2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Implementation of the multi-level security (MLS) policy.
|
|
|
|
*
|
|
|
|
* Author : Stephen Smalley, <sds@epoch.ncsc.mil>
|
|
|
|
*/
|
|
|
|
/*
|
|
|
|
* Updated: Trusted Computer Solutions, Inc. <dgoeddel@trustedcs.com>
|
|
|
|
*
|
|
|
|
* Support for enhanced MLS infrastructure.
|
|
|
|
*
|
2006-02-24 14:44:05 -07:00
|
|
|
* Copyright (C) 2004-2006 Trusted Computer Solutions, Inc.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2006-08-05 00:17:57 -06:00
|
|
|
/*
|
|
|
|
* Updated: Hewlett-Packard <paul.moore@hp.com>
|
|
|
|
*
|
2006-11-29 11:18:18 -07:00
|
|
|
* Added support to import/export the MLS label from NetLabel
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
* (c) Copyright Hewlett-Packard Development Company, L.P., 2006
|
|
|
|
*/
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/errno.h>
|
2006-11-29 11:18:18 -07:00
|
|
|
#include <net/netlabel.h>
|
2005-07-28 02:07:37 -06:00
|
|
|
#include "sidtab.h"
|
2005-04-16 16:20:36 -06:00
|
|
|
#include "mls.h"
|
|
|
|
#include "policydb.h"
|
|
|
|
#include "services.h"
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Return the length in bytes for the MLS fields of the
|
|
|
|
* security context string representation of `context'.
|
|
|
|
*/
|
2008-04-18 15:38:32 -06:00
|
|
|
int mls_compute_context_len(struct context *context)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
int i, l, len, head, prev;
|
|
|
|
char *nm;
|
|
|
|
struct ebitmap *e;
|
2005-09-03 16:55:16 -06:00
|
|
|
struct ebitmap_node *node;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
len = 1; /* for the beginning ":" */
|
|
|
|
for (l = 0; l < 2; l++) {
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
int index_sens = context->range.level[l].sens;
|
2010-11-29 13:47:09 -07:00
|
|
|
len += strlen(sym_name(&policydb, SYM_LEVELS, index_sens - 1));
|
2005-04-16 16:20:36 -06:00
|
|
|
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
/* categories */
|
|
|
|
head = -2;
|
|
|
|
prev = -2;
|
|
|
|
e = &context->range.level[l].cat;
|
|
|
|
ebitmap_for_each_positive_bit(e, node, i) {
|
|
|
|
if (i - prev > 1) {
|
|
|
|
/* one or more negative bits are skipped */
|
|
|
|
if (head != prev) {
|
2010-11-29 13:47:09 -07:00
|
|
|
nm = sym_name(&policydb, SYM_CATS, prev);
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
len += strlen(nm) + 1;
|
|
|
|
}
|
2010-11-29 13:47:09 -07:00
|
|
|
nm = sym_name(&policydb, SYM_CATS, i);
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
len += strlen(nm) + 1;
|
|
|
|
head = i;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
prev = i;
|
|
|
|
}
|
|
|
|
if (prev != head) {
|
2010-11-29 13:47:09 -07:00
|
|
|
nm = sym_name(&policydb, SYM_CATS, prev);
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
len += strlen(nm) + 1;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
if (l == 0) {
|
|
|
|
if (mls_level_eq(&context->range.level[0],
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
&context->range.level[1]))
|
2005-04-16 16:20:36 -06:00
|
|
|
break;
|
|
|
|
else
|
|
|
|
len++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return len;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Write the security context string representation of
|
|
|
|
* the MLS fields of `context' into the string `*scontext'.
|
|
|
|
* Update `*scontext' to point to the end of the MLS fields.
|
|
|
|
*/
|
|
|
|
void mls_sid_to_context(struct context *context,
|
2008-04-18 15:38:32 -06:00
|
|
|
char **scontext)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
char *scontextp, *nm;
|
|
|
|
int i, l, head, prev;
|
|
|
|
struct ebitmap *e;
|
2005-09-03 16:55:16 -06:00
|
|
|
struct ebitmap_node *node;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2005-04-16 16:20:36 -06:00
|
|
|
return;
|
|
|
|
|
|
|
|
scontextp = *scontext;
|
|
|
|
|
|
|
|
*scontextp = ':';
|
|
|
|
scontextp++;
|
|
|
|
|
|
|
|
for (l = 0; l < 2; l++) {
|
2010-11-29 13:47:09 -07:00
|
|
|
strcpy(scontextp, sym_name(&policydb, SYM_LEVELS,
|
|
|
|
context->range.level[l].sens - 1));
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
scontextp += strlen(scontextp);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* categories */
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
head = -2;
|
|
|
|
prev = -2;
|
|
|
|
e = &context->range.level[l].cat;
|
|
|
|
ebitmap_for_each_positive_bit(e, node, i) {
|
|
|
|
if (i - prev > 1) {
|
|
|
|
/* one or more negative bits are skipped */
|
|
|
|
if (prev != head) {
|
|
|
|
if (prev - head > 1)
|
2005-04-16 16:20:36 -06:00
|
|
|
*scontextp++ = '.';
|
|
|
|
else
|
|
|
|
*scontextp++ = ',';
|
2010-11-29 13:47:09 -07:00
|
|
|
nm = sym_name(&policydb, SYM_CATS, prev);
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
strcpy(scontextp, nm);
|
|
|
|
scontextp += strlen(nm);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
if (prev < 0)
|
|
|
|
*scontextp++ = ':';
|
|
|
|
else
|
|
|
|
*scontextp++ = ',';
|
2010-11-29 13:47:09 -07:00
|
|
|
nm = sym_name(&policydb, SYM_CATS, i);
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
strcpy(scontextp, nm);
|
|
|
|
scontextp += strlen(nm);
|
|
|
|
head = i;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
prev = i;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
if (prev != head) {
|
|
|
|
if (prev - head > 1)
|
2005-04-16 16:20:36 -06:00
|
|
|
*scontextp++ = '.';
|
|
|
|
else
|
|
|
|
*scontextp++ = ',';
|
2010-11-29 13:47:09 -07:00
|
|
|
nm = sym_name(&policydb, SYM_CATS, prev);
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
strcpy(scontextp, nm);
|
|
|
|
scontextp += strlen(nm);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
if (l == 0) {
|
|
|
|
if (mls_level_eq(&context->range.level[0],
|
2008-04-18 15:38:32 -06:00
|
|
|
&context->range.level[1]))
|
2005-04-16 16:20:36 -06:00
|
|
|
break;
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
else
|
|
|
|
*scontextp++ = '-';
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
*scontext = scontextp;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2007-11-07 08:08:00 -07:00
|
|
|
int mls_level_isvalid(struct policydb *p, struct mls_level *l)
|
|
|
|
{
|
|
|
|
struct level_datum *levdatum;
|
|
|
|
struct ebitmap_node *node;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
if (!l->sens || l->sens > p->p_levels.nprim)
|
|
|
|
return 0;
|
|
|
|
levdatum = hashtab_search(p->p_levels.table,
|
2010-11-29 13:47:09 -07:00
|
|
|
sym_name(p, SYM_LEVELS, l->sens - 1));
|
2007-11-07 08:08:00 -07:00
|
|
|
if (!levdatum)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
ebitmap_for_each_positive_bit(&l->cat, node, i) {
|
|
|
|
if (i > p->p_cats.nprim)
|
|
|
|
return 0;
|
|
|
|
if (!ebitmap_get_bit(&levdatum->level->cat, i)) {
|
|
|
|
/*
|
|
|
|
* Category may not be associated with
|
|
|
|
* sensitivity.
|
|
|
|
*/
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
int mls_range_isvalid(struct policydb *p, struct mls_range *r)
|
|
|
|
{
|
|
|
|
return (mls_level_isvalid(p, &r->level[0]) &&
|
|
|
|
mls_level_isvalid(p, &r->level[1]) &&
|
|
|
|
mls_level_dom(&r->level[1], &r->level[0]));
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Return 1 if the MLS fields in the security context
|
|
|
|
* structure `c' are valid. Return 0 otherwise.
|
|
|
|
*/
|
|
|
|
int mls_context_isvalid(struct policydb *p, struct context *c)
|
|
|
|
{
|
|
|
|
struct user_datum *usrdatum;
|
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!p->mls_enabled)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 1;
|
|
|
|
|
2007-11-07 08:08:00 -07:00
|
|
|
if (!mls_range_isvalid(p, &c->range))
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (c->role == OBJECT_R_VAL)
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* User must be authorized for the MLS range.
|
|
|
|
*/
|
|
|
|
if (!c->user || c->user > p->p_users.nprim)
|
|
|
|
return 0;
|
|
|
|
usrdatum = p->user_val_to_struct[c->user - 1];
|
|
|
|
if (!mls_range_contains(usrdatum->range, c->range))
|
|
|
|
return 0; /* user may not be associated with range */
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Set the MLS fields in the security context structure
|
|
|
|
* `context' based on the string representation in
|
|
|
|
* the string `*scontext'. Update `*scontext' to
|
|
|
|
* point to the end of the string representation of
|
|
|
|
* the MLS fields.
|
|
|
|
*
|
|
|
|
* This function modifies the string in place, inserting
|
|
|
|
* NULL characters to terminate the MLS fields.
|
2005-07-28 02:07:37 -06:00
|
|
|
*
|
|
|
|
* If a def_sid is provided and no MLS field is present,
|
|
|
|
* copy the MLS field of the associated default context.
|
|
|
|
* Used for upgraded to MLS systems where objects may lack
|
|
|
|
* MLS fields.
|
|
|
|
*
|
|
|
|
* Policy read-lock must be held for sidtab lookup.
|
|
|
|
*
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2008-05-07 11:03:20 -06:00
|
|
|
int mls_context_to_sid(struct policydb *pol,
|
|
|
|
char oldc,
|
2005-04-16 16:20:36 -06:00
|
|
|
char **scontext,
|
2005-07-28 02:07:37 -06:00
|
|
|
struct context *context,
|
|
|
|
struct sidtab *s,
|
|
|
|
u32 def_sid)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
|
|
|
|
char delim;
|
|
|
|
char *scontextp, *p, *rngptr;
|
|
|
|
struct level_datum *levdatum;
|
|
|
|
struct cat_datum *catdatum, *rngdatum;
|
|
|
|
int l, rc = -EINVAL;
|
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!pol->mls_enabled) {
|
2005-11-08 22:34:32 -07:00
|
|
|
if (def_sid != SECSID_NULL && oldc)
|
2010-04-09 05:30:29 -06:00
|
|
|
*scontext += strlen(*scontext) + 1;
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
2005-11-08 22:34:32 -07:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2005-07-28 02:07:37 -06:00
|
|
|
/*
|
|
|
|
* No MLS component to the security context, try and map to
|
|
|
|
* default if provided.
|
|
|
|
*/
|
|
|
|
if (!oldc) {
|
|
|
|
struct context *defcon;
|
|
|
|
|
|
|
|
if (def_sid == SECSID_NULL)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
defcon = sidtab_search(s, def_sid);
|
|
|
|
if (!defcon)
|
|
|
|
goto out;
|
|
|
|
|
2006-12-12 12:02:41 -07:00
|
|
|
rc = mls_context_cpy(context, defcon);
|
2005-04-16 16:20:36 -06:00
|
|
|
goto out;
|
2005-07-28 02:07:37 -06:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* Extract low sensitivity. */
|
|
|
|
scontextp = p = *scontext;
|
|
|
|
while (*p && *p != ':' && *p != '-')
|
|
|
|
p++;
|
|
|
|
|
|
|
|
delim = *p;
|
2008-07-20 14:57:01 -06:00
|
|
|
if (delim != '\0')
|
|
|
|
*p++ = '\0';
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
for (l = 0; l < 2; l++) {
|
2008-05-07 11:03:20 -06:00
|
|
|
levdatum = hashtab_search(pol->p_levels.table, scontextp);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (!levdatum) {
|
|
|
|
rc = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
context->range.level[l].sens = levdatum->level->sens;
|
|
|
|
|
|
|
|
if (delim == ':') {
|
|
|
|
/* Extract category set. */
|
|
|
|
while (1) {
|
|
|
|
scontextp = p;
|
|
|
|
while (*p && *p != ',' && *p != '-')
|
|
|
|
p++;
|
|
|
|
delim = *p;
|
2008-07-20 14:57:01 -06:00
|
|
|
if (delim != '\0')
|
|
|
|
*p++ = '\0';
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* Separate into range if exists */
|
2008-04-18 15:38:32 -06:00
|
|
|
rngptr = strchr(scontextp, '.');
|
|
|
|
if (rngptr != NULL) {
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Remove '.' */
|
2008-07-20 14:57:01 -06:00
|
|
|
*rngptr++ = '\0';
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2008-05-07 11:03:20 -06:00
|
|
|
catdatum = hashtab_search(pol->p_cats.table,
|
2008-04-18 15:38:32 -06:00
|
|
|
scontextp);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (!catdatum) {
|
|
|
|
rc = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
rc = ebitmap_set_bit(&context->range.level[l].cat,
|
2008-04-18 15:38:32 -06:00
|
|
|
catdatum->value - 1, 1);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (rc)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* If range, set all categories in range */
|
|
|
|
if (rngptr) {
|
|
|
|
int i;
|
|
|
|
|
2008-05-07 11:03:20 -06:00
|
|
|
rngdatum = hashtab_search(pol->p_cats.table, rngptr);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (!rngdatum) {
|
|
|
|
rc = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (catdatum->value >= rngdatum->value) {
|
|
|
|
rc = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = catdatum->value; i < rngdatum->value; i++) {
|
|
|
|
rc = ebitmap_set_bit(&context->range.level[l].cat, i, 1);
|
|
|
|
if (rc)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (delim != ',')
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (delim == '-') {
|
|
|
|
/* Extract high sensitivity. */
|
|
|
|
scontextp = p;
|
|
|
|
while (*p && *p != ':')
|
|
|
|
p++;
|
|
|
|
|
|
|
|
delim = *p;
|
2008-07-20 14:57:01 -06:00
|
|
|
if (delim != '\0')
|
|
|
|
*p++ = '\0';
|
2005-04-16 16:20:36 -06:00
|
|
|
} else
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (l == 0) {
|
|
|
|
context->range.level[1].sens = context->range.level[0].sens;
|
|
|
|
rc = ebitmap_cpy(&context->range.level[1].cat,
|
|
|
|
&context->range.level[0].cat);
|
|
|
|
if (rc)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
*scontext = ++p;
|
|
|
|
rc = 0;
|
|
|
|
out:
|
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
2006-02-24 14:44:05 -07:00
|
|
|
/*
|
|
|
|
* Set the MLS fields in the security context structure
|
|
|
|
* `context' based on the string representation in
|
|
|
|
* the string `str'. This function will allocate temporary memory with the
|
|
|
|
* given constraints of gfp_mask.
|
|
|
|
*/
|
|
|
|
int mls_from_string(char *str, struct context *context, gfp_t gfp_mask)
|
|
|
|
{
|
|
|
|
char *tmpstr, *freestr;
|
|
|
|
int rc;
|
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2006-02-24 14:44:05 -07:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
/* we need freestr because mls_context_to_sid will change
|
|
|
|
the value of tmpstr */
|
|
|
|
tmpstr = freestr = kstrdup(str, gfp_mask);
|
|
|
|
if (!tmpstr) {
|
|
|
|
rc = -ENOMEM;
|
|
|
|
} else {
|
2008-05-07 11:03:20 -06:00
|
|
|
rc = mls_context_to_sid(&policydb, ':', &tmpstr, context,
|
2008-04-18 15:38:32 -06:00
|
|
|
NULL, SECSID_NULL);
|
2006-02-24 14:44:05 -07:00
|
|
|
kfree(freestr);
|
|
|
|
}
|
|
|
|
|
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Copies the MLS range `range' into `context'.
|
|
|
|
*/
|
2010-02-03 08:40:20 -07:00
|
|
|
int mls_range_set(struct context *context,
|
2008-04-18 15:38:32 -06:00
|
|
|
struct mls_range *range)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
int l, rc = 0;
|
|
|
|
|
|
|
|
/* Copy the MLS range into the context */
|
|
|
|
for (l = 0; l < 2; l++) {
|
|
|
|
context->range.level[l].sens = range->level[l].sens;
|
|
|
|
rc = ebitmap_cpy(&context->range.level[l].cat,
|
|
|
|
&range->level[l].cat);
|
|
|
|
if (rc)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
|
|
|
int mls_setup_user_range(struct context *fromcon, struct user_datum *user,
|
2008-04-18 15:38:32 -06:00
|
|
|
struct context *usercon)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2010-02-03 08:40:20 -07:00
|
|
|
if (policydb.mls_enabled) {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct mls_level *fromcon_sen = &(fromcon->range.level[0]);
|
|
|
|
struct mls_level *fromcon_clr = &(fromcon->range.level[1]);
|
|
|
|
struct mls_level *user_low = &(user->range.level[0]);
|
|
|
|
struct mls_level *user_clr = &(user->range.level[1]);
|
|
|
|
struct mls_level *user_def = &(user->dfltlevel);
|
|
|
|
struct mls_level *usercon_sen = &(usercon->range.level[0]);
|
|
|
|
struct mls_level *usercon_clr = &(usercon->range.level[1]);
|
|
|
|
|
|
|
|
/* Honor the user's default level if we can */
|
2008-05-14 09:27:45 -06:00
|
|
|
if (mls_level_between(user_def, fromcon_sen, fromcon_clr))
|
2005-04-16 16:20:36 -06:00
|
|
|
*usercon_sen = *user_def;
|
2008-05-14 09:27:45 -06:00
|
|
|
else if (mls_level_between(fromcon_sen, user_def, user_clr))
|
2005-04-16 16:20:36 -06:00
|
|
|
*usercon_sen = *fromcon_sen;
|
2008-05-14 09:27:45 -06:00
|
|
|
else if (mls_level_between(fromcon_clr, user_low, user_def))
|
2005-04-16 16:20:36 -06:00
|
|
|
*usercon_sen = *user_low;
|
2008-05-14 09:27:45 -06:00
|
|
|
else
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
/* Lower the clearance of available contexts
|
|
|
|
if the clearance of "fromcon" is lower than
|
|
|
|
that of the user's default clearance (but
|
|
|
|
only if the "fromcon" clearance dominates
|
|
|
|
the user's computed sensitivity level) */
|
2008-04-18 15:38:32 -06:00
|
|
|
if (mls_level_dom(user_clr, fromcon_clr))
|
2005-04-16 16:20:36 -06:00
|
|
|
*usercon_clr = *fromcon_clr;
|
2008-04-18 15:38:32 -06:00
|
|
|
else if (mls_level_dom(fromcon_clr, user_clr))
|
2005-04-16 16:20:36 -06:00
|
|
|
*usercon_clr = *user_clr;
|
2008-04-18 15:38:32 -06:00
|
|
|
else
|
2005-04-16 16:20:36 -06:00
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Convert the MLS fields in the security context
|
|
|
|
* structure `c' from the values specified in the
|
|
|
|
* policy `oldp' to the values specified in the policy `newp'.
|
|
|
|
*/
|
|
|
|
int mls_convert_context(struct policydb *oldp,
|
|
|
|
struct policydb *newp,
|
|
|
|
struct context *c)
|
|
|
|
{
|
|
|
|
struct level_datum *levdatum;
|
|
|
|
struct cat_datum *catdatum;
|
|
|
|
struct ebitmap bitmap;
|
2005-09-03 16:55:16 -06:00
|
|
|
struct ebitmap_node *node;
|
2005-04-16 16:20:36 -06:00
|
|
|
int l, i;
|
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
for (l = 0; l < 2; l++) {
|
|
|
|
levdatum = hashtab_search(newp->p_levels.table,
|
2010-11-29 13:47:09 -07:00
|
|
|
sym_name(oldp, SYM_LEVELS,
|
|
|
|
c->range.level[l].sens - 1));
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
if (!levdatum)
|
|
|
|
return -EINVAL;
|
|
|
|
c->range.level[l].sens = levdatum->level->sens;
|
|
|
|
|
|
|
|
ebitmap_init(&bitmap);
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
ebitmap_for_each_positive_bit(&c->range.level[l].cat, node, i) {
|
|
|
|
int rc;
|
|
|
|
|
|
|
|
catdatum = hashtab_search(newp->p_cats.table,
|
2010-11-29 13:47:09 -07:00
|
|
|
sym_name(oldp, SYM_CATS, i));
|
SELinux: improve performance when AVC misses.
* We add ebitmap_for_each_positive_bit() which enables to walk on
any positive bit on the given ebitmap, to improve its performance
using common bit-operations defined in linux/bitops.h.
In the previous version, this logic was implemented using a combination
of ebitmap_for_each_bit() and ebitmap_node_get_bit(), but is was worse
in performance aspect.
This logic is most frequestly used to compute a new AVC entry,
so this patch can improve SELinux performance when AVC misses are happen.
* struct ebitmap_node is redefined as an array of "unsigned long", to get
suitable for using find_next_bit() which is fasted than iteration of
shift and logical operation, and to maximize memory usage allocated
from general purpose slab.
* Any ebitmap_for_each_bit() are repleced by the new implementation
in ss/service.c and ss/mls.c. Some of related implementation are
changed, however, there is no incompatibility with the previous
version.
* The width of any new line are less or equal than 80-chars.
The following benchmark shows the effect of this patch, when we
access many files which have different security context one after
another. The number is more than /selinux/avc/cache_threshold, so
any access always causes AVC misses.
selinux-2.6 selinux-2.6-ebitmap
AVG: 22.763 [s] 8.750 [s]
STD: 0.265 0.019
------------------------------------------
1st: 22.558 [s] 8.786 [s]
2nd: 22.458 [s] 8.750 [s]
3rd: 22.478 [s] 8.754 [s]
4th: 22.724 [s] 8.745 [s]
5th: 22.918 [s] 8.748 [s]
6th: 22.905 [s] 8.764 [s]
7th: 23.238 [s] 8.726 [s]
8th: 22.822 [s] 8.729 [s]
Signed-off-by: KaiGai Kohei <kaigai@ak.jp.nec.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2007-09-28 11:20:55 -06:00
|
|
|
if (!catdatum)
|
|
|
|
return -EINVAL;
|
|
|
|
rc = ebitmap_set_bit(&bitmap, catdatum->value - 1, 1);
|
|
|
|
if (rc)
|
|
|
|
return rc;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
ebitmap_destroy(&c->range.level[l].cat);
|
|
|
|
c->range.level[l].cat = bitmap;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int mls_compute_sid(struct context *scontext,
|
|
|
|
struct context *tcontext,
|
|
|
|
u16 tclass,
|
|
|
|
u32 specified,
|
|
|
|
struct context *newcontext)
|
|
|
|
{
|
2010-01-07 13:55:16 -07:00
|
|
|
struct range_trans rtr;
|
|
|
|
struct mls_range *r;
|
2006-09-26 00:31:59 -06:00
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2005-04-16 16:20:36 -06:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
switch (specified) {
|
|
|
|
case AVTAB_TRANSITION:
|
2006-09-26 00:31:59 -06:00
|
|
|
/* Look for a range transition rule. */
|
2010-01-07 13:55:16 -07:00
|
|
|
rtr.source_type = scontext->type;
|
|
|
|
rtr.target_type = tcontext->type;
|
|
|
|
rtr.target_class = tclass;
|
|
|
|
r = hashtab_search(policydb.range_tr, &rtr);
|
|
|
|
if (r)
|
|
|
|
return mls_range_set(newcontext, r);
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Fallthrough */
|
|
|
|
case AVTAB_CHANGE:
|
selinux: dynamic class/perm discovery
Modify SELinux to dynamically discover class and permission values
upon policy load, based on the dynamic object class/perm discovery
logic from libselinux. A mapping is created between kernel-private
class and permission indices used outside the security server and the
policy values used within the security server.
The mappings are only applied upon kernel-internal computations;
similar mappings for the private indices of userspace object managers
is handled on a per-object manager basis by the userspace AVC. The
interfaces for compute_av and transition_sid are split for kernel
vs. userspace; the userspace functions are distinguished by a _user
suffix.
The kernel-private class indices are no longer tied to the policy
values and thus do not need to skip indices for userspace classes;
thus the kernel class index values are compressed. The flask.h
definitions were regenerated by deleting the userspace classes from
refpolicy's definitions and then regenerating the headers. Going
forward, we can just maintain the flask.h, av_permissions.h, and
classmap.h definitions separately from policy as they are no longer
tied to the policy values. The next patch introduces a utility to
automate generation of flask.h and av_permissions.h from the
classmap.h definitions.
The older kernel class and permission string tables are removed and
replaced by a single security class mapping table that is walked at
policy load to generate the mapping. The old kernel class validation
logic is completely replaced by the mapping logic.
The handle unknown logic is reworked. reject_unknown=1 is handled
when the mappings are computed at policy load time, similar to the old
handling by the class validation logic. allow_unknown=1 is handled
when computing and mapping decisions - if the permission was not able
to be mapped (i.e. undefined, mapped to zero), then it is
automatically added to the allowed vector. If the class was not able
to be mapped (i.e. undefined, mapped to zero), then all permissions
are allowed for it if allow_unknown=1.
avc_audit leverages the new security class mapping table to lookup the
class and permission names from the kernel-private indices.
The mdp program is updated to use the new table when generating the
class definitions and allow rules for a minimal boot policy for the
kernel. It should be noted that this policy will not include any
userspace classes, nor will its policy index values for the kernel
classes correspond with the ones in refpolicy (they will instead match
the kernel-private indices).
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>
2009-09-30 11:37:50 -06:00
|
|
|
if (tclass == policydb.process_class)
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Use the process MLS attributes. */
|
2006-12-12 12:02:41 -07:00
|
|
|
return mls_context_cpy(newcontext, scontext);
|
2005-04-16 16:20:36 -06:00
|
|
|
else
|
|
|
|
/* Use the process effective MLS attributes. */
|
2006-12-12 12:02:41 -07:00
|
|
|
return mls_context_cpy_low(newcontext, scontext);
|
2005-04-16 16:20:36 -06:00
|
|
|
case AVTAB_MEMBER:
|
2008-01-24 13:30:52 -07:00
|
|
|
/* Use the process effective MLS attributes. */
|
|
|
|
return mls_context_cpy_low(newcontext, scontext);
|
2009-12-03 01:48:28 -07:00
|
|
|
|
|
|
|
/* fall through */
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
2006-11-29 11:18:18 -07:00
|
|
|
#ifdef CONFIG_NETLABEL
|
2006-08-05 00:17:57 -06:00
|
|
|
/**
|
2006-11-29 11:18:18 -07:00
|
|
|
* mls_export_netlbl_lvl - Export the MLS sensitivity levels to NetLabel
|
2006-08-05 00:17:57 -06:00
|
|
|
* @context: the security context
|
2006-11-29 11:18:18 -07:00
|
|
|
* @secattr: the NetLabel security attributes
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
* Description:
|
2006-11-29 11:18:18 -07:00
|
|
|
* Given the security context copy the low MLS sensitivity level into the
|
|
|
|
* NetLabel MLS sensitivity level field.
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
*/
|
2006-11-29 11:18:18 -07:00
|
|
|
void mls_export_netlbl_lvl(struct context *context,
|
|
|
|
struct netlbl_lsm_secattr *secattr)
|
2006-08-05 00:17:57 -06:00
|
|
|
{
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2006-08-05 00:17:57 -06:00
|
|
|
return;
|
|
|
|
|
2008-01-29 06:37:59 -07:00
|
|
|
secattr->attr.mls.lvl = context->range.level[0].sens - 1;
|
2006-11-29 11:18:18 -07:00
|
|
|
secattr->flags |= NETLBL_SECATTR_MLS_LVL;
|
2006-08-05 00:17:57 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2006-11-29 11:18:18 -07:00
|
|
|
* mls_import_netlbl_lvl - Import the NetLabel MLS sensitivity levels
|
2006-08-05 00:17:57 -06:00
|
|
|
* @context: the security context
|
2006-11-29 11:18:18 -07:00
|
|
|
* @secattr: the NetLabel security attributes
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
* Description:
|
2006-11-29 11:18:18 -07:00
|
|
|
* Given the security context and the NetLabel security attributes, copy the
|
|
|
|
* NetLabel MLS sensitivity level into the context.
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
*/
|
2006-11-29 11:18:18 -07:00
|
|
|
void mls_import_netlbl_lvl(struct context *context,
|
|
|
|
struct netlbl_lsm_secattr *secattr)
|
2006-08-05 00:17:57 -06:00
|
|
|
{
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2006-08-05 00:17:57 -06:00
|
|
|
return;
|
|
|
|
|
2008-01-29 06:37:59 -07:00
|
|
|
context->range.level[0].sens = secattr->attr.mls.lvl + 1;
|
2006-11-29 11:18:18 -07:00
|
|
|
context->range.level[1].sens = context->range.level[0].sens;
|
2006-08-05 00:17:57 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2006-11-29 11:18:18 -07:00
|
|
|
* mls_export_netlbl_cat - Export the MLS categories to NetLabel
|
2006-08-05 00:17:57 -06:00
|
|
|
* @context: the security context
|
2006-11-29 11:18:18 -07:00
|
|
|
* @secattr: the NetLabel security attributes
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
* Description:
|
2006-11-29 11:18:18 -07:00
|
|
|
* Given the security context copy the low MLS categories into the NetLabel
|
|
|
|
* MLS category field. Returns zero on success, negative values on failure.
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
*/
|
2006-11-29 11:18:18 -07:00
|
|
|
int mls_export_netlbl_cat(struct context *context,
|
|
|
|
struct netlbl_lsm_secattr *secattr)
|
2006-08-05 00:17:57 -06:00
|
|
|
{
|
2006-11-29 11:18:18 -07:00
|
|
|
int rc;
|
2006-08-05 00:17:57 -06:00
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2006-08-05 00:17:57 -06:00
|
|
|
return 0;
|
|
|
|
|
2006-11-29 11:18:18 -07:00
|
|
|
rc = ebitmap_netlbl_export(&context->range.level[0].cat,
|
2008-01-29 06:37:59 -07:00
|
|
|
&secattr->attr.mls.cat);
|
|
|
|
if (rc == 0 && secattr->attr.mls.cat != NULL)
|
2006-11-29 11:18:18 -07:00
|
|
|
secattr->flags |= NETLBL_SECATTR_MLS_CAT;
|
2006-08-05 00:17:57 -06:00
|
|
|
|
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2006-11-29 11:18:18 -07:00
|
|
|
* mls_import_netlbl_cat - Import the MLS categories from NetLabel
|
2006-08-05 00:17:57 -06:00
|
|
|
* @context: the security context
|
2006-11-29 11:18:18 -07:00
|
|
|
* @secattr: the NetLabel security attributes
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
* Description:
|
2006-11-29 11:18:18 -07:00
|
|
|
* Copy the NetLabel security attributes into the SELinux context; since the
|
|
|
|
* NetLabel security attribute only contains a single MLS category use it for
|
|
|
|
* both the low and high categories of the context. Returns zero on success,
|
|
|
|
* negative values on failure.
|
2006-08-05 00:17:57 -06:00
|
|
|
*
|
|
|
|
*/
|
2006-11-29 11:18:18 -07:00
|
|
|
int mls_import_netlbl_cat(struct context *context,
|
|
|
|
struct netlbl_lsm_secattr *secattr)
|
2006-08-05 00:17:57 -06:00
|
|
|
{
|
2006-11-29 11:18:18 -07:00
|
|
|
int rc;
|
2006-08-05 00:17:57 -06:00
|
|
|
|
2010-02-03 08:40:20 -07:00
|
|
|
if (!policydb.mls_enabled)
|
2006-08-05 00:17:57 -06:00
|
|
|
return 0;
|
|
|
|
|
2006-11-29 11:18:18 -07:00
|
|
|
rc = ebitmap_netlbl_import(&context->range.level[0].cat,
|
2008-01-29 06:37:59 -07:00
|
|
|
secattr->attr.mls.cat);
|
2006-11-29 11:18:18 -07:00
|
|
|
if (rc != 0)
|
|
|
|
goto import_netlbl_cat_failure;
|
|
|
|
|
|
|
|
rc = ebitmap_cpy(&context->range.level[1].cat,
|
|
|
|
&context->range.level[0].cat);
|
|
|
|
if (rc != 0)
|
|
|
|
goto import_netlbl_cat_failure;
|
2006-08-05 00:17:57 -06:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
2006-11-29 11:18:18 -07:00
|
|
|
import_netlbl_cat_failure:
|
2006-08-05 00:17:57 -06:00
|
|
|
ebitmap_destroy(&context->range.level[0].cat);
|
|
|
|
ebitmap_destroy(&context->range.level[1].cat);
|
|
|
|
return rc;
|
|
|
|
}
|
2006-11-29 11:18:18 -07:00
|
|
|
#endif /* CONFIG_NETLABEL */
|