Return-path: Received: from mail-wi0-f170.google.com ([209.85.212.170]:36467 "EHLO mail-wi0-f170.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751470AbbGNFNS convert rfc822-to-8bit (ORCPT ); Tue, 14 Jul 2015 01:13:18 -0400 Received: by widjy10 with SMTP id jy10so88123177wid.1 for ; Mon, 13 Jul 2015 22:13:17 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <55A4446C.4090803@candelatech.com> References: <55A4446C.4090803@candelatech.com> Date: Tue, 14 Jul 2015 07:13:17 +0200 Message-ID: (sfid-20150714_071322_096032_64F33BEA) Subject: Re: Don't enable PROMISC when adding AP to a bridge? From: Michal Kazior To: Ben Greear Cc: "linux-wireless@vger.kernel.org" , ath10k Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 14 July 2015 at 01:06, Ben Greear wrote: > I notice some of my WLE900VX cards will lock up and not receive any packets > after a few seconds of being in PROMISC mode. This happens on my own > CT firmware as well as stock firmware-2.bin. This breaks monitor mode, of > course, but it also breaks bridged AP mode. > > So, I found this patch from way back when: > http://lists.celinuxforum.org/pipermail/bridge/2008-June/005906.html > > I hacked a similar thing into my 4.0.4 kernel, and now bridged mode > AP seems much more stable (and the radio does not go into PROMISC mode, > so it should be more efficient in a busy RF environment). > > Maybe this feature could be revisited? This has already been done and mac80211 in 4.2-rc no longer propagates promisc to drivers[1]. [1]: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/net/mac80211?id=df1404650ccbfeb76a84f301f22316be0d00a864 Michał