From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gerd Hoffmann Subject: Re: [PATCH] kvmtool: handle x86 firmware smaller than 128k Date: Tue, 07 Nov 2017 08:33:51 +0100 Message-ID: <1510040031.24424.3.camel@redhat.com> References: <20171106114837.6882-1-kraxel@redhat.com> <3b1b32c5-7b71-7b9d-ab00-a7610b564fcc@arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit To: Jean-Philippe Brucker , "kvm@vger.kernel.org" Return-path: Received: from mx1.redhat.com ([209.132.183.28]:60114 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755205AbdKGHdx (ORCPT ); Tue, 7 Nov 2017 02:33:53 -0500 In-Reply-To: <3b1b32c5-7b71-7b9d-ab00-a7610b564fcc@arm.com> Sender: kvm-owner@vger.kernel.org List-ID: On Mon, 2017-11-06 at 17:09 +0000, Jean-Philippe Brucker wrote: > Hi Gerd, > > This change deserves a comment, as it's not obvious what we're doing > here. > How about: > > """ > When using an external firmware on x86, kvmtool follows the legacy > BIOS > standard, by entering firmware at address 0xffff0 in 16bit real mode. This has nothing to do with the legacy bios standards. The cpu starts executing at 0xffff0 after reset, no matter what the firmware is. It's the same for uefi and coreboot. But, yes, mentioning the reset vector is reasonable, it you don't know that x86 detail it isn't obvious indeed. Sending v2 in a moment. cheers, Gerd