From patchwork Sat Jun 10 10:52:43 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Luca Coelho X-Patchwork-Id: 9779781 X-Patchwork-Delegate: johannes@sipsolutions.net Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id DAA5F6034B for ; Sat, 10 Jun 2017 10:53:30 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id D597B2870F for ; Sat, 10 Jun 2017 10:53:30 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id C9E1328712; Sat, 10 Jun 2017 10:53:30 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 0286B2871B for ; Sat, 10 Jun 2017 10:53:29 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752009AbdFJKw5 (ORCPT ); Sat, 10 Jun 2017 06:52:57 -0400 Received: from paleale.coelho.fi ([176.9.41.70]:60204 "EHLO farmhouse.coelho.fi" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751941AbdFJKwz (ORCPT ); Sat, 10 Jun 2017 06:52:55 -0400 Received: from a91-156-4-241.elisa-laajakaista.fi ([91.156.4.241] helo=redipa.ger.corp.intel.com) by farmhouse.coelho.fi with esmtpsa (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.89) (envelope-from ) id 1dJe0j-0003L3-0A; Sat, 10 Jun 2017 13:52:53 +0300 From: Luca Coelho To: johannes@sipsolutions.net Cc: linux-wireless@vger.kernel.org, Johannes Berg Date: Sat, 10 Jun 2017 13:52:43 +0300 Message-Id: <20170610105246.26455-2-luca@coelho.fi> X-Mailer: git-send-email 2.11.0 In-Reply-To: <20170610105246.26455-1-luca@coelho.fi> References: <20170610105246.26455-1-luca@coelho.fi> X-SA-Exim-Connect-IP: 91.156.4.241 X-SA-Exim-Mail-From: luca@coelho.fi Subject: [PATCH 1/4] mac80211: set bss_info data before configuring the channel X-SA-Exim-Version: 4.2.1 (built Tue, 02 Aug 2016 21:08:31 +0000) X-SA-Exim-Scanned: Yes (on farmhouse.coelho.fi) Sender: linux-wireless-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-wireless@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Johannes Berg When mac80211 changes the channel, it also calls into the driver's bss_info_changed() callback, e.g. with BSS_CHANGED_IDLE. The driver may, like iwlwifi does, access more data from bss_info in that case and iwlwifi accesses the basic_rates bitmap, but if changing from a band with more (basic) rates to one with fewer, an out-of-bounds access of the rate array may result. While we can't avoid having invalid data at some point in time, we can avoid having it while we call the driver - so set up all the data before configuring the channel, and then apply it afterwards. This fixes https://bugzilla.kernel.org/show_bug.cgi?id=195677 Reported-by: Johannes Hirte Tested-by: Johannes Hirte Debugged-by: Emmanuel Grumbach Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 38 ++++++++++++++++++++++++++++---------- 1 file changed, 28 insertions(+), 10 deletions(-) diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 1ae9be090309..570d9ab61950 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -4397,15 +4397,19 @@ static int ieee80211_prep_connection(struct ieee80211_sub_if_data *sdata, return -ENOMEM; } - if (new_sta || override) { - err = ieee80211_prep_channel(sdata, cbss); - if (err) { - if (new_sta) - sta_info_free(local, new_sta); - return -EINVAL; - } - } - + /* + * Set up the information for the new channel before setting the + * new channel. We can't - completely race-free - change the basic + * rates bitmap and the channel (sband) that it refers to, but if + * we set it up before we at least avoid calling into the driver's + * bss_info_changed() method with invalid information (since we do + * call that from changing the channel - only for IDLE and perhaps + * some others, but ...). + * + * So to avoid that, just set up all the new information before the + * channel, but tell the driver to apply it only afterwards, since + * it might need the new channel for that. + */ if (new_sta) { u32 rates = 0, basic_rates = 0; bool have_higher_than_11mbit; @@ -4488,8 +4492,22 @@ static int ieee80211_prep_connection(struct ieee80211_sub_if_data *sdata, sdata->vif.bss_conf.sync_dtim_count = 0; } rcu_read_unlock(); + } - /* tell driver about BSSID, basic rates and timing */ + if (new_sta || override) { + err = ieee80211_prep_channel(sdata, cbss); + if (err) { + if (new_sta) + sta_info_free(local, new_sta); + return -EINVAL; + } + } + + if (new_sta) { + /* + * tell driver about BSSID, basic rates and timing + * this was set up above, before setting the channel + */ ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_BSSID | BSS_CHANGED_BASIC_RATES | BSS_CHANGED_BEACON_INT);