[RFC PATCH v1] ima: verify mprotect change is consistent with mmap policy

[RFC PATCH v1] ima: verify mprotect change is consistent with mmap policy

[Mimi Zohar]

Files can be mmap'ed read/write and later changed to execute to circumvent
IMA's mmap appraise policy rules.  Due to locking issues (mmap semaphore
would be taken prior to i_mutex), files can not be measured or appraised at
this point.  Eliminate this integrity gap, by denying the mprotect
PROT_EXECUTE change, if an mmap appraise policy rule exists.

On mprotect change success, return 0.  On failure, return -EACESS.

Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
---
Changelog v1:
- Reverse tests to remove code indentation. (Lakshmi Ramasubramanian)
- General code cleanup, including adding comments.

 include/linux/ima.h               |  7 ++++++
 security/integrity/ima/ima_main.c | 51 +++++++++++++++++++++++++++++++++++++++
 security/security.c               |  7 +++++-
 3 files changed, 64 insertions(+), 1 deletion(-)

diff --git a/include/linux/ima.h b/include/linux/ima.h
index aefe758f4466..9164e1534ec9 100644
--- a/include/linux/ima.h
+++ b/include/linux/ima.h
@@ -18,6 +18,7 @@ extern int ima_file_check(struct file *file, int mask);
 extern void ima_post_create_tmpfile(struct inode *inode);
 extern void ima_file_free(struct file *file);
 extern int ima_file_mmap(struct file *file, unsigned long prot);
+extern int ima_file_mprotect(struct vm_area_struct *vma, unsigned long prot);
 extern int ima_load_data(enum kernel_load_data_id id);
 extern int ima_read_file(struct file *file, enum kernel_read_file_id id);
 extern int ima_post_read_file(struct file *file, void *buf, loff_t size,
@@ -70,6 +71,12 @@ static inline int ima_file_mmap(struct file *file, unsigned long prot)
 	return 0;
 }
 
+static inline int ima_file_mprotect(struct vm_area_struct *vma,
+				    unsigned long prot)
+{
+	return 0;
+}
+
 static inline int ima_load_data(enum kernel_load_data_id id)
 {
 	return 0;
diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
index f96f151294e6..800fb3bba418 100644
--- a/security/integrity/ima/ima_main.c
+++ b/security/integrity/ima/ima_main.c
@@ -394,6 +394,57 @@ int ima_file_mmap(struct file *file, unsigned long prot)
 }
 
 /**
+ * ima_file_mprotect - based on policy, limit mprotect change
+ * @prot: contains the protection that will be applied by the kernel.
+ *
+ * Files can be mmap'ed read/write and later changed to execute to circumvent
+ * IMA's mmap appraisal policy rules.  Due to locking issues (mmap semaphore
+ * would be taken before i_mutex), files can not be measured or appraised at
+ * this point.  Eliminate this integrity gap by denying the mprotect
+ * PROT_EXECUTE change, if an mmap appraise policy rule exists.
+ *
+ * On mprotect change success, return 0.  On failure, return -EACESS.
+ */
+int ima_file_mprotect(struct vm_area_struct *vma, unsigned long prot)
+{
+	struct ima_template_desc *template;
+	struct file *file = vma->vm_file;
+	char filename[NAME_MAX];
+	char *pathbuf = NULL;
+	const char *pathname = NULL;
+	struct inode *inode;
+	int result = 0;
+	int action;
+	u32 secid;
+	int pcr;
+
+	/* Is mprotect making an mmap'ed file executable? */
+	if (!vma->vm_file || !(prot & PROT_EXEC) || (vma->vm_flags & VM_EXEC))
+		return 0;
+
+	security_task_getsecid(current, &secid);
+	inode = file_inode(vma->vm_file);
+	action = ima_get_action(inode, current_cred(), secid, MAY_EXEC,
+				MMAP_CHECK, &pcr, &template, 0);
+
+	/* Is the mmap'ed file in policy? */
+	if (!(action & (IMA_MEASURE | IMA_APPRAISE_SUBMASK)))
+		return 0;
+
+	if (action & IMA_APPRAISE_SUBMASK)
+		result = -EPERM;
+
+	file = vma->vm_file;
+	pathname = ima_d_path(&file->f_path, &pathbuf, filename);
+	integrity_audit_msg(AUDIT_INTEGRITY_DATA, inode, pathname,
+			    "collect_data", "failed-mprotect", result, 0);
+	if (pathbuf)
+		__putname(pathbuf);
+
+	return result;
+}
+
+/**
  * ima_bprm_check - based on policy, collect/store measurement.
  * @bprm: contains the linux_binprm structure
  *
diff --git a/security/security.c b/security/security.c
index 7fed24b9d57e..dd0917c5bfe9 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1512,7 +1512,12 @@ int security_mmap_addr(unsigned long addr)
 int security_file_mprotect(struct vm_area_struct *vma, unsigned long reqprot,
 			    unsigned long prot)
 {
-	return call_int_hook(file_mprotect, 0, vma, reqprot, prot);
+	int ret;
+
+	ret = call_int_hook(file_mprotect, 0, vma, reqprot, prot);
+	if (ret)
+		return ret;
+	return ima_file_mprotect(vma, prot);
 }
 
 int security_file_lock(struct file *file, unsigned int cmd)
-- 
2.7.5

Re: [RFC PATCH v1] ima: verify mprotect change is consistent with mmap policy

[Lakshmi Ramasubramanian]

On 5/5/20 10:30 AM, Mimi Zohar wrote:

> Files can be mmap'ed read/write and later changed to execute to circumvent
> IMA's mmap appraise policy rules.  Due to locking issues (mmap semaphore
> would be taken prior to i_mutex), files can not be measured or appraised at
> this point.  Eliminate this integrity gap, by denying the mprotect
> PROT_EXECUTE change, if an mmap appraise policy rule exists.
> 
> On mprotect change success, return 0.  On failure, return -EACESS.
> 
> Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
> ---
> Changelog v1:
> - Reverse tests to remove code indentation. (Lakshmi Ramasubramanian)
> - General code cleanup, including adding comments.
> 
>   include/linux/ima.h               |  7 ++++++
>   security/integrity/ima/ima_main.c | 51 +++++++++++++++++++++++++++++++++++++++
>   security/security.c               |  7 +++++-
>   3 files changed, 64 insertions(+), 1 deletion(-)

Reviewed-by: Lakshmi Ramasubramanian <nramas@linux.microsoft.com>