[Bridge] [PATCH 2.4] Bridge STP message age

[Bridge] [PATCH 2.4] Bridge STP message age

[Stephen Hemminger]

This is the 2.4 version which incorporates the observations from Kishore
about 2.6. The existing 2.4 code set the age properly, but did not increment
correctly for HZ=1000 and could send a stale (expired) configuration.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>

diff -Nru a/net/bridge/br_stp.c b/net/bridge/br_stp.c
--- a/net/bridge/br_stp.c	2004-06-29 15:30:32 -07:00
+++ b/net/bridge/br_stp.c	2004-06-29 15:30:32 -07:00
@@ -20,7 +20,10 @@
 #include "br_private.h"
 #include "br_private_stp.h"
 
-
+/* since time values in bpdu are in jiffies and then scaled (1/256)
+ * before sending, make sure that is at least one.
+ */
+#define MESSAGE_AGE_INCR	((HZ < 256) ? 1 : (HZ/256))
 
 /* called under ioctl_lock or bridge lock */
 int br_is_root_bridge(struct net_bridge *br)
@@ -160,24 +163,26 @@
 	bpdu.root_path_cost = br->root_path_cost;
 	bpdu.bridge_id = br->bridge_id;
 	bpdu.port_id = p->port_id;
-	bpdu.message_age = 0;
-	if (!br_is_root_bridge(br)) {
+	if (br_is_root_bridge(br)) 
+		bpdu.message_age = 0;
+	else {
 		struct net_bridge_port *root;
-		unsigned long age;
 
 		root = br_get_port(br, br->root_port);
-		age = br_timer_get_residue(&root->message_age_timer) + 1;
-		bpdu.message_age = age;
+		bpdu.message_age =  br_timer_get_residue(&root->message_age_timer)
+			+ MESSAGE_AGE_INCR;
 	}
 	bpdu.max_age = br->max_age;
 	bpdu.hello_time = br->hello_time;
 	bpdu.forward_delay = br->forward_delay;
 
-	br_send_config_bpdu(p, &bpdu);
+	if (bpdu.message_age < br->max_age) {
+		br_send_config_bpdu(p, &bpdu);
 
-	p->topology_change_ack = 0;
-	p->config_pending = 0;
-	br_timer_set(&p->hold_timer, jiffies);
+		p->topology_change_ack = 0;
+		p->config_pending = 0;
+		br_timer_set(&p->hold_timer, jiffies);
+	}
 }
 
 /* called under bridge lock */

[Bridge] Re: [PATCH 2.4] Bridge STP message age

[David S. Miller]

On Tue, 29 Jun 2004 15:37:40 -0700
Stephen Hemminger <shemminger@osdl.org> wrote:

> This is the 2.4 version which incorporates the observations from Kishore
> about 2.6. The existing 2.4 code set the age properly, but did not increment
> correctly for HZ=1000 and could send a stale (expired) configuration.

Both 2.4.x and 2.6.x variants applied, thanks Stephen.