From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [net-next 02/11] igb: Use node specific allocations for the q_vectors and rings Date: Mon, 10 Oct 2011 18:32:28 +0200 Message-ID: <20111010163228.GA14482@one.firstfloor.org> References: <1318056461-19562-1-git-send-email-jeffrey.t.kirsher@intel.com> <1318056461-19562-3-git-send-email-jeffrey.t.kirsher@intel.com> <4E931C61.7040204@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Andi Kleen , Jeff Kirsher , davem@davemloft.net, netdev@vger.kernel.org, gospo@redhat.com, sassmann@redhat.com To: Alexander Duyck Return-path: Received: from one.firstfloor.org ([213.235.205.2]:34320 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750912Ab1JJQcc (ORCPT ); Mon, 10 Oct 2011 12:32:32 -0400 Content-Disposition: inline In-Reply-To: <4E931C61.7040204@intel.com> Sender: netdev-owner@vger.kernel.org List-ID: > The RR configuration is somewhat arbitrary. However it is still better > than dumping everyting on a single node, and it works with the > configuration when the rings numbers line up with the CPU numbers since > normally the CPUs are RR on the nodes. From what I have seen it does > work quite well and it prevents almost all cross-node memory accesses > when running a routing workload. Ok so it's optimized for one specific workload. I'm sure you'll find some other workload where it doesn't work out. I suppose it's hard to get right in the general case, but best would be if ethtool had a nice and easy interface to set it at least. However one disadvantage of that patch over the existing state of the art (numactl modprobe ...) is that there's no way to override the placement now. So if you do the forced RR I think you need the ethtool part too, or at least some parameter to turn it off. -Andi