From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: [PATCH 1/4] i8254: Factor out base class for KVM reuse Date: Wed, 01 Feb 2012 13:43:13 +0100 Message-ID: <4F293361.2000300@siemens.com> References: <1e206a3b5dbd396606099fdbd9abd8c00bbdda31.1328035554.git.jan.kiszka@siemens.com> <4F285FBA.5050607@codemonkey.ws> <4F2861F4.1000604@web.de> <4F292ED1.3060901@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Marcelo Tosatti , qemu-devel , "kvm@vger.kernel.org" To: Paolo Bonzini Return-path: In-Reply-To: <4F292ED1.3060901@redhat.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org Sender: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org List-Id: kvm.vger.kernel.org On 2012-02-01 13:23, Paolo Bonzini wrote: > On 01/31/2012 10:49 PM, Jan Kiszka wrote: >>> Just make the methods that you want to override virtual with the default >>> implementation and then make a KVMPIT that inherits from the PIT and >>> then overrides whatever virtual functions it needs to. >> >> That doesn't sound like the proper design for this purpose. Rather, we >> have an abstract base class that both implementations are derived from. > > Absolutely! > >> If I'm not using QOM properly to achieve this, please tell me. > > It looks good, besides the need to rebase to Anthony's "part 3" changes. Given that this part is not upstream and that the relevant conversion is scripted, I would leave this to Anthony so far. Or what is the schedule? Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux