svn commit: r225106 - user/adrian/if_ath_tx/sys/dev/ath
Adrian Chadd
adrian at FreeBSD.org
Tue Aug 23 11:53:09 UTC 2011
Author: adrian
Date: Tue Aug 23 11:53:08 2011
New Revision: 225106
URL: http://svn.freebsd.org/changeset/base/225106
Log:
Undo some of the stuff I've been tinkering with.
TX aggregation seems to be working on faster hardware; it's just
running iperf locally on these MIPS boards is causing headaches.
* Always schedule the TID in ath_tx_swq()
* Only schedule the TID when packets are thrown on the end of that
TID software queue.
This matches what the linux and reference code driver does.
Modified:
user/adrian/if_ath_tx/sys/dev/ath/if_ath_tx.c
Modified: user/adrian/if_ath_tx/sys/dev/ath/if_ath_tx.c
==============================================================================
--- user/adrian/if_ath_tx/sys/dev/ath/if_ath_tx.c Tue Aug 23 09:42:29 2011 (r225105)
+++ user/adrian/if_ath_tx/sys/dev/ath/if_ath_tx.c Tue Aug 23 11:53:08 2011 (r225106)
@@ -1948,18 +1948,11 @@ ath_tx_swq(struct ath_softc *sc, struct
/* Queue frame to the tail of the software queue */
ATH_TXQ_LOCK(atid);
ATH_TXQ_INSERT_TAIL(atid, bf, bf_list);
+ ATH_TXQ_UNLOCK(atid);
- /*
- * Don't queue frames if the TID has a handful
- * of hardware queued frames already.
- */
- if (atid->hwq_depth < sc->sc_tid_hwq_hi) {
- ATH_TXQ_UNLOCK(atid);
- ATH_TXQ_LOCK(txq);
- ath_tx_tid_sched(sc, an, tid);
- ATH_TXQ_UNLOCK(txq);
- } else
- ATH_TXQ_UNLOCK(atid);
+ ATH_TXQ_LOCK(txq);
+ ath_tx_tid_sched(sc, an, tid);
+ ATH_TXQ_UNLOCK(txq);
}
/*
@@ -2407,10 +2400,6 @@ ath_tx_aggr_retry_unaggr(struct ath_soft
}
#endif
- ATH_TXQ_LOCK(sc->sc_ac2q[atid->ac]);
- ath_tx_tid_sched(sc, an, atid->tid);
- ATH_TXQ_UNLOCK(sc->sc_ac2q[atid->ac]);
-
/* Free buffer, bf is free after this call */
ath_tx_default_comp(sc, bf, 0);
@@ -2836,10 +2825,6 @@ ath_tx_aggr_comp_unaggr(struct ath_softc
bf->bf_state.bfs_pktlen,
1, (ts->ts_status == 0) ? 0 : 1);
- ATH_TXQ_LOCK(sc->sc_ac2q[atid->ac]);
- ath_tx_tid_sched(sc, an, atid->tid);
- ATH_TXQ_UNLOCK(sc->sc_ac2q[atid->ac]);
-
/*
* If a cleanup is in progress, punt to comp_cleanup;
* rather than handling it here. It's thus their
More information about the svn-src-user
mailing list