Commit 27edb1ac authored by Emmanuel Grumbach's avatar Emmanuel Grumbach Committed by Johannes Berg

iwlwifi: silently ignore fw flaws in Tx path

We know that we have issues with the fw in the reclaim path.
This is why iwl_reclaim doesn't complain too loud when it
happens since it is recoverable. Somehow, the caller of
iwl_reclaim however WARNed when it happens. This doesn't
make any sense.

When I digged into the history of that code, I discovered
that this bug occurs only when we receive a BA notification.
So move the W/A in the BA notification handling code where
it was before.

This patch addresses:
http://bugzilla.intellinuxwireless.org/show_bug.cgi?id=2387

Cc: stable@vger.kernel.org
Reported-by: default avatarFlorian Reitmeir <florian@reitmeir.org>
Signed-off-by: default avatarEmmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: default avatarJohannes Berg <johannes.berg@intel.com>
parent 8907a6fb
...@@ -1097,29 +1097,6 @@ static void iwl_check_abort_status(struct iwl_priv *priv, ...@@ -1097,29 +1097,6 @@ static void iwl_check_abort_status(struct iwl_priv *priv,
} }
} }
static int iwl_reclaim(struct iwl_priv *priv, int sta_id, int tid,
int txq_id, int ssn, struct sk_buff_head *skbs)
{
if (unlikely(txq_id >= IWLAGN_FIRST_AMPDU_QUEUE &&
tid != IWL_TID_NON_QOS &&
txq_id != priv->tid_data[sta_id][tid].agg.txq_id)) {
/*
* FIXME: this is a uCode bug which need to be addressed,
* log the information and return for now.
* Since it is can possibly happen very often and in order
* not to fill the syslog, don't use IWL_ERR or IWL_WARN
*/
IWL_DEBUG_TX_QUEUES(priv,
"Bad queue mapping txq_id=%d, agg_txq[sta:%d,tid:%d]=%d\n",
txq_id, sta_id, tid,
priv->tid_data[sta_id][tid].agg.txq_id);
return 1;
}
iwl_trans_reclaim(priv->trans, txq_id, ssn, skbs);
return 0;
}
int iwlagn_rx_reply_tx(struct iwl_priv *priv, struct iwl_rx_cmd_buffer *rxb, int iwlagn_rx_reply_tx(struct iwl_priv *priv, struct iwl_rx_cmd_buffer *rxb,
struct iwl_device_cmd *cmd) struct iwl_device_cmd *cmd)
{ {
...@@ -1181,9 +1158,8 @@ int iwlagn_rx_reply_tx(struct iwl_priv *priv, struct iwl_rx_cmd_buffer *rxb, ...@@ -1181,9 +1158,8 @@ int iwlagn_rx_reply_tx(struct iwl_priv *priv, struct iwl_rx_cmd_buffer *rxb,
next_reclaimed); next_reclaimed);
} }
/*we can free until ssn % q.n_bd not inclusive */ iwl_trans_reclaim(priv->trans, txq_id, ssn, &skbs);
WARN_ON_ONCE(iwl_reclaim(priv, sta_id, tid,
txq_id, ssn, &skbs));
iwlagn_check_ratid_empty(priv, sta_id, tid); iwlagn_check_ratid_empty(priv, sta_id, tid);
freed = 0; freed = 0;
...@@ -1308,16 +1284,27 @@ int iwlagn_rx_reply_compressed_ba(struct iwl_priv *priv, ...@@ -1308,16 +1284,27 @@ int iwlagn_rx_reply_compressed_ba(struct iwl_priv *priv,
return 0; return 0;
} }
if (unlikely(scd_flow != agg->txq_id)) {
/*
* FIXME: this is a uCode bug which need to be addressed,
* log the information and return for now.
* Since it is can possibly happen very often and in order
* not to fill the syslog, don't use IWL_ERR or IWL_WARN
*/
IWL_DEBUG_TX_QUEUES(priv,
"Bad queue mapping txq_id=%d, agg_txq[sta:%d,tid:%d]=%d\n",
scd_flow, sta_id, tid, agg->txq_id);
spin_unlock(&priv->sta_lock);
return 0;
}
__skb_queue_head_init(&reclaimed_skbs); __skb_queue_head_init(&reclaimed_skbs);
/* Release all TFDs before the SSN, i.e. all TFDs in front of /* Release all TFDs before the SSN, i.e. all TFDs in front of
* block-ack window (we assume that they've been successfully * block-ack window (we assume that they've been successfully
* transmitted ... if not, it's too late anyway). */ * transmitted ... if not, it's too late anyway). */
if (iwl_reclaim(priv, sta_id, tid, scd_flow, iwl_trans_reclaim(priv->trans, scd_flow, ba_resp_scd_ssn,
ba_resp_scd_ssn, &reclaimed_skbs)) { &reclaimed_skbs);
spin_unlock(&priv->sta_lock);
return 0;
}
IWL_DEBUG_TX_REPLY(priv, "REPLY_COMPRESSED_BA [%d] Received from %pM, " IWL_DEBUG_TX_REPLY(priv, "REPLY_COMPRESSED_BA [%d] Received from %pM, "
"sta_id = %d\n", "sta_id = %d\n",
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment