From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: [PATCH 0/2] range timer support Date: Tue, 28 Oct 2008 10:54:18 +0000 Message-ID: References: <49582C73AC36CC4C8C6C42390304E81C092F97F902@pdsmsx502.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <49582C73AC36CC4C8C6C42390304E81C092F97F902@pdsmsx502.ccr.corp.intel.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: "Yu, Ke" , "xen-devel@lists.xensource.com" Cc: "Liu, Jinsong" , "Tian, Kevin" , "Wei, Gang" List-Id: xen-devel@lists.xenproject.org On 28/10/08 06:57, "Yu, Ke" wrote: > The major changes are: > - rebase with latest upstream: sync with the timer link list chagne (changeset > 18381), sync with the heap memory allocation change (changeset 18671) > - Per Keir's comment, more usage is found: use the range timer in HVM virtual > periodic timer (xen/arch/x86/hvm/vpt.c). Since vpt has the logic to handle > missing ticks, it is an ideal place to use range timer. > - the original deferrable timer is renamed to range timer, to better describe > the conecpt. This renaming is inspired by a similar proposal in linux kernel > community (http://kerneltrap.org/mailarchive/linux-kernel/2008/9/1/3160404) > > A preliminary measurement is done in a dual core mobile platform (Montivina). > When creating 4 RHEL5 HVM UP guests, the range timer patch can reduce ~5% > power consumption. More measurement is onging. This post is inteneded to get > more feedback to refine it. You'll have to explain how the implementation in timer.c works. I'm not sure I believe it really does. -- Keir