From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Palmer Subject: Lenovo S10e, charge_full == 0 ?!? Date: Mon, 1 Jun 2009 07:42:20 +1000 Message-ID: <20090531214220.GK27997@hezmatt.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from sasquatch.hezmatt.org ([70.85.129.92]:38930 "EHLO sasquatch.hezmatt.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751563AbZEaVv5 (ORCPT ); Sun, 31 May 2009 17:51:57 -0400 Received: from hezmatt.org (unknown [10.6.66.6]) by sasquatch.hezmatt.org (Postfix) with ESMTP id 49A65282E2B for ; Mon, 1 Jun 2009 07:42:18 +1000 (EST) Content-Disposition: inline Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: linux-acpi@vger.kernel.org [Please Cc me; I'm not subscribed to the list] Hi, I've just bought a Lenovo S10e ideapad, and it's going nicely except that /sys/class/power_supply/BAT0/charge_full is 0, rather than something pleasantly large and reassuring. This is causing battery monitors to produce all sorts of crazy results. The rest of the battery parameters seem OK (nothing abnormally 0 in uevent), including charge_full_design. I've charged and dischaged the battery a couple of times, just to make sure it's not some sort of "initialisation" problem. I've tried with both 2.6.26-2 (out of Debian Lenny), and also 2.6.29-2 (from Debian Sid), and the results are the same. I'd try a git tree kernel, but wihch one? Linus' tree? The Linux ACPI tree? If anyone recognises this problem and can point me at a likely candidate for a fix, I'm happy to try it out, but randomly firing away at kernel trees, given the likely build speed on this little netbook, is likely to be an exercise in frustration. Failing that, does anyone have any pointers as to where I could start in tracking this down? It seems like it'll be a fairly straightforward bug -- the driver's probably just looking in the wrong place for the value, and needs a re-education. I'm completely unfamiliar with ACPI and the kernel code relating to ACPI, though, so perhaps it's a lot more complicated than that. Worst case, if all else fails, I'm thinking that a "fix" might be to check if charge_full == 0, and if so use charge_full_design. It's ugly, but better than the current (hah!) situation. - Matt