From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [ kvm-Bugs-1802227 ] network speed Date: Fri, 28 Sep 2007 18:59:50 +0200 Message-ID: <46FD3306.7070007@qumranet.com> References: <10EA09EFD8728347A513008B6B0DA77A022AEA3C@pdsmsx411.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org To: "Dong, Eddie" Return-path: In-Reply-To: <10EA09EFD8728347A513008B6B0DA77A022AEA3C-wq7ZOvIWXbNpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Errors-To: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: kvm.vger.kernel.org Dong, Eddie wrote: > kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org wrote: > >> Bugs item #1802227, was opened at 2007-09-25 23:07 >> Message generated for change (Tracker Item Submitted) made by Item >> Submitter You can respond by visiting: >> https://sourceforge.net/tracker/?func=detail&atid=893831&aid=18 >> 02227&group_id=180599 >> >> Please note that this message will contain a full copy of the comment >> thread, including the initial issue submission, for this request, >> not just the latest update. >> Category: None >> Group: None >> Status: Open >> Resolution: None >> Priority: 5 >> Private: No >> Submitted By: xeb (xebd) >> Assigned to: Nobody/Anonymous (nobody) >> Summary: network speed >> >> Initial Comment: >> Hello. >> >> Host:Linux 2.6.22-gentoo-r2 #2 SMP Fri Aug 3 07:01:46 MSD 2007 x86_64 >> AMD Athlon(tm) 64 X2 Dual Core Processor 5600+ AuthenticAMD >> GNU/Linux,Gentoo, kvm-44 >> >> Guest:Linux 2.6.22-hardened-r4 >> >> I have measured network speed host<->guest and it is 200-400 KB/s. >> Is so low speed proper for kvm ? >> >> >> > > I did a try on my box, but I get consistant scp performance of 1.5MB/s > w/ or w/o -no-kvm-irqchip. My platform is old, but host is 2.6.23 from > kvm.git some time ago. > In the meantime, we do see issues in some configuration, while > didn't find the common point yet. > > Is this due to host linux version difference? Stranger ... Any new > finding? > hmm, if the host doesn't have hrtimers you can get better timing in userspace (using rtc) that in the kernel. However better timing shouldn't lead to better networking performance. -- Do not meddle in the internals of kernels, for they are subtle and quick to panic. ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/