Commit 5dadbe4e authored by Wen Gong's avatar Wen Gong Committed by Kalle Valo

ath10k: add atomic protection for device recovery

When it has more than one restart_work queued meanwhile, the 2nd
restart_work is very easy to break the 1st restart work and lead
recovery fail.

Add a flag to allow only one restart work running untill
device successfully recovered.

It already has flag ATH10K_FLAG_CRASH_FLUSH, but it can not use this
flag again, because it is clear in ath10k_core_start. The function
ieee80211_reconfig(called by ieee80211_restart_work) of mac80211 do
many things and drv_start(call to ath10k_core_start) is 1st thing,
when drv_start complete, it does not mean restart complete. So it
add new flag and clear it in ath10k_reconfig_complete, because it
is the last thing called from drv_reconfig_complete of function
ieee80211_reconfig, after it, the restart process finished.

Tested-on: QCA6174 hw3.2 SDIO WLAN.RMH.4.4.1-00049
Signed-off-by: default avatarWen Gong <wgong@codeaurora.org>
Signed-off-by: default avatarKalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/010101746bead6a0-d5e97c66-dedd-4b92-810e-c2e4840fafc9-000000@us-west-2.amazonses.com
parent 2bc2b87b
......@@ -2294,6 +2294,17 @@ static int ath10k_init_hw_params(struct ath10k *ar)
return 0;
}
void ath10k_core_start_recovery(struct ath10k *ar)
{
if (test_and_set_bit(ATH10K_FLAG_RESTARTING, &ar->dev_flags)) {
ath10k_warn(ar, "already restarting\n");
return;
}
queue_work(ar->workqueue, &ar->restart_work);
}
EXPORT_SYMBOL(ath10k_core_start_recovery);
static void ath10k_core_restart(struct work_struct *work)
{
struct ath10k *ar = container_of(work, struct ath10k, restart_work);
......
......@@ -865,6 +865,9 @@ enum ath10k_dev_flags {
/* Per Station statistics service */
ATH10K_FLAG_PEER_STATS,
/* Indicates that ath10k device is during recovery process and not complete */
ATH10K_FLAG_RESTARTING,
};
enum ath10k_cal_mode {
......@@ -1320,6 +1323,7 @@ int ath10k_core_start(struct ath10k *ar, enum ath10k_firmware_mode mode,
const struct ath10k_fw_components *fw_components);
int ath10k_wait_for_suspend(struct ath10k *ar, u32 suspend_opt);
void ath10k_core_stop(struct ath10k *ar);
void ath10k_core_start_recovery(struct ath10k *ar);
int ath10k_core_register(struct ath10k *ar,
const struct ath10k_bus_params *bus_params);
void ath10k_core_unregister(struct ath10k *ar);
......
......@@ -583,7 +583,7 @@ static ssize_t ath10k_write_simulate_fw_crash(struct file *file,
ret = ath10k_debug_fw_assert(ar);
} else if (!strcmp(buf, "hw-restart")) {
ath10k_info(ar, "user requested hw restart\n");
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
ret = 0;
} else {
ret = -EINVAL;
......@@ -2005,7 +2005,7 @@ static ssize_t ath10k_write_btcoex(struct file *file,
}
} else {
ath10k_info(ar, "restarting firmware due to btcoex change");
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
}
if (val)
......@@ -2136,7 +2136,7 @@ static ssize_t ath10k_write_peer_stats(struct file *file,
ath10k_info(ar, "restarting firmware due to Peer stats change");
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
ret = count;
exit:
......
......@@ -7932,6 +7932,7 @@ static void ath10k_reconfig_complete(struct ieee80211_hw *hw,
ath10k_info(ar, "device successfully recovered\n");
ar->state = ATH10K_STATE_ON;
ieee80211_wake_queues(ar->hw);
clear_bit(ATH10K_FLAG_RESTARTING, &ar->dev_flags);
}
mutex_unlock(&ar->conf_mutex);
......
......@@ -1774,7 +1774,7 @@ static void ath10k_pci_fw_dump_work(struct work_struct *work)
mutex_unlock(&ar->dump_mutex);
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
}
static void ath10k_pci_fw_crashed_dump(struct ath10k *ar)
......
......@@ -561,7 +561,7 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,
ATH10K_HTC_MBOX_MAX_PAYLOAD_LENGTH);
ret = -ENOMEM;
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
ath10k_warn(ar, "exceeds length, start recovery\n");
goto err;
......@@ -960,7 +960,7 @@ static int ath10k_sdio_mbox_read_int_status(struct ath10k *ar,
ret = ath10k_sdio_read(ar, MBOX_HOST_INT_STATUS_ADDRESS,
irq_proc_reg, sizeof(*irq_proc_reg));
if (ret) {
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
ath10k_warn(ar, "read int status fail, start recovery\n");
goto out;
}
......@@ -2501,7 +2501,7 @@ void ath10k_sdio_fw_crashed_dump(struct ath10k *ar)
ath10k_sdio_enable_intrs(ar);
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
}
static int ath10k_sdio_probe(struct sdio_func *func,
......
......@@ -1305,7 +1305,7 @@ int ath10k_snoc_fw_indication(struct ath10k *ar, u64 type)
switch (type) {
case ATH10K_QMI_EVENT_FW_READY_IND:
if (test_bit(ATH10K_SNOC_FLAG_REGISTERED, &ar_snoc->flags)) {
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
break;
}
......
......@@ -1937,7 +1937,7 @@ int ath10k_wmi_cmd_send(struct ath10k *ar, struct sk_buff *skb, u32 cmd_id)
if (ret == -EAGAIN) {
ath10k_warn(ar, "wmi command %d timeout, restarting hardware\n",
cmd_id);
queue_work(ar->workqueue, &ar->restart_work);
ath10k_core_start_recovery(ar);
}
return ret;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment