linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH RESEND] audit: fix wrong timeout value in audit_log_start
@ 2012-09-03 10:52 Andrea Righi
  0 siblings, 0 replies; only message in thread
From: Andrea Righi @ 2012-09-03 10:52 UTC (permalink / raw)
  To: Al Viro, Eric Paris; +Cc: Sean Jenkins, linux-kernel

audit_log_start() does not properly check timeout values when
audit_backlog_limit is exceeded, for example it is possible to use
negative timeout values with schedule_timeout() triggering error
messages like this:

  schedule_timeout: wrong timeout value ffffffffffff54e4

Be sure to never set negative values for the timeout.

Also, do a small refactoring to improve code readability.

Reported-by: Sean Jenkins <sean@bluehost.com>
Signed-off-by: Andrea Righi <andrea@betterlinux.com>
---
 kernel/audit.c |   46 ++++++++++++++++++++++++++++++----------------
 1 file changed, 30 insertions(+), 16 deletions(-)

diff --git a/kernel/audit.c b/kernel/audit.c
index ea3b7b6..362d9b4 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -1123,6 +1123,28 @@ static inline void audit_get_stamp(struct audit_context *ctx,
 	}
 }
 
+/* Wait for auditd to drain the queue a little */
+static void audit_log_congestion_wait(unsigned long timeout_start)
+{
+	DECLARE_WAITQUEUE(wait, current);
+
+	set_current_state(TASK_INTERRUPTIBLE);
+	add_wait_queue(&audit_backlog_wait, &wait);
+
+	if (audit_backlog_limit &&
+	    skb_queue_len(&audit_skb_queue) > audit_backlog_limit) {
+		unsigned long deadline, now = ACCESS_ONCE(jiffies);
+
+		WARN_ON(time_after(timeout_start, now));
+		deadline = timeout_start + audit_backlog_wait_time;
+		if (time_before(now, deadline))
+			schedule_timeout(deadline - now);
+	}
+
+	__set_current_state(TASK_RUNNING);
+	remove_wait_queue(&audit_backlog_wait, &wait);
+}
+
 /* Obtain an audit buffer.  This routine does locking to obtain the
  * audit buffer, but then no locking is required for calls to
  * audit_log_*format.  If the tsk is a task that is currently in a
@@ -1166,22 +1188,14 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask,
 		reserve = 5; /* Allow atomic callers to go up to five
 				entries over the normal backlog limit */
 
-	while (audit_backlog_limit
-	       && skb_queue_len(&audit_skb_queue) > audit_backlog_limit + reserve) {
-		if (gfp_mask & __GFP_WAIT && audit_backlog_wait_time
-		    && time_before(jiffies, timeout_start + audit_backlog_wait_time)) {
-
-			/* Wait for auditd to drain the queue a little */
-			DECLARE_WAITQUEUE(wait, current);
-			set_current_state(TASK_INTERRUPTIBLE);
-			add_wait_queue(&audit_backlog_wait, &wait);
-
-			if (audit_backlog_limit &&
-			    skb_queue_len(&audit_skb_queue) > audit_backlog_limit)
-				schedule_timeout(timeout_start + audit_backlog_wait_time - jiffies);
-
-			__set_current_state(TASK_RUNNING);
-			remove_wait_queue(&audit_backlog_wait, &wait);
+	while (audit_backlog_limit &&
+	       skb_queue_len(&audit_skb_queue) >
+				audit_backlog_limit + reserve) {
+		if ((gfp_mask & __GFP_WAIT) &&
+		    audit_backlog_wait_time &&
+		    time_is_after_jiffies(timeout_start +
+					audit_backlog_wait_time)) {
+			audit_log_congestion_wait(timeout_start);
 			continue;
 		}
 		if (audit_rate_check() && printk_ratelimit())
-- 
1.7.9.5


^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2012-09-03 10:53 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-09-03 10:52 [PATCH RESEND] audit: fix wrong timeout value in audit_log_start Andrea Righi

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).