diff options
| author | Bryan O'Donoghue <bryan.odonoghue@linaro.org> | 2021-11-09 12:48:18 +0000 |
|---|---|---|
| committer | Bryan O'Donoghue <bryan.odonoghue@linaro.org> | 2021-12-10 12:48:03 +0000 |
| commit | 60447cbac82e70776bca6971358479c6be941140 (patch) | |
| tree | f3cf9ed25b575db030356a79e277fe85f8bd5593 | |
| parent | 7eb9e3f4152c126f3f7b1226024fcca9f1b772bb (diff) | |
wcn36xx: Fix maximum TX power to 30 dbm not 25 dbm.ath.git-wcn36xx-10-12-21-base
Downstream sets this value to 30 dbm not 25 dbm. This ends up making a
difference we can see in the field with a Google Nest. We deauthenticate
due to local choice to roam to a 5GHz channel with too low RSSI rather
than stay on 30 dbm indoor US 2.4GHz channels supported by Nest.
The firmware actually cares about the parameter value we pass, which is
not obvious at all except via trial and error.
Fixes: 8e84c2582169 ("wcn36xx: mac80211 driver for Qualcomm WCN3660/WCN3680 hardware")
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
| -rw-r--r-- | drivers/net/wireless/ath/wcn36xx/main.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c index 4074398eafef..f59eb1119da8 100644 --- a/drivers/net/wireless/ath/wcn36xx/main.c +++ b/drivers/net/wireless/ath/wcn36xx/main.c @@ -37,14 +37,14 @@ MODULE_PARM_DESC(debug_mask, "Debugging mask"); .band = NL80211_BAND_2GHZ, \ .center_freq = (_freq), \ .hw_value = (_idx), \ - .max_power = 25, \ + .max_power = 30, \ } #define CHAN5G(_freq, _idx, _phy_val) { \ .band = NL80211_BAND_5GHZ, \ .center_freq = (_freq), \ .hw_value = (_phy_val) << HW_VALUE_PHY_SHIFT | HW_VALUE_CHANNEL(_idx), \ - .max_power = 25, \ + .max_power = 30, \ } /* The wcn firmware expects channel values to matching |