From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754746AbYJBOdf (ORCPT ); Thu, 2 Oct 2008 10:33:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753455AbYJBOd2 (ORCPT ); Thu, 2 Oct 2008 10:33:28 -0400 Received: from relay2.sgi.com ([192.48.171.30]:52605 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753441AbYJBOd1 (ORCPT ); Thu, 2 Oct 2008 10:33:27 -0400 Message-ID: <48E4DBC8.3070202@sgi.com> Date: Thu, 02 Oct 2008 07:33:44 -0700 From: Mike Travis User-Agent: Thunderbird 2.0.0.6 (X11/20070801) MIME-Version: 1.0 To: Pavel Machek CC: Ingo Molnar , Andrew Morton , rpurdie@rpsys.net, Jack Steiner , linux-kernel@vger.kernel.org, Thomas Gleixner , "H. Peter Anvin" Subject: Re: [PATCH 1/1] SGI X86 UV: Provide a System Activity Indicator driver References: <20080919143744.009764000@polaris-admin.engr.sgi.com> <20080919143744.190005000@polaris-admin.engr.sgi.com> <20080922114820.GA13990@elte.hu> <48D7AEE6.7080804@sgi.com> <20080928194201.GA1310@ucw.cz> <48E3BE2D.4020606@sgi.com> <20081002083700.GB19993@atrey.karlin.mff.cuni.cz> In-Reply-To: <20081002083700.GB19993@atrey.karlin.mff.cuni.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: >> Pavel Machek wrote: >>>> Another relevant point is that I will be adding a bit more functionality >>>> to disable the timer interrupt on truly "idle" cpus (like have been idle >>>> for some amount of seconds). We would then use the "exit from idle" >>>> callback to reestablish the timer interrupt. [This would allow them to >>>> enter power down states if appropriate.] >>> Should you look at nohz instead of reinventing it? >> Thanks, I did look at it. Quite complex. Maybe I'm missing something >> but I don't see how it fits in? Are you saying I should be using > > I seen your remark above about disabling timer interrupt on idle > cpus. That's exactly nohz functionality, right? Maybe I misunderstood > you and you meant "my monitoring interrupt" and not "generic system > timer interrupt"? > > Pavel Ahh, yes, now I see... ;-) And thanks for the pointer, I hadn't looked at this before. In regards to "powering down" a UV system, it's not clear yet that it will help much unless we can power down a whole lot more of the system than just the cpus... ;-) [but in reality, reducing power use any time you can is real necessity.] Btw, are you ok with the remainder of the patch? Thanks, Mike