[RFC] mac80211: do not restart ps timer during scan

[RFC] mac80211: do not restart ps timer during scan

[Rajkumar Manoharan]

While leaving oper channel, STA informs sleep state to AP to
stop sending data. Before sending ack for the nullfunc, AP is
still sending the data to STA which restarts ps_timer that
is causing unnecessary nullfunc exchange on timer expiry
when the STA is on offchannel. So don't restart ps_timer
on data reception during scan. This issue was identified by
the following warning.

WARNING: at net/mac80211/tx.c:661 invoke_tx_handlers+0xf07/0x1330 [mac80211]
wlan0: Dropped data frame as no usable bitrate found while scanning and
associated. Target station: 00:03:7f:0b:a6:1b on 5 GHz band
Call Trace:
  [<ffffffffa0413ba7>] invoke_tx_handlers+0xf07/0x1330 [mac80211]
  [<ffffffffa0414056>] ieee80211_tx+0x86/0x2c0 [mac80211]
  [<ffffffffa0414345>] ieee80211_xmit+0xb5/0x1d0 [mac80211]
  [<ffffffffa04037e0>] ieee80211_dynamic_ps_enable_work+0x0/0xb0 [mac80211]
  [<ffffffffa04158cf>] ieee80211_tx_skb+0x4f/0x60 [mac80211]
  [<ffffffffa04026e6>] ieee80211_send_nullfunc+0x46/0x60 [mac80211]
  [<ffffffffa0403885>] ieee80211_dynamic_ps_enable_work+0xa5/0xb0 [mac80211]

Signed-off-by: Rajkumar Manoharan <rmanoharan@atheros.com>
---
 net/mac80211/rx.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 7185c93..0e02ce9 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1929,7 +1929,9 @@ ieee80211_rx_h_data(struct ieee80211_rx_data *rx)
 	dev->stats.rx_bytes += rx->skb->len;
 
 	if (local->ps_sdata && local->hw.conf.dynamic_ps_timeout > 0 &&
-	    !is_multicast_ether_addr(((struct ethhdr *)rx->skb->data)->h_dest)) {
+	    !is_multicast_ether_addr(
+		    ((struct ethhdr *)rx->skb->data)->h_dest) &&
+	    !local->scanning) {
 			mod_timer(&local->dynamic_ps_timer, jiffies +
 			 msecs_to_jiffies(local->hw.conf.dynamic_ps_timeout));
 	}
-- 
1.7.4

[PATCH] ath9k: wakeup hw before stopping beacon queue

[Rajkumar Manoharan]

Before stopping beacon queue ensure that hw is in AWAKE state.
No doing so, causes failure to stop tx dma error message.

Signed-off-by: Rajkumar Manoharan <rmanoharan@atheros.com>
---
 drivers/net/wireless/ath/ath9k/main.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index c9925e9..bfc70c1 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -1411,6 +1411,7 @@ static void ath9k_do_vif_add_setup(struct ieee80211_hw *hw,
 
 	if (ath9k_uses_beacons(vif->type)) {
 		int error;
+		ath9k_ps_wakeup(sc);
 		ath9k_hw_stoptxdma(sc->sc_ah, sc->beacon.beaconq);
 		/* This may fail because upper levels do not have beacons
 		 * properly configured yet.  That's OK, we assume it
@@ -1423,6 +1424,7 @@ static void ath9k_do_vif_add_setup(struct ieee80211_hw *hw,
 			ath9k_reclaim_beacon(sc, vif);
 		else
 			ath_beacon_config(sc, vif);
+		ath9k_ps_restore(sc);
 	}
 }
 
-- 
1.7.4

Re: [RFC] mac80211: do not restart ps timer during scan

[Johannes Berg]

On Tue, 2011-02-01 at 20:04 +0530, Rajkumar Manoharan wrote:
> While leaving oper channel, STA informs sleep state to AP to
> stop sending data. Before sending ack for the nullfunc, AP is
> still sending the data to STA which restarts ps_timer that
> is causing unnecessary nullfunc exchange on timer expiry
> when the STA is on offchannel. So don't restart ps_timer
> on data reception during scan. This issue was identified by
> the following warning.
> 
> WARNING: at net/mac80211/tx.c:661 invoke_tx_handlers+0xf07/0x1330 [mac80211]
> wlan0: Dropped data frame as no usable bitrate found while scanning and
> associated. Target station: 00:03:7f:0b:a6:1b on 5 GHz band
> Call Trace:
>   [<ffffffffa0413ba7>] invoke_tx_handlers+0xf07/0x1330 [mac80211]
>   [<ffffffffa0414056>] ieee80211_tx+0x86/0x2c0 [mac80211]
>   [<ffffffffa0414345>] ieee80211_xmit+0xb5/0x1d0 [mac80211]
>   [<ffffffffa04037e0>] ieee80211_dynamic_ps_enable_work+0x0/0xb0 [mac80211]
>   [<ffffffffa04158cf>] ieee80211_tx_skb+0x4f/0x60 [mac80211]
>   [<ffffffffa04026e6>] ieee80211_send_nullfunc+0x46/0x60 [mac80211]
>   [<ffffffffa0403885>] ieee80211_dynamic_ps_enable_work+0xa5/0xb0 [mac80211]
> 
> Signed-off-by: Rajkumar Manoharan <rmanoharan@atheros.com>
> ---
>  net/mac80211/rx.c |    4 +++-
>  1 files changed, 3 insertions(+), 1 deletions(-)
> 
> diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
> index 7185c93..0e02ce9 100644
> --- a/net/mac80211/rx.c
> +++ b/net/mac80211/rx.c
> @@ -1929,7 +1929,9 @@ ieee80211_rx_h_data(struct ieee80211_rx_data *rx)
>  	dev->stats.rx_bytes += rx->skb->len;
>  
>  	if (local->ps_sdata && local->hw.conf.dynamic_ps_timeout > 0 &&
> -	    !is_multicast_ether_addr(((struct ethhdr *)rx->skb->data)->h_dest)) {
> +	    !is_multicast_ether_addr(
> +		    ((struct ethhdr *)rx->skb->data)->h_dest) &&
> +	    !local->scanning) {

What if we're off-channel due to other work like P2P?

johannes

Re: [RFC] mac80211: do not restart ps timer during scan

[Ben Greear]

On 02/01/2011 06:36 AM, Johannes Berg wrote:
> On Tue, 2011-02-01 at 20:04 +0530, Rajkumar Manoharan wrote:
>> While leaving oper channel, STA informs sleep state to AP to
>> stop sending data. Before sending ack for the nullfunc, AP is
>> still sending the data to STA which restarts ps_timer that
>> is causing unnecessary nullfunc exchange on timer expiry
>> when the STA is on offchannel. So don't restart ps_timer
>> on data reception during scan. This issue was identified by
>> the following warning.
>>
>> WARNING: at net/mac80211/tx.c:661 invoke_tx_handlers+0xf07/0x1330 [mac80211]
>> wlan0: Dropped data frame as no usable bitrate found while scanning and
>> associated. Target station: 00:03:7f:0b:a6:1b on 5 GHz band
>> Call Trace:
>>    [<ffffffffa0413ba7>] invoke_tx_handlers+0xf07/0x1330 [mac80211]
>>    [<ffffffffa0414056>] ieee80211_tx+0x86/0x2c0 [mac80211]
>>    [<ffffffffa0414345>] ieee80211_xmit+0xb5/0x1d0 [mac80211]
>>    [<ffffffffa04037e0>] ieee80211_dynamic_ps_enable_work+0x0/0xb0 [mac80211]
>>    [<ffffffffa04158cf>] ieee80211_tx_skb+0x4f/0x60 [mac80211]
>>    [<ffffffffa04026e6>] ieee80211_send_nullfunc+0x46/0x60 [mac80211]
>>    [<ffffffffa0403885>] ieee80211_dynamic_ps_enable_work+0xa5/0xb0 [mac80211]
>>
>> Signed-off-by: Rajkumar Manoharan<rmanoharan@atheros.com>
>> ---
>>   net/mac80211/rx.c |    4 +++-
>>   1 files changed, 3 insertions(+), 1 deletions(-)
>>
>> diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
>> index 7185c93..0e02ce9 100644
>> --- a/net/mac80211/rx.c
>> +++ b/net/mac80211/rx.c
>> @@ -1929,7 +1929,9 @@ ieee80211_rx_h_data(struct ieee80211_rx_data *rx)
>>   	dev->stats.rx_bytes += rx->skb->len;
>>
>>   	if (local->ps_sdata&&  local->hw.conf.dynamic_ps_timeout>  0&&
>> -	    !is_multicast_ether_addr(((struct ethhdr *)rx->skb->data)->h_dest)) {
>> +	    !is_multicast_ether_addr(
>> +		    ((struct ethhdr *)rx->skb->data)->h_dest)&&
>> +	    !local->scanning) {
>
> What if we're off-channel due to other work like P2P?

And, if I can get the scan-on-channel optimization patch in, this check
would need to be changed to check for scanning & off-channel instead of
just scanning.  I have a helper method in my patch to check if we're on
the oper-channel....

Thanks,
Ben

>
> johannes
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html


-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com

Re: [RFC] mac80211: do not restart ps timer during scan

[Rajkumar Manoharan]

On Tue, Feb 01, 2011 at 08:06:58PM +0530, Johannes Berg wrote:
> On Tue, 2011-02-01 at 20:04 +0530, Rajkumar Manoharan wrote:
> > While leaving oper channel, STA informs sleep state to AP to
> > stop sending data. Before sending ack for the nullfunc, AP is
> > still sending the data to STA which restarts ps_timer that
> > is causing unnecessary nullfunc exchange on timer expiry
> > when the STA is on offchannel. So don't restart ps_timer
> > on data reception during scan. This issue was identified by
> > the following warning.
> > 
> > WARNING: at net/mac80211/tx.c:661 invoke_tx_handlers+0xf07/0x1330 [mac80211]
> > wlan0: Dropped data frame as no usable bitrate found while scanning and
> > associated. Target station: 00:03:7f:0b:a6:1b on 5 GHz band
> > Call Trace:
> >   [<ffffffffa0413ba7>] invoke_tx_handlers+0xf07/0x1330 [mac80211]
> >   [<ffffffffa0414056>] ieee80211_tx+0x86/0x2c0 [mac80211]
> >   [<ffffffffa0414345>] ieee80211_xmit+0xb5/0x1d0 [mac80211]
> >   [<ffffffffa04037e0>] ieee80211_dynamic_ps_enable_work+0x0/0xb0 [mac80211]
> >   [<ffffffffa04158cf>] ieee80211_tx_skb+0x4f/0x60 [mac80211]
> >   [<ffffffffa04026e6>] ieee80211_send_nullfunc+0x46/0x60 [mac80211]
> >   [<ffffffffa0403885>] ieee80211_dynamic_ps_enable_work+0xa5/0xb0 [mac80211]
> > 
> > Signed-off-by: Rajkumar Manoharan <rmanoharan@atheros.com>
> > ---
> >  net/mac80211/rx.c |    4 +++-
> >  1 files changed, 3 insertions(+), 1 deletions(-)
> > 
> > diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
> > index 7185c93..0e02ce9 100644
> > --- a/net/mac80211/rx.c
> > +++ b/net/mac80211/rx.c
> > @@ -1929,7 +1929,9 @@ ieee80211_rx_h_data(struct ieee80211_rx_data *rx)
> >  	dev->stats.rx_bytes += rx->skb->len;
> >  
> >  	if (local->ps_sdata && local->hw.conf.dynamic_ps_timeout > 0 &&
> > -	    !is_multicast_ether_addr(((struct ethhdr *)rx->skb->data)->h_dest)) {
> > +	    !is_multicast_ether_addr(
> > +		    ((struct ethhdr *)rx->skb->data)->h_dest) &&
> > +	    !local->scanning) {
> 
> What if we're off-channel due to other work like P2P?
>
Yes. we need to handle offchannel state too. Is it fine to have a check against sdata->state 
SDATA_STATE_OFFCHANNEL.

--
Rajkumar