> > diff --git a/wpa_supplicant/sme.c b/wpa_supplicant/sme.c @@ -971,7
> > +971,8 @@ static void sme_handle_external_auth_start(struct
> wpa_supplicant *wpa_s,
> > for (ssid = wpa_s->conf->ssid; ssid; ssid = ssid->next) {
> > if (!wpas_network_disabled(wpa_s, ssid) &&
> > ssid_str_len == ssid->ssid_len &&
> > - os_memcmp(ssid_str, ssid->ssid, ssid_str_len) == 0)
> > + os_memcmp(ssid_str, ssid->ssid, ssid_str_len) == 0 &&
> > + ssid->key_mgmt == WPA_KEY_MGMT_SAE)
> > break;
>
> This does not look correct to me.. ssid->key_mgmt is a bitfield of enabled
> options. Comparing it to a single one (WPA_KEY_MGMT_SAE) would mean that
> any network profile with both SAE and something else enabled, e.g., the
> common transition mode option with WPA_KEY_MGMT_PSK, would be skipped.
> Shouldn't this be (ssid->key_mgmt & WPA_KEY_MGMT_SAE) instead?
You're right it should be (ssid->key_mgmt & WPA_KEY_MGMT_SAE).
I'll resend the patch
cedric
_______________________________________________
Hostap mailing list
Hostap@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/hostap
