From mboxrd@z Thu Jan 1 00:00:00 1970 From: john cooper Subject: Re: [Qemu-devel] [PATCH 2/2] Add serial number support for virtio_blk, V3 Date: Wed, 27 May 2009 11:06:26 -0400 Message-ID: <4A1D56F2.5090207@redhat.com> References: <4A1C88DE.6050608@redhat.com> <20090527074919.GB7356@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: KVM list , qemu-devel@nongnu.org, rusty@rustcorp.com.au, john.cooper@redhat.com To: Christoph Hellwig Return-path: Received: from mx2.redhat.com ([66.187.237.31]:38007 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1762995AbZE0PIq (ORCPT ); Wed, 27 May 2009 11:08:46 -0400 In-Reply-To: <20090527074919.GB7356@lst.de> Sender: kvm-owner@vger.kernel.org List-ID: Christoph Hellwig wrote: > This looks functionally correct, but pretty far from normal kernel coding > style. I tend to avoid 'goto's. Christoph Hellwig wrote: >> /* >> * IDE-compatible identify ioctl. >> * >> * Currenlyt only returns the serial number and leaves all other fields >> * zero. >> */ > > Btw, thinking about it the rest of the information in the ioctl should > probably be filled up with faked data, similar to how we do it for > the ide emulation inside qemu. Doing so crossed my mind but thought it may be better to start here and provide data on an as-needed basis. But as you point out there is precedent (and likely reason) for hw/ide.c:ide_identify() doing as such. I don't have a strong bias either way. Comments from others? -john -- john.cooper@redhat.com