From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail2.candelatech.com (mail2.candelatech.com [208.74.158.173]) by huchra.bufferbloat.net (Postfix) with ESMTP id BB40C21F22A for ; Sun, 13 Jul 2014 12:44:42 -0700 (PDT) Received: from [192.168.1.32] (93-137-137-233.adsl.net.t-com.hr [93.137.137.233]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by mail2.candelatech.com (Postfix) with ESMTPSA id 8C28240B146; Sun, 13 Jul 2014 12:44:40 -0700 (PDT) Message-ID: <53C2E1A6.1070001@candelatech.com> Date: Sun, 13 Jul 2014 12:44:38 -0700 From: Ben Greear Organization: Candela Technologies User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.5.0 MIME-Version: 1.0 To: Dave Taht , Sebastian Moeller References: In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: "ath9k-devel@lists.ath9k.org" , cerowrt-devel Subject: Re: [Cerowrt-devel] [ath9k-devel] periodic hang of ath9k X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 13 Jul 2014 19:44:42 -0000 On 07/13/2014 12:18 PM, Dave Taht wrote: > cc-ing ath9k-devel for this update on http://www.bufferbloat.net/issues/442 > > this bug, which some people (usually on macs with low signal strength) > can get to occur fairly rapidly, but I can't, is driving me 9 kinds of > crazy... I've mentioned this before..but I have a patch that works around this problem: I had merge issues, so now it's two patches until I have time to clean up my tree: http://dmz2.candelatech.com/git/gitweb.cgi?p=linux-3.14.dev.y/.git;a=commit;h=579035d4b2b7110d67ff451e728709f5b2510ca0 http://dmz2.candelatech.com/git/gitweb.cgi?p=linux-3.14.dev.y/.git;a=commit;h=ae6f95e8a35578e0668e47becb053b29009eec9a I suspect this may cause memory leaks, as a test that causes these work-arounds to hit often will run out of memory after a few days (200+ station vifs constantly doing traffic). Could be lots of other code of course, and kmemleak shows no leaks, so I'm at a loss on the mem leak issue. If you can, try patching these in and see if that works around your tx hang problem. I'd love to have a better fix, btw. Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com