From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vojtech Pavlik Subject: Re: Generic battery interface Date: Fri, 28 Jul 2006 01:31:54 +0200 Message-ID: <20060727233154.GB4907@suse.cz> References: <20060727221632.GE3797@elf.ucw.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from styx.suse.cz ([82.119.242.94]:25809 "EHLO mail.suse.cz") by vger.kernel.org with ESMTP id S1750707AbWG0XcI (ORCPT ); Thu, 27 Jul 2006 19:32:08 -0400 Content-Disposition: inline In-Reply-To: <20060727221632.GE3797@elf.ucw.cz> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: Pavel Machek Cc: "Brown, Len" , Shem Multinymous , Matthew Garrett , kernel list , linux-thinkpad@linux-thinkpad.org, linux-acpi@vger.kernel.org On Fri, Jul 28, 2006 at 12:16:32AM +0200, Pavel Machek wrote: > /dev/XXX > + battery layer will need to be invented > - that layer will need maintainer > + maintainer ensures this is not a mess, allocates numbers and figures out which features are identical, which not, etc. This will actually be a lot easier than with keyboards: The number of different battery drivers/types is rather limited fortunately. note: It's absolutely necessary to limit the API to a well usable SUBSET of a superset of the features of all drivers/devices, even sacrificing obscure features to keep the API sane. One example would be the HID Power spec, which simply can't be supported to full extent by any sane API. > - does not quite suit zaurus-style batteries, that can _only_ be > polled. Can be worked-around by polling from kernel (and we > are often doing that, anyway) + and the kernel can change the polling frequency based on power saving state changes > + userspace backdoor interface will need to be invented, /dev/uinput style > - hard to handle obscure features > (do_not_charge_battery_for_X_minutes), and we'll probably want > to unify batteries a bit, probably loosing precision. -- Vojtech Pavlik Director SuSE Labs