[PATCH 0/3] kexec: limit kexec_load syscall

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

IMA-appraisal is mostly being used in the embedded or single purpose
closed system environments.  In these environments, both the Kconfig
options and the userspace tools can be modified appropriately to limit
syscalls.  For stock kernels, userspace applications need to continue to
work with older kernels as well as with newer kernels.

In this environment, the customer needs the ability to define a system
wide IMA runtime policy, such as requiring all kexec'ed images (or
firmware) to be signed, without being dependent on either the Kconfig
options or the userspace tools.

This patch set allows the customer to define a policy which requires
kexec'ed kernels to be signed.

Mimi Zohar (3):
  ima: based on the "secure_boot" policy limit syscalls
  kexec: call LSM hook for kexec_load syscall
  ima: based on policy require signed kexec kernel images

 include/linux/security.h            |  6 ++++++
 kernel/kexec.c                      | 11 +++++++++++
 security/integrity/ima/ima.h        |  1 +
 security/integrity/ima/ima_main.c   |  9 +++++++++
 security/integrity/ima/ima_policy.c | 27 ++++++++++++++++++++-------
 security/security.c                 |  6 ++++++
 6 files changed, 53 insertions(+), 7 deletions(-)

-- 
2.7.5

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

IMA-appraisal is mostly being used in the embedded or single purpose
closed system environments.  In these environments, both the Kconfig
options and the userspace tools can be modified appropriately to limit
syscalls.  For stock kernels, userspace applications need to continue to
work with older kernels as well as with newer kernels.

In this environment, the customer needs the ability to define a system
wide IMA runtime policy, such as requiring all kexec'ed images (or
firmware) to be signed, without being dependent on either the Kconfig
options or the userspace tools.

This patch set allows the customer to define a policy which requires
kexec'ed kernels to be signed.

Mimi Zohar (3):
  ima: based on the "secure_boot" policy limit syscalls
  kexec: call LSM hook for kexec_load syscall
  ima: based on policy require signed kexec kernel images

 include/linux/security.h            |  6 ++++++
 kernel/kexec.c                      | 11 +++++++++++
 security/integrity/ima/ima.h        |  1 +
 security/integrity/ima/ima_main.c   |  9 +++++++++
 security/integrity/ima/ima_policy.c | 27 ++++++++++++++++++++-------
 security/security.c                 |  6 ++++++
 6 files changed, 53 insertions(+), 7 deletions(-)

-- 
2.7.5


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

IMA-appraisal is mostly being used in the embedded or single purpose
closed system environments.  In these environments, both the Kconfig
options and the userspace tools can be modified appropriately to limit
syscalls.  For stock kernels, userspace applications need to continue to
work with older kernels as well as with newer kernels.

In this environment, the customer needs the ability to define a system
wide IMA runtime policy, such as requiring all kexec'ed images (or
firmware) to be signed, without being dependent on either the Kconfig
options or the userspace tools.

This patch set allows the customer to define a policy which requires
kexec'ed kernels to be signed.

Mimi Zohar (3):
  ima: based on the "secure_boot" policy limit syscalls
  kexec: call LSM hook for kexec_load syscall
  ima: based on policy require signed kexec kernel images

 include/linux/security.h            |  6 ++++++
 kernel/kexec.c                      | 11 +++++++++++
 security/integrity/ima/ima.h        |  1 +
 security/integrity/ima/ima_main.c   |  9 +++++++++
 security/integrity/ima/ima_policy.c | 27 ++++++++++++++++++++-------
 security/security.c                 |  6 ++++++
 6 files changed, 53 insertions(+), 7 deletions(-)

-- 
2.7.5

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 1/3] ima: based on the "secure_boot" policy limit syscalls

[Mimi Zohar]

The builtin "secure_boot" policy adds IMA appraisal rules requiring kernel
modules (finit_module syscall), direct firmware load, kexec kernel image
(kexec_file_load syscall), and the IMA policy to be signed, but did not
prevent the other syscalls/methods from working.  Loading an equivalent
custom policy containing these same rules would have prevented the other
syscalls/methods from working.

This patch refactors the code to load custom policies, defining a new
function named ima_appraise_flag().  The new function is called either
when loading the builtin "secure_boot" or custom policies.

Fixes: 503ceaef8e2e ("ima: define a set of appraisal rules requiring file signatures")
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
---
 security/integrity/ima/ima_policy.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index 03cbba423e59..df3e45878a87 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -440,6 +440,17 @@ void ima_update_policy_flag(void)
 		ima_policy_flag &= ~IMA_APPRAISE;
 }
 
+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;
+	return 0;
+}
+
 /**
  * ima_init_policy - initialize the default measure rules.
  *
@@ -478,9 +489,12 @@ void __init ima_init_policy(void)
 	 * Insert the appraise rules requiring file signatures, prior to
 	 * any other appraise rules.
 	 */
-	for (i = 0; i < secure_boot_entries; i++)
+	for (i = 0; i < secure_boot_entries; i++) {
 		list_add_tail(&secure_boot_rules[i].list,
 			      &ima_default_rules);
+		temp_ima_appraise |=
+		    ima_appraise_flag(secure_boot_rules[i].func);
+	}
 
 	for (i = 0; i < appraise_entries; i++) {
 		list_add_tail(&default_appraise_rules[i].list,
@@ -934,12 +948,9 @@ static int ima_parse_rule(char *rule, struct ima_rule_entry *entry)
 	}
 	if (!result && (entry->action == UNKNOWN))
 		result = -EINVAL;
-	else if (entry->func == MODULE_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_MODULES;
-	else if (entry->func == FIRMWARE_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_FIRMWARE;
-	else if (entry->func == POLICY_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_POLICY;
+	else if (entry->action == APPRAISE)
+		temp_ima_appraise |= ima_appraise_flag(entry->func);
+
 	audit_log_format(ab, "res=%d", !result);
 	audit_log_end(ab);
 	return result;
-- 
2.7.5

[PATCH 1/3] ima: based on the "secure_boot" policy limit syscalls

[Mimi Zohar]

The builtin "secure_boot" policy adds IMA appraisal rules requiring kernel
modules (finit_module syscall), direct firmware load, kexec kernel image
(kexec_file_load syscall), and the IMA policy to be signed, but did not
prevent the other syscalls/methods from working.  Loading an equivalent
custom policy containing these same rules would have prevented the other
syscalls/methods from working.

This patch refactors the code to load custom policies, defining a new
function named ima_appraise_flag().  The new function is called either
when loading the builtin "secure_boot" or custom policies.

Fixes: 503ceaef8e2e ("ima: define a set of appraisal rules requiring file signatures")
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
---
 security/integrity/ima/ima_policy.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index 03cbba423e59..df3e45878a87 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -440,6 +440,17 @@ void ima_update_policy_flag(void)
 		ima_policy_flag &= ~IMA_APPRAISE;
 }
 
+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;
+	return 0;
+}
+
 /**
  * ima_init_policy - initialize the default measure rules.
  *
@@ -478,9 +489,12 @@ void __init ima_init_policy(void)
 	 * Insert the appraise rules requiring file signatures, prior to
 	 * any other appraise rules.
 	 */
-	for (i = 0; i < secure_boot_entries; i++)
+	for (i = 0; i < secure_boot_entries; i++) {
 		list_add_tail(&secure_boot_rules[i].list,
 			      &ima_default_rules);
+		temp_ima_appraise |=
+		    ima_appraise_flag(secure_boot_rules[i].func);
+	}
 
 	for (i = 0; i < appraise_entries; i++) {
 		list_add_tail(&default_appraise_rules[i].list,
@@ -934,12 +948,9 @@ static int ima_parse_rule(char *rule, struct ima_rule_entry *entry)
 	}
 	if (!result && (entry->action == UNKNOWN))
 		result = -EINVAL;
-	else if (entry->func == MODULE_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_MODULES;
-	else if (entry->func == FIRMWARE_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_FIRMWARE;
-	else if (entry->func == POLICY_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_POLICY;
+	else if (entry->action == APPRAISE)
+		temp_ima_appraise |= ima_appraise_flag(entry->func);
+
 	audit_log_format(ab, "res=%d", !result);
 	audit_log_end(ab);
 	return result;
-- 
2.7.5


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 1/3] ima: based on the "secure_boot" policy limit syscalls

[Mimi Zohar]

The builtin "secure_boot" policy adds IMA appraisal rules requiring kernel
modules (finit_module syscall), direct firmware load, kexec kernel image
(kexec_file_load syscall), and the IMA policy to be signed, but did not
prevent the other syscalls/methods from working.  Loading an equivalent
custom policy containing these same rules would have prevented the other
syscalls/methods from working.

This patch refactors the code to load custom policies, defining a new
function named ima_appraise_flag().  The new function is called either
when loading the builtin "secure_boot" or custom policies.

Fixes: 503ceaef8e2e ("ima: define a set of appraisal rules requiring file signatures")
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
---
 security/integrity/ima/ima_policy.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index 03cbba423e59..df3e45878a87 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -440,6 +440,17 @@ void ima_update_policy_flag(void)
 		ima_policy_flag &= ~IMA_APPRAISE;
 }
 
+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;
+	return 0;
+}
+
 /**
  * ima_init_policy - initialize the default measure rules.
  *
@@ -478,9 +489,12 @@ void __init ima_init_policy(void)
 	 * Insert the appraise rules requiring file signatures, prior to
 	 * any other appraise rules.
 	 */
-	for (i = 0; i < secure_boot_entries; i++)
+	for (i = 0; i < secure_boot_entries; i++) {
 		list_add_tail(&secure_boot_rules[i].list,
 			      &ima_default_rules);
+		temp_ima_appraise |=
+		    ima_appraise_flag(secure_boot_rules[i].func);
+	}
 
 	for (i = 0; i < appraise_entries; i++) {
 		list_add_tail(&default_appraise_rules[i].list,
@@ -934,12 +948,9 @@ static int ima_parse_rule(char *rule, struct ima_rule_entry *entry)
 	}
 	if (!result && (entry->action == UNKNOWN))
 		result = -EINVAL;
-	else if (entry->func == MODULE_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_MODULES;
-	else if (entry->func == FIRMWARE_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_FIRMWARE;
-	else if (entry->func == POLICY_CHECK)
-		temp_ima_appraise |= IMA_APPRAISE_POLICY;
+	else if (entry->action == APPRAISE)
+		temp_ima_appraise |= ima_appraise_flag(entry->func);
+
 	audit_log_format(ab, "res=%d", !result);
 	audit_log_end(ab);
 	return result;
-- 
2.7.5

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 2/3] kexec: call LSM hook for kexec_load syscall

[Mimi Zohar]

In order for LSMs and IMA-appraisal to differentiate between the
kexec_load and kexec_file_load_syscalls, an LSM call needs to be added
to the original kexec_load syscall.  From a technical perspective there
is no need for defining a new LSM hook, as the existing
security_kernel_kexec_load() works just fine.  However, the name is
confusing.  For this reason, instead of defining a new LSM hook, this
patch defines security_kexec_load() as a wrapper for the existing LSM
security_kernel_file_read() hook.

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Matthew Garrett <mjg59@google.com>
Cc: Casey Schaufler <casey@schaufler-ca.com>

Changelog v1:
- Define and call security_kexec_load(), a wrapper for
security_kernel_read_file().
---
 include/linux/security.h |  6 ++++++
 kernel/kexec.c           | 11 +++++++++++
 security/security.c      |  6 ++++++
 3 files changed, 23 insertions(+)

diff --git a/include/linux/security.h b/include/linux/security.h
index 63030c85ee19..26f6d85903ed 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -323,6 +323,7 @@ int security_kernel_module_request(char *kmod_name);
 int security_kernel_read_file(struct file *file, enum kernel_read_file_id id);
 int security_kernel_post_read_file(struct file *file, char *buf, loff_t size,
 				   enum kernel_read_file_id id);
+int security_kexec_load(void);
 int security_task_fix_setuid(struct cred *new, const struct cred *old,
 			     int flags);
 int security_task_setpgid(struct task_struct *p, pid_t pgid);
@@ -922,6 +923,11 @@ static inline int security_kernel_post_read_file(struct file *file,
 	return 0;
 }
 
+static inline int security_kexec_load(void)
+{
+	return 0;
+}
+
 static inline int security_task_fix_setuid(struct cred *new,
 					   const struct cred *old,
 					   int flags)
diff --git a/kernel/kexec.c b/kernel/kexec.c
index aed8fb2564b3..6b44b0e9a60b 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -11,6 +11,7 @@
 #include <linux/capability.h>
 #include <linux/mm.h>
 #include <linux/file.h>
+#include <linux/security.h>
 #include <linux/kexec.h>
 #include <linux/mutex.h>
 #include <linux/list.h>
@@ -195,11 +196,21 @@ static int do_kexec_load(unsigned long entry, unsigned long nr_segments,
 static inline int kexec_load_check(unsigned long nr_segments,
 				   unsigned long flags)
 {
+	int result;
+
 	/* We only trust the superuser with rebooting the system. */
 	if (!capable(CAP_SYS_BOOT) || kexec_load_disabled)
 		return -EPERM;
 
 	/*
+	 * Allow LSMs and IMA to differentiate between kexec_load and
+	 * kexec_file_load syscalls.
+	 */
+	result = security_kexec_load();
+	if (result < 0)
+		return result;
+
+	/*
 	 * Verify we have a legal set of flags
 	 * This leaves us room for future extensions.
 	 */
diff --git a/security/security.c b/security/security.c
index 68f46d849abe..0f3390000156 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1044,6 +1044,12 @@ int security_kernel_read_file(struct file *file, enum kernel_read_file_id id)
 }
 EXPORT_SYMBOL_GPL(security_kernel_read_file);
 
+int security_kexec_load()
+{
+	return security_kernel_read_file(NULL, READING_KEXEC_IMAGE);
+}
+EXPORT_SYMBOL_GPL(security_kexec_load);
+
 int security_kernel_post_read_file(struct file *file, char *buf, loff_t size,
 				   enum kernel_read_file_id id)
 {
-- 
2.7.5

[PATCH 2/3] kexec: call LSM hook for kexec_load syscall

[Mimi Zohar]

In order for LSMs and IMA-appraisal to differentiate between the
kexec_load and kexec_file_load_syscalls, an LSM call needs to be added
to the original kexec_load syscall.  From a technical perspective there
is no need for defining a new LSM hook, as the existing
security_kernel_kexec_load() works just fine.  However, the name is
confusing.  For this reason, instead of defining a new LSM hook, this
patch defines security_kexec_load() as a wrapper for the existing LSM
security_kernel_file_read() hook.

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Matthew Garrett <mjg59@google.com>
Cc: Casey Schaufler <casey@schaufler-ca.com>

Changelog v1:
- Define and call security_kexec_load(), a wrapper for
security_kernel_read_file().
---
 include/linux/security.h |  6 ++++++
 kernel/kexec.c           | 11 +++++++++++
 security/security.c      |  6 ++++++
 3 files changed, 23 insertions(+)

diff --git a/include/linux/security.h b/include/linux/security.h
index 63030c85ee19..26f6d85903ed 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -323,6 +323,7 @@ int security_kernel_module_request(char *kmod_name);
 int security_kernel_read_file(struct file *file, enum kernel_read_file_id id);
 int security_kernel_post_read_file(struct file *file, char *buf, loff_t size,
 				   enum kernel_read_file_id id);
+int security_kexec_load(void);
 int security_task_fix_setuid(struct cred *new, const struct cred *old,
 			     int flags);
 int security_task_setpgid(struct task_struct *p, pid_t pgid);
@@ -922,6 +923,11 @@ static inline int security_kernel_post_read_file(struct file *file,
 	return 0;
 }
 
+static inline int security_kexec_load(void)
+{
+	return 0;
+}
+
 static inline int security_task_fix_setuid(struct cred *new,
 					   const struct cred *old,
 					   int flags)
diff --git a/kernel/kexec.c b/kernel/kexec.c
index aed8fb2564b3..6b44b0e9a60b 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -11,6 +11,7 @@
 #include <linux/capability.h>
 #include <linux/mm.h>
 #include <linux/file.h>
+#include <linux/security.h>
 #include <linux/kexec.h>
 #include <linux/mutex.h>
 #include <linux/list.h>
@@ -195,11 +196,21 @@ static int do_kexec_load(unsigned long entry, unsigned long nr_segments,
 static inline int kexec_load_check(unsigned long nr_segments,
 				   unsigned long flags)
 {
+	int result;
+
 	/* We only trust the superuser with rebooting the system. */
 	if (!capable(CAP_SYS_BOOT) || kexec_load_disabled)
 		return -EPERM;
 
 	/*
+	 * Allow LSMs and IMA to differentiate between kexec_load and
+	 * kexec_file_load syscalls.
+	 */
+	result = security_kexec_load();
+	if (result < 0)
+		return result;
+
+	/*
 	 * Verify we have a legal set of flags
 	 * This leaves us room for future extensions.
 	 */
diff --git a/security/security.c b/security/security.c
index 68f46d849abe..0f3390000156 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1044,6 +1044,12 @@ int security_kernel_read_file(struct file *file, enum kernel_read_file_id id)
 }
 EXPORT_SYMBOL_GPL(security_kernel_read_file);
 
+int security_kexec_load()
+{
+	return security_kernel_read_file(NULL, READING_KEXEC_IMAGE);
+}
+EXPORT_SYMBOL_GPL(security_kexec_load);
+
 int security_kernel_post_read_file(struct file *file, char *buf, loff_t size,
 				   enum kernel_read_file_id id)
 {
-- 
2.7.5


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 2/3] kexec: call LSM hook for kexec_load syscall

[Mimi Zohar]

In order for LSMs and IMA-appraisal to differentiate between the
kexec_load and kexec_file_load_syscalls, an LSM call needs to be added
to the original kexec_load syscall.  From a technical perspective there
is no need for defining a new LSM hook, as the existing
security_kernel_kexec_load() works just fine.  However, the name is
confusing.  For this reason, instead of defining a new LSM hook, this
patch defines security_kexec_load() as a wrapper for the existing LSM
security_kernel_file_read() hook.

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Matthew Garrett <mjg59@google.com>
Cc: Casey Schaufler <casey@schaufler-ca.com>

Changelog v1:
- Define and call security_kexec_load(), a wrapper for
security_kernel_read_file().
---
 include/linux/security.h |  6 ++++++
 kernel/kexec.c           | 11 +++++++++++
 security/security.c      |  6 ++++++
 3 files changed, 23 insertions(+)

diff --git a/include/linux/security.h b/include/linux/security.h
index 63030c85ee19..26f6d85903ed 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -323,6 +323,7 @@ int security_kernel_module_request(char *kmod_name);
 int security_kernel_read_file(struct file *file, enum kernel_read_file_id id);
 int security_kernel_post_read_file(struct file *file, char *buf, loff_t size,
 				   enum kernel_read_file_id id);
+int security_kexec_load(void);
 int security_task_fix_setuid(struct cred *new, const struct cred *old,
 			     int flags);
 int security_task_setpgid(struct task_struct *p, pid_t pgid);
@@ -922,6 +923,11 @@ static inline int security_kernel_post_read_file(struct file *file,
 	return 0;
 }
 
+static inline int security_kexec_load(void)
+{
+	return 0;
+}
+
 static inline int security_task_fix_setuid(struct cred *new,
 					   const struct cred *old,
 					   int flags)
diff --git a/kernel/kexec.c b/kernel/kexec.c
index aed8fb2564b3..6b44b0e9a60b 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -11,6 +11,7 @@
 #include <linux/capability.h>
 #include <linux/mm.h>
 #include <linux/file.h>
+#include <linux/security.h>
 #include <linux/kexec.h>
 #include <linux/mutex.h>
 #include <linux/list.h>
@@ -195,11 +196,21 @@ static int do_kexec_load(unsigned long entry, unsigned long nr_segments,
 static inline int kexec_load_check(unsigned long nr_segments,
 				   unsigned long flags)
 {
+	int result;
+
 	/* We only trust the superuser with rebooting the system. */
 	if (!capable(CAP_SYS_BOOT) || kexec_load_disabled)
 		return -EPERM;
 
 	/*
+	 * Allow LSMs and IMA to differentiate between kexec_load and
+	 * kexec_file_load syscalls.
+	 */
+	result = security_kexec_load();
+	if (result < 0)
+		return result;
+
+	/*
 	 * Verify we have a legal set of flags
 	 * This leaves us room for future extensions.
 	 */
diff --git a/security/security.c b/security/security.c
index 68f46d849abe..0f3390000156 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1044,6 +1044,12 @@ int security_kernel_read_file(struct file *file, enum kernel_read_file_id id)
 }
 EXPORT_SYMBOL_GPL(security_kernel_read_file);
 
+int security_kexec_load()
+{
+	return security_kernel_read_file(NULL, READING_KEXEC_IMAGE);
+}
+EXPORT_SYMBOL_GPL(security_kexec_load);
+
 int security_kernel_post_read_file(struct file *file, char *buf, loff_t size,
 				   enum kernel_read_file_id id)
 {
-- 
2.7.5

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info@ http://vger.kernel.org/majordomo-info.html

[PATCH 3/3] ima: based on policy require signed kexec kernel images

[Mimi Zohar]

The original kexec_load syscall can not verify file signatures.  This
patch differentiates between the kexec_load and kexec_file_load
syscalls.

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Matthew Garrett <mjg59@google.com>
---
 security/integrity/ima/ima.h        | 1 +
 security/integrity/ima/ima_main.c   | 9 +++++++++
 security/integrity/ima/ima_policy.c | 2 ++
 3 files changed, 12 insertions(+)

diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h
index 35fe91aa1fc9..03c9c37ee345 100644
--- a/security/integrity/ima/ima.h
+++ b/security/integrity/ima/ima.h
@@ -233,6 +233,7 @@ int ima_policy_show(struct seq_file *m, void *v);
 #define IMA_APPRAISE_MODULES	0x08
 #define IMA_APPRAISE_FIRMWARE	0x10
 #define IMA_APPRAISE_POLICY	0x20
+#define IMA_APPRAISE_KEXEC	0x40
 
 #ifdef CONFIG_IMA_APPRAISE
 int ima_appraise_measurement(enum ima_hooks func,
diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
index 74d0bd7e76d7..754ece08e1c6 100644
--- a/security/integrity/ima/ima_main.c
+++ b/security/integrity/ima/ima_main.c
@@ -444,6 +444,15 @@ int ima_read_file(struct file *file, enum kernel_read_file_id read_id)
 		}
 		return 0;	/* We rely on module signature checking */
 	}
+
+	if (!file && read_id == READING_KEXEC_IMAGE) {
+		if ((ima_appraise & IMA_APPRAISE_KEXEC) &&
+		    (ima_appraise & IMA_APPRAISE_ENFORCE)) {
+			pr_err("impossible to appraise a kernel image without a file descriptor; try using kexec_file syscall.\n");
+			return -EACCES;	/* INTEGRITY_UNKNOWN */
+		}
+		return 0;
+	}
 	return 0;
 }
 
diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index df3e45878a87..a9e96a884867 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -448,6 +448,8 @@ static int ima_appraise_flag(enum ima_hooks func)
 		return IMA_APPRAISE_FIRMWARE;
 	else if (func == POLICY_CHECK)
 		return IMA_APPRAISE_POLICY;
+	else if (func == KEXEC_KERNEL_CHECK)
+		return IMA_APPRAISE_KEXEC;
 	return 0;
 }
 
-- 
2.7.5

[PATCH 3/3] ima: based on policy require signed kexec kernel images

[Mimi Zohar]

The original kexec_load syscall can not verify file signatures.  This
patch differentiates between the kexec_load and kexec_file_load
syscalls.

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Matthew Garrett <mjg59@google.com>
---
 security/integrity/ima/ima.h        | 1 +
 security/integrity/ima/ima_main.c   | 9 +++++++++
 security/integrity/ima/ima_policy.c | 2 ++
 3 files changed, 12 insertions(+)

diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h
index 35fe91aa1fc9..03c9c37ee345 100644
--- a/security/integrity/ima/ima.h
+++ b/security/integrity/ima/ima.h
@@ -233,6 +233,7 @@ int ima_policy_show(struct seq_file *m, void *v);
 #define IMA_APPRAISE_MODULES	0x08
 #define IMA_APPRAISE_FIRMWARE	0x10
 #define IMA_APPRAISE_POLICY	0x20
+#define IMA_APPRAISE_KEXEC	0x40
 
 #ifdef CONFIG_IMA_APPRAISE
 int ima_appraise_measurement(enum ima_hooks func,
diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
index 74d0bd7e76d7..754ece08e1c6 100644
--- a/security/integrity/ima/ima_main.c
+++ b/security/integrity/ima/ima_main.c
@@ -444,6 +444,15 @@ int ima_read_file(struct file *file, enum kernel_read_file_id read_id)
 		}
 		return 0;	/* We rely on module signature checking */
 	}
+
+	if (!file && read_id == READING_KEXEC_IMAGE) {
+		if ((ima_appraise & IMA_APPRAISE_KEXEC) &&
+		    (ima_appraise & IMA_APPRAISE_ENFORCE)) {
+			pr_err("impossible to appraise a kernel image without a file descriptor; try using kexec_file syscall.\n");
+			return -EACCES;	/* INTEGRITY_UNKNOWN */
+		}
+		return 0;
+	}
 	return 0;
 }
 
diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index df3e45878a87..a9e96a884867 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -448,6 +448,8 @@ static int ima_appraise_flag(enum ima_hooks func)
 		return IMA_APPRAISE_FIRMWARE;
 	else if (func == POLICY_CHECK)
 		return IMA_APPRAISE_POLICY;
+	else if (func == KEXEC_KERNEL_CHECK)
+		return IMA_APPRAISE_KEXEC;
 	return 0;
 }
 
-- 
2.7.5


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 3/3] ima: based on policy require signed kexec kernel images

[Mimi Zohar]

The original kexec_load syscall can not verify file signatures.  This
patch differentiates between the kexec_load and kexec_file_load
syscalls.

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Matthew Garrett <mjg59@google.com>
---
 security/integrity/ima/ima.h        | 1 +
 security/integrity/ima/ima_main.c   | 9 +++++++++
 security/integrity/ima/ima_policy.c | 2 ++
 3 files changed, 12 insertions(+)

diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h
index 35fe91aa1fc9..03c9c37ee345 100644
--- a/security/integrity/ima/ima.h
+++ b/security/integrity/ima/ima.h
@@ -233,6 +233,7 @@ int ima_policy_show(struct seq_file *m, void *v);
 #define IMA_APPRAISE_MODULES	0x08
 #define IMA_APPRAISE_FIRMWARE	0x10
 #define IMA_APPRAISE_POLICY	0x20
+#define IMA_APPRAISE_KEXEC	0x40
 
 #ifdef CONFIG_IMA_APPRAISE
 int ima_appraise_measurement(enum ima_hooks func,
diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
index 74d0bd7e76d7..754ece08e1c6 100644
--- a/security/integrity/ima/ima_main.c
+++ b/security/integrity/ima/ima_main.c
@@ -444,6 +444,15 @@ int ima_read_file(struct file *file, enum kernel_read_file_id read_id)
 		}
 		return 0;	/* We rely on module signature checking */
 	}
+
+	if (!file && read_id == READING_KEXEC_IMAGE) {
+		if ((ima_appraise & IMA_APPRAISE_KEXEC) &&
+		    (ima_appraise & IMA_APPRAISE_ENFORCE)) {
+			pr_err("impossible to appraise a kernel image without a file descriptor; try using kexec_file syscall.\n");
+			return -EACCES;	/* INTEGRITY_UNKNOWN */
+		}
+		return 0;
+	}
 	return 0;
 }
 
diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index df3e45878a87..a9e96a884867 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -448,6 +448,8 @@ static int ima_appraise_flag(enum ima_hooks func)
 		return IMA_APPRAISE_FIRMWARE;
 	else if (func == POLICY_CHECK)
 		return IMA_APPRAISE_POLICY;
+	else if (func == KEXEC_KERNEL_CHECK)
+		return IMA_APPRAISE_KEXEC;
 	return 0;
 }
 
-- 
2.7.5

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

In environments that require the kexec kernel image to be signed, prevent
using the kexec_load syscall.  In order for LSMs and IMA to differentiate
between kexec_load and kexec_file_load syscalls, this patch set adds a
call to security_kernel_read_file() in kexec_load_check().

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>

Mimi Zohar (3):
  ima: based on the "secure_boot" policy limit syscalls
  kexec: call LSM hook for kexec_load syscall
  ima: based on policy require signed kexec kernel images

 kernel/kexec.c                      | 11 +++++++++++
 security/integrity/ima/ima.h        |  1 +
 security/integrity/ima/ima_main.c   |  9 +++++++++
 security/integrity/ima/ima_policy.c | 27 ++++++++++++++++++++-------
 4 files changed, 41 insertions(+), 7 deletions(-)

-- 
2.7.5

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

In environments that require the kexec kernel image to be signed, prevent
using the kexec_load syscall.  In order for LSMs and IMA to differentiate
between kexec_load and kexec_file_load syscalls, this patch set adds a
call to security_kernel_read_file() in kexec_load_check().

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>

Mimi Zohar (3):
  ima: based on the "secure_boot" policy limit syscalls
  kexec: call LSM hook for kexec_load syscall
  ima: based on policy require signed kexec kernel images

 kernel/kexec.c                      | 11 +++++++++++
 security/integrity/ima/ima.h        |  1 +
 security/integrity/ima/ima_main.c   |  9 +++++++++
 security/integrity/ima/ima_policy.c | 27 ++++++++++++++++++++-------
 4 files changed, 41 insertions(+), 7 deletions(-)

-- 
2.7.5


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

In environments that require the kexec kernel image to be signed, prevent
using the kexec_load syscall.  In order for LSMs and IMA to differentiate
between kexec_load and kexec_file_load syscalls, this patch set adds a
call to security_kernel_read_file() in kexec_load_check().

Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>

Mimi Zohar (3):
  ima: based on the "secure_boot" policy limit syscalls
  kexec: call LSM hook for kexec_load syscall
  ima: based on policy require signed kexec kernel images

 kernel/kexec.c                      | 11 +++++++++++
 security/integrity/ima/ima.h        |  1 +
 security/integrity/ima/ima_main.c   |  9 +++++++++
 security/integrity/ima/ima_policy.c | 27 ++++++++++++++++++++-------
 4 files changed, 41 insertions(+), 7 deletions(-)

-- 
2.7.5

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> In environments that require the kexec kernel image to be signed, prevent
> using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> between kexec_load and kexec_file_load syscalls, this patch set adds a
> call to security_kernel_read_file() in kexec_load_check().

Having thought about it some more this justification for these changes
does not work.  The functionality of kexec_load is already root-only.
So in environments that require the kernel image to be signed just don't
use kexec_load.  Possibly even compile kexec_load out to save space
because you will never need it.  You don't need a new security hook to
do any of that.  Userspace is a very fine mechanism for being the
instrument of policy.

If you don't trust userspace that needs to be spelled out very clearly.
You need to talk about what your threat models are.

If the only justification is so that that we can't boot windows if
someone hacks into userspace it has my nack because that is another kind
of complete non-sense.

Given that you are not trusting userspace this changeset also probably
needs to have the kernel-hardening list cc'd.  Because the only possible
justification I can imagine for something like this is kernel-hardening.

Eric

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> In environments that require the kexec kernel image to be signed, prevent
> using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> between kexec_load and kexec_file_load syscalls, this patch set adds a
> call to security_kernel_read_file() in kexec_load_check().

Having thought about it some more this justification for these changes
does not work.  The functionality of kexec_load is already root-only.
So in environments that require the kernel image to be signed just don't
use kexec_load.  Possibly even compile kexec_load out to save space
because you will never need it.  You don't need a new security hook to
do any of that.  Userspace is a very fine mechanism for being the
instrument of policy.

If you don't trust userspace that needs to be spelled out very clearly.
You need to talk about what your threat models are.

If the only justification is so that that we can't boot windows if
someone hacks into userspace it has my nack because that is another kind
of complete non-sense.

Given that you are not trusting userspace this changeset also probably
needs to have the kernel-hardening list cc'd.  Because the only possible
justification I can imagine for something like this is kernel-hardening.

Eric

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> In environments that require the kexec kernel image to be signed, prevent
> using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> between kexec_load and kexec_file_load syscalls, this patch set adds a
> call to security_kernel_read_file() in kexec_load_check().

Having thought about it some more this justification for these changes
does not work.  The functionality of kexec_load is already root-only.
So in environments that require the kernel image to be signed just don't
use kexec_load.  Possibly even compile kexec_load out to save space
because you will never need it.  You don't need a new security hook to
do any of that.  Userspace is a very fine mechanism for being the
instrument of policy.

If you don't trust userspace that needs to be spelled out very clearly.
You need to talk about what your threat models are.

If the only justification is so that that we can't boot windows if
someone hacks into userspace it has my nack because that is another kind
of complete non-sense.

Given that you are not trusting userspace this changeset also probably
needs to have the kernel-hardening list cc'd.  Because the only possible
justification I can imagine for something like this is kernel-hardening.

Eric
--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Matthew Garrett]

On Thu, May 3, 2018 at 1:13 PM Eric W. Biederman <ebiederm@xmission.com>
wrote:

> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> > In environments that require the kexec kernel image to be signed,
prevent
> > using the kexec_load syscall.  In order for LSMs and IMA to
differentiate
> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> > call to security_kernel_read_file() in kexec_load_check().

> Having thought about it some more this justification for these changes
> does not work.  The functionality of kexec_load is already root-only.
> So in environments that require the kernel image to be signed just don't
> use kexec_load.  Possibly even compile kexec_load out to save space
> because you will never need it.  You don't need a new security hook to
> do any of that.  Userspace is a very fine mechanism for being the
> instrument of policy.

> If you don't trust userspace that needs to be spelled out very clearly.
> You need to talk about what your threat models are.

kexec_load gives root arbitrary power to modify the running kernel image,
including the ability to disable enforcement of module signatures. Given
that it weakens other security mechanisms that are designed to prevent root
from disabling them, it makes sense to allow the imposition of an
equivalent restriction.

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Matthew Garrett]

On Thu, May 3, 2018 at 1:13 PM Eric W. Biederman <ebiederm@xmission.com>
wrote:

> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> > In environments that require the kexec kernel image to be signed,
prevent
> > using the kexec_load syscall.  In order for LSMs and IMA to
differentiate
> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> > call to security_kernel_read_file() in kexec_load_check().

> Having thought about it some more this justification for these changes
> does not work.  The functionality of kexec_load is already root-only.
> So in environments that require the kernel image to be signed just don't
> use kexec_load.  Possibly even compile kexec_load out to save space
> because you will never need it.  You don't need a new security hook to
> do any of that.  Userspace is a very fine mechanism for being the
> instrument of policy.

> If you don't trust userspace that needs to be spelled out very clearly.
> You need to talk about what your threat models are.

kexec_load gives root arbitrary power to modify the running kernel image,
including the ability to disable enforcement of module signatures. Given
that it weakens other security mechanisms that are designed to prevent root
from disabling them, it makes sense to allow the imposition of an
equivalent restriction.

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 0/3] kexec: limit kexec_load syscall

[Matthew Garrett]

On Thu, May 3, 2018 at 1:13 PM Eric W. Biederman <ebiederm@xmission.com>
wrote:

> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> > In environments that require the kexec kernel image to be signed,
prevent
> > using the kexec_load syscall.  In order for LSMs and IMA to
differentiate
> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> > call to security_kernel_read_file() in kexec_load_check().

> Having thought about it some more this justification for these changes
> does not work.  The functionality of kexec_load is already root-only.
> So in environments that require the kernel image to be signed just don't
> use kexec_load.  Possibly even compile kexec_load out to save space
> because you will never need it.  You don't need a new security hook to
> do any of that.  Userspace is a very fine mechanism for being the
> instrument of policy.

> If you don't trust userspace that needs to be spelled out very clearly.
> You need to talk about what your threat models are.

kexec_load gives root arbitrary power to modify the running kernel image,
including the ability to disable enforcement of module signatures. Given
that it weakens other security mechanisms that are designed to prevent root
from disabling them, it makes sense to allow the imposition of an
equivalent restriction.
--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Matthew Garrett <mjg59@google.com> writes:

> On Thu, May 3, 2018 at 1:13 PM Eric W. Biederman <ebiederm@xmission.com>
> wrote:
>
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>
>> > In environments that require the kexec kernel image to be signed,
> prevent
>> > using the kexec_load syscall.  In order for LSMs and IMA to
> differentiate
>> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> > call to security_kernel_read_file() in kexec_load_check().
>
>> Having thought about it some more this justification for these changes
>> does not work.  The functionality of kexec_load is already root-only.
>> So in environments that require the kernel image to be signed just don't
>> use kexec_load.  Possibly even compile kexec_load out to save space
>> because you will never need it.  You don't need a new security hook to
>> do any of that.  Userspace is a very fine mechanism for being the
>> instrument of policy.
>
>> If you don't trust userspace that needs to be spelled out very clearly.
>> You need to talk about what your threat models are.
>
> kexec_load gives root arbitrary power to modify the running kernel image,
> including the ability to disable enforcement of module signatures.

No.  It does absolutely nothing to the running kernel image.
Combined with reboot(..., LINUX_REBOOT_CMD_KEXE, ...) it does allow
booting something different.  It is argubably a little more efficient
than writing to a file to direct the bootloader to boot something
different and then calling reboot.  But it is not fundamentally
different.

> Given
> that it weakens other security mechanisms that are designed to prevent root
> from disabling them, it makes sense to allow the imposition of an
> equivalent restriction.

Say what.  You are saying a lot of words without any specifics.  Not a
specific threat mode.  Not which security mecahnisms you are worried
about weakening.  Not what classes of problems you are trying to defend
against.

I absolutely hate this nonsense.  I thought you already went 20 rounds
with Linus and learned you need to be upfront with what you are
concerned about.

I believe reasonable situations can be constructed.  But I am not seeing
that happen here.

My hand wavy argument to go with yours is that code paths that are root
only are not audited for security properties.  As such the number of
exploitable bus you can find in them is larger than normal.  It might be
a little harder to mount xfs or another filesystem with an exploitable
file system image but I expect it exists.

Further nothing I have seen you involved with has been about truly
hardening the system against a hostile root.  I have for the last
several years been chipping away at that and you have been nowhere to be
found.

So please be specific.  Talk about which threat you are worried about.
Because so far this looks like someones effort to look like they were
doing something without actually caring about real world threats.

Eric

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Matthew Garrett <mjg59@google.com> writes:

> On Thu, May 3, 2018 at 1:13 PM Eric W. Biederman <ebiederm@xmission.com>
> wrote:
>
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>
>> > In environments that require the kexec kernel image to be signed,
> prevent
>> > using the kexec_load syscall.  In order for LSMs and IMA to
> differentiate
>> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> > call to security_kernel_read_file() in kexec_load_check().
>
>> Having thought about it some more this justification for these changes
>> does not work.  The functionality of kexec_load is already root-only.
>> So in environments that require the kernel image to be signed just don't
>> use kexec_load.  Possibly even compile kexec_load out to save space
>> because you will never need it.  You don't need a new security hook to
>> do any of that.  Userspace is a very fine mechanism for being the
>> instrument of policy.
>
>> If you don't trust userspace that needs to be spelled out very clearly.
>> You need to talk about what your threat models are.
>
> kexec_load gives root arbitrary power to modify the running kernel image,
> including the ability to disable enforcement of module signatures.

No.  It does absolutely nothing to the running kernel image.
Combined with reboot(..., LINUX_REBOOT_CMD_KEXE, ...) it does allow
booting something different.  It is argubably a little more efficient
than writing to a file to direct the bootloader to boot something
different and then calling reboot.  But it is not fundamentally
different.

> Given
> that it weakens other security mechanisms that are designed to prevent root
> from disabling them, it makes sense to allow the imposition of an
> equivalent restriction.

Say what.  You are saying a lot of words without any specifics.  Not a
specific threat mode.  Not which security mecahnisms you are worried
about weakening.  Not what classes of problems you are trying to defend
against.

I absolutely hate this nonsense.  I thought you already went 20 rounds
with Linus and learned you need to be upfront with what you are
concerned about.

I believe reasonable situations can be constructed.  But I am not seeing
that happen here.

My hand wavy argument to go with yours is that code paths that are root
only are not audited for security properties.  As such the number of
exploitable bus you can find in them is larger than normal.  It might be
a little harder to mount xfs or another filesystem with an exploitable
file system image but I expect it exists.

Further nothing I have seen you involved with has been about truly
hardening the system against a hostile root.  I have for the last
several years been chipping away at that and you have been nowhere to be
found.

So please be specific.  Talk about which threat you are worried about.
Because so far this looks like someones effort to look like they were
doing something without actually caring about real world threats.

Eric






_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Matthew Garrett <mjg59@google.com> writes:

> On Thu, May 3, 2018 at 1:13 PM Eric W. Biederman <ebiederm@xmission.com>
> wrote:
>
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>
>> > In environments that require the kexec kernel image to be signed,
> prevent
>> > using the kexec_load syscall.  In order for LSMs and IMA to
> differentiate
>> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> > call to security_kernel_read_file() in kexec_load_check().
>
>> Having thought about it some more this justification for these changes
>> does not work.  The functionality of kexec_load is already root-only.
>> So in environments that require the kernel image to be signed just don't
>> use kexec_load.  Possibly even compile kexec_load out to save space
>> because you will never need it.  You don't need a new security hook to
>> do any of that.  Userspace is a very fine mechanism for being the
>> instrument of policy.
>
>> If you don't trust userspace that needs to be spelled out very clearly.
>> You need to talk about what your threat models are.
>
> kexec_load gives root arbitrary power to modify the running kernel image,
> including the ability to disable enforcement of module signatures.

No.  It does absolutely nothing to the running kernel image.
Combined with reboot(..., LINUX_REBOOT_CMD_KEXE, ...) it does allow
booting something different.  It is argubably a little more efficient
than writing to a file to direct the bootloader to boot something
different and then calling reboot.  But it is not fundamentally
different.

> Given
> that it weakens other security mechanisms that are designed to prevent root
> from disabling them, it makes sense to allow the imposition of an
> equivalent restriction.

Say what.  You are saying a lot of words without any specifics.  Not a
specific threat mode.  Not which security mecahnisms you are worried
about weakening.  Not what classes of problems you are trying to defend
against.

I absolutely hate this nonsense.  I thought you already went 20 rounds
with Linus and learned you need to be upfront with what you are
concerned about.

I believe reasonable situations can be constructed.  But I am not seeing
that happen here.

My hand wavy argument to go with yours is that code paths that are root
only are not audited for security properties.  As such the number of
exploitable bus you can find in them is larger than normal.  It might be
a little harder to mount xfs or another filesystem with an exploitable
file system image but I expect it exists.

Further nothing I have seen you involved with has been about truly
hardening the system against a hostile root.  I have for the last
several years been chipping away at that and you have been nowhere to be
found.

So please be specific.  Talk about which threat you are worried about.
Because so far this looks like someones effort to look like they were
doing something without actually caring about real world threats.

Eric





--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Matthew Garrett]

On Thu, May 3, 2018 at 2:59 PM Eric W. Biederman <ebiederm@xmission.com>
wrote:

> Matthew Garrett <mjg59@google.com> writes:
> > kexec_load gives root arbitrary power to modify the running kernel
image,
> > including the ability to disable enforcement of module signatures.

> No.  It does absolutely nothing to the running kernel image.
> Combined with reboot(..., LINUX_REBOOT_CMD_KEXE, ...) it does allow
> booting something different.  It is argubably a little more efficient
> than writing to a file to direct the bootloader to boot something
> different and then calling reboot.  But it is not fundamentally
> different.

It absolutely does - https://mjg59.dreamwidth.org/28746.html gives an
example. The payload just needs to return.

> > Given
> > that it weakens other security mechanisms that are designed to prevent
root
> > from disabling them, it makes sense to allow the imposition of an
> > equivalent restriction.

> Say what.  You are saying a lot of words without any specifics.  Not a
> specific threat mode.  Not which security mecahnisms you are worried
> about weakening.  Not what classes of problems you are trying to defend
> against.

I have a kernel configured with module.sig_enforce enabled - root is unable
to load kernel modules that are unsigned, and since sig_enforce is
bool_enable_only, root is unable to flip that back. Any number of security
models may be implemented with that assumption. However, root still has
access to kexec_load(), and can therefore kexec into a dummy payload that
flips that byte back to 0 and permits loading unsigned module code.

There may well be other mechanisms that permit root to gain arbitrary
ability to modify kernel code. My argument is that we should treat those as
bugs, not use their existence as a justification for leaving open known
cases.

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Matthew Garrett]

On Thu, May 3, 2018 at 2:59 PM Eric W. Biederman <ebiederm@xmission.com>
wrote:

> Matthew Garrett <mjg59@google.com> writes:
> > kexec_load gives root arbitrary power to modify the running kernel
image,
> > including the ability to disable enforcement of module signatures.

> No.  It does absolutely nothing to the running kernel image.
> Combined with reboot(..., LINUX_REBOOT_CMD_KEXE, ...) it does allow
> booting something different.  It is argubably a little more efficient
> than writing to a file to direct the bootloader to boot something
> different and then calling reboot.  But it is not fundamentally
> different.

It absolutely does - https://mjg59.dreamwidth.org/28746.html gives an
example. The payload just needs to return.

> > Given
> > that it weakens other security mechanisms that are designed to prevent
root
> > from disabling them, it makes sense to allow the imposition of an
> > equivalent restriction.

> Say what.  You are saying a lot of words without any specifics.  Not a
> specific threat mode.  Not which security mecahnisms you are worried
> about weakening.  Not what classes of problems you are trying to defend
> against.

I have a kernel configured with module.sig_enforce enabled - root is unable
to load kernel modules that are unsigned, and since sig_enforce is
bool_enable_only, root is unable to flip that back. Any number of security
models may be implemented with that assumption. However, root still has
access to kexec_load(), and can therefore kexec into a dummy payload that
flips that byte back to 0 and permits loading unsigned module code.

There may well be other mechanisms that permit root to gain arbitrary
ability to modify kernel code. My argument is that we should treat those as
bugs, not use their existence as a justification for leaving open known
cases.

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

[PATCH 0/3] kexec: limit kexec_load syscall

[Matthew Garrett]

On Thu, May 3, 2018 at 2:59 PM Eric W. Biederman <ebiederm@xmission.com>
wrote:

> Matthew Garrett <mjg59@google.com> writes:
> > kexec_load gives root arbitrary power to modify the running kernel
image,
> > including the ability to disable enforcement of module signatures.

> No.  It does absolutely nothing to the running kernel image.
> Combined with reboot(..., LINUX_REBOOT_CMD_KEXE, ...) it does allow
> booting something different.  It is argubably a little more efficient
> than writing to a file to direct the bootloader to boot something
> different and then calling reboot.  But it is not fundamentally
> different.

It absolutely does - https://mjg59.dreamwidth.org/28746.html gives an
example. The payload just needs to return.

> > Given
> > that it weakens other security mechanisms that are designed to prevent
root
> > from disabling them, it makes sense to allow the imposition of an
> > equivalent restriction.

> Say what.  You are saying a lot of words without any specifics.  Not a
> specific threat mode.  Not which security mecahnisms you are worried
> about weakening.  Not what classes of problems you are trying to defend
> against.

I have a kernel configured with module.sig_enforce enabled - root is unable
to load kernel modules that are unsigned, and since sig_enforce is
bool_enable_only, root is unable to flip that back. Any number of security
models may be implemented with that assumption. However, root still has
access to kexec_load(), and can therefore kexec into a dummy payload that
flips that byte back to 0 and permits loading unsigned module code.

There may well be other mechanisms that permit root to gain arbitrary
ability to modify kernel code. My argument is that we should treat those as
bugs, not use their existence as a justification for leaving open known
cases.
--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

[Cc'ing Kees and kernel-hardening]

On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > In environments that require the kexec kernel image to be signed, prevent
> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> > call to security_kernel_read_file() in kexec_load_check().
> 
> Having thought about it some more this justification for these changes
> does not work.  The functionality of kexec_load is already root-only.
> So in environments that require the kernel image to be signed just don't
> use kexec_load.  Possibly even compile kexec_load out to save space
> because you will never need it.  You don't need a new security hook to
> do any of that.  Userspace is a very fine mechanism for being the
> instrument of policy.

True, for those building their own kernel, they can disable the old
syscalls.  The concern is not for those building their own kernels,
but for those using stock kernels.  

By adding an LSM hook here in the kexec_load syscall, as opposed to an
IMA specific hook, other LSMs can piggy back on top of it.  Currently,
both load_pin and SELinux are gating the kernel module syscalls based
on security_kernel_read_file.

If there was a similar option for the kernel image, I'm pretty sure
other LSMs would use it.

>From an IMA perspective, there needs to be some method for only
allowing signed code to be loaded, executed, etc. - kernel modules,
kernel image/initramfs, firmware, policies.

> If you don't trust userspace that needs to be spelled out very clearly.
> You need to talk about what your threat models are.
> 
> If the only justification is so that that we can't boot windows if
> someone hacks into userspace it has my nack because that is another kind
> of complete non-sense.

The usecase is the ability to gate the kexec_load usage in stock
kernels.

> 
> Given that you are not trusting userspace this changeset also probably
> needs to have the kernel-hardening list cc'd.  Because the only possible
> justification I can imagine for something like this is kernel-hardening.

Sure, Cc'ing linux-hardening and Kees.

Mimi

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

[Cc'ing Kees and kernel-hardening]

On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > In environments that require the kexec kernel image to be signed, prevent
> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> > call to security_kernel_read_file() in kexec_load_check().
> 
> Having thought about it some more this justification for these changes
> does not work.  The functionality of kexec_load is already root-only.
> So in environments that require the kernel image to be signed just don't
> use kexec_load.  Possibly even compile kexec_load out to save space
> because you will never need it.  You don't need a new security hook to
> do any of that.  Userspace is a very fine mechanism for being the
> instrument of policy.

True, for those building their own kernel, they can disable the old
syscalls.  The concern is not for those building their own kernels,
but for those using stock kernels.  

By adding an LSM hook here in the kexec_load syscall, as opposed to an
IMA specific hook, other LSMs can piggy back on top of it.  Currently,
both load_pin and SELinux are gating the kernel module syscalls based
on security_kernel_read_file.

If there was a similar option for the kernel image, I'm pretty sure
other LSMs would use it.

From an IMA perspective, there needs to be some method for only
allowing signed code to be loaded, executed, etc. - kernel modules,
kernel image/initramfs, firmware, policies.

> If you don't trust userspace that needs to be spelled out very clearly.
> You need to talk about what your threat models are.
> 
> If the only justification is so that that we can't boot windows if
> someone hacks into userspace it has my nack because that is another kind
> of complete non-sense.

The usecase is the ability to gate the kexec_load usage in stock
kernels.

> 
> Given that you are not trusting userspace this changeset also probably
> needs to have the kernel-hardening list cc'd.  Because the only possible
> justification I can imagine for something like this is kernel-hardening.

Sure, Cc'ing linux-hardening and Kees.

Mimi


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

[Cc'ing Kees and kernel-hardening]

On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > In environments that require the kexec kernel image to be signed, prevent
> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> > call to security_kernel_read_file() in kexec_load_check().
> 
> Having thought about it some more this justification for these changes
> does not work.  The functionality of kexec_load is already root-only.
> So in environments that require the kernel image to be signed just don't
> use kexec_load.  Possibly even compile kexec_load out to save space
> because you will never need it.  You don't need a new security hook to
> do any of that.  Userspace is a very fine mechanism for being the
> instrument of policy.

True, for those building their own kernel, they can disable the old
syscalls.  The concern is not for those building their own kernels,
but for those using stock kernels.  

By adding an LSM hook here in the kexec_load syscall, as opposed to an
IMA specific hook, other LSMs can piggy back on top of it.  Currently,
both load_pin and SELinux are gating the kernel module syscalls based
on security_kernel_read_file.

If there was a similar option for the kernel image, I'm pretty sure
other LSMs would use it.

>From an IMA perspective, there needs to be some method for only
allowing signed code to be loaded, executed, etc. - kernel modules,
kernel image/initramfs, firmware, policies.

> If you don't trust userspace that needs to be spelled out very clearly.
> You need to talk about what your threat models are.
> 
> If the only justification is so that that we can't boot windows if
> someone hacks into userspace it has my nack because that is another kind
> of complete non-sense.

The usecase is the ability to gate the kexec_load usage in stock
kernels.

> 
> Given that you are not trusting userspace this changeset also probably
> needs to have the kernel-hardening list cc'd.  Because the only possible
> justification I can imagine for something like this is kernel-hardening.

Sure, Cc'ing linux-hardening and Kees.

Mimi

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

[Cc'ing Kees and kernel-hardening]

On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > In environments that require the kexec kernel image to be signed, prevent
> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> > call to security_kernel_read_file() in kexec_load_check().
> 
> Having thought about it some more this justification for these changes
> does not work.  The functionality of kexec_load is already root-only.
> So in environments that require the kernel image to be signed just don't
> use kexec_load.  Possibly even compile kexec_load out to save space
> because you will never need it.  You don't need a new security hook to
> do any of that.  Userspace is a very fine mechanism for being the
> instrument of policy.

True, for those building their own kernel, they can disable the old
syscalls. ?The concern is not for those building their own kernels,
but for those using stock kernels. ?

By adding an LSM hook here in the kexec_load syscall, as opposed to an
IMA specific hook, other LSMs can piggy back on top of it. ?Currently,
both load_pin and SELinux are gating the kernel module syscalls based
on security_kernel_read_file.

If there was a similar option for the kernel image, I'm pretty sure
other LSMs would use it.

>From an IMA perspective, there needs to be some method for only
allowing signed code to be loaded, executed, etc. - kernel modules,
kernel image/initramfs, firmware, policies.

> If you don't trust userspace that needs to be spelled out very clearly.
> You need to talk about what your threat models are.
> 
> If the only justification is so that that we can't boot windows if
> someone hacks into userspace it has my nack because that is another kind
> of complete non-sense.

The usecase is the ability to gate the kexec_load usage in stock
kernels.

> 
> Given that you are not trusting userspace this changeset also probably
> needs to have the kernel-hardening list cc'd.  Because the only possible
> justification I can imagine for something like this is kernel-hardening.

Sure, Cc'ing linux-hardening and Kees.

Mimi

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> [Cc'ing Kees and kernel-hardening]
>
> On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > In environments that require the kexec kernel image to be signed, prevent
>> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> > call to security_kernel_read_file() in kexec_load_check().
>> 
>> Having thought about it some more this justification for these changes
>> does not work.  The functionality of kexec_load is already root-only.
>> So in environments that require the kernel image to be signed just don't
>> use kexec_load.  Possibly even compile kexec_load out to save space
>> because you will never need it.  You don't need a new security hook to
>> do any of that.  Userspace is a very fine mechanism for being the
>> instrument of policy.
>
> True, for those building their own kernel, they can disable the old
> syscalls.  The concern is not for those building their own kernels,
> but for those using stock kernels.  
>
> By adding an LSM hook here in the kexec_load syscall, as opposed to an
> IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> both load_pin and SELinux are gating the kernel module syscalls based
> on security_kernel_read_file.
>
> If there was a similar option for the kernel image, I'm pretty sure
> other LSMs would use it.
>
> From an IMA perspective, there needs to be some method for only
> allowing signed code to be loaded, executed, etc. - kernel modules,
> kernel image/initramfs, firmware, policies.

What is the IMA perspective.  Why can't IMA trust appropriately
authorized userspace?

>> If you don't trust userspace that needs to be spelled out very clearly.
>> You need to talk about what your threat models are.
>> 
>> If the only justification is so that that we can't boot windows if
>> someone hacks into userspace it has my nack because that is another kind
>> of complete non-sense.
>
> The usecase is the ability to gate the kexec_load usage in stock
> kernels.

But kexec_load is already gated.  It requires CAP_SYS_BOOT.

>> Given that you are not trusting userspace this changeset also probably
>> needs to have the kernel-hardening list cc'd.  Because the only possible
>> justification I can imagine for something like this is kernel-hardening.
>
> Sure, Cc'ing linux-hardening and Kees.
>
> Mimi

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> [Cc'ing Kees and kernel-hardening]
>
> On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > In environments that require the kexec kernel image to be signed, prevent
>> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> > call to security_kernel_read_file() in kexec_load_check().
>> 
>> Having thought about it some more this justification for these changes
>> does not work.  The functionality of kexec_load is already root-only.
>> So in environments that require the kernel image to be signed just don't
>> use kexec_load.  Possibly even compile kexec_load out to save space
>> because you will never need it.  You don't need a new security hook to
>> do any of that.  Userspace is a very fine mechanism for being the
>> instrument of policy.
>
> True, for those building their own kernel, they can disable the old
> syscalls.  The concern is not for those building their own kernels,
> but for those using stock kernels.  
>
> By adding an LSM hook here in the kexec_load syscall, as opposed to an
> IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> both load_pin and SELinux are gating the kernel module syscalls based
> on security_kernel_read_file.
>
> If there was a similar option for the kernel image, I'm pretty sure
> other LSMs would use it.
>
> From an IMA perspective, there needs to be some method for only
> allowing signed code to be loaded, executed, etc. - kernel modules,
> kernel image/initramfs, firmware, policies.

What is the IMA perspective.  Why can't IMA trust appropriately
authorized userspace?

>> If you don't trust userspace that needs to be spelled out very clearly.
>> You need to talk about what your threat models are.
>> 
>> If the only justification is so that that we can't boot windows if
>> someone hacks into userspace it has my nack because that is another kind
>> of complete non-sense.
>
> The usecase is the ability to gate the kexec_load usage in stock
> kernels.

But kexec_load is already gated.  It requires CAP_SYS_BOOT.

>> Given that you are not trusting userspace this changeset also probably
>> needs to have the kernel-hardening list cc'd.  Because the only possible
>> justification I can imagine for something like this is kernel-hardening.
>
> Sure, Cc'ing linux-hardening and Kees.
>
> Mimi

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> [Cc'ing Kees and kernel-hardening]
>
> On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > In environments that require the kexec kernel image to be signed, prevent
>> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> > call to security_kernel_read_file() in kexec_load_check().
>> 
>> Having thought about it some more this justification for these changes
>> does not work.  The functionality of kexec_load is already root-only.
>> So in environments that require the kernel image to be signed just don't
>> use kexec_load.  Possibly even compile kexec_load out to save space
>> because you will never need it.  You don't need a new security hook to
>> do any of that.  Userspace is a very fine mechanism for being the
>> instrument of policy.
>
> True, for those building their own kernel, they can disable the old
> syscalls.  The concern is not for those building their own kernels,
> but for those using stock kernels.  
>
> By adding an LSM hook here in the kexec_load syscall, as opposed to an
> IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> both load_pin and SELinux are gating the kernel module syscalls based
> on security_kernel_read_file.
>
> If there was a similar option for the kernel image, I'm pretty sure
> other LSMs would use it.
>
> From an IMA perspective, there needs to be some method for only
> allowing signed code to be loaded, executed, etc. - kernel modules,
> kernel image/initramfs, firmware, policies.

What is the IMA perspective.  Why can't IMA trust appropriately
authorized userspace?

>> If you don't trust userspace that needs to be spelled out very clearly.
>> You need to talk about what your threat models are.
>> 
>> If the only justification is so that that we can't boot windows if
>> someone hacks into userspace it has my nack because that is another kind
>> of complete non-sense.
>
> The usecase is the ability to gate the kexec_load usage in stock
> kernels.

But kexec_load is already gated.  It requires CAP_SYS_BOOT.

>> Given that you are not trusting userspace this changeset also probably
>> needs to have the kernel-hardening list cc'd.  Because the only possible
>> justification I can imagine for something like this is kernel-hardening.
>
> Sure, Cc'ing linux-hardening and Kees.
>
> Mimi

[PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> [Cc'ing Kees and kernel-hardening]
>
> On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > In environments that require the kexec kernel image to be signed, prevent
>> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> > call to security_kernel_read_file() in kexec_load_check().
>> 
>> Having thought about it some more this justification for these changes
>> does not work.  The functionality of kexec_load is already root-only.
>> So in environments that require the kernel image to be signed just don't
>> use kexec_load.  Possibly even compile kexec_load out to save space
>> because you will never need it.  You don't need a new security hook to
>> do any of that.  Userspace is a very fine mechanism for being the
>> instrument of policy.
>
> True, for those building their own kernel, they can disable the old
> syscalls. ?The concern is not for those building their own kernels,
> but for those using stock kernels. ?
>
> By adding an LSM hook here in the kexec_load syscall, as opposed to an
> IMA specific hook, other LSMs can piggy back on top of it. ?Currently,
> both load_pin and SELinux are gating the kernel module syscalls based
> on security_kernel_read_file.
>
> If there was a similar option for the kernel image, I'm pretty sure
> other LSMs would use it.
>
> From an IMA perspective, there needs to be some method for only
> allowing signed code to be loaded, executed, etc. - kernel modules,
> kernel image/initramfs, firmware, policies.

What is the IMA perspective.  Why can't IMA trust appropriately
authorized userspace?

>> If you don't trust userspace that needs to be spelled out very clearly.
>> You need to talk about what your threat models are.
>> 
>> If the only justification is so that that we can't boot windows if
>> someone hacks into userspace it has my nack because that is another kind
>> of complete non-sense.
>
> The usecase is the ability to gate the kexec_load usage in stock
> kernels.

But kexec_load is already gated.  It requires CAP_SYS_BOOT.

>> Given that you are not trusting userspace this changeset also probably
>> needs to have the kernel-hardening list cc'd.  Because the only possible
>> justification I can imagine for something like this is kernel-hardening.
>
> Sure, Cc'ing linux-hardening and Kees.
>
> Mimi
--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > [Cc'ing Kees and kernel-hardening]
> >
> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > In environments that require the kexec kernel image to be signed, prevent
> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> > call to security_kernel_read_file() in kexec_load_check().
> >> 
> >> Having thought about it some more this justification for these changes
> >> does not work.  The functionality of kexec_load is already root-only.
> >> So in environments that require the kernel image to be signed just don't
> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> because you will never need it.  You don't need a new security hook to
> >> do any of that.  Userspace is a very fine mechanism for being the
> >> instrument of policy.
> >
> > True, for those building their own kernel, they can disable the old
> > syscalls.  The concern is not for those building their own kernels,
> > but for those using stock kernels.  
> >
> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> > both load_pin and SELinux are gating the kernel module syscalls based
> > on security_kernel_read_file.
> >
> > If there was a similar option for the kernel image, I'm pretty sure
> > other LSMs would use it.
> >
> > From an IMA perspective, there needs to be some method for only
> > allowing signed code to be loaded, executed, etc. - kernel modules,
> > kernel image/initramfs, firmware, policies.
> 
> What is the IMA perspective.  Why can't IMA trust appropriately
> authorized userspace?

Suppose a system owner wants to define a system wide policy that
requires all code be signed - kernel modules, firmware, kexec image &
initramfs, executables, mmapped files, etc - without having to rebuild
the kernel.  Without a call in kexec_load that isn't possible.

> 
> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> You need to talk about what your threat models are.
> >> 
> >> If the only justification is so that that we can't boot windows if
> >> someone hacks into userspace it has my nack because that is another kind
> >> of complete non-sense.
> >
> > The usecase is the ability to gate the kexec_load usage in stock
> > kernels.
> 
> But kexec_load is already gated.  It requires CAP_SYS_BOOT.

It isn't a matter of kexec_load already being gated, but of wanting a
single place for defining a system wide policy, as described above.

Mimi

> 
> >> Given that you are not trusting userspace this changeset also probably
> >> needs to have the kernel-hardening list cc'd.  Because the only possible
> >> justification I can imagine for something like this is kernel-hardening.
> >
> > Sure, Cc'ing linux-hardening and Kees.
> >
> > Mimi
> 

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > [Cc'ing Kees and kernel-hardening]
> >
> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > In environments that require the kexec kernel image to be signed, prevent
> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> > call to security_kernel_read_file() in kexec_load_check().
> >> 
> >> Having thought about it some more this justification for these changes
> >> does not work.  The functionality of kexec_load is already root-only.
> >> So in environments that require the kernel image to be signed just don't
> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> because you will never need it.  You don't need a new security hook to
> >> do any of that.  Userspace is a very fine mechanism for being the
> >> instrument of policy.
> >
> > True, for those building their own kernel, they can disable the old
> > syscalls.  The concern is not for those building their own kernels,
> > but for those using stock kernels.  
> >
> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> > both load_pin and SELinux are gating the kernel module syscalls based
> > on security_kernel_read_file.
> >
> > If there was a similar option for the kernel image, I'm pretty sure
> > other LSMs would use it.
> >
> > From an IMA perspective, there needs to be some method for only
> > allowing signed code to be loaded, executed, etc. - kernel modules,
> > kernel image/initramfs, firmware, policies.
> 
> What is the IMA perspective.  Why can't IMA trust appropriately
> authorized userspace?

Suppose a system owner wants to define a system wide policy that
requires all code be signed - kernel modules, firmware, kexec image &
initramfs, executables, mmapped files, etc - without having to rebuild
the kernel.  Without a call in kexec_load that isn't possible.

> 
> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> You need to talk about what your threat models are.
> >> 
> >> If the only justification is so that that we can't boot windows if
> >> someone hacks into userspace it has my nack because that is another kind
> >> of complete non-sense.
> >
> > The usecase is the ability to gate the kexec_load usage in stock
> > kernels.
> 
> But kexec_load is already gated.  It requires CAP_SYS_BOOT.

It isn't a matter of kexec_load already being gated, but of wanting a
single place for defining a system wide policy, as described above.

Mimi

> 
> >> Given that you are not trusting userspace this changeset also probably
> >> needs to have the kernel-hardening list cc'd.  Because the only possible
> >> justification I can imagine for something like this is kernel-hardening.
> >
> > Sure, Cc'ing linux-hardening and Kees.
> >
> > Mimi
> 


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > [Cc'ing Kees and kernel-hardening]
> >
> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > In environments that require the kexec kernel image to be signed, prevent
> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> > call to security_kernel_read_file() in kexec_load_check().
> >> 
> >> Having thought about it some more this justification for these changes
> >> does not work.  The functionality of kexec_load is already root-only.
> >> So in environments that require the kernel image to be signed just don't
> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> because you will never need it.  You don't need a new security hook to
> >> do any of that.  Userspace is a very fine mechanism for being the
> >> instrument of policy.
> >
> > True, for those building their own kernel, they can disable the old
> > syscalls.  The concern is not for those building their own kernels,
> > but for those using stock kernels.  
> >
> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> > both load_pin and SELinux are gating the kernel module syscalls based
> > on security_kernel_read_file.
> >
> > If there was a similar option for the kernel image, I'm pretty sure
> > other LSMs would use it.
> >
> > From an IMA perspective, there needs to be some method for only
> > allowing signed code to be loaded, executed, etc. - kernel modules,
> > kernel image/initramfs, firmware, policies.
> 
> What is the IMA perspective.  Why can't IMA trust appropriately
> authorized userspace?

Suppose a system owner wants to define a system wide policy that
requires all code be signed - kernel modules, firmware, kexec image &
initramfs, executables, mmapped files, etc - without having to rebuild
the kernel.  Without a call in kexec_load that isn't possible.

> 
> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> You need to talk about what your threat models are.
> >> 
> >> If the only justification is so that that we can't boot windows if
> >> someone hacks into userspace it has my nack because that is another kind
> >> of complete non-sense.
> >
> > The usecase is the ability to gate the kexec_load usage in stock
> > kernels.
> 
> But kexec_load is already gated.  It requires CAP_SYS_BOOT.

It isn't a matter of kexec_load already being gated, but of wanting a
single place for defining a system wide policy, as described above.

Mimi

> 
> >> Given that you are not trusting userspace this changeset also probably
> >> needs to have the kernel-hardening list cc'd.  Because the only possible
> >> justification I can imagine for something like this is kernel-hardening.
> >
> > Sure, Cc'ing linux-hardening and Kees.
> >
> > Mimi
> 

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > [Cc'ing Kees and kernel-hardening]
> >
> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > In environments that require the kexec kernel image to be signed, prevent
> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> > call to security_kernel_read_file() in kexec_load_check().
> >> 
> >> Having thought about it some more this justification for these changes
> >> does not work.  The functionality of kexec_load is already root-only.
> >> So in environments that require the kernel image to be signed just don't
> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> because you will never need it.  You don't need a new security hook to
> >> do any of that.  Userspace is a very fine mechanism for being the
> >> instrument of policy.
> >
> > True, for those building their own kernel, they can disable the old
> > syscalls. ?The concern is not for those building their own kernels,
> > but for those using stock kernels. ?
> >
> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> > IMA specific hook, other LSMs can piggy back on top of it. ?Currently,
> > both load_pin and SELinux are gating the kernel module syscalls based
> > on security_kernel_read_file.
> >
> > If there was a similar option for the kernel image, I'm pretty sure
> > other LSMs would use it.
> >
> > From an IMA perspective, there needs to be some method for only
> > allowing signed code to be loaded, executed, etc. - kernel modules,
> > kernel image/initramfs, firmware, policies.
> 
> What is the IMA perspective.  Why can't IMA trust appropriately
> authorized userspace?

Suppose a system owner wants to define a system wide policy that
requires all code be signed - kernel modules, firmware, kexec image &
initramfs, executables, mmapped files, etc - without having to rebuild
the kernel. ?Without a call in kexec_load that isn't possible.

> 
> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> You need to talk about what your threat models are.
> >> 
> >> If the only justification is so that that we can't boot windows if
> >> someone hacks into userspace it has my nack because that is another kind
> >> of complete non-sense.
> >
> > The usecase is the ability to gate the kexec_load usage in stock
> > kernels.
> 
> But kexec_load is already gated.  It requires CAP_SYS_BOOT.

It isn't a matter of kexec_load already being gated, but of wanting a
single place for defining a system wide policy, as described above.

Mimi

> 
> >> Given that you are not trusting userspace this changeset also probably
> >> needs to have the kernel-hardening list cc'd.  Because the only possible
> >> justification I can imagine for something like this is kernel-hardening.
> >
> > Sure, Cc'ing linux-hardening and Kees.
> >
> > Mimi
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > [Cc'ing Kees and kernel-hardening]
>> >
>> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> >> 
>> >> > In environments that require the kexec kernel image to be signed, prevent
>> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> >> > call to security_kernel_read_file() in kexec_load_check().
>> >> 
>> >> Having thought about it some more this justification for these changes
>> >> does not work.  The functionality of kexec_load is already root-only.
>> >> So in environments that require the kernel image to be signed just don't
>> >> use kexec_load.  Possibly even compile kexec_load out to save space
>> >> because you will never need it.  You don't need a new security hook to
>> >> do any of that.  Userspace is a very fine mechanism for being the
>> >> instrument of policy.
>> >
>> > True, for those building their own kernel, they can disable the old
>> > syscalls.  The concern is not for those building their own kernels,
>> > but for those using stock kernels.  
>> >
>> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
>> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
>> > both load_pin and SELinux are gating the kernel module syscalls based
>> > on security_kernel_read_file.
>> >
>> > If there was a similar option for the kernel image, I'm pretty sure
>> > other LSMs would use it.
>> >
>> > From an IMA perspective, there needs to be some method for only
>> > allowing signed code to be loaded, executed, etc. - kernel modules,
>> > kernel image/initramfs, firmware, policies.
>> 
>> What is the IMA perspective.  Why can't IMA trust appropriately
>> authorized userspace?
>
> Suppose a system owner wants to define a system wide policy that
> requires all code be signed - kernel modules, firmware, kexec image &
> initramfs, executables, mmapped files, etc - without having to rebuild
> the kernel.  Without a call in kexec_load that isn't possible.

Of course it is.  You just make it a requirement that before an
executable will be signed it will be audited to see that it doesn't
call sys_kexec_load.  Signing presumably means something.  So it should
not be hard to enforce a policy like that on a specialty system call
that most applications will never call.

>> >> If you don't trust userspace that needs to be spelled out very clearly.
>> >> You need to talk about what your threat models are.
>> >> 
>> >> If the only justification is so that that we can't boot windows if
>> >> someone hacks into userspace it has my nack because that is another kind
>> >> of complete non-sense.
>> >
>> > The usecase is the ability to gate the kexec_load usage in stock
>> > kernels.
>> 
>> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
>
> It isn't a matter of kexec_load already being gated, but of wanting a
> single place for defining a system wide policy, as described above.

Signing is only a tool to enforce a policy.  Signing by itself is not a
policy.  Enforcing any quality controls in the signed executables should
trivially prevent kexec_load from being used.

Eric

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > [Cc'ing Kees and kernel-hardening]
>> >
>> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> >> 
>> >> > In environments that require the kexec kernel image to be signed, prevent
>> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> >> > call to security_kernel_read_file() in kexec_load_check().
>> >> 
>> >> Having thought about it some more this justification for these changes
>> >> does not work.  The functionality of kexec_load is already root-only.
>> >> So in environments that require the kernel image to be signed just don't
>> >> use kexec_load.  Possibly even compile kexec_load out to save space
>> >> because you will never need it.  You don't need a new security hook to
>> >> do any of that.  Userspace is a very fine mechanism for being the
>> >> instrument of policy.
>> >
>> > True, for those building their own kernel, they can disable the old
>> > syscalls.  The concern is not for those building their own kernels,
>> > but for those using stock kernels.  
>> >
>> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
>> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
>> > both load_pin and SELinux are gating the kernel module syscalls based
>> > on security_kernel_read_file.
>> >
>> > If there was a similar option for the kernel image, I'm pretty sure
>> > other LSMs would use it.
>> >
>> > From an IMA perspective, there needs to be some method for only
>> > allowing signed code to be loaded, executed, etc. - kernel modules,
>> > kernel image/initramfs, firmware, policies.
>> 
>> What is the IMA perspective.  Why can't IMA trust appropriately
>> authorized userspace?
>
> Suppose a system owner wants to define a system wide policy that
> requires all code be signed - kernel modules, firmware, kexec image &
> initramfs, executables, mmapped files, etc - without having to rebuild
> the kernel.  Without a call in kexec_load that isn't possible.

Of course it is.  You just make it a requirement that before an
executable will be signed it will be audited to see that it doesn't
call sys_kexec_load.  Signing presumably means something.  So it should
not be hard to enforce a policy like that on a specialty system call
that most applications will never call.

>> >> If you don't trust userspace that needs to be spelled out very clearly.
>> >> You need to talk about what your threat models are.
>> >> 
>> >> If the only justification is so that that we can't boot windows if
>> >> someone hacks into userspace it has my nack because that is another kind
>> >> of complete non-sense.
>> >
>> > The usecase is the ability to gate the kexec_load usage in stock
>> > kernels.
>> 
>> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
>
> It isn't a matter of kexec_load already being gated, but of wanting a
> single place for defining a system wide policy, as described above.

Signing is only a tool to enforce a policy.  Signing by itself is not a
policy.  Enforcing any quality controls in the signed executables should
trivially prevent kexec_load from being used.

Eric

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > [Cc'ing Kees and kernel-hardening]
>> >
>> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> >> 
>> >> > In environments that require the kexec kernel image to be signed, prevent
>> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> >> > call to security_kernel_read_file() in kexec_load_check().
>> >> 
>> >> Having thought about it some more this justification for these changes
>> >> does not work.  The functionality of kexec_load is already root-only.
>> >> So in environments that require the kernel image to be signed just don't
>> >> use kexec_load.  Possibly even compile kexec_load out to save space
>> >> because you will never need it.  You don't need a new security hook to
>> >> do any of that.  Userspace is a very fine mechanism for being the
>> >> instrument of policy.
>> >
>> > True, for those building their own kernel, they can disable the old
>> > syscalls.  The concern is not for those building their own kernels,
>> > but for those using stock kernels.  
>> >
>> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
>> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
>> > both load_pin and SELinux are gating the kernel module syscalls based
>> > on security_kernel_read_file.
>> >
>> > If there was a similar option for the kernel image, I'm pretty sure
>> > other LSMs would use it.
>> >
>> > From an IMA perspective, there needs to be some method for only
>> > allowing signed code to be loaded, executed, etc. - kernel modules,
>> > kernel image/initramfs, firmware, policies.
>> 
>> What is the IMA perspective.  Why can't IMA trust appropriately
>> authorized userspace?
>
> Suppose a system owner wants to define a system wide policy that
> requires all code be signed - kernel modules, firmware, kexec image &
> initramfs, executables, mmapped files, etc - without having to rebuild
> the kernel.  Without a call in kexec_load that isn't possible.

Of course it is.  You just make it a requirement that before an
executable will be signed it will be audited to see that it doesn't
call sys_kexec_load.  Signing presumably means something.  So it should
not be hard to enforce a policy like that on a specialty system call
that most applications will never call.

>> >> If you don't trust userspace that needs to be spelled out very clearly.
>> >> You need to talk about what your threat models are.
>> >> 
>> >> If the only justification is so that that we can't boot windows if
>> >> someone hacks into userspace it has my nack because that is another kind
>> >> of complete non-sense.
>> >
>> > The usecase is the ability to gate the kexec_load usage in stock
>> > kernels.
>> 
>> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
>
> It isn't a matter of kexec_load already being gated, but of wanting a
> single place for defining a system wide policy, as described above.

Signing is only a tool to enforce a policy.  Signing by itself is not a
policy.  Enforcing any quality controls in the signed executables should
trivially prevent kexec_load from being used.

Eric

[PATCH 0/3] kexec: limit kexec_load syscall

[Eric W. Biederman]

Mimi Zohar <zohar@linux.vnet.ibm.com> writes:

> On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
>> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> 
>> > [Cc'ing Kees and kernel-hardening]
>> >
>> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
>> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
>> >> 
>> >> > In environments that require the kexec kernel image to be signed, prevent
>> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
>> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
>> >> > call to security_kernel_read_file() in kexec_load_check().
>> >> 
>> >> Having thought about it some more this justification for these changes
>> >> does not work.  The functionality of kexec_load is already root-only.
>> >> So in environments that require the kernel image to be signed just don't
>> >> use kexec_load.  Possibly even compile kexec_load out to save space
>> >> because you will never need it.  You don't need a new security hook to
>> >> do any of that.  Userspace is a very fine mechanism for being the
>> >> instrument of policy.
>> >
>> > True, for those building their own kernel, they can disable the old
>> > syscalls. ?The concern is not for those building their own kernels,
>> > but for those using stock kernels. ?
>> >
>> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
>> > IMA specific hook, other LSMs can piggy back on top of it. ?Currently,
>> > both load_pin and SELinux are gating the kernel module syscalls based
>> > on security_kernel_read_file.
>> >
>> > If there was a similar option for the kernel image, I'm pretty sure
>> > other LSMs would use it.
>> >
>> > From an IMA perspective, there needs to be some method for only
>> > allowing signed code to be loaded, executed, etc. - kernel modules,
>> > kernel image/initramfs, firmware, policies.
>> 
>> What is the IMA perspective.  Why can't IMA trust appropriately
>> authorized userspace?
>
> Suppose a system owner wants to define a system wide policy that
> requires all code be signed - kernel modules, firmware, kexec image &
> initramfs, executables, mmapped files, etc - without having to rebuild
> the kernel. ?Without a call in kexec_load that isn't possible.

Of course it is.  You just make it a requirement that before an
executable will be signed it will be audited to see that it doesn't
call sys_kexec_load.  Signing presumably means something.  So it should
not be hard to enforce a policy like that on a specialty system call
that most applications will never call.

>> >> If you don't trust userspace that needs to be spelled out very clearly.
>> >> You need to talk about what your threat models are.
>> >> 
>> >> If the only justification is so that that we can't boot windows if
>> >> someone hacks into userspace it has my nack because that is another kind
>> >> of complete non-sense.
>> >
>> > The usecase is the ability to gate the kexec_load usage in stock
>> > kernels.
>> 
>> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
>
> It isn't a matter of kexec_load already being gated, but of wanting a
> single place for defining a system wide policy, as described above.

Signing is only a tool to enforce a policy.  Signing by itself is not a
policy.  Enforcing any quality controls in the signed executables should
trivially prevent kexec_load from being used.

Eric
--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 18:03 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > [Cc'ing Kees and kernel-hardening]
> >> >
> >> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> >> 
> >> >> > In environments that require the kexec kernel image to be signed, prevent
> >> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> >> > call to security_kernel_read_file() in kexec_load_check().
> >> >> 
> >> >> Having thought about it some more this justification for these changes
> >> >> does not work.  The functionality of kexec_load is already root-only.
> >> >> So in environments that require the kernel image to be signed just don't
> >> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> >> because you will never need it.  You don't need a new security hook to
> >> >> do any of that.  Userspace is a very fine mechanism for being the
> >> >> instrument of policy.
> >> >
> >> > True, for those building their own kernel, they can disable the old
> >> > syscalls.  The concern is not for those building their own kernels,
> >> > but for those using stock kernels.  
> >> >
> >> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> >> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> >> > both load_pin and SELinux are gating the kernel module syscalls based
> >> > on security_kernel_read_file.
> >> >
> >> > If there was a similar option for the kernel image, I'm pretty sure
> >> > other LSMs would use it.
> >> >
> >> > From an IMA perspective, there needs to be some method for only
> >> > allowing signed code to be loaded, executed, etc. - kernel modules,
> >> > kernel image/initramfs, firmware, policies.
> >> 
> >> What is the IMA perspective.  Why can't IMA trust appropriately
> >> authorized userspace?
> >
> > Suppose a system owner wants to define a system wide policy that
> > requires all code be signed - kernel modules, firmware, kexec image &
> > initramfs, executables, mmapped files, etc - without having to rebuild
> > the kernel.  Without a call in kexec_load that isn't possible.
> 
> Of course it is.  You just make it a requirement that before an
> executable will be signed it will be audited to see that it doesn't
> call sys_kexec_load.  Signing presumably means something.  So it should
> not be hard to enforce a policy like that on a specialty system call
> that most applications will never call.

Initially I'm hoping that files will simply come signed, providing
file provenance.  Anything else is gravy.

> >> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> >> You need to talk about what your threat models are.
> >> >> 
> >> >> If the only justification is so that that we can't boot windows if
> >> >> someone hacks into userspace it has my nack because that is another kind
> >> >> of complete non-sense.
> >> >
> >> > The usecase is the ability to gate the kexec_load usage in stock
> >> > kernels.
> >> 
> >> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
> >
> > It isn't a matter of kexec_load already being gated, but of wanting a
> > single place for defining a system wide policy, as described above.
> 
> Signing is only a tool to enforce a policy.  Signing by itself is not a
> policy.  Enforcing any quality controls in the signed executables should
> trivially prevent kexec_load from being used.

Existing kernels might not support the newer kexec_file_load syscall,
so packages are currently being built to try one syscall and fallback
to using the other one.  In this case, it has nothing to do with
quality control.

Mimi

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 18:03 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > [Cc'ing Kees and kernel-hardening]
> >> >
> >> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> >> 
> >> >> > In environments that require the kexec kernel image to be signed, prevent
> >> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> >> > call to security_kernel_read_file() in kexec_load_check().
> >> >> 
> >> >> Having thought about it some more this justification for these changes
> >> >> does not work.  The functionality of kexec_load is already root-only.
> >> >> So in environments that require the kernel image to be signed just don't
> >> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> >> because you will never need it.  You don't need a new security hook to
> >> >> do any of that.  Userspace is a very fine mechanism for being the
> >> >> instrument of policy.
> >> >
> >> > True, for those building their own kernel, they can disable the old
> >> > syscalls.  The concern is not for those building their own kernels,
> >> > but for those using stock kernels.  
> >> >
> >> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> >> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> >> > both load_pin and SELinux are gating the kernel module syscalls based
> >> > on security_kernel_read_file.
> >> >
> >> > If there was a similar option for the kernel image, I'm pretty sure
> >> > other LSMs would use it.
> >> >
> >> > From an IMA perspective, there needs to be some method for only
> >> > allowing signed code to be loaded, executed, etc. - kernel modules,
> >> > kernel image/initramfs, firmware, policies.
> >> 
> >> What is the IMA perspective.  Why can't IMA trust appropriately
> >> authorized userspace?
> >
> > Suppose a system owner wants to define a system wide policy that
> > requires all code be signed - kernel modules, firmware, kexec image &
> > initramfs, executables, mmapped files, etc - without having to rebuild
> > the kernel.  Without a call in kexec_load that isn't possible.
> 
> Of course it is.  You just make it a requirement that before an
> executable will be signed it will be audited to see that it doesn't
> call sys_kexec_load.  Signing presumably means something.  So it should
> not be hard to enforce a policy like that on a specialty system call
> that most applications will never call.

Initially I'm hoping that files will simply come signed, providing
file provenance.  Anything else is gravy.

> >> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> >> You need to talk about what your threat models are.
> >> >> 
> >> >> If the only justification is so that that we can't boot windows if
> >> >> someone hacks into userspace it has my nack because that is another kind
> >> >> of complete non-sense.
> >> >
> >> > The usecase is the ability to gate the kexec_load usage in stock
> >> > kernels.
> >> 
> >> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
> >
> > It isn't a matter of kexec_load already being gated, but of wanting a
> > single place for defining a system wide policy, as described above.
> 
> Signing is only a tool to enforce a policy.  Signing by itself is not a
> policy.  Enforcing any quality controls in the signed executables should
> trivially prevent kexec_load from being used.

Existing kernels might not support the newer kexec_file_load syscall,
so packages are currently being built to try one syscall and fallback
to using the other one.  In this case, it has nothing to do with
quality control.

Mimi


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 18:03 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > [Cc'ing Kees and kernel-hardening]
> >> >
> >> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> >> 
> >> >> > In environments that require the kexec kernel image to be signed, prevent
> >> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> >> > call to security_kernel_read_file() in kexec_load_check().
> >> >> 
> >> >> Having thought about it some more this justification for these changes
> >> >> does not work.  The functionality of kexec_load is already root-only.
> >> >> So in environments that require the kernel image to be signed just don't
> >> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> >> because you will never need it.  You don't need a new security hook to
> >> >> do any of that.  Userspace is a very fine mechanism for being the
> >> >> instrument of policy.
> >> >
> >> > True, for those building their own kernel, they can disable the old
> >> > syscalls.  The concern is not for those building their own kernels,
> >> > but for those using stock kernels.  
> >> >
> >> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> >> > IMA specific hook, other LSMs can piggy back on top of it.  Currently,
> >> > both load_pin and SELinux are gating the kernel module syscalls based
> >> > on security_kernel_read_file.
> >> >
> >> > If there was a similar option for the kernel image, I'm pretty sure
> >> > other LSMs would use it.
> >> >
> >> > From an IMA perspective, there needs to be some method for only
> >> > allowing signed code to be loaded, executed, etc. - kernel modules,
> >> > kernel image/initramfs, firmware, policies.
> >> 
> >> What is the IMA perspective.  Why can't IMA trust appropriately
> >> authorized userspace?
> >
> > Suppose a system owner wants to define a system wide policy that
> > requires all code be signed - kernel modules, firmware, kexec image &
> > initramfs, executables, mmapped files, etc - without having to rebuild
> > the kernel.  Without a call in kexec_load that isn't possible.
> 
> Of course it is.  You just make it a requirement that before an
> executable will be signed it will be audited to see that it doesn't
> call sys_kexec_load.  Signing presumably means something.  So it should
> not be hard to enforce a policy like that on a specialty system call
> that most applications will never call.

Initially I'm hoping that files will simply come signed, providing
file provenance.  Anything else is gravy.

> >> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> >> You need to talk about what your threat models are.
> >> >> 
> >> >> If the only justification is so that that we can't boot windows if
> >> >> someone hacks into userspace it has my nack because that is another kind
> >> >> of complete non-sense.
> >> >
> >> > The usecase is the ability to gate the kexec_load usage in stock
> >> > kernels.
> >> 
> >> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
> >
> > It isn't a matter of kexec_load already being gated, but of wanting a
> > single place for defining a system wide policy, as described above.
> 
> Signing is only a tool to enforce a policy.  Signing by itself is not a
> policy.  Enforcing any quality controls in the signed executables should
> trivially prevent kexec_load from being used.

Existing kernels might not support the newer kexec_file_load syscall,
so packages are currently being built to try one syscall and fallback
to using the other one.  In this case, it has nothing to do with
quality control.

Mimi

[PATCH 0/3] kexec: limit kexec_load syscall

[Mimi Zohar]

On Thu, 2018-05-03 at 18:03 -0500, Eric W. Biederman wrote:
> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> 
> > On Thu, 2018-05-03 at 16:38 -0500, Eric W. Biederman wrote:
> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> 
> >> > [Cc'ing Kees and kernel-hardening]
> >> >
> >> > On Thu, 2018-05-03 at 15:13 -0500, Eric W. Biederman wrote:
> >> >> Mimi Zohar <zohar@linux.vnet.ibm.com> writes:
> >> >> 
> >> >> > In environments that require the kexec kernel image to be signed, prevent
> >> >> > using the kexec_load syscall.  In order for LSMs and IMA to differentiate
> >> >> > between kexec_load and kexec_file_load syscalls, this patch set adds a
> >> >> > call to security_kernel_read_file() in kexec_load_check().
> >> >> 
> >> >> Having thought about it some more this justification for these changes
> >> >> does not work.  The functionality of kexec_load is already root-only.
> >> >> So in environments that require the kernel image to be signed just don't
> >> >> use kexec_load.  Possibly even compile kexec_load out to save space
> >> >> because you will never need it.  You don't need a new security hook to
> >> >> do any of that.  Userspace is a very fine mechanism for being the
> >> >> instrument of policy.
> >> >
> >> > True, for those building their own kernel, they can disable the old
> >> > syscalls. ?The concern is not for those building their own kernels,
> >> > but for those using stock kernels. ?
> >> >
> >> > By adding an LSM hook here in the kexec_load syscall, as opposed to an
> >> > IMA specific hook, other LSMs can piggy back on top of it. ?Currently,
> >> > both load_pin and SELinux are gating the kernel module syscalls based
> >> > on security_kernel_read_file.
> >> >
> >> > If there was a similar option for the kernel image, I'm pretty sure
> >> > other LSMs would use it.
> >> >
> >> > From an IMA perspective, there needs to be some method for only
> >> > allowing signed code to be loaded, executed, etc. - kernel modules,
> >> > kernel image/initramfs, firmware, policies.
> >> 
> >> What is the IMA perspective.  Why can't IMA trust appropriately
> >> authorized userspace?
> >
> > Suppose a system owner wants to define a system wide policy that
> > requires all code be signed - kernel modules, firmware, kexec image &
> > initramfs, executables, mmapped files, etc - without having to rebuild
> > the kernel. ?Without a call in kexec_load that isn't possible.
> 
> Of course it is.  You just make it a requirement that before an
> executable will be signed it will be audited to see that it doesn't
> call sys_kexec_load.  Signing presumably means something.  So it should
> not be hard to enforce a policy like that on a specialty system call
> that most applications will never call.

Initially I'm hoping that files will simply come signed, providing
file provenance. ?Anything else is gravy.

> >> >> If you don't trust userspace that needs to be spelled out very clearly.
> >> >> You need to talk about what your threat models are.
> >> >> 
> >> >> If the only justification is so that that we can't boot windows if
> >> >> someone hacks into userspace it has my nack because that is another kind
> >> >> of complete non-sense.
> >> >
> >> > The usecase is the ability to gate the kexec_load usage in stock
> >> > kernels.
> >> 
> >> But kexec_load is already gated.  It requires CAP_SYS_BOOT.
> >
> > It isn't a matter of kexec_load already being gated, but of wanting a
> > single place for defining a system wide policy, as described above.
> 
> Signing is only a tool to enforce a policy.  Signing by itself is not a
> policy.  Enforcing any quality controls in the signed executables should
> trivially prevent kexec_load from being used.

Existing kernels might not support the newer kexec_file_load syscall,
so packages are currently being built to try one syscall and fallback
to using the other one. ?In this case, it has nothing to do with
quality control.

Mimi

--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html