From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: [RFC][PATCH -mm 0/4] Hibernation: Arbitrary boot kernel support on x86_64 Date: Mon, 20 Aug 2007 15:10:03 +0200 Message-ID: <200708201510.03734.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-pm-bounces@lists.linux-foundation.org Errors-To: linux-pm-bounces@lists.linux-foundation.org To: pm list Cc: discuss@x86-64.org, Nigel Cunningham List-Id: linux-pm@vger.kernel.org Hi, After the kexec hibernation discussion I started to think what it would take to change the restore code so that a hibernation image could be loaded and restored with the help of a kernel different from the image one. As a result, I have created the following series of patches. Except for that being an interesting problem, my motivation to do it is two-fold. First, it will allow us to eliminate the (annoying) requirement to use the same kernel for hibernation and restore. Second, it will allow us to carry out the restore in a more or less ACPI-compliant way, by using an ACPI-less kernel as the boot one. As you can see in the patches, it doesn't take a lot of code to do that, although conceptually it is a bit compilcated. The first patch is the essential one, as it contains all of the necessary bits. The second one makes the ACPI restore code work if ACPI has not been enabled during boot before loading the image. The last two patches are auxiliary, although they improve the flexibility of the restore code. Comments welcome. Greetings, Rafael -- "Premature optimization is the root of all evil." - Donald Knuth