Return-path: Received: from mail-pz0-f46.google.com ([209.85.210.46]:58637 "EHLO mail-pz0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755393Ab1EDTz5 (ORCPT ); Wed, 4 May 2011 15:55:57 -0400 Received: by pzk9 with SMTP id 9so647750pzk.19 for ; Wed, 04 May 2011 12:55:57 -0700 (PDT) Message-ID: <4DC1AF49.7080204@lwfinger.net> (sfid-20110504_215600_352107_FAC3642B) Date: Wed, 04 May 2011 14:55:53 -0500 From: Larry Finger MIME-Version: 1.0 To: Johannes Berg , John Linville CC: wireless Subject: WARNING from mac80211 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: Johannes and John, A regression in recent versions of mac80211 are generating the WARNING at net/mac80211/wep.c:101, because the skb headroom is less than WEP_IV_LEN or the tailroom is less than WEP_ICV_LEN. My testing indicates that it is a headroom problem as the following hack "fixes" it: Index: wireless-testing-new/net/mac80211/tx.c =================================================================== --- wireless-testing-new.orig/net/mac80211/tx.c +++ wireless-testing-new/net/mac80211/tx.c @@ -1935,6 +1935,8 @@ netdev_tx_t ieee80211_subif_start_xmit(s head_need += IEEE80211_ENCRYPT_HEADROOM; head_need += local->tx_headroom; head_need = max_t(int, 0, head_need); + if (head_need < WEP_IV_LEN) + head_need = WEP_IV_LEN; if (ieee80211_skb_resize(local, skb, head_need, true)) goto fail; } I'm certain that this is too simple minded, but it at least points to a place to look. I have not bisected the problem to see which commit caused the regression, but I can if you want. Larry