From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sridhar Samudrala Subject: Re: [WIP] [PATCH] WAS Re: [RFC] New driver API to speed up small packets xmits Date: Wed, 16 May 2007 14:44:41 -0700 Message-ID: <464B7B49.4070003@us.ibm.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: ak@suse.de, ashwin.chaugule@celunite.com, David Miller , Gagan Arneja , hadi@cyberus.ca, Evgeniy Polyakov , netdev@vger.kernel.org, netdev-owner@vger.kernel.org, rdreier@cisco.com, xma@us.ibm.com To: Krishna Kumar2 Return-path: Received: from e6.ny.us.ibm.com ([32.97.182.146]:42978 "EHLO e6.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756106AbXEQEpN (ORCPT ); Thu, 17 May 2007 00:45:13 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Krishna Kumar2 wrote: > Hi Sridhar, > > Sridhar Samudrala wrote on 05/17/2007 03:42:03 AM: > >> AFAIK, gso_skb can be a list of skb's. Can we add a list >> to another list using __skb_queue_head()? >> Also, if gso_skb is a list of multiple skb's, i think the >> count needs to be decremented by the number of segments in >> gso_skb. > > gso_skb is the last GSO skb that failed to be sent. This already > segmented skb is kept "cached" and whenever the next xmit happens, > this skb is first sent before any packets from the queue are taken > out (otherwise out-of-order packets). So there can atmost be one > gso_skb per device. Yes. There can be only one gso_skb per device. But it can have more than one segments linked together via skb->next. Thanks Sridhar