From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hongkaixing Subject: Re: [PATCH] xenpaging:add a new array to speed up page-in in xenpaging Date: Sat, 07 Jan 2012 16:55:49 +0800 Message-ID: <002c01cccd1a$279c9a00$76d5ce00$@com> References: <052727b8165ce6e05002.1325732917@h00166998.china.huawei.com> <20120105205910.GA29311@aepfle.de> <20120106130737.GB1694@aepfle.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-reply-to: <20120106130737.GB1694@aepfle.de> Content-language: zh-cn List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: 'Olaf Hering' Cc: bicky.shi@huawei.com, xiaowei.yang@huawei.com, xen-devel@lists.xensource.com, yanqiangjun@huawei.com, hanweidong@huawei.com List-Id: xen-devel@lists.xenproject.org > -----Original Message----- > From: Olaf Hering [mailto:olaf@aepfle.de] > Sent: Friday, January 06, 2012 9:08 PM > To: hongkaixing@huawei.com > Cc: xen-devel@lists.xensource.com > Subject: Re: [PATCH] xenpaging:add a new array to speed up page-in in xenpaging > > On Thu, Jan 05, Olaf Hering wrote: > > > Is the page-out time for 512M really that fast? For me page-out is still > > really slow even when the pagefile is in tmpfs. It takes several > > minutes, I get around 4MB/s. page-in is around 20MB/s. > > Wait, the slow page-out is due to the poll() timeout in > xenpaging_wait_for_event_or_timeout(). This can be fixed with some sort > of dynamic timeout while there is still something to page-out. I don't think so. Our approach still use 100ms in poll(), Using 10ms seems no significant change. In my opinion, In order to improve speed , 2 points are necessary: 1. put page-in request fast 2. request should be in disk order. Especially for a large swap file. > > Olaf