From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kiyoshi Ueda Subject: Re: Re: [PATCH 0/0] dm-mpath: queue-length oriented dynamic load balancer Date: Fri, 30 Jan 2009 17:10:38 +0900 Message-ID: <4982B5FE.5040807@ct.jp.nec.com> References: <498153BE.60104@ct.jp.nec.com> <20090129140335.GA27390@agk-dp.fab.redhat.com> Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20090129140335.GA27390@agk-dp.fab.redhat.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: Alasdair Kergon Cc: dm-devel@redhat.com, stefan.bader@canonical.com List-Id: dm-devel.ids Hi Alasdair, Sorry for confusing you. On 01/29/2009 11:03 PM +0900, Alasdair G Kergon wrote: > On Thu, Jan 29, 2009 at 03:59:10PM +0900, Kiyoshi Ueda wrote: >> This patch-set can be applied on top of 2.6.29-rc2. > > It's not clear: are you reordering the dependencies now? > > Previously there were three sets of patches taken in this order: > 1 change to request-based > 2 add queue length ps > 3 add service time ps > i.e. 2 and 3 depend on 1 > > Are you now changing that so that 2 and 3 could go upstream before 1? I had sent those three sets in the order above since 1 enhances 2 and 3. But 2 and 3 don't functionally depend on 1 (although 1 and 3 have a few cosmetic dependencies), so we can reorder that as we want. 2 and 3 are simpler patches than 1, and I found some problems in 1, so I think reordering might be a good idea if you are OK. As for 3, I still want to get feedback from others as it'll involve discussion about the optimal interface for load balancer. Thanks, Kiyoshi Ueda