The ADM8211B seems to stall after some time of constant

activity at DS11 speeds.  Check for this in the watchdog
using seemingly undocumented bits of the TEST1 register
used in Linux driver and restart if a stall is detected.

1 files changed, 19 insertions, 1 deletions

diff --git a/sys/dev/ic/atw.c b/sys/dev/ic/atw.c
index efae7ce4e96..aefcb7226a3 100644
--- a/sys/dev/ic/atw.c
+++ b/sys/dev/ic/atw.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: atw.c,v 1.39 2005/09/24 09:22:26 jsg Exp $	*/
+/*	$OpenBSD: atw.c,v 1.40 2005/09/24 11:19:14 jsg Exp $	*/
 /*	$NetBSD: atw.c,v 1.69 2004/07/23 07:07:55 dyoung Exp $	*/
 
 /*-
@@ -3365,6 +3365,7 @@ atw_watchdog(struct ifnet *ifp)
 {
 	struct atw_softc *sc = ifp->if_softc;
 	struct ieee80211com *ic = &sc->sc_ic;
+	uint32_t test1, rra, rwa;
 
 	ifp->if_timer = 0;
 	if (ATW_IS_ENABLED(sc) == 0)
@@ -3385,6 +3386,23 @@ atw_watchdog(struct ifnet *ifp)
 	}
 	if (sc->sc_tx_timer != 0 || sc->sc_rescan_timer != 0)
 		ifp->if_timer = 1;
+
+	/*
+	 * ADM8211B seems to stall every so often, check for this.
+	 * These bits are what the Linux driver checks, they don't
+	 * seem to be documented by ADMTek/Infineon?
+	 */
+	if (sc->sc_rev == ATW_REVISION_BA) {
+		test1 = ATW_READ(sc, ATW_TEST1);
+		rra = (test1 >> 12) & 0x1ff;
+		rwa = (test1 >> 2) & 0x1ff;
+
+		if ((rra != rwa) && !(test1 & 0x2)) {
+			atw_init(ifp);
+			atw_start(ifp);
+		}
+	}
+
 	ieee80211_watchdog(ifp);
 }