svn commit: r240926 - head/sys/dev/ath

[Adrian Chadd]

Author: adrian
Date: Tue Sep 25 20:41:43 2012
New Revision: 240926
URL: http://svn.freebsd.org/changeset/base/240926

Log:
  Oops - don't do the clrdmask check in ath_tx_xmit_normal() - the wrong
  lock may be held.
  
  Kim reported that the TID lock wasn't held when ath_tx_update_clrdmask()
  was called. Well, the underlying hardware TXQ for that TID.
  
  I'm betting it's the cabq stuff. ath_tx_xmit_normal() can be called
  for both real and software cabq.  For software cabq, the real destination
  txq is different to the txq. So, the lock check will fail.
  
  Reported by:	Kim Culhan <w8hdkim at gmail.com>

Modified:
  head/sys/dev/ath/if_ath_tx.c

Modified: head/sys/dev/ath/if_ath_tx.c
==============================================================================
--- head/sys/dev/ath/if_ath_tx.c	Tue Sep 25 20:27:42 2012	(r240925)
+++ head/sys/dev/ath/if_ath_tx.c	Tue Sep 25 20:41:43 2012	(r240926)
@@ -1356,9 +1356,6 @@ ath_tx_xmit_normal(struct ath_softc *sc,
 	 */
 	bf->bf_state.bfs_txflags |= HAL_TXDESC_CLRDMASK;
 
-	/* See if clrdmask needs to be set */
-	ath_tx_update_clrdmask(sc, tid, bf);
-
 	/* Setup the descriptor before handoff */
 	ath_tx_do_ratelookup(sc, bf);
 	ath_tx_calc_duration(sc, bf);
@@ -2820,6 +2817,8 @@ ath_tx_swq(struct ath_softc *sc, struct 
 	} else if (txq->axq_depth < sc->sc_hwq_limit) {
 		/* AMPDU not running, attempt direct dispatch */
 		DPRINTF(sc, ATH_DEBUG_SW_TX, "%s: xmit_normal\n", __func__);
+		/* See if clrdmask needs to be set */
+		ath_tx_update_clrdmask(sc, atid, bf);
 		ath_tx_xmit_normal(sc, txq, bf);
 	} else {
 		/* Busy; queue */