From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Jan Beulich" <jbeulich@suse.com>
Subject: Re: [PATCH 4/4] xen: use idle vcpus to scrub pages
Date: Wed, 18 Jun 2014 11:42:25 +0100
Message-ID: <53A17B2102000078000B6994@mail.emea.novell.com>
References: <1403005783-30746-1-git-send-email-bob.liu@oracle.com>
	<1403005783-30746-4-git-send-email-bob.liu@oracle.com>
	<53A0583B020000780001B0E0@mail.emea.novell.com>
	<53A0E8CE.7050304@oracle.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xen.org>
Received: from mail6.bemta5.messagelabs.com ([195.245.231.135])
	by lists.xen.org with esmtp (Exim 4.72)
	(envelope-from <jbeulich@suse.com>) id 1WxDK1-0007hK-2n
	for xen-devel@lists.xenproject.org; Wed, 18 Jun 2014 10:42:29 +0000
In-Reply-To: <53A0E8CE.7050304@oracle.com>
Content-Disposition: inline
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: bob.liu@oracle.com
Cc: lliubbo@gmail.com, keir@xen.org, ian.campbell@citrix.com, George.Dunlap@eu.citrix.com, andrew.cooper3@citrix.com, xen-devel@lists.xenproject.org
List-Id: xen-devel@lists.xenproject.org

>>> Bob Liu <bob.liu@oracle.com> 06/18/14 3:18 AM >>>
>On 06/17/2014 09:01 PM, Jan Beulich wrote:
>>>>> On 17.06.14 at 13:49, <lliubbo@gmail.com> wrote:
>>> --- a/xen/common/domain.c
>>> +++ b/xen/common/domain.c
>>> @@ -587,12 +587,14 @@ int domain_kill(struct domain *d)
>>>          d->tmem_client = NULL;
>>>          /* fallthrough */
>>>      case DOMDYING_dying:
>>> +        enable_idle_scrub = 0;
>>>          rc = domain_relinquish_resources(d);
>>>          if ( rc != 0 )
>>>          {
>>>              BUG_ON(rc != -EAGAIN);
>>>              break;
>>>          }
>>> +        enable_idle_scrub = 1;
>>>          for_each_vcpu ( d, v )
>>>              unmap_vcpu_info(v);
>>>          d->is_dying = DOMDYING_dead;
>> 
>> ???
>
>enable_idle_scrub is a rough way to disable idle vcpu scrubbing, so that
>domain_relinquish_resources() can get "&heap_lock" faster and make xl
>destroy return quickly.

If such a hack is really necessary, it requires a decent code comment. But I
don't think it's needed, as long as you keep down the lock holding time on the
scrubber side.

Jan