From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH 2/7] KVM: Introduce __KVM_HAVE_IRQCHIP Date: Wed, 17 Apr 2013 14:15:27 +0200 Message-ID: <516E925F.6000708@redhat.com> References: <1366133175-23986-1-git-send-email-agraf@suse.de> <1366133175-23986-3-git-send-email-agraf@suse.de> <516E8C49.2040304@redhat.com> <516E8D72.7020907@redhat.com> <516E9033.2030904@redhat.com> <45E3F0D8-DC30-4E79-BCE2-E0840FA392E5@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: kvm-ppc@vger.kernel.org, "kvm@vger.kernel.org mailing list" , Scott Wood , Marcelo Tosatti , Gleb Natapov To: Alexander Graf Return-path: In-Reply-To: <45E3F0D8-DC30-4E79-BCE2-E0840FA392E5@suse.de> Sender: kvm-ppc-owner@vger.kernel.org List-Id: kvm.vger.kernel.org Il 17/04/2013 14:10, Alexander Graf ha scritto: >>> Still, __KVM_HAVE_IRQCHIP is clearly a subset of >>> __KVM_HAVE_IOAPIC; defining one without the other makes no sense >>> and will cause compilation or link errors for trace_kvm_ack_irq. >>> >>> Either we drop it altogether, or we should not break compilation >>> consciously---especially if the problem is so trivial and obvious >>> that you had to think of leaving it out. > I disagree. I actually _want_ to break it on purpose, so we have even > more reason to remove all that useless code if nobody complains. Then remove it in patch 1, and move all the remaining IOAPIC code back to arch/x86 at the end. As things are, you're just leaving someone else to do the work (no matter if it is to fix it, or to zap it). Paolo