[PATCH 0/3] wifi: rtw89: correct ieee80211_restart_hw() flow caused by SER L2 event

[PATCH 0/3] wifi: rtw89: correct ieee80211_restart_hw() flow caused by SER L2 event

[Ping-Ke Shih]

When SER L2 event is triggered by firmware, driver will call
ieee80211_restart_hw() to restart whole hardware and firmware. Reset some
software counters properly by patch 1/3 during this situation. Also,
resolve H2C waiting problem by patches 2/3 and 3/3, or it causes
disconnection from AP.

Zong-Zhe Yang (3):
  wifi: rtw89: ser: reset total_sta_assoc and tdls_peer when L2
  wifi: rtw89: tweak H2C TX waiting function for SER
  wifi: rtw89: refine packet offload handling under SER

 drivers/net/wireless/realtek/rtw89/core.h |  1 +
 drivers/net/wireless/realtek/rtw89/fw.c   | 12 ++++++++++--
 drivers/net/wireless/realtek/rtw89/mac.c  |  3 +++
 drivers/net/wireless/realtek/rtw89/ser.c  |  5 +++++
 4 files changed, 19 insertions(+), 2 deletions(-)

-- 
2.25.1

[PATCH 1/3] wifi: rtw89: ser: reset total_sta_assoc and tdls_peer when L2

[Ping-Ke Shih]

From: Zong-Zhe Yang <kevin_yang@realtek.com>

The total_sta_assoc and the tdls_peer are used for statistics accodring
to stations' information. L2 (Level 2) SER (system error recovery) will
call ieee80211_restart_hw() which re-invokes sta_state ops. And then,
the total_sta_assoc and tdls_peer will be re-increased. In case wrong
statistics results, we reset them in SER L2 handling.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
---
 drivers/net/wireless/realtek/rtw89/ser.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/ser.c b/drivers/net/wireless/realtek/rtw89/ser.c
index 9ba99f3764e7e..54b314b8b3292 100644
--- a/drivers/net/wireless/realtek/rtw89/ser.c
+++ b/drivers/net/wireless/realtek/rtw89/ser.c
@@ -303,6 +303,7 @@ static void ser_reset_vif(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 	rtw89_core_release_bit_map(rtwdev->hw_port, rtwvif->port);
 	rtwvif->net_type = RTW89_NET_TYPE_NO_LINK;
 	rtwvif->trigger = false;
+	rtwvif->tdls_peer = 0;
 }
 
 static void ser_sta_deinit_cam_iter(void *data, struct ieee80211_sta *sta)
@@ -341,6 +342,8 @@ static void ser_reset_mac_binding(struct rtw89_dev *rtwdev)
 	rtw89_core_release_all_bits_map(rtwdev->mac_id_map, RTW89_MAX_MAC_ID_NUM);
 	rtw89_for_each_rtwvif(rtwdev, rtwvif)
 		ser_reset_vif(rtwdev, rtwvif);
+
+	rtwdev->total_sta_assoc = 0;
 }
 
 /* hal function */
-- 
2.25.1

[PATCH 2/3] wifi: rtw89: tweak H2C TX waiting function for SER

[Ping-Ke Shih]

From: Zong-Zhe Yang <kevin_yang@realtek.com>

Some specific H2C (host to chip command) needs waiting until FW ACK by
C2H (chip to host event). However, during SER (system error recovery),
most interrupts are disabled, so we can't receive C2H immediately. It
causes this kind of H2C TX waits will always time out during SER.

To save time spent by SER, we don't do these redundant waits. And, to
make a difference from -ETIMEDOUT in other cases, we make the function
return 1 for SER case. When some H2C callers really catch `ret == 1` at
runtime, they can determine whether it's reasonable or not, and consider
how to resolve their flow if needed.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
---
 drivers/net/wireless/realtek/rtw89/core.h | 1 +
 drivers/net/wireless/realtek/rtw89/fw.c   | 8 ++++++++
 drivers/net/wireless/realtek/rtw89/ser.c  | 2 ++
 3 files changed, 11 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index b60cd9852259e..5893ac9eb3f9b 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3504,6 +3504,7 @@ enum rtw89_flags {
 	RTW89_FLAG_LOW_POWER_MODE,
 	RTW89_FLAG_INACTIVE_PS,
 	RTW89_FLAG_CRASH_SIMULATING,
+	RTW89_FLAG_SER_HANDLING,
 	RTW89_FLAG_WOWLAN,
 	RTW89_FLAG_FORBIDDEN_TRACK_WROK,
 	RTW89_FLAG_CHANGING_INTERFACE,
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index ad277f22b1973..b7befcfd4a913 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -3792,6 +3792,11 @@ int rtw89_fw_wow_cam_update(struct rtw89_dev *rtwdev,
 	return ret;
 }
 
+/* Return < 0, if failures happen during waiting for the condition.
+ * Return 0, when waiting for the condition succeeds.
+ * Return > 0, if the wait is considered unreachable due to driver/FW design,
+ * where 1 means during SER.
+ */
 static int rtw89_h2c_tx_and_wait(struct rtw89_dev *rtwdev, struct sk_buff *skb,
 				 struct rtw89_wait_info *wait, unsigned int cond)
 {
@@ -3804,6 +3809,9 @@ static int rtw89_h2c_tx_and_wait(struct rtw89_dev *rtwdev, struct sk_buff *skb,
 		return -EBUSY;
 	}
 
+	if (test_bit(RTW89_FLAG_SER_HANDLING, rtwdev->flags))
+		return 1;
+
 	return rtw89_wait_for_cond(wait, cond);
 }
 
diff --git a/drivers/net/wireless/realtek/rtw89/ser.c b/drivers/net/wireless/realtek/rtw89/ser.c
index 54b314b8b3292..0462ba693f6fd 100644
--- a/drivers/net/wireless/realtek/rtw89/ser.c
+++ b/drivers/net/wireless/realtek/rtw89/ser.c
@@ -409,6 +409,7 @@ static void ser_idle_st_hdl(struct rtw89_ser *ser, u8 evt)
 	switch (evt) {
 	case SER_EV_STATE_IN:
 		rtw89_hci_recovery_complete(rtwdev);
+		clear_bit(RTW89_FLAG_SER_HANDLING, rtwdev->flags);
 		clear_bit(RTW89_FLAG_CRASH_SIMULATING, rtwdev->flags);
 		break;
 	case SER_EV_L1_RESET_PREPARE:
@@ -421,6 +422,7 @@ static void ser_idle_st_hdl(struct rtw89_ser *ser, u8 evt)
 		ser_state_goto(ser, SER_L2_RESET_ST);
 		break;
 	case SER_EV_STATE_OUT:
+		set_bit(RTW89_FLAG_SER_HANDLING, rtwdev->flags);
 		rtw89_hci_recovery_start(rtwdev);
 		break;
 	default:
-- 
2.25.1

[PATCH 3/3] wifi: rtw89: refine packet offload handling under SER

[Ping-Ke Shih]

From: Zong-Zhe Yang <kevin_yang@realtek.com>

H2C of packet offload needs to wait FW ACK by C2H. But, it's possible
that packet offload happens during SER (system error recovery), e.g.
SER L2 which restarts HW. More, packet offload flow isn't deferrable.
So, the H2C wait may get `ret == 1` (unreachable).

However, the logic FW deals with packet offload is simple enough, just
clone content. It means that as long as the H2C is issued successfully,
the thing will succeed sooner or later. Therefore, after we add a debug
log when receiving ACK to packet offload, it would be acceptable that
during SER, packet offload don't really wait for ACK. And, if debugging,
we can still check its debug logs. Besides, we can expect that if we see
SER before receiving ACK to packet offload, those debug logs of the ACK
have a time difference.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
---
 drivers/net/wireless/realtek/rtw89/fw.c  | 4 ++--
 drivers/net/wireless/realtek/rtw89/mac.c | 3 +++
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index b7befcfd4a913..f7d4f9d14736d 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2467,7 +2467,7 @@ int rtw89_fw_h2c_del_pkt_offload(struct rtw89_dev *rtwdev, u8 id)
 	cond = RTW89_FW_OFLD_WAIT_COND_PKT_OFLD(id, RTW89_PKT_OFLD_OP_DEL);
 
 	ret = rtw89_h2c_tx_and_wait(rtwdev, skb, wait, cond);
-	if (ret) {
+	if (ret < 0) {
 		rtw89_debug(rtwdev, RTW89_DBG_FW,
 			    "failed to del pkt ofld: id %d, ret %d\n",
 			    id, ret);
@@ -2517,7 +2517,7 @@ int rtw89_fw_h2c_add_pkt_offload(struct rtw89_dev *rtwdev, u8 *id,
 	cond = RTW89_FW_OFLD_WAIT_COND_PKT_OFLD(alloc_id, RTW89_PKT_OFLD_OP_ADD);
 
 	ret = rtw89_h2c_tx_and_wait(rtwdev, skb, wait, cond);
-	if (ret) {
+	if (ret < 0) {
 		rtw89_debug(rtwdev, RTW89_DBG_FW,
 			    "failed to add pkt ofld: id %d, ret %d\n",
 			    alloc_id, ret);
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index acba53cf7cc31..f637fdafc7f10 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -4456,6 +4456,9 @@ rtw89_mac_c2h_pkt_ofld_rsp(struct rtw89_dev *rtwdev, struct sk_buff *skb_c2h,
 	struct rtw89_completion_data data = {};
 	unsigned int cond;
 
+	rtw89_debug(rtwdev, RTW89_DBG_FW, "pkt ofld rsp: id %d op %d len %d\n",
+		    pkt_id, pkt_op, pkt_len);
+
 	data.err = !pkt_len;
 	cond = RTW89_FW_OFLD_WAIT_COND_PKT_OFLD(pkt_id, pkt_op);
 
-- 
2.25.1

Re: [PATCH 1/3] wifi: rtw89: ser: reset total_sta_assoc and tdls_peer when L2

[Kalle Valo]

Ping-Ke Shih <pkshih@realtek.com> wrote:

> From: Zong-Zhe Yang <kevin_yang@realtek.com>
> 
> The total_sta_assoc and the tdls_peer are used for statistics accodring
> to stations' information. L2 (Level 2) SER (system error recovery) will
> call ieee80211_restart_hw() which re-invokes sta_state ops. And then,
> the total_sta_assoc and tdls_peer will be re-increased. In case wrong
> statistics results, we reset them in SER L2 handling.
> 
> Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
> Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>

3 patches applied to wireless-next.git, thanks.

cda66049bab5 wifi: rtw89: ser: reset total_sta_assoc and tdls_peer when L2
b79a84fbbdb0 wifi: rtw89: tweak H2C TX waiting function for SER
8b21c08ef7df wifi: rtw89: refine packet offload handling under SER

-- 
https://patchwork.kernel.org/project/linux-wireless/patch/20230516082441.11154-2-pkshih@realtek.com/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches