audit/stable-5.17 PR 20220131

-----BEGIN PGP SIGNATURE-----
 
 iQJIBAABCAAyFiEES0KozwfymdVUl37v6iDy2pc3iXMFAmH4I8oUHHBhdWxAcGF1
 bC1tb29yZS5jb20ACgkQ6iDy2pc3iXP9zw//ZAU42ylQXvGfLJbCsbZHKq7H/Ljo
 QKbwlSKh+aP+IXcRxnh9fV3vTkc7LkHZfrhGx35aHlS1HK8XIa/XHHNTPHfaEKai
 iDiXwICP5j1JysWCKJXu0uhe7juW3ko7+bQyI8MYBAeLKrbC1F04oswSgVNuX2bq
 aMyD9+GdAv7PiVVw0oc5+hKFS/8q/VRdbCsJmESKUDhthS3hqH8wZJqy37aYIpF1
 /yqkvEIts+GzeQrSWvsGL+O720GIyZ8V2/cEH2y+pnGYgoYdlXhD1CccXOOCocb8
 M/6uQZTgQiEVf1LMuu/WIW6CzrRQIjOt4SNU0cXLkWlgxAN1p5b9sP+YcncWU23N
 zbIACFiFVz1ZhxHT0AEVs+thdZrF6CJX0xfsb+GvJJeYy9aw11s7VGlYhaM+1haG
 8oeYmtjQ+rjjkEKMUcOWQYxRvCZIsI6z5JlqoFC0zuJda1k3418LyDZARwCUrm1c
 6QY35M0HHxa8k7TWtvJ6aopxM4pg+ZL8WS0shULHRqw/NprvYc6KeZc0/VNyFojJ
 S4wa+Z6rXoYIvGReeDkUOiJjigW78/kyQR2rsxHWaGlDfL8+bALdnorkTbU2G4oG
 Jl1HQdOSuAeAT/D7w/UmKFXDBPXHU77sLnjxiCDMmbmVY6Vxja/hZD/wy9RyyK4K
 UAXDJNwhndHsgX4=
 =sFi4
 -----END PGP SIGNATURE-----

Merge tag 'audit-pr-20220131' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit

Pull audit fix from Paul Moore:
 "A single audit patch to fix problems relating to audit queuing and
  system responsiveness when "audit=1" is specified on the kernel
  command line and the audit daemon is SIGSTOP'd for an extended period
  of time"

* tag 'audit-pr-20220131' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit:
  audit: improve audit queue handling when "audit=1" on cmdline

kernel/audit.c

@@ -541,20 +541,22 @@ static void kauditd_printk_skb(struct sk_buff *skb)
 /**
  * kauditd_rehold_skb - Handle a audit record send failure in the hold queue
  * @skb: audit record
+ * @error: error code (unused)
  *
  * Description:
  * This should only be used by the kauditd_thread when it fails to flush the
  * hold queue.
  */
-static void kauditd_rehold_skb(struct sk_buff *skb)
+static void kauditd_rehold_skb(struct sk_buff *skb, __always_unused int error)
 {
-	/* put the record back in the queue at the same place */
-	skb_queue_head(&audit_hold_queue, skb);
+	/* put the record back in the queue */
+	skb_queue_tail(&audit_hold_queue, skb);
 }
 
 /**
  * kauditd_hold_skb - Queue an audit record, waiting for auditd
  * @skb: audit record
+ * @error: error code
  *
  * Description:
  * Queue the audit record, waiting for an instance of auditd.  When this
@@ -564,19 +566,31 @@ static void kauditd_rehold_skb(struct sk_buff *skb)
  * and queue it, if we have room.  If we want to hold on to the record, but we
  * don't have room, record a record lost message.
  */
-static void kauditd_hold_skb(struct sk_buff *skb)
+static void kauditd_hold_skb(struct sk_buff *skb, int error)
 {
 	/* at this point it is uncertain if we will ever send this to auditd so
 	 * try to send the message via printk before we go any further */
 	kauditd_printk_skb(skb);
 
 	/* can we just silently drop the message? */
-	if (!audit_default) {
-		kfree_skb(skb);
+	if (!audit_default)
+		goto drop;
+
+	/* the hold queue is only for when the daemon goes away completely,
+	 * not -EAGAIN failures; if we are in a -EAGAIN state requeue the
+	 * record on the retry queue unless it's full, in which case drop it
+	 */
+	if (error == -EAGAIN) {
+		if (!audit_backlog_limit ||
+		    skb_queue_len(&audit_retry_queue) < audit_backlog_limit) {
+			skb_queue_tail(&audit_retry_queue, skb);
 			return;
 		}
+		audit_log_lost("kauditd retry queue overflow");
+		goto drop;
+	}
 
-	/* if we have room, queue the message */
+	/* if we have room in the hold queue, queue the message */
 	if (!audit_backlog_limit ||
 	    skb_queue_len(&audit_hold_queue) < audit_backlog_limit) {
 		skb_queue_tail(&audit_hold_queue, skb);
@@ -585,24 +599,32 @@ static void kauditd_hold_skb(struct sk_buff *skb)
 
 	/* we have no other options - drop the message */
 	audit_log_lost("kauditd hold queue overflow");
+drop:
 	kfree_skb(skb);
 }
 
 /**
  * kauditd_retry_skb - Queue an audit record, attempt to send again to auditd
  * @skb: audit record
+ * @error: error code (unused)
  *
  * Description:
  * Not as serious as kauditd_hold_skb() as we still have a connected auditd,
  * but for some reason we are having problems sending it audit records so
  * queue the given record and attempt to resend.
  */
-static void kauditd_retry_skb(struct sk_buff *skb)
+static void kauditd_retry_skb(struct sk_buff *skb, __always_unused int error)
 {
-	/* NOTE: because records should only live in the retry queue for a
-	 * short period of time, before either being sent or moved to the hold
-	 * queue, we don't currently enforce a limit on this queue */
+	if (!audit_backlog_limit ||
+	    skb_queue_len(&audit_retry_queue) < audit_backlog_limit) {
 		skb_queue_tail(&audit_retry_queue, skb);
+		return;
+	}
+
+	/* we have to drop the record, send it via printk as a last effort */
+	kauditd_printk_skb(skb);
+	audit_log_lost("kauditd retry queue overflow");
+	kfree_skb(skb);
 }
 
 /**
@@ -640,7 +662,7 @@ static void auditd_reset(const struct auditd_connection *ac)
 	/* flush the retry queue to the hold queue, but don't touch the main
 	 * queue since we need to process that normally for multicast */
 	while ((skb = skb_dequeue(&audit_retry_queue)))
-		kauditd_hold_skb(skb);
+		kauditd_hold_skb(skb, -ECONNREFUSED);
 }
 
 /**
@@ -714,16 +736,18 @@ static int kauditd_send_queue(struct sock *sk, u32 portid,
 			      struct sk_buff_head *queue,
 			      unsigned int retry_limit,
 			      void (*skb_hook)(struct sk_buff *skb),
-			      void (*err_hook)(struct sk_buff *skb))
+			      void (*err_hook)(struct sk_buff *skb, int error))
 {
 	int rc = 0;
-	struct sk_buff *skb;
+	struct sk_buff *skb = NULL;
+	struct sk_buff *skb_tail;
 	unsigned int failed = 0;
 
 	/* NOTE: kauditd_thread takes care of all our locking, we just use
 	 *       the netlink info passed to us (e.g. sk and portid) */
 
-	while ((skb = skb_dequeue(queue))) {
+	skb_tail = skb_peek_tail(queue);
+	while ((skb != skb_tail) && (skb = skb_dequeue(queue))) {
 		/* call the skb_hook for each skb we touch */
 		if (skb_hook)
 			(*skb_hook)(skb);
@@ -731,7 +755,7 @@ static int kauditd_send_queue(struct sock *sk, u32 portid,
 		/* can we send to anyone via unicast? */
 		if (!sk) {
 			if (err_hook)
-				(*err_hook)(skb);
+				(*err_hook)(skb, -ECONNREFUSED);
 			continue;
 		}
 
@@ -745,7 +769,7 @@ retry:
 			    rc == -ECONNREFUSED || rc == -EPERM) {
 				sk = NULL;
 				if (err_hook)
-					(*err_hook)(skb);
+					(*err_hook)(skb, rc);
 				if (rc == -EAGAIN)
 					rc = 0;
 				/* continue to drain the queue */