From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hannes Frederic Sowa Subject: Re: [PATCH RFC] ipv6: use stronger hash for reassembly queue hash table Date: Fri, 8 Mar 2013 17:18:16 +0100 Message-ID: <20130308161816.GF28531@order.stressinduktion.org> References: <20130307214211.GP7941@order.stressinduktion.org> <20130308055718.GA28531@order.stressinduktion.org> <20130308130433.GB28531@order.stressinduktion.org> <1362754386.15793.226.camel@edumazet-glaptop> <20130308150831.GD28531@order.stressinduktion.org> <1362756219.15793.240.camel@edumazet-glaptop> <20130308155404.GE28531@order.stressinduktion.org> <1362759310.15793.244.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Cc: netdev@vger.kernel.org, yoshfuji@linux-ipv6.org To: Eric Dumazet Return-path: Received: from order.stressinduktion.org ([87.106.68.36]:51605 "EHLO order.stressinduktion.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933631Ab3CHQSS (ORCPT ); Fri, 8 Mar 2013 11:18:18 -0500 Content-Disposition: inline In-Reply-To: <1362759310.15793.244.camel@edumazet-glaptop> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, Mar 08, 2013 at 08:15:10AM -0800, Eric Dumazet wrote: > On Fri, 2013-03-08 at 16:54 +0100, Hannes Frederic Sowa wrote: > > > Btw. did s.o. have a look at skb->rxhash? I just started looking after > > its many users but perhaps someone did this job already. > > You mean for IPv6 ? Yes, because of the usage of ipv6_addr_hash in skb_flow_dissect. > Most of us use the rxhash provided by the NIC, our software fallback is > a best effort as well for RPS or some qdisc. > > If some hacker specifically cook packets hashing to same rxhash, thats > fine. Ok, good to know. Thanks!