From mboxrd@z Thu Jan 1 00:00:00 1970 From: George Dunlap Subject: Re: [PATCH] SEDF: avoid gathering vCPU-s on pCPU0 Date: Mon, 4 Mar 2013 12:11:34 +0000 Message-ID: <51348F76.3090204@eu.citrix.com> References: <5130D8D002000078000C26C6@nat28.tlf.novell.com> <5134959502000078000C2C60@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <5134959502000078000C2C60@nat28.tlf.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: xen-devel List-Id: xen-devel@lists.xenproject.org On 04/03/13 11:37, Jan Beulich wrote: > While you might argue that affinity changes ought to be done > in a "sensible" way, I think you still realize that there is behavior > here that one wouldn't be able to control without explicitly > adding an intermediate step when doing this for DomU-s. And > you should also keep in mind that there are certain things that > Dom0 needs to change its vCPU affinities for (pv-ops doesn't > do that, which is while certain things there just can't work); > this is how I noticed in the first place that the second half of > the fix is necessary. Oh right -- so the problem is that when you *expand* the mask, you want the scheduler to take another look at where might be a good place to run the vcpu given the changed constraints. It's probably not a bad idea to just extend that idea to all the schedulers really. It's not like we expect people to be changing the affinity masks hundreds of times a second. -George