From mboxrd@z Thu Jan  1 00:00:00 1970
From: Stephen Hemminger <stephen@networkplumber.org>
Subject: Re: [PATCH 04/13] eal: introduce driver type
Date: Tue, 20 Dec 2016 09:09:47 -0800
Message-ID: <20161220090947.2543c09e@xeon-e3>
References: <20161219215944.17226-1-sthemmin@microsoft.com>
 <20161219215944.17226-5-sthemmin@microsoft.com>
 <CALe+Z02fh5iw4ZAf5Qxikz9i32Cyig0OOZBqoUQJcq+6JB2zUg@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Cc: dev@dpdk.org, Stephen Hemminger <sthemmin@microsoft.com>
To: Jan Blunck <jblunck@infradead.org>
Return-path: <dev-bounces@dpdk.org>
Received: from mail-pg0-f51.google.com (mail-pg0-f51.google.com [74.125.83.51])
 by dpdk.org (Postfix) with ESMTP id 3C127FBC5
 for <dev@dpdk.org>; Tue, 20 Dec 2016 18:09:58 +0100 (CET)
Received: by mail-pg0-f51.google.com with SMTP id f188so74890703pgc.3
 for <dev@dpdk.org>; Tue, 20 Dec 2016 09:09:58 -0800 (PST)
In-Reply-To: <CALe+Z02fh5iw4ZAf5Qxikz9i32Cyig0OOZBqoUQJcq+6JB2zUg@mail.gmail.com>
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On Tue, 20 Dec 2016 14:00:55 +0100
Jan Blunck <jblunck@infradead.org> wrote:

> On Mon, Dec 19, 2016 at 10:59 PM, Stephen Hemminger
> <stephen@networkplumber.org> wrote:
> > Since multiple buses and device types need to be supported.
> > Provide type field in driver.
> > ---
> >  lib/librte_eal/common/include/rte_dev.h  | 15 ++++++++++++---
> >  lib/librte_eal/common/include/rte_pci.h  |  1 +
> >  lib/librte_eal/common/include/rte_vdev.h |  1 +
> >  3 files changed, 14 insertions(+), 3 deletions(-)
> >
> > diff --git a/lib/librte_eal/common/include/rte_dev.h b/lib/librte_eal/common/include/rte_dev.h
> > index e5471a22..3f4e26e6 100644
> > --- a/lib/librte_eal/common/include/rte_dev.h
> > +++ b/lib/librte_eal/common/include/rte_dev.h
> > @@ -144,12 +144,21 @@ void rte_eal_device_insert(struct rte_device *dev);
> >  void rte_eal_device_remove(struct rte_device *dev);
> >
> >  /**
> > + * Type of device driver
> > + */
> > +enum rte_driver_type {
> > +       PMD_VIRTUAL,
> > +       PMD_PCI,
> > +};
> > +
> > +/**
> >   * A structure describing a device driver.
> >   */
> >  struct rte_driver {
> >         TAILQ_ENTRY(rte_driver) next;  /**< Next in list. */
> > -       const char *name;                   /**< Driver name. */
> > -       const char *alias;              /**< Driver alias. */
> > +       const char *name;              /**< Driver name. */
> > +       const char *alias;             /**< Driver alias. */
> > +       enum rte_driver_type type;     /**< Driver type. */
> >  };  
> 
> I wonder who is the user of the type. It can't be the driver itself
> because it must be aware of what kind of low-level interface it
> serves. So this seems to be a helper to allow users of rte_driver to
> upcast to the object that embeds the rte_driver at runtime. I don't
> believe that this is a good interface because it often leads to ugly
> and error prone code.

The thing I ran into was the test code. The testpmd (and related code) is an
example of how some code might want to use dev_info_get to find out about a device
then take some action based on the driver type.  Testpmd has code that manipulates
PCI registers. Right now it is broken when run on virtual devices.