From mboxrd@z Thu Jan 1 00:00:00 1970 From: Francois Romieu Subject: Re: [PATCH v2 1/5] net: Support ethtool ops for rx of errored frames. Date: Mon, 20 Jun 2011 00:48:00 +0200 Message-ID: <20110619224800.GA2861@electric-eye.fr.zoreil.com> References: <1308430045-24816-1-git-send-email-greearb@candelatech.com> <1308430045-24816-2-git-send-email-greearb@candelatech.com> <20110618213400.GA2741@electric-eye.fr.zoreil.com> <4DFE57DA.8090404@candelatech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netdev@vger.kernel.org To: Ben Greear Return-path: Received: from violet.fr.zoreil.com ([92.243.8.30]:39278 "EHLO violet.fr.zoreil.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755013Ab1FSXAs (ORCPT ); Sun, 19 Jun 2011 19:00:48 -0400 Content-Disposition: inline In-Reply-To: <4DFE57DA.8090404@candelatech.com> Sender: netdev-owner@vger.kernel.org List-ID: Ben Greear : > On 06/18/2011 02:34 PM, Francois Romieu wrote: > >greearb@candelatech.com : > >[...] > >>This can be useful when sniffing dodgy networks. > > > >Do you plan to add something similar - i.e. not per packet - for the Tx path ? > > Ability to tx errored frames? I posted a patch to enable sending > frames with custom (ie, invalid) Ethernet FCS, but you have to > enable it per-socket, and it will only work with AF_PACKET sockets. > > Can you offer more details on what you are asking for? Realtek's 816[89] and 810[23] have no room in their Tx descriptors to control ethernet checksum generation. There is a global register (TxConfig) though. Before looking at the per-socket information, the driver would need to be instructed to disable Tx hardware ethernet checksumming globally. -- Ueimor