On Wed, 2025-01-29 at 21:22 +0530, Roopni Devanathan wrote:
>
> --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
> +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
> @@ -126,7 +126,7 @@ struct iwl_mvm_time_event_data {
> /* Power management */
>
> /**
> - * enum iwl_power_scheme
> + * enum iwl_power_scheme - enum iwl power sceme set by debugfs
How does that belong here?!
> +++ b/net/mac80211/cfg.c
> @@ -3038,7 +3038,13 @@ static int ieee80211_set_wiphy_params(struct wiphy *wiphy, u8 radio_id, u32 chan
> }
>
> if (changed & WIPHY_PARAM_RTS_THRESHOLD) {
> - err = drv_set_rts_threshold(local, wiphy->rts_threshold);
> + u32 rts_threshold;
> +
> + if (radio_id >= wiphy->n_radio)
> + rts_threshold = wiphy->rts_threshold;
> + else
> + rts_threshold = wiphy->radio_cfg[radio_id].rts_threshold;
> + err = drv_set_rts_threshold(local, radio_id, rts_threshold);
Should we really just leave it all up to the driver, or perhaps call it
multiple times for each radio? Dunno.
> @@ -715,8 +716,14 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx)
> tx->sdata->vif.type == NL80211_IFTYPE_OCB);
>
> /* set up RTS protection if desired */
> - if (len > tx->local->hw.wiphy->rts_threshold) {
> - txrc.rts = true;
> + if (tx->local->hw.wiphy->n_radio) {
> + for (i = 0; i < tx->local->hw.wiphy->n_radio; i++) {
> + if (len > tx->local->hw.wiphy->radio_cfg[i].rts_threshold)
> + txrc.rts = true;
> + }
> + } else {
> + if (len > tx->local->hw.wiphy->rts_threshold)
> + txrc.rts = true;
> }
Are you sure you need this? Seems odd to me.
> +static ssize_t rts_threshold_read(struct file *file, char __user *user_buf,
> + size_t count, loff_t *ppos)
I'm not convinced it's worth keeping this at all since you can retrieve
it via nl80211?
johannes
