ath9k: add stability fixes for long standing hang issues (FS#13, #34, #…

…373, #383)

The radio would stop communicating completely. This issue was easiest to
trigger on AR913x devices, e.g. the TP-Link TL-WR1043ND, but other
hardware was occasionally affected as well.

The most critical issue was a race condition in disabling/enabling IRQs
between the IRQ handler and the IRQ processing tasklet

Signed-off-by: Felix Fietkau <nbd@nbd.name>

package/kernel/mac80211/patches/354-ath9k-rename-tx_complete_work-to-hw_check_work.patch

@@ -0,0 +1,175 @@
+From: Felix Fietkau <nbd@nbd.name>
+Date: Wed, 25 Jan 2017 12:57:05 +0100
+Subject: [PATCH] ath9k: rename tx_complete_work to hw_check_work
+
+Also include common MAC alive check. This should make the hang checks
+more reliable for modes where beacons are not sent and is used as a
+starting point for further hang check improvements
+
+Signed-off-by: Felix Fietkau <nbd@nbd.name>
+---
+
+--- a/drivers/net/wireless/ath/ath9k/ath9k.h
++++ b/drivers/net/wireless/ath/ath9k/ath9k.h
+@@ -108,7 +108,7 @@ int ath_descdma_setup(struct ath_softc *
+ #define ATH_AGGR_MIN_QDEPTH        2
+ /* minimum h/w qdepth for non-aggregated traffic */
+ #define ATH_NON_AGGR_MIN_QDEPTH    8
+-#define ATH_TX_COMPLETE_POLL_INT   1000
++#define ATH_HW_CHECK_POLL_INT      1000
+ #define ATH_TXFIFO_DEPTH           8
+ #define ATH_TX_ERROR               0x01
+
+@@ -745,7 +745,7 @@ void ath9k_csa_update(struct ath_softc *
+ #define ATH_PAPRD_TIMEOUT         100 /* msecs */
+ #define ATH_PLL_WORK_INTERVAL     100
+
+-void ath_tx_complete_poll_work(struct work_struct *work);
++void ath_hw_check_work(struct work_struct *work);
+ void ath_reset_work(struct work_struct *work);
+ bool ath_hw_check(struct ath_softc *sc);
+ void ath_hw_pll_work(struct work_struct *work);
+@@ -1053,7 +1053,7 @@ struct ath_softc {
+ #ifdef CPTCFG_ATH9K_DEBUGFS
+ 	struct ath9k_debug debug;
+ #endif
+-	struct delayed_work tx_complete_work;
++	struct delayed_work hw_check_work;
+ 	struct delayed_work hw_pll_work;
+ 	struct timer_list sleep_timer;
+
+--- a/drivers/net/wireless/ath/ath9k/init.c
++++ b/drivers/net/wireless/ath/ath9k/init.c
+@@ -681,6 +681,7 @@ static int ath9k_init_softc(u16 devid, s
+ 	INIT_WORK(&sc->hw_reset_work, ath_reset_work);
+ 	INIT_WORK(&sc->paprd_work, ath_paprd_calibrate);
+ 	INIT_DELAYED_WORK(&sc->hw_pll_work, ath_hw_pll_work);
++	INIT_DELAYED_WORK(&sc->hw_check_work, ath_hw_check_work);
+
+ 	ath9k_init_channel_context(sc);
+
+--- a/drivers/net/wireless/ath/ath9k/link.c
++++ b/drivers/net/wireless/ath/ath9k/link.c
+@@ -20,20 +20,13 @@
+  * TX polling - checks if the TX engine is stuck somewhere
+  * and issues a chip reset if so.
+  */
+-void ath_tx_complete_poll_work(struct work_struct *work)
++static bool ath_tx_complete_check(struct ath_softc *sc)
+ {
+-	struct ath_softc *sc = container_of(work, struct ath_softc,
+-					    tx_complete_work.work);
+ 	struct ath_txq *txq;
+ 	int i;
+-	bool needreset = false;
+-
+
+-	if (sc->tx99_state) {
+-		ath_dbg(ath9k_hw_common(sc->sc_ah), RESET,
+-			"skip tx hung detection on tx99\n");
+-		return;
+-	}
++	if (sc->tx99_state)
++		return true;
+
+ 	for (i = 0; i < IEEE80211_NUM_ACS; i++) {
+ 		txq = sc->tx.txq_map[i];
+@@ -41,25 +34,36 @@ void ath_tx_complete_poll_work(struct wo
+ 		ath_txq_lock(sc, txq);
+ 		if (txq->axq_depth) {
+ 			if (txq->axq_tx_inprogress) {
+-				needreset = true;
+ 				ath_txq_unlock(sc, txq);
+-				break;
+-			} else {
+-				txq->axq_tx_inprogress = true;
++				goto reset;
+ 			}
++
++			txq->axq_tx_inprogress = true;
+ 		}
+ 		ath_txq_unlock(sc, txq);
+ 	}
+
+-	if (needreset) {
+-		ath_dbg(ath9k_hw_common(sc->sc_ah), RESET,
+-			"tx hung, resetting the chip\n");
+-		ath9k_queue_reset(sc, RESET_TYPE_TX_HANG);
++	return true;
++
++reset:
++	ath_dbg(ath9k_hw_common(sc->sc_ah), RESET,
++		"tx hung, resetting the chip\n");
++	ath9k_queue_reset(sc, RESET_TYPE_TX_HANG);
++	return false;
++
++}
++
++void ath_hw_check_work(struct work_struct *work)
++{
++	struct ath_softc *sc = container_of(work, struct ath_softc,
++					    hw_check_work.work);
++
++	if (!ath_hw_check(sc) ||
++	    !ath_tx_complete_check(sc))
+ 		return;
+-	}
+
+-	ieee80211_queue_delayed_work(sc->hw, &sc->tx_complete_work,
+-				     msecs_to_jiffies(ATH_TX_COMPLETE_POLL_INT));
++	ieee80211_queue_delayed_work(sc->hw, &sc->hw_check_work,
++				     msecs_to_jiffies(ATH_HW_CHECK_POLL_INT));
+ }
+
+ /*
+--- a/drivers/net/wireless/ath/ath9k/main.c
++++ b/drivers/net/wireless/ath/ath9k/main.c
+@@ -181,7 +181,7 @@ void ath9k_ps_restore(struct ath_softc *
+ static void __ath_cancel_work(struct ath_softc *sc)
+ {
+ 	cancel_work_sync(&sc->paprd_work);
+-	cancel_delayed_work_sync(&sc->tx_complete_work);
++	cancel_delayed_work_sync(&sc->hw_check_work);
+ 	cancel_delayed_work_sync(&sc->hw_pll_work);
+
+ #ifdef CPTCFG_ATH9K_BTCOEX_SUPPORT
+@@ -198,7 +198,8 @@ void ath_cancel_work(struct ath_softc *s
+
+ void ath_restart_work(struct ath_softc *sc)
+ {
+-	ieee80211_queue_delayed_work(sc->hw, &sc->tx_complete_work, 0);
++	ieee80211_queue_delayed_work(sc->hw, &sc->hw_check_work,
++				     ATH_HW_CHECK_POLL_INT);
+
+ 	if (AR_SREV_9340(sc->sc_ah) || AR_SREV_9330(sc->sc_ah))
+ 		ieee80211_queue_delayed_work(sc->hw, &sc->hw_pll_work,
+@@ -2091,7 +2092,7 @@ void __ath9k_flush(struct ieee80211_hw *
+ 	int timeout;
+ 	bool drain_txq;
+
+-	cancel_delayed_work_sync(&sc->tx_complete_work);
++	cancel_delayed_work_sync(&sc->hw_check_work);
+
+ 	if (ah->ah_flags & AH_UNPLUGGED) {
+ 		ath_dbg(common, ANY, "Device has been unplugged!\n");
+@@ -2129,7 +2130,8 @@ void __ath9k_flush(struct ieee80211_hw *
+ 		ath9k_ps_restore(sc);
+ 	}
+
+-	ieee80211_queue_delayed_work(hw, &sc->tx_complete_work, 0);
++	ieee80211_queue_delayed_work(hw, &sc->hw_check_work,
++				     ATH_HW_CHECK_POLL_INT);
+ }
+
+ static bool ath9k_tx_frames_pending(struct ieee80211_hw *hw)
+--- a/drivers/net/wireless/ath/ath9k/xmit.c
++++ b/drivers/net/wireless/ath/ath9k/xmit.c
+@@ -2915,8 +2915,6 @@ int ath_tx_init(struct ath_softc *sc, in
+ 		return error;
+ 	}
+
+-	INIT_DELAYED_WORK(&sc->tx_complete_work, ath_tx_complete_poll_work);
+-
+ 	if (sc->sc_ah->caps.hw_caps & ATH9K_HW_CAP_EDMA)
+ 		error = ath_tx_edma_init(sc);
+

package/kernel/mac80211/patches/355-ath9k_hw-check-if-the-chip-failed-to-wake-up.patch

@@ -0,0 +1,30 @@
+From: Felix Fietkau <nbd@nbd.name>
+Date: Wed, 25 Jan 2017 12:58:17 +0100
+Subject: [PATCH] ath9k_hw: check if the chip failed to wake up
+
+In an RFC patch, Sven Eckelmann and Simon Wunderlich reported:
+
+"QCA 802.11n chips (especially AR9330/AR9340) sometimes end up in a
+state in which a read of AR_CFG always returns 0xdeadbeef.
+This should not happen when when the power_mode of the device is
+ATH9K_PM_AWAKE."
+
+Include the check for the default register state in the existing MAC
+hang check.
+
+Signed-off-by: Felix Fietkau <nbd@nbd.name>
+---
+
+--- a/drivers/net/wireless/ath/ath9k/hw.c
++++ b/drivers/net/wireless/ath/ath9k/hw.c
+@@ -1624,6 +1624,10 @@ bool ath9k_hw_check_alive(struct ath_hw
+ 	int count = 50;
+ 	u32 reg, last_val;
+
++	/* Check if chip failed to wake up */
++	if (REG_READ(ah, AR_CFG) == 0xdeadbeef)
++		return false;
++
+ 	if (AR_SREV_9300(ah))
+ 		return !ath9k_hw_detect_mac_hang(ah);
+

package/kernel/mac80211/patches/356-ath9k-check-for-deaf-rx-path-state.patch

@@ -0,0 +1,100 @@
+From: Felix Fietkau <nbd@nbd.name>
+Date: Wed, 25 Jan 2017 13:00:58 +0100
+Subject: [PATCH] ath9k: check for deaf rx path state
+
+Various chips occasionally run into a state where the tx path still
+appears to be working normally, but the rx path is deaf.
+
+There is no known register signature to check for this state explicitly,
+so use the lack of rx interrupts as an indicator.
+
+This detection is prone to false positives, since a device could also
+simply be in an environment where there are no frames on the air.
+However, in this case doing a reset should be harmless since it's
+obviously not interrupting any real activity. To avoid confusion, call
+the reset counters in this case "Rx path inactive" instead of something
+like "Rx path deaf", since it may not be an indication of a real
+hardware failure.
+
+Signed-off-by: Felix Fietkau <nbd@nbd.name>
+---
+
+--- a/drivers/net/wireless/ath/ath9k/ath9k.h
++++ b/drivers/net/wireless/ath/ath9k/ath9k.h
+@@ -1027,6 +1027,7 @@ struct ath_softc {
+
+ 	u8 gtt_cnt;
+ 	u32 intrstatus;
++	u32 rx_active;
+ 	u16 ps_flags; /* PS_* */
+ 	bool ps_enabled;
+ 	bool ps_idle;
+--- a/drivers/net/wireless/ath/ath9k/debug.c
++++ b/drivers/net/wireless/ath/ath9k/debug.c
+@@ -763,6 +763,7 @@ static int read_file_reset(struct seq_fi
+ 		[RESET_TYPE_BEACON_STUCK] = "Stuck Beacon",
+ 		[RESET_TYPE_MCI] = "MCI Reset",
+ 		[RESET_TYPE_CALIBRATION] = "Calibration error",
++		[RESET_TYPE_RX_INACTIVE] = "Rx path inactive",
+ 		[RESET_TX_DMA_ERROR] = "Tx DMA stop error",
+ 		[RESET_RX_DMA_ERROR] = "Rx DMA stop error",
+ 	};
+--- a/drivers/net/wireless/ath/ath9k/debug.h
++++ b/drivers/net/wireless/ath/ath9k/debug.h
+@@ -50,6 +50,7 @@ enum ath_reset_type {
+ 	RESET_TYPE_BEACON_STUCK,
+ 	RESET_TYPE_MCI,
+ 	RESET_TYPE_CALIBRATION,
++	RESET_TYPE_RX_INACTIVE,
+ 	RESET_TX_DMA_ERROR,
+ 	RESET_RX_DMA_ERROR,
+ 	__RESET_TYPE_MAX
+--- a/drivers/net/wireless/ath/ath9k/link.c
++++ b/drivers/net/wireless/ath/ath9k/link.c
+@@ -53,13 +53,27 @@ reset:
+
+ }
+
++static bool ath_rx_active_check(struct ath_softc *sc)
++{
++	if (sc->rx_active) {
++		sc->rx_active = 0;
++		return true;
++	}
++
++	ath_dbg(ath9k_hw_common(sc->sc_ah), RESET,
++		"rx path inactive, resetting the chip\n");
++	ath9k_queue_reset(sc, RESET_TYPE_RX_INACTIVE);
++	return false;
++}
++
+ void ath_hw_check_work(struct work_struct *work)
+ {
+ 	struct ath_softc *sc = container_of(work, struct ath_softc,
+ 					    hw_check_work.work);
+
+ 	if (!ath_hw_check(sc) ||
+-	    !ath_tx_complete_check(sc))
++	    !ath_tx_complete_check(sc) ||
++	    !ath_rx_active_check(sc))
+ 		return;
+
+ 	ieee80211_queue_delayed_work(sc->hw, &sc->hw_check_work,
+--- a/drivers/net/wireless/ath/ath9k/main.c
++++ b/drivers/net/wireless/ath/ath9k/main.c
+@@ -269,6 +269,7 @@ static bool ath_complete_reset(struct at
+ 	}
+
+ 	sc->gtt_cnt = 0;
++	sc->rx_active = 1;
+
+ 	ath9k_hw_set_interrupts(ah);
+ 	ath9k_hw_enable_interrupts(ah);
+@@ -452,6 +453,7 @@ void ath9k_tasklet(unsigned long data)
+ 			ath_rx_tasklet(sc, 0, true);
+
+ 		ath_rx_tasklet(sc, 0, false);
++		sc->rx_active = 1;
+ 	}
+
+ 	if (status & ATH9K_INT_TX) {