Skip to contentSkip to conversation

SELinux: Measure state and hash of policy using IMA

PATCH In review Lakshmi Ramasubramanian 4 participants 12 messages Aug 22, 2020 – Aug 31, 2020

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 SELinux: Measure state and hash of policy using IMA

Critical data structures of security modules are currently not measured. Therefore an attestation service, for instance, would not be able to attest whether the security modules are always operating with the policies and configuration that the system administrator had setup. The policies and configuration for the security modules could be tampered with by rogue user mode agents or modified through some inadvertent actions on the system. Measuring such critical data would enable an attestation service to reliably assess the security configuration of the system.

SELinux configuration and policy are some of the critical data for this security module that needs to be measured. This measurement can be used by an attestation service, for instance, to verify if the configuration and policies have been setup correctly and that they haven't been tampered with at runtime.

Measure SELinux configuration, policy capabilities settings, and the hash of the loaded policy by calling the IMA hook ima_measure_critical_data(). Since the size of the loaded policy can be quite large, hash of the policy is measured instead of the entire policy to avoid bloating the IMA log.

Enable early boot measurement for SELinux in IMA since SELinux initializes its state and policy before custom IMA policy is loaded.

Sample measurement of SELinux state and hash of the policy:

10 e32e...5ac3 ima-buf sha256:86e8...4594 selinux-state-1595389364:287899386 696e697469616c697a65643d313b656e61626c65643d313b656e666f7263696e673d303b636865636b72657170726f743d313b6e6574776f726b5f706565725f636f6e74726f6c733d313b6f70656e5f7065726d733d313b657874656e6465645f736f636b65745f636c6173733d313b616c776179735f636865636b5f6e6574776f726b3d303b6367726f75705f7365636c6162656c3d313b6e6e705f6e6f737569645f7472616e736974696f6e3d313b67656e66735f7365636c6162656c5f73796d6c696e6b733d303 10 9e81...0857 ima-buf sha256:4941...68fc selinux-policy-hash-1597335667:462051628 8d1d...1834

To verify the measurement check the following:

Execute the following command to extract the measured data from the IMA log for SELinux configuration (selinux-state).

grep -m 1 "selinux-state" /sys/kernel/security/integrity/ima/ascii_runtime_measurements | cut -d' ' -f 6 | xxd -r -p

The output should be the list of key-value pairs. For example,
initialized=1;enabled=1;enforcing=0;checkreqprot=1;network_peer_controls=1;open_perms=1;extended_socket_class=1;always_check_network=0;cgroup_seclabel=1;nnp_nosuid_transition=1;genfs_seclabel_symlinks=0;

To verify the measured data with the current SELinux state:

=> enabled should be set to 1 if /sys/fs/selinux folder exists,
0 otherwise

For other entries, compare the integer value in the files
=> /sys/fs/selinux/enforce
=> /sys/fs/selinux/checkreqprot
And, each of the policy capabilities files under
=> /sys/fs/selinux/policy_capabilities

For selinux-policy-hash, the hash of SELinux policy is included in the IMA log entry.

To verify the measured data with the current SELinux policy run the following commands and verify the output hash values match.

sha256sum /sys/fs/selinux/policy | cut -d' ' -f 1

grep -m 1 "selinux-policy-hash" /sys/kernel/security/integrity/ima/ascii_runtime_measurements | cut -d' ' -f 6

This patch is dependent on the following patch series:

	https://patchwork.kernel.org/patch/11709527/
	https://patchwork.kernel.org/patch/11730193/
	https://patchwork.kernel.org/patch/11730757/
---
 security/integrity/ima/Kconfig      |   3 +-
 security/selinux/Makefile           |   2 +
 security/selinux/hooks.c            |   1 +
 security/selinux/include/security.h |  15 +++
 security/selinux/measure.c          | 149 ++++++++++++++++++++++++++++
 security/selinux/selinuxfs.c        |   3 +
 security/selinux/ss/services.c      |  71 +++++++++++--
 7 files changed, 233 insertions(+), 11 deletions(-)
 create mode 100644 security/selinux/measure.c
security/integrity/ima/Kconfig +1 -2
index bc2adab7bae2..e4fb1761d64a 100644
@@ -324,8 +324,7 @@ config IMA_MEASURE_ASYMMETRIC_KEYS
 
 config IMA_QUEUE_EARLY_BOOT_DATA
 	bool
-	depends on IMA_MEASURE_ASYMMETRIC_KEYS
-	depends on SYSTEM_TRUSTED_KEYRING
+	depends on SECURITY_SELINUX || (IMA_MEASURE_ASYMMETRIC_KEYS && SYSTEM_TRUSTED_KEYRING)
 	default y
 
 config IMA_SECURE_AND_OR_TRUSTED_BOOT
security/selinux/Makefile +2 -0
index 4d8e0e8adf0b..83d512116341 100644
@@ -16,6 +16,8 @@ selinux-$(CONFIG_NETLABEL) += netlabel.o
 
 selinux-$(CONFIG_SECURITY_INFINIBAND) += ibpkey.o
 
+selinux-$(CONFIG_IMA) += measure.o
+
 ccflags-y := -I$(srctree)/security/selinux -I$(srctree)/security/selinux/include
 
 $(addprefix $(obj)/,$(selinux-y)): $(obj)/flask.h
security/selinux/hooks.c +1 -0
index efa6108b1ce9..5521dfc1900b 100644
@@ -7394,6 +7394,7 @@ int selinux_disable(struct selinux_state *state)
 	}
 
 	selinux_mark_disabled(state);
+	selinux_measure_state(state);
 
 	pr_info("SELinux:  Disabled at runtime.\n");
 
security/selinux/include/security.h +15 -0
index b0e02cfe3ce1..77f42d9b544b 100644
@@ -222,16 +222,31 @@ static inline bool selinux_policycap_genfs_seclabel_symlinks(void)
 	return state->policycap[POLICYDB_CAPABILITY_GENFS_SECLABEL_SYMLINKS];
 }
 
+static inline bool selinux_checkreqprot(const struct selinux_state *state)
+{
+	return READ_ONCE(state->checkreqprot);
+}
+
 int security_mls_enabled(struct selinux_state *state);
 int security_load_policy(struct selinux_state *state,
 			 void *data, size_t len);
 int security_read_policy(struct selinux_state *state,
 			 void **data, size_t *len);
+int security_read_policy_kernel(struct selinux_state *state,
+				void **data, size_t *len);
 size_t security_policydb_len(struct selinux_state *state);
 
 int security_policycap_supported(struct selinux_state *state,
 				 unsigned int req_cap);
 
+#ifdef CONFIG_IMA
+extern void selinux_measure_state(struct selinux_state *selinux_state);
+#else
+static inline void selinux_measure_state(struct selinux_state *selinux_state)
+{
+}
+#endif
+
 #define SEL_VEC_MAX 32
 struct av_decision {
 	u32 allowed;
security/selinux/measure.c new file +149 -0

149 lines not on this page · raw message

security/selinux/selinuxfs.c +3 -0
index 4781314c2510..6d46eaef5c92 100644
@@ -173,6 +173,7 @@ static ssize_t sel_write_enforce(struct file *file, const char __user *buf,
 			from_kuid(&init_user_ns, audit_get_loginuid(current)),
 			audit_get_sessionid(current));
 		enforcing_set(state, new_value);
+		selinux_measure_state(state);
 		if (new_value)
 			avc_ss_reset(state->avc, 0);
 		selnl_notify_setenforce(new_value);
@@ -678,6 +679,8 @@ static ssize_t sel_write_checkreqprot(struct file *file, const char __user *buf,
 
 	fsi->state->checkreqprot = new_value ? 1 : 0;
 	length = count;
+
+	selinux_measure_state(fsi->state);
 out:
 	kfree(page);
 	return length;
security/selinux/ss/services.c +62 -9

111 lines not on this page · raw message

Signature · 1 line
2.28.0
maintainer
#1 lore raw
Quoted 77 lines · Lakshmi Ramasubramanian

Critical data structures of security modules are currently not measured. Therefore an attestation service, for instance, would not be able to attest whether the security modules are always operating with the policies and configuration that the system administrator had setup. The policies and configuration for the security modules could be tampered with by rogue user mode agents or modified through some inadvertent actions on the system. Measuring such critical data would enable an attestation service to reliably assess the security configuration of the system.

SELinux configuration and policy are some of the critical data for this security module that needs to be measured. This measurement can be used by an attestation service, for instance, to verify if the configuration and policies have been setup correctly and that they haven't been tampered with at runtime.

Measure SELinux configuration, policy capabilities settings, and the hash of the loaded policy by calling the IMA hook ima_measure_critical_data(). Since the size of the loaded policy can be quite large, hash of the policy is measured instead of the entire policy to avoid bloating the IMA log.

Enable early boot measurement for SELinux in IMA since SELinux initializes its state and policy before custom IMA policy is loaded.

Sample measurement of SELinux state and hash of the policy:

10 e32e...5ac3 ima-buf sha256:86e8...4594 selinux-state-1595389364:287899386 696e697469616c697a65643d313b656e61626c65643d313b656e666f7263696e673d303b636865636b72657170726f743d313b6e6574776f726b5f706565725f636f6e74726f6c733d313b6f70656e5f7065726d733d313b657874656e6465645f736f636b65745f636c6173733d313b616c776179735f636865636b5f6e6574776f726b3d303b6367726f75705f7365636c6162656c3d313b6e6e705f6e6f737569645f7472616e736974696f6e3d313b67656e66735f7365636c6162656c5f73796d6c696e6b733d303 10 9e81...0857 ima-buf sha256:4941...68fc selinux-policy-hash-1597335667:462051628 8d1d...1834

To verify the measurement check the following:

Execute the following command to extract the measured data from the IMA log for SELinux configuration (selinux-state).

grep -m 1 "selinux-state" /sys/kernel/security/integrity/ima/ascii_runtime_measurements | cut -d' ' -f 6 | xxd -r -p

The output should be the list of key-value pairs. For example,
initialized=1;enabled=1;enforcing=0;checkreqprot=1;network_peer_controls=1;open_perms=1;extended_socket_class=1;always_check_network=0;cgroup_seclabel=1;nnp_nosuid_transition=1;genfs_seclabel_symlinks=0;

To verify the measured data with the current SELinux state:

=> enabled should be set to 1 if /sys/fs/selinux folder exists,
0 otherwise

For other entries, compare the integer value in the files
=> /sys/fs/selinux/enforce
=> /sys/fs/selinux/checkreqprot
And, each of the policy capabilities files under
=> /sys/fs/selinux/policy_capabilities

For selinux-policy-hash, the hash of SELinux policy is included in the IMA log entry.

To verify the measured data with the current SELinux policy run the following commands and verify the output hash values match.

sha256sum /sys/fs/selinux/policy | cut -d' ' -f 1

grep -m 1 "selinux-policy-hash" /sys/kernel/security/integrity/ima/ascii_runtime_measurements | cut -d' ' -f 6

This patch is dependent on the following patch series:

        https://patchwork.kernel.org/patch/11709527/
        https://patchwork.kernel.org/patch/11730193/
        https://patchwork.kernel.org/patch/11730757/

---

security/selinux/ss/services.c +5 -0
+int security_read_policy_kernel(struct selinux_state *state,
+                               void **data, size_t *len)
+{
+       int rc;
+

security/selinux/ss/services.c int security_read_policy(struct selinux_state *state, · line 3811

Quoted 10 lines
security/selinux/ss/services.c +8 -0
+       rc = security_read_policy_len(state, len);
+       if (rc)
+               return rc;
+
+       *data = vmalloc(*len);
+       if (!*data)
+               return -ENOMEM;
 
+       return security_read_selinux_policy(state, data, len);
 }

See the discussion here:
https://lore.kernel.org/selinux/[email protected]/T/#t

In order for this to be safe, you need to ensure that all callers of security_read_policy_kernel() have taken fsi->mutex in selinuxfs and any use of security_read_policy_len() occurs while holding the mutex. Otherwise, the length can change between security_read_policy_len() and security_read_selinux_policy() if policy is reloaded.

#2 lore raw
Quoted 22 lines · Stephen Smalley

On Fri, Aug 21, 2020 at 9:00 PM Lakshmi Ramasubramanian

Quoted 5 lines
security/selinux/ss/services.c +5 -0
+int security_read_policy_kernel(struct selinux_state *state,
+                               void **data, size_t *len)
+{
+       int rc;
+

security/selinux/ss/services.c int security_read_policy(struct selinux_state *state, · line 3811

Quoted 10 lines
security/selinux/ss/services.c +8 -0
+       rc = security_read_policy_len(state, len);
+       if (rc)
+               return rc;
+
+       *data = vmalloc(*len);
+       if (!*data)
+               return -ENOMEM;
 
+       return security_read_selinux_policy(state, data, len);
  }

See the discussion here:
https://lore.kernel.org/selinux/[email protected]/T/#t

Quoted 6 lines

In order for this to be safe, you need to ensure that all callers of security_read_policy_kernel() have taken fsi->mutex in selinuxfs and any use of security_read_policy_len() occurs while holding the mutex. Otherwise, the length can change between security_read_policy_len() and security_read_selinux_policy() if policy is reloaded.

Thanks. I'll make this change.

-lakshmi

#3 lore raw
In reply to Stephen Smalley
Quoted 19 lines · Stephen Smalley
Quoted 5 lines
security/selinux/ss/services.c +5 -0
+int security_read_policy_kernel(struct selinux_state *state,
+                               void **data, size_t *len)
+{
+       int rc;
+

security/selinux/ss/services.c int security_read_policy(struct selinux_state *state, · line 3811

Quoted 10 lines
security/selinux/ss/services.c +8 -0
+       rc = security_read_policy_len(state, len);
+       if (rc)
+               return rc;
+
+       *data = vmalloc(*len);
+       if (!*data)
+               return -ENOMEM;
 
+       return security_read_selinux_policy(state, data, len);
  }

See the discussion here:
https://lore.kernel.org/selinux/[email protected]/T/#t

Quoted 6 lines

In order for this to be safe, you need to ensure that all callers of security_read_policy_kernel() have taken fsi->mutex in selinuxfs and any use of security_read_policy_len() occurs while holding the mutex. Otherwise, the length can change between security_read_policy_len() and security_read_selinux_policy() if policy is reloaded.

"struct selinux_fs_info" is available when calling security_read_policy_kernel() - currently in measure.c. Only "struct selinux_state" is.

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

thanks,
-lakshmi

maintainer
#4 lore raw
Quoted 29 lines · Lakshmi Ramasubramanian
Quoted 19 lines · Stephen Smalley
Quoted 5 lines
security/selinux/ss/services.c +5 -0
+int security_read_policy_kernel(struct selinux_state *state,
+                               void **data, size_t *len)
+{
+       int rc;
+

security/selinux/ss/services.c int security_read_policy(struct selinux_state *state, · line 3811

Quoted 10 lines
security/selinux/ss/services.c +8 -0
+       rc = security_read_policy_len(state, len);
+       if (rc)
+               return rc;
+
+       *data = vmalloc(*len);
+       if (!*data)
+               return -ENOMEM;
 
+       return security_read_selinux_policy(state, data, len);
  }

See the discussion here:
https://lore.kernel.org/selinux/[email protected]/T/#t

Quoted 6 lines

In order for this to be safe, you need to ensure that all callers of security_read_policy_kernel() have taken fsi->mutex in selinuxfs and any use of security_read_policy_len() occurs while holding the mutex. Otherwise, the length can change between security_read_policy_len() and security_read_selinux_policy() if policy is reloaded.

Quoted 5 lines

"struct selinux_fs_info" is available when calling security_read_policy_kernel() - currently in measure.c. Only "struct selinux_state" is.

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

No, I think perhaps we should move the mutex to selinux_state instead of selinux_fs_info. selinux_fs_info has a pointer to selinux_state so it can then use it indirectly. Note that your patches are going to conflict with other ongoing work in the selinux next branch that is refactoring policy load and converting the policy rwlock to RCU.

#5 lore raw
Quoted 35 lines · Stephen Smalley
Quoted 33 lines · Lakshmi Ramasubramanian
Quoted 19 lines · Stephen Smalley
Quoted 5 lines
security/selinux/ss/services.c +5 -0
+int security_read_policy_kernel(struct selinux_state *state,
+                               void **data, size_t *len)
+{
+       int rc;
+

security/selinux/ss/services.c int security_read_policy(struct selinux_state *state, · line 3811

Quoted 10 lines
security/selinux/ss/services.c +8 -0
+       rc = security_read_policy_len(state, len);
+       if (rc)
+               return rc;
+
+       *data = vmalloc(*len);
+       if (!*data)
+               return -ENOMEM;
 
+       return security_read_selinux_policy(state, data, len);
  }

See the discussion here:
https://lore.kernel.org/selinux/[email protected]/T/#t

Quoted 6 lines

In order for this to be safe, you need to ensure that all callers of security_read_policy_kernel() have taken fsi->mutex in selinuxfs and any use of security_read_policy_len() occurs while holding the mutex. Otherwise, the length can change between security_read_policy_len() and security_read_selinux_policy() if policy is reloaded.

"struct selinux_fs_info" is available when calling security_read_policy_kernel() - currently in measure.c. Only "struct selinux_state" is.

Quoted 7 lines
Quoted 1 line

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

No, I think perhaps we should move the mutex to selinux_state instead of selinux_fs_info. selinux_fs_info has a pointer to selinux_state so it can then use it indirectly. Note that your patches are going to conflict with other ongoing work in the selinux next branch that is refactoring policy load and converting the policy rwlock to RCU.

Yeah, and I'm experimenting with a patch on top of Stephen's RCU work that would allow you to do this in a straightforward way without even messing with the fsi->mutex. My patch may or may not be eventually committed, but either way I'd recommend holding off on this for a while until the dust settles around the RCU conversion.

Signature · 3 lines
Ondrej Mosnacek
Software Engineer, Platform Security - SELinux kernel
Red Hat, Inc.
#6 lore raw
Quoted 45 lines · Ondrej Mosnacek
Quoted 35 lines · Stephen Smalley
Quoted 33 lines · Lakshmi Ramasubramanian
Quoted 19 lines · Stephen Smalley
Quoted 5 lines
security/selinux/ss/services.c +5 -0
+int security_read_policy_kernel(struct selinux_state *state,
+                               void **data, size_t *len)
+{
+       int rc;
+

security/selinux/ss/services.c int security_read_policy(struct selinux_state *state, · line 3811

Quoted 10 lines
security/selinux/ss/services.c +8 -0
+       rc = security_read_policy_len(state, len);
+       if (rc)
+               return rc;
+
+       *data = vmalloc(*len);
+       if (!*data)
+               return -ENOMEM;
 
+       return security_read_selinux_policy(state, data, len);
   }

See the discussion here:
https://lore.kernel.org/selinux/[email protected]/T/#t

Quoted 6 lines

In order for this to be safe, you need to ensure that all callers of security_read_policy_kernel() have taken fsi->mutex in selinuxfs and any use of security_read_policy_len() occurs while holding the mutex. Otherwise, the length can change between security_read_policy_len() and security_read_selinux_policy() if policy is reloaded.

"struct selinux_fs_info" is available when calling security_read_policy_kernel() - currently in measure.c. Only "struct selinux_state" is.

Quoted 7 lines
Quoted 1 line

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

No, I think perhaps we should move the mutex to selinux_state instead of selinux_fs_info. selinux_fs_info has a pointer to selinux_state so it can then use it indirectly. Note that your patches are going to conflict with other ongoing work in the selinux next branch that is refactoring policy load and converting the policy rwlock to RCU.

Quoted 6 lines

Yeah, and I'm experimenting with a patch on top of Stephen's RCU work that would allow you to do this in a straightforward way without even messing with the fsi->mutex. My patch may or may not be eventually committed, but either way I'd recommend holding off on this for a while until the dust settles around the RCU conversion.

I can make the SELinux\IMA changes in "selinux next branch" taking dependencies on Stephen's patches + relevant IMA patches.

Could you please let me know the URL to the "selinux next branch"?

thanks,
-lakshmi

maintainer
#7 lore raw
Quoted 6 lines · Lakshmi Ramasubramanian
Quoted 5 lines · Ondrej Mosnacek
Quoted 3 lines · Stephen Smalley
Quoted 1 line · Lakshmi Ramasubramanian

On 8/24/20 7:00 AM, Stephen Smalley wrote:

...

Quoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 1 line

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

No, I think perhaps we should move the mutex to selinux_state instead of selinux_fs_info. selinux_fs_info has a pointer to selinux_state so it can then use it indirectly. Note that your patches are going to conflict with other ongoing work in the selinux next branch that is refactoring policy load and converting the policy rwlock to RCU.

Quoted 8 lines
Quoted 5 lines

Yeah, and I'm experimenting with a patch on top of Stephen's RCU work that would allow you to do this in a straightforward way without even messing with the fsi->mutex. My patch may or may not be eventually committed, but either way I'd recommend holding off on this for a while until the dust settles around the RCU conversion.

I can make the SELinux\IMA changes in "selinux next branch" taking dependencies on Stephen's patches + relevant IMA patches.

I know it can be frustrating to hear what I'm about to say, but the best option is probably just to wait a little to let things settle in the SELinux -next branch. There is a lot of stuff going on right now with patches flooding in (at least "flooding" from a SELinux kernel development perspective) and we/I've haven't gotten through all of them yet.

Quoted 1 line

Could you please let me know the URL to the "selinux next branch"?

git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux.git next

Signature · 2 lines
paul moore
www.paul-moore.com
#8 lore raw

On 8/24/20 3:18 PM, Paul Moore wrote:

Hi Paul,

Quoted 17 lines
Quoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 1 line

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

No, I think perhaps we should move the mutex to selinux_state instead of selinux_fs_info. selinux_fs_info has a pointer to selinux_state so it can then use it indirectly. Note that your patches are going to conflict with other ongoing work in the selinux next branch that is refactoring policy load and converting the policy rwlock to RCU.

Quoted 8 lines
Quoted 5 lines

Yeah, and I'm experimenting with a patch on top of Stephen's RCU work that would allow you to do this in a straightforward way without even messing with the fsi->mutex. My patch may or may not be eventually committed, but either way I'd recommend holding off on this for a while until the dust settles around the RCU conversion.

I can make the SELinux\IMA changes in "selinux next branch" taking dependencies on Stephen's patches + relevant IMA patches.

Quoted 7 lines

I know it can be frustrating to hear what I'm about to say, but the best option is probably just to wait a little to let things settle in the SELinux -next branch. There is a lot of stuff going on right now with patches flooding in (at least "flooding" from a SELinux kernel development perspective) and we/I've haven't gotten through all of them yet.

Could you please let me know when the current set of changes in SELinux next branch would be completed and be ready to take new changes?

I mean, roughly - would it be a month from now or you expect that to take longer?

thanks,
-lakshmi

maintainer
#9 lore raw
Quoted 30 lines · Lakshmi Ramasubramanian

On 8/24/20 3:18 PM, Paul Moore wrote:

Hi Paul,

Quoted 17 lines
Quoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 1 line

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

No, I think perhaps we should move the mutex to selinux_state instead of selinux_fs_info. selinux_fs_info has a pointer to selinux_state so it can then use it indirectly. Note that your patches are going to conflict with other ongoing work in the selinux next branch that is refactoring policy load and converting the policy rwlock to RCU.

Quoted 8 lines
Quoted 5 lines

Yeah, and I'm experimenting with a patch on top of Stephen's RCU work that would allow you to do this in a straightforward way without even messing with the fsi->mutex. My patch may or may not be eventually committed, but either way I'd recommend holding off on this for a while until the dust settles around the RCU conversion.

I can make the SELinux\IMA changes in "selinux next branch" taking dependencies on Stephen's patches + relevant IMA patches.

Quoted 7 lines

I know it can be frustrating to hear what I'm about to say, but the best option is probably just to wait a little to let things settle in the SELinux -next branch. There is a lot of stuff going on right now with patches flooding in (at least "flooding" from a SELinux kernel development perspective) and we/I've haven't gotten through all of them yet.

Quoted 5 lines

Could you please let me know when the current set of changes in SELinux next branch would be completed and be ready to take new changes?

I mean, roughly - would it be a month from now or you expect that to take longer?

I can't speak for Paul but I would expect it to be sooner rather than later. Ondrej has some follow ups on top of my policy rcu conversion but then it should be good to go.

maintainer
#10 lore raw
Quoted 36 lines · Stephen Smalley
Quoted 33 lines · Lakshmi Ramasubramanian

On 8/24/20 3:18 PM, Paul Moore wrote:

Hi Paul,

Quoted 17 lines
Quoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 1 line

Is Ondrej's re-try approach I need to use to workaround policy reload issue?

No, I think perhaps we should move the mutex to selinux_state instead of selinux_fs_info. selinux_fs_info has a pointer to selinux_state so it can then use it indirectly. Note that your patches are going to conflict with other ongoing work in the selinux next branch that is refactoring policy load and converting the policy rwlock to RCU.

Quoted 8 lines
Quoted 5 lines

Yeah, and I'm experimenting with a patch on top of Stephen's RCU work that would allow you to do this in a straightforward way without even messing with the fsi->mutex. My patch may or may not be eventually committed, but either way I'd recommend holding off on this for a while until the dust settles around the RCU conversion.

I can make the SELinux\IMA changes in "selinux next branch" taking dependencies on Stephen's patches + relevant IMA patches.

Quoted 7 lines

I know it can be frustrating to hear what I'm about to say, but the best option is probably just to wait a little to let things settle in the SELinux -next branch. There is a lot of stuff going on right now with patches flooding in (at least "flooding" from a SELinux kernel development perspective) and we/I've haven't gotten through all of them yet.

Could you please let me know when the current set of changes in SELinux next branch would be completed and be ready to take new changes?

Quoted 6 lines
Quoted 2 lines

I mean, roughly - would it be a month from now or you expect that to take longer?

I can't speak for Paul but I would expect it to be sooner rather than later. Ondrej has some follow ups on top of my policy rcu conversion but then it should be good to go.

I think the major changes are now merged although there are still a couple of changes coming from Ondrej that could affect your code. For your purposes, the important things to note are:

1) The mutex has moved from selinux_fs_info to selinux_state and is now named policy_mutex. You will need to take it around your call to security_read_policy_kernel().

2) security_policydb_len() was removed and security_read_policy() just directly reads the policydb len. You can do the same from your security_read_policy_kernel() variant.

3) Ondrej has a pending change to move the policycap[] array from selinux_state to selinux_policy so that it can be atomically updated with the policy.

4) Ondrej has a pending change to eliminate the separate initialized boolean from selinux_state and just test whether selinux_state.policy is non-NULL but as long as you are using selinux_initialized() to test, your code should be unaffected.

#11 lore raw
Quoted 24 lines · Stephen Smalley
Quoted 11 lines
Quoted 7 lines
Quoted 1 line

Could you please let me know when the current set of changes in SELinux next branch would be completed and be ready to take new changes?

I mean, roughly - would it be a month from now or you expect that to take longer?

I can't speak for Paul but I would expect it to be sooner rather than later. Ondrej has some follow ups on top of my policy rcu conversion but then it should be good to go.

I think the major changes are now merged although there are still a couple of changes coming from Ondrej that could affect your code. For your purposes, the important things to note are:

1) The mutex has moved from selinux_fs_info to selinux_state and is now named policy_mutex. You will need to take it around your call to security_read_policy_kernel().

2) security_policydb_len() was removed and security_read_policy() just directly reads the policydb len. You can do the same from your security_read_policy_kernel() variant.

Quoted 9 lines

3) Ondrej has a pending change to move the policycap[] array from selinux_state to selinux_policy so that it can be atomically updated with the policy.

4) Ondrej has a pending change to eliminate the separate initialized boolean from selinux_state and just test whether selinux_state.policy is non-NULL but as long as you are using selinux_initialized() to test, your code should be unaffected.

Thanks a lot for the update Stephen.

I will start updating the IMA measurement changes in selinux next branch. Will post the patches this week.

-lakshmi