From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: [PATCH] only set scheduler timer for non-idle CPU Date: Thu, 02 Apr 2009 14:18:41 +0100 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Keir Fraser , Thomas Pfeuffer , "Yu, Ke" , "xen-devel@lists.xensource.com" List-Id: xen-devel@lists.xenproject.org On 02/04/2009 14:06, "Keir Fraser" wrote: >> I think your patch is not good in case sedf-scheduler is used. If idle V= CPU >> is the current "running" VCPU, the scheduler timer is set to the next "p= eriod >> begin" of the first VCPU in the wait queue. >> Your patch prevents sedf from taking the VCPUs waiting for their next pe= riod >> into the runnable queue again. >=20 > It=B9s probably cleaner always to respect the specific scheduler=B9s scheduli= ng > quantum in schedule.c, but then have sched_credit.c return a very large > quantum (large as possible) for the idle VCPU. I checked in something to this effect as c/s 19500. -- Keir