Re: [Ilw] Re: 6300agn: queue stuck and driver doesn't recover

["Guy, Wey-Yi" <wey-yi.w.guy@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 4599 bytes --]

Hi,

On Sun, 2012-02-05 at 16:45 +0100, Alexander Schnaidt wrote:
> On Sat, Feb 4, 2012 at 11:42 PM, Andrew Lutomirski <luto@mit.edu> wrote:
> >
> > I've recently started to notice wireless failures -- after being
> > connected for a few minutes, the connection dies.  Other devices on
> > the same network continue to work.  'iw dev wlan0 disconnect' will fix
> > it.
> >
> > I'm not at all sure, but I think this is a 3.2 regression.  My kernel
> >  is 3.2.2-1.fc16.x86_64.
> Hi!
> 
> I am experiencing a similar regression on every kernel >=3.2.
> Connected to a wpa protected AP, every application
> loses it's connection after a period of time. If I let it be, it
> eventually reconnects
> and continues for a while until the cycle repeats itself.
> 
> 03:00.0 Network controller: Intel Corporation Ultimate N WiFi Link 5300
>         Subsystem: Intel Corporation Device 1011
>         Flags: bus master, fast devsel, latency 0, IRQ 48
>         Memory at f2500000 (64-bit, non-prefetchable) [size=8K]
>         Capabilities: <access denied>
>         Kernel driver in use: iwlwifi
> 
> Networkmanager, netcfg or wicd do not report any problems. The logs
> are clean and only ever mention the reconnection-event itself.
> 
> Now, running wpa_supplicant interactively spits out some info:
> 
> [alex@lx200s ~]$ sudo wpa_supplicant -D wext -i wlan0 -c wpstest.conf
> Password:
> Trying to associate with 00:1f:3f:13:47:1d (SSID='MyFancyAP' freq=2417 MHz)
> Associated with 00:1f:3f:13:47:1d
> WPA: Key negotiation completed with 00:1f:3f:13:47:1d [PTK=TKIP GTK=TKIP]
> CTRL-EVENT-CONNECTED - Connection to 00:1f:3f:13:47:1d
> completed (auth) [id=0 id_str=]
> WPA: Group rekeying completed with 00:1f:3f:13:47:1d [GTK=TKIP]
> WPA: Group rekeying completed with 00:1f:3f:13:47:1d [GTK=TKIP]
> CTRL-EVENT-DISCONNECTED bssid=00:1f:3f:13:47:1d reason=0
> Trying to associate with 00:1f:3f:13:47:1d (SSID='MyFancyAP' freq=2417 MHz)
> Associated with 00:1f:3f:13:47:1d
> WPA: Key negotiation completed with 00:1f:3f:13:47:1d [PTK=TKIP GTK=TKIP]
> CTRL-EVENT-CONNECTED - Connection to 00:1f:3f:13:47:1d
> completed (reauth) [id=0 id_str=]
> ^CCTRL-EVENT-TERMINATING - signal 2 received
> 
> The connection *always* stalls at the second group rekeying event.
> When the third group rekeying happens wpa_supplicant(?) re-associates the
> connection and the cycle repeats.
> 
> Here's the the dmesg output during the time frame:
> 
> [ 126.172145] iwlwifi 0000:03:00.0: L1 Disabled; Enabling L0S
> [ 126.172530] iwlwifi 0000:03:00.0: Radio type=0x0-0x2-0x0
> [ 126.322917] iwlwifi 0000:03:00.0: L1 Disabled; Enabling L0S
> [ 126.323315] iwlwifi 0000:03:00.0: Radio type=0x0-0x2-0x0
> [ 129.644682] wlan0: authenticate with 00:1f:3f:13:47:1d (try 1)
> [ 129.647687] wlan0: authenticated
> [ 129.649798] wlan0: associate with 00:1f:3f:13:47:1d (try 1)
> [ 129.653886] wlan0: RX AssocResp from 00:1f:3f:13:47:1d
> (capab=0x31 status=0 aid=2)
> [ 129.653895] wlan0: associated
> [ 1506.536175] wlan0: deauthenticated from 00:1f:3f:13:47:1d (Reason: 2)
> [ 1506.600035] cfg80211: Calling CRDA to update world regulatory domain
> [ 1509.857439] wlan0: authenticate with 00:1f:3f:13:47:1d (try 1)
> [ 1509.860511] wlan0: authenticated
> [ 1509.862438] wlan0: associate with 00:1f:3f:13:47:1d (try 1)
> [ 1509.866443] wlan0: RX AssocResp from 00:1f:3f:13:47:1d
> (capab=0x31 status=0 aid=2)
> [ 1509.866451] wlan0: associated
> 
> At 1506.5 the re-association happens.
> 
> I can't influence the interval of the wpa rekeying of my wlan-router, so I'm
> not sure if this is related.
> The amount of network traffic doesn't seem to influence this behavior, either.
> 
> I tried to bisect it but ended up at:
> 
> commit 3c607d27c818cf4a5d28f2c73b18a88f8fbdfa33
> Author: Don Fry <donald.h.fry@intel.com>
> Date:   Fri Sep 30 11:40:20 2011 -0700
> 
>     iwlagn: rename iwlagn module iwlwifi and alias to iwlagn.
> 
>     Rename the iwlagn module as iwlwifi in preparation for future
>     changes.  Add an alias to iwlagn for backward compatibility.
> 
>     Signed-off-by: Don Fry <donald.h.fry@intel.com>
>     Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
>     Signed-off-by: John W. Linville <linville@tuxdriver.com>
> 
> That doesn't make a lot of sense to me but I wanted to mention it
> 
> I'll gladly provide more info.

I agree that does not make much sense, but we will take a look into it.
btw, could you try the attach patch and see if it help?

thanks
Wey
> 
> _______________________________________________
> ilw mailing list
> ilw@linux.intel.com
> http://linux.intel.com/mailman/listinfo/ilw


[-- Attachment #2: tid.patch --]
[-- Type: text/x-patch, Size: 1998 bytes --]

diff --git a/drivers/net/wireless/iwlwifi/iwl-agn-tx.c b/drivers/net/wireless/iwlwifi/iwl-agn-tx.c
index d9d758e..1b70048 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-agn-tx.c
@@ -95,6 +95,7 @@ static void iwlagn_tx_cmd_build_basic(struct iwl_priv *priv,
 			tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
 		else
 			tx_flags &= ~TX_CMD_FLG_SEQ_CTL_MSK;
+		tx_cmd->tid_tspec = IWL_MAX_TID_COUNT;
 	}
 
 	iwlagn_tx_cmd_protection(priv, info, fc, &tx_flags);
@@ -808,6 +809,8 @@ static void iwl_rx_reply_tx_agg(struct iwl_priv *priv,
 	u32 status = le16_to_cpu(tx_resp->status.status);
 	int i;
 
+        WARN_ON(tid == IWL_MAX_TID_COUNT);
+
 	if (agg->wait_for_ba)
 		IWL_DEBUG_TX_REPLY(priv,
 			"got tx response w/o block-ack\n");
@@ -1035,10 +1038,12 @@ int iwlagn_rx_reply_tx(struct iwl_priv *priv, struct iwl_rx_mem_buffer *rxb,
 		}
 
 		__skb_queue_head_init(&skbs);
-		priv->tid_data[sta_id][tid].next_reclaimed = next_reclaimed;
 
-		IWL_DEBUG_TX_REPLY(priv, "Next reclaimed packet:%d",
-					  next_reclaimed);
+		if (tid != IWL_MAX_TID_COUNT) {
+			priv->tid_data[sta_id][tid].next_reclaimed = next_reclaimed;
+			IWL_DEBUG_TX_REPLY(priv, "Next reclaimed packet:%d",
+						  next_reclaimed);
+		}
 
 		/*we can free until ssn % q.n_bd not inclusive */
 		WARN_ON(iwl_trans_reclaim(trans(priv), sta_id, tid, txq_id,
diff --git a/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c b/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c
index aa87994..fa1b369 100644
--- a/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c
+++ b/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c
@@ -1551,7 +1551,7 @@ static int iwl_trans_pcie_reclaim(struct iwl_trans *trans, int sta_id, int tid,
 
 	txq->time_stamp = jiffies;
 
-	if (unlikely(txq_id >= IWLAGN_FIRST_AMPDU_QUEUE &&
+	if (unlikely(tid != IWL_MAX_TID_COUNT && txq_id >= IWLAGN_FIRST_AMPDU_QUEUE &&
 		     txq_id != trans_pcie->agg_txq[sta_id][tid])) {
 		/*
 		 * FIXME: this is a uCode bug which need to be addressed,