From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753278Ab0IFKnH (ORCPT ); Mon, 6 Sep 2010 06:43:07 -0400 Received: from cantor2.suse.de ([195.135.220.15]:46235 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750885Ab0IFKnG (ORCPT ); Mon, 6 Sep 2010 06:43:06 -0400 From: Thomas Renninger Organization: SUSE Products GmbH To: Jean Pihet , discuss@lesswatts.org, linux-pm@lists.linux-foundation.org Subject: Re: [PATCH] tracing, perf: add more power related events Date: Mon, 6 Sep 2010 12:42:59 +0200 User-Agent: KMail/1.13.3 (Linux/2.6.31.5-0.1-desktop; KDE/4.4.3; x86_64; ; ) Cc: Ingo Molnar , Arjan van de Ven , Peter Zijlstra , Len Brown , arjan@infradead.org, Kevin Hilman , linux-kernel@vger.kernel.org References: <20100904081801.GA19876@elte.hu> In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201009061243.00479.trenn@suse.de> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Monday 06 September 2010 10:56:50 Jean Pihet wrote: > Hi, > > Thanks for the feedback! > > > On Friday 03 September 2010 21:55:55 Ingo Molnar wrote: > >Well, the patch that got posted: > > > > [PATCH] tracing, perf: add more power related events > > > >Only adds tracepoints to power.h, but doesnt actually use them anywhere > >... > > > >Is there a companion patch i missed? > I will re-start a thread with the full patch and with both lkml and > linux-omap in Cc. discuss@lesswatts.org and linux-pm@lists.linux-foundation.org are also lists where people could be interested in such stuff. Could you explain a bit what these ARM specific power domains are then, please. One idea I had after seeing these: Would it make sense to additionally add one or several device power perf event(s)? bluetooth, wireless, usb, cameras, whatever, could register it by name, like you've done it with the power domains. E.g. I had a problem with 2.6.31 where bluetooth was powered up/activated at boot time, while it got activated ondemand before. Cost me some Watts and quite some time to find that out. Showing perf power events when bluetooth gets activated and put into lower power states again (and the same for other devices), sounds like a good idea? Thomas