lkml.org 
[lkml]   [2011]   [Mar]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH v2 resend 10/12] proc: hold cred_guard_mutex in check_mem_permission()
Date
Avoid a potential race when task exec's and we get a new ->mm but check against
the old credentials in ptrace_may_access().

Holding of the mutex is implemented by factoring out the body of the code into a
helper function __check_mem_permission(). Performing this factorization now
simplifies upcoming changes and minimizes churn in the diff's.

Signed-off-by: Stephen Wilson <wilsons@start.ca>
---
fs/proc/base.c | 26 ++++++++++++++++++++++----
1 files changed, 22 insertions(+), 4 deletions(-)
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 13c5e8c..084d046 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -191,10 +191,7 @@ static int proc_root_link(struct inode *inode, struct path *path)
return result;
}

-/*
- * Return zero if current may access user memory in @task, -error if not.
- */
-static int check_mem_permission(struct task_struct *task)
+static int __check_mem_permission(struct task_struct *task)
{
/*
* A task can always look at itself, in case it chooses
@@ -222,6 +219,27 @@ static int check_mem_permission(struct task_struct *task)
return -EPERM;
}

+/*
+ * Return zero if current may access user memory in @task, -error if not.
+ */
+static int check_mem_permission(struct task_struct *task)
+{
+ int err;
+
+ /*
+ * Avoid racing if task exec's as we might get a new mm but validate
+ * against old credentials.
+ */
+ err = mutex_lock_killable(&task->signal->cred_guard_mutex);
+ if (err)
+ return err;
+
+ err = __check_mem_permission(task);
+ mutex_unlock(&task->signal->cred_guard_mutex);
+
+ return err;
+}
+
struct mm_struct *mm_for_maps(struct task_struct *task)
{
struct mm_struct *mm;
--
1.7.3.5


\
 
 \ /
  Last update: 2011-03-23 15:53    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site