Skip to contentSkip to conversation

mac80211/ath9k: Support AMPDU with multiple VIFs.

PATCH In review [email protected] 3 participants 3 messages Sep 23, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 mac80211/ath9k: Support AMPDU with multiple VIFs.

From: Ben Greear <[email protected]>

The old ieee80211_find_sta_by_hw method didn't properly find VIFS when there was more than one per AP. This caused AMPDU logic in ath9k to get the wrong VIF when trying to account for transmitted SKBs.

This patch changes ieee80211_find_sta_by_hw to take a localaddr argument to distinguish between VIFs with the same AP but different local addresses.

---
:100644 100644 c5e7af4... 9165ac8... M drivers/net/wireless/ath/ath9k/recv.c
:100644 100644 85a7323... 6543828... M drivers/net/wireless/ath/ath9k/xmit.c
:100644 100644 12a49f0... 9b25334... M include/net/mac80211.h
:100644 100644 44e10a9... f543331... M net/mac80211/sta_info.c

 drivers/net/wireless/ath/ath9k/recv.c |    6 +++++-
 drivers/net/wireless/ath/ath9k/xmit.c |    2 +-
 include/net/mac80211.h                |   14 ++++----------
 net/mac80211/sta_info.c               |   11 +++++++++--
 4 files changed, 19 insertions(+), 14 deletions(-)
drivers/net/wireless/ath/ath9k/recv.c +5 -1
index c5e7af4..9165ac8 100644
@@ -977,7 +977,11 @@ static void ath9k_process_rssi(struct ath_common *common,
 	 * at least one sdata of a wiphy on mac80211 but with ath9k virtual
 	 * wiphy you'd have to iterate over every wiphy and each sdata.
 	 */
-	sta = ieee80211_find_sta_by_hw(hw, hdr->addr2);
+	if (is_multicast_ether_addr(hdr->addr1))
+		sta = ieee80211_find_sta_by_hw(hw, hdr->addr2, NULL);
+	else
+		sta = ieee80211_find_sta_by_hw(hw, hdr->addr2, hdr->addr1);
+		
 	if (sta) {
 		an = (struct ath_node *) sta->drv_priv;
 		if (rx_stats->rs_rssi != ATH9K_RSSI_BAD &&
drivers/net/wireless/ath/ath9k/xmit.c +1 -1
index 85a7323..6543828 100644
@@ -329,7 +329,7 @@ static void ath_tx_complete_aggr(struct ath_softc *sc, struct ath_txq *txq,
 	rcu_read_lock();
 
 	/* XXX: use ieee80211_find_sta! */
-	sta = ieee80211_find_sta_by_hw(hw, hdr->addr1);
+	sta = ieee80211_find_sta_by_hw(hw, hdr->addr1, hdr->addr2);
 	if (!sta) {
 		rcu_read_unlock();
 
include/net/mac80211.h +4 -10
index 12a49f0..9b25334 100644
@@ -2419,22 +2419,16 @@ struct ieee80211_sta *ieee80211_find_sta(struct ieee80211_vif *vif,
  * ieee80211_find_sta_by_hw - find a station on hardware
  *
  * @hw: pointer as obtained from ieee80211_alloc_hw()
- * @addr: station's address
+ * @addr: remote station's address
+ * @localaddr: local address (vif->sdata->vif.addr). Use NULL for 'any'.
  *
  * This function must be called under RCU lock and the
  * resulting pointer is only valid under RCU lock as well.
  *
- * NOTE: This function should not be used! When mac80211 is converted
- *	 internally to properly keep track of stations on multiple
- *	 virtual interfaces, it will not always know which station to
- *	 return here since a single address might be used by multiple
- *	 logical stations (e.g. consider a station connecting to another
- *	 BSSID on the same AP hardware without disconnecting first).
- *
- * DO NOT USE THIS FUNCTION.
  */
 struct ieee80211_sta *ieee80211_find_sta_by_hw(struct ieee80211_hw *hw,
-					       const u8 *addr);
+					       const u8 *addr,
+					       const u8 *localaddr);
 
 /**
  * ieee80211_sta_block_awake - block station from waking up
net/mac80211/sta_info.c +9 -2
index 44e10a9..f543331 100644
@@ -839,12 +839,19 @@ void ieee80211_sta_expire(struct ieee80211_sub_if_data *sdata,
 }
 
 struct ieee80211_sta *ieee80211_find_sta_by_hw(struct ieee80211_hw *hw,
-					       const u8 *addr)
+					       const u8 *addr,
+					       const u8 *localaddr)
 {
 	struct sta_info *sta, *nxt;
 
-	/* Just return a random station ... first in list ... */
+	/*
+	 * Just return a random station if localaddr is NULL
+	 * ... first in list.
+	 */
 	for_each_sta_info(hw_to_local(hw), addr, sta, nxt) {
+		if (localaddr &&
+		    compare_ether_addr(sta->sdata->vif.addr, localaddr) != 0)
+			continue;
 		if (!sta->uploaded)
 			return NULL;
 		return &sta->sta;
Signature · 1 line
1.7.2.3
maintainer
#1 lore raw
Quoted 9 lines · On Thu, 2010-09-23 at 06:54 -0700, [email protected] wrote:

*
- * NOTE: This function should not be used! When mac80211 is converted
- * internally to properly keep track of stations on multiple
- * virtual interfaces, it will not always know which station to
- * return here since a single address might be used by multiple
- * logical stations (e.g. consider a station connecting to another
- * BSSID on the same AP hardware without disconnecting first).
- *
- * DO NOT USE THIS FUNCTION.

Also, I'm not sure you really should remove this comment, since it still applies when you pass NULL, and I'm still not convinced that ath9k couldn't keep track of the virtual interface accurately?

johannes

#2 lore raw
Quoted 12 lines · Johannes Berg
Quoted 9 lines · On Thu, 2010-09-23 at 06:54 -0700, [email protected] wrote:
   *

- * NOTE: This function should not be used! When mac80211 is converted
- * internally to properly keep track of stations on multiple
- * virtual interfaces, it will not always know which station to
- * return here since a single address might be used by multiple
- * logical stations (e.g. consider a station connecting to another
- * BSSID on the same AP hardware without disconnecting first).
- *
- * DO NOT USE THIS FUNCTION.

Quoted 3 lines

Also, I'm not sure you really should remove this comment, since it still applies when you pass NULL, and I'm still not convinced that ath9k couldn't keep track of the virtual interface accurately?

It seems ath9k would have to selectively remove buffers from it's tx queues with vifs dissappeared. I had a previous patch that flushed all when interfaces came and went, but it seems I would also have to do something similar for AP mode when remote stations came and went. And for remote stations, Felix didn't like flushing the queues entirely, so it would have to be for just pkts belonging to that VIF.

That would mean quite a bit of re-write of the buffer flush logic as far as I can tell.

As for the by_hw() method, it's only used in ath9k, and the only time it gets passed a NULL is in the ath9k_process_rssi code when the destination address is multicast (so we cannot reliably match on a destination interface).

I'll update the warning comment to mention about passing in NULL as localaddr.

Thanks,
Ben

Signature · 2 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com