Return-path: Received: from an-out-0708.google.com ([209.85.132.251]:20272 "EHLO an-out-0708.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756578AbYLLPiW (ORCPT ); Fri, 12 Dec 2008 10:38:22 -0500 Received: by an-out-0708.google.com with SMTP id d40so689892and.1 for ; Fri, 12 Dec 2008 07:38:21 -0800 (PST) Message-ID: (sfid-20081212_163826_194763_32C4EC02) Date: Fri, 12 Dec 2008 10:38:20 -0500 From: "Bob Copeland" To: "Benoit PAPILLAULT" Subject: Re: [ath5k-devel] [PATCH] ath5k : Fix correct padding Cc: "Johannes Berg" , ath5k-devel@lists.ath5k.org, linux-wireless@vger.kernel.org, linville@tuxdriver.com, "Harvey Harrison" In-Reply-To: <49422598.5050204@free.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 References: <49418D23.8060905@free.fr> <1229033298.15776.18.camel@brick> <1229033411.8081.91.camel@johannes.berg> <49422598.5050204@free.fr> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Fri, Dec 12, 2008 at 3:49 AM, Benoit PAPILLAULT wrote: > Nice catch! I wonder how it was working in my preliminary tests. Sorry > for the duplicates with the same patch for ath9k. The padding/unpadding > stuff is needed by the hardware, ie the hardware insert padding on RX > and expect padding on TX, as mentioned in earlier ath5k and madwifi > source code. I should clarify: I meant Patrick already posted a fix for ath5k (and possibly the ensuing discussion led to the ath9k patch). Anyway, let's just get a patch in :) The extra padding on TX side shouldn't hurt anything, OTOH, yeah it is probably unnecessary to do it for ACK/CTS. Also take a look at desc.c when setting up TX descriptor: frame_len = pkt_len - (hdr_len & 3) + FCS_LEN; What we have now will be 10 - 2 + 4 = 12 vs what should probably be 14. -- Bob Copeland %% www.bobcopeland.com