When BT coexistence is enabled (eg oper mode 13, which is the default)
the initialisation on startup sometimes silently fails.
In a normal initialisation we see
usb 1-1.3: Product: Wireless USB Network Module
usb 1-1.3: Manufacturer: Redpine Signals, Inc.
usb 1-1.3: SerialNumber: 000000000001
rsi_91x: rsi_probe: Initialized os intf ops
rsi_91x: rsi_load_9116_firmware: Loading chunk 0
rsi_91x: rsi_load_9116_firmware: Loading chunk 1
rsi_91x: rsi_load_9116_firmware: Loading chunk 2
rsi_91x: Max Stations Allowed = 1
But sometimes the last log is missing and the wlan net device is
not created.
Running a userspace loop that resets the hardware via a GPIO shows the
problem occurring ~5/100 resets.
The problem does not occur in oper mode 1 (wifi only).
Adding logs shows that the initialisation state machine requests a MAC
reset via rsi_send_reset_mac() but the firmware does not reply, leading
to the initialisation sequence being incomplete.
Fix this by delaying attaching the BT adapter until the wifi
initialisation has completed.
With this applied I have done > 300 reset loops with no errors.
Fixes: 716b840c7641 ("rsi: handle BT traffic in driver")
Signed-off-by: Martin Fuzzey <[email protected]>
CC: [email protected]
---
drivers/net/wireless/rsi/rsi_91x_main.c | 16 +++++++++++++---
drivers/net/wireless/rsi/rsi_91x_mgmt.c | 3 +++
drivers/net/wireless/rsi/rsi_main.h | 3 ++-
3 files changed, 18 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/rsi/rsi_91x_main.c b/drivers/net/wireless/rsi/rsi_91x_main.c
index 29d8304..aece1d3 100644
--- a/drivers/net/wireless/rsi/rsi_91x_main.c
+++ b/drivers/net/wireless/rsi/rsi_91x_main.c
@@ -210,9 +210,10 @@ int rsi_read_pkt(struct rsi_common *common, u8 *rx_pkt, s32 rcv_pkt_len)
bt_pkt_type = frame_desc[offset + BT_RX_PKT_TYPE_OFST];
if (bt_pkt_type == BT_CARD_READY_IND) {
rsi_dbg(INFO_ZONE, "BT Card ready recvd\n");
- if (rsi_bt_ops.attach(common, &g_proto_ops))
- rsi_dbg(ERR_ZONE,
- "Failed to attach BT module\n");
+ if (common->fsm_state == FSM_MAC_INIT_DONE)
+ rsi_attach_bt(common);
+ else
+ common->bt_defer_attach = true;
} else {
if (common->bt_adapter)
rsi_bt_ops.recv_pkt(common->bt_adapter,
@@ -277,6 +278,15 @@ void rsi_set_bt_context(void *priv, void *bt_context)
}
#endif
+void rsi_attach_bt(struct rsi_common *common)
+{
+#ifdef CONFIG_RSI_COEX
+ if (rsi_bt_ops.attach(common, &g_proto_ops))
+ rsi_dbg(ERR_ZONE,
+ "Failed to attach BT module\n");
+#endif
+}
+
/**
* rsi_91x_init() - This function initializes os interface operations.
* @void: Void.
diff --git a/drivers/net/wireless/rsi/rsi_91x_mgmt.c b/drivers/net/wireless/rsi/rsi_91x_mgmt.c
index ed67f65..50c4b9b 100644
--- a/drivers/net/wireless/rsi/rsi_91x_mgmt.c
+++ b/drivers/net/wireless/rsi/rsi_91x_mgmt.c
@@ -2056,6 +2056,9 @@ static int rsi_handle_ta_confirm_type(struct rsi_common *common,
if (common->reinit_hw) {
complete(&common->wlan_init_completion);
} else {
+ if (common->bt_defer_attach)
+ rsi_attach_bt(common);
+
return rsi_mac80211_attach(common);
}
}
diff --git a/drivers/net/wireless/rsi/rsi_main.h b/drivers/net/wireless/rsi/rsi_main.h
index b3e25bc..b983cc5 100644
--- a/drivers/net/wireless/rsi/rsi_main.h
+++ b/drivers/net/wireless/rsi/rsi_main.h
@@ -320,6 +320,7 @@ struct rsi_common {
struct ieee80211_vif *roc_vif;
bool eapol4_confirm;
+ bool bt_defer_attach;
void *bt_adapter;
struct cfg80211_scan_request *hwscan;
@@ -401,5 +402,5 @@ struct rsi_host_intf_ops {
enum rsi_host_intf rsi_get_host_intf(void *priv);
void rsi_set_bt_context(void *priv, void *bt_context);
-
+void rsi_attach_bt(struct rsi_common *common);
#endif
--
1.9.1
Martin Fuzzey <[email protected]> wrote:
> When BT coexistence is enabled (eg oper mode 13, which is the default)
> the initialisation on startup sometimes silently fails.
>
> In a normal initialisation we see
> usb 1-1.3: Product: Wireless USB Network Module
> usb 1-1.3: Manufacturer: Redpine Signals, Inc.
> usb 1-1.3: SerialNumber: 000000000001
> rsi_91x: rsi_probe: Initialized os intf ops
> rsi_91x: rsi_load_9116_firmware: Loading chunk 0
> rsi_91x: rsi_load_9116_firmware: Loading chunk 1
> rsi_91x: rsi_load_9116_firmware: Loading chunk 2
> rsi_91x: Max Stations Allowed = 1
>
> But sometimes the last log is missing and the wlan net device is
> not created.
>
> Running a userspace loop that resets the hardware via a GPIO shows the
> problem occurring ~5/100 resets.
>
> The problem does not occur in oper mode 1 (wifi only).
>
> Adding logs shows that the initialisation state machine requests a MAC
> reset via rsi_send_reset_mac() but the firmware does not reply, leading
> to the initialisation sequence being incomplete.
>
> Fix this by delaying attaching the BT adapter until the wifi
> initialisation has completed.
>
> With this applied I have done > 300 reset loops with no errors.
>
> Fixes: 716b840c7641 ("rsi: handle BT traffic in driver")
> Signed-off-by: Martin Fuzzey <[email protected]>
> CC: [email protected]
Failed to apply, please rebase on top of wireless-drivers-next.
error: sha1 information is lacking or useless (drivers/net/wireless/rsi/rsi_91x_mgmt.c).
error: could not build fake ancestor
hint: Use 'git am --show-current-patch' to see the failed patch
Applying: rsi: fix occasional initialisation failure with BT coex
Patch failed at 0001 rsi: fix occasional initialisation failure with BT coex
3 patches set to Changes Requested.
12423167 [1/3] rsi: fix occasional initialisation failure with BT coex
12423171 [2/3] rsi: fix key enabled check causing unwanted encryption for vap_id > 0
12423173 [3/3] rsi: fix rate mask set leading to P2P failure
--
https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches