Sergey Matyukevich <sergey.matyukevich.os@xxxxxxxxxxxxx> writes:
> This patch implements all the necessary features needed to keep
> operating channel information in sync between firmware and driver:
>
> - implement cfg80211 qtnf_get_channel handler
> - implement QLINK_EVENT_FREQ_CHANGE event handling
> using this event firmware notifies host when operating channel is changed
> - update qtnf_dump_survey
> specify current channel in use in survey report
> - update qtnf_connect
> pass to firmware channel number rather than frequency
>
> Signed-off-by: Sergey Matyukevich <sergey.matyukevich.os@xxxxxxxxxxxxx>
[...]
> +static int
> +qtnf_event_handle_freq_change(struct qtnf_wmac *mac,
> + const struct qlink_event_freq_change *data,
> + u16 len)
> +{
> + struct wiphy *wiphy = priv_to_wiphy(mac);
> + struct cfg80211_chan_def chandef;
> + struct ieee80211_channel *chan;
> + struct qtnf_bss_config *bss_cfg;
> + struct qtnf_vif *vif;
> + int freq;
> + int i;
> +
> + if (len < sizeof(*data)) {
> + pr_err("payload is too short\n");
> + return -EINVAL;
> + }
> +
> + freq = le32_to_cpu(data->freq);
> + chan = ieee80211_get_channel(wiphy, freq);
> + if (!chan) {
> + pr_err("channel at %d MHz not found\n", freq);
> + return -EINVAL;
> + }
> +
> + pr_info("MAC%d switch to new channel %u MHz\n", mac->macid, freq);
Should this be a debug message? I don't see why the driver would need to
spam this every time.
--
Kalle Valo
