From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mailfront3.rz.tu-bs.de ([134.169.12.134] helo=tu-bs.de) by bombadil.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1awBnt-0007l5-2l for ath10k@lists.infradead.org; Fri, 29 Apr 2016 17:02:09 +0000 Received: from [217.247.128.186] (account hanharms@tu-braunschweig.de HELO [172.17.10.21]) by tu-braunschweig.de (CommuniGate Pro SMTP 6.1.10f) with ESMTPA id 45907094 for ath10k@lists.infradead.org; Fri, 29 Apr 2016 19:01:45 +0200 Subject: Re: ath10k IBSS and A-MSDU throuput References: <5723831A.7010607@tu-bs.de> <572389ED.6080306@candelatech.com> From: Hannes Harms Message-ID: <57239376.2090701@tu-bs.de> Date: Fri, 29 Apr 2016 19:01:42 +0200 MIME-Version: 1.0 In-Reply-To: <572389ED.6080306@candelatech.com> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: "ath10k" Errors-To: ath10k-bounces+kvalo=adurom.com@lists.infradead.org To: ath10k@lists.infradead.org To sniff frames on air I would need additional hardware, right? Or maybe there is a possibility to debug in / outgoing amsdu frames in the ath10k driver? Unfortunately my kernel programming skills are very limited, but maybe I will continue testing next week. Thanks, Hannes Am 29.04.2016 um 18:21 schrieb Ben Greear: > > > On 04/29/2016 08:51 AM, Harms, Hannes wrote: >> Hi everyone, >> >> I have set up two Compex WLE900 in ad-hoc mode, 80MHz channel width, >> channel 100. >> I am using kernel 3.17 from Gateworks for imx6 and >> my own compat_wirless (4.4.2) with all patches from Candela for >> ath10k and all patches for mac80211 from the >> Candela 4.4 kernel tree. >> >> The upload for TCP is about 200 MBit/s and the download is about 250 >> MBit/s, tested with iperf3. >> >> Now I am trying to enable A-MSDU as described in the firmware release >> 10.1.467-ct-15 from Candela, >> but I can not measure anydifference in throughput. >> >> Could anyone confirm that A-MSDU is really working with IBSS? > > You can sniff it and see if you can sniff amsdu frames on air? > > I only tested it a bit when developing that feature for another user > (who eventually decided to not use it anyway since it required linux > kernel hacks to make it work), so possibly there was a regression... > > Thanks, > Ben > _______________________________________________ ath10k mailing list ath10k@lists.infradead.org http://lists.infradead.org/mailman/listinfo/ath10k