From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756453AbZBJV3R (ORCPT ); Tue, 10 Feb 2009 16:29:17 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755363AbZBJV3B (ORCPT ); Tue, 10 Feb 2009 16:29:01 -0500 Received: from mail-ew0-f21.google.com ([209.85.219.21]:54430 "EHLO mail-ew0-f21.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755457AbZBJV3A convert rfc822-to-8bit (ORCPT ); Tue, 10 Feb 2009 16:29:00 -0500 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=u2mCtLKf0q3jzVp4ZLmcjE1hyvDkNA4YRjxguKdd11VhPn1dY+YyjuGJshih1Z+MJf iLCht8URY9jbSuHj2vP4Zyz+zydWy5J0+xqE/PQRmvqoot/CHEQgduz/Kc7qo6JpssSG bacO8PVy5UBPcpabft1e46kt2xcuhN+RrU9aE= MIME-Version: 1.0 In-Reply-To: <49917650.1060208@tudelft.nl> References: <1233967681-6378-1-git-send-email-giuseppe.bilotta@gmail.com> <498D9F02.9000107@tremplin-utc.net> <498E46F2.1020401@gmail.com> <4990BE23.8080301@tremplin-utc.net> <4990DEFD.4010700@gmail.com> <49915832.1090501@tremplin-utc.net> <49917650.1060208@tudelft.nl> Date: Tue, 10 Feb 2009 22:28:58 +0100 Message-ID: Subject: Re: [PATCH] lis3lv02d: add axes knowledge of HP Pavilion dv5 models From: Giuseppe Bilotta To: =?ISO-8859-1?Q?=C9ric_Piel?= Cc: =?ISO-8859-1?Q?=C9ric_Piel?= , Palatis Tseng , Pavel Machek , LKML , Andrew Morton Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Feb 10, 2009 at 1:42 PM, Éric Piel wrote: > Can you try something like this in lis3lv02d_poweron(): > adev.read(handle, CTRL_REG2, &val); > val |= CTRL2_BDU | CTRL2_IEN; > + val &= ~CTRL2_BLE; > adev.write(handle, CTRL_REG2, val); Ok, this is getting crazier and crazier. Apparently, setting the CTRL2_BLE bit this way makes no difference, and so does setting it with val |= (i.e. enabling it): in both cases, if I debug lis3lv02d_read_16 I always get 0 in the lower byte and something in the upper byte. By forcefully swapping lo and hi I actually get very little fluctuations in the Y and Z axes (so small that jstest doesn't detect them), but I still get huge (4k) fluctuations in the X axis: apparently, this axis has very small SIGNED fluctuations around the 0, which translate to fluctuations between 255 and 1 Now, the lo/hi thing might be solved some other way (e.g. by checking the BLE bit and relying on its setting instead of trying to force it, and then swapping hi and lo as needed). But how do we solve the SIGNEDNESS of lo ? -- Giuseppe "Oblomov" Bilotta