From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wanpeng Li Subject: Re: [question] Is there a plan to introduce a unified co-scheduling mechanism to CFS ? Date: Fri, 10 Oct 2014 19:44:34 +0800 Message-ID: <5437C6A2.3020908@gmail.com> References: <5437C505.4010504@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE To: Zhang Haoyu , kvm , "qemu-devel@nongnu.org Developers" Return-path: Received: from mail-pa0-f48.google.com ([209.85.220.48]:49890 "EHLO mail-pa0-f48.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751268AbaJJLoj (ORCPT ); Fri, 10 Oct 2014 07:44:39 -0400 Received: by mail-pa0-f48.google.com with SMTP id eu11so1612317pac.35 for ; Fri, 10 Oct 2014 04:44:39 -0700 (PDT) In-Reply-To: <5437C505.4010504@gmail.com> Sender: kvm-owner@vger.kernel.org List-ID: =E4=BA=8E 10/10/14, 7:37 PM, Zhang Haoyu =E5=86=99=E9=81=93: > Hi, > > Is it worthy to introduce a unified co-scheduling mechanism to CFS ? > Because multiple cooperating threads or tasks frequently synchronize=20 > with each other, > not executing them concurrently would only increase the latency of=20 > synchronization. > For example, a thread blocking in spinlock to waiting for another=20 > thread to release the same spinlock > might reduce its waiting time by being executed concurrently with the= =20 > thread which hold the same spinlock. > In virtualization scenario, multiple vcpus (which belong to the same=20 > vm) co-scheduling is more desired > when several cooperating threads/task is running in guest. > > Is there a plane for this work? Please refer to gang scheduler. Regards, Wanpeng Li > > Thanks, > Zhang Haoyu > --=20 > To unsubscribe from this list: send the line "unsubscribe kvm" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html