From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754912Ab1A2VQn (ORCPT ); Sat, 29 Jan 2011 16:16:43 -0500 Received: from ogre.sisk.pl ([217.79.144.158]:49425 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751513Ab1A2VQm convert rfc822-to-8bit (ORCPT ); Sat, 29 Jan 2011 16:16:42 -0500 From: "Rafael J. Wysocki" To: matthieu castet Subject: Re: [RFC] merge 32/64 bits realmode wakeup code Date: Sat, 29 Jan 2011 22:15:57 +0100 User-Agent: KMail/1.13.5 (Linux/2.6.38-rc2+; KDE/4.4.4; x86_64; ; ) Cc: Linux Kernel list , linux-acpi@vger.kernel.org, "H. Peter Anvin" , Thomas Gleixner , x86@kernel.org References: <4D41EA12.1030402@free.fr> <201101272354.00042.rjw@sisk.pl> <4D447C00.70709@free.fr> In-Reply-To: <4D447C00.70709@free.fr> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-2" Content-Transfer-Encoding: 8BIT Message-Id: <201101292215.57292.rjw@sisk.pl> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Saturday, January 29, 2011, matthieu castet wrote: > Rafael J. Wysocki a écrit : > > On Thursday, January 27, 2011, matthieu castet wrote: > >> Hi, > >> > >> this patch merge 32 and 64 realmode wakeup code : > >> - this mean less ifdef in code > >> - we could remove now unused field in wakeup_header (pmode_*) > >> - this allow to set the first 1 MB NX. > >> > >> Do you think it is safe on all X86 32 bit machine ? > >> > >> Matthieu > >> > >> > >> Signed-off-by: Matthieu CASTET > > > > It _looks_ sane, but it will require some serious testing. I'm not sure, > > though, if we can get enough testing coverage without actually committing > > that change. > > > > How much testing has it received so far? > > > Not too much only qemu with 1 and 2 core and a laptop with 1 cpu. > > Please note that with this patch we can also remove the gdt, idt, ldt stuff from > arch/x86/kernel/acpi/wakeup_32.S That's correct, but those changes can wait. The most urgent thing is to fix the resume regression at hand. I guess we should ask the reporter's in https://bugzilla.kernel.org/show_bug.cgi?id=27472 to test that patch, right? Rafael