From: Herbert Xu Subject: Re: [PATCH 2/3] xfrm: Traffic Flow Confidentiality for IPv4 ESP Date: Wed, 8 Dec 2010 17:24:32 +0800 Message-ID: <20101208092432.GA15610@gondor.apana.org.au> References: <1291717744-30111-1-git-send-email-martin@strongswan.org> <1291717744-30111-3-git-send-email-martin@strongswan.org> <20101208084954.GA15252@gondor.apana.org.au> <1291800041.2005.25.camel@martin> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-crypto@vger.kernel.org, netdev@vger.kernel.org To: Martin Willi Return-path: Received: from helcar.apana.org.au ([209.40.204.226]:60942 "EHLO fornost.hengli.com.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752133Ab0LHJYe (ORCPT ); Wed, 8 Dec 2010 04:24:34 -0500 Content-Disposition: inline In-Reply-To: <1291800041.2005.25.camel@martin> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Wed, Dec 08, 2010 at 10:20:41AM +0100, Martin Willi wrote: > > > In particular, why would we need a boundary at all? Setting it to > > anything other than the PMTU would seem to defeat the purpose of > > TFC for packets between the boundary and the PMTU. > > I don't agree, this highly depends on the traffic on the SA. For a > general purpose tunnel with TCP flows, PMTU padding is fine. But if > there are only small packets (maybe SIP+RTP), padding to the PMTU is > very expensive. > > The administrator setting up the SAs probably knows (or even controls > directly) what traffic it is used for, and might lower the boundary > accordingly. OK, that's a good reason. But you should probably get rid of that unused flag field in the user-interface and just provide a pad length. Thanks, -- Email: Herbert Xu Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt