From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261168AbVCEVTF (ORCPT ); Sat, 5 Mar 2005 16:19:05 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261175AbVCEVTE (ORCPT ); Sat, 5 Mar 2005 16:19:04 -0500 Received: from gprs189-60.eurotel.cz ([160.218.189.60]:17370 "EHLO amd.ucw.cz") by vger.kernel.org with ESMTP id S261168AbVCEVSl (ORCPT ); Sat, 5 Mar 2005 16:18:41 -0500 Date: Sat, 5 Mar 2005 22:17:47 +0100 From: Pavel Machek To: Brice Goglin Cc: kernel list , ACPI mailing list , seife@suse.de, Len Brown Subject: Re: s4bios: does anyone use it? Message-ID: <20050305211747.GF1424@elf.ucw.cz> References: <20050305191405.GA1463@elf.ucw.cz> <422A1FB6.3000504@ens-lyon.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <422A1FB6.3000504@ens-lyon.org> X-Warning: Reading this can be dangerous to your mental health. User-Agent: Mutt/1.5.6+20040907i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi! > >Is there single user of s4bios? It used to work for me 4 notebooks > >ago, but I never really used it. I think I'm the only person that ever > >seen it working, but I could be wrong. Is there anyone using s4bios in > >2.6.11? > > > >If not, I guess we should remove that code from kernel. It is not > >usefull, and it is likely broken. > > Pavel > I always suspend my Compaq Evo N6OOc to disk using "echo 4b > > /proc/acpi/sleep". > I don't remember the reason why I originally did choose this one instead of > S4. > I just checked that S4 and S4Bios work the same on 2.6.11-mm1 (even with my > PCMCIA wireless card connected, great!). > From what I remember, I didn't see any difference between S4 and S4Bios in > recent vanilla kernels. Can you try cat /proc/acpi/sleep? If there's no difference between S4 and S4bios, than you are probably just using plain S4... > By the way, it seems that Radeon makes suspend slower because it needs > to be blanked or something like that. Is there any way to avoid this ? Yes, but it will take quite long to do it properly. pm_message_t framework needs to go in, first. Pavel -- People were complaining that M$ turns users into beta-testers... ...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!