[PATCH] mac80211: split out and decrease probe wait time

[PATCH] mac80211: split out and decrease probe wait time

[Johannes Berg]

The time we wait for a probe response after probing an AP due to
beacon loss is currently the same as the monitoring interval, 2s.
This is far too long, APs should respond to probes within a
fraction of that time. To be able to adjust both values, add a
new constant IEEE80211_PROBE_WAIT, use it for checking the probe
response, and adjust it down to 200ms instead of 2 seconds.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
---
 net/mac80211/mlme.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

--- wireless-testing.orig/net/mac80211/mlme.c	2009-05-12 19:58:27.000000000 +0200
+++ wireless-testing/net/mac80211/mlme.c	2009-05-12 19:59:44.000000000 +0200
@@ -33,6 +33,7 @@
 #define IEEE80211_ASSOC_TIMEOUT (HZ / 5)
 #define IEEE80211_ASSOC_MAX_TRIES 3
 #define IEEE80211_MONITORING_INTERVAL (2 * HZ)
+#define IEEE80211_PROBE_WAIT (HZ / 5)
 #define IEEE80211_PROBE_IDLE_TIME (60 * HZ)
 #define IEEE80211_RETRY_AUTH_INTERVAL (1 * HZ)
 
@@ -1205,7 +1206,7 @@ void ieee80211_beacon_loss_work(struct w
 	ieee80211_send_probe_req(sdata, ifmgd->bssid, ifmgd->ssid,
 				 ifmgd->ssid_len, NULL, 0);
 
-	mod_timer(&ifmgd->timer, jiffies + IEEE80211_MONITORING_INTERVAL);
+	mod_timer(&ifmgd->timer, jiffies + IEEE80211_PROBE_WAIT);
 }
 
 void ieee80211_beacon_loss(struct ieee80211_vif *vif)
@@ -1242,7 +1243,7 @@ static void ieee80211_associated(struct 
 	}
 
 	if ((ifmgd->flags & IEEE80211_STA_PROBEREQ_POLL) &&
-	    time_after(jiffies, sta->last_rx + IEEE80211_MONITORING_INTERVAL)) {
+	    time_after(jiffies, sta->last_rx + IEEE80211_PROBE_WAIT)) {
 		printk(KERN_DEBUG "%s: no probe response from AP %pM "
 		       "- disassociating\n",
 		       sdata->dev->name, ifmgd->bssid);
@@ -1258,7 +1259,7 @@ static void ieee80211_associated(struct 
 	if (!((local->hw.flags & IEEE80211_HW_BEACON_FILTER) &&
 	      (local->hw.conf.flags & IEEE80211_CONF_PS)) &&
 	    time_after(jiffies,
-		       ifmgd->last_beacon + IEEE80211_MONITORING_INTERVAL)) {
+		       ifmgd->last_beacon + IEEE80211_PROBE_WAIT)) {
 #ifdef CONFIG_MAC80211_VERBOSE_DEBUG
 		if (net_ratelimit()) {
 			printk(KERN_DEBUG "%s: beacon loss from AP %pM "

[PATCH v2] mac80211: split out and decrease probe wait time

[Johannes Berg]

The time we wait for a probe response after probing an AP due to
beacon loss is currently the same as the monitoring interval, 2s.
This is far too long, APs should respond to probes within a
fraction of that time. To be able to adjust both values, add a
new constant IEEE80211_PROBE_WAIT, use it for checking the probe
response, and adjust it down to 200ms instead of 2 seconds.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
---
Oops, replaced one too many, although we probably should make that bit
smarter too.

 net/mac80211/mlme.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

--- wireless-testing.orig/net/mac80211/mlme.c	2009-05-12 19:58:27.000000000 +0200
+++ wireless-testing/net/mac80211/mlme.c	2009-05-12 20:10:30.000000000 +0200
@@ -33,6 +33,7 @@
 #define IEEE80211_ASSOC_TIMEOUT (HZ / 5)
 #define IEEE80211_ASSOC_MAX_TRIES 3
 #define IEEE80211_MONITORING_INTERVAL (2 * HZ)
+#define IEEE80211_PROBE_WAIT (HZ / 20)
 #define IEEE80211_PROBE_IDLE_TIME (60 * HZ)
 #define IEEE80211_RETRY_AUTH_INTERVAL (1 * HZ)
 
@@ -1205,7 +1206,7 @@ void ieee80211_beacon_loss_work(struct w
 	ieee80211_send_probe_req(sdata, ifmgd->bssid, ifmgd->ssid,
 				 ifmgd->ssid_len, NULL, 0);
 
-	mod_timer(&ifmgd->timer, jiffies + IEEE80211_MONITORING_INTERVAL);
+	mod_timer(&ifmgd->timer, jiffies + IEEE80211_PROBE_WAIT);
 }
 
 void ieee80211_beacon_loss(struct ieee80211_vif *vif)
@@ -1242,7 +1243,7 @@ static void ieee80211_associated(struct 
 	}
 
 	if ((ifmgd->flags & IEEE80211_STA_PROBEREQ_POLL) &&
-	    time_after(jiffies, sta->last_rx + IEEE80211_MONITORING_INTERVAL)) {
+	    time_after(jiffies, sta->last_rx + IEEE80211_PROBE_WAIT)) {
 		printk(KERN_DEBUG "%s: no probe response from AP %pM "
 		       "- disassociating\n",
 		       sdata->dev->name, ifmgd->bssid);

Re: [PATCH v2] mac80211: split out and decrease probe wait time

[Johannes Berg]

You know ... somehow I managed to send out v2 with a wrong interval.

This was supposed to be / 5 not / 20, should I send a separate patch or
would you roll this into the original? The commit log says 200ms, but
the code uses 50ms which is far too short.

johannes

---
 net/mac80211/mlme.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- wireless-testing.orig/net/mac80211/mlme.c	2009-05-19 19:36:00.564057102 +0200
+++ wireless-testing/net/mac80211/mlme.c	2009-05-19 19:36:04.773067791 +0200
@@ -33,7 +33,7 @@
 #define IEEE80211_ASSOC_TIMEOUT (HZ / 5)
 #define IEEE80211_ASSOC_MAX_TRIES 3
 #define IEEE80211_MONITORING_INTERVAL (2 * HZ)
-#define IEEE80211_PROBE_WAIT (HZ / 20)
+#define IEEE80211_PROBE_WAIT (HZ / 5)
 #define IEEE80211_PROBE_IDLE_TIME (60 * HZ)
 #define IEEE80211_RETRY_AUTH_INTERVAL (1 * HZ)
 

Re: [PATCH v2] mac80211: split out and decrease probe wait time

[John W. Linville]

On Tue, May 19, 2009 at 07:37:28PM +0200, Johannes Berg wrote:
> You know ... somehow I managed to send out v2 with a wrong interval.
> 
> This was supposed to be / 5 not / 20, should I send a separate patch or
> would you roll this into the original? The commit log says 200ms, but
> the code uses 50ms which is far too short.

Ok, I'll roll this up...

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.

Re: [PATCH v2] mac80211: split out and decrease probe wait time

[John W. Linville]

On Tue, May 19, 2009 at 07:37:28PM +0200, Johannes Berg wrote:
> You know ... somehow I managed to send out v2 with a wrong interval.
> 
> This was supposed to be / 5 not / 20, should I send a separate patch or
> would you roll this into the original? The commit log says 200ms, but
> the code uses 50ms which is far too short.
> 
> johannes
> 
> ---
>  net/mac80211/mlme.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> --- wireless-testing.orig/net/mac80211/mlme.c	2009-05-19 19:36:00.564057102 +0200
> +++ wireless-testing/net/mac80211/mlme.c	2009-05-19 19:36:04.773067791 +0200
> @@ -33,7 +33,7 @@
>  #define IEEE80211_ASSOC_TIMEOUT (HZ / 5)
>  #define IEEE80211_ASSOC_MAX_TRIES 3
>  #define IEEE80211_MONITORING_INTERVAL (2 * HZ)
> -#define IEEE80211_PROBE_WAIT (HZ / 20)
> +#define IEEE80211_PROBE_WAIT (HZ / 5)
>  #define IEEE80211_PROBE_IDLE_TIME (60 * HZ)
>  #define IEEE80211_RETRY_AUTH_INTERVAL (1 * HZ)

Crud, I didn't realize I had already pushed that patch to Dave M.
Could you submit this as a separate patch?

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.

Re: [PATCH] mac80211: split out and decrease probe wait time

[Kalle Valo]

Johannes Berg <johannes@sipsolutions.net> writes:

> The time we wait for a probe response after probing an AP due to
> beacon loss is currently the same as the monitoring interval, 2s.
> This is far too long, APs should respond to probes within a
> fraction of that time. To be able to adjust both values, add a
> new constant IEEE80211_PROBE_WAIT, use it for checking the probe
> response, and adjust it down to 200ms instead of 2 seconds.

Looks good to me.

> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>

Reviewed-by: Kalle Valo <kalle.valo@nokia.com>

-- 
Kalle Valo