mpo_create_mbuf_from_ifnet = mac_mls_create_mbuf_from_ifnet,
.mpo_create_mbuf_multicast_encap = mac_mls_create_mbuf_multicast_encap,
.mpo_create_mbuf_netlayer = mac_mls_create_mbuf_netlayer,
.mpo_fragment_match = mac_mls_fragment_match,
.mpo_relabel_ifnet = mac_mls_relabel_ifnet,
.mpo_update_ipq = mac_mls_update_ipq,
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
.mpo_inpcb_sosetlabel = mac_mls_inpcb_sosetlabel,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_create_proc0 = mac_mls_create_proc0,
.mpo_create_proc1 = mac_mls_create_proc1,
.mpo_relabel_cred = mac_mls_relabel_cred,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_cleanup_sysv_msgmsg = mac_mls_cleanup_sysv_msgmsg,
.mpo_cleanup_sysv_msgqueue = mac_mls_cleanup_sysv_msgqueue,
Gratuitous renaming of four System V Semaphore MAC Framework entry points to convert _sema() to _sem() for consistency purposes with respect to the other semaphore-related entry points: mac_init_sysv_sema() -> mac_init_sysv_sem() mac_destroy_sysv_sem() -> mac_destroy_sysv_sem() mac_create_sysv_sema() -> mac_create_sysv_sem() mac_cleanup_sysv_sema() -> mac_cleanup_sysv_sem() Congruent changes are made to the policy interface to support this. Obtained from: TrustedBSD Project Sponsored by: SPAWAR, SPARTA
2005-06-07 05:03:28 +00:00
.mpo_cleanup_sysv_sem = mac_mls_cleanup_sysv_sem,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_cleanup_sysv_shm = mac_mls_cleanup_sysv_shm,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_check_bpfdesc_receive = mac_mls_check_bpfdesc_receive,
.mpo_check_cred_relabel = mac_mls_check_cred_relabel,
.mpo_check_cred_visible = mac_mls_check_cred_visible,
.mpo_check_ifnet_relabel = mac_mls_check_ifnet_relabel,
.mpo_check_ifnet_transmit = mac_mls_check_ifnet_transmit,
Introduce a MAC label reference in 'struct inpcb', which caches the MAC label referenced from 'struct socket' in the IPv4 and IPv6-based protocols. This permits MAC labels to be checked during network delivery operations without dereferencing inp->inp_socket to get to so->so_label, which will eventually avoid our having to grab the socket lock during delivery at the network layer. This change introduces 'struct inpcb' as a labeled object to the MAC Framework, along with the normal circus of entry points: initialization, creation from socket, destruction, as well as a delivery access control check. For most policies, the inpcb label will simply be a cache of the socket label, so a new protocol switch method is introduced, pr_sosetlabel() to notify protocols that the socket layer label has been updated so that the cache can be updated while holding appropriate locks. Most protocols implement this using pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use the the worker function in_pcbsosetlabel(), which calls into the MAC Framework to perform a cache update. Biba, LOMAC, and MLS implement these entry points, as do the stub policy, and test policy. Reviewed by: sam, bms Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-11-18 00:39:07 +00:00
.mpo_check_inpcb_deliver = mac_mls_check_inpcb_deliver,
Implement MLS confidentiality protection for System V IPC objects (message queues, semaphores, shared memory). Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Obtained from: TrustedBSD Project Sponsored by: DARPA, SPAWAR, McAfee Research
2005-01-22 20:11:16 +00:00
.mpo_check_sysv_msgrcv = mac_mls_check_sysv_msgrcv,
.mpo_check_sysv_msgrmid = mac_mls_check_sysv_msgrmid,
.mpo_check_sysv_msqget = mac_mls_check_sysv_msqget,
.mpo_check_sysv_msqsnd = mac_mls_check_sysv_msqsnd,
.mpo_check_sysv_msqrcv = mac_mls_check_sysv_msqrcv,
.mpo_check_sysv_msqctl = mac_mls_check_sysv_msqctl,
.mpo_check_sysv_semctl = mac_mls_check_sysv_semctl,
.mpo_check_sysv_semget = mac_mls_check_sysv_semget,
.mpo_check_sysv_semop = mac_mls_check_sysv_semop,
.mpo_check_sysv_shmat = mac_mls_check_sysv_shmat,
.mpo_check_sysv_shmctl = mac_mls_check_sysv_shmctl,
.mpo_check_sysv_shmget = mac_mls_check_sysv_shmget,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_check_mount_stat = mac_mls_check_mount_stat,
.mpo_check_pipe_ioctl = mac_mls_check_pipe_ioctl,
.mpo_check_pipe_poll = mac_mls_check_pipe_poll,
.mpo_check_pipe_read = mac_mls_check_pipe_read,
.mpo_check_pipe_relabel = mac_mls_check_pipe_relabel,
.mpo_check_pipe_stat = mac_mls_check_pipe_stat,
.mpo_check_pipe_write = mac_mls_check_pipe_write,
Introduce MAC Framework and MAC Policy entry points to label and control access to POSIX Semaphores: mac_init_posix_sem() Initialize label for POSIX semaphore mac_create_posix_sem() Create POSIX semaphore mac_destroy_posix_sem() Destroy POSIX semaphore mac_check_posix_sem_destroy() Check whether semaphore may be destroyed mac_check_posix_sem_getvalue() Check whether semaphore may be queried mac_check_possix_sem_open() Check whether semaphore may be opened mac_check_posix_sem_post() Check whether semaphore may be posted to mac_check_posix_sem_unlink() Check whether semaphore may be unlinked mac_check_posix_sem_wait() Check whether may wait on semaphore Update Biba, MLS, Stub, and Test policies to implement these entry points. For information flow policies, most semaphore operations are effectively read/write. Submitted by: Dandekar Hrishikesh <rishi_dandekar at sbcglobal dot net> Sponsored by: DARPA, McAfee, SPARTA Obtained from: TrustedBSD Project
2005-05-04 10:39:15 +00:00
.mpo_check_posix_sem_destroy = mac_mls_check_posix_sem_write,
.mpo_check_posix_sem_getvalue = mac_mls_check_posix_sem_rdonly,
.mpo_check_posix_sem_open = mac_mls_check_posix_sem_write,
.mpo_check_posix_sem_post = mac_mls_check_posix_sem_write,
.mpo_check_posix_sem_unlink = mac_mls_check_posix_sem_write,
.mpo_check_posix_sem_wait = mac_mls_check_posix_sem_write,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_check_proc_debug = mac_mls_check_proc_debug,
.mpo_check_proc_sched = mac_mls_check_proc_sched,
.mpo_check_proc_signal = mac_mls_check_proc_signal,
.mpo_check_socket_deliver = mac_mls_check_socket_deliver,
.mpo_check_socket_relabel = mac_mls_check_socket_relabel,
.mpo_check_socket_visible = mac_mls_check_socket_visible,
Expand scope of the MLS policy to include a new entry point available for enforcement: mac_mls_check_system_swapon() - Require that the subject and the swapfile target vnode labels dominate one another. An additional check is probably needed here to require that the swapfile target has a label of mls/high to prevent information leakage through swapfiles. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-03-25 01:16:10 +00:00
.mpo_check_system_swapon = mac_mls_check_system_swapon,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_check_vnode_access = mac_mls_check_vnode_open,
.mpo_check_vnode_chdir = mac_mls_check_vnode_chdir,
.mpo_check_vnode_chroot = mac_mls_check_vnode_chroot,
.mpo_check_vnode_create = mac_mls_check_vnode_create,
.mpo_check_vnode_delete = mac_mls_check_vnode_delete,
.mpo_check_vnode_deleteacl = mac_mls_check_vnode_deleteacl,
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
.mpo_check_vnode_deleteextattr = mac_mls_check_vnode_deleteextattr,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_check_vnode_exec = mac_mls_check_vnode_exec,
.mpo_check_vnode_getacl = mac_mls_check_vnode_getacl,
.mpo_check_vnode_getextattr = mac_mls_check_vnode_getextattr,
.mpo_check_vnode_link = mac_mls_check_vnode_link,
Implementations of mpo_check_vnode_deleteextattr() and mpo_check_vnode_listextattr() for Biba, MLS, and BSD Extended. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-08-21 14:34:54 +00:00
.mpo_check_vnode_listextattr = mac_mls_check_vnode_listextattr,
Move to C99 sparse structure initialization for the mac_policy_ops structure definition, rather than using an operation vector we translate into the structure. Originally, we used a vector for two reasons: (1) We wanted to define the structure sparsely, which wasn't supported by the C compiler for structures. For a policy with five entry points, you don't want to have to stick in a few hundred NULL function pointers. (2) We thought it would improve ABI compatibility allowing modules to work with kernels that had a superset of the entry points defined in the module, even if the kernel had changed its entry point set. Both of these no longer apply: (1) C99 gives us a way to sparsely define a static structure. (2) The ABI problems existed anyway, due to enumeration numbers, argument changes, and semantic mismatches. Since the going rule for FreeBSD is that you really need your modules to pretty closely match your kernel, it's not worth the complexity. This submit eliminates the operation vector, dynamic allocation of the operation structure, copying of the vector to the structure, and redoes the vectors in each policy to direct structure definitions. One enourmous benefit of this change is that we now get decent type checking on policy entry point implementation arguments. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-30 18:48:51 +00:00
.mpo_check_vnode_lookup = mac_mls_check_vnode_lookup,
.mpo_check_vnode_mmap = mac_mls_check_vnode_mmap,
.mpo_check_vnode_open = mac_mls_check_vnode_open,
.mpo_check_vnode_poll = mac_mls_check_vnode_poll,
.mpo_check_vnode_read = mac_mls_check_vnode_read,
.mpo_check_vnode_readdir = mac_mls_check_vnode_readdir,
.mpo_check_vnode_readlink = mac_mls_check_vnode_readlink,
.mpo_check_vnode_relabel = mac_mls_check_vnode_relabel,
.mpo_check_vnode_rename_from = mac_mls_check_vnode_rename_from,
.mpo_check_vnode_rename_to = mac_mls_check_vnode_rename_to,
.mpo_check_vnode_revoke = mac_mls_check_vnode_revoke,
.mpo_check_vnode_setacl = mac_mls_check_vnode_setacl,
.mpo_check_vnode_setextattr = mac_mls_check_vnode_setextattr,
.mpo_check_vnode_setflags = mac_mls_check_vnode_setflags,
.mpo_check_vnode_setmode = mac_mls_check_vnode_setmode,
.mpo_check_vnode_setowner = mac_mls_check_vnode_setowner,
.mpo_check_vnode_setutimes = mac_mls_check_vnode_setutimes,
.mpo_check_vnode_stat = mac_mls_check_vnode_stat,
.mpo_check_vnode_write = mac_mls_check_vnode_write,
Fix panic associated with file creation via RPC/NFS when the MLS policy is loaded. This problem stems from the fact that the policy is not properly initializing the mac label associated with the NFS daemon. Obtained from: TrustedBSD Project Discussed with: rwatson
2006-08-26 20:13:35 +00:00
.mpo_associate_nfsd_label = mac_mls_associate_nfsd_label,
Introduce a new entry point, mac_create_mbuf_from_firewall. This entry point exists to allow the mandatory access control policy to properly initialize mbufs generated by the firewall. An example where this might happen is keep alive packets, or ICMP error packets in response to other packets. This takes care of kernel panics associated with un-initialize mbuf labels when the firewall generates packets. [1] I modified this patch from it's original version, the initial patch introduced a number of entry points which were programmatically equivalent. So I introduced only one. Instead, we should leverage mac_create_mbuf_netlayer() which is used for similar situations, an example being icmp_error() This will minimize the impact associated with the MFC Submitted by: mlaier [1] MFC after: 1 week This is a RELENG_6 candidate
2006-09-12 04:25:13 +00:00
.mpo_create_mbuf_from_firewall = mac_mls_create_mbuf_from_firewall,
Introduce support for Mandatory Access Control and extensible kernel access control. Provide implementations of some sample operating system security policy extensions. These are not yet hooked up to the build as other infrastructure is still being committed. Most of these work fairly well and are in daily use in our development and (limited) production environments. Some are not yet in their final form, and a number of the labeled policies waste a lot of kernel memory and will be fixed over the next month or so to be more conservative. They do give good examples of the flexibility of the MAC framework for implementing a variety of security policies. mac_biba: Implementation of fixed-label Biba integrity policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned integrity levels, and information flow is controlled based on a read-up, write-down policy. Currently, purely hierarchal. mac_bsdextended: Implementation of a "file system firewall", which allows the administrator to specify a series of rules limiting access by users and groups to objects owned by other users and groups. This policy is unlabeled, relying on existing system security labeling (file permissions/ownership, process credentials). mac_ifoff: Secure interface silencing. Special-purpose module to limit inappropriate out-going network traffic for silent monitoring scenarios. Prevents the various network stacks from generating any output despite an interface being live for reception. mac_mls: Implementation of fixed-label Multi-Level Security confidentiality policy, similar to those found in a number of commercial trusted operating systems. All subjects and objects are assigned confidentiality levels, and information flow is controlled based on a write-up, read-down policy. Currently, purely hiearchal, although non-hierarchal support is in the works. mac_none: Policy module implementing all MAC policy entry points with empty stubs. A good place to start if you want all the prototypes types in for you, and don't mind a bit of pruning. Can be loaded, but has no access control impact. Useful also for performance measurements. mac_seeotheruids: Policy module implementing a security service similar to security.bsd.seeotheruids, only a slightly more detailed policy involving exceptions for members of specific groups, etc. This policy is unlabeled, relying on existing system security labeling (process credentials). mac_test: Policy module implementing basic sanity tests for label handling. Attempts to ensure that labels are not freed multiple times, etc, etc. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-07-31 18:07:45 +00:00
};
Trim "trustedbsd_" from the front of the policy module "short names"; the vendor is only included in the long name currently, reducing verbosity when modules are registered and unregistered. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-03-27 19:26:39 +00:00
MAC_POLICY_SET(&mac_mls_ops, mac_mls, "TrustedBSD MAC/MLS",
Enable the MAC_ALWAYS_LABEL_MBUF flag for the Biba, LOMAC, MLS, and Test policies. Missed in earlier merge. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-04-15 20:51:18 +00:00
MPC_LOADTIME_FLAG_NOTLATE | MPC_LOADTIME_FLAG_LABELMBUFS, &mac_mls_slot);
Reference in New Issue Copy Permalink