From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757436AbYFBGJw (ORCPT ); Mon, 2 Jun 2008 02:09:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751611AbYFBGJn (ORCPT ); Mon, 2 Jun 2008 02:09:43 -0400 Received: from E23SMTP04.au.ibm.com ([202.81.18.173]:47662 "EHLO e23smtp04.au.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751110AbYFBGJn (ORCPT ); Mon, 2 Jun 2008 02:09:43 -0400 Message-ID: <48438E54.1000404@linux.vnet.ibm.com> Date: Mon, 02 Jun 2008 11:38:20 +0530 From: Balbir Singh Reply-To: balbir@linux.vnet.ibm.com Organization: IBM User-Agent: Thunderbird 2.0.0.14 (X11/20080501) MIME-Version: 1.0 To: Pavel Machek CC: Arjan van de Ven , Vaidyanathan Srinivasan , Linux Kernel , venkatesh.pallipadi@intel.com, suresh.b.siddha@intel.com, Michael Neuling , "Amit K. Arora" Subject: Re: [RFC PATCH v1 0/3] Scaled statistics using APERF/MPERF in x86 References: <20080526142513.24680.97164.stgit@drishya.in.ibm.com> <20080526085000.33787eac@infradead.org> <20080531211306.GM5405@ucw.cz> In-Reply-To: <20080531211306.GM5405@ucw.cz> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Pavel Machek wrote: > Hi! > >>> entitlement for the process as per the current CPU frequency. This >>> technique is used in powerpc architecture with the help of hardware >>> registers that accurately capture the entitlement. >>> >> there are some issues with this unfortunately, and these make it >> a very complex thing to do. >> Just to mention a few: >> 1) What if the BIOS no longer allows us to go to the max frequency for >> a period (for example as a result of overheating); with the approach >> above, the admin would THINK he can go faster, but he cannot in reality, >> so there's misleading information (the system looks half busy, while in > > Plus time one-second-computation-job returning anything else is just > wrong. Even when it only happens when overheated... or only on battery > power. > > If you want scaled utime, you need new interface. We do have a new interface, two new parameters per-task utimescaled and stimescaled. They already exist in task_struct. Ditto for the delay accounting pieces. Did I misunderstand your need for a new interface? We still keep utime and stime around. -- Warm Regards, Balbir Singh Linux Technology Center IBM, ISTL