On Thu, Sep 12, 2019 at 04:04:29PM -0600, Ahmed Zaki wrote: > The check for the ESS or IBSS bits in the BSS caps is done via equality > to IEEE80211_CAP_ESS (0x01). This will only be true for AP/ESS and will > fail in case of IBSS (0x02). What do you mean with "fail" in this context? An IBSS is not an ESS, so it sounds correct to me for bss_is_ess() to return 0 if the BSS is an IBSS. > diff --git a/wpa_supplicant/events.c b/wpa_supplicant/events.c > @@ -906,8 +906,7 @@ static int bss_is_ess(struct wpa_bss *bss) > IEEE80211_CAP_DMG_AP; > } > > - return ((bss->caps & (IEEE80211_CAP_ESS | IEEE80211_CAP_IBSS)) == > - IEEE80211_CAP_ESS); > + return (bss->caps & IEEE80211_CAP_ESS || bss->caps & IEEE80211_CAP_IBSS); This would break bss_is_ess() by making it incorrectly claim an IBSS to be an IBSS. Please note that the only caller of bss_is_ess() used to originally do "if (bss->caps & IEEE80211_CAP_IBSS) continue" which is what is still happening with the current bss_is_ess() implementation while this proposed change would negate that. What are you trying to fix with this? -- Jouni Malinen PGP id EFC895FA _______________________________________________ Hostap mailing list Hostap@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/hostap
