From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx0b-001b2d01.pphosted.com ([148.163.158.5]:46778 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1725914AbgEGMj7 (ORCPT ); Thu, 7 May 2020 08:39:59 -0400 Subject: Re: [kvm-unit-tests RFC] s390x: Add Protected VM support References: <20200506124636.21876-1-mhartmay@linux.ibm.com> <877dxo6i6v.fsf@linux.ibm.com> From: Christian Borntraeger Message-ID: <2900c255-5ccb-2c99-b376-0b502e7becac@de.ibm.com> Date: Thu, 7 May 2020 14:34:51 +0200 MIME-Version: 1.0 In-Reply-To: <877dxo6i6v.fsf@linux.ibm.com> Content-Type: text/plain; charset="utf-8" Content-Language: en-US Content-Transfer-Encoding: 8bit Sender: linux-s390-owner@vger.kernel.org List-ID: To: Marc Hartmayer , Janosch Frank , kvm@vger.kernel.org Cc: Thomas Huth , David Hildenbrand , Cornelia Huck , Andrew Jones , Paolo Bonzini , linux-s390@vger.kernel.org On 07.05.20 14:30, Marc Hartmayer wrote: > On Wed, May 06, 2020 at 04:03 PM +0200, Janosch Frank wrote: >> On 5/6/20 2:46 PM, Marc Hartmayer wrote: >>> Add support for Protected Virtual Machine (PVM) tests. For starting a >>> PVM guest we must be able to generate a PVM image by using the >>> `genprotimg` tool from the s390-tools collection. This requires the >>> ability to pass a machine-specific host-key document, so the option >>> `--host-key-document` is added to the configure script. >>> >>> Signed-off-by: Marc Hartmayer > > […snip…] > >>> [intercept] >>> file = intercept.elf >>> +pv_support = 1 >> >> So, let's do this discussion once more: >> Why would we need a opt-in for something which works on all our current >> tests? I'd much rather have a opt-out or just a bail-out when running >> the test like I already implemented for the storage key related >> tests... >> >> I don't see any benefit for this right now other than forcing me to add >> another line to this file that was not needed before.. >> > > Okay. So shall I add an option ’pv_not_supported’? Or simply assume that > the actual test cases will handle it? I would suggest to fix the testcases (e.g. do an early exit when we are running secure and we know it does not work)