From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ie0-x22f.google.com (mail-ie0-x22f.google.com [IPv6:2607:f8b0:4001:c03::22f]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id B14CC202102 for ; Tue, 28 May 2013 06:18:11 -0700 (PDT) Received: by mail-ie0-f175.google.com with SMTP id tp5so4564170ieb.34 for ; Tue, 28 May 2013 06:18:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=LVMo4y4txeTQn49Evw+udrJXUXo9b7oypAxfksdjJEI=; b=vWEHFshW39J+eCxoE/z8WCZw4IlhItV9OZYivnUowP/XqyDf8aoHdCAfDWzqAJY+Ur /7lHqEIKuEeOZu7T7u9/m8VA6/t06JWZ8WxitxzZTunsuNPIV/NQSnLQsmAcXpdkSos1 ruFFMHgrP/XMhAZ2PQu4pysOG+WJNrMZjZApIkE7Pf+hD/LwfDxB8UDQVp6FTko9J1h0 VfYDarypBQJRPiIfxsJup56Cj4/3ECHxjOHv0L8U8EY7nOANZfX1oRo0UlAdepwCq/Og swDsgv+IQPFwiWJNkYPkjHbvLODw9EeoC8NKajohDFmq9fWnaGTcGzxigqKEkDvzC2+a JRew== MIME-Version: 1.0 X-Received: by 10.50.141.161 with SMTP id rp1mr6522828igb.11.1369747090812; Tue, 28 May 2013 06:18:10 -0700 (PDT) Received: by 10.64.35.44 with HTTP; Tue, 28 May 2013 06:18:10 -0700 (PDT) Date: Tue, 28 May 2013 06:18:10 -0700 Message-ID: From: Dave Taht To: cerowrt-devel@lists.bufferbloat.net Content-Type: multipart/alternative; boundary=089e013cbdcac7cd7204ddc7186b Subject: [Cerowrt-devel] a bit of kernel hacking for ipv6 encapsulation on skb_flow_dissector? X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 28 May 2013 13:18:11 -0000 --089e013cbdcac7cd7204ddc7186b Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I have been fiddling with WAY too many things simultaneously. At the moment, for example, I'm poking into why transmission doesn't successfully set the tos field on utp packets while waiting for a profile enabled build for the tplink while doing a bunch of other things. ... anyway... While doing that I saw that fq_codel's treatment of IPv6 through a 6in4 tunnel didn't appear to be doing the "right thing", that ipv4 was outcompeting it left and right. It looks like the skb_flow_dissector doesn't "do the right thing" on proto 41 (6in4), where it should peel off the ipv4 header and look inside it for the ipv6 5 tuple. It does do the right thing for GRE... anybody want to take a stab at fixing net/core/flow_dissector.c ? --=20 Dave T=E4ht Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html --089e013cbdcac7cd7204ddc7186b Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I have been fiddling with WAY too many things simultaneously. At the moment= , for example, I'm poking into why transmission doesn't successfull= y set the tos field on utp packets while waiting for a profile enabled buil= d for the tplink while doing a bunch of other things.

... anyway...

While doing that I saw that fq_codel's treatm= ent of IPv6 through a 6in4 tunnel didn't appear to be doing the "r= ight thing", that ipv4 was outcompeting it left and right.

It looks like the skb_flow_dissector doesn't "do the right thing&q= uot; on proto 41 (6in4), where it should peel off the ipv4 header and look = inside it for the ipv6 5 tuple. It does do the right thing for GRE... anybo= dy want to take a stab at fixing
net/core/flow_dissector.c ?
--
Dave T=E4ht

Fixing bufferbloat= with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html=20 --089e013cbdcac7cd7204ddc7186b--