From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Dillow Subject: Re: [PATCH RFC] r8169: straighten out overlength frame detection Date: Mon, 28 Dec 2009 19:24:51 -0500 Message-ID: <1262046291.8998.6.camel@obelisk.thedillows.org> References: <20091228194834.GA18422@hmsreliant.think-freely.org> <20091228195053.GB18422@hmsreliant.think-freely.org> <20091228213114.GA24285@zoreil.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Neil Horman , netdev@vger.kernel.org, davem@davemloft.net, eric.dumazet@gmail.com, nhorman@redhat.com To: =?ISO-8859-1?Q?Fran=E7ois?= romieu Return-path: Received: from smtp.knology.net ([24.214.63.101]:50255 "EHLO smtp.knology.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752078AbZL2AY6 (ORCPT ); Mon, 28 Dec 2009 19:24:58 -0500 In-Reply-To: <20091228213114.GA24285@zoreil.com> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, 2009-12-28 at 22:31 +0100, Fran=C3=A7ois romieu wrote: > (I'm back) >=20 > The Mon, Dec 28, 2009 at 02:50:53PM -0500, Neil Horman wrote : > [...] > I doubt that we will be able to allocate that much memory reliably fo= r long. >=20 > I'd rather go for static buffers + copy (+ src mac address of our new= friend). The driver doesn't support fragmented receives, but it seems like the H= W does -- is this known to work on the different revisions? Or alternatively, is it known to be broken on any of them? It seems like i= t would be preferable to implement this and use RxMaxSize to tell the NIC how big the allocated buffer fragments are. Or I'm misreading the capabilities of the NIC? In any event, I'm probably not going to have time to write/test it on the HW I have anytime soon, so perhaps the static buffers/copy is the safest option for now, albeit non-optimal. Dave