From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:58593) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Ur1jf-0004L7-JV for qemu-devel@nongnu.org; Mon, 24 Jun 2013 04:02:52 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Ur1jc-0000VP-0i for qemu-devel@nongnu.org; Mon, 24 Jun 2013 04:02:51 -0400 Received: from mx1.redhat.com ([209.132.183.28]:33061) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Ur1jb-0000VA-QS for qemu-devel@nongnu.org; Mon, 24 Jun 2013 04:02:47 -0400 Date: Mon, 24 Jun 2013 10:02:42 +0200 From: Kevin Wolf Message-ID: <20130624080242.GA3338@dhcp-200-207.str.redhat.com> References: <1372018066-21822-1-git-send-email-mail@kevin-wolf.de> <87zjugwlp2.fsf@codemonkey.ws> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <87zjugwlp2.fsf@codemonkey.ws> Subject: Re: [Qemu-devel] [PATCH 0/3] multiboot: Fix memory information List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Anthony Liguori Cc: pbonzini@redhat.com, Kevin Wolf , qemu-devel@nongnu.org Am 23.06.2013 um 23:39 hat Anthony Liguori geschrieben: > Kevin Wolf writes: > > > Kevin Wolf (3): > > multiboot: Don't forget last mmap entry > > multiboot: Calculate upper_mem in the ROM > > multiboot: Updated ROM binary > > Reviewed-by: Anthony Liguori > > Do you have a test case that triggered this that you can share? I haven't seen an actual kernel failure myself, this was reported by someone else. But it's easy enough to check with a simple Multiboot kernel that just outputs lower_mem/upper_mem and the mmap. For debugging and fixing I added some throw-away debug code to an existing kernel and compared the fixed version with the output when loaded by GRUB and it matches now. If we can have a real test case for this somewhere, I can write a small kernel to do the check. Not sure where it would fit though - probably kvm-unittests is the closest, even though the test wouldn't really have anything to do with KVM. > I'll apply this after a day or so when others have had a chance to review. Sounds good, thanks. Kevin