auth_mode in cyw43_ev_scan_result_t is wrong

[peterharperuk]

The auth_mode field in cyw43_ev_scan_result_t isn't right. (Did I add the doxygen?)

/*!
 * \brief Structure to return wifi scan results
 */
//!\{
typedef struct _cyw43_ev_scan_result_t {
    uint32_t _0[5];
    uint8_t bssid[6];   ///< access point mac address
    uint16_t _1[2];
    uint8_t ssid_len;   ///< length of wlan access point name
    uint8_t ssid[32];   ///< wlan access point name
    uint32_t _2[5];
    uint16_t channel;   ///< wifi channel
    uint16_t _3;
    uint8_t auth_mode;  ///< wifi auth mode \ref CYW43_AUTH_
    int16_t rssi;       ///< signal strength
} cyw43_ev_scan_result_t;
//!\}

It's only 8 bits when it needs to be 32 bits. I think it's actually the dtim period if you look where it's pulled from the frame.
I have some code that extracts this value from the RSN information element if you would like this to be fixed? I think it's useful to know if an AP is WPA2 or WPA3?

[dpgeorge]

It's only 8 bits when it needs to be 32 bits.

Hmm, the values put into auth_mode are just bitwise or of 1, 2 and 4, so fit in 8 bits.

But, they don't properly correspond to the standard CYW43_AUTH_xxx constants (it looks like they match the lower 8-bits of those constants only).

I have some code that extracts this value from the RSN information element if you would like this to be fixed? I think it's useful to know if an AP is WPA2 or WPA3?

Is there enough information in the scan result to reconstruct the correct CYW43_AUTH_xxx constant for each result? If so that sounds good, thanks!

[broughtong]

Just got caught out by this too!

Thought I could take the scan results auth_mode field and then feed it back in to attempt to make a connection...nope!

My scan result can back as '5', lucky I found this thread as even with a bit of digging + docs I really couldn't see how it would map to 0x00200002, 0x00400004, or 0x00400006