Hi, Actually, I think my suggestion _would_ in fact fix the whole issue. > diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c > index 51622333d460..70d6de29425b 100644 > --- a/net/mac80211/cfg.c > +++ b/net/mac80211/cfg.c > @@ -2934,19 +2934,20 @@ static int > ieee80211_start_radar_detection(struct wiphy *wiphy, > memcpy(pos, beacon->probe_resp, beacon->probe_resp_len); > pos += beacon->probe_resp_len; > } > - if (beacon->ftm_responder) > + if (beacon->ftm_responder != -1) { This doesn't make sense without the change I suggested, since if we don't do the change I suggested, beacon->ftm_responder will never actually be -1. I'd change it like this, fixing the memory overflow bug along the way: diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 51622333d460..818aa0060349 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -2891,7 +2891,7 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon) len = beacon->head_len + beacon->tail_len + beacon->beacon_ies_len + beacon->proberesp_ies_len + beacon->assocresp_ies_len + - beacon->probe_resp_len; + beacon->probe_resp_len + beacon->lci_len + beacon->civicloc_len; new_beacon = kzalloc(sizeof(*new_beacon) + len, GFP_KERNEL); if (!new_beacon) @@ -2934,8 +2934,9 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon) memcpy(pos, beacon->probe_resp, beacon->probe_resp_len); pos += beacon->probe_resp_len; } - if (beacon->ftm_responder) - new_beacon->ftm_responder = beacon->ftm_responder; + + /* might copy -1, meaning no changes requested */ + new_beacon->ftm_responder = beacon->ftm_responder; if (beacon->lci) { new_beacon->lci_len = beacon->lci_len; new_beacon->lci = pos; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 744b5851bbf9..8d763725498c 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -7870,6 +7870,7 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) } memset(¶ms, 0, sizeof(params)); + params.beacon_csa.ftm_responder = -1; if (!info->attrs[NL80211_ATTR_WIPHY_FREQ] || !info->attrs[NL80211_ATTR_CH_SWITCH_COUNT]) If that seems good to you I'll submit the patch. johannes