From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arvid Brodin Subject: Re: [PATCH v4] net/hsr: Add support for the High-availability Seamless Redundancy protocol (HSRv0) Date: Sat, 21 Sep 2013 03:03:51 +0200 Message-ID: <523CF077.8010607@xdin.com> References: <523A4F5E.9070503@xdin.com> <20130920.151017.2125552741552759738.davem@davemloft.net> <523CEBCD.10808@xdin.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Arvid Brodin , , , , , , To: David Miller Return-path: Received: from spam1.webland.se ([91.207.112.90]:63134 "EHLO spam1.webland.se" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751702Ab3IUBDy (ORCPT ); Fri, 20 Sep 2013 21:03:54 -0400 In-Reply-To: <523CEBCD.10808@xdin.com> Sender: netdev-owner@vger.kernel.org List-ID: On 2013-09-21 02:43, Arvid Brodin wrote: > It's weird though, the "len" bug doesn't show up in my practical test= s -=20 > scp of 16 MiB file (sha1sum compared afterwards) and ping with defaul= t=20 > and 65 KB packets work just fine before and after this change - even > between one fixed system and one with the bug remaining. Could it be > that none of the network device drivers we use (e1000e and macb) make= =20 > use of nonlinear skbs? =46orget about this, e1000e is running on an architecture with=20 HAVE_EFFICIENT_UNALIGNED_ACCESS, and on macb it does affect things -=20 with the "fix" nothing works. :-|=20 --=20 Arvid Brodin | Consultant (Linux) XDIN AB | Knarrarn=E4sgatan 7 | SE-164 40 Kista | Sweden | xdin.com