From mboxrd@z Thu Jan 1 00:00:00 1970 From: "ezequiel.garcia@free-electrons.com" Subject: Re: [PATCH v1 4/6] net: fec: Increase buffer descriptor entry number Date: Fri, 30 May 2014 11:01:37 -0300 Message-ID: <20140530140137.GC783@arch.cereza> References: <1401415552-2263-1-git-send-email-b38611@freescale.com> <1401415552-2263-5-git-send-email-b38611@freescale.com> <063D6719AE5E284EB5DD2968C1650D6D1725053E@AcuExch.aculab.com> <1401455586.3645.96.camel@edumazet-glaptop2.roam.corp.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "fugang.duan@freescale.com" , David Laight , "Frank.Li@freescale.com" , "davem@davemloft.net" , "netdev@vger.kernel.org" , "shawn.guo@linaro.org" , "bhutchings@solarflare.com" , "stephen@networkplumber.org" To: Eric Dumazet Return-path: Received: from top.free-electrons.com ([176.31.233.9]:57621 "EHLO mail.free-electrons.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S933250AbaE3OFc (ORCPT ); Fri, 30 May 2014 10:05:32 -0400 Content-Disposition: inline In-Reply-To: <1401455586.3645.96.camel@edumazet-glaptop2.roam.corp.google.com> Sender: netdev-owner@vger.kernel.org List-ID: Hello Eric, On 30 May 06:13 AM, Eric Dumazet wrote: > On Fri, 2014-05-30 at 09:42 +0000, fugang.duan@freescale.com wrote: > > From: David Laight Data: Friday, May 30, = 2014 5:11 PM > > > > > >Does the driver support BQL (or similar) in order to limit the amo= unt of > > >queued tx traffic? > > >Otherwise you've significantly increased the latency for connectio= ns other > > >than one doing bulk tx. > > > > > > David > > > > > The driver still don't support BQL. > > I will add the feature to support FEC. Thanks for your advise. >=20 > Note that a full size TSO packet (44 or 45 MSS) requires about 88 or = 90 > descriptors. >=20 What's the rationale behing those numbers? =46ollowing the Solarflare commit you pointed me a while ago, I've set gso_max_segs to limit the number of descriptors in the mvneta and mv643xx_eth drivers. My understanding reading the sfc commit 7e6d06f0de3f7 is that the maximum number of required descriptors would be around ~250: /* Header and payload descriptor for each output segment, plus * one for every input fragment boundary within a segment */ unsigned int max_descs =3D EFX_TSO_MAX_SEGS * 2 + MAX_SKB_FRAGS; Maybe I've misunderstood something? PS: I was also about to add BQL support to the drivers... --=20 Ezequiel Garc=EDa, Free Electrons Embedded Linux, Kernel and Android Engineering http://free-electrons.com