From mboxrd@z Thu Jan 1 00:00:00 1970 From: Greg KH Subject: Re: [PATCH 03/15] samsung-laptop: don't handle backlight if handled by acpi/video Date: Wed, 23 Nov 2011 12:30:12 -0800 Message-ID: <20111123203012.GB1355@suse.de> References: <1321999366-23784-1-git-send-email-corentincj@iksaif.net> <1321999366-23784-4-git-send-email-corentincj@iksaif.net> <20111122221633.GA24283@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from cantor2.suse.de ([195.135.220.15]:59126 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756338Ab1KWW2h (ORCPT ); Wed, 23 Nov 2011 17:28:37 -0500 Content-Disposition: inline In-Reply-To: Sender: platform-driver-x86-owner@vger.kernel.org List-ID: To: Corentin Chary Cc: platform-driver-x86@vger.kernel.org, Matthew Garrett , linux-kernel@vger.kernel.org On Wed, Nov 23, 2011 at 07:58:20AM +0100, Corentin Chary wrote: > On Tue, Nov 22, 2011 at 11:16 PM, Greg KH wrote: > > On Tue, Nov 22, 2011 at 11:02:34PM +0100, Corentin Chary wrote: > >> samsung-laptop is not at all related to ACPI, but since this interface > >> is not documented at all, and the driver has to use it at load to > >> understand how it works on the laptop, I think it's a good idea to > >> disable it if a better solution is available. > > > > I wish this would work, but on the machine that I had, there was a valid > > ACPI table for the video device, yet I was told to "ignore it, it > > doesn't work and use the sabi interface instead", so this patch would > > break that machine :( > > Well, it's why acpi_backlight=vendor is here: to enable and use the > vendor backlight instead of using the standard ACPI interface. > And nowadays, even gpu drivers provide a backlight class (at least for > intel, nvidia/nouveau, and maybe radeon. > > > We might want to trigger off of the machine type for this kind of > > thing, that's the only way I can see this working properly. > > But, yes, this is a "regression". Maybe we should do that: > - enable the backlight by default for all SECLINUX models > - disable it by default for swsmi, but let the user be able to enable > it with acpi_backlight=vendor (this is what is done with this patch). That sounds reasonable. thanks, greg k-h