From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rick Jones Subject: Re: [RFC] bnx2x: Insane RX rings Date: Fri, 10 Sep 2010 08:46:06 -0700 Message-ID: <4C8A52BE.9040105@hp.com> References: <1284065105.4782.11.camel@edumazet-laptop> <4C894FBD.2020109@ans.pl> <20100909.143001.104050644.davem@davemloft.net> <4C8953D3.9060204@hp.com> <1284117374.30831.2.camel@lb-tlvb-eilong.il.broadcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: David Miller , "ole@ans.pl" , "eric.dumazet@gmail.com" , "netdev@vger.kernel.org" To: eilong@broadcom.com Return-path: Received: from g5t0009.atlanta.hp.com ([15.192.0.46]:29255 "EHLO g5t0009.atlanta.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751941Ab0IJPqJ (ORCPT ); Fri, 10 Sep 2010 11:46:09 -0400 In-Reply-To: <1284117374.30831.2.camel@lb-tlvb-eilong.il.broadcom.com> Sender: netdev-owner@vger.kernel.org List-ID: >>>I think simply that the RX queue size should be scaled by the number >>>of queues we have. > > > There are few factors that can be considered when scaling the ring > sizes: > - Number of queues per device > - Number of devices > - Available amount of memory > - Others... > > I'm thinking about adding a factor only according to the number of > queues - this will still cause issues for systems with many ports. Does > that sound reasonable or not enough? Do you think the number of devices > or even the amount of free memory should be considered? At one level we are talking about horses and barn doors - for example, the minimum memory requirements for ProLiants have already been set (and communicated for some time) taking memory usage of their LOMs (Lan On Motherboard) into account. rick jones