Date: Sun, 31 Mar 2013 05:35:14 -0700 (PDT) From: Barney Cordoba <barney_cordoba@yahoo.com> To: Jeffrey EPieper <jeffrey.e.pieper@intel.com>, Jack Vogel <jfvogel@gmail.com> Cc: Nick Rogers <ncrogers@gmail.com>, "freebsd-net@freebsd.org" <freebsd-net@freebsd.org>, "Clement Hermann \(nodens\)" <nodens2099@gmail.com> Subject: Re: igb and ALTQ in 9.1-rc3 Message-ID: <1364733314.80531.YahooMailClassic@web121603.mail.ne1.yahoo.com> In-Reply-To: <CAFOYbc==t%2B=riaMpQiWg0dg=3QgsCOnsiBxTBqq%2BW08nGra6MA@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
The reason that Jack is a no better programmer now than he was in 2009 migh= t have=A0something to do with the fact that he hides when his work is criti= cized.=A0 Why not release the benchmarks you did while designing the igb driver, Jack= ? Say what,you didn't do any benchmarking? How does the default driver perf= orm, say in a firewall,with 1000 user load? What's the optimum number of qu= eues to use in such a system?What's the effect of CPU binding? What's the e= ffect with multiple cards when you havemore queues than you have physical c= pus?=A0 What made you decide to use buf_ring? Something new to play with? I'm guessing that you have no idea. BC--- On Fri, 3/29/13, Jack Vogel <jfvogel@gmail.com> wrote: From: Jack Vogel <jfvogel@gmail.com> Subject: Re: igb and ALTQ in 9.1-rc3 To: "Pieper, Jeffrey E" <jeffrey.e.pieper@intel.com> Cc: "Barney Cordoba" <barney_cordoba@yahoo.com>, "Nick Rogers" <ncrogers@gm= ail.com>, "freebsd-net@freebsd.org" <freebsd-net@freebsd.org>, "Clement Her= mann (nodens)" <nodens2099@gmail.com> Date: Friday, March 29, 2013, 12:36 PM Fortunately, Barney doesn't speak for me, or for Intel, and I've long ago r= ealized its pointless to attempt anything like a fair conversation with him. The only thing he's eve= r contributed is slander =0Aand pseudo-critique... another poison thread I'm done with. Jack On Fri, Mar 29, 2013 at 8:45 AM, Pieper, Jeffrey E <jeffrey.e.pieper@intel.= com> wrote: =0A =0A =0A-----Original Message----- =0AFrom: owner-freebsd-net@freebsd.org [mailto:owner-freebsd-net@freebsd.or= g] On Behalf Of Barney Cordoba =0ASent: Friday, March 29, 2013 5:51 AM =0ATo: Jack Vogel; Nick Rogers =0ACc: freebsd-net@freebsd.org; Clement Hermann (nodens) =0ASubject: Re: igb and ALTQ in 9.1-rc3 =0A =0A =0A =0A--- On Thu, 3/28/13, Nick Rogers <ncrogers@gmail.com> wrote: =0A =0A> From: Nick Rogers <ncrogers@gmail.com> =0A> Subject: Re: igb and ALTQ in 9.1-rc3 =0A> To: "Jack Vogel" <jfvogel@gmail.com> =0A> Cc: "Barney Cordoba" <barney_cordoba@yahoo.com>, "Clement Hermann (nod= ens)" <nodens2099@gmail.com>, "freebsd-net@freebsd.org" <freebsd-net@freebs= d.org> =0A=0A> Date: Thursday, March 28, 2013, 9:29 PM =0A> On Thu, Mar 28, 2013 at 4:16 PM, Jack =0A> Vogel <jfvogel@gmail.com> =0A> wrote: =0A> > Have been kept fairly busy with other matters, one =0A> thing I could do short =0A> > term is =0A> > change the defines in igb the way I did in the em =0A> driver so you could still =0A> > define =0A> > the older if_start entry. Right now those are based on =0A> OS version and so you =0A> > will =0A> > automatically get if_transmit, but I could change it to =0A> be IGB_LEGACY_TX or =0A> > so, =0A> > and that could be defined in the Makefile. =0A> > =0A> > Would this help? =0A> =0A> I'm currently using ALTQ successfully with the em driver, so =0A> if igb =0A> behaved the same with respect to using if_start instead of =0A> if_transmit =0A> when ALTQ is in play, that would be great. I do not =0A> completely =0A> understand the change you propose as I am not very familiar =0A> with the =0A> driver internals. Any kind of patch or extra =0A> Makefile/make.conf =0A> definition that would allow me to build a 9-STABLE kernel =0A> with an igb =0A> driver that works again with ALTQ, ASAP, would be much =0A> appreciated. =0A> =0A> > =0A> > Jack =0A> > =0A> > =0A> > =0A> > On Thu, Mar 28, 2013 at 2:31 PM, Nick Rogers <ncrogers@gmail.com> =0A> wrote: =0A> >> =0A> >> On Tue, Dec 11, 2012 at 1:09 AM, Jack Vogel <jfvogel@gmail.com> =0A> wrote: =0A> >> > On Mon, Dec 10, 2012 at 11:58 PM, Gleb =0A> Smirnoff <glebius@freebsd.org> =0A> >> > wrote: =0A> >> > =0A> >> >> On Mon, Dec 10, 2012 at 03:31:19PM -0800, =0A> Jack Vogel wrote: =0A> >> >> J> UH, maybe asking the owner of the =0A> driver would help :) =0A> >> >> J> =0A> >> >> J> ... and no, I've never been aware of =0A> doing anything to stop =0A> >> >> supporting =0A> >> >> altq =0A> >> >> J> so you wouldn't see any commits. If =0A> there's something in the altq =0A> >> >> code =0A> >> >> or =0A> >> >> J> support (which I have nothing to do =0A> with) that caused this no-one =0A> >> >> informed =0A> >> >> J> me. =0A> >> >> =0A> >> >> Switching from if_start to if_transmit =0A> effectively disables ALTQ =0A> >> >> support. =0A> >> >> =0A> >> >> AFAIR, there is some magic implemented in =0A> other drivers that makes them =0A> >> >> modern (that means using if_transmit), but =0A> still capable to switch to =0A> >> >> queueing =0A> >> >> mode if SIOCADDALTQ was casted upon them. =0A> >> >> =0A> >> >> =0A> >> > Oh, hmmm, I'll look into the matter after my =0A> vacation. =0A> >> > =0A> >> > Jack =0A> >> =0A> >> Has there been any progress on resolving this =0A> issue? I recently ran =0A> >> into this problem upgrading my servers from 8.3 to =0A> 9.1-RELEASE and am =0A> >> wondering what the latest recommendation is. I've =0A> used ALTQ and igb =0A> >> successfully for years and it is unfortunate it no =0A> longer works. =0A> >> Appreciate any advice. =0A> >> =0A> =0A>Do yourself a favor and either get a cheap dual port 82571 card or =0A>2 cards and disable the IGB ports. The igb driver is defective, and unt= il =0A>they back out the new, untested multi-queue stuff you're just neutering =0A>your system trying to use it. =0A> =0A>Frankly this project made a huge mistake by moving forward with multi =0A>queue just for the sake of saying that you support it; without having =0A>any credible plan for implementing it. That nonsense that Bill Macy did =0A>should have been tarballed up and deposited in the trash folder. The =0A>biggest mess in programming history. =0A> =0A>That being said, the solution is not to hack the igb driver; its to mak= e =0A>ALTQ if_transmit compatible, which shouldn't be all that difficult. =0A> =0A>BC =0A =0AI may be misunderstanding what you are saying, but if the solution is, a= s you say "not to hack the igb driver", then how is it defective in this ca= se? Or are you just directing vitriol toward Intel? Multi-queue is working = fine in igb. =0A=0A =0AJeff =0A =0A_______________________________________________ =0Afreebsd-net@freebsd.org mailing list =0Ahttp://lists.freebsd.org/mailman/listinfo/freebsd-net =0ATo unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org" =0A_______________________________________________ =0Afreebsd-net@freebsd.org mailing list =0Ahttp://lists.freebsd.org/mailman/listinfo/freebsd-net =0ATo unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org" =0A =0A From owner-freebsd-net@FreeBSD.ORG Sun Mar 31 12:42:48 2013 Return-Path: <owner-freebsd-net@FreeBSD.ORG> Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.FreeBSD.org [8.8.178.115]) by hub.freebsd.org (Postfix) with ESMTP id D4A6F168 for <freebsd-net@freebsd.org>; Sun, 31 Mar 2013 12:42:48 +0000 (UTC) (envelope-from barney_cordoba@yahoo.com) Received: from nm32-vm5.bullet.mail.ne1.yahoo.com (nm32-vm5.bullet.mail.ne1.yahoo.com [98.138.229.53]) by mx1.freebsd.org (Postfix) with ESMTP id 9D6D3AF7 for <freebsd-net@freebsd.org>; Sun, 31 Mar 2013 12:42:48 +0000 (UTC) Received: from [98.138.90.57] by nm32.bullet.mail.ne1.yahoo.com with NNFMP; 31 Mar 2013 12:42:42 -0000 Received: from [98.138.89.197] by tm10.bullet.mail.ne1.yahoo.com with NNFMP; 31 Mar 2013 12:42:42 -0000 Received: from [127.0.0.1] by omp1055.mail.ne1.yahoo.com with NNFMP; 31 Mar 2013 12:42:42 -0000 X-Yahoo-Newman-Property: ymail-3 X-Yahoo-Newman-Id: 239295.23083.bm@omp1055.mail.ne1.yahoo.com Received: (qmail 16077 invoked by uid 60001); 31 Mar 2013 12:42:42 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1364733762; bh=+gRRIodxTu9U0bx7Y8YesfxVmfbI/h8A7b634aRkut0=; h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:Message-ID:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type; b=s1WoxAr4Drx/w5hFweAG8At5Fa9+tY63b9auD6qddiYOa+p0Lb41s1HqHoZoVAkSN62Jyr7/xtCiRQqASp3fTZ3aahZwC8Znb57+zBWHICWYv3D8z7/NIme8ln3ZPitbCtUlC8iJyAzuB8aoAOHNm97Et+ZyCd8Kriw28eak/YA= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:Message-ID:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type; b=dVYXCOlnUUTBtf74iGyGWhudreZW2ZJfG6LjBGViIkidr6LnLClhExkhMIJIVgKaYe3PgiL1298I+BwNUUkRomA0scrQKUTjd4KwhAHopIFSSP+EY7Iis2G9V4lNArL0Qb0kXUo2RfGSDfvLkyKgb5g7iqg9NsV0ufUVdxE8GEU=; X-YMail-OSG: HRkucTUVM1keDvWEUwsjvSgjM2EJvEg.BBv_MBtqAkoLyo_ iZ5sM52n2W60aYPsmZYLx_7R_geEQdI8TNIkDzgW.KhrTExBTndR5eIRgJ43 GWs4sm_hsAG1oUGXhcWrfPtBVmZUfbO3tm6nd8qUEbj7I6zcXTs4P7vBnpl4 HaEI3hdxiUZqLhkq8E5FVwV6.Y0Ngmp2g7C5YljfwVc5bKrd1dq5jJhhlkmX 2syIKscLNut_MdLzDgMhoAhYhzBOH9TlxoO4dMrnbKAL3Vw7HErUO5WmC1Ba qt7XvZ2J1jzH0Tl8.Zj9vwfcEwDpg2FRwb05GiUyx3iKERcnuhVO3dVKoFVn 9cvTXYDGTGOMaEBnqzB_QFqpz0X8C_ZAA1DHZz.J5PrA3J3Tz2yUgiLuk7su QLbJkXcr30wpEoaluh_mY63Xe16TU5jNiIdmgiFvXaWV7mU3LlDeVybWNJOr hpYznb0C0ikSdfjcNYbSiMikl2qK6ToCWD0ZYbSj2J1fwOaVlcWkJqhd30m7 StL6fpO3I.jPRrHOD15CbuKzYYS15Xg-- Received: from [98.203.118.124] by web121606.mail.ne1.yahoo.com via HTTP; Sun, 31 Mar 2013 05:42:39 PDT X-Rocket-MIMEInfo: 002.001, CgotLS0gT24gRnJpLCAzLzI5LzEzLCBBZHJpYW4gQ2hhZGQgPGFkcmlhbkBmcmVlYnNkLm9yZz4gd3JvdGU6Cgo.IEZyb206IEFkcmlhbiBDaGFkZCA8YWRyaWFuQGZyZWVic2Qub3JnPgo.IFN1YmplY3Q6IFJlOiBpZ2IgYW5kIEFMVFEgaW4gOS4xLXJjMwo.IFRvOiAiTmljayBSb2dlcnMiIDxuY3JvZ2Vyc0BnbWFpbC5jb20.Cj4gQ2M6ICJQaWVwZXIsIEplZmZyZXkgRSIgPGplZmZyZXkuZS5waWVwZXJAaW50ZWwuY29tPiwgImZyZWVic2QtbmV0QGZyZWVic2Qub3JnIiA8ZnJlZWJzZC1uZXRAZnJlZWJzZC4BMAEBAQE- X-Mailer: YahooMailClassic/15.1.7 YahooMailWebService/0.8.139.530 Message-ID: <1364733759.8517.YahooMailClassic@web121606.mail.ne1.yahoo.com> Date: Sun, 31 Mar 2013 05:42:39 -0700 (PDT) From: Barney Cordoba <barney_cordoba@yahoo.com> Subject: Re: igb and ALTQ in 9.1-rc3 To: Nick Rogers <ncrogers@gmail.com>, Adrian Chadd <adrian@freebsd.org> In-Reply-To: <CAJ-Vmok5uVG1hQh2FQc=8LMCJPmFgxy=SGzuA4mjivY9cnGUAw@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jeffrey EPieper <jeffrey.e.pieper@intel.com>, "freebsd-net@freebsd.org" <freebsd-net@freebsd.org>, "Clement Hermann \(nodens\)" <nodens2099@gmail.com>, Jack Vogel <jfvogel@gmail.com> X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Networking and TCP/IP with FreeBSD <freebsd-net.freebsd.org> List-Unsubscribe: <http://lists.freebsd.org/mailman/options/freebsd-net>, <mailto:freebsd-net-request@freebsd.org?subject=unsubscribe> List-Archive: <http://lists.freebsd.org/pipermail/freebsd-net>; List-Post: <mailto:freebsd-net@freebsd.org> List-Help: <mailto:freebsd-net-request@freebsd.org?subject=help> List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-net>, <mailto:freebsd-net-request@freebsd.org?subject=subscribe> X-List-Received-Date: Sun, 31 Mar 2013 12:42:48 -0000 --- On Fri, 3/29/13, Adrian Chadd <adrian@freebsd.org> wrote: > From: Adrian Chadd <adrian@freebsd.org> > Subject: Re: igb and ALTQ in 9.1-rc3 > To: "Nick Rogers" <ncrogers@gmail.com> > Cc: "Pieper, Jeffrey E" <jeffrey.e.pieper@intel.com>, "freebsd-net@freebsd.org" <freebsd-net@freebsd.org>, "Clement Hermann (nodens)" <nodens2099@gmail.com>, "Jack Vogel" <jfvogel@gmail.com> > Date: Friday, March 29, 2013, 1:10 PM > On 29 March 2013 10:04, Nick Rogers > <ncrogers@gmail.com> > wrote: > > > > Multiqueue or not, I would appreciate any help with > this thread's > > original issue. Whether or not its the ideal thing to > do, I cannot > > simply just replace the NICs with an em(4) variant, as > I have hundreds > > of customers/systems already in production running 8.3 > and relying on > > the igb driver + ALTQ. I need to be able to upgrade > these systems to > > 9.1 without making hardware changes. > > > > > If it's that critical, have you thought about contracting > out that task to > a developer? You have 100s of systems/customers using 1990s-class traffic shaping and you have no programmer on staff with the skills to patch and test an ethernet driver? the igb driver has always sucked rocks, why did you use them in the first place. Or did they just happen to be on the MB you use? BC
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1364733314.80531.YahooMailClassic>