From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [kvm-unit-tests PATCH v1] s390x: add a pgm irq handler and a way to expect them Date: Tue, 30 May 2017 12:59:28 +0200 Message-ID: References: <20170529121736.3055-1-david@redhat.com> <67569aa2-c95f-e7ac-47dd-ed2939a6c3ae@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= , Christian Borntraeger To: David Hildenbrand , Thomas Huth , kvm@vger.kernel.org Return-path: Received: from mx1.redhat.com ([209.132.183.28]:52438 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751270AbdE3K7b (ORCPT ); Tue, 30 May 2017 06:59:31 -0400 In-Reply-To: <67569aa2-c95f-e7ac-47dd-ed2939a6c3ae@redhat.com> Content-Language: en-US Sender: kvm-owner@vger.kernel.org List-ID: On 30/05/2017 10:03, David Hildenbrand wrote: >>> We need "-fno-delete-null-pointer-checks", otherwise trying to access the >>> lowcore at address 0 makes GCC generate very weird code. >> I wonder whether you could get rid of that by using a global variable >> for lc instead? > As far as I remember, I tried that and it didn't change a thing. Will do > a quick test. > -fno-delete-null-pointer-checks is the right thing to do, trying to hide undefined behavior from the compiler leads to misery. :) Paolo