[PATCH AUTOSEL 6.1 13/17] wifi: mt76: mt7921s: fix potential hung tasks during chip recovery

From: Sasha Levin
Date: Mon May 27 2024 - 10:58:28 EST


From: Leon Yen <leon.yen@xxxxxxxxxxxx>

[ Upstream commit ecf0b2b8a37c8464186620bef37812a117ff6366 ]

During chip recovery (e.g. chip reset), there is a possible situation that
kernel worker reset_work is holding the lock and waiting for kernel thread
stat_worker to be parked, while stat_worker is waiting for the release of
the same lock.
It causes a deadlock resulting in the dumping of hung tasks messages and
possible rebooting of the device.

This patch prevents the execution of stat_worker during the chip recovery.

Signed-off-by: Leon Yen <leon.yen@xxxxxxxxxxxx>
Signed-off-by: Ming Yen Hsieh <MingYen.Hsieh@xxxxxxxxxxxx>
Signed-off-by: Felix Fietkau <nbd@xxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/net/wireless/mediatek/mt76/mt7921/mac.c | 2 ++
drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c | 2 --
drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c | 2 --
drivers/net/wireless/mediatek/mt76/sdio.c | 3 ++-
4 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
index 49ddca84f7862..cae7c21ca1f8b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
@@ -883,6 +883,7 @@ void mt7921_mac_reset_work(struct work_struct *work)
int i, ret;

dev_dbg(dev->mt76.dev, "chip reset\n");
+ set_bit(MT76_RESET, &dev->mphy.state);
dev->hw_full_reset = true;
ieee80211_stop_queues(hw);

@@ -911,6 +912,7 @@ void mt7921_mac_reset_work(struct work_struct *work)
}

dev->hw_full_reset = false;
+ clear_bit(MT76_RESET, &dev->mphy.state);
pm->suspended = false;
ieee80211_wake_queues(hw);
ieee80211_iterate_active_interfaces(hw,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
index 8dd60408b117e..cb20ddcad137c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
@@ -78,7 +78,6 @@ int mt7921e_mac_reset(struct mt7921_dev *dev)
mt76_wr(dev, MT_WFDMA0_HOST_INT_ENA, 0);
mt76_wr(dev, MT_PCIE_MAC_INT_ENABLE, 0x0);

- set_bit(MT76_RESET, &dev->mphy.state);
set_bit(MT76_MCU_RESET, &dev->mphy.state);
wake_up(&dev->mt76.mcu.wait);
skb_queue_purge(&dev->mt76.mcu.res_q);
@@ -129,7 +128,6 @@ int mt7921e_mac_reset(struct mt7921_dev *dev)

err = __mt7921_start(&dev->phy);
out:
- clear_bit(MT76_RESET, &dev->mphy.state);

local_bh_disable();
napi_enable(&dev->mt76.tx_napi);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c
index fd07b66233920..46af03803de73 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/sdio_mac.c
@@ -98,7 +98,6 @@ int mt7921s_mac_reset(struct mt7921_dev *dev)
mt76_connac_free_pending_tx_skbs(&dev->pm, NULL);
mt76_txq_schedule_all(&dev->mphy);
mt76_worker_disable(&dev->mt76.tx_worker);
- set_bit(MT76_RESET, &dev->mphy.state);
set_bit(MT76_MCU_RESET, &dev->mphy.state);
wake_up(&dev->mt76.mcu.wait);
skb_queue_purge(&dev->mt76.mcu.res_q);
@@ -135,7 +134,6 @@ int mt7921s_mac_reset(struct mt7921_dev *dev)

err = __mt7921_start(&dev->phy);
out:
- clear_bit(MT76_RESET, &dev->mphy.state);

mt76_worker_enable(&dev->mt76.tx_worker);

diff --git a/drivers/net/wireless/mediatek/mt76/sdio.c b/drivers/net/wireless/mediatek/mt76/sdio.c
index fc4fb94635645..2a81c2f663449 100644
--- a/drivers/net/wireless/mediatek/mt76/sdio.c
+++ b/drivers/net/wireless/mediatek/mt76/sdio.c
@@ -499,7 +499,8 @@ static void mt76s_tx_status_data(struct mt76_worker *worker)
dev = container_of(sdio, struct mt76_dev, sdio);

while (true) {
- if (test_bit(MT76_REMOVED, &dev->phy.state))
+ if (test_bit(MT76_RESET, &dev->phy.state) ||
+ test_bit(MT76_REMOVED, &dev->phy.state))
break;

if (!dev->drv->tx_status_data(dev, &update))
--
2.43.0