From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mikulas Patocka Subject: [PATCH 0/4] Integrate dm-latency functionality to dm-statistics Date: Tue, 9 Jun 2015 17:20:46 -0400 (EDT) Message-ID: Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: 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 G. Kergon" , Mike Snitzer , Coly Li Cc: Tao Ma , dm-devel@redhat.com, Laurence Oberman , Robin Dong List-Id: dm-devel.ids Hi Here I'm sending patches that add funcionality provided by dm-latency into dm-statistics framework. Dm-latency was introduced in this post https://www.redhat.com/archives/dm-devel/2015-February/msg00158.html , however, there is already similar framework dm-statistics in the kernel. In order to have cleaner and easier to maintain code, it is better to integrate dm-latency functionality into dm-statistics, rather than having two frameworks that do similar things. This patch series makes these changes: * it is possible to use precise time measurement with nanosecond granularity (previously, time was measured in jiffies) * it is possible to collect histogram of latencies. The histogram boundaries may be selected by the user and there is unlimited number of possible boundaries (in dm-latency, the histogram boundaries were fixed) * it is possible to use dm-statistics on multipath The documentation of these extensions is in Documentation/device-mapper/statistics.txt I'd like to ask people who designed dm-latency to check this patchset, try to use it and tell us if it provides sufficient functionality to them (the indended functinality of dm-latency was to predict hard disk failures from i/o request latency histogram). Mikulas