Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 9 Feb 2021 04:23:21 -0500
From:   "Michael S. Tsirkin" <mst@redhat.com>
To:     Jason Wang <jasowang@redhat.com>
Cc:     virtualization@lists.linux-foundation.org,
        linux-kernel@vger.kernel.org, shahafs@mellanox.com,
        lulu@redhat.com, sgarzare@redhat.com, rdunlap@infradead.org
Subject: Re: [PATCH V3 16/19] virtio-pci: introduce modern device module
Message-ID: <20210209042000-mutt-send-email-mst@kernel.org>
References: <20210104065503.199631-1-jasowang@redhat.com>
 <20210104065503.199631-17-jasowang@redhat.com>
 <20210205103214-mutt-send-email-mst@kernel.org>
 <24cb3ebe-1248-3e31-0716-cf498cf1d728@redhat.com>
 <20210208070253-mutt-send-email-mst@kernel.org>
 <fe639f0f-d639-3c3c-e297-042127788aca@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <fe639f0f-d639-3c3c-e297-042127788aca@redhat.com>
Precedence: bulk

On Tue, Feb 09, 2021 at 11:29:46AM +0800, Jason Wang wrote:
> 
> On 2021/2/8 下午8:04, Michael S. Tsirkin wrote:
> > On Mon, Feb 08, 2021 at 01:42:27PM +0800, Jason Wang wrote:
> > > On 2021/2/5 下午11:34, Michael S. Tsirkin wrote:
> > > > On Mon, Jan 04, 2021 at 02:55:00PM +0800, Jason Wang wrote:
> > > > > Signed-off-by: Jason Wang<jasowang@redhat.com>
> > > > I don't exactly get why we need to split the modern driver out,
> > > > and it can confuse people who are used to be seeing virtio-pci.
> > > 
> > > The virtio-pci module still there. No user visible changes. Just some codes
> > > that could be shared with other driver were split out.
> > > 
> > What I am saying is this: we can have virtio-vdpa depend on
> > virtio-pci without splitting the common code out to an
> > extra module.
> 
> 
> Ok.
> 
> 
> > 
> > > > The vdpa thing so far looks like a development tool, why do
> > > > we care that it depends on a bit of extra code?
> > > 
> > > If I'm not misunderstanding, trying to share codes is proposed by you here:
> > > 
> > > https://lkml.org/lkml/2020/6/10/232
> > > 
> > > We also had the plan to convert IFCVF to use this library.
> > > 
> > > Thanks
> > If that happens then an extra module might become useful.
> 
> 
> So does it make sense that I post a new version and let's merge it first.
> Then Intel or I can convert IFCVF to use the library?
> 
> Thanks

Generally it's best if we actually have a couple of users before we bother
with refactoring - it's hard to predict the future,
so we don't really know what kind of refactoring will work for IFCVF ...

> 
> >