ath9k: recover ar9380 chips from rare stuck state
In the experiment with Azimuth ADEPT-n testbed where the APs transmit
power was reduced to 25% and the signal strength was futher attenuated
by 20dB and induced a path loss of ~7dB, the station was reporting
beacon losses and the following issue were observed.
* rx clear is stuck at low for more than 300ms
* dcu chain and complete state is stuck at one of the hang signature
This patch triggers the hang detection logic that recovers the chip
from any of the above conditions. As the issue was originally reported
in ChromeOs with AR9382 chips, this detection logic is enabled only for
AR9380/2 chips.
Cc: Paul Stewart <pstew@google.com>
Reported-by: Gary Morain <gmorain@google.com>
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
diff --git a/drivers/net/wireless/ath/ath9k/ath9k.h b/drivers/net/wireless/ath/ath9k/ath9k.h
index 8c84049..0792d87 100644
--- a/drivers/net/wireless/ath/ath9k/ath9k.h
+++ b/drivers/net/wireless/ath/ath9k/ath9k.h
@@ -430,6 +430,8 @@
void ath_reset_work(struct work_struct *work);
void ath_hw_check(struct work_struct *work);
void ath_hw_pll_work(struct work_struct *work);
+void ath_rx_poll(unsigned long data);
+void ath_start_rx_poll(struct ath_softc *sc, u8 nbeacon);
void ath_paprd_calibrate(struct work_struct *work);
void ath_ani_calibrate(unsigned long data);
void ath_start_ani(struct ath_common *common);
@@ -670,6 +672,7 @@
struct ath_beacon_config cur_beacon_conf;
struct delayed_work tx_complete_work;
struct delayed_work hw_pll_work;
+ struct timer_list rx_poll_timer;
#ifdef CONFIG_ATH9K_BTCOEX_SUPPORT
struct ath_btcoex btcoex;