From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: [RFC PATCH V2] New Xen netback implementation Date: Mon, 30 Jan 2012 10:21:09 -0500 Message-ID: <20120130152109.GA17923@phenom.dumpdata.com> References: <1326808024-3744-1-git-send-email-wei.liu2@citrix.com> <20120127192214.GA14437@phenom.dumpdata.com> <1327936037.26983.241.camel@zakaz.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "Wei Liu (Intern)" , "netdev@vger.kernel.org" , "xen-devel@lists.xensource.com" , David Vrabel , Paul Durrant To: Ian Campbell Return-path: Received: from acsinet15.oracle.com ([141.146.126.227]:22769 "EHLO acsinet15.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751742Ab2A3PXp (ORCPT ); Mon, 30 Jan 2012 10:23:45 -0500 Content-Disposition: inline In-Reply-To: <1327936037.26983.241.camel@zakaz.uk.xensource.com> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, Jan 30, 2012 at 03:07:16PM +0000, Ian Campbell wrote: > On Fri, 2012-01-27 at 19:22 +0000, Konrad Rzeszutek Wilk wrote: > > > > - This patch set solves the CPU banding problem I've seen with the > > older netback. The older one I could see X netback threads eating > > 80% > > of CPU. With this one, the number is down to 13-14%. > > "CPU banding problem"? > > If you had X threads using 80% before do you now see Y threads using > 13-14% where Y is bigger or smaller than X? Is 80*X ~= 13*Y? Yes. ~=. The count looked to be the same. > > Ian.