From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757785Ab3BXM7Z (ORCPT ); Sun, 24 Feb 2013 07:59:25 -0500 Received: from cantor2.suse.de ([195.135.220.15]:38672 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756223Ab3BXM7Y (ORCPT ); Sun, 24 Feb 2013 07:59:24 -0500 Date: Sun, 24 Feb 2013 13:59:17 +0100 From: Borislav Petkov To: Peter Feuerer Cc: LKML , Zhang Rui , Durgadoss R , Borislav Petkov , Andreas Mohr , Alexander Lam Subject: Re: [PATCH] acerhdf: Fix fan activation with new thermal governor Message-ID: <20130224125917.GD19609@pd.tnic> Mail-Followup-To: Borislav Petkov , Peter Feuerer , LKML , Zhang Rui , Durgadoss R , Andreas Mohr , Alexander Lam References: <20130222111521.GA25473@pd.tnic> <1361647210-12983-1-git-send-email-bp@alien8.de> <20130224112851.GA19609@pd.tnic> <20130224120952.GB19609@pd.tnic> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20130224120952.GB19609@pd.tnic> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Feb 24, 2013 at 01:09:52PM +0100, Borislav Petkov wrote: > On Sun, Feb 24, 2013 at 12:42:55PM +0100, Peter Feuerer wrote: > > Hi Boris, > > > > thanks for your best wishes in the last mail, I'm feeling little better now. > > Nice :) > > > Please test my last patch with the 4 trip points ;) - even if you > > don't really like it, it is working great! - And to be honest, I still > > prefer this solution! > > Right, but 4 trip points for this simple driver is a bit of a overkill, > don't you think? If we want to be really accurate, we'd only need two, > think of three temperature intervals here: > > 0 - temp <= fanon > 1 - fanon =< temp < crit > 2 - temp >= crit > > I need to go stare at it a bit more. Oh, and then there's thermal_cdev_update which calls down ->set_cur_state, i.e. acerhdf_set_cur_state() handing us down a state which we're supposed to take into consideration but we completely ignore and no our own thing. And, AFAICT, when this thing gets called, we need to set the fan to either AUTO or OFF, based on the state it's giving us from above, no? Because with acerhdf_get_max_state() we tell it that we have two states, 0 and 1. Hmm. -- Regards/Gruss, Boris. Sent from a fat crate under my desk. Formatting is fine. --