2019-06-01 10:08:55 +02:00
// SPDX-License-Identifier: GPL-2.0-only
2009-02-04 09:06:58 -05:00
/*
* Copyright ( C ) 2008 IBM Corporation
* Author : Mimi Zohar < zohar @ us . ibm . com >
*
* ima_policy . c
2014-03-04 18:04:20 +02:00
* - initialize default measure policy rules
2009-02-04 09:06:58 -05:00
*/
2019-06-27 23:19:32 -03:00
2018-12-09 15:36:29 -05:00
# include <linux/init.h>
2009-02-04 09:06:58 -05:00
# include <linux/list.h>
2020-10-02 10:38:15 -07:00
# include <linux/kernel_read_file.h>
2016-01-14 17:57:47 -05:00
# include <linux/fs.h>
2009-02-04 09:06:58 -05:00
# include <linux/security.h>
# include <linux/magic.h>
2009-02-04 09:07:00 -05:00
# include <linux/parser.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 17:04:11 +09:00
# include <linux/slab.h>
2015-12-02 17:47:54 +02:00
# include <linux/rculist.h>
2012-09-03 23:23:13 +03:00
# include <linux/genhd.h>
2015-12-02 17:47:56 +02:00
# include <linux/seq_file.h>
2018-10-09 23:00:36 +05:30
# include <linux/ima.h>
2009-02-04 09:06:58 -05:00
# include "ima.h"
/* flags definitions */
2014-03-04 18:04:20 +02:00
# define IMA_FUNC 0x0001
# define IMA_MASK 0x0002
2009-02-04 09:06:58 -05:00
# define IMA_FSMAGIC 0x0004
# define IMA_UID 0x0008
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
# define IMA_FOWNER 0x0010
2012-09-03 23:23:13 +03:00
# define IMA_FSUUID 0x0020
2014-11-05 07:53:55 -05:00
# define IMA_INMASK 0x0040
2014-11-05 07:48:36 -05:00
# define IMA_EUID 0x0080
2016-06-01 13:14:01 -05:00
# define IMA_PCR 0x0100
2018-01-15 11:20:36 -05:00
# define IMA_FSNAME 0x0200
2019-12-11 08:47:07 -08:00
# define IMA_KEYRINGS 0x0400
2021-01-07 20:07:05 -08:00
# define IMA_LABEL 0x0800
2021-08-16 08:11:00 +00:00
# define IMA_VALIDATE_ALGOS 0x1000
2021-10-07 14:03:02 -06:00
# define IMA_GID 0x2000
# define IMA_EGID 0x4000
# define IMA_FGROUP 0x8000
2009-02-04 09:06:58 -05:00
2012-09-12 20:51:32 +03:00
# define UNKNOWN 0
# define MEASURE 0x0001 /* same as IMA_MEASURE */
# define DONT_MEASURE 0x0002
# define APPRAISE 0x0004 /* same as IMA_APPRAISE */
# define DONT_APPRAISE 0x0008
2012-06-14 10:04:36 -07:00
# define AUDIT 0x0040
ima: support new "hash" and "dont_hash" policy actions
The builtin ima_appraise_tcb policy, which is specified on the boot
command line, can be replaced with a custom policy, normally early in
the boot process. Custom policies can be more restrictive in some ways,
like requiring file signatures, but can be less restrictive in other
ways, like not appraising mutable files. With a less restrictive policy
in place, files in the builtin policy might not be hashed and labeled
with a security.ima hash. On reboot, files which should be labeled in
the ima_appraise_tcb are not labeled, possibly preventing the system
from booting properly.
To resolve this problem, this patch extends the existing IMA policy
actions "measure", "dont_measure", "appraise", "dont_appraise", and
"audit" with "hash" and "dont_hash". The new "hash" action will write
the file hash as security.ima, but without requiring the file to be
appraised as well.
For example, the builtin ima_appraise_tcb policy includes the rule,
"appraise fowner=0". Adding the "hash fowner=0" rule to a custom
policy, will cause the needed file hashes to be calculated and written
as security.ima xattrs.
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
2016-09-29 10:04:52 -04:00
# define HASH 0x0100
# define DONT_HASH 0x0200
2009-02-04 09:07:00 -05:00
2016-06-01 13:14:01 -05:00
# define INVALID_PCR(a) (((a) < 0) || \
2019-12-09 10:31:43 -08:00
( a ) > = ( sizeof_field ( struct integrity_iint_cache , measured_pcrs ) * 8 ) )
2016-06-01 13:14:01 -05:00
2014-09-12 19:35:54 +02:00
int ima_policy_flag ;
2015-12-07 14:35:47 -05:00
static int temp_ima_appraise ;
2018-07-13 14:06:01 -04:00
static int build_ima_appraise __ro_after_init ;
2014-09-12 19:35:54 +02:00
2021-08-16 08:11:01 +00:00
atomic_t ima_setxattr_allowed_hash_algorithms ;
2009-02-04 09:07:00 -05:00
# define MAX_LSM_RULES 6
enum lsm_rule_types { LSM_OBJ_USER , LSM_OBJ_ROLE , LSM_OBJ_TYPE ,
LSM_SUBJ_USER , LSM_SUBJ_ROLE , LSM_SUBJ_TYPE
} ;
2009-02-04 09:06:58 -05:00
2015-06-11 20:48:33 -04:00
enum policy_types { ORIGINAL_TCB = 1 , DEFAULT_TCB } ;
2018-10-09 23:00:35 +05:30
enum policy_rule_list { IMA_DEFAULT_POLICY = 1 , IMA_CUSTOM_POLICY } ;
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
struct ima_rule_opt_list {
size_t count ;
char * items [ ] ;
} ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
struct ima_rule_entry {
2009-02-04 09:06:58 -05:00
struct list_head list ;
ima: integrity appraisal extension
IMA currently maintains an integrity measurement list used to assert the
integrity of the running system to a third party. The IMA-appraisal
extension adds local integrity validation and enforcement of the
measurement against a "good" value stored as an extended attribute
'security.ima'. The initial methods for validating 'security.ima' are
hashed based, which provides file data integrity, and digital signature
based, which in addition to providing file data integrity, provides
authenticity.
This patch creates and maintains the 'security.ima' xattr, containing
the file data hash measurement. Protection of the xattr is provided by
EVM, if enabled and configured.
Based on policy, IMA calls evm_verifyxattr() to verify a file's metadata
integrity and, assuming success, compares the file's current hash value
with the one stored as an extended attribute in 'security.ima'.
Changelov v4:
- changed iint cache flags to hex values
Changelog v3:
- change appraisal default for filesystems without xattr support to fail
Changelog v2:
- fix audit msg 'res' value
- removed unused 'ima_appraise=' values
Changelog v1:
- removed unused iint mutex (Dmitry Kasatkin)
- setattr hook must not reset appraised (Dmitry Kasatkin)
- evm_verifyxattr() now differentiates between no 'security.evm' xattr
(INTEGRITY_NOLABEL) and no EVM 'protected' xattrs included in the
'security.evm' (INTEGRITY_NOXATTRS).
- replace hash_status with ima_status (Dmitry Kasatkin)
- re-initialize slab element ima_status on free (Dmitry Kasatkin)
- include 'security.ima' in EVM if CONFIG_IMA_APPRAISE, not CONFIG_IMA
- merged half "ima: ima_must_appraise_or_measure API change" (Dmitry Kasatkin)
- removed unnecessary error variable in process_measurement() (Dmitry Kasatkin)
- use ima_inode_post_setattr() stub function, if IMA_APPRAISE not configured
(moved ima_inode_post_setattr() to ima_appraise.c)
- make sure ima_collect_measurement() can read file
Changelog:
- add 'iint' to evm_verifyxattr() call (Dimitry Kasatkin)
- fix the race condition between chmod, which takes the i_mutex and then
iint->mutex, and ima_file_free() and process_measurement(), which take
the locks in the reverse order, by eliminating iint->mutex. (Dmitry Kasatkin)
- cleanup of ima_appraise_measurement() (Dmitry Kasatkin)
- changes as a result of the iint not allocated for all regular files, but
only for those measured/appraised.
- don't try to appraise new/empty files
- expanded ima_appraisal description in ima/Kconfig
- IMA appraise definitions required even if IMA_APPRAISE not enabled
- add return value to ima_must_appraise() stub
- unconditionally set status = INTEGRITY_PASS *after* testing status,
not before. (Found by Joe Perches)
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2012-02-13 10:15:05 -05:00
int action ;
2009-02-04 09:06:58 -05:00
unsigned int flags ;
enum ima_hooks func ;
int mask ;
unsigned long fsmagic ;
2017-06-01 07:00:26 +02:00
uuid_t fsuuid ;
2012-05-25 18:24:12 -06:00
kuid_t uid ;
2021-10-07 14:03:02 -06:00
kgid_t gid ;
2012-10-02 21:38:48 -07:00
kuid_t fowner ;
2021-10-07 14:03:02 -06:00
kgid_t fgroup ;
2021-10-07 14:03:01 -06:00
bool ( * uid_op ) ( kuid_t cred_uid , kuid_t rule_uid ) ; /* Handlers for operators */
2021-10-07 14:03:02 -06:00
bool ( * gid_op ) ( kgid_t cred_gid , kgid_t rule_gid ) ;
2021-10-07 14:03:01 -06:00
bool ( * fowner_op ) ( kuid_t cred_uid , kuid_t rule_uid ) ; /* uid_eq(), uid_gt(), uid_lt() */
2021-10-07 14:03:02 -06:00
bool ( * fgroup_op ) ( kgid_t cred_gid , kgid_t rule_gid ) ; /* gid_eq(), gid_gt(), gid_lt() */
2016-06-01 13:14:01 -05:00
int pcr ;
2021-08-16 08:11:00 +00:00
unsigned int allowed_algos ; /* bitfield of allowed hash algorithms */
2009-02-04 09:07:00 -05:00
struct {
void * rule ; /* LSM file metadata specific */
2020-07-09 01:19:08 -05:00
char * args_p ; /* audit value */
2009-02-04 09:07:00 -05:00
int type ; /* audit type */
} lsm [ MAX_LSM_RULES ] ;
2018-01-15 11:20:36 -05:00
char * fsname ;
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
struct ima_rule_opt_list * keyrings ; /* Measure keys added to these keyrings */
2021-01-07 20:07:05 -08:00
struct ima_rule_opt_list * label ; /* Measure data grouped under this label */
2019-06-19 15:46:11 -07:00
struct ima_template_desc * template ;
2009-02-04 09:06:58 -05:00
} ;
2021-08-16 08:11:00 +00:00
/*
* sanity check in case the kernels gains more hash algorithms that can
* fit in an unsigned int
*/
static_assert (
8 * sizeof ( unsigned int ) > = HASH_ALGO__LAST ,
" The bitfield allowed_algos in ima_rule_entry is too small to contain all the supported hash algorithms, consider using a bigger type " ) ;
2009-05-21 15:47:06 -04:00
/*
* Without LSM specific knowledge , the default policy can only be
2021-10-07 14:03:02 -06:00
* written in terms of . action , . func , . mask , . fsmagic , . uid , . gid ,
* . fowner , and . fgroup
2009-02-04 09:07:00 -05:00
*/
2009-05-21 15:47:06 -04:00
/*
* The minimum rule set to allow for full TCB coverage . Measures all files
* opened or mmap for exec and everything read by root . Dangerous because
* normal users can easily run the machine out of memory simply building
* and running executables .
*/
2017-02-13 16:34:35 +11:00
static struct ima_rule_entry dont_measure_rules [ ] __ro_after_init = {
2014-03-04 18:04:20 +02:00
{ . action = DONT_MEASURE , . fsmagic = PROC_SUPER_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = SYSFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = DEBUGFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = TMPFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = DEVPTS_SUPER_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = BINFMTFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = SECURITYFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = SELINUX_MAGIC , . flags = IMA_FSMAGIC } ,
2018-03-22 13:46:01 +00:00
{ . action = DONT_MEASURE , . fsmagic = SMACK_MAGIC , . flags = IMA_FSMAGIC } ,
2015-04-11 17:13:06 +02:00
{ . action = DONT_MEASURE , . fsmagic = CGROUP_SUPER_MAGIC ,
. flags = IMA_FSMAGIC } ,
2017-05-09 11:25:27 -07:00
{ . action = DONT_MEASURE , . fsmagic = CGROUP2_SUPER_MAGIC ,
. flags = IMA_FSMAGIC } ,
2018-11-14 17:24:13 -05:00
{ . action = DONT_MEASURE , . fsmagic = NSFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_MEASURE , . fsmagic = EFIVARFS_MAGIC , . flags = IMA_FSMAGIC }
2015-06-11 20:48:33 -04:00
} ;
2017-02-13 16:34:35 +11:00
static struct ima_rule_entry original_measurement_rules [ ] __ro_after_init = {
2015-06-11 20:48:33 -04:00
{ . action = MEASURE , . func = MMAP_CHECK , . mask = MAY_EXEC ,
. flags = IMA_FUNC | IMA_MASK } ,
{ . action = MEASURE , . func = BPRM_CHECK , . mask = MAY_EXEC ,
. flags = IMA_FUNC | IMA_MASK } ,
{ . action = MEASURE , . func = FILE_CHECK , . mask = MAY_READ ,
2017-01-27 19:23:01 +03:00
. uid = GLOBAL_ROOT_UID , . uid_op = & uid_eq ,
. flags = IMA_FUNC | IMA_MASK | IMA_UID } ,
2015-06-11 20:48:33 -04:00
{ . action = MEASURE , . func = MODULE_CHECK , . flags = IMA_FUNC } ,
{ . action = MEASURE , . func = FIRMWARE_CHECK , . flags = IMA_FUNC } ,
} ;
2017-02-13 16:34:35 +11:00
static struct ima_rule_entry default_measurement_rules [ ] __ro_after_init = {
2014-03-04 18:04:20 +02:00
{ . action = MEASURE , . func = MMAP_CHECK , . mask = MAY_EXEC ,
2009-02-04 09:06:58 -05:00
. flags = IMA_FUNC | IMA_MASK } ,
2014-03-04 18:04:20 +02:00
{ . action = MEASURE , . func = BPRM_CHECK , . mask = MAY_EXEC ,
2009-02-04 09:06:58 -05:00
. flags = IMA_FUNC | IMA_MASK } ,
2015-06-11 20:48:33 -04:00
{ . action = MEASURE , . func = FILE_CHECK , . mask = MAY_READ ,
2017-01-27 19:23:01 +03:00
. uid = GLOBAL_ROOT_UID , . uid_op = & uid_eq ,
. flags = IMA_FUNC | IMA_INMASK | IMA_EUID } ,
2015-06-11 20:48:33 -04:00
{ . action = MEASURE , . func = FILE_CHECK , . mask = MAY_READ ,
2017-01-27 19:23:01 +03:00
. uid = GLOBAL_ROOT_UID , . uid_op = & uid_eq ,
. flags = IMA_FUNC | IMA_INMASK | IMA_UID } ,
2014-03-04 18:04:20 +02:00
{ . action = MEASURE , . func = MODULE_CHECK , . flags = IMA_FUNC } ,
2014-07-22 10:39:48 -04:00
{ . action = MEASURE , . func = FIRMWARE_CHECK , . flags = IMA_FUNC } ,
2016-01-15 10:17:12 -05:00
{ . action = MEASURE , . func = POLICY_CHECK , . flags = IMA_FUNC } ,
2009-02-04 09:06:58 -05:00
} ;
2017-02-13 16:34:35 +11:00
static struct ima_rule_entry default_appraise_rules [ ] __ro_after_init = {
2014-03-04 18:04:20 +02:00
{ . action = DONT_APPRAISE , . fsmagic = PROC_SUPER_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = SYSFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = DEBUGFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = TMPFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = RAMFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = DEVPTS_SUPER_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = BINFMTFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = SECURITYFS_MAGIC , . flags = IMA_FSMAGIC } ,
{ . action = DONT_APPRAISE , . fsmagic = SELINUX_MAGIC , . flags = IMA_FSMAGIC } ,
2018-03-22 13:46:01 +00:00
{ . action = DONT_APPRAISE , . fsmagic = SMACK_MAGIC , . flags = IMA_FSMAGIC } ,
2015-04-21 16:54:24 -04:00
{ . action = DONT_APPRAISE , . fsmagic = NSFS_MAGIC , . flags = IMA_FSMAGIC } ,
2018-11-14 17:24:13 -05:00
{ . action = DONT_APPRAISE , . fsmagic = EFIVARFS_MAGIC , . flags = IMA_FSMAGIC } ,
2014-03-04 18:04:20 +02:00
{ . action = DONT_APPRAISE , . fsmagic = CGROUP_SUPER_MAGIC , . flags = IMA_FSMAGIC } ,
2017-05-09 11:25:27 -07:00
{ . action = DONT_APPRAISE , . fsmagic = CGROUP2_SUPER_MAGIC , . flags = IMA_FSMAGIC } ,
2015-12-07 15:08:01 -05:00
# ifdef CONFIG_IMA_WRITE_POLICY
{ . action = APPRAISE , . func = POLICY_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
# endif
2014-11-05 17:01:16 +02:00
# ifndef CONFIG_IMA_APPRAISE_SIGNED_INIT
2017-01-27 19:23:01 +03:00
{ . action = APPRAISE , . fowner = GLOBAL_ROOT_UID , . fowner_op = & uid_eq ,
. flags = IMA_FOWNER } ,
2014-11-05 17:01:16 +02:00
# else
/* force signature */
2017-01-27 19:23:01 +03:00
{ . action = APPRAISE , . fowner = GLOBAL_ROOT_UID , . fowner_op = & uid_eq ,
2014-11-05 17:01:16 +02:00
. flags = IMA_FOWNER | IMA_DIGSIG_REQUIRED } ,
# endif
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
} ;
2018-07-13 14:06:01 -04:00
static struct ima_rule_entry build_appraise_rules [ ] __ro_after_init = {
# ifdef CONFIG_IMA_APPRAISE_REQUIRE_MODULE_SIGS
{ . action = APPRAISE , . func = MODULE_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
# endif
# ifdef CONFIG_IMA_APPRAISE_REQUIRE_FIRMWARE_SIGS
{ . action = APPRAISE , . func = FIRMWARE_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
# endif
# ifdef CONFIG_IMA_APPRAISE_REQUIRE_KEXEC_SIGS
{ . action = APPRAISE , . func = KEXEC_KERNEL_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
# endif
# ifdef CONFIG_IMA_APPRAISE_REQUIRE_POLICY_SIGS
{ . action = APPRAISE , . func = POLICY_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
# endif
} ;
2017-04-21 18:58:27 -04:00
static struct ima_rule_entry secure_boot_rules [ ] __ro_after_init = {
{ . action = APPRAISE , . func = MODULE_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
{ . action = APPRAISE , . func = FIRMWARE_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
{ . action = APPRAISE , . func = KEXEC_KERNEL_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
{ . action = APPRAISE , . func = POLICY_CHECK ,
. flags = IMA_FUNC | IMA_DIGSIG_REQUIRED } ,
} ;
2021-01-07 20:07:07 -08:00
static struct ima_rule_entry critical_data_rules [ ] __ro_after_init = {
{ . action = MEASURE , . func = CRITICAL_DATA , . flags = IMA_FUNC } ,
} ;
2018-10-09 23:00:36 +05:30
/* An array of architecture specific rules */
2019-06-11 21:40:32 +08:00
static struct ima_rule_entry * arch_policy_entry __ro_after_init ;
2018-10-09 23:00:36 +05:30
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
static LIST_HEAD ( ima_default_rules ) ;
static LIST_HEAD ( ima_policy_rules ) ;
2015-12-02 17:47:54 +02:00
static LIST_HEAD ( ima_temp_rules ) ;
2021-10-09 18:38:21 +08:00
static struct list_head __rcu * ima_rules = ( struct list_head __rcu * ) ( & ima_default_rules ) ;
2009-02-04 09:06:58 -05:00
2015-06-11 20:48:33 -04:00
static int ima_policy __initdata ;
2015-12-02 17:47:54 +02:00
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
static int __init default_measure_policy_setup ( char * str )
2009-05-21 15:47:06 -04:00
{
2015-06-11 20:48:33 -04:00
if ( ima_policy )
return 1 ;
ima_policy = ORIGINAL_TCB ;
2009-05-21 15:47:06 -04:00
return 1 ;
}
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
__setup ( " ima_tcb " , default_measure_policy_setup ) ;
2017-04-24 12:04:09 -04:00
static bool ima_use_appraise_tcb __initdata ;
2017-04-21 18:58:27 -04:00
static bool ima_use_secure_boot __initdata ;
2021-01-07 20:07:07 -08:00
static bool ima_use_critical_data __initdata ;
2018-02-21 11:36:32 -05:00
static bool ima_fail_unverifiable_sigs __ro_after_init ;
2015-06-11 20:48:33 -04:00
static int __init policy_setup ( char * str )
{
2017-04-24 12:04:09 -04:00
char * p ;
2015-06-11 20:48:33 -04:00
2017-04-24 12:04:09 -04:00
while ( ( p = strsep ( & str , " | \n " ) ) ! = NULL ) {
if ( * p = = ' ' )
continue ;
if ( ( strcmp ( p , " tcb " ) = = 0 ) & & ! ima_policy )
ima_policy = DEFAULT_TCB ;
else if ( strcmp ( p , " appraise_tcb " ) = = 0 )
2017-10-07 16:02:21 +02:00
ima_use_appraise_tcb = true ;
2017-04-21 18:58:27 -04:00
else if ( strcmp ( p , " secure_boot " ) = = 0 )
2017-10-07 16:02:21 +02:00
ima_use_secure_boot = true ;
2021-01-07 20:07:07 -08:00
else if ( strcmp ( p , " critical_data " ) = = 0 )
ima_use_critical_data = true ;
2018-02-21 11:36:32 -05:00
else if ( strcmp ( p , " fail_securely " ) = = 0 )
ima_fail_unverifiable_sigs = true ;
2020-09-04 16:40:58 -03:00
else
pr_err ( " policy \" %s \" not found " , p ) ;
2017-04-24 12:04:09 -04:00
}
2015-06-11 20:48:33 -04:00
return 1 ;
}
__setup ( " ima_policy= " , policy_setup ) ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
static int __init default_appraise_policy_setup ( char * str )
{
2017-10-07 16:02:21 +02:00
ima_use_appraise_tcb = true ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
return 1 ;
}
__setup ( " ima_appraise_tcb " , default_appraise_policy_setup ) ;
2009-05-21 15:47:06 -04:00
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
static struct ima_rule_opt_list * ima_alloc_rule_opt_list ( const substring_t * src )
{
struct ima_rule_opt_list * opt_list ;
size_t count = 0 ;
char * src_copy ;
char * cur , * next ;
size_t i ;
src_copy = match_strdup ( src ) ;
if ( ! src_copy )
return ERR_PTR ( - ENOMEM ) ;
next = src_copy ;
while ( ( cur = strsep ( & next , " | " ) ) ) {
/* Don't accept an empty list item */
if ( ! ( * cur ) ) {
kfree ( src_copy ) ;
return ERR_PTR ( - EINVAL ) ;
}
count + + ;
}
/* Don't accept an empty list */
if ( ! count ) {
kfree ( src_copy ) ;
return ERR_PTR ( - EINVAL ) ;
}
opt_list = kzalloc ( struct_size ( opt_list , items , count ) , GFP_KERNEL ) ;
if ( ! opt_list ) {
kfree ( src_copy ) ;
return ERR_PTR ( - ENOMEM ) ;
}
/*
* strsep ( ) has already replaced all instances of ' | ' with ' \0 ' ,
* leaving a byte sequence of NUL - terminated strings . Reference each
* string with the array of items .
*
* IMPORTANT : Ownership of the allocated buffer is transferred from
* src_copy to the first element in the items array . To free the
* buffer , kfree ( ) must only be called on the first element of the
* array .
*/
for ( i = 0 , cur = src_copy ; i < count ; i + + ) {
opt_list - > items [ i ] = cur ;
cur = strchr ( cur , ' \0 ' ) + 1 ;
}
opt_list - > count = count ;
return opt_list ;
}
static void ima_free_rule_opt_list ( struct ima_rule_opt_list * opt_list )
{
if ( ! opt_list )
return ;
if ( opt_list - > count ) {
kfree ( opt_list - > items [ 0 ] ) ;
opt_list - > count = 0 ;
}
kfree ( opt_list ) ;
}
2019-06-14 15:20:15 +03:00
static void ima_lsm_free_rule ( struct ima_rule_entry * entry )
{
int i ;
for ( i = 0 ; i < MAX_LSM_RULES ; i + + ) {
2020-07-10 15:37:50 -05:00
ima_filter_rule_free ( entry - > lsm [ i ] . rule ) ;
2019-06-14 15:20:15 +03:00
kfree ( entry - > lsm [ i ] . args_p ) ;
}
2020-07-09 01:19:01 -05:00
}
static void ima_free_rule ( struct ima_rule_entry * entry )
{
if ( ! entry )
return ;
/*
* entry - > template - > fields may be allocated in ima_parse_rule ( ) but that
* reference is owned by the corresponding ima_template_desc element in
* the defined_templates list and cannot be freed here
*/
kfree ( entry - > fsname ) ;
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
ima_free_rule_opt_list ( entry - > keyrings ) ;
2020-07-09 01:19:01 -05:00
ima_lsm_free_rule ( entry ) ;
2019-06-14 15:20:15 +03:00
kfree ( entry ) ;
}
static struct ima_rule_entry * ima_lsm_copy_rule ( struct ima_rule_entry * entry )
{
struct ima_rule_entry * nentry ;
2020-01-15 17:42:30 +02:00
int i ;
2019-06-14 15:20:15 +03:00
/*
* Immutable elements are copied over as pointers and data ; only
* lsm rules can change
*/
2020-09-09 20:09:06 +01:00
nentry = kmemdup ( entry , sizeof ( * nentry ) , GFP_KERNEL ) ;
if ( ! nentry )
return NULL ;
2019-12-09 10:31:43 -08:00
memset ( nentry - > lsm , 0 , sizeof_field ( struct ima_rule_entry , lsm ) ) ;
2019-06-14 15:20:15 +03:00
for ( i = 0 ; i < MAX_LSM_RULES ; i + + ) {
2020-01-15 17:42:30 +02:00
if ( ! entry - > lsm [ i ] . args_p )
2019-06-14 15:20:15 +03:00
continue ;
nentry - > lsm [ i ] . type = entry - > lsm [ i ] . type ;
2020-07-09 01:19:07 -05:00
nentry - > lsm [ i ] . args_p = entry - > lsm [ i ] . args_p ;
/*
* Remove the reference from entry so that the associated
* memory will not be freed during a later call to
* ima_lsm_free_rule ( entry ) .
*/
entry - > lsm [ i ] . args_p = NULL ;
2019-06-14 15:20:15 +03:00
2020-07-10 15:37:50 -05:00
ima_filter_rule_init ( nentry - > lsm [ i ] . type , Audit_equal ,
nentry - > lsm [ i ] . args_p ,
& nentry - > lsm [ i ] . rule ) ;
2020-01-15 17:42:30 +02:00
if ( ! nentry - > lsm [ i ] . rule )
pr_warn ( " rule for LSM \' %s \' is undefined \n " ,
2020-07-09 01:19:08 -05:00
nentry - > lsm [ i ] . args_p ) ;
2019-06-14 15:20:15 +03:00
}
return nentry ;
}
static int ima_lsm_update_rule ( struct ima_rule_entry * entry )
{
struct ima_rule_entry * nentry ;
nentry = ima_lsm_copy_rule ( entry ) ;
if ( ! nentry )
return - ENOMEM ;
list_replace_rcu ( & entry - > list , & nentry - > list ) ;
synchronize_rcu ( ) ;
2020-07-09 01:19:01 -05:00
/*
* ima_lsm_copy_rule ( ) shallow copied all references , except for the
* LSM references , from entry to nentry so we only want to free the LSM
2021-11-24 21:44:18 +00:00
* references and the entry itself . All other memory references will now
2020-07-09 01:19:01 -05:00
* be owned by nentry .
*/
2019-06-14 15:20:15 +03:00
ima_lsm_free_rule ( entry ) ;
2020-07-09 01:19:01 -05:00
kfree ( entry ) ;
2019-06-14 15:20:15 +03:00
return 0 ;
}
2020-07-09 01:19:04 -05:00
static bool ima_rule_contains_lsm_cond ( struct ima_rule_entry * entry )
{
int i ;
for ( i = 0 ; i < MAX_LSM_RULES ; i + + )
if ( entry - > lsm [ i ] . args_p )
return true ;
return false ;
}
2014-03-04 18:04:20 +02:00
/*
2015-12-02 17:47:54 +02:00
* The LSM policy can be reloaded , leaving the IMA LSM based rules referring
* to the old , stale LSM policy . Update the IMA LSM based rules to reflect
2019-06-14 15:20:15 +03:00
* the reloaded LSM policy .
2013-01-03 14:19:09 -05:00
*/
static void ima_lsm_update_rules ( void )
{
2019-06-14 15:20:15 +03:00
struct ima_rule_entry * entry , * e ;
2020-07-09 01:19:10 -05:00
int result ;
2013-01-03 14:19:09 -05:00
2019-06-14 15:20:15 +03:00
list_for_each_entry_safe ( entry , e , & ima_policy_rules , list ) {
2020-07-09 01:19:10 -05:00
if ( ! ima_rule_contains_lsm_cond ( entry ) )
2019-06-14 15:20:15 +03:00
continue ;
result = ima_lsm_update_rule ( entry ) ;
if ( result ) {
2020-01-15 17:42:30 +02:00
pr_err ( " lsm rule update error %d \n " , result ) ;
2019-06-14 15:20:15 +03:00
return ;
2013-01-03 14:19:09 -05:00
}
}
}
2019-06-14 15:20:15 +03:00
int ima_lsm_policy_change ( struct notifier_block * nb , unsigned long event ,
void * lsm_data )
{
if ( event ! = LSM_POLICY_CHANGE )
return NOTIFY_DONE ;
ima_lsm_update_rules ( ) ;
return NOTIFY_OK ;
}
2009-02-04 09:06:58 -05:00
/**
2021-01-07 20:07:01 -08:00
* ima_match_rule_data - determine whether func_data matches the policy rule
2019-12-11 08:47:06 -08:00
* @ rule : a pointer to a rule
2021-01-07 20:07:01 -08:00
* @ func_data : data to match against the measure rule data
2019-12-11 08:47:06 -08:00
* @ cred : a pointer to a credentials structure for user validation
*
2021-01-07 20:07:01 -08:00
* Returns true if func_data matches one in the rule , false otherwise .
2019-12-11 08:47:06 -08:00
*/
2021-01-07 20:07:01 -08:00
static bool ima_match_rule_data ( struct ima_rule_entry * rule ,
const char * func_data ,
const struct cred * cred )
2019-12-11 08:47:06 -08:00
{
2021-01-07 20:07:01 -08:00
const struct ima_rule_opt_list * opt_list = NULL ;
2019-12-11 08:47:06 -08:00
bool matched = false ;
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
size_t i ;
2019-12-11 08:47:06 -08:00
if ( ( rule - > flags & IMA_UID ) & & ! rule - > uid_op ( cred - > uid , rule - > uid ) )
return false ;
2021-01-07 20:07:01 -08:00
switch ( rule - > func ) {
case KEY_CHECK :
if ( ! rule - > keyrings )
return true ;
opt_list = rule - > keyrings ;
break ;
2021-01-07 20:07:04 -08:00
case CRITICAL_DATA :
2021-01-07 20:07:05 -08:00
if ( ! rule - > label )
return true ;
opt_list = rule - > label ;
break ;
2021-01-07 20:07:01 -08:00
default :
return false ;
}
2019-12-11 08:47:06 -08:00
2021-01-07 20:07:01 -08:00
if ( ! func_data )
2019-12-11 08:47:06 -08:00
return false ;
2021-01-07 20:07:01 -08:00
for ( i = 0 ; i < opt_list - > count ; i + + ) {
if ( ! strcmp ( opt_list - > items [ i ] , func_data ) ) {
2019-12-11 08:47:06 -08:00
matched = true ;
break ;
}
}
return matched ;
}
2009-02-04 09:06:58 -05:00
/**
2020-01-15 17:42:30 +02:00
* ima_match_rules - determine whether an inode matches the policy rule .
2009-02-04 09:06:58 -05:00
* @ rule : a pointer to a rule
2021-01-21 14:19:45 +01:00
* @ mnt_userns : user namespace of the mount the inode was found from
2009-02-04 09:06:58 -05:00
* @ inode : a pointer to an inode
2018-01-08 13:36:20 -08:00
* @ cred : a pointer to a credentials structure for user validation
* @ secid : the secid of the task to be validated
2009-02-04 09:06:58 -05:00
* @ func : LIM hook identifier
* @ mask : requested action ( MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC )
2021-01-07 20:07:01 -08:00
* @ func_data : func specific data , may be NULL
2009-02-04 09:06:58 -05:00
*
* Returns true on rule match , false on failure .
*/
2021-01-21 14:19:45 +01:00
static bool ima_match_rules ( struct ima_rule_entry * rule ,
struct user_namespace * mnt_userns ,
struct inode * inode , const struct cred * cred ,
u32 secid , enum ima_hooks func , int mask ,
2021-01-07 20:07:01 -08:00
const char * func_data )
2009-02-04 09:06:58 -05:00
{
2009-02-04 09:07:00 -05:00
int i ;
2009-02-04 09:06:58 -05:00
2013-11-13 23:42:39 +02:00
if ( ( rule - > flags & IMA_FUNC ) & &
( rule - > func ! = func & & func ! = POST_SETATTR ) )
2009-02-04 09:06:58 -05:00
return false ;
2021-01-07 20:07:04 -08:00
switch ( func ) {
case KEY_CHECK :
case CRITICAL_DATA :
return ( ( rule - > func = = func ) & &
ima_match_rule_data ( rule , func_data , cred ) ) ;
default :
break ;
}
2013-11-13 23:42:39 +02:00
if ( ( rule - > flags & IMA_MASK ) & &
( rule - > mask ! = mask & & func ! = POST_SETATTR ) )
2009-02-04 09:06:58 -05:00
return false ;
2014-11-05 07:53:55 -05:00
if ( ( rule - > flags & IMA_INMASK ) & &
( ! ( rule - > mask & mask ) & & func ! = POST_SETATTR ) )
return false ;
2009-02-04 09:06:58 -05:00
if ( ( rule - > flags & IMA_FSMAGIC )
& & rule - > fsmagic ! = inode - > i_sb - > s_magic )
return false ;
2018-01-15 11:20:36 -05:00
if ( ( rule - > flags & IMA_FSNAME )
& & strcmp ( rule - > fsname , inode - > i_sb - > s_type - > name ) )
return false ;
2012-09-03 23:23:13 +03:00
if ( ( rule - > flags & IMA_FSUUID ) & &
2017-05-10 15:06:33 +02:00
! uuid_equal ( & rule - > fsuuid , & inode - > i_sb - > s_uuid ) )
2012-09-03 23:23:13 +03:00
return false ;
2017-01-27 19:23:01 +03:00
if ( ( rule - > flags & IMA_UID ) & & ! rule - > uid_op ( cred - > uid , rule - > uid ) )
2009-02-04 09:06:58 -05:00
return false ;
2014-11-05 07:48:36 -05:00
if ( rule - > flags & IMA_EUID ) {
if ( has_capability_noaudit ( current , CAP_SETUID ) ) {
2017-01-27 19:23:01 +03:00
if ( ! rule - > uid_op ( cred - > euid , rule - > uid )
& & ! rule - > uid_op ( cred - > suid , rule - > uid )
& & ! rule - > uid_op ( cred - > uid , rule - > uid ) )
2014-11-05 07:48:36 -05:00
return false ;
2017-01-27 19:23:01 +03:00
} else if ( ! rule - > uid_op ( cred - > euid , rule - > uid ) )
2014-11-05 07:48:36 -05:00
return false ;
}
2021-10-07 14:03:02 -06:00
if ( ( rule - > flags & IMA_GID ) & & ! rule - > gid_op ( cred - > gid , rule - > gid ) )
return false ;
if ( rule - > flags & IMA_EGID ) {
if ( has_capability_noaudit ( current , CAP_SETGID ) ) {
if ( ! rule - > gid_op ( cred - > egid , rule - > gid )
& & ! rule - > gid_op ( cred - > sgid , rule - > gid )
& & ! rule - > gid_op ( cred - > gid , rule - > gid ) )
return false ;
} else if ( ! rule - > gid_op ( cred - > egid , rule - > gid ) )
return false ;
}
2017-01-27 19:23:01 +03:00
if ( ( rule - > flags & IMA_FOWNER ) & &
2021-01-21 14:19:45 +01:00
! rule - > fowner_op ( i_uid_into_mnt ( mnt_userns , inode ) , rule - > fowner ) )
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
return false ;
2021-10-07 14:03:02 -06:00
if ( ( rule - > flags & IMA_FGROUP ) & &
! rule - > fgroup_op ( i_gid_into_mnt ( mnt_userns , inode ) , rule - > fgroup ) )
return false ;
2009-02-04 09:07:00 -05:00
for ( i = 0 ; i < MAX_LSM_RULES ; i + + ) {
2009-05-05 13:12:48 -04:00
int rc = 0 ;
2018-01-08 13:36:20 -08:00
u32 osid ;
2009-02-04 09:07:00 -05:00
2020-01-15 17:42:30 +02:00
if ( ! rule - > lsm [ i ] . rule ) {
if ( ! rule - > lsm [ i ] . args_p )
continue ;
else
return false ;
}
2009-02-04 09:07:00 -05:00
switch ( i ) {
case LSM_OBJ_USER :
case LSM_OBJ_ROLE :
case LSM_OBJ_TYPE :
security_inode_getsecid ( inode , & osid ) ;
2020-07-10 15:37:50 -05:00
rc = ima_filter_rule_match ( osid , rule - > lsm [ i ] . type ,
Audit_equal ,
rule - > lsm [ i ] . rule ) ;
2009-02-04 09:07:00 -05:00
break ;
case LSM_SUBJ_USER :
case LSM_SUBJ_ROLE :
case LSM_SUBJ_TYPE :
2020-07-10 15:37:50 -05:00
rc = ima_filter_rule_match ( secid , rule - > lsm [ i ] . type ,
Audit_equal ,
rule - > lsm [ i ] . rule ) ;
2020-11-20 12:25:46 -06:00
break ;
2009-02-04 09:07:00 -05:00
default :
break ;
}
if ( ! rc )
return false ;
}
2009-02-04 09:06:58 -05:00
return true ;
}
2012-12-03 17:08:11 -05:00
/*
* In addition to knowing that we need to appraise the file in general ,
2012-12-05 15:14:38 -05:00
* we need to differentiate between calling hooks , for hook specific rules .
2012-12-03 17:08:11 -05:00
*/
2016-01-14 20:59:14 -05:00
static int get_subaction ( struct ima_rule_entry * rule , enum ima_hooks func )
2012-12-03 17:08:11 -05:00
{
2012-12-05 15:14:38 -05:00
if ( ! ( rule - > flags & IMA_FUNC ) )
return IMA_FILE_APPRAISE ;
2014-03-04 18:04:20 +02:00
switch ( func ) {
2012-12-03 17:08:11 -05:00
case MMAP_CHECK :
return IMA_MMAP_APPRAISE ;
case BPRM_CHECK :
return IMA_BPRM_APPRAISE ;
2018-01-08 13:36:20 -08:00
case CREDS_CHECK :
return IMA_CREDS_APPRAISE ;
2012-12-03 17:08:11 -05:00
case FILE_CHECK :
2015-11-19 12:39:22 -05:00
case POST_SETATTR :
2012-12-03 17:08:11 -05:00
return IMA_FILE_APPRAISE ;
2015-11-19 12:39:22 -05:00
case MODULE_CHECK . . . MAX_CHECK - 1 :
default :
return IMA_READ_APPRAISE ;
2012-12-03 17:08:11 -05:00
}
}
2009-02-04 09:06:58 -05:00
/**
* ima_match_policy - decision based on LSM and other conditions
2021-01-21 14:19:45 +01:00
* @ mnt_userns : user namespace of the mount the inode was found from
2009-02-04 09:06:58 -05:00
* @ inode : pointer to an inode for which the policy decision is being made
2018-01-08 13:36:20 -08:00
* @ cred : pointer to a credentials structure for which the policy decision is
* being made
* @ secid : LSM secid of the task to be validated
2009-02-04 09:06:58 -05:00
* @ func : IMA hook identifier
* @ mask : requested action ( MAY_READ | MAY_WRITE | MAY_APPEND | MAY_EXEC )
2016-06-01 13:14:02 -05:00
* @ pcr : set the pcr to extend
2019-06-19 15:46:11 -07:00
* @ template_desc : the template that should be used for this rule
2021-01-07 20:07:01 -08:00
* @ func_data : func specific data , may be NULL
2021-08-16 08:11:00 +00:00
* @ allowed_algos : allowlist of hash algorithms for the IMA xattr
2009-02-04 09:06:58 -05:00
*
* Measure decision based on func / mask / fsmagic and LSM ( subj / obj / type )
* conditions .
*
2015-12-02 17:47:54 +02:00
* Since the IMA policy may be updated multiple times we need to lock the
* list when walking it . Reads are many orders of magnitude more numerous
* than writes so ima_match_policy ( ) is classical RCU candidate .
2009-02-04 09:06:58 -05:00
*/
2021-01-21 14:19:45 +01:00
int ima_match_policy ( struct user_namespace * mnt_userns , struct inode * inode ,
const struct cred * cred , u32 secid , enum ima_hooks func ,
int mask , int flags , int * pcr ,
2019-12-11 08:47:06 -08:00
struct ima_template_desc * * template_desc ,
2021-08-16 08:11:00 +00:00
const char * func_data , unsigned int * allowed_algos )
2009-02-04 09:06:58 -05:00
{
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
struct ima_rule_entry * entry ;
ima: integrity appraisal extension
IMA currently maintains an integrity measurement list used to assert the
integrity of the running system to a third party. The IMA-appraisal
extension adds local integrity validation and enforcement of the
measurement against a "good" value stored as an extended attribute
'security.ima'. The initial methods for validating 'security.ima' are
hashed based, which provides file data integrity, and digital signature
based, which in addition to providing file data integrity, provides
authenticity.
This patch creates and maintains the 'security.ima' xattr, containing
the file data hash measurement. Protection of the xattr is provided by
EVM, if enabled and configured.
Based on policy, IMA calls evm_verifyxattr() to verify a file's metadata
integrity and, assuming success, compares the file's current hash value
with the one stored as an extended attribute in 'security.ima'.
Changelov v4:
- changed iint cache flags to hex values
Changelog v3:
- change appraisal default for filesystems without xattr support to fail
Changelog v2:
- fix audit msg 'res' value
- removed unused 'ima_appraise=' values
Changelog v1:
- removed unused iint mutex (Dmitry Kasatkin)
- setattr hook must not reset appraised (Dmitry Kasatkin)
- evm_verifyxattr() now differentiates between no 'security.evm' xattr
(INTEGRITY_NOLABEL) and no EVM 'protected' xattrs included in the
'security.evm' (INTEGRITY_NOXATTRS).
- replace hash_status with ima_status (Dmitry Kasatkin)
- re-initialize slab element ima_status on free (Dmitry Kasatkin)
- include 'security.ima' in EVM if CONFIG_IMA_APPRAISE, not CONFIG_IMA
- merged half "ima: ima_must_appraise_or_measure API change" (Dmitry Kasatkin)
- removed unnecessary error variable in process_measurement() (Dmitry Kasatkin)
- use ima_inode_post_setattr() stub function, if IMA_APPRAISE not configured
(moved ima_inode_post_setattr() to ima_appraise.c)
- make sure ima_collect_measurement() can read file
Changelog:
- add 'iint' to evm_verifyxattr() call (Dimitry Kasatkin)
- fix the race condition between chmod, which takes the i_mutex and then
iint->mutex, and ima_file_free() and process_measurement(), which take
the locks in the reverse order, by eliminating iint->mutex. (Dmitry Kasatkin)
- cleanup of ima_appraise_measurement() (Dmitry Kasatkin)
- changes as a result of the iint not allocated for all regular files, but
only for those measured/appraised.
- don't try to appraise new/empty files
- expanded ima_appraisal description in ima/Kconfig
- IMA appraise definitions required even if IMA_APPRAISE not enabled
- add return value to ima_must_appraise() stub
- unconditionally set status = INTEGRITY_PASS *after* testing status,
not before. (Found by Joe Perches)
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2012-02-13 10:15:05 -05:00
int action = 0 , actmask = flags | ( flags < < 1 ) ;
2021-10-09 18:38:21 +08:00
struct list_head * ima_rules_tmp ;
2009-02-04 09:06:58 -05:00
2020-11-12 12:39:59 -08:00
if ( template_desc & & ! * template_desc )
2019-07-19 07:16:57 -04:00
* template_desc = ima_template_desc_current ( ) ;
2015-12-02 17:47:54 +02:00
rcu_read_lock ( ) ;
2021-10-09 18:38:21 +08:00
ima_rules_tmp = rcu_dereference ( ima_rules ) ;
list_for_each_entry_rcu ( entry , ima_rules_tmp , list ) {
2009-02-04 09:06:58 -05:00
ima: integrity appraisal extension
IMA currently maintains an integrity measurement list used to assert the
integrity of the running system to a third party. The IMA-appraisal
extension adds local integrity validation and enforcement of the
measurement against a "good" value stored as an extended attribute
'security.ima'. The initial methods for validating 'security.ima' are
hashed based, which provides file data integrity, and digital signature
based, which in addition to providing file data integrity, provides
authenticity.
This patch creates and maintains the 'security.ima' xattr, containing
the file data hash measurement. Protection of the xattr is provided by
EVM, if enabled and configured.
Based on policy, IMA calls evm_verifyxattr() to verify a file's metadata
integrity and, assuming success, compares the file's current hash value
with the one stored as an extended attribute in 'security.ima'.
Changelov v4:
- changed iint cache flags to hex values
Changelog v3:
- change appraisal default for filesystems without xattr support to fail
Changelog v2:
- fix audit msg 'res' value
- removed unused 'ima_appraise=' values
Changelog v1:
- removed unused iint mutex (Dmitry Kasatkin)
- setattr hook must not reset appraised (Dmitry Kasatkin)
- evm_verifyxattr() now differentiates between no 'security.evm' xattr
(INTEGRITY_NOLABEL) and no EVM 'protected' xattrs included in the
'security.evm' (INTEGRITY_NOXATTRS).
- replace hash_status with ima_status (Dmitry Kasatkin)
- re-initialize slab element ima_status on free (Dmitry Kasatkin)
- include 'security.ima' in EVM if CONFIG_IMA_APPRAISE, not CONFIG_IMA
- merged half "ima: ima_must_appraise_or_measure API change" (Dmitry Kasatkin)
- removed unnecessary error variable in process_measurement() (Dmitry Kasatkin)
- use ima_inode_post_setattr() stub function, if IMA_APPRAISE not configured
(moved ima_inode_post_setattr() to ima_appraise.c)
- make sure ima_collect_measurement() can read file
Changelog:
- add 'iint' to evm_verifyxattr() call (Dimitry Kasatkin)
- fix the race condition between chmod, which takes the i_mutex and then
iint->mutex, and ima_file_free() and process_measurement(), which take
the locks in the reverse order, by eliminating iint->mutex. (Dmitry Kasatkin)
- cleanup of ima_appraise_measurement() (Dmitry Kasatkin)
- changes as a result of the iint not allocated for all regular files, but
only for those measured/appraised.
- don't try to appraise new/empty files
- expanded ima_appraisal description in ima/Kconfig
- IMA appraise definitions required even if IMA_APPRAISE not enabled
- add return value to ima_must_appraise() stub
- unconditionally set status = INTEGRITY_PASS *after* testing status,
not before. (Found by Joe Perches)
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2012-02-13 10:15:05 -05:00
if ( ! ( entry - > action & actmask ) )
continue ;
2021-01-21 14:19:45 +01:00
if ( ! ima_match_rules ( entry , mnt_userns , inode , cred , secid ,
idmapped-mounts-v5.12
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCYCegywAKCRCRxhvAZXjc
ouJ6AQDlf+7jCQlQdeKKoN9QDFfMzG1ooemat36EpRRTONaGuAD8D9A4sUsG4+5f
4IU5Lj9oY4DEmF8HenbWK2ZHsesL2Qg=
=yPaw
-----END PGP SIGNATURE-----
Merge tag 'idmapped-mounts-v5.12' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux
Pull idmapped mounts from Christian Brauner:
"This introduces idmapped mounts which has been in the making for some
time. Simply put, different mounts can expose the same file or
directory with different ownership. This initial implementation comes
with ports for fat, ext4 and with Christoph's port for xfs with more
filesystems being actively worked on by independent people and
maintainers.
Idmapping mounts handle a wide range of long standing use-cases. Here
are just a few:
- Idmapped mounts make it possible to easily share files between
multiple users or multiple machines especially in complex
scenarios. For example, idmapped mounts will be used in the
implementation of portable home directories in
systemd-homed.service(8) where they allow users to move their home
directory to an external storage device and use it on multiple
computers where they are assigned different uids and gids. This
effectively makes it possible to assign random uids and gids at
login time.
- It is possible to share files from the host with unprivileged
containers without having to change ownership permanently through
chown(2).
- It is possible to idmap a container's rootfs and without having to
mangle every file. For example, Chromebooks use it to share the
user's Download folder with their unprivileged containers in their
Linux subsystem.
- It is possible to share files between containers with
non-overlapping idmappings.
- Filesystem that lack a proper concept of ownership such as fat can
use idmapped mounts to implement discretionary access (DAC)
permission checking.
- They allow users to efficiently changing ownership on a per-mount
basis without having to (recursively) chown(2) all files. In
contrast to chown (2) changing ownership of large sets of files is
instantenous with idmapped mounts. This is especially useful when
ownership of a whole root filesystem of a virtual machine or
container is changed. With idmapped mounts a single syscall
mount_setattr syscall will be sufficient to change the ownership of
all files.
- Idmapped mounts always take the current ownership into account as
idmappings specify what a given uid or gid is supposed to be mapped
to. This contrasts with the chown(2) syscall which cannot by itself
take the current ownership of the files it changes into account. It
simply changes the ownership to the specified uid and gid. This is
especially problematic when recursively chown(2)ing a large set of
files which is commong with the aforementioned portable home
directory and container and vm scenario.
- Idmapped mounts allow to change ownership locally, restricting it
to specific mounts, and temporarily as the ownership changes only
apply as long as the mount exists.
Several userspace projects have either already put up patches and
pull-requests for this feature or will do so should you decide to pull
this:
- systemd: In a wide variety of scenarios but especially right away
in their implementation of portable home directories.
https://systemd.io/HOME_DIRECTORY/
- container runtimes: containerd, runC, LXD:To share data between
host and unprivileged containers, unprivileged and privileged
containers, etc. The pull request for idmapped mounts support in
containerd, the default Kubernetes runtime is already up for quite
a while now: https://github.com/containerd/containerd/pull/4734
- The virtio-fs developers and several users have expressed interest
in using this feature with virtual machines once virtio-fs is
ported.
- ChromeOS: Sharing host-directories with unprivileged containers.
I've tightly synced with all those projects and all of those listed
here have also expressed their need/desire for this feature on the
mailing list. For more info on how people use this there's a bunch of
talks about this too. Here's just two recent ones:
https://www.cncf.io/wp-content/uploads/2020/12/Rootless-Containers-in-Gitpod.pdf
https://fosdem.org/2021/schedule/event/containers_idmap/
This comes with an extensive xfstests suite covering both ext4 and
xfs:
https://git.kernel.org/brauner/xfstests-dev/h/idmapped_mounts
It covers truncation, creation, opening, xattrs, vfscaps, setid
execution, setgid inheritance and more both with idmapped and
non-idmapped mounts. It already helped to discover an unrelated xfs
setgid inheritance bug which has since been fixed in mainline. It will
be sent for inclusion with the xfstests project should you decide to
merge this.
In order to support per-mount idmappings vfsmounts are marked with
user namespaces. The idmapping of the user namespace will be used to
map the ids of vfs objects when they are accessed through that mount.
By default all vfsmounts are marked with the initial user namespace.
The initial user namespace is used to indicate that a mount is not
idmapped. All operations behave as before and this is verified in the
testsuite.
Based on prior discussions we want to attach the whole user namespace
and not just a dedicated idmapping struct. This allows us to reuse all
the helpers that already exist for dealing with idmappings instead of
introducing a whole new range of helpers. In addition, if we decide in
the future that we are confident enough to enable unprivileged users
to setup idmapped mounts the permission checking can take into account
whether the caller is privileged in the user namespace the mount is
currently marked with.
The user namespace the mount will be marked with can be specified by
passing a file descriptor refering to the user namespace as an
argument to the new mount_setattr() syscall together with the new
MOUNT_ATTR_IDMAP flag. The system call follows the openat2() pattern
of extensibility.
The following conditions must be met in order to create an idmapped
mount:
- The caller must currently have the CAP_SYS_ADMIN capability in the
user namespace the underlying filesystem has been mounted in.
- The underlying filesystem must support idmapped mounts.
- The mount must not already be idmapped. This also implies that the
idmapping of a mount cannot be altered once it has been idmapped.
- The mount must be a detached/anonymous mount, i.e. it must have
been created by calling open_tree() with the OPEN_TREE_CLONE flag
and it must not already have been visible in the filesystem.
The last two points guarantee easier semantics for userspace and the
kernel and make the implementation significantly simpler.
By default vfsmounts are marked with the initial user namespace and no
behavioral or performance changes are observed.
The manpage with a detailed description can be found here:
https://git.kernel.org/brauner/man-pages/c/1d7b902e2875a1ff342e036a9f866a995640aea8
In order to support idmapped mounts, filesystems need to be changed
and mark themselves with the FS_ALLOW_IDMAP flag in fs_flags. The
patches to convert individual filesystem are not very large or
complicated overall as can be seen from the included fat, ext4, and
xfs ports. Patches for other filesystems are actively worked on and
will be sent out separately. The xfstestsuite can be used to verify
that port has been done correctly.
The mount_setattr() syscall is motivated independent of the idmapped
mounts patches and it's been around since July 2019. One of the most
valuable features of the new mount api is the ability to perform
mounts based on file descriptors only.
Together with the lookup restrictions available in the openat2()
RESOLVE_* flag namespace which we added in v5.6 this is the first time
we are close to hardened and race-free (e.g. symlinks) mounting and
path resolution.
While userspace has started porting to the new mount api to mount
proper filesystems and create new bind-mounts it is currently not
possible to change mount options of an already existing bind mount in
the new mount api since the mount_setattr() syscall is missing.
With the addition of the mount_setattr() syscall we remove this last
restriction and userspace can now fully port to the new mount api,
covering every use-case the old mount api could. We also add the
crucial ability to recursively change mount options for a whole mount
tree, both removing and adding mount options at the same time. This
syscall has been requested multiple times by various people and
projects.
There is a simple tool available at
https://github.com/brauner/mount-idmapped
that allows to create idmapped mounts so people can play with this
patch series. I'll add support for the regular mount binary should you
decide to pull this in the following weeks:
Here's an example to a simple idmapped mount of another user's home
directory:
u1001@f2-vm:/$ sudo ./mount --idmap both:1000:1001:1 /home/ubuntu/ /mnt
u1001@f2-vm:/$ ls -al /home/ubuntu/
total 28
drwxr-xr-x 2 ubuntu ubuntu 4096 Oct 28 22:07 .
drwxr-xr-x 4 root root 4096 Oct 28 04:00 ..
-rw------- 1 ubuntu ubuntu 3154 Oct 28 22:12 .bash_history
-rw-r--r-- 1 ubuntu ubuntu 220 Feb 25 2020 .bash_logout
-rw-r--r-- 1 ubuntu ubuntu 3771 Feb 25 2020 .bashrc
-rw-r--r-- 1 ubuntu ubuntu 807 Feb 25 2020 .profile
-rw-r--r-- 1 ubuntu ubuntu 0 Oct 16 16:11 .sudo_as_admin_successful
-rw------- 1 ubuntu ubuntu 1144 Oct 28 00:43 .viminfo
u1001@f2-vm:/$ ls -al /mnt/
total 28
drwxr-xr-x 2 u1001 u1001 4096 Oct 28 22:07 .
drwxr-xr-x 29 root root 4096 Oct 28 22:01 ..
-rw------- 1 u1001 u1001 3154 Oct 28 22:12 .bash_history
-rw-r--r-- 1 u1001 u1001 220 Feb 25 2020 .bash_logout
-rw-r--r-- 1 u1001 u1001 3771 Feb 25 2020 .bashrc
-rw-r--r-- 1 u1001 u1001 807 Feb 25 2020 .profile
-rw-r--r-- 1 u1001 u1001 0 Oct 16 16:11 .sudo_as_admin_successful
-rw------- 1 u1001 u1001 1144 Oct 28 00:43 .viminfo
u1001@f2-vm:/$ touch /mnt/my-file
u1001@f2-vm:/$ setfacl -m u:1001:rwx /mnt/my-file
u1001@f2-vm:/$ sudo setcap -n 1001 cap_net_raw+ep /mnt/my-file
u1001@f2-vm:/$ ls -al /mnt/my-file
-rw-rwxr--+ 1 u1001 u1001 0 Oct 28 22:14 /mnt/my-file
u1001@f2-vm:/$ ls -al /home/ubuntu/my-file
-rw-rwxr--+ 1 ubuntu ubuntu 0 Oct 28 22:14 /home/ubuntu/my-file
u1001@f2-vm:/$ getfacl /mnt/my-file
getfacl: Removing leading '/' from absolute path names
# file: mnt/my-file
# owner: u1001
# group: u1001
user::rw-
user:u1001:rwx
group::rw-
mask::rwx
other::r--
u1001@f2-vm:/$ getfacl /home/ubuntu/my-file
getfacl: Removing leading '/' from absolute path names
# file: home/ubuntu/my-file
# owner: ubuntu
# group: ubuntu
user::rw-
user:ubuntu:rwx
group::rw-
mask::rwx
other::r--"
* tag 'idmapped-mounts-v5.12' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux: (41 commits)
xfs: remove the possibly unused mp variable in xfs_file_compat_ioctl
xfs: support idmapped mounts
ext4: support idmapped mounts
fat: handle idmapped mounts
tests: add mount_setattr() selftests
fs: introduce MOUNT_ATTR_IDMAP
fs: add mount_setattr()
fs: add attr_flags_to_mnt_flags helper
fs: split out functions to hold writers
namespace: only take read lock in do_reconfigure_mnt()
mount: make {lock,unlock}_mount_hash() static
namespace: take lock_mount_hash() directly when changing flags
nfs: do not export idmapped mounts
overlayfs: do not mount on top of idmapped mounts
ecryptfs: do not mount on top of idmapped mounts
ima: handle idmapped mounts
apparmor: handle idmapped mounts
fs: make helpers idmap mount aware
exec: handle idmapped mounts
would_dump: handle idmapped mounts
...
2021-02-23 13:39:45 -08:00
func , mask , func_data ) )
ima: integrity appraisal extension
IMA currently maintains an integrity measurement list used to assert the
integrity of the running system to a third party. The IMA-appraisal
extension adds local integrity validation and enforcement of the
measurement against a "good" value stored as an extended attribute
'security.ima'. The initial methods for validating 'security.ima' are
hashed based, which provides file data integrity, and digital signature
based, which in addition to providing file data integrity, provides
authenticity.
This patch creates and maintains the 'security.ima' xattr, containing
the file data hash measurement. Protection of the xattr is provided by
EVM, if enabled and configured.
Based on policy, IMA calls evm_verifyxattr() to verify a file's metadata
integrity and, assuming success, compares the file's current hash value
with the one stored as an extended attribute in 'security.ima'.
Changelov v4:
- changed iint cache flags to hex values
Changelog v3:
- change appraisal default for filesystems without xattr support to fail
Changelog v2:
- fix audit msg 'res' value
- removed unused 'ima_appraise=' values
Changelog v1:
- removed unused iint mutex (Dmitry Kasatkin)
- setattr hook must not reset appraised (Dmitry Kasatkin)
- evm_verifyxattr() now differentiates between no 'security.evm' xattr
(INTEGRITY_NOLABEL) and no EVM 'protected' xattrs included in the
'security.evm' (INTEGRITY_NOXATTRS).
- replace hash_status with ima_status (Dmitry Kasatkin)
- re-initialize slab element ima_status on free (Dmitry Kasatkin)
- include 'security.ima' in EVM if CONFIG_IMA_APPRAISE, not CONFIG_IMA
- merged half "ima: ima_must_appraise_or_measure API change" (Dmitry Kasatkin)
- removed unnecessary error variable in process_measurement() (Dmitry Kasatkin)
- use ima_inode_post_setattr() stub function, if IMA_APPRAISE not configured
(moved ima_inode_post_setattr() to ima_appraise.c)
- make sure ima_collect_measurement() can read file
Changelog:
- add 'iint' to evm_verifyxattr() call (Dimitry Kasatkin)
- fix the race condition between chmod, which takes the i_mutex and then
iint->mutex, and ima_file_free() and process_measurement(), which take
the locks in the reverse order, by eliminating iint->mutex. (Dmitry Kasatkin)
- cleanup of ima_appraise_measurement() (Dmitry Kasatkin)
- changes as a result of the iint not allocated for all regular files, but
only for those measured/appraised.
- don't try to appraise new/empty files
- expanded ima_appraisal description in ima/Kconfig
- IMA appraise definitions required even if IMA_APPRAISE not enabled
- add return value to ima_must_appraise() stub
- unconditionally set status = INTEGRITY_PASS *after* testing status,
not before. (Found by Joe Perches)
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2012-02-13 10:15:05 -05:00
continue ;
2009-02-04 09:06:58 -05:00
ima: rename IMA_ACTION_FLAGS to IMA_NONACTION_FLAGS
Simple policy rule options, such as fowner, uid, or euid, can be checked
immediately, while other policy rule options, such as requiring a file
signature, need to be deferred.
The 'flags' field in the integrity_iint_cache struct contains the policy
action', 'subaction', and non action/subaction.
action: measure/measured, appraise/appraised, (collect)/collected,
audit/audited
subaction: appraise status for each hook (e.g. file, mmap, bprm, read,
creds)
non action/subaction: deferred policy rule options and state
Rename the IMA_ACTION_FLAGS to IMA_NONACTION_FLAGS.
Reviewed-by: Stefan Berger <stefanb@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2021-12-28 09:53:14 -05:00
action | = entry - > flags & IMA_NONACTION_FLAGS ;
2012-06-08 13:58:49 +03:00
2012-09-12 20:51:32 +03:00
action | = entry - > action & IMA_DO_MASK ;
ima: support new "hash" and "dont_hash" policy actions
The builtin ima_appraise_tcb policy, which is specified on the boot
command line, can be replaced with a custom policy, normally early in
the boot process. Custom policies can be more restrictive in some ways,
like requiring file signatures, but can be less restrictive in other
ways, like not appraising mutable files. With a less restrictive policy
in place, files in the builtin policy might not be hashed and labeled
with a security.ima hash. On reboot, files which should be labeled in
the ima_appraise_tcb are not labeled, possibly preventing the system
from booting properly.
To resolve this problem, this patch extends the existing IMA policy
actions "measure", "dont_measure", "appraise", "dont_appraise", and
"audit" with "hash" and "dont_hash". The new "hash" action will write
the file hash as security.ima, but without requiring the file to be
appraised as well.
For example, the builtin ima_appraise_tcb policy includes the rule,
"appraise fowner=0". Adding the "hash fowner=0" rule to a custom
policy, will cause the needed file hashes to be calculated and written
as security.ima xattrs.
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
2016-09-29 10:04:52 -04:00
if ( entry - > action & IMA_APPRAISE ) {
2012-12-05 15:14:38 -05:00
action | = get_subaction ( entry , func ) ;
2018-03-10 23:07:34 -05:00
action & = ~ IMA_HASH ;
2018-02-21 11:36:32 -05:00
if ( ima_fail_unverifiable_sigs )
action | = IMA_FAIL_UNVERIFIABLE_SIGS ;
2012-12-03 17:08:11 -05:00
2021-08-16 08:11:00 +00:00
if ( allowed_algos & &
entry - > flags & IMA_VALIDATE_ALGOS )
* allowed_algos = entry - > allowed_algos ;
}
2019-07-19 07:16:57 -04:00
2012-09-12 20:51:32 +03:00
if ( entry - > action & IMA_DO_MASK )
actmask & = ~ ( entry - > action | entry - > action < < 1 ) ;
else
actmask & = ~ ( entry - > action | entry - > action > > 1 ) ;
2009-02-04 09:06:58 -05:00
2016-06-01 13:14:02 -05:00
if ( ( pcr ) & & ( entry - > flags & IMA_PCR ) )
* pcr = entry - > pcr ;
2019-06-19 15:46:11 -07:00
if ( template_desc & & entry - > template )
* template_desc = entry - > template ;
ima: integrity appraisal extension
IMA currently maintains an integrity measurement list used to assert the
integrity of the running system to a third party. The IMA-appraisal
extension adds local integrity validation and enforcement of the
measurement against a "good" value stored as an extended attribute
'security.ima'. The initial methods for validating 'security.ima' are
hashed based, which provides file data integrity, and digital signature
based, which in addition to providing file data integrity, provides
authenticity.
This patch creates and maintains the 'security.ima' xattr, containing
the file data hash measurement. Protection of the xattr is provided by
EVM, if enabled and configured.
Based on policy, IMA calls evm_verifyxattr() to verify a file's metadata
integrity and, assuming success, compares the file's current hash value
with the one stored as an extended attribute in 'security.ima'.
Changelov v4:
- changed iint cache flags to hex values
Changelog v3:
- change appraisal default for filesystems without xattr support to fail
Changelog v2:
- fix audit msg 'res' value
- removed unused 'ima_appraise=' values
Changelog v1:
- removed unused iint mutex (Dmitry Kasatkin)
- setattr hook must not reset appraised (Dmitry Kasatkin)
- evm_verifyxattr() now differentiates between no 'security.evm' xattr
(INTEGRITY_NOLABEL) and no EVM 'protected' xattrs included in the
'security.evm' (INTEGRITY_NOXATTRS).
- replace hash_status with ima_status (Dmitry Kasatkin)
- re-initialize slab element ima_status on free (Dmitry Kasatkin)
- include 'security.ima' in EVM if CONFIG_IMA_APPRAISE, not CONFIG_IMA
- merged half "ima: ima_must_appraise_or_measure API change" (Dmitry Kasatkin)
- removed unnecessary error variable in process_measurement() (Dmitry Kasatkin)
- use ima_inode_post_setattr() stub function, if IMA_APPRAISE not configured
(moved ima_inode_post_setattr() to ima_appraise.c)
- make sure ima_collect_measurement() can read file
Changelog:
- add 'iint' to evm_verifyxattr() call (Dimitry Kasatkin)
- fix the race condition between chmod, which takes the i_mutex and then
iint->mutex, and ima_file_free() and process_measurement(), which take
the locks in the reverse order, by eliminating iint->mutex. (Dmitry Kasatkin)
- cleanup of ima_appraise_measurement() (Dmitry Kasatkin)
- changes as a result of the iint not allocated for all regular files, but
only for those measured/appraised.
- don't try to appraise new/empty files
- expanded ima_appraisal description in ima/Kconfig
- IMA appraise definitions required even if IMA_APPRAISE not enabled
- add return value to ima_must_appraise() stub
- unconditionally set status = INTEGRITY_PASS *after* testing status,
not before. (Found by Joe Perches)
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2012-02-13 10:15:05 -05:00
if ( ! actmask )
break ;
2009-02-04 09:06:58 -05:00
}
2015-12-02 17:47:54 +02:00
rcu_read_unlock ( ) ;
ima: integrity appraisal extension
IMA currently maintains an integrity measurement list used to assert the
integrity of the running system to a third party. The IMA-appraisal
extension adds local integrity validation and enforcement of the
measurement against a "good" value stored as an extended attribute
'security.ima'. The initial methods for validating 'security.ima' are
hashed based, which provides file data integrity, and digital signature
based, which in addition to providing file data integrity, provides
authenticity.
This patch creates and maintains the 'security.ima' xattr, containing
the file data hash measurement. Protection of the xattr is provided by
EVM, if enabled and configured.
Based on policy, IMA calls evm_verifyxattr() to verify a file's metadata
integrity and, assuming success, compares the file's current hash value
with the one stored as an extended attribute in 'security.ima'.
Changelov v4:
- changed iint cache flags to hex values
Changelog v3:
- change appraisal default for filesystems without xattr support to fail
Changelog v2:
- fix audit msg 'res' value
- removed unused 'ima_appraise=' values
Changelog v1:
- removed unused iint mutex (Dmitry Kasatkin)
- setattr hook must not reset appraised (Dmitry Kasatkin)
- evm_verifyxattr() now differentiates between no 'security.evm' xattr
(INTEGRITY_NOLABEL) and no EVM 'protected' xattrs included in the
'security.evm' (INTEGRITY_NOXATTRS).
- replace hash_status with ima_status (Dmitry Kasatkin)
- re-initialize slab element ima_status on free (Dmitry Kasatkin)
- include 'security.ima' in EVM if CONFIG_IMA_APPRAISE, not CONFIG_IMA
- merged half "ima: ima_must_appraise_or_measure API change" (Dmitry Kasatkin)
- removed unnecessary error variable in process_measurement() (Dmitry Kasatkin)
- use ima_inode_post_setattr() stub function, if IMA_APPRAISE not configured
(moved ima_inode_post_setattr() to ima_appraise.c)
- make sure ima_collect_measurement() can read file
Changelog:
- add 'iint' to evm_verifyxattr() call (Dimitry Kasatkin)
- fix the race condition between chmod, which takes the i_mutex and then
iint->mutex, and ima_file_free() and process_measurement(), which take
the locks in the reverse order, by eliminating iint->mutex. (Dmitry Kasatkin)
- cleanup of ima_appraise_measurement() (Dmitry Kasatkin)
- changes as a result of the iint not allocated for all regular files, but
only for those measured/appraised.
- don't try to appraise new/empty files
- expanded ima_appraisal description in ima/Kconfig
- IMA appraise definitions required even if IMA_APPRAISE not enabled
- add return value to ima_must_appraise() stub
- unconditionally set status = INTEGRITY_PASS *after* testing status,
not before. (Found by Joe Perches)
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2012-02-13 10:15:05 -05:00
return action ;
2009-02-04 09:06:58 -05:00
}
2021-08-16 08:11:01 +00:00
/**
* ima_update_policy_flags ( ) - Update global IMA variables
*
* Update ima_policy_flag and ima_setxattr_allowed_hash_algorithms
* based on the currently loaded policy .
*
* With ima_policy_flag , the decision to short circuit out of a function
* or not call the function in the first place can be made earlier .
*
* With ima_setxattr_allowed_hash_algorithms , the policy can restrict the
* set of hash algorithms accepted when updating the security . ima xattr of
* a file .
*
* Context : called after a policy update and at system initialization .
2014-09-12 19:35:54 +02:00
*/
2021-08-16 08:11:01 +00:00
void ima_update_policy_flags ( void )
2014-09-12 19:35:54 +02:00
{
struct ima_rule_entry * entry ;
2021-08-16 08:11:01 +00:00
int new_policy_flag = 0 ;
2021-10-09 18:38:21 +08:00
struct list_head * ima_rules_tmp ;
2014-09-12 19:35:54 +02:00
2021-08-16 08:11:01 +00:00
rcu_read_lock ( ) ;
2021-10-09 18:38:21 +08:00
ima_rules_tmp = rcu_dereference ( ima_rules ) ;
list_for_each_entry_rcu ( entry , ima_rules_tmp , list ) {
2021-08-16 08:11:01 +00:00
/*
* SETXATTR_CHECK rules do not implement a full policy check
* because rule checking would probably have an important
* performance impact on setxattr ( ) . As a consequence , only one
* SETXATTR_CHECK can be active at a given time .
* Because we want to preserve that property , we set out to use
* atomic_cmpxchg . Either :
* - the atomic was non - zero : a setxattr hash policy is
* already enforced , we do nothing
* - the atomic was zero : no setxattr policy was set , enable
* the setxattr hash policy
*/
if ( entry - > func = = SETXATTR_CHECK ) {
atomic_cmpxchg ( & ima_setxattr_allowed_hash_algorithms ,
0 , entry - > allowed_algos ) ;
/* SETXATTR_CHECK doesn't impact ima_policy_flag */
continue ;
}
2014-09-12 19:35:54 +02:00
if ( entry - > action & IMA_DO_MASK )
2021-08-16 08:11:01 +00:00
new_policy_flag | = entry - > action ;
2014-09-12 19:35:54 +02:00
}
2021-08-16 08:11:01 +00:00
rcu_read_unlock ( ) ;
2014-09-12 19:35:54 +02:00
2018-07-13 14:06:01 -04:00
ima_appraise | = ( build_ima_appraise | temp_ima_appraise ) ;
2014-09-12 19:35:54 +02:00
if ( ! ima_appraise )
2021-08-16 08:11:01 +00:00
new_policy_flag & = ~ IMA_APPRAISE ;
ima_policy_flag = new_policy_flag ;
2014-09-12 19:35:54 +02:00
}
2018-04-12 00:15:22 -04:00
static int ima_appraise_flag ( enum ima_hooks func )
{
if ( func = = MODULE_CHECK )
return IMA_APPRAISE_MODULES ;
else if ( func = = FIRMWARE_CHECK )
return IMA_APPRAISE_FIRMWARE ;
else if ( func = = POLICY_CHECK )
return IMA_APPRAISE_POLICY ;
2018-07-13 14:05:58 -04:00
else if ( func = = KEXEC_KERNEL_CHECK )
return IMA_APPRAISE_KEXEC ;
2018-04-12 00:15:22 -04:00
return 0 ;
}
2018-10-09 23:00:35 +05:30
static void add_rules ( struct ima_rule_entry * entries , int count ,
enum policy_rule_list policy_rule )
{
int i = 0 ;
for ( i = 0 ; i < count ; i + + ) {
struct ima_rule_entry * entry ;
if ( policy_rule & IMA_DEFAULT_POLICY )
list_add_tail ( & entries [ i ] . list , & ima_default_rules ) ;
if ( policy_rule & IMA_CUSTOM_POLICY ) {
entry = kmemdup ( & entries [ i ] , sizeof ( * entry ) ,
GFP_KERNEL ) ;
if ( ! entry )
continue ;
list_add_tail ( & entry - > list , & ima_policy_rules ) ;
}
2020-04-27 12:28:59 +02:00
if ( entries [ i ] . action = = APPRAISE ) {
if ( entries ! = build_appraise_rules )
temp_ima_appraise | =
ima_appraise_flag ( entries [ i ] . func ) ;
else
build_ima_appraise | =
ima_appraise_flag ( entries [ i ] . func ) ;
}
2018-10-09 23:00:35 +05:30
}
}
2018-10-09 23:00:36 +05:30
static int ima_parse_rule ( char * rule , struct ima_rule_entry * entry ) ;
static int __init ima_init_arch_policy ( void )
{
const char * const * arch_rules ;
const char * const * rules ;
int arch_entries = 0 ;
int i = 0 ;
arch_rules = arch_get_ima_policy ( ) ;
if ( ! arch_rules )
return arch_entries ;
/* Get number of rules */
for ( rules = arch_rules ; * rules ! = NULL ; rules + + )
arch_entries + + ;
arch_policy_entry = kcalloc ( arch_entries + 1 ,
sizeof ( * arch_policy_entry ) , GFP_KERNEL ) ;
if ( ! arch_policy_entry )
return 0 ;
/* Convert each policy string rules to struct ima_rule_entry format */
for ( rules = arch_rules , i = 0 ; * rules ! = NULL ; rules + + ) {
char rule [ 255 ] ;
int result ;
2021-10-08 11:14:30 +02:00
result = strscpy ( rule , * rules , sizeof ( rule ) ) ;
2018-10-09 23:00:36 +05:30
INIT_LIST_HEAD ( & arch_policy_entry [ i ] . list ) ;
result = ima_parse_rule ( rule , & arch_policy_entry [ i ] ) ;
if ( result ) {
pr_warn ( " Skipping unknown architecture policy rule: %s \n " ,
rule ) ;
memset ( & arch_policy_entry [ i ] , 0 ,
sizeof ( * arch_policy_entry ) ) ;
continue ;
}
i + + ;
}
return i ;
}
2009-02-04 09:06:58 -05:00
/**
* ima_init_policy - initialize the default measure rules .
*
2021-10-08 11:14:29 +02:00
* ima_rules points to either the ima_default_rules or the new ima_policy_rules .
2009-02-04 09:06:58 -05:00
*/
2009-05-21 15:43:32 -04:00
void __init ima_init_policy ( void )
2009-02-04 09:06:58 -05:00
{
2018-10-09 23:00:36 +05:30
int build_appraise_entries , arch_entries ;
2014-03-04 18:04:20 +02:00
2018-10-09 23:00:35 +05:30
/* if !ima_policy, we load NO default rules */
if ( ima_policy )
add_rules ( dont_measure_rules , ARRAY_SIZE ( dont_measure_rules ) ,
IMA_DEFAULT_POLICY ) ;
2015-06-11 20:48:33 -04:00
switch ( ima_policy ) {
case ORIGINAL_TCB :
2018-10-09 23:00:35 +05:30
add_rules ( original_measurement_rules ,
ARRAY_SIZE ( original_measurement_rules ) ,
IMA_DEFAULT_POLICY ) ;
2015-06-11 20:48:33 -04:00
break ;
case DEFAULT_TCB :
2018-10-09 23:00:35 +05:30
add_rules ( default_measurement_rules ,
ARRAY_SIZE ( default_measurement_rules ) ,
IMA_DEFAULT_POLICY ) ;
2020-11-20 12:25:46 -06:00
break ;
2015-06-11 20:48:33 -04:00
default :
break ;
}
2015-04-07 12:22:11 +03:00
2018-10-09 23:00:36 +05:30
/*
* Based on runtime secure boot flags , insert arch specific measurement
* and appraise rules requiring file signatures for both the initial
* and custom policies , prior to other appraise rules .
* ( Highest priority )
*/
arch_entries = ima_init_arch_policy ( ) ;
if ( ! arch_entries )
pr_info ( " No architecture policies found \n " ) ;
else
add_rules ( arch_policy_entry , arch_entries ,
IMA_DEFAULT_POLICY | IMA_CUSTOM_POLICY ) ;
2017-04-21 18:58:27 -04:00
/*
2018-07-13 14:06:01 -04:00
* Insert the builtin " secure_boot " policy rules requiring file
2018-10-09 23:00:36 +05:30
* signatures , prior to other appraise rules .
2017-04-21 18:58:27 -04:00
*/
2018-10-09 23:00:35 +05:30
if ( ima_use_secure_boot )
add_rules ( secure_boot_rules , ARRAY_SIZE ( secure_boot_rules ) ,
IMA_DEFAULT_POLICY ) ;
2017-04-21 18:58:27 -04:00
2018-07-13 14:06:01 -04:00
/*
* Insert the build time appraise rules requiring file signatures
* for both the initial and custom policies , prior to other appraise
2018-10-09 23:00:35 +05:30
* rules . As the secure boot rules includes all of the build time
* rules , include either one or the other set of rules , but not both .
2018-07-13 14:06:01 -04:00
*/
2018-10-09 23:00:35 +05:30
build_appraise_entries = ARRAY_SIZE ( build_appraise_rules ) ;
if ( build_appraise_entries ) {
if ( ima_use_secure_boot )
add_rules ( build_appraise_rules , build_appraise_entries ,
IMA_CUSTOM_POLICY ) ;
else
add_rules ( build_appraise_rules , build_appraise_entries ,
IMA_DEFAULT_POLICY | IMA_CUSTOM_POLICY ) ;
2018-07-13 14:06:01 -04:00
}
2018-10-09 23:00:35 +05:30
if ( ima_use_appraise_tcb )
add_rules ( default_appraise_rules ,
ARRAY_SIZE ( default_appraise_rules ) ,
IMA_DEFAULT_POLICY ) ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
2021-01-07 20:07:07 -08:00
if ( ima_use_critical_data )
add_rules ( critical_data_rules ,
ARRAY_SIZE ( critical_data_rules ) ,
IMA_DEFAULT_POLICY ) ;
2021-08-16 08:11:01 +00:00
atomic_set ( & ima_setxattr_allowed_hash_algorithms , 0 ) ;
ima_update_policy_flags ( ) ;
2009-02-04 09:06:58 -05:00
}
2009-02-04 09:07:00 -05:00
2015-12-22 08:51:23 -05:00
/* Make sure we have a valid policy, at least containing some rules. */
2016-01-20 11:13:46 +00:00
int ima_check_policy ( void )
2015-12-22 08:51:23 -05:00
{
if ( list_empty ( & ima_temp_rules ) )
return - EINVAL ;
return 0 ;
}
2009-02-04 09:07:00 -05:00
/**
* ima_update_policy - update default_rules with new measure rules
*
* Called on file . release to update the default rules with a complete new
2015-12-02 17:47:54 +02:00
* policy . What we do here is to splice ima_policy_rules and ima_temp_rules so
* they make a queue . The policy may be updated multiple times and this is the
* RCU updater .
*
* Policy rules are never deleted so ima_policy_flag gets zeroed only once when
* we switch from the default policy to user defined .
2009-02-04 09:07:00 -05:00
*/
void ima_update_policy ( void )
{
2018-05-22 17:06:55 +03:00
struct list_head * policy = & ima_policy_rules ;
2015-12-02 17:47:54 +02:00
2018-05-22 17:06:55 +03:00
list_splice_tail_init_rcu ( & ima_temp_rules , policy , synchronize_rcu ) ;
2015-12-02 17:47:54 +02:00
2021-10-09 18:38:21 +08:00
if ( ima_rules ! = ( struct list_head __rcu * ) policy ) {
2015-12-02 17:47:54 +02:00
ima_policy_flag = 0 ;
2018-10-09 23:00:36 +05:30
2021-10-09 18:38:21 +08:00
rcu_assign_pointer ( ima_rules , policy ) ;
2018-10-09 23:00:36 +05:30
/*
* IMA architecture specific policy rules are specified
* as strings and converted to an array of ima_entry_rules
* on boot . After loading a custom policy , free the
* architecture specific rules stored as an array .
*/
kfree ( arch_policy_entry ) ;
2015-12-02 17:47:54 +02:00
}
2021-08-16 08:11:01 +00:00
ima_update_policy_flags ( ) ;
2020-01-22 17:32:05 -08:00
/* Custom IMA policy has been loaded */
ima_process_queued_keys ( ) ;
2009-02-04 09:07:00 -05:00
}
2018-12-17 19:14:49 -05:00
/* Keep the enumeration in sync with the policy_tokens! */
2021-10-07 14:03:02 -06:00
enum policy_opt {
2018-12-17 19:14:49 -05:00
Opt_measure , Opt_dont_measure ,
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
Opt_appraise , Opt_dont_appraise ,
ima: support new "hash" and "dont_hash" policy actions
The builtin ima_appraise_tcb policy, which is specified on the boot
command line, can be replaced with a custom policy, normally early in
the boot process. Custom policies can be more restrictive in some ways,
like requiring file signatures, but can be less restrictive in other
ways, like not appraising mutable files. With a less restrictive policy
in place, files in the builtin policy might not be hashed and labeled
with a security.ima hash. On reboot, files which should be labeled in
the ima_appraise_tcb are not labeled, possibly preventing the system
from booting properly.
To resolve this problem, this patch extends the existing IMA policy
actions "measure", "dont_measure", "appraise", "dont_appraise", and
"audit" with "hash" and "dont_hash". The new "hash" action will write
the file hash as security.ima, but without requiring the file to be
appraised as well.
For example, the builtin ima_appraise_tcb policy includes the rule,
"appraise fowner=0". Adding the "hash fowner=0" rule to a custom
policy, will cause the needed file hashes to be calculated and written
as security.ima xattrs.
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
2016-09-29 10:04:52 -04:00
Opt_audit , Opt_hash , Opt_dont_hash ,
2009-02-04 09:07:00 -05:00
Opt_obj_user , Opt_obj_role , Opt_obj_type ,
Opt_subj_user , Opt_subj_role , Opt_subj_type ,
2021-10-07 14:03:02 -06:00
Opt_func , Opt_mask , Opt_fsmagic , Opt_fsname , Opt_fsuuid ,
Opt_uid_eq , Opt_euid_eq , Opt_gid_eq , Opt_egid_eq ,
Opt_fowner_eq , Opt_fgroup_eq ,
Opt_uid_gt , Opt_euid_gt , Opt_gid_gt , Opt_egid_gt ,
Opt_fowner_gt , Opt_fgroup_gt ,
Opt_uid_lt , Opt_euid_lt , Opt_gid_lt , Opt_egid_lt ,
Opt_fowner_lt , Opt_fgroup_lt ,
2021-08-16 08:11:00 +00:00
Opt_appraise_type , Opt_appraise_flag , Opt_appraise_algos ,
2019-12-11 08:47:07 -08:00
Opt_permit_directio , Opt_pcr , Opt_template , Opt_keyrings ,
2021-01-07 20:07:05 -08:00
Opt_label , Opt_err
2009-02-04 09:07:00 -05:00
} ;
2018-12-17 19:14:49 -05:00
static const match_table_t policy_tokens = {
2009-02-04 09:07:00 -05:00
{ Opt_measure , " measure " } ,
{ Opt_dont_measure , " dont_measure " } ,
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
{ Opt_appraise , " appraise " } ,
{ Opt_dont_appraise , " dont_appraise " } ,
2012-06-14 10:04:36 -07:00
{ Opt_audit , " audit " } ,
ima: support new "hash" and "dont_hash" policy actions
The builtin ima_appraise_tcb policy, which is specified on the boot
command line, can be replaced with a custom policy, normally early in
the boot process. Custom policies can be more restrictive in some ways,
like requiring file signatures, but can be less restrictive in other
ways, like not appraising mutable files. With a less restrictive policy
in place, files in the builtin policy might not be hashed and labeled
with a security.ima hash. On reboot, files which should be labeled in
the ima_appraise_tcb are not labeled, possibly preventing the system
from booting properly.
To resolve this problem, this patch extends the existing IMA policy
actions "measure", "dont_measure", "appraise", "dont_appraise", and
"audit" with "hash" and "dont_hash". The new "hash" action will write
the file hash as security.ima, but without requiring the file to be
appraised as well.
For example, the builtin ima_appraise_tcb policy includes the rule,
"appraise fowner=0". Adding the "hash fowner=0" rule to a custom
policy, will cause the needed file hashes to be calculated and written
as security.ima xattrs.
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
2016-09-29 10:04:52 -04:00
{ Opt_hash , " hash " } ,
{ Opt_dont_hash , " dont_hash " } ,
2009-02-04 09:07:00 -05:00
{ Opt_obj_user , " obj_user=%s " } ,
{ Opt_obj_role , " obj_role=%s " } ,
{ Opt_obj_type , " obj_type=%s " } ,
{ Opt_subj_user , " subj_user=%s " } ,
{ Opt_subj_role , " subj_role=%s " } ,
{ Opt_subj_type , " subj_type=%s " } ,
{ Opt_func , " func=%s " } ,
{ Opt_mask , " mask=%s " } ,
{ Opt_fsmagic , " fsmagic=%s " } ,
2018-01-15 11:20:36 -05:00
{ Opt_fsname , " fsname=%s " } ,
2012-09-03 23:23:13 +03:00
{ Opt_fsuuid , " fsuuid=%s " } ,
2017-01-27 19:23:01 +03:00
{ Opt_uid_eq , " uid=%s " } ,
{ Opt_euid_eq , " euid=%s " } ,
2021-10-07 14:03:02 -06:00
{ Opt_gid_eq , " gid=%s " } ,
{ Opt_egid_eq , " egid=%s " } ,
2017-01-27 19:23:01 +03:00
{ Opt_fowner_eq , " fowner=%s " } ,
2021-10-07 14:03:02 -06:00
{ Opt_fgroup_eq , " fgroup=%s " } ,
2017-01-27 19:23:01 +03:00
{ Opt_uid_gt , " uid>%s " } ,
{ Opt_euid_gt , " euid>%s " } ,
2021-10-07 14:03:02 -06:00
{ Opt_gid_gt , " gid>%s " } ,
{ Opt_egid_gt , " egid>%s " } ,
2017-01-27 19:23:01 +03:00
{ Opt_fowner_gt , " fowner>%s " } ,
2021-10-07 14:03:02 -06:00
{ Opt_fgroup_gt , " fgroup>%s " } ,
2017-01-27 19:23:01 +03:00
{ Opt_uid_lt , " uid<%s " } ,
{ Opt_euid_lt , " euid<%s " } ,
2021-10-07 14:03:02 -06:00
{ Opt_gid_lt , " gid<%s " } ,
{ Opt_egid_lt , " egid<%s " } ,
2017-01-27 19:23:01 +03:00
{ Opt_fowner_lt , " fowner<%s " } ,
2021-10-07 14:03:02 -06:00
{ Opt_fgroup_lt , " fgroup<%s " } ,
2012-06-08 13:58:49 +03:00
{ Opt_appraise_type , " appraise_type=%s " } ,
2019-10-30 23:31:32 -04:00
{ Opt_appraise_flag , " appraise_flag=%s " } ,
2021-08-16 08:11:00 +00:00
{ Opt_appraise_algos , " appraise_algos=%s " } ,
ima: audit log files opened with O_DIRECT flag
Files are measured or appraised based on the IMA policy. When a
file, in policy, is opened with the O_DIRECT flag, a deadlock
occurs.
The first attempt at resolving this lockdep temporarily removed the
O_DIRECT flag and restored it, after calculating the hash. The
second attempt introduced the O_DIRECT_HAVELOCK flag. Based on this
flag, do_blockdev_direct_IO() would skip taking the i_mutex a second
time. The third attempt, by Dmitry Kasatkin, resolves the i_mutex
locking issue, by re-introducing the IMA mutex, but uncovered
another problem. Reading a file with O_DIRECT flag set, writes
directly to userspace pages. A second patch allocates a user-space
like memory. This works for all IMA hooks, except ima_file_free(),
which is called on __fput() to recalculate the file hash.
Until this last issue is addressed, do not 'collect' the
measurement for measuring, appraising, or auditing files opened
with the O_DIRECT flag set. Based on policy, permit or deny file
access. This patch defines a new IMA policy rule option named
'permit_directio'. Policy rules could be defined, based on LSM
or other criteria, to permit specific applications to open files
with the O_DIRECT flag set.
Changelog v1:
- permit or deny file access based IMA policy rules
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Acked-by: Dmitry Kasatkin <d.kasatkin@samsung.com>
Cc: <stable@vger.kernel.org>
2014-05-12 09:28:11 -04:00
{ Opt_permit_directio , " permit_directio " } ,
2016-06-01 13:14:01 -05:00
{ Opt_pcr , " pcr=%s " } ,
2019-06-19 15:46:11 -07:00
{ Opt_template , " template=%s " } ,
2019-12-11 08:47:07 -08:00
{ Opt_keyrings , " keyrings=%s " } ,
2021-01-07 20:07:05 -08:00
{ Opt_label , " label=%s " } ,
2009-02-04 09:07:00 -05:00
{ Opt_err , NULL }
} ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
static int ima_lsm_rule_init ( struct ima_rule_entry * entry ,
2013-01-03 14:19:09 -05:00
substring_t * args , int lsm_rule , int audit_type )
2009-02-04 09:07:00 -05:00
{
int result ;
2010-04-20 10:21:01 -04:00
if ( entry - > lsm [ lsm_rule ] . rule )
return - EINVAL ;
2013-01-03 14:19:09 -05:00
entry - > lsm [ lsm_rule ] . args_p = match_strdup ( args ) ;
if ( ! entry - > lsm [ lsm_rule ] . args_p )
return - ENOMEM ;
2009-02-04 09:07:00 -05:00
entry - > lsm [ lsm_rule ] . type = audit_type ;
2020-07-10 15:37:50 -05:00
result = ima_filter_rule_init ( entry - > lsm [ lsm_rule ] . type , Audit_equal ,
entry - > lsm [ lsm_rule ] . args_p ,
& entry - > lsm [ lsm_rule ] . rule ) ;
2013-01-03 14:19:09 -05:00
if ( ! entry - > lsm [ lsm_rule ] . rule ) {
2020-01-15 17:42:30 +02:00
pr_warn ( " rule for LSM \' %s \' is undefined \n " ,
2020-07-09 01:19:08 -05:00
entry - > lsm [ lsm_rule ] . args_p ) ;
2020-01-15 17:42:30 +02:00
2021-10-09 18:38:21 +08:00
if ( ima_rules = = ( struct list_head __rcu * ) ( & ima_default_rules ) ) {
2020-01-15 17:42:30 +02:00
kfree ( entry - > lsm [ lsm_rule ] . args_p ) ;
2020-07-09 01:19:02 -05:00
entry - > lsm [ lsm_rule ] . args_p = NULL ;
2020-01-15 17:42:30 +02:00
result = - EINVAL ;
} else
result = 0 ;
2013-01-03 14:19:09 -05:00
}
2009-02-04 09:07:00 -05:00
return result ;
}
2017-01-27 19:23:01 +03:00
static void ima_log_string_op ( struct audit_buffer * ab , char * key , char * value ,
2021-10-07 14:03:02 -06:00
enum policy_opt rule_operator )
2010-04-20 10:21:30 -04:00
{
2018-06-04 16:54:54 -04:00
if ( ! ab )
return ;
2021-10-07 14:03:02 -06:00
switch ( rule_operator ) {
case Opt_uid_gt :
case Opt_euid_gt :
case Opt_gid_gt :
case Opt_egid_gt :
case Opt_fowner_gt :
case Opt_fgroup_gt :
2017-01-27 19:23:01 +03:00
audit_log_format ( ab , " %s> " , key ) ;
2021-10-07 14:03:02 -06:00
break ;
case Opt_uid_lt :
case Opt_euid_lt :
case Opt_gid_lt :
case Opt_egid_lt :
case Opt_fowner_lt :
case Opt_fgroup_lt :
2017-01-27 19:23:01 +03:00
audit_log_format ( ab , " %s< " , key ) ;
2021-10-07 14:03:02 -06:00
break ;
default :
2017-01-27 19:23:01 +03:00
audit_log_format ( ab , " %s= " , key ) ;
2021-10-07 14:03:02 -06:00
}
2018-06-04 16:54:53 -04:00
audit_log_format ( ab , " %s " , value ) ;
2010-04-20 10:21:30 -04:00
}
2017-01-27 19:23:01 +03:00
static void ima_log_string ( struct audit_buffer * ab , char * key , char * value )
{
2021-10-07 14:03:02 -06:00
ima_log_string_op ( ab , key , value , Opt_err ) ;
2017-01-27 19:23:01 +03:00
}
2010-04-20 10:21:30 -04:00
2019-06-27 23:19:32 -03:00
/*
* Validating the appended signature included in the measurement list requires
* the file hash calculated without the appended signature ( i . e . , the ' d - modsig '
* field ) . Therefore , notify the user if they have the ' modsig ' field but not
* the ' d - modsig ' field in the template .
*/
static void check_template_modsig ( const struct ima_template_desc * template )
{
# define MSG "template with 'modsig' field also needs 'd-modsig' field\n"
bool has_modsig , has_dmodsig ;
static bool checked ;
int i ;
/* We only need to notify the user once. */
if ( checked )
return ;
has_modsig = has_dmodsig = false ;
for ( i = 0 ; i < template - > num_fields ; i + + ) {
if ( ! strcmp ( template - > fields [ i ] - > field_id , " modsig " ) )
has_modsig = true ;
else if ( ! strcmp ( template - > fields [ i ] - > field_id , " d-modsig " ) )
has_dmodsig = true ;
}
if ( has_modsig & & ! has_dmodsig )
pr_notice ( MSG ) ;
checked = true ;
# undef MSG
}
2020-07-09 01:19:03 -05:00
static bool ima_validate_rule ( struct ima_rule_entry * entry )
{
2020-07-09 01:19:09 -05:00
/* Ensure that the action is set and is compatible with the flags */
2020-07-09 01:19:03 -05:00
if ( entry - > action = = UNKNOWN )
return false ;
2020-07-09 01:19:09 -05:00
if ( entry - > action ! = MEASURE & & entry - > flags & IMA_PCR )
return false ;
if ( entry - > action ! = APPRAISE & &
2021-08-16 08:11:00 +00:00
entry - > flags & ( IMA_DIGSIG_REQUIRED | IMA_MODSIG_ALLOWED |
IMA_CHECK_BLACKLIST | IMA_VALIDATE_ALGOS ) )
2020-07-09 01:19:09 -05:00
return false ;
/*
* The IMA_FUNC bit must be set if and only if there ' s a valid hook
* function specified , and vice versa . Enforcing this property allows
* for the NONE case below to validate a rule without an explicit hook
* function .
*/
if ( ( ( entry - > flags & IMA_FUNC ) & & entry - > func = = NONE ) | |
( ! ( entry - > flags & IMA_FUNC ) & & entry - > func ! = NONE ) )
return false ;
2020-07-09 01:19:03 -05:00
/*
* Ensure that the hook function is compatible with the other
* components of the rule
*/
switch ( entry - > func ) {
case NONE :
case FILE_CHECK :
case MMAP_CHECK :
case BPRM_CHECK :
case CREDS_CHECK :
case POST_SETATTR :
case FIRMWARE_CHECK :
2020-07-09 01:19:09 -05:00
case POLICY_CHECK :
if ( entry - > flags & ~ ( IMA_FUNC | IMA_MASK | IMA_FSMAGIC |
IMA_UID | IMA_FOWNER | IMA_FSUUID |
IMA_INMASK | IMA_EUID | IMA_PCR |
2021-10-07 14:03:02 -06:00
IMA_FSNAME | IMA_GID | IMA_EGID |
IMA_FGROUP | IMA_DIGSIG_REQUIRED |
2021-08-16 08:11:00 +00:00
IMA_PERMIT_DIRECTIO | IMA_VALIDATE_ALGOS ) )
2020-07-09 01:19:09 -05:00
return false ;
break ;
case MODULE_CHECK :
2020-07-09 01:19:03 -05:00
case KEXEC_KERNEL_CHECK :
case KEXEC_INITRAMFS_CHECK :
2020-07-09 01:19:09 -05:00
if ( entry - > flags & ~ ( IMA_FUNC | IMA_MASK | IMA_FSMAGIC |
IMA_UID | IMA_FOWNER | IMA_FSUUID |
IMA_INMASK | IMA_EUID | IMA_PCR |
2021-10-07 14:03:02 -06:00
IMA_FSNAME | IMA_GID | IMA_EGID |
IMA_FGROUP | IMA_DIGSIG_REQUIRED |
2020-07-09 01:19:09 -05:00
IMA_PERMIT_DIRECTIO | IMA_MODSIG_ALLOWED |
2021-08-16 08:11:00 +00:00
IMA_CHECK_BLACKLIST | IMA_VALIDATE_ALGOS ) )
2020-07-09 01:19:09 -05:00
return false ;
2020-07-09 01:19:03 -05:00
break ;
case KEXEC_CMDLINE :
2020-07-09 01:19:04 -05:00
if ( entry - > action & ~ ( MEASURE | DONT_MEASURE ) )
return false ;
2020-07-09 01:19:11 -05:00
if ( entry - > flags & ~ ( IMA_FUNC | IMA_FSMAGIC | IMA_UID |
IMA_FOWNER | IMA_FSUUID | IMA_EUID |
2021-10-07 14:03:02 -06:00
IMA_PCR | IMA_FSNAME | IMA_GID | IMA_EGID |
IMA_FGROUP ) )
2020-07-09 01:19:04 -05:00
return false ;
break ;
2020-07-09 01:19:03 -05:00
case KEY_CHECK :
if ( entry - > action & ~ ( MEASURE | DONT_MEASURE ) )
return false ;
2021-10-07 14:03:02 -06:00
if ( entry - > flags & ~ ( IMA_FUNC | IMA_UID | IMA_GID | IMA_PCR |
2020-07-09 01:19:05 -05:00
IMA_KEYRINGS ) )
return false ;
if ( ima_rule_contains_lsm_cond ( entry ) )
return false ;
2021-01-07 20:07:04 -08:00
break ;
case CRITICAL_DATA :
if ( entry - > action & ~ ( MEASURE | DONT_MEASURE ) )
return false ;
2021-10-07 14:03:02 -06:00
if ( entry - > flags & ~ ( IMA_FUNC | IMA_UID | IMA_GID | IMA_PCR |
2021-01-07 20:07:05 -08:00
IMA_LABEL ) )
2021-01-07 20:07:04 -08:00
return false ;
if ( ima_rule_contains_lsm_cond ( entry ) )
return false ;
2021-08-16 08:11:01 +00:00
break ;
case SETXATTR_CHECK :
/* any action other than APPRAISE is unsupported */
if ( entry - > action ! = APPRAISE )
return false ;
/* SETXATTR_CHECK requires an appraise_algos parameter */
if ( ! ( entry - > flags & IMA_VALIDATE_ALGOS ) )
return false ;
/*
* full policies are not supported , they would have too
* much of a performance impact
*/
if ( entry - > flags & ~ ( IMA_FUNC | IMA_VALIDATE_ALGOS ) )
return false ;
2020-07-09 01:19:03 -05:00
break ;
default :
return false ;
}
2020-07-09 01:19:06 -05:00
/* Ensure that combinations of flags are compatible with each other */
if ( entry - > flags & IMA_CHECK_BLACKLIST & &
! ( entry - > flags & IMA_MODSIG_ALLOWED ) )
return false ;
2020-07-09 01:19:03 -05:00
return true ;
}
2021-08-16 08:11:00 +00:00
static unsigned int ima_parse_appraise_algos ( char * arg )
{
unsigned int res = 0 ;
int idx ;
char * token ;
while ( ( token = strsep ( & arg , " , " ) ) ! = NULL ) {
idx = match_string ( hash_algo_name , HASH_ALGO__LAST , token ) ;
if ( idx < 0 ) {
pr_err ( " unknown hash algorithm \" %s \" " ,
token ) ;
return 0 ;
}
2021-08-16 08:11:01 +00:00
if ( ! crypto_has_alg ( hash_algo_name [ idx ] , 0 , 0 ) ) {
pr_err ( " unavailable hash algorithm \" %s \" , check your kernel configuration " ,
token ) ;
return 0 ;
}
2021-08-16 08:11:00 +00:00
/* Add the hash algorithm to the 'allowed' bitfield */
res | = ( 1U < < idx ) ;
}
return res ;
}
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
static int ima_parse_rule ( char * rule , struct ima_rule_entry * entry )
2009-02-04 09:07:00 -05:00
{
struct audit_buffer * ab ;
2014-11-05 07:53:55 -05:00
char * from ;
2009-02-04 09:07:00 -05:00
char * p ;
2021-10-07 14:03:02 -06:00
bool eid_token ; /* either euid or egid */
2019-06-19 15:46:11 -07:00
struct ima_template_desc * template_desc ;
2009-02-04 09:07:00 -05:00
int result = 0 ;
2018-06-04 16:54:55 -04:00
ab = integrity_audit_log_start ( audit_context ( ) , GFP_KERNEL ,
AUDIT_INTEGRITY_POLICY_RULE ) ;
2009-02-04 09:07:00 -05:00
2012-05-25 18:24:12 -06:00
entry - > uid = INVALID_UID ;
2021-10-07 14:03:02 -06:00
entry - > gid = INVALID_GID ;
2012-10-02 21:38:48 -07:00
entry - > fowner = INVALID_UID ;
2021-10-07 14:03:02 -06:00
entry - > fgroup = INVALID_GID ;
2017-01-27 19:23:01 +03:00
entry - > uid_op = & uid_eq ;
2021-10-07 14:03:02 -06:00
entry - > gid_op = & gid_eq ;
2017-01-27 19:23:01 +03:00
entry - > fowner_op = & uid_eq ;
2021-10-07 14:03:02 -06:00
entry - > fgroup_op = & gid_eq ;
2010-04-20 10:21:07 -04:00
entry - > action = UNKNOWN ;
2010-04-20 10:21:18 -04:00
while ( ( p = strsep ( & rule , " \t " ) ) ! = NULL ) {
2009-02-04 09:07:00 -05:00
substring_t args [ MAX_OPT_ARGS ] ;
int token ;
unsigned long lnum ;
if ( result < 0 )
break ;
2010-04-20 10:21:18 -04:00
if ( ( * p = = ' \0 ' ) | | ( * p = = ' ' ) | | ( * p = = ' \t ' ) )
continue ;
2009-02-04 09:07:00 -05:00
token = match_token ( p , policy_tokens , args ) ;
switch ( token ) {
case Opt_measure :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " action " , " measure " ) ;
2010-04-20 10:21:01 -04:00
if ( entry - > action ! = UNKNOWN )
result = - EINVAL ;
2009-02-04 09:07:00 -05:00
entry - > action = MEASURE ;
break ;
case Opt_dont_measure :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " action " , " dont_measure " ) ;
2010-04-20 10:21:01 -04:00
if ( entry - > action ! = UNKNOWN )
result = - EINVAL ;
2009-02-04 09:07:00 -05:00
entry - > action = DONT_MEASURE ;
break ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
case Opt_appraise :
ima_log_string ( ab , " action " , " appraise " ) ;
if ( entry - > action ! = UNKNOWN )
result = - EINVAL ;
entry - > action = APPRAISE ;
break ;
case Opt_dont_appraise :
ima_log_string ( ab , " action " , " dont_appraise " ) ;
if ( entry - > action ! = UNKNOWN )
result = - EINVAL ;
entry - > action = DONT_APPRAISE ;
break ;
2012-06-14 10:04:36 -07:00
case Opt_audit :
ima_log_string ( ab , " action " , " audit " ) ;
if ( entry - > action ! = UNKNOWN )
result = - EINVAL ;
entry - > action = AUDIT ;
break ;
ima: support new "hash" and "dont_hash" policy actions
The builtin ima_appraise_tcb policy, which is specified on the boot
command line, can be replaced with a custom policy, normally early in
the boot process. Custom policies can be more restrictive in some ways,
like requiring file signatures, but can be less restrictive in other
ways, like not appraising mutable files. With a less restrictive policy
in place, files in the builtin policy might not be hashed and labeled
with a security.ima hash. On reboot, files which should be labeled in
the ima_appraise_tcb are not labeled, possibly preventing the system
from booting properly.
To resolve this problem, this patch extends the existing IMA policy
actions "measure", "dont_measure", "appraise", "dont_appraise", and
"audit" with "hash" and "dont_hash". The new "hash" action will write
the file hash as security.ima, but without requiring the file to be
appraised as well.
For example, the builtin ima_appraise_tcb policy includes the rule,
"appraise fowner=0". Adding the "hash fowner=0" rule to a custom
policy, will cause the needed file hashes to be calculated and written
as security.ima xattrs.
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
2016-09-29 10:04:52 -04:00
case Opt_hash :
ima_log_string ( ab , " action " , " hash " ) ;
if ( entry - > action ! = UNKNOWN )
result = - EINVAL ;
entry - > action = HASH ;
break ;
case Opt_dont_hash :
ima_log_string ( ab , " action " , " dont_hash " ) ;
if ( entry - > action ! = UNKNOWN )
result = - EINVAL ;
entry - > action = DONT_HASH ;
break ;
2009-02-04 09:07:00 -05:00
case Opt_func :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " func " , args [ 0 ] . from ) ;
2010-04-20 10:21:01 -04:00
if ( entry - > func )
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
result = - EINVAL ;
2010-04-20 10:21:01 -04:00
2010-01-26 17:02:41 -05:00
if ( strcmp ( args [ 0 ] . from , " FILE_CHECK " ) = = 0 )
entry - > func = FILE_CHECK ;
/* PATH_CHECK is for backwards compat */
else if ( strcmp ( args [ 0 ] . from , " PATH_CHECK " ) = = 0 )
entry - > func = FILE_CHECK ;
2012-10-16 12:40:08 +10:30
else if ( strcmp ( args [ 0 ] . from , " MODULE_CHECK " ) = = 0 )
entry - > func = MODULE_CHECK ;
2014-07-22 10:39:48 -04:00
else if ( strcmp ( args [ 0 ] . from , " FIRMWARE_CHECK " ) = = 0 )
entry - > func = FIRMWARE_CHECK ;
2012-12-13 11:15:04 -05:00
else if ( ( strcmp ( args [ 0 ] . from , " FILE_MMAP " ) = = 0 )
| | ( strcmp ( args [ 0 ] . from , " MMAP_CHECK " ) = = 0 ) )
entry - > func = MMAP_CHECK ;
2009-02-04 09:07:00 -05:00
else if ( strcmp ( args [ 0 ] . from , " BPRM_CHECK " ) = = 0 )
entry - > func = BPRM_CHECK ;
2018-01-08 13:36:20 -08:00
else if ( strcmp ( args [ 0 ] . from , " CREDS_CHECK " ) = = 0 )
entry - > func = CREDS_CHECK ;
ima: support for kexec image and initramfs
Add IMA policy support for measuring/appraising the kexec image and
initramfs. Two new IMA policy identifiers KEXEC_KERNEL_CHECK and
KEXEC_INITRAMFS_CHECK are defined.
Example policy rules:
measure func=KEXEC_KERNEL_CHECK
appraise func=KEXEC_KERNEL_CHECK appraise_type=imasig
measure func=KEXEC_INITRAMFS_CHECK
appraise func=KEXEC_INITRAMFS_CHECK appraise_type=imasig
Moving the enumeration to the vfs layer simplified the patches, allowing
the IMA changes, for the most part, to be separated from the other
changes. Unfortunately, passing either a kernel_read_file_id or a
ima_hooks enumeration within IMA is messy.
Option 1: duplicate kernel_read_file enumeration in ima_hooks
enum kernel_read_file_id {
...
READING_KEXEC_IMAGE,
READING_KEXEC_INITRAMFS,
READING_MAX_ID
enum ima_hooks {
...
KEXEC_KERNEL_CHECK
KEXEC_INITRAMFS_CHECK
Option 2: define ima_hooks as extension of kernel_read_file
eg: enum ima_hooks {
FILE_CHECK = READING_MAX_ID,
MMAP_CHECK,
In order to pass both kernel_read_file_id and ima_hooks values, we
would need to specify a struct containing a union.
struct caller_id {
union {
enum ima_hooks func_id;
enum kernel_read_file_id read_id;
};
};
Option 3: incorportate the ima_hooks enumeration into kernel_read_file_id,
perhaps changing the enumeration name.
For now, duplicate the new READING_KEXEC_IMAGE/INITRAMFS in the ima_hooks.
Changelog v4:
- replaced switch statement with a kernel_read_file_id to an ima_hooks
id mapping array - Dmitry
- renamed ima_hook tokens KEXEC_CHECK and INITRAMFS_CHECK to
KEXEC_KERNEL_CHECK and KEXEC_INITRAMFS_CHECK respectively - Dave Young
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Acked-by: Petko Manolov <petkan@mip-labs.com>
Acked-by: Dmitry Kasatkin <dmitry.kasatkin@huawei.com>
Cc: Dave Young <dyoung@redhat.com>
2016-01-14 20:59:14 -05:00
else if ( strcmp ( args [ 0 ] . from , " KEXEC_KERNEL_CHECK " ) = =
0 )
entry - > func = KEXEC_KERNEL_CHECK ;
else if ( strcmp ( args [ 0 ] . from , " KEXEC_INITRAMFS_CHECK " )
= = 0 )
entry - > func = KEXEC_INITRAMFS_CHECK ;
2016-01-15 10:17:12 -05:00
else if ( strcmp ( args [ 0 ] . from , " POLICY_CHECK " ) = = 0 )
entry - > func = POLICY_CHECK ;
2019-06-23 23:23:29 -07:00
else if ( strcmp ( args [ 0 ] . from , " KEXEC_CMDLINE " ) = = 0 )
entry - > func = KEXEC_CMDLINE ;
2020-08-11 14:26:21 -05:00
else if ( IS_ENABLED ( CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS ) & &
strcmp ( args [ 0 ] . from , " KEY_CHECK " ) = = 0 )
2019-12-11 08:47:03 -08:00
entry - > func = KEY_CHECK ;
2021-01-07 20:07:04 -08:00
else if ( strcmp ( args [ 0 ] . from , " CRITICAL_DATA " ) = = 0 )
entry - > func = CRITICAL_DATA ;
2021-08-16 08:11:01 +00:00
else if ( strcmp ( args [ 0 ] . from , " SETXATTR_CHECK " ) = = 0 )
entry - > func = SETXATTR_CHECK ;
2009-02-04 09:07:00 -05:00
else
result = - EINVAL ;
if ( ! result )
entry - > flags | = IMA_FUNC ;
break ;
case Opt_mask :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " mask " , args [ 0 ] . from ) ;
2010-04-20 10:21:01 -04:00
if ( entry - > mask )
result = - EINVAL ;
2014-11-05 07:53:55 -05:00
from = args [ 0 ] . from ;
if ( * from = = ' ^ ' )
from + + ;
if ( ( strcmp ( from , " MAY_EXEC " ) ) = = 0 )
2009-02-04 09:07:00 -05:00
entry - > mask = MAY_EXEC ;
2014-11-05 07:53:55 -05:00
else if ( strcmp ( from , " MAY_WRITE " ) = = 0 )
2009-02-04 09:07:00 -05:00
entry - > mask = MAY_WRITE ;
2014-11-05 07:53:55 -05:00
else if ( strcmp ( from , " MAY_READ " ) = = 0 )
2009-02-04 09:07:00 -05:00
entry - > mask = MAY_READ ;
2014-11-05 07:53:55 -05:00
else if ( strcmp ( from , " MAY_APPEND " ) = = 0 )
2009-02-04 09:07:00 -05:00
entry - > mask = MAY_APPEND ;
else
result = - EINVAL ;
if ( ! result )
2014-11-05 07:53:55 -05:00
entry - > flags | = ( * args [ 0 ] . from = = ' ^ ' )
? IMA_INMASK : IMA_MASK ;
2009-02-04 09:07:00 -05:00
break ;
case Opt_fsmagic :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " fsmagic " , args [ 0 ] . from ) ;
2010-04-20 10:21:01 -04:00
if ( entry - > fsmagic ) {
result = - EINVAL ;
break ;
}
2014-03-04 18:04:20 +02:00
result = kstrtoul ( args [ 0 ] . from , 16 , & entry - > fsmagic ) ;
2009-02-04 09:07:00 -05:00
if ( ! result )
entry - > flags | = IMA_FSMAGIC ;
break ;
2018-01-15 11:20:36 -05:00
case Opt_fsname :
ima_log_string ( ab , " fsname " , args [ 0 ] . from ) ;
entry - > fsname = kstrdup ( args [ 0 ] . from , GFP_KERNEL ) ;
if ( ! entry - > fsname ) {
result = - ENOMEM ;
break ;
}
result = 0 ;
entry - > flags | = IMA_FSNAME ;
break ;
2019-12-11 08:47:07 -08:00
case Opt_keyrings :
ima_log_string ( ab , " keyrings " , args [ 0 ] . from ) ;
2020-08-11 14:26:21 -05:00
if ( ! IS_ENABLED ( CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS ) | |
entry - > keyrings ) {
2019-12-11 08:47:07 -08:00
result = - EINVAL ;
break ;
}
2020-01-16 18:18:21 -08:00
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
entry - > keyrings = ima_alloc_rule_opt_list ( args ) ;
if ( IS_ERR ( entry - > keyrings ) ) {
result = PTR_ERR ( entry - > keyrings ) ;
entry - > keyrings = NULL ;
2019-12-11 08:47:07 -08:00
break ;
}
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
2019-12-11 08:47:07 -08:00
entry - > flags | = IMA_KEYRINGS ;
break ;
2021-01-07 20:07:05 -08:00
case Opt_label :
ima_log_string ( ab , " label " , args [ 0 ] . from ) ;
if ( entry - > label ) {
result = - EINVAL ;
break ;
}
entry - > label = ima_alloc_rule_opt_list ( args ) ;
if ( IS_ERR ( entry - > label ) ) {
result = PTR_ERR ( entry - > label ) ;
entry - > label = NULL ;
break ;
}
entry - > flags | = IMA_LABEL ;
break ;
2012-09-03 23:23:13 +03:00
case Opt_fsuuid :
ima_log_string ( ab , " fsuuid " , args [ 0 ] . from ) ;
2018-01-17 20:27:11 +02:00
if ( ! uuid_is_null ( & entry - > fsuuid ) ) {
2012-09-03 23:23:13 +03:00
result = - EINVAL ;
break ;
}
2017-06-01 07:00:26 +02:00
result = uuid_parse ( args [ 0 ] . from , & entry - > fsuuid ) ;
2013-02-24 23:42:37 -05:00
if ( ! result )
entry - > flags | = IMA_FSUUID ;
2012-09-03 23:23:13 +03:00
break ;
2017-01-27 19:23:01 +03:00
case Opt_uid_gt :
case Opt_euid_gt :
entry - > uid_op = & uid_gt ;
2020-08-23 17:36:59 -05:00
fallthrough ;
2017-01-27 19:23:01 +03:00
case Opt_uid_lt :
case Opt_euid_lt :
if ( ( token = = Opt_uid_lt ) | | ( token = = Opt_euid_lt ) )
entry - > uid_op = & uid_lt ;
2020-08-23 17:36:59 -05:00
fallthrough ;
2017-01-27 19:23:01 +03:00
case Opt_uid_eq :
case Opt_euid_eq :
2021-10-07 14:03:02 -06:00
eid_token = ( token = = Opt_euid_eq ) | |
( token = = Opt_euid_gt ) | |
( token = = Opt_euid_lt ) ;
2017-01-27 19:23:01 +03:00
2021-10-07 14:03:02 -06:00
ima_log_string_op ( ab , eid_token ? " euid " : " uid " ,
args [ 0 ] . from , token ) ;
2010-04-20 10:21:01 -04:00
2012-05-25 18:24:12 -06:00
if ( uid_valid ( entry - > uid ) ) {
2010-04-20 10:21:01 -04:00
result = - EINVAL ;
break ;
}
2014-02-05 15:13:14 +09:00
result = kstrtoul ( args [ 0 ] . from , 10 , & lnum ) ;
2009-02-04 09:07:00 -05:00
if ( ! result ) {
2014-11-05 07:48:36 -05:00
entry - > uid = make_kuid ( current_user_ns ( ) ,
( uid_t ) lnum ) ;
if ( ! uid_valid ( entry - > uid ) | |
( uid_t ) lnum ! = lnum )
2009-02-04 09:07:00 -05:00
result = - EINVAL ;
else
2021-10-07 14:03:02 -06:00
entry - > flags | = eid_token
? IMA_EUID : IMA_UID ;
}
break ;
case Opt_gid_gt :
case Opt_egid_gt :
entry - > gid_op = & gid_gt ;
fallthrough ;
case Opt_gid_lt :
case Opt_egid_lt :
if ( ( token = = Opt_gid_lt ) | | ( token = = Opt_egid_lt ) )
entry - > gid_op = & gid_lt ;
fallthrough ;
case Opt_gid_eq :
case Opt_egid_eq :
eid_token = ( token = = Opt_egid_eq ) | |
( token = = Opt_egid_gt ) | |
( token = = Opt_egid_lt ) ;
ima_log_string_op ( ab , eid_token ? " egid " : " gid " ,
args [ 0 ] . from , token ) ;
if ( gid_valid ( entry - > gid ) ) {
result = - EINVAL ;
break ;
}
result = kstrtoul ( args [ 0 ] . from , 10 , & lnum ) ;
if ( ! result ) {
entry - > gid = make_kgid ( current_user_ns ( ) ,
( gid_t ) lnum ) ;
if ( ! gid_valid ( entry - > gid ) | |
( ( ( gid_t ) lnum ) ! = lnum ) )
result = - EINVAL ;
else
entry - > flags | = eid_token
? IMA_EGID : IMA_GID ;
2009-02-04 09:07:00 -05:00
}
break ;
2017-01-27 19:23:01 +03:00
case Opt_fowner_gt :
entry - > fowner_op = & uid_gt ;
2020-08-23 17:36:59 -05:00
fallthrough ;
2017-01-27 19:23:01 +03:00
case Opt_fowner_lt :
if ( token = = Opt_fowner_lt )
entry - > fowner_op = & uid_lt ;
2020-08-23 17:36:59 -05:00
fallthrough ;
2017-01-27 19:23:01 +03:00
case Opt_fowner_eq :
2021-10-07 14:03:02 -06:00
ima_log_string_op ( ab , " fowner " , args [ 0 ] . from , token ) ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
2012-10-02 21:38:48 -07:00
if ( uid_valid ( entry - > fowner ) ) {
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
result = - EINVAL ;
break ;
}
2014-02-05 15:13:14 +09:00
result = kstrtoul ( args [ 0 ] . from , 10 , & lnum ) ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
if ( ! result ) {
2021-10-07 14:03:01 -06:00
entry - > fowner = make_kuid ( current_user_ns ( ) ,
( uid_t ) lnum ) ;
if ( ! uid_valid ( entry - > fowner ) | |
( ( ( uid_t ) lnum ) ! = lnum ) )
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
result = - EINVAL ;
else
entry - > flags | = IMA_FOWNER ;
}
break ;
2021-10-07 14:03:02 -06:00
case Opt_fgroup_gt :
entry - > fgroup_op = & gid_gt ;
fallthrough ;
case Opt_fgroup_lt :
if ( token = = Opt_fgroup_lt )
entry - > fgroup_op = & gid_lt ;
fallthrough ;
case Opt_fgroup_eq :
ima_log_string_op ( ab , " fgroup " , args [ 0 ] . from , token ) ;
if ( gid_valid ( entry - > fgroup ) ) {
result = - EINVAL ;
break ;
}
result = kstrtoul ( args [ 0 ] . from , 10 , & lnum ) ;
if ( ! result ) {
entry - > fgroup = make_kgid ( current_user_ns ( ) ,
( gid_t ) lnum ) ;
if ( ! gid_valid ( entry - > fgroup ) | |
( ( ( gid_t ) lnum ) ! = lnum ) )
result = - EINVAL ;
else
entry - > flags | = IMA_FGROUP ;
}
break ;
2009-02-04 09:07:00 -05:00
case Opt_obj_user :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " obj_user " , args [ 0 ] . from ) ;
2013-01-03 14:19:09 -05:00
result = ima_lsm_rule_init ( entry , args ,
2009-02-04 09:07:00 -05:00
LSM_OBJ_USER ,
AUDIT_OBJ_USER ) ;
break ;
case Opt_obj_role :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " obj_role " , args [ 0 ] . from ) ;
2013-01-03 14:19:09 -05:00
result = ima_lsm_rule_init ( entry , args ,
2009-02-04 09:07:00 -05:00
LSM_OBJ_ROLE ,
AUDIT_OBJ_ROLE ) ;
break ;
case Opt_obj_type :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " obj_type " , args [ 0 ] . from ) ;
2013-01-03 14:19:09 -05:00
result = ima_lsm_rule_init ( entry , args ,
2009-02-04 09:07:00 -05:00
LSM_OBJ_TYPE ,
AUDIT_OBJ_TYPE ) ;
break ;
case Opt_subj_user :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " subj_user " , args [ 0 ] . from ) ;
2013-01-03 14:19:09 -05:00
result = ima_lsm_rule_init ( entry , args ,
2009-02-04 09:07:00 -05:00
LSM_SUBJ_USER ,
AUDIT_SUBJ_USER ) ;
break ;
case Opt_subj_role :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " subj_role " , args [ 0 ] . from ) ;
2013-01-03 14:19:09 -05:00
result = ima_lsm_rule_init ( entry , args ,
2009-02-04 09:07:00 -05:00
LSM_SUBJ_ROLE ,
AUDIT_SUBJ_ROLE ) ;
break ;
case Opt_subj_type :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " subj_type " , args [ 0 ] . from ) ;
2013-01-03 14:19:09 -05:00
result = ima_lsm_rule_init ( entry , args ,
2009-02-04 09:07:00 -05:00
LSM_SUBJ_TYPE ,
AUDIT_SUBJ_TYPE ) ;
break ;
2012-06-08 13:58:49 +03:00
case Opt_appraise_type :
ima_log_string ( ab , " appraise_type " , args [ 0 ] . from ) ;
if ( ( strcmp ( args [ 0 ] . from , " imasig " ) ) = = 0 )
entry - > flags | = IMA_DIGSIG_REQUIRED ;
2020-07-09 01:19:09 -05:00
else if ( IS_ENABLED ( CONFIG_IMA_APPRAISE_MODSIG ) & &
2019-06-27 23:19:28 -03:00
strcmp ( args [ 0 ] . from , " imasig|modsig " ) = = 0 )
entry - > flags | = IMA_DIGSIG_REQUIRED |
IMA_MODSIG_ALLOWED ;
2012-06-08 13:58:49 +03:00
else
result = - EINVAL ;
break ;
2019-10-30 23:31:32 -04:00
case Opt_appraise_flag :
ima_log_string ( ab , " appraise_flag " , args [ 0 ] . from ) ;
2020-07-09 01:19:06 -05:00
if ( IS_ENABLED ( CONFIG_IMA_APPRAISE_MODSIG ) & &
strstr ( args [ 0 ] . from , " blacklist " ) )
2019-10-30 23:31:32 -04:00
entry - > flags | = IMA_CHECK_BLACKLIST ;
2020-07-09 01:19:06 -05:00
else
result = - EINVAL ;
2019-10-30 23:31:32 -04:00
break ;
2021-08-16 08:11:00 +00:00
case Opt_appraise_algos :
ima_log_string ( ab , " appraise_algos " , args [ 0 ] . from ) ;
if ( entry - > allowed_algos ) {
result = - EINVAL ;
break ;
}
entry - > allowed_algos =
ima_parse_appraise_algos ( args [ 0 ] . from ) ;
/* invalid or empty list of algorithms */
if ( ! entry - > allowed_algos ) {
result = - EINVAL ;
break ;
}
entry - > flags | = IMA_VALIDATE_ALGOS ;
break ;
ima: audit log files opened with O_DIRECT flag
Files are measured or appraised based on the IMA policy. When a
file, in policy, is opened with the O_DIRECT flag, a deadlock
occurs.
The first attempt at resolving this lockdep temporarily removed the
O_DIRECT flag and restored it, after calculating the hash. The
second attempt introduced the O_DIRECT_HAVELOCK flag. Based on this
flag, do_blockdev_direct_IO() would skip taking the i_mutex a second
time. The third attempt, by Dmitry Kasatkin, resolves the i_mutex
locking issue, by re-introducing the IMA mutex, but uncovered
another problem. Reading a file with O_DIRECT flag set, writes
directly to userspace pages. A second patch allocates a user-space
like memory. This works for all IMA hooks, except ima_file_free(),
which is called on __fput() to recalculate the file hash.
Until this last issue is addressed, do not 'collect' the
measurement for measuring, appraising, or auditing files opened
with the O_DIRECT flag set. Based on policy, permit or deny file
access. This patch defines a new IMA policy rule option named
'permit_directio'. Policy rules could be defined, based on LSM
or other criteria, to permit specific applications to open files
with the O_DIRECT flag set.
Changelog v1:
- permit or deny file access based IMA policy rules
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Acked-by: Dmitry Kasatkin <d.kasatkin@samsung.com>
Cc: <stable@vger.kernel.org>
2014-05-12 09:28:11 -04:00
case Opt_permit_directio :
entry - > flags | = IMA_PERMIT_DIRECTIO ;
2016-06-01 13:14:01 -05:00
break ;
case Opt_pcr :
ima_log_string ( ab , " pcr " , args [ 0 ] . from ) ;
result = kstrtoint ( args [ 0 ] . from , 10 , & entry - > pcr ) ;
if ( result | | INVALID_PCR ( entry - > pcr ) )
result = - EINVAL ;
else
entry - > flags | = IMA_PCR ;
2019-06-19 15:46:11 -07:00
break ;
case Opt_template :
ima_log_string ( ab , " template " , args [ 0 ] . from ) ;
if ( entry - > action ! = MEASURE ) {
result = - EINVAL ;
break ;
}
template_desc = lookup_template_desc ( args [ 0 ] . from ) ;
if ( ! template_desc | | entry - > template ) {
result = - EINVAL ;
break ;
}
/*
* template_desc_init_fields ( ) does nothing if
* the template is already initialised , so
* it ' s safe to do this unconditionally
*/
template_desc_init_fields ( template_desc - > fmt ,
& ( template_desc - > fields ) ,
& ( template_desc - > num_fields ) ) ;
entry - > template = template_desc ;
ima: audit log files opened with O_DIRECT flag
Files are measured or appraised based on the IMA policy. When a
file, in policy, is opened with the O_DIRECT flag, a deadlock
occurs.
The first attempt at resolving this lockdep temporarily removed the
O_DIRECT flag and restored it, after calculating the hash. The
second attempt introduced the O_DIRECT_HAVELOCK flag. Based on this
flag, do_blockdev_direct_IO() would skip taking the i_mutex a second
time. The third attempt, by Dmitry Kasatkin, resolves the i_mutex
locking issue, by re-introducing the IMA mutex, but uncovered
another problem. Reading a file with O_DIRECT flag set, writes
directly to userspace pages. A second patch allocates a user-space
like memory. This works for all IMA hooks, except ima_file_free(),
which is called on __fput() to recalculate the file hash.
Until this last issue is addressed, do not 'collect' the
measurement for measuring, appraising, or auditing files opened
with the O_DIRECT flag set. Based on policy, permit or deny file
access. This patch defines a new IMA policy rule option named
'permit_directio'. Policy rules could be defined, based on LSM
or other criteria, to permit specific applications to open files
with the O_DIRECT flag set.
Changelog v1:
- permit or deny file access based IMA policy rules
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Acked-by: Dmitry Kasatkin <d.kasatkin@samsung.com>
Cc: <stable@vger.kernel.org>
2014-05-12 09:28:11 -04:00
break ;
2009-02-04 09:07:00 -05:00
case Opt_err :
2010-04-20 10:21:30 -04:00
ima_log_string ( ab , " UNKNOWN " , p ) ;
2010-04-20 10:21:13 -04:00
result = - EINVAL ;
2009-02-04 09:07:00 -05:00
break ;
}
}
2020-07-09 01:19:03 -05:00
if ( ! result & & ! ima_validate_rule ( entry ) )
2009-02-04 09:07:00 -05:00
result = - EINVAL ;
2018-04-12 00:15:22 -04:00
else if ( entry - > action = = APPRAISE )
temp_ima_appraise | = ima_appraise_flag ( entry - > func ) ;
2019-06-27 23:19:32 -03:00
if ( ! result & & entry - > flags & IMA_MODSIG_ALLOWED ) {
template_desc = entry - > template ? entry - > template :
ima_template_desc_current ( ) ;
check_template_modsig ( template_desc ) ;
}
2012-02-14 17:11:07 -05:00
audit_log_format ( ab , " res=%d " , ! result ) ;
2009-02-04 09:07:00 -05:00
audit_log_end ( ab ) ;
return result ;
}
/**
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
* ima_parse_add_rule - add a rule to ima_policy_rules
2009-02-04 09:07:00 -05:00
* @ rule - ima measurement policy rule
*
2015-12-02 17:47:54 +02:00
* Avoid locking by allowing just one writer at a time in ima_write_policy ( )
2010-04-20 10:20:54 -04:00
* Returns the length of the rule parsed , an error code on failure
2009-02-04 09:07:00 -05:00
*/
2010-04-20 10:20:54 -04:00
ssize_t ima_parse_add_rule ( char * rule )
2009-02-04 09:07:00 -05:00
{
2013-12-11 14:44:04 -05:00
static const char op [ ] = " update_policy " ;
2010-04-20 10:20:54 -04:00
char * p ;
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
struct ima_rule_entry * entry ;
2010-04-20 10:20:54 -04:00
ssize_t result , len ;
2009-02-04 09:07:00 -05:00
int audit_info = 0 ;
2014-10-03 14:40:19 +03:00
p = strsep ( & rule , " \n " ) ;
len = strlen ( p ) + 1 ;
2014-10-03 14:40:20 +03:00
p + = strspn ( p , " \t " ) ;
2014-10-03 14:40:19 +03:00
2014-10-03 14:40:20 +03:00
if ( * p = = ' # ' | | * p = = ' \0 ' )
2014-10-03 14:40:19 +03:00
return len ;
2009-02-04 09:07:00 -05:00
entry = kzalloc ( sizeof ( * entry ) , GFP_KERNEL ) ;
if ( ! entry ) {
integrity_audit_msg ( AUDIT_INTEGRITY_STATUS , NULL ,
NULL , op , " -ENOMEM " , - ENOMEM , audit_info ) ;
return - ENOMEM ;
}
INIT_LIST_HEAD ( & entry - > list ) ;
2010-04-20 10:20:54 -04:00
result = ima_parse_rule ( p , entry ) ;
2010-04-20 10:21:24 -04:00
if ( result ) {
2020-07-09 01:19:02 -05:00
ima_free_rule ( entry ) ;
2009-02-11 11:12:28 -05:00
integrity_audit_msg ( AUDIT_INTEGRITY_STATUS , NULL ,
2014-06-16 15:52:07 -04:00
NULL , op , " invalid-policy " , result ,
2009-02-11 11:12:28 -05:00
audit_info ) ;
2010-04-20 10:21:24 -04:00
return result ;
2009-02-11 11:12:28 -05:00
}
2010-04-20 10:21:24 -04:00
2015-12-02 17:47:54 +02:00
list_add_tail ( & entry - > list , & ima_temp_rules ) ;
2010-04-20 10:21:24 -04:00
return len ;
2009-02-04 09:07:00 -05:00
}
2015-12-02 17:47:54 +02:00
/**
* ima_delete_rules ( ) called to cleanup invalid in - flight policy .
* We don ' t need locking as we operate on the temp list , which is
* different from the active one . There is also only one user of
* ima_delete_rules ( ) at a time .
*/
2009-02-05 09:28:26 +11:00
void ima_delete_rules ( void )
2009-02-04 09:07:00 -05:00
{
ima: add appraise action keywords and default rules
Unlike the IMA measurement policy, the appraise policy can not be dependent
on runtime process information, such as the task uid, as the 'security.ima'
xattr is written on file close and must be updated each time the file changes,
regardless of the current task uid.
This patch extends the policy language with 'fowner', defines an appraise
policy, which appraises all files owned by root, and defines 'ima_appraise_tcb',
a new boot command line option, to enable the appraise policy.
Changelog v3:
- separate the measure from the appraise rules in order to support measuring
without appraising and appraising without measuring.
- change appraisal default for filesystems without xattr support to fail
- update default appraise policy for cgroups
Changelog v1:
- don't appraise RAMFS (Dmitry Kasatkin)
- merged rest of "ima: ima_must_appraise_or_measure API change" commit
(Dmtiry Kasatkin)
ima_must_appraise_or_measure() called ima_match_policy twice, which
searched the policy for a matching rule. Once for a matching measurement
rule and subsequently for an appraisal rule. Searching the policy twice
is unnecessary overhead, which could be noticeable with a large policy.
The new version of ima_must_appraise_or_measure() does everything in a
single iteration using a new version of ima_match_policy(). It returns
IMA_MEASURE, IMA_APPRAISE mask.
With the use of action mask only one efficient matching function
is enough. Removed other specific versions of matching functions.
Changelog:
- change 'owner' to 'fowner' to conform to the new LSM conditions posted by
Roberto Sassu.
- fix calls to ima_log_string()
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@intel.com>
2011-03-09 22:25:48 -05:00
struct ima_rule_entry * entry , * tmp ;
2009-02-04 09:07:00 -05:00
2015-12-07 14:35:47 -05:00
temp_ima_appraise = 0 ;
2015-12-02 17:47:54 +02:00
list_for_each_entry_safe ( entry , tmp , & ima_temp_rules , list ) {
2009-02-04 09:07:00 -05:00
list_del ( & entry - > list ) ;
2020-07-09 01:19:01 -05:00
ima_free_rule ( entry ) ;
2009-02-04 09:07:00 -05:00
}
}
2015-12-02 17:47:56 +02:00
2020-06-18 14:10:12 -07:00
# define __ima_hook_stringify(func, str) (#func),
2019-06-27 23:19:30 -03:00
const char * const func_tokens [ ] = {
__ima_hooks ( __ima_hook_stringify )
} ;
2015-12-02 17:47:56 +02:00
# ifdef CONFIG_IMA_READ_POLICY
enum {
mask_exec = 0 , mask_write , mask_read , mask_append
} ;
2017-06-07 22:49:10 -03:00
static const char * const mask_tokens [ ] = {
2019-05-29 15:30:35 +02:00
" ^MAY_EXEC " ,
" ^MAY_WRITE " ,
" ^MAY_READ " ,
" ^MAY_APPEND "
2015-12-02 17:47:56 +02:00
} ;
void * ima_policy_start ( struct seq_file * m , loff_t * pos )
{
loff_t l = * pos ;
struct ima_rule_entry * entry ;
2021-10-09 18:38:21 +08:00
struct list_head * ima_rules_tmp ;
2015-12-02 17:47:56 +02:00
rcu_read_lock ( ) ;
2021-10-09 18:38:21 +08:00
ima_rules_tmp = rcu_dereference ( ima_rules ) ;
list_for_each_entry_rcu ( entry , ima_rules_tmp , list ) {
2015-12-02 17:47:56 +02:00
if ( ! l - - ) {
rcu_read_unlock ( ) ;
return entry ;
}
}
rcu_read_unlock ( ) ;
return NULL ;
}
void * ima_policy_next ( struct seq_file * m , void * v , loff_t * pos )
{
struct ima_rule_entry * entry = v ;
rcu_read_lock ( ) ;
entry = list_entry_rcu ( entry - > list . next , struct ima_rule_entry , list ) ;
rcu_read_unlock ( ) ;
( * pos ) + + ;
2021-10-09 18:38:21 +08:00
return ( & entry - > list = = & ima_default_rules | |
& entry - > list = = & ima_policy_rules ) ? NULL : entry ;
2015-12-02 17:47:56 +02:00
}
void ima_policy_stop ( struct seq_file * m , void * v )
{
}
2018-12-17 19:14:49 -05:00
# define pt(token) policy_tokens[token].pattern
2015-12-02 17:47:56 +02:00
# define mt(token) mask_tokens[token]
2016-01-28 13:10:36 -05:00
/*
* policy_func_show - display the ima_hooks policy rule
*/
static void policy_func_show ( struct seq_file * m , enum ima_hooks func )
{
2017-06-07 22:49:11 -03:00
if ( func > 0 & & func < MAX_CHECK )
seq_printf ( m , " func=%s " , func_tokens [ func ] ) ;
else
seq_printf ( m , " func=%d " , func ) ;
2016-01-28 13:10:36 -05:00
}
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
static void ima_show_rule_opt_list ( struct seq_file * m ,
const struct ima_rule_opt_list * opt_list )
{
size_t i ;
for ( i = 0 ; i < opt_list - > count ; i + + )
seq_printf ( m , " %s%s " , i ? " | " : " " , opt_list - > items [ i ] ) ;
}
2021-08-16 08:11:00 +00:00
static void ima_policy_show_appraise_algos ( struct seq_file * m ,
unsigned int allowed_hashes )
{
int idx , list_size = 0 ;
for ( idx = 0 ; idx < HASH_ALGO__LAST ; idx + + ) {
if ( ! ( allowed_hashes & ( 1U < < idx ) ) )
continue ;
/* only add commas if the list contains multiple entries */
if ( list_size + + )
seq_puts ( m , " , " ) ;
seq_puts ( m , hash_algo_name [ idx ] ) ;
}
}
2015-12-02 17:47:56 +02:00
int ima_policy_show ( struct seq_file * m , void * v )
{
struct ima_rule_entry * entry = v ;
2016-05-20 17:00:57 -07:00
int i ;
2015-12-02 17:47:56 +02:00
char tbuf [ 64 ] = { 0 , } ;
2019-05-29 15:30:35 +02:00
int offset = 0 ;
2015-12-02 17:47:56 +02:00
rcu_read_lock ( ) ;
2022-02-01 15:37:10 -05:00
/* Do not print rules with inactive LSM labels */
for ( i = 0 ; i < MAX_LSM_RULES ; i + + ) {
if ( entry - > lsm [ i ] . args_p & & ! entry - > lsm [ i ] . rule ) {
rcu_read_unlock ( ) ;
return 0 ;
}
}
2015-12-02 17:47:56 +02:00
if ( entry - > action & MEASURE )
seq_puts ( m , pt ( Opt_measure ) ) ;
if ( entry - > action & DONT_MEASURE )
seq_puts ( m , pt ( Opt_dont_measure ) ) ;
if ( entry - > action & APPRAISE )
seq_puts ( m , pt ( Opt_appraise ) ) ;
if ( entry - > action & DONT_APPRAISE )
seq_puts ( m , pt ( Opt_dont_appraise ) ) ;
if ( entry - > action & AUDIT )
seq_puts ( m , pt ( Opt_audit ) ) ;
ima: support new "hash" and "dont_hash" policy actions
The builtin ima_appraise_tcb policy, which is specified on the boot
command line, can be replaced with a custom policy, normally early in
the boot process. Custom policies can be more restrictive in some ways,
like requiring file signatures, but can be less restrictive in other
ways, like not appraising mutable files. With a less restrictive policy
in place, files in the builtin policy might not be hashed and labeled
with a security.ima hash. On reboot, files which should be labeled in
the ima_appraise_tcb are not labeled, possibly preventing the system
from booting properly.
To resolve this problem, this patch extends the existing IMA policy
actions "measure", "dont_measure", "appraise", "dont_appraise", and
"audit" with "hash" and "dont_hash". The new "hash" action will write
the file hash as security.ima, but without requiring the file to be
appraised as well.
For example, the builtin ima_appraise_tcb policy includes the rule,
"appraise fowner=0". Adding the "hash fowner=0" rule to a custom
policy, will cause the needed file hashes to be calculated and written
as security.ima xattrs.
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com>
2016-09-29 10:04:52 -04:00
if ( entry - > action & HASH )
seq_puts ( m , pt ( Opt_hash ) ) ;
if ( entry - > action & DONT_HASH )
seq_puts ( m , pt ( Opt_dont_hash ) ) ;
2015-12-02 17:47:56 +02:00
seq_puts ( m , " " ) ;
2016-01-28 13:10:36 -05:00
if ( entry - > flags & IMA_FUNC )
policy_func_show ( m , entry - > func ) ;
2015-12-02 17:47:56 +02:00
2019-05-29 15:30:35 +02:00
if ( ( entry - > flags & IMA_MASK ) | | ( entry - > flags & IMA_INMASK ) ) {
if ( entry - > flags & IMA_MASK )
offset = 1 ;
2015-12-02 17:47:56 +02:00
if ( entry - > mask & MAY_EXEC )
2019-05-29 15:30:35 +02:00
seq_printf ( m , pt ( Opt_mask ) , mt ( mask_exec ) + offset ) ;
2015-12-02 17:47:56 +02:00
if ( entry - > mask & MAY_WRITE )
2019-05-29 15:30:35 +02:00
seq_printf ( m , pt ( Opt_mask ) , mt ( mask_write ) + offset ) ;
2015-12-02 17:47:56 +02:00
if ( entry - > mask & MAY_READ )
2019-05-29 15:30:35 +02:00
seq_printf ( m , pt ( Opt_mask ) , mt ( mask_read ) + offset ) ;
2015-12-02 17:47:56 +02:00
if ( entry - > mask & MAY_APPEND )
2019-05-29 15:30:35 +02:00
seq_printf ( m , pt ( Opt_mask ) , mt ( mask_append ) + offset ) ;
2015-12-02 17:47:56 +02:00
seq_puts ( m , " " ) ;
}
if ( entry - > flags & IMA_FSMAGIC ) {
snprintf ( tbuf , sizeof ( tbuf ) , " 0x%lx " , entry - > fsmagic ) ;
seq_printf ( m , pt ( Opt_fsmagic ) , tbuf ) ;
seq_puts ( m , " " ) ;
}
2018-01-15 11:20:36 -05:00
if ( entry - > flags & IMA_FSNAME ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %s " , entry - > fsname ) ;
seq_printf ( m , pt ( Opt_fsname ) , tbuf ) ;
seq_puts ( m , " " ) ;
}
2019-12-11 08:47:07 -08:00
if ( entry - > flags & IMA_KEYRINGS ) {
ima: Pre-parse the list of keyrings in a KEY_CHECK rule
The ima_keyrings buffer was used as a work buffer for strsep()-based
parsing of the "keyrings=" option of an IMA policy rule. This parsing
was re-performed each time an asymmetric key was added to a kernel
keyring for each loaded policy rule that contained a "keyrings=" option.
An example rule specifying this option is:
measure func=KEY_CHECK keyrings=a|b|c
The rule says to measure asymmetric keys added to any of the kernel
keyrings named "a", "b", or "c". The size of the buffer size was
equal to the size of the largest "keyrings=" value seen in a previously
loaded rule (5 + 1 for the NUL-terminator in the previous example) and
the buffer was pre-allocated at the time of policy load.
The pre-allocated buffer approach suffered from a couple bugs:
1) There was no locking around the use of the buffer so concurrent key
add operations, to two different keyrings, would result in the
strsep() loop of ima_match_keyring() to modify the buffer at the same
time. This resulted in unexpected results from ima_match_keyring()
and, therefore, could cause unintended keys to be measured or keys to
not be measured when IMA policy intended for them to be measured.
2) If the kstrdup() that initialized entry->keyrings in ima_parse_rule()
failed, the ima_keyrings buffer was freed and set to NULL even when a
valid KEY_CHECK rule was previously loaded. The next KEY_CHECK event
would trigger a call to strcpy() with a NULL destination pointer and
crash the kernel.
Remove the need for a pre-allocated global buffer by parsing the list of
keyrings in a KEY_CHECK rule at the time of policy load. The
ima_rule_entry will contain an array of string pointers which point to
the name of each keyring specified in the rule. No string processing
needs to happen at the time of asymmetric key add so iterating through
the list and doing a string comparison is all that's required at the
time of policy check.
In the process of changing how the "keyrings=" policy option is handled,
a couple additional bugs were fixed:
1) The rule parser accepted rules containing invalid "keyrings=" values
such as "a|b||c", "a|b|", or simply "|".
2) The /sys/kernel/security/ima/policy file did not display the entire
"keyrings=" value if the list of keyrings was longer than what could
fit in the fixed size tbuf buffer in ima_policy_show().
Fixes: 5c7bac9fb2c5 ("IMA: pre-allocate buffer to hold keyrings string")
Fixes: 2b60c0ecedf8 ("IMA: Read keyrings= option from the IMA policy")
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>
Reviewed-by: Nayna Jain <nayna@linux.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
2020-08-11 14:26:20 -05:00
seq_puts ( m , " keyrings= " ) ;
ima_show_rule_opt_list ( m , entry - > keyrings ) ;
2019-12-11 08:47:07 -08:00
seq_puts ( m , " " ) ;
}
2021-01-07 20:07:05 -08:00
if ( entry - > flags & IMA_LABEL ) {
seq_puts ( m , " label= " ) ;
ima_show_rule_opt_list ( m , entry - > label ) ;
seq_puts ( m , " " ) ;
}
2016-06-01 13:14:01 -05:00
if ( entry - > flags & IMA_PCR ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %d " , entry - > pcr ) ;
seq_printf ( m , pt ( Opt_pcr ) , tbuf ) ;
seq_puts ( m , " " ) ;
}
2015-12-02 17:47:56 +02:00
if ( entry - > flags & IMA_FSUUID ) {
2017-06-01 07:00:26 +02:00
seq_printf ( m , " fsuuid=%pU " , & entry - > fsuuid ) ;
2015-12-02 17:47:56 +02:00
seq_puts ( m , " " ) ;
}
if ( entry - > flags & IMA_UID ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %d " , __kuid_val ( entry - > uid ) ) ;
2017-01-27 19:23:01 +03:00
if ( entry - > uid_op = = & uid_gt )
seq_printf ( m , pt ( Opt_uid_gt ) , tbuf ) ;
else if ( entry - > uid_op = = & uid_lt )
seq_printf ( m , pt ( Opt_uid_lt ) , tbuf ) ;
else
seq_printf ( m , pt ( Opt_uid_eq ) , tbuf ) ;
2015-12-02 17:47:56 +02:00
seq_puts ( m , " " ) ;
}
if ( entry - > flags & IMA_EUID ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %d " , __kuid_val ( entry - > uid ) ) ;
2017-01-27 19:23:01 +03:00
if ( entry - > uid_op = = & uid_gt )
seq_printf ( m , pt ( Opt_euid_gt ) , tbuf ) ;
else if ( entry - > uid_op = = & uid_lt )
seq_printf ( m , pt ( Opt_euid_lt ) , tbuf ) ;
else
seq_printf ( m , pt ( Opt_euid_eq ) , tbuf ) ;
2015-12-02 17:47:56 +02:00
seq_puts ( m , " " ) ;
}
2021-10-07 14:03:02 -06:00
if ( entry - > flags & IMA_GID ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %d " , __kgid_val ( entry - > gid ) ) ;
if ( entry - > gid_op = = & gid_gt )
seq_printf ( m , pt ( Opt_gid_gt ) , tbuf ) ;
else if ( entry - > gid_op = = & gid_lt )
seq_printf ( m , pt ( Opt_gid_lt ) , tbuf ) ;
else
seq_printf ( m , pt ( Opt_gid_eq ) , tbuf ) ;
seq_puts ( m , " " ) ;
}
if ( entry - > flags & IMA_EGID ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %d " , __kgid_val ( entry - > gid ) ) ;
if ( entry - > gid_op = = & gid_gt )
seq_printf ( m , pt ( Opt_egid_gt ) , tbuf ) ;
else if ( entry - > gid_op = = & gid_lt )
seq_printf ( m , pt ( Opt_egid_lt ) , tbuf ) ;
else
seq_printf ( m , pt ( Opt_egid_eq ) , tbuf ) ;
seq_puts ( m , " " ) ;
}
2015-12-02 17:47:56 +02:00
if ( entry - > flags & IMA_FOWNER ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %d " , __kuid_val ( entry - > fowner ) ) ;
2017-01-27 19:23:01 +03:00
if ( entry - > fowner_op = = & uid_gt )
seq_printf ( m , pt ( Opt_fowner_gt ) , tbuf ) ;
else if ( entry - > fowner_op = = & uid_lt )
seq_printf ( m , pt ( Opt_fowner_lt ) , tbuf ) ;
else
seq_printf ( m , pt ( Opt_fowner_eq ) , tbuf ) ;
2015-12-02 17:47:56 +02:00
seq_puts ( m , " " ) ;
}
2021-10-07 14:03:02 -06:00
if ( entry - > flags & IMA_FGROUP ) {
snprintf ( tbuf , sizeof ( tbuf ) , " %d " , __kgid_val ( entry - > fgroup ) ) ;
if ( entry - > fgroup_op = = & gid_gt )
seq_printf ( m , pt ( Opt_fgroup_gt ) , tbuf ) ;
else if ( entry - > fgroup_op = = & gid_lt )
seq_printf ( m , pt ( Opt_fgroup_lt ) , tbuf ) ;
else
seq_printf ( m , pt ( Opt_fgroup_eq ) , tbuf ) ;
seq_puts ( m , " " ) ;
}
2021-08-16 08:11:00 +00:00
if ( entry - > flags & IMA_VALIDATE_ALGOS ) {
seq_puts ( m , " appraise_algos= " ) ;
ima_policy_show_appraise_algos ( m , entry - > allowed_algos ) ;
seq_puts ( m , " " ) ;
}
2015-12-02 17:47:56 +02:00
for ( i = 0 ; i < MAX_LSM_RULES ; i + + ) {
if ( entry - > lsm [ i ] . rule ) {
switch ( i ) {
case LSM_OBJ_USER :
seq_printf ( m , pt ( Opt_obj_user ) ,
2020-07-09 01:19:08 -05:00
entry - > lsm [ i ] . args_p ) ;
2015-12-02 17:47:56 +02:00
break ;
case LSM_OBJ_ROLE :
seq_printf ( m , pt ( Opt_obj_role ) ,
2020-07-09 01:19:08 -05:00
entry - > lsm [ i ] . args_p ) ;
2015-12-02 17:47:56 +02:00
break ;
case LSM_OBJ_TYPE :
seq_printf ( m , pt ( Opt_obj_type ) ,
2020-07-09 01:19:08 -05:00
entry - > lsm [ i ] . args_p ) ;
2015-12-02 17:47:56 +02:00
break ;
case LSM_SUBJ_USER :
seq_printf ( m , pt ( Opt_subj_user ) ,
2020-07-09 01:19:08 -05:00
entry - > lsm [ i ] . args_p ) ;
2015-12-02 17:47:56 +02:00
break ;
case LSM_SUBJ_ROLE :
seq_printf ( m , pt ( Opt_subj_role ) ,
2020-07-09 01:19:08 -05:00
entry - > lsm [ i ] . args_p ) ;
2015-12-02 17:47:56 +02:00
break ;
case LSM_SUBJ_TYPE :
seq_printf ( m , pt ( Opt_subj_type ) ,
2020-07-09 01:19:08 -05:00
entry - > lsm [ i ] . args_p ) ;
2015-12-02 17:47:56 +02:00
break ;
}
2020-01-05 09:18:13 +08:00
seq_puts ( m , " " ) ;
2015-12-02 17:47:56 +02:00
}
}
2019-06-19 15:46:11 -07:00
if ( entry - > template )
seq_printf ( m , " template=%s " , entry - > template - > name ) ;
2019-06-27 23:19:28 -03:00
if ( entry - > flags & IMA_DIGSIG_REQUIRED ) {
if ( entry - > flags & IMA_MODSIG_ALLOWED )
seq_puts ( m , " appraise_type=imasig|modsig " ) ;
else
seq_puts ( m , " appraise_type=imasig " ) ;
}
2019-10-30 23:31:32 -04:00
if ( entry - > flags & IMA_CHECK_BLACKLIST )
seq_puts ( m , " appraise_flag=check_blacklist " ) ;
2015-12-02 17:47:56 +02:00
if ( entry - > flags & IMA_PERMIT_DIRECTIO )
seq_puts ( m , " permit_directio " ) ;
rcu_read_unlock ( ) ;
seq_puts ( m , " \n " ) ;
return 0 ;
}
# endif /* CONFIG_IMA_READ_POLICY */
2019-08-19 17:18:01 -07:00
# if defined(CONFIG_IMA_APPRAISE) && defined(CONFIG_INTEGRITY_TRUSTED_KEYRING)
/*
* ima_appraise_signature : whether IMA will appraise a given function using
* an IMA digital signature . This is restricted to cases where the kernel
* has a set of built - in trusted keys in order to avoid an attacker simply
* loading additional keys .
*/
bool ima_appraise_signature ( enum kernel_read_file_id id )
{
struct ima_rule_entry * entry ;
bool found = false ;
enum ima_hooks func ;
2021-10-09 18:38:21 +08:00
struct list_head * ima_rules_tmp ;
2019-08-19 17:18:01 -07:00
if ( id > = READING_MAX_ID )
return false ;
func = read_idmap [ id ] ? : FILE_CHECK ;
rcu_read_lock ( ) ;
2021-10-09 18:38:21 +08:00
ima_rules_tmp = rcu_dereference ( ima_rules ) ;
list_for_each_entry_rcu ( entry , ima_rules_tmp , list ) {
2019-08-19 17:18:01 -07:00
if ( entry - > action ! = APPRAISE )
continue ;
/*
* A generic entry will match , but otherwise require that it
* match the func we ' re looking for
*/
if ( entry - > func & & entry - > func ! = func )
continue ;
/*
* We require this to be a digital signature , not a raw IMA
* hash .
*/
if ( entry - > flags & IMA_DIGSIG_REQUIRED )
found = true ;
/*
* We ' ve found a rule that matches , so break now even if it
* didn ' t require a digital signature - a later rule that does
* won ' t override it , so would be a false positive .
*/
break ;
}
rcu_read_unlock ( ) ;
return found ;
}
# endif /* CONFIG_IMA_APPRAISE && CONFIG_INTEGRITY_TRUSTED_KEYRING */