Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Felix Fietkau <nbd@nbd.name>
To:     linux-wireless@vger.kernel.org
Subject: [PATCH 2/5] mt76: mt76x02: improve tx hang detection
Date:   Thu,  3 Feb 2022 14:35:57 +0100
Message-Id: <20220203133600.92211-2-nbd@nbd.name>
In-Reply-To: <20220203133600.92211-1-nbd@nbd.name>
References: <20220203133600.92211-1-nbd@nbd.name>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Instead of checking if any queue has not made progress since the last run,
only trigger hang detection if one of the queues has not made any progress
in 10 subsequent runs. This should reduce false positive firmware restarts

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76x02.h  |  2 +-
 .../net/wireless/mediatek/mt76/mt76x02_mmio.c | 30 +++++++------------
 2 files changed, 12 insertions(+), 20 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02.h b/drivers/net/wireless/mediatek/mt76/mt76x02.h
index 3232083e9916..f76fd22ee035 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02.h
@@ -103,7 +103,7 @@ struct mt76x02_dev {
 	u8 tbtt_count;
 
 	u32 tx_hang_reset;
-	u8 tx_hang_check;
+	u8 tx_hang_check[4];
 	u8 beacon_hang_check;
 	u8 mcu_timeout;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
index ec0de691129a..8bcd8afa0d3a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
@@ -348,18 +348,20 @@ static bool mt76x02_tx_hang(struct mt76x02_dev *dev)
 	for (i = 0; i < 4; i++) {
 		q = dev->mphy.q_tx[i];
 
-		if (!q->queued)
-			continue;
-
 		prev_dma_idx = dev->mt76.tx_dma_idx[i];
 		dma_idx = readl(&q->regs->dma_idx);
 		dev->mt76.tx_dma_idx[i] = dma_idx;
 
-		if (prev_dma_idx == dma_idx)
-			break;
+		if (!q->queued || prev_dma_idx != dma_idx) {
+			dev->tx_hang_check[i] = 0;
+			continue;
+		}
+
+		if (++dev->tx_hang_check[i] >= MT_TX_HANG_TH)
+			return true;
 	}
 
-	return i < 4;
+	return false;
 }
 
 static void mt76x02_key_sync(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
@@ -530,23 +532,13 @@ static void mt76x02_check_tx_hang(struct mt76x02_dev *dev)
 	if (test_bit(MT76_RESTART, &dev->mphy.state))
 		return;
 
-	if (mt76x02_tx_hang(dev)) {
-		if (++dev->tx_hang_check >= MT_TX_HANG_TH)
-			goto restart;
-	} else {
-		dev->tx_hang_check = 0;
-	}
-
-	if (dev->mcu_timeout)
-		goto restart;
-
-	return;
+	if (!mt76x02_tx_hang(dev) && !dev->mcu_timeout)
+		return;
 
-restart:
 	mt76x02_watchdog_reset(dev);
 
 	dev->tx_hang_reset++;
-	dev->tx_hang_check = 0;
+	memset(dev->tx_hang_check, 0, sizeof(dev->tx_hang_check));
 	memset(dev->mt76.tx_dma_idx, 0xff,
 	       sizeof(dev->mt76.tx_dma_idx));
 }
-- 
2.32.0 (Apple Git-132)